CHAPTER 1

INTRODUCTION
1.1. Introduction
Remote sensing (RS) is the science of using electromagnetic radiations (EMR) to identify
earth surface features and estimation of their geo- physical nature during analysis and
interpretation of its spectral, spatial, and temporal signature without doing physical contact with
the object. RS uses satellite imagery technology to sense the landcover of Earth. At the early of
stage 21st century, satellite imagery became widely available. Currently, various type of orbiting
satellites are found to collect crucial data for improving knowledge about Earth's atmosphere,
oceans, ice, and land. Orbital platforms used different regions of Electromagnetic Spectrum
(EMS) to provide interest data, which contributed with larger scale aerial or ground-based
sensing and analysis to prepare satellite images that picture enough information needed to
monitor trends of natural phenomena. Satellite images are rich material that plays a vital role in
providing geographical information. There are specialized RS applications are employed for
interpreting and analyzing the satellite images in order to understand the behavior of considered
phenomena.
Satellite Image Classification (SIC) is the most significant technique used in RS for the
computerized study and pattern recognition of satellite information, which is based on diversity
structures of the image that involving rigorous validation of the training samples depending on
the used classification algorithm [1]. The powerful of such algorithms is depends on the way of
extracting the information from huge number of data found in images. Then, according to these
information, the pixels are grouping into meaningful classes that enable to interpret and studying
various types of regions that included in the image [2].
1.2. Remote Sensing
Remote sensing is relevant process that permits to extract information about the surface
of land without actual contact with the area being observed [3]. The remotely sensed images are
the aerial and satellite images that allow exact mapping of landcover and make landscape traits
comprehensible on regional, continental, and even global scales [4]. Figure 1.1 shows the
principle of remote sensing, the solar radiation incident on objects found at the landcover, which
are reflect these radiation toward the satellite, the satellite technology receive these radiation and
record the image of the considered area [5].
Remote sensing operates in several region of the electromagnetic spectrum. The
ultraviolent (UV) portion of the spectrum has the shortest wavelength that is practical use of
remote sensing. It is used to dig out a general picture for the object embedded underground.
Passing through the visible spectrum, the infrared radiation (IR) is also used to record thermal
images that are useful in climate consideration. Actually, satellites images are taken through the
visible bands are usually employed in the appearance of digital images containing valuable
information used to assist developers to understand the information of the landcover. The aspires
of satellite images categorization is to split image into discrete classes. The consequential
classified image is a thematic map of the landscape image [5]. The satellite imagery is
categorized into active and passive remote sensing when information is merely recorded [6];
more details about them are given by the following subsections:
1.2.1. Active Remote Sensing
Active remote sensing emits radiation to scan objects and areas whereupon a sensor then
detect and measures the radiation that is reflected or backscattered from the target as shown in
Figure (1.2) [7]. (Radio Detection and Ranging) RADAR and (Light Detection and Ranging)
LiDAR are examples of active remote sensing where the time delay between emission and return
signal is measured, which help to establish the location, speed and direction of the target object
on the ground [7].

Figure 1.1: Principle of remote sensing

 1.2.2. Passive Remote Sensing
Passive sensors gather natural radiation that is emitted or reflected by the object or
surrounding areas. The most common source of radiation measured by passive sensors is
reflected sunlight as shown in Figure 1.3 [7]. The common examples of passive remote sensing
are satellite imagery, film photography, and radiometers [8].

Figure 1.2: Active sensor system of remote sensing

Figure 1.3: Passive sensor system of remote sensing

1.3. Satellite Imagery
Satellites are greatly used in the imagery purposes of remote sensing; they have several
unique characteristics enable to sense the Earth's surface remotely. The satellite images are
produced by sensing the electromagnetic energy at different wavelengths that reflected by
objects. Visible satellite images are made via satellite signals received by visible channels that
senses reflected solar radiation, such that, it is available only during daylight [9]. The major
benefit of using the visible imagery is due to its ability to give higher resolution images than
other imagery bands. Thus, smaller features can be distinguished at visible imagery. The problem
faces visible imagery is that clouds are white, while land and water surfaces are shaded, which
impedes the process of satellite imagery. This problem is overcome by using radar sensors that
uses microwaves to create satellite image, such sensor is enabled to see through clouds and in
night [9]. On the other hand, optical sensors uses multispectral, hyper-spectral and multi-
polarization technique that operated at different bands, which employed to improve the detection
of objects under the sea or ground.
Many satellite platforms that used for imagery purposes are designed to follow a
determined orbit in direction from north to the south of the earth, which corresponding to the
Earth's rotation from the west to east. This setting of satellites allows them to cover most of the
earth's surface over a certain period of time, where the coverage area is called swath [8]. The
Landsat (Land Satellite) provides the longest continuous record of satellite based observations. It
is a primary source of medium spatial resolution Earth observations and monitoring global
change .Observation requires revealing both natural and human-induced landscape changes,
Landsat provides the only inventory of the global land surface over time on a seasonal basis [10].
1.4. Landsat Satellites
Landsat was designed in the 1960s and launched in 1972, The Landsat system is
important in its own right as a remote sensing system that has contributed greatly to Earth
resources studies. Landsat was proposed by scientists and implemented by U.S. government to
satisfy applications of remote sensing to survey of the Earth’s land areas. Then, Landsat became
a project to lunch series of satellites as a part of National Aeronautics and Space Administration
(NASA) Mission to Planet Earth that managed by the Goddard Space Flight Center in Greenbelt,
which is a one of NASA's high priority programs to help further the understanding of the Earth
system. The history of Landsat series is shown in Figure 1.4 [11].
The first Landsat sensor (Landsat-1) was known as the Earth Resources Technology
Satellite (ERTS), it recorded energy in the visible and near infrared spectral. ERTS was practical
for observation of Earth resources from satellite altitudes; it gives sufficient proportion of scenes
that free of cloud cover. The Landsat system consists of spacecraft borne sensors that observe the
Earth and then transmit information by microwave signals to ground stations that receive and
process data for dissemination to a community of data users. Early Landsat vehicles carried two
sensor systems: The Return Beam Vidicon (RBV) and the Multispectral Scanner Subsystem
(MSS). The RBV was a camera like instrument designed to provide relative to the MSS, high
spatial resolution and geometric accuracy but lower spectral and radiometric detail.

Figure 1.4: History of Landsat

1.5. Satellite Image Resolution
Satellite image resolution is the ability of sensor to monitor the smallest object clearly
with distinct boundaries [12]. Resolution often refers to a pixel count in digital imaging. It is
described by the number of pixel columns (i.e., width of image) and the number of pixel rows
(i.e., height of image). Pixel resolution indicates the squared area on the ground that covered by
each pixel in the image, which is measured by one side of this square area. In addition, there are
four types of satellite image resolution may be taken in account, they are: spatial, spectral,
radiometric, and temporal resolutions. More details for each type are given in the following
subsections:
1.5.1. Spatial Resolution
Spatial resolution indicates the size of the smallest possible feature that can be pictured.
In a digital image, the resolution is limited via the pixel size, i.e. the smallest resolvable object
cannot be smaller than the pixel size. The essential resolution of an imaging system depends on
their Instantaneous Field of View (IFOV) of the sensor, which is a measure of the ground area
viewed by a single detector element in a given instant in time. The pixel size is determined by
the sampling distance [13].
 1.5.2. Spectral resolution
Spectral resolution can be defined as the number of spectral bands used to record
spectrally split radiative energy received from the target. Many remote sensing systems record
energy over several separate wavelength ranges at various spectral resolutions. Advanced multi-
spectral sensors called hyperspecteral sensors, which detect hundreds of very narrow spectral
bands throughout the visible, nearinfrared, and mid-infrared portions of the electromagnetic
spectrum.
1.5.3. Radiometric Resolution
Radiometric characteristics refer to the smallest change in intensity level that can be
detected by the sensing system. A fine radiometric resolution is critical in studying targets that
have only a subtle variation in their reflectance, such as detection of different kinds of minerals
in the soil and varying levels of vegetation stress caused by drought and diseases. Also, remotely
sensed data of a fine radiometric resolution are especially critical in quantitative applications in
which a ground parameter (e.g., sea surface temperature and concentration level of suspended
solids in a water body) is retrieved from pixel values directly. Data of a higher quantization level
enable the retrieval to be achieved more accurately while a coarse radiometric resolution causes
the pixels to look similar to each other.
1.5.4. Temporal resolution
Temporal resolution indicates the period at which the same ground area is sensed by the
same sensing system. The actual temporal resolution is typically measured by days, it depends on
three factors: satellite capabilities, swath overlap, and latitude. Since remote sensing satellites are
revolving around the Earth 24 hours a day and 365 days a year, the temporal resolution is
directly related to the satellite orbital period. A short period means more revolutions per day and
is equivalent to a high temporal resolution.
1.6. Motivation and challenges
The continuous proliferation and improvement of remote (satellite and aerial) data
sensors yields a huge volume of Earth’s images with high spatial and temporal resolution, as well
as with rich spectral information. For instance, the constellation of Pléiades satellites revisits the
entire Earth every day, acquiring optical data with a spatial resolution of less than a meter. These
data contain a valuable source of information, which opens the door to a large range of important
applications, such as the monitoring of natural disasters, the planning of urban environments and
precision agriculture. However, petabytes of these massive images are stored into binary files as
unstructured raw data. As a result, a large part of them is never used. Thus, it has become crucial
to develop methods to automatically analyze these data.
One of the most important problems for the automatic interpretation of remote sensing
data is image classification, which consists in the assignment of a class, or thematic label (for
example, building, tree, car, etc.), to each pixel in the image. Figure 1.5 shows an example of
such pixel-wise classification of an aerial image into six classes of interest. In this work, I focus
on supervised classification, in which the learning approaches are used to train the classifier from
labeled reference data. In the manuscript, the terms labeling and classification are used
interchangeably. I also refer to dense classification or semantic classification to denote the
labeling of individual pixels in the image. The term semantic is typically used to emphasize that
the classes of interest represent object types (e.g., car, building) rather than low-level physical
materials (e.g., metal sheets).

Figure 1.5 Example of pixel-wise classification of an aerial image.

While the first classification approaches have been designed more than forty years ago
with the appearance of the first satellites [14], recent technological developments raise new
challenges. The first challenge consists in adapting classification methods to the characteristics
of new sensors, since the state-of-the-art methods may be not directly applicable or may not give
satisfactory performances. For instance, first satellite sensors captured data of a low spatial
resolution, and significant research was focused on discovering the proportions (or abundances)
of physical materials within each single pixel [15]. The recently developed sensors acquire
images with a high spatial resolution, where pixels become small when compared to the objects.
This implies the need of object-based analysis, i.e., the development of the algorithms which
learn to analyze the consistency and relationship between groups of pixels in the image, aiming
at identifying the whole objects.
The second challenge is the scalability of the systems. With the increasing amount and
openness of the data covering a large geographic extent, there is a growing interest in developing
scalable methods able to produce classification maps of large areas, in the scale of cities,
countries, continents and up to the entire Earth [16]. While there is an urgent need to design
computationally efficient algorithms to process large-scale data, it is also a challenge to design
methods that would produce accurate maps for dissimilar areas of the Earth, given the large
variability of the appearance of landscapes across the planet, i.e. large intra-class variability (see
Figure 1.6). While the first classification methods in remote sensing tried to distinguish different
classes of low-level materials and substances (e.g., water, soil, gravel), current research studies
are in a big part focused on semantic segmentation, where high-level contextual reasoning (e.g.,
shape and context analysis) must be performed to discriminate more abstract semantic classes. In
this manuscript, I present my research studies on the development of learning methods for
remote sensing image classification. In the following, I first summarize the state-of-the-art
approaches, and then detail my recent contributions.
1.7. Approaches, trends and contributions
The pioneering classification approaches for remote sensing imagery consider each pixel
as a pattern and all the data collected by sensors for this pixel are considered as the initial set of
features (for instance, spectrum values in the case of optical imagery). Then, the standard pattern
recognition scheme [17] is applied to classify each pixel:
1. The classification model and its specifications are selected (e.g., neural network,
support vector machines (SVM), etc.).
2. Since the initial set of features may be redundant, feature extraction/selection is often
performed, aiming at obtaining the features which discriminate at best classes of
interest.
3. The next step, called “training of the classifier,” consists in partitioning the entire
feature space into K exhaustive, non-overlapping regions, so that every point in the
feature space is uniquely associated with one of the K classes. This is done by feeding
into the classification model examples of samples for each of the classes (so called
training samples).
 Once the training step is accomplished, every new pixel can be classified according to its
feature set. Such classification approach is often referred to as pixel-wise, where the term pixel-
wise has the different meaning than in the previous section: here, by pixel-wise we mean that to
classify any pixel, no information from its neighborhood is taken into account. Different models
have been applied for pixel-wise classification in the remote sensing community; among them,
probably the most widely used ones are neural networks [18, 19, and 20], SVM [21, 22] and
random forests [23, 24].
In a real image, neighboring pixels are related or correlated, both because imaging
sensors acquire significant amount of energy from adjacent pixels [25] and because objects
occurred in the image scene are in most cases large when compared to the size of a pixel. For
instance, if a given pixel in the agricultural image scene represents a corn, its adjacent pixels
belong with the high probability to the corn field. This spatial contextual information
significantly helps to interpret image scene. Therefore, most recent classification approaches are
spectral-spatial, i.e. they classify each image pixel based on: 1) its own spectral values (the
spectral information) and 2) information extracted from its neighborhood (the spatial
information). The use of spectral-spatial classifiers is especially rewarding when processing
images with high spatial resolution; that is why these techniques are especially relevant to
analyze data captured by modern and near-future sensors.

Figure 1.6: Examples of aerial images around the Earth, where large intra-class variability can be
appreciated for ‘building’ class.
 In the following, I will present three important families/strategies of spectral-spatial
classification, which are successfully used until nowadays and can also be used in combination
with each other, and will position my contributions with respect to these strategies:
1. Graph-based Methods: In these approaches, an image is represented as a graph, where
the nodes typically represent image pixels or regions, while edges either connect adjacent
regions [26, 27], or keep track of a hierarchy in a multi-scale image representation [28,
29]. By applying different graph construction and pruning strategies, these techniques
typically output a disconnected graph, where each connected sub graph corresponds to
the classified object in the resulting classification map. A popular graph-cut technique
applies a min-cut algorithm on the image graph to regularize outputs of the pixel-wise
classifier, by minimizing the Markov Random Field-based energy function associated
with the image and the initial classification result [30]. Another approach consists in
building first a hierarchy of image regions at different levels of details, based on standard
homogeneity/dissimilarity criteria, and then selecting from this hierarchy the regions at
different scales that correspond to the objects in the final classification map [28, 31].
Such hierarchical data structure can be very flexible, for instance, it can be easily
augmented to include textural or shape information for more accurate image analysis.
2. Feature Engineering: The spatial information can be included into a classifier by
analyzing the similarity/correlations between the data in the neighborhood of an image
pixel. The initial approaches would typically use a well-defined algorithm, which can be
often described by a (set of) pre-defined convolutional mask(s), for this purpose. For
instance, Tsai et al. [32] and Huang and Zhang [33] have investigated the use of texture
measures derived from the gray level co-occurrence matrix, such as angular second
moment, entropy and homogeneity, for spectral spatial classification. A popular approach
in remote sensing explores the principles of mathematical morphology, which by
definition aims at analyzing spatial relations between sets of pixels, i.e., extracting
information about the size, shape and orientation of structures. For this purpose,
morphological profiles [34], extended morphological profiles [35], attribute profiles [36]
and extinction profiles [37] have been successfully applied. In all cases, the core idea is
to progressively simplify image based on a pre-defined rule/attribute, which allows
analyzing a particular characteristic (for instant, area or standard deviation of regions)
 within this image. The extracted morphological features typically lay in a very-high-
dimensional space, and additional feature extraction must be applied to reduce the
dimensionality of the feature space. Another group of techniques introduces strong shape
priors into classification: for example, in remote sensing images buildings can be
modeled as rectangles [38, 39] and roads can be modeled by line segments [40].
3. Deep learned features: Even though feature engineering gives satisfactory results for
many applications, when the complexity of the data and the number of classes increases,
it is not straightforward to design hand-crafted features that would exhibit high
discriminative power for all classes. The recent trend in the state-of-the-art consists in
designing classification systems that would be able to automatically learn hundreds of
expressive high-level contextual features. In this context, the use of deep learning, in
particular convolutional neural networks, is intensively investigated by both image
analysis and remote sensing communities. While some works use deep learning-based
features as the input to a conventional classifier, such as SVM or logistic regression
classifier [41], most recent studies design an end-to-end system which jointly learns to
extract relevant contextual features and conduct the classification [42, 43].
 CHAPTER 2
LITERATURE SURVEY
In this chapter, we discuss about the various types of satellite image classification
approaches and also discuss few notable works.
2.1. Satellite Image Classification
There are several methods and techniques for satellite image classification. Satellite
image classification methods can be broadly classified into three categories: automated, manual,
and hybrid as shown in Figure 2.1 [44]: The following subsections illustrate each category of
classification:

Figure 2.1: Satellite image classifications methods hierarchy

2.1.1. Automated Methods
Automated satellite image classification methods use algorithms that applied on the
satellite image to systematically group pixels into meaningful categories. Majority of
classification methods fall under this category. Automated satellite image classification methods
further classified into two categories supervised and unsupervised classification methods [45];
Figure 2.2 shows sequenced stages of both supervised and unsupervised that given in the
following subsection.

Figure 2.2: Unsupervised and supervised classification principle

A. Unsupervised Classification
Unsupervised classification is a process of clustering satellite image pixels into unlabeled
classes/clusters. The image classification concepts aids in unsupervised classification to provide
an access into image features for clustering, which can be analyzed in term of classification
quality. The result of an unsupervised classification is an image of statistical features clusters.
There is several different unsupervised classification algorithms are commonly used in remote
sensing. Most common ones of them are: isodata, support vector machine (SVM), and K-Means
classification method that may be used to establish the dataset of the images that classified for
first time [45].
B. Supervised Classification
Supervised classification is a process of sorting pixels into a number of individual classes
based on the comparison with training sets [46]. Such methods require training set to be
implemented, the data set predefined region of image belong to different classes in the image.
The enrollment of the data set is an important primary stage in the supervised satellite image
classification methods. The quality of a supervised classification depends on the quality of the
training set. The supervised classification usually has the following sequence of operations:
defining of the training sites, dataset (or, signatures) extraction, and image classification [47].
Major supervised classification methods use the statistical techniques like Artificial Neural
Network (ANN), Binary Decision Tree (BDT) and Image Segmentation [45].
2.1.2 Manual Methods
Manual satellite image classification methods are robust, effective and efficient methods.
This type of methods consumes more running time. In manual methods the efficiency and
accuracy of the classification depend on knowledge and familiarity of the analyst towards the
field of study [45].
2.1.3. Hybrid Methods
Hybrid satellite image classification methods combines the advantages of automated and
manual methods. Hybrid approach uses automated satellite image classification methods to get
initial classification, further manual methods are used to refine classification and correct errors
[44].
2.2. Literature Survey
Several studies have been published in the field of interest; they mostly cover the use of
satellite image classification methods. The following selected researches are the most interesting
ones:
Rhonda, et al (2009) introduced a feature reduction method based on the singular value
decomposition (SVD). This SVD-based feature reduction method reduces the storage and
processing requirements of the SVD by utilizing a training dataset. This feature reduction
technique was applied to training data from two multitemporal datasets of Landsat TM/ETM+
imagery acquired over a forested area in Virginia, USA and Rondônia, Brazil. Subsequent
parallel iterative guided spectral class rejection (pIGSCR) forest/non-forest classifications were
performed to determine the quality of the feature reduction. The classifications of the Virginia
data were five times faster using SVD-based feature reduction without affecting the classification
accuracy. Feature reduction using the SVD was also compared to feature reduction using
principal components analysis (PCA). The highest average accuracies for the Virginia dataset
(88.34%) and for the Rondônia dataset (93.31%) were achieved using the SVD. The results
presented here indicate that SVD-based feature reduction can produce statistically significantly
better classifications than PCA [48].
Sathya P. and Malathi L. (2011) established a segmentation and classification of remote
sensing images. This classified image is given to KMeans algorithm and Back Propagation
algorithm of ANN to calculate the density count. The density count is stored in database for
future reference and for other applications; also it has the capability to show the comparison of
the results of both the algorithms. The experimental result found that K-means algorithm gives
very high accuracy, but it is useful for single database at a time. The neural network provides
good accuracy, and it found useful for multiple databases [45].
Kumar et.al. (2012) proposed an improved contrast enhancement technique of multiband
satellite image based on the singular value decomposition (SVD) and discrete cosine transform
(DCT). Normalized difference vegetation index (NDVI) technique was used to extract features
of low-contrast satellite images. The visual and quantitative results proved the increased
efficiency and flexibility of the proposed method. The simulation results show that the
enhancement-based NDVI using DCT-SVD technique is highly useful to detect the surface
features of the visible area which are extremely beneficial for municipal planning and
management [49].
Aras (2014) Classified Landsat satellite images of the area of Laylan district in Kirkuk
province, for the years 1990, 2000 and 2013 by using (ENVI 4.7, Matlab), several different
methods of classification have been used to classify area of these images; like unsupervised
method (ISODATA and K-Means), supervised methods (Mahalanibos distance classifier and
Maximum likelihood Classifier) and scatter plot method to get the most accurate results and then
to compare the results of each method. The result showed a growth in urban land between 1990
and 2013 while the vegetation witnessed a reduction in this period. The important observation is
that the water bodies was not exist in selected area at the considered period but it appeared in
2013 image where the positive change happened in the top-left part of the selected area after the
build of Shirin dam and collecting water there.
Habib et al. (2014) used a Zernike moments-based descriptor as a measure of shape
information for the detection of buildings from Very High Spatial Resolution (VHSR) satellite
images. The proposed approach comprises three steps. First; the image is segmented into
homogeneous objects based on the spectral and spatial information. Second; Zernike feature
vector is computed for each segment. Finally, a Support Vector Machines (SVM) based
classification using the feature vectors as inputs is performed. Experimental results and
comparison with Environment for Visualizing Images (ENVI) commercial package confirm the
effectiveness of the proposed approach [50].
Harikrishnan and Poongodi S. (2015) implemented a cellular with fuzzy rules for
classifying the satellite image and analyzed the quality of classified image. cellular automata are
implemented for simulation of satellite images and also cellular automata relates to
categorization in satellite image is used simultaneously. Where the accuracy rate of cellular
automata is very much increased when fuzzy rules are implemented to this systems, uncertain
pixels which are still present in the classification process is eliminated and uncertain pixel is
classified to each class its give paths to well classified image with high accuracy [51].
Akkacha et al (2015) proposed a combination of three classification methods which are
K-means, LVQ (linear vector quantization) and SVM (support vector machine). The
unsupervised k-means technique to get an initial view of the image site that will help to choose
the best training area for SVM classification, which is the second kind. The last kind is the
neuronal supervised LVQ technique. LANDSAT7 remote sensing images of the Ouargla oasis
area were used. The results showed the vegetation separated from urban landscape. The neuronal
LVQ method is also used in order to improve the opinion of base classifier when ensemble
classification is done. It was noticeable that SVM gives the best results when using simple
majority vote technique [52].
Noor Z. (2015) established a satellite image classification based on semantic indexing
techniques. After assurance the acceptable behavior of the classification, the results are evaluated
and discussed for Landsat satellite image of Razaza Lake and the area surrounding district in
Karbala province are classified for years 1990,1999 and 2014. It was found that water and hills
in decrease, while vegetation, wet land and barren land in increase for years 1999 and 2014;
comparable with 1990 and the results of proposed methods proved it's superiority, where the
classification accuracies for the proposed method Singular Value Decomposition are 92.5%,
89.5% and 90% for years 1990,1999,2014, respectively, while it were 92%, 89% and 91% for
unsupervised classification method based mean value in years 1990,1999,2014 receptively.
Brindha S. (2015) proposed a satellite image enhancement using DWTSVD, in which the
image is segmented using the multiregional Resolution and the Markov Random Field (MRR-
MRF) Model, this technique decomposes the input image into four frequency sub band by using
DWT and also calculate the Singular Value Matrix of the Low-Low Sub band images and
equalize the enhanced images through Inverse Discrete Wavelet Transform. After the
enhancement process, image segmentation technique is applied to the image. MRR-MRF
techniques to improve the segmentation accuracy, quality of the image, and reduce the noise of
the satellite images. The experimental result showed better performance and high accuracy when
compared with other methods [53].
High-resolution remote sensing image segmentation with fuzzy classification technique is
introduced in the paper [54]. The technique is based on multi-thresholds method. During the
Image segmentation edges of the image are identified and derived with the algorithms having
homogeneity definitions. This technique binds the items based on an optimization function by
using the region merging method.
In paper [55] ANN is used to perform the Edge detection. Most frequently used Initial
step in Image processing to perform action recognition is Edge detection. A NN learns by
altering the weights between the dissimilar layers. The ANN model used has 1 input layer, 1
output layer, and 1 hidden layer. The input layer consists of 4 neurons presented as a 2X2 input
window. The hidden layer has 12 hidden neurons with a single neuron in the output layer. The
output layer gives the edge detected. A 2X2 sliding window is used to perform the Edge
detection. The 2X2 window is provided with 4 crisp values all these crisp values are later
mapped as four inputs for the FIS. The pixel is grouped as Edge / Non-edge by the ANFIS
method. The performance is evaluated using PSNR () and PR () parameters. The best quality of
edge detection happens when PR value is highest. This technique results in better classification
and edge detection than those of Sobel and Canny. The method results in well-defined and clear
edges.
Paper [56] presents a hybrid method for training the classifier. It uses a Fuzzy-fusion
Interface to classify the satellite images. The technique follows certain steps. The first step is
fuzzification- It includes the creation of membership function using the training data set and the
histogram of them. This set is used to perform the fuzzification of the bands. The second step
detects the land cover in the images by creating a set of rules which use all the input variable.
The last step is the formation of the new fuzzy-interference system. This system uses a Sugeno
logic to test the developed scheme with 4 operators. This scheme results in the evaluation of
operators, with high accuracy, precise classification when compared to other methods.
This paper [57] compares the two classification techniques for land cover classification
the Maximum Likelihood classification and the artificial neural networks. The study classifies
five land cover classes as Forest, Grassland, Urban, Water, and cloud. The performance of the
classification is done in terms of Accuracy, Error matrix and Kappa coefficients. Maximum
Likelihood is a popular classification technique using the Gaussian data distribution of data. The
ANN is a recent algorithm for classification and is a non-parametric classifier. The results
include Accuracy, Error matrix and Kappa coefficients. The Accuracy is defined as the
probability of error in classification. The Error matrix is defined as the accuracy measure
between the classified images and the training set data for the same image. The overall
classification for ANN is 93.5% and Kappa value is 0.909. The Maximum likelihood has an
overall accuracy of 80.5% and 0.722 kappa coefficient.
In the paper uses the Backpropagation(BP) neural network for designing of an electric
field classifier. This electric field classifier is used to extract the ultra-low frequency (ULF)
section waveform data from the Wenchuan earthquake. They obtained the mean value, mean
square error, skewness, and kurtosis of an electric field component [58]. The results of this can
be summarized as follows:
1) The BP neural network classifier is used to classify the ULF electric signal data.
2) The effective classification of the electric field signal takes place.
4) BP classifiers employ multiple learning rates hence has high-speed convergence and less
target error.
The paper [59] presents a study which aims to evaluate the capability of Artificial Neural
Network system (ANNs) with the backpropagation algorithm and K-means algorithm with
different approaches for classification of the satellite images. The ANN’s classifier is compared
with two conventional classifiers which is Maximum Likelihood (ML) and unsupervised
(ISODATA). Neural network classification depends on the training data set for the correct
classification. The image is classified into LU/LC (Land use /Land cover). The feed-forward
network with one hidden layer and Back-propagation Architecture is implemented for Image
Classification. The results are determined by the Accuracy Assessments and Kappa coefficient
value. The confusion matrix helps in calculating the percentage of accuracy.
 In the paper [60] it involves two aims that have emerged in neural networks i.e the
increase in the ability to understand the behavior of nervous system and thus taking inspiration
from this knowledge and building up of new systems to perform some related tasks presents a
comprehensive overview of modeling, simulation and implementation of neural networks. It also
includes the growth and evolution in different aspects of neural networks. The papers enlighten
the fact that how neural network has contributed to the development of various concepts such as
Neuro-engineering, Computational Intelligence, Computational Neuroscience and Machine
Learning. It depicts that how the interest in Artificial Neural Network is increasing and ANN’s
algorithms and models are used as standard tools in various fields such as information
engineering. In the field of information processing, neural network is used to solve some real
time problems as it has a robust set of computation procedures along with theoretical bases.
A Boosted Genetic Fuzzy Classifier (BGFC) for multispectral satellite images
classification is used in the paper [61]. The fuzzy rules used for classification are generated in
the iterative fashion, incrementally covering subspaces of the feature space, and resemble
reasoning employed by humans. After completing the rule generation A genetic tuning stage is
employed in order to improve the cooperation among the fuzzy rules, also maximize the
performance obtained after the stage one. The IKONOS multispectral VHR image was used for
testing. The Interpretability obtained by the BGFC method is of high level when compared to
Fuzzy classifiers. The introduced system effectively and efficiently handles the multi-
dimensional feature sources.
In the paper [62] includes the Artificial Neural Network method to perform the image
classification. The images used are obtained from IRS-6 and LISS-III satellites. The LISS-III
image classification process involves supervised learning method. A pixel-based classification
takes place. The efficiency is measured using the kappa coefficient. The methodology includes
the feature extraction, designing of ANN and then testing and classification. The two types of
feed forward ANN are implemented with data set freely available for LISS-III Indian satellite.
The first kind includes a single hidden layer with variations in the number of neurons present in
the hidden layer. The second kind has two hidden layers with the first hidden layer having a
constant number of neurons and the second hidden having variation in the number of neurons.
Jawak and al [63] have studied that the classifier for information extraction from satellite
images is based on two types of learning: Supervised learning and unsupervised learning, and
they have proposed several approaches to classification: Hybrid classification, Pixel-based
classification (PBC) and Object-oriented classification (OOC). These authors consider that
object-oriented classification overcomes the limitations of traditional methods based on the pixel
approach because pixel-based classification is a traditional classification that does not work well
on images that contain objects such as snow and ice, water, rock, and shadow.
Jabari and al [64] presented the main problems in high-resolution image classification,
such as uncertainties in the position of object boundaries in satellite images and the complex
similarity of segments with different classes.
Another supervised classification algorithm has been proposed by researchers Firozjaei
and al [65]; this algorithm is called Homogeneity Distance Classification Algorithm (HDCA). It
classifies satellite images using texture and spectrum information to classify these images in two
additional iterative computational steps. These researchers also proposed an improvement to the
Gravitational Search Algorithm (IGSA) to select features and determine the scale functionality of
the space in HDCA.
Pelletier and al [66] proposed an extensive study on temporal convolutional neural
networks (TempCNN) and recurrent neural networks (RNN). TempCNN is a deep learning
approach applying convolutions in the temporal dimension to learn temporal and spectral
features automatically.
Liang Zhang et al. [67] say that in the field of aerospace measurement and control field,
optical equipment generates a large amount of data as image. Thus, it has great research value for
how to process a huge number of image data quickly and effectively. With the development of
deep learning, great progress has been made in the task of image classification. The task images
are generated by optical measurement equipment are classified using the deep learning method.
Firstly, based on residual network, a general deep learning image classification framework, a
binary image classification network namely rocket image and other image is built. Secondly, on
the basis of the binary cross entropy loss function, the modified loss function is used to achieves
a better generalization effect on those images difficult to classify. Then, the visible image data
downloaded from optical equipment is randomly divided into training set, validation set and test
set. The data augmentation method is used to train the binary classification model on a relatively
small training set. The optimal model weight is selected according to the loss value on the
validation set. This method has certain value for exploring the application of deep learning
method in the intelligent and rapid processing of optical equipment task image in aerospace
measurement and control field.
T. Postadjiana et al. [68] proposes that supervised classification is the basic task for
landcover map generation. From semantic segmentation to speech recognition deep neural
networks has outperformed the state-of-the-art classifiers in many machine learning challenges.
Such strategies are now commonly employed in the literature for the purpose of land-cover
mapping. The system develops the strategy for the use of deep networks to label very high
resolution satellite images, with the perspective of mapping regions at country scale. Therefore, a
super pixel based method is introduced in order to (i) ensure correct delineation of objects and
(ii) perform the classification in a dense way but with decent computing times.
Chaomin Shen et al. [69] discuss that the percentage of cloud cover is one of the key
indices for satellite imagery analysis. To date, cloud cover assessment has performed manually
in most ground stations. To facilitate the process, a deep learning approach for cloud cover
assessment in quick look satellite images is proposed. Same as the manual operation, given a
quick look image, the algorithm returns 8 labels ranging from A to E and *, indicating the cloud
percentages in different areas of the image. This is achieved by constructing 8 improved VGG-
16 models, where parameters such as the loss function, learning rate and dropout are tailored for
better performance. The procedure of manual assessment can be summarized as follows. First,
determine whether there is cloud cover in the scene by visual inspection. Some prior knowledge,
e.g., shape, color and shadow, may be used. Second, estimate the percentage of cloud presence.
Although in reality, the labels are often determined as follows. If there is no cloud, then A; If a
very small amount of clouds exist, then B; C and D are given to escalating levels of clouds; and
E is given when the whole part is almost covered by clouds. There is also a label * for no-data.
This mostly happens when the sensor switches, causing no data for several seconds. The
disadvantages of manual assessment are obvious. First of all, it is tedious work. Second, results
may be inaccurate due to subjective judgment.
A novel deep learning application in remote sensing by assessing the cloud cover in an
optical scene is considered. This algorithm uses parallel VGG 16 networks and returns 8 labels
indicating the percentage of cloud cover in respective subscenes. The proposed algorithm
combines several state-of-the-art techniques and achieves reasonable results.
 Qingshan Liu et al. [70] discuss about a multiscale deep feature learning method for high-
resolution satellite image scene classification. However, satellite images with high spatial
resolution pose many challenging issues in image classification. First, the enhanced resolution
brings more details; thus, simple low-level features (e.g., intensity and textures) widely used in
the case of low-resolution images are insufficient in capturing efficiently discriminative
information. Second, objects in the same type of scene might have different scales and
orientations. Besides, high-resolution satellite images often consist of many different semantic
classes, which make further classification more difficult. Taking the commercial scene comprises
roads, buildings, trees, parking lots, and so on. Thus, developing effective feature representations
is critical for solving these issues.
Specifically, at first warp the original satellite image into multiple different scales. The
images in each scale are employed to train a deep convolutional neural network (DCNN).
However, simultaneously training multiple DCNNs is time consuming. To address this issue, a
DCNN with spatial pyramid pooling is explored. Since different SPP-nets have the same number
of parameters, which share the identical initial values, and only fine-tuning the parameters in
fully connected layers ensures the effectiveness of each network, thereby greatly accelerating the
training process. Then, the multiscale satellite images are fed into their corresponding SPP-nets,
respectively, to extract multiscale deep features. Finally, a multiple kernel learning method is
developed to automatically learn the optimal combination of such features.
T. Postadjiana et al. [71] discuss about semantic segmentation of remote sensing images
enables in particular land-cover map generation for a given set of classes. Very recent literature
has shown the superior performance of DCNN for many tasks, from object recognition to
semantic labelling, including the classification of VHR satellite images. A simple yet effective
architecture is DCNN. Input images are of size of 65*65*4 (number of bands). Only
convolutions of size 3*3 are used to limit the number of parameters. The second line of the table
displays the number of filters per convolution layer. The ReLU activation function is set after
each convolution in order to introduce non-linearity and maxpooling layers increase the receptive
field (the spatial information taken into account by the filter). Finally, a fully connected layer on
top sums up the information contained in all features in the last convolution layer.
However, while plethora of works aim at improving object delineation on geographically
restricted areas, few tend to solve this classification task at very large scales. New issues occur
such as intra-class class variability, diachronic between surveys, and the appearance of new
classes in a specific area, that do not exist in the predefined set of labels. Therefore, this work
intends to (i) perform large scale classification and to (ii) expand a set of land-cover classes,
using the off-the shelf model learnt in a specific area of interest and adapting it to unseen areas.
Patrick Helber et al. [72] discuss about the challenge of land use and land cover
classification using Sentinel-2 satellite images. The key contributions are as follows. A novel
dataset based on Sentinel-2 satellite images covering 13 different spectral bands and consisting
of 10 classes with in total 27,000 labeled images are presented. The state-of-the-art CNN on this
novel dataset with its different spectral bands are considered. Also evaluate deep CNNs on
existing remote sensing datasets. With the proposed novel dataset, a better overall classification
accuracy is gained. The classification system resulting from the proposed research opens a gate
towards various Earth observation applications.
The challenge of land use and land cover classification is considered. For this task,
dataset based on Sentinel-2 satellite images are used. The proposed dataset consists of 10 classes
covering 13 different spectral bands with in total 27,000 labeled images. The evaluated state of
the art deep CNNs on this novel dataset. Also evaluated deep CNNs on existing remote sensing
datasets and compared the obtained results. For the novel dataset, the performance is analyzed
based on different spectral bands. The proposed research can be leveraged for multiple real-
world Earth observation applications. Possible applications are land use and land cover change
detection and the improvement of geographical maps.
Keyang Cai et al. [73] proposes about cloud classification of satellite image in
meteorological forecast. Traditional machine learning methods need to manually design and
extract a large number of image features, while the utilization of satellite image features is not
high. CNN is the first truly successful learning algorithm for multi-layer network structure.
Compared with the general forward BP algorithm, CNN can reduce the number of parameters
needed in learning in spatial relations, so as to improve the training performance. A small piece
of local feel in the picture as the bottom of the hierarchical structure of the input, continue to
transfer to the next layer, each layer through the digital filter to obtain the characteristics of the
data. This method has significant effect on the observed data such as scaling, rotation and so on.
The cloud classification process of satellite based on deep convolution neural network includes
pre-processing, feature extraction, classification and other steps. A convolution neural network
for cloud classification, which can automatically learn features and obtain classification results is
constructed. The method has high precision and good robustness.
Okan Bilge Ozdemir et al. [74] says that the stacked autoencoders which are widely
utilized in deep learning research are applied to remote sensing domain for hyperspectral
classification. High dimensional hyperspectral data is an excellent candidate for deep learning
methods. However, there are no works in literature that focuses on such deep learning
approaches for hyperspectral imagery. This aims to fill this gap by utilizing stacked
autoencoders. Using stacked autoencoders, intrinsic representations of the data are learned in an
unsupervised way. Using labeled data, these representations are fine tuned. Then, using a soft-
max activation function, hyperspectral classification is done. Parameter optimization of Stacked
Autoencoders (SAE) is done with extensive experiments. It focuses on utilization of
representation learning methods for hyperspectral classification task on high definition
hyperspectral images. Strong side of representation learning methods are its unsupervised
automatic feature learning step which makes it possible to omit the feature extraction step that
requires domain knowledge.
Proposed method utilizes SAEs to learn discriminative representations in the data. Using
the label information in the training set, representations are fine tuned. Then, the test set is
classified using these representations. Effects of different parameters on the performance results
are also investigated. Most effective parameter is found as the epoch number of the supervised
classification step. For each parameter (learning rate, batch size and epoch numbers), optimal
values are approximated with extensive testing. From 432 different parameter configurations,
optimal parameter values are found as 25, 60, 100 and 5 for batch size, epoch numbers
(representation & classification) and learning rate, respectively. This score is greatly promising
when compared to state-of-the-art methods. This study can be enhanced by two important
additions, neighborhood factors and utilization of unlabelled data. With these additions as future
study, more general and robust results can be achieved.
Mykola Lavreniuk et al. [75] says that crop classification maps based on high resolution
remote sensing data are essential for supporting sustainable land management. The most
challenging problems for their producing are collecting of ground-based training and validation
datasets, non regular satellite data acquisition and cloudiness. To increase the efficiency of
ground data utilization it is important to develop classifiers able to be trained on the data
collected in the previous year. A deep learning method is analyzed for providing crop
classification maps using in-situ data that has been collected in the previous year. Main idea is to
utilize deep learning approach based on sparse auto encoder. At the first stage it is trained on
satellite data only and then neural network fine-tuning is conducted based on in-situ data form
the previous year. Taking into account that collecting ground truth data is very time consuming
and challenging task, the proposed approach allows us to avoid necessity for annual collecting
in-situ data for the same territory.
This approach is based on converting time series of observations into sparse vectors from
the unified hyperspace with similar representation. This idea allows further utilization of the auto
encoder to learn dependencies in data without labels and fine-tuning final neural network using
available in-situ data from previous year. This methodology was utilized to provide crop
classification map for central part of Ukraine for 2017 based on multi-temporal Sentinel-1
images with different acquisition dates.
Lior Bragilevsky et al. [76] says that satellite image analysis has become very important
in a lot of areas such as monitoring floods and other natural disasters, earthquake and tsunami
prediction, ship tracking and navigation, monitoring the effects of climate change etc. Large size
and inaccessibility of some of the regions of Amazon wants best satellite imaging system. Useful
informations are extracted from satellite images will help to observe and understand the
changing nature of the Amazon basin, and help better to manage deforestation and its
consequences. Some images were labelled and could be used to train various machine learning
algorithms. Another set of test image was provided without labels and was used to make
predictions. The predicted labels for test images were submitted to the competition for scoring.
The hope was that well-trained models arising from this competition would allow for a better
understanding of the deforestation process in the Amazon basin and give insight on how to better
manage this process.
Machine learning can be the key to saving the world from losing football field-sized
forest areas each second. As deforestation in the Amazon basin causes devastating effects both
on the ecosystem and the environment, there is urgent need to better understand and manage its
changing landscape. A competition was recently conducted to develop algorithms to analyze
satellite images of the Amazon. Successful algorithms will be able to detect subtle features in
different image scenes, giving us the crucial data needed to be able to manage deforestation and
its consequences more effectively.
Rodrigo F. Berriel et al. [77] proposes about a zebra crossing classification and detection
system. It is one of the main tasks for mobility autonomy. Although important, there are few data
available on where crosswalks are in the world. The automatic annotation of crosswalks
locations worldwide can be very useful for online maps, GPS applications and many others. In
addition, the availability of zebra crossing locations on these applications can be of great use to
people with disabilities, to road management, and to autonomous vehicles. However,
automatically annotating this kind of data is a challenging task. They are often aging (painting
fading away), occluded by vehicle and pedestrians, darkened by strong shadows, and many other
factors.
High-resolution satellite imagery has been increasingly used on remote sensing
classification problems. One of the main factors is the availability of this kind of data. Despite
the high availability, very little effort has been placed on the zebra crossing classification
problem. Crowd sourcing systems are exploited in order to enable the automatic acquisition and
annotation of a large-scale satellite imagery database for crosswalks related tasks.
Unnikrishnan, A., Sowmya, V. and Soman, K.P [78] presented a deep learning
architecture with hypertuning of the network, reducing the input bands to two (i.e., red and near-
infrared (NIR)). Deep learning architectures were designed for three networks, namely VGG,
AlexNet and ConvNet. Here, the hypertuning was accomplished over the filters of each
convolutional layer. The classification of different classes was accomplished using the modified
architecture with a reduced number of filters and two-band information. However, the
hypertuned deep learning architecture obtained less accuracy during the classification.
Jiang, J et al. [79] developed the Double-Channel Convolutional Neural Network
(DCCNN) model for classifying RGB-NIR images using the correlation among the R, G, B and
NIR bands. For describing the RGB and NIR image features, the DCCNN had two independent
CNN networks. Next, feature fusion was performed at the fully connected layer and the
classification was performed at the final layer. This configuration was useful for the effective
utilization of various features of RGB-NIR images. Moreover, overfitting was avoided using the
net dropout technique, which eliminated 60% of the neurons in the fully connected layer.
However, the classification accuracy was lower for the double-channel CNN model with the net
dropout technique.
N. Yang et al. [80] presented a training approach, namely greedy DropSample, for
increasing the speed of the Convolutional Neural Network (CNN)’s optimization process during
image classification. This method was mainly focused on the samples that generated the highest
gradients. Moreover, the activations of the network were biased in the training process due to the
absence of certain training samples. The samples with less losses were filtered out to increase the
speed of the CNN. However, the developed DropSample failed to consider the similarity
between the classes.
Weng et al. [81] developed the Multi-Dimensional Multi-Grained Scanning Structure
(MGSS) method to classify the land cover/land use over remote sensing images. The developed
MGSS was used to extract the spatial and spectral information from the images. Next, the
prediction was obtained by mapping the probability feature vectors in the residual forest
structure. The number of parameters required for the optimization of MGSS was lower when
compared to the CNN. However, the gradient passed from the high level affected by a certain
event in MGSS.
Zhong, Y et al. [82] presented an agile CNN structure, namely SatCNN, for obtaining the
classification over High-Spatial-Resolution Remote-Sensing (HSR-RS) images. Here, the
intrinsic features from the HSR-RS images were captured using preprocessing with the z-score
methods. The developed SatCNN was used to balance the training efficiency and generalization
ability of the model. However, the performance of the SatCNN was sensitive to the testing ratio.
Specifically, the SatCNN’s performance was affected due to the intra- and inter-class complexity
of the HSR-RS.
S. A. Yamashkin et al. [83] solved the issue of classification over HSR-RS images using
deep learning methods along with the conditions of labeled data scarcity. The GeoSystemNet
model solved the classification issue based on the genetic uniformity of spatially neighboring
objects of various scales and hierarchical levels. However, the GeoSystemNet model required a
huge amount of freedom degrees to maintain the classification performance.
Syrris et al. [84] developed the SatImNet, which is a group of open training, structured
and harmonized data with respect to certain rules. Further, the CNN was modeled to obtain the
classification of satellite images. Finally, the satellite image classification carried out on the
Office-31 dataset and the NWPU-Merced-Land satellite image dataset was performed as follows:
Hu et al. [85] presented the Coordinate Partial Adversarial Domain Adaptation (CPADA) to
perform an unsupervised satellite image classification. The CPADA was used to develop a
partial transfer learning technique, and the negative transfer was discarded by coordinate loss
using the down-weighting outlier satellite images. The classification of CPADA improved
because of the domain-invariant features obtained from the CPADA. However, the features were
misaligned, due to the deviation that occurred among the predicted and ideal weights.
 CHAPTER 3
PROPOSED METHOD
In this section, we discuss about the proposed method for satellite image classification.
The presented work includes: Pre-processing, feature extraction, and classification. Figure 3.1
represents the basic block diagram of the suggested framework.

Satellite Images Predicted Results

Pre-processing Feature Extraction Classification by
by ELBP SVM

Figure 3.1: Block diagram of the proposed satellite image classification

Figure 3.1 underneath shows the stream cycle of the strategy embraced for this work.
Here this work has taken four diverse classes of pictures and train the framework with highlights
of those pictures. Highlights of those preparation pictures are Extended Local Binary Patterns
(ELBP) and Linear Kernel base Support Vector Machine (LKSVM) and Radial Kernel-based
Support Vector Machine (RKSVM). This work utilizes at least five preparing pictures for one
class. Next is the choice of the test picture the test picture can be some other picture, however, it
must be unique concerning preparing pictures. At that point highlights of the test picture
extricated as was separated from preparing pictures.
3.1Pre-processing
Pre-processing is a broad word for activities that employ images at a low degree of
complexity, with both output data being intensity images. The iconic visuals are all of the same
kind as the original sensor signals, with an intensity picture often represented by an image
matrix. Action values (artistry). However, because analogous techniques are now used,
geometric hope of developing images (for example, booty shaking, sizing, and reinterpretation)
has been designated among also before the styles. The end result of also before that would be an
advancement of something like the dataset, which thus inhibits accidental nonlinear behavior or
elevates some photograph feature extraction.
3.2 Feature Extraction
Proposed technique is a gridding technique that subdivides big amounts of information
into separate groups for programming languages reasons of course. There seem to be plenty of
dependent variables, converting such large datasets typically generally require a notable number
of online resources. Techniques for choosing but instead integrating data points to create features
ensure that the data that must be prepared whilst still effective manner or rather totally labelling
this same empirical distribution.
Object recognition is handy because we need to cut the levels of necessary funds for
processing without losing crucial or important data. Proposed technique may also aid in reducing
the amount of duplicated data in an inquiry. Consequently, content analysis or the user's efforts
in creating variable combinations (features) speed up the computational component's teaching or
generalization procedures. This same Glcm Trend (LBP) texturing operator is a basic but
effective shading operator that labels actual pixels of an image by thresholding the neighborhood
from each pixel as well as interpret the outcome as more than just a bitwise integers.
With its racially prejudiced capability and processing economy, the RGB image operation
is becoming a popular approach in a wide range of applications. It may also be considered as a
unifying approach to the generally heterogeneous mathematical and geometric notions of pattern
recognition. The LBP pilot's resilience to periodic grey and black changes caused, for one, by
changes in illumination may be its most important feature in implementations. Another notable
aspect is its operational ease, which enables it to analyze images in challenging real-time
contexts. The ELBP approach is an enhanced variant of LBP in which the picture is separated
into three separate images based on the color configuration (red, blue, and green). The pattern
values are retrieved from all three photos and then put together.
The ELBP operator not only performs binary comparison between the central pixel and
its neighbours, but also encodes their exact gray value differences using some additional binary
units. Specifically, the ELBP feature consists of several LBP codes at multiple layers which
encode the gray-value difference (GD) between the central pixel and its neighbouring pixels. As
shown in Figure 3.2, the first layer of ELBP is actually the original LBP code encoding the sign
of GD. The following layers of ELBP then encode the absolute value of GD. Basically, each
absolute GD value is first encoded in its binary representation, and then all the binary values at a
given layer result in an additional local binary pattern. For example, in Figure 3.2 the first layer
is the original LBP code that encodes the sign of GD, thus yielding a decimal number of 211
from its binary form (11010011)2. The absolute values of GD, i.e. 1, 5, 3, 2, 1, 2, 3, 0, are first
encoded in their binary numbers: (001)2, (101)2, (011) 2, (010) 2 , …, etc. Using a same weight
scheme of LBP on all the binary bits, its ELBP code of the corresponding layer can be generated,
e.g., L2 is composed of (01000000)2 and its decimal value is 64; L3 is composed of
(00110110)2 and its decimal value is 54; finally L4 is composed of (11101010)2 and its decimal
value is 234. As a result, when describing similar local textures, although the first layer LBP is
not discriminative enough, the information encoded in the other additional layers can be utilized
to distinguish them. Its downside is that ELBP greatly increases feature dimensionality

Figure 3.2: Example of ELBP feature descriptor

3.3 Support Vector Machines
Nowadays, SVM [86] is commonly used in the neuro-imaging analysis, particularly in
classification problems. The primary objective of this approach is to estimate an optimal hyper-
plane by maximizing the margin between data points and the decision surface. However, in most
cases, the data points within the classes are contaminated by noise, where SVM (linear SVM)
cannot distinguish the data points completely. To limit this issue, introducing a soft margin and a
cost function C into SVM helps to mitigate the training error [87].
In some scenarios, especially when we map a feature set that contains many features into
a higher dimensional feature space, the data points within the classes are not separable linearly.
For this reason, a non-linear SVM is initiated with the help of a kernel trick [88]. In this paper,
we have chosen a radial basis function (RBF) kernel because:
1. It provides good performance when the feature set contains fewer numbers of features.
2. It needs less numbers of hyper-parameters compared to the polynomial kernel.
 The RBF kernel is defined by the following equation

(3.1)

where, and denote the objects and ; denotes the kernel variable which is used to
evaluate the smoothness of the boundary between the classes in the original object space.
3.4 Performance Measures
To evaluate the performance of learning approaches, we consider the following performance
measures:

(3.2)

(3.3)

(3.4)

(3.5)

(3.6)

(3.7)

where, = True Positive, = False Negative, = False Positive and = True Negative.
REFERENCES
1. Baboo, S. Santhosh, and S. Thirunavukkarasu. "Image segmentation using high
resolution multispectral satellite imagery implemented by fcm clustering
techniques." International Journal of Computer Science Issues (IJCSI) 11.3 (2014): 154.
2. R. L. Kettig and D. A. Landgrebe. Classification of multispectral image data by
extraction and classification of homogeneous objects. IEEE Trans. Geoscience
Electronics, 14(1):19–26, Jan. 1976.
3. Goel, Samiksha, Arpita Sharma, and V. K. Panchal. "A hybrid algorithm for satellite
image classification." International Conference on Advances in Computing,
Communication and Control. Springer, Berlin, Heidelberg, 2011.
4. LENI, EZIL SAM. "A TECHNIQUE FOR CLASSIFICATION OF HIGH
RESOLUTION SATELLITE IMAGES USING OBJECT-BASED
SEGMENTATION." Journal of Theoretical & Applied Information Technology 68.2
(2014).
5. Poongodi, S., and Dr B. Kalaavathi. "Satellite image classification based on fuzzy with
cellular automata." International Journal of Electronics and Communication
Engineering 2.3 (2015): 137-141.
6. Toshniwal, Mayank. "Satellite image classification using neural networks." 3rd
International Conference: Sciences of Electronic, Technologies of Information and
Telecommunications. 2005.
7. Rees, William Gareth. Physical principles of remote sensing. Cambridge university press,
2013.
8. Joseph, George. Fundamentals of remote sensing. Universities Press, 2005.
9. Rencz, A. N., and R. A. Ryerson. "Manual of Remote Sensing for the Earth Sciences."
10. Chander, Gyanesh, Brian L. Markham, and Dennis L. Helder. "Summary of current
radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI
sensors." Remote sensing of environment 113.5 (2009): 893-903.
11. Andary, Jim, et al. "Landsat 7 System design overview." IGARSS'96. 1996 International
Geoscience and Remote Sensing Symposium. Vol. 4. IEEE, 1996.
12. ERDAS (Firm). ERDAS field guide. Erdas, 1997.
13. Daptardar, A. A., and Vishal J. Kesti. "Introduction to remote sensors and image
processing and its applications." IJLTET 2 (2013): 107-114.
14. Nirmal Keshava. A survey of spectral unmixing algorithms. Lincoln Lab. J., 14:55– 78,
2003.
15. Gellert Mattyus, Shenlong Wang, Sanja Fidler, and Raquel Urtasun. Enhancing road
maps by parsing aerial images around the world. In IEEE ICCV, 2015.
16. R. O. Duda, P. E. Hart, and D. G. Stork. Pattern Classification. Wiley, 2001.
17. J. A. Benediktsson and P. H. Swain. Statistical Methods and Neural Network Approaches
for Classification of Data from Multiple Sources. PhD thesis, Purdue
Univ., School of Elect. Eng., West Lafayette, IN, 1990.
18. J. A. Benediktsson, P. H. Swain, and O. K. Ersoy. Conjugate gradient neural networks in
classification of very high dimensional remote sensing data. International
Journal of Remote Sensing, 14(15):2883 – 2903, 1993.
19. E. Merényi. Intelligent understanding of hyperspectral images through selforganizing neural
maps. In Proc. 2nd Int. Conf. Cybernetics and Information Technologies, Systems and
Applications (CITSA 2005), pages 30–35, Orlando, FL, USA, 2005.
20. V. Vapnik. Statistical Learning Theory. New York: Wiley, 1998.
21. G. Camps-Valls, L. Gomez-Chova, J. Muñoz-Marí, J. Vila-Francés, and J.
CalpeMaravilla. Composite kernels for hyperspectral image classification. IEEE Geosci.
Remote Sens. Lett., 3(1):93–97, 2006.
22. J. Ham, Yangchi Chen, M. M. Crawford, and J. Ghosh. Investigation of the random forest
framework for classification of hyperspectral data. IEEE Trans. Geosci. Remote Sens.,
43(3):492–501, March 2005.
23. R. Hansch and O. Hellwich. When to fuse what? random forest based fusion of low-,
mid-, and high-level information for land cover classification from optical and sar
images. In IEEE IGARSS, pages 3587–3590, July 2016.
24. J. A. Richards y X. Jia. Remote Sensing Digital Image Analysis. In SpringerVerlag,
1999.
25. Y. Tarabalka, J. Chanussot, and J. A. Benediktsson. Segmentation and classification of
hyperspectral images using minimum spanning forest grown from automatically selected
 markers. IEEE Trans. Systems, Man, and Cybernetics: Part B, 40(5):1267–1279, Oct.
2010.
26. K. Bernard, Y. Tarabalka, J. Angulo, J. Chanussot, and J. A. Benediktsson. Spectral-
spatial classification of hyperspectral data based on a stochastic minimum spanning forest
approach. IEEE Transactions on Image Processing, 21(4):2008– 2021, April 2012.
27. S. Valero, P. Salembier, and J. Chanussot. Hyperspectral image representation and
processing with binary partition trees. IEEE Trans. Image Process. 22(4):1430– 1443,
2013.
28. E. Maggiori, Y. Tarabalka, and G. Charpiat. Improved partition trees for multiclass
segmentation of remote sensing images. In IEEE IGARSS, pages 1016–1019. IEEE,
2015.
29. Y. Tarabalka and A. Rana. Graph-cut-based model for spectral-spatial classification of
hyperspectral images. In igarss, pages 3418–3421, July 2014.
30. Emmanuel Maggiori, Yuliya Tarabalka, and Guillaume Charpiat. Optimizing partition
trees for multi-object segmentation with shape prior. In 26th British Machine Vision
Conference (BMVC), 2015.
31. F. Tsai, C.-K. Chang, and G.-R. Liu. Texture analysis for three dimension remote sensing
data by 3D GLCM. In Proc. of 27th Asian Conference on Remote Sensing, pages 1–6,
August 2006.
32. X. Huang and L. Zhang. A comparative study of spatial approaches for urban mapping
using hyperspectral rosis images over pavia city, northern italy. International Journal of
Remote Sensing, 30(12):3205–3221, 2009.
33. J.A. Benediktsson, M. Pesaresi, and K. Arnason. Classification and feature extraction for
remote sensing images from urban areas based on morphological transformations. IEEE
Trans. Geosci. Remote Sens., 41(9):1940–1949, Sept. 2003.
34. M. Fauvel, J. A. Benediktsson, J. Chanussot, and J. R. Sveinsson. Spectral and spatial
classification of hyperspectral data using svms and morphological profiles. IEEE Trans.
Geosci. Remote Sens., 46(11):3804–3814, Nov 2008.
35. M. Dalla Mura, J. A. Benediktsson, B. Waske, and L. Bruzzone. Morphological attribute
profiles for the analysis of very high resolution images. IEEE Trans. Geosci. Remote
Sens., 48(10):3747–3762, Oct 2010.
36. P. Ghamisi, R. Souza, J. A. Benediktsson, X. X. Zhu, L. Rittner, and R. A. Lotufo.
Extinction profiles for the classification of remote sensing data. IEEE Trans. Geosci.
Remote Sens., 54(10):5631–5645, Oct 2016.
37. Mathias Ortner, Xavier Descombes, and Josiane Zerubia. Building outline extraction
from digital elevation models using marked point processes. IJCV, 72(2):107–132, 2007.
38. Yinghua Wang, Florence Tupin, and Chongzhao Han. Building detection from high-
resolution polsar data at the rectangle level by combining region and edge information.
Pattern Recognition Letters, 20, July 2010.
39. C. Lacoste, X. Descombes, and J. Zerubia. Point processes for unsupervised line network
extraction in remote sensing. IEEE TPAMI, 27(10):1568–1579, October 2005.
40. Wenzhi Zhao, Zhou Guo, Jun Yue, Xiuyuan Zhang, and Liqun Luo. On combining
multiscale deep learning features for the classification of hyperspectral remote sensing
imagery. International Journal of Remote Sensing, 36(13):3368–3379, 2015.
41. Volodymyr Mnih. Machine learning for aerial image labeling. PhD thesis, University of
Toronto, 2013.
42. Emmanuel Maggiori, Yuliya Tarabalka, Guillaume Charpiat, and Pierre Alliez.
Convolutional neural networks for large-scale remote-sensing image classification. IEEE
Trans. Geosci. Remote Sens., 55(2):645–657, 2017.
43. Dimitrios Marmanis, Konrad Schindler, Jan Dirk Wegner, Mihai Datcu, and Uwe Stilla.
Semantic segmentation of aerial images with explicit class-boundary modeling. In IEEE
IGARSS, pages 5071–5074, 2017.
44. Abburu, Sunitha, and Suresh Babu Golla. "Satellite image classification methods and
techniques: A review." International journal of computer applications 119.8 (2015).
45. Sathya, P., and L. Malathi. "Classification and segmentation in satellite imagery using
back propagation algorithm of ann and k-means algorithm." International Journal of
Machine Learning and Computing 1.4 (2011): 422.
46. Shivali, A. K., and V. K. Vishakha. "Supervised and Unsupervised Neural Network for
Classification of Satellite Images." Int. J. Comp. Appl 25 (2013).
47. Chriskos, Panteleimon, et al. "De-identifying facial images using singular value
decomposition." 2014 37th International Convention on Information and Communication
Technology, Electronics and Microelectronics (MIPRO). IEEE, 2014.
48. Phillips, Rhonda D., et al. "Feature reduction using a singular value decomposition for
the iterative guided spectral class rejection hybrid classifier." ISPRS Journal of
Photogrammetry and Remote Sensing 64.1 (2009): 107-116.
49. Kumar, Anil, Ashish Kumar Bhandari, and P. Padhy. "Improved normalised difference
vegetation index method based on discrete cosine transform and singular value
decomposition for satellite image processing." IET Signal Processing 6.7 (2012): 617-
625.
50. Mahi, Habib, Hadria Isabaten, and Chahira Serief. "Zernike moments and svm for shape
classification in very high resolution satellite images." Int. Arab J. Inf. Technol. 11.1
(2014): 43-51.
51. Mahi, Habib, Hadria Isabaten, and Chahira Serief. "Zernike moments and svm for shape
classification in very high resolution satellite images." Int. Arab J. Inf. Technol. 11.1
(2014): 43-51.
52. Bekaddour, Akkacha, Abdelhafid Bessaid, and Fethi Tarek Bendimerad. "Multi spectral
satellite image ensembles classification combining k-means, LVQ and SVM
classification techniques." Journal of the Indian Society of Remote Sensing 43.4 (2015):
671-686.
53. Brindha, S. "Satellite image enhancement using dwt–svd and segmentation using mrr–
mrf model." Journal of Network Communications and Emerging Technologies
(JNCET) 1.1 (2015): 6-10.
54. Chengfan Li, Jingyuan Yin, Junj uan Zhao, Extraction of Urban Vegetation from High-
Resolution Remote Sensing Image, International Conference On Computer Design And
Applications, IEEE, 2010
55. Shamama Anwar, Sugandh Raj, A Neural Network Approach to Edge Detection using
Adaptive Neuro-Fuzzy Inference System, IEEE, 2014.
56. Tiago M. A. Santos André Mora, Rita A. Ribeiro, and João M. N. Silva, Fuzzy-Fusion
approach for Land Cover Classification, IEEE International Conference on Intelligent
Engineering Systems, June 30-July 2, 2016
57. K. C. Tan, H. S. Lim and M. Z. Mat Jafri, Comparison of Neural Network and Maximum
Likelihood Classifiers for Land Cover Classification Using Landsat Multispectral Data,
IEEE Conference on Open Systems, 2011
58. Wei Zhang, Zhong Li, Weidong Xu, Haiquan Zhou, A Classifier of Satellite Signals
Based on the Back-Propagation Neural Network, International Congress on Image and
Signal Processing, 2015
59. Nur Anis Mahmon1 and Norsuzila Ya’acob, A Review on Classification of Satellite
Image Using Artificial Neural Network (ANN), Control and System Graduate Research
Colloquium, IEEE, 2015
60. Alberto Prieton, Beatriz Prieto, Eva Martinez Ortigosa, Eduardo Ros, Francisco Pelayo,
Julio Ortega, Ignacio Rojas , Neural networks: An overview of early research, current
frameworks and new challenges, Elsevier B.V, 2016.
61. D.G. Stavrakoudis, J.B. Theocharis , G.C. Zalidis, A Boosted Genetic Fuzzy Classifier
for land cover classification of remote sensing imagery, International Society for
Photogrammetry and Remote Sensing, Inc , Elsevier B.V , 2011
62. Mr.Anand Upadhyay, Dr.S.K.Singh , Ms.Pooja Singh, Ms.Priya Singh, Comparative
study of Artificial Neural Network-based classification of IRS LISS-III satellite images,
IEEE, 2015.
63. S. D. Jawak, P. Devliyal, and A. J. Luis, “A Comprehensive Review on Pixel Oriented
and Object Oriented Methods for Information Extraction From Remotely Sensed Satellite
Images with a Special Emphasis on Cryospheric Applications,” Adv. Remote Sens.,
vol.04, no. 03, pp. 177–195, 2015.
64. S. Jabari and Y. Zhang, “Very High Resolution Satellite Image Classification Using
Fuzzy Rule-Based Systems,” Algorithms, vol.6, no. 4, pp. 762–781, 2013.
65. M. K. Firozjaei, I. Daryaei, A. Sedighi, Q. Weng, and S. K. Alavipanah, “Homogeneity
Distance Classification Algorithm (Hdca): A novel Algorithm for Satellite Image
Classification,” Remote Sens., vol.11, no. 5, 2019.
66. C. Pelletier, G. I. Webb, and F. Petitjean, “Temporal Convolutional Neural Network for
the Classification of Satellite Image Time Series,” Remote Sens., vol.11, no. 5, pp. 1–25,
2019.
67. L. Zhang, Z. Chen, J. Wang and Z. Huang, "Rocket Image Classification Based on Deep
Convolutional Neural Network," 2018 10th International Conference on
Communications, Circuits and Systems (ICCCAS), Chengdu, China, 2018, pp. 383-386.
68. C. Shen, C. Zhao, M. Yu and Y. Peng, "Cloud Cover Assessment in Satellite Images Via
Deep Ordinal Classification," IGARSS 2018 - 2018 IEEE International Geoscience and
Remote Sensing Symposium, Valencia, 2018, pp. 3509-3512.
69. T. Postadjian, A. L. Bris, C. Mallet and H. Sahbi, "Superpixel Partitioning of Very High
Resolution Satellite Images for Large-Scale Classification Perspectives with Deep
Convolutional Neural Networks," IGARSS 2018 - 2018 IEEE International Geoscience
and Remote Sensing Symposium, Valencia, 2018, pp. 1328-1331.
70. Q. Liu, R. Hang, H. Song and Z. Li, "Learning Multiscale Deep Features for High-
Resolution Satellite Image Scene Classification," in IEEE Transactions on Geoscience
and Remote Sensing, vol. 56, no. 1, pp. 117- 126, Jan. 2018.
71. T. Postadjian, A. L. Bris, H. Sahbi and C. Malle, "Domain Adaptation for Large Scale
Classification of Very High Resolution Satellite Images with Deep Convolutional Neural
Networks," IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing
Symposium, Valencia, 2018, pp. 3623-3626.
72. P. Helber, B. Bischke, A. Dengel and D. Borth, "Introducing Eurosat: A Novel Dataset
and Deep Learning Benchmark for Land Use and Land Cover Classification," IGARSS
2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia,
2018, pp. 204-207.
73. K. Cai and H. Wang, "Cloud classification of satellite image based on convolutional
neural networks," 2017 8th IEEE International Conference on Software Engineering and
Service Science (ICSESS), Beijing, 2017, pp. 874-877.
74. A. O. B. Özdemir, B. E. Gedik and C. Y. Y. Çetin, "Hyperspectral classification using
stacked autoencoders with deep learning," 2014 6th Workshop on Hyperspectral Image
and Signal Processing: Evolution in Remote Sensing (WHISPERS), Lausanne, 2014, pp.
1-4.
75. M. Lavreniuk, N. Kussul and A. Novikov, "Deep Learning Crop Classification Approach
Based on Sparse Coding of Time Series of Satellite Data," IGARSS 2018 - 2018 IEEE
International Geoscience and Remote Sensing Symposium, Valencia, 2018, pp. 4812-
4815.
76. L. Bragilevsky and I. V. Bajić, "Deep learning for Amazon satellite image analysis,"
2017 IEEE Pacific Rim Conference on Communications, Computers and Signal
Processing (PACRIM), Victoria, BC, 2017, pp. 1-5.
77. R. F. Berriel, A. T. Lopes, A. F. de Souza and T. OliveiraSantos, "Deep Learning-Based
Large-Scale Automatic Satellite Crosswalk Classification," in IEEE Geoscience and
Remote Sensing Letters, vol. 14, no. 9, pp. 1513- 1517, Sept. 2017.
78. Unnikrishnan, A.; Sowmya, V.; Soman, K.P. Deep learning architectures for land cover
classification using red and near-infrared satellite images. Multimed. Tools
Appl. 2019, 78, 18379–18394. [Google Scholar] [CrossRef]
79. Jiang, J.; Liu, F.; Xu, Y.; Huang, H. Multi-spectral RGB-NIR image classification using
double-channel CNN. IEEE Access 2019, 7, 20607–20613. [Google Scholar] [CrossRef]
80. Yang, N.; Tang, H.; Yue, J.; Yang, X.; Xu, Z. Accelerating the Training Process of
Convolutional Neural Networks for Image Classification by Dropping Training Samples
Out. IEEE Access 2020, 8, 142393–142403. [Google Scholar] [CrossRef]
81. Weng, L.; Qian, M.; Xia, M.; Xu, Y.; Li, C. Land Use/Land Cover Recognition in Arid
Zone Using A Multi-dimensional Multi-grained Residual Forest. Comput.
Geosci. 2020, 144, 104557. [Google Scholar] [CrossRef]
82. Zhong, Y.; Fei, F.; Liu, Y.; Zhao, B.; Jiao, H.; Zhang, L. SatCNN: Satellite image dataset
classification using agile convolutional neural networks. Remote Sens. Lett. 2017, 8,
136–145. [Google Scholar] [CrossRef]
83. Yamashkin, S.A.; Yamashkin, A.A.; Zanozin, V.V.; Radovanovic, M.M.; Barmin, A.N.
Improving the Efficiency of Deep Learning Methods in Remote Sensing Data Analysis:
Geosystem Approach. IEEE Access 2020, 8, 179516–179529. [Google Scholar]
[CrossRef]
84. Syrris, V.; Pesek, O.; Soille, P. SatImNet: Structured and Harmonised Training Data for
Enhanced Satellite Imagery Classification. Remote Sens. 2020, 12, 3358. [Google
Scholar] [CrossRef]
85. Hu, J.; Tuo, H.; Wang, C.; Zhong, H.; Pan, H.; Jing, Z. Unsupervised satellite image
classification based on partial transfer learning. Aerosp. Syst. 2019, 3, 21–28.
86. Vapnik, Vladimir, The nature of statistical learning theory (Springer science & business
media, 2013).
87. Cortes, Corinna and Vapnik, Vladimir, "Support-vector networks", Machine learning 20,
3 (1995), pp. 273--297.
88. Cristianini, Nello, Shawe-Taylor, John et al., An introduction to support vector machines
and other kernel-based learning methods (Cambridge university press, 2000).
 Appendix A

MATLAB
A.1 Introduction

MATLAB is a high-performance language for technical computing. It integrates

computation, visualization, and programming in an easy-to-use environment where problems and
solutions are expressed in familiar mathematical notation. MATLAB stands for matrix
laboratory, and was written originally to provide easy access to matrix software developed by
LINPACK (linear system package) and EISPACK (Eigen system package) projects. MATLAB
is therefore built on a foundation of sophisticated matrix software in which the basic element is
array that does not require pre dimensioning which to solve many technical computing problems,
especially those with matrix and vector formulations, in a fraction of time.
MATLAB features a family of applications specific solutions called toolboxes. Very
important to most users of MATLAB, toolboxes allow learning and applying specialized
technology. These are comprehensive collections of MATLAB functions (M-files) that extend
the MATLAB environment to solve particular classes of problems. Areas in which toolboxes are
available include signal processing, control system, neural networks, fuzzy logic, wavelets,
simulation and many others.
Typical uses of MATLAB include: Math and computation, Algorithm development, Data
acquisition, Modeling, simulation, prototyping, Data analysis, exploration, visualization,
Scientific and engineering graphics, Application development, including graphical user interface
building.

A.2 Basic Building Blocks of MATLAB

The basic building block of MATLAB is MATRIX. The fundamental data type is the
array. Vectors, scalars, real matrices and complex matrix are handled as specific class of this
basic data type. The built in functions are optimized for vector operations. No dimension
statements are required for vectors or arrays.
A.2.1 MATLAB Window

The MATLAB works based on five windows: Command window, Workspace window,
Current directory window, Command history window, Editor Window, Graphics window and
Online-help window.

A.2.1.1 Command Window

The command window is where the user types MATLAB commands and expressions at
the prompt (>>) and where the output of those commands is displayed. It is opened when the
application program is launched. All commands including user-written programs are typed in
this window at MATLAB prompt for execution.

A.2.1.2 Work Space Window

MATLAB defines the workspace as the set of variables that the user creates in a work
session. The workspace browser shows these variables and some information about them.
Double clicking on a variable in the workspace browser launches the Array Editor, which can be
used to obtain information.

A.2.1.3 Current Directory Window

The current Directory tab shows the contents of the current directory, whose path is
shown in the current directory window. For example, in the windows operating system the path
might be as follows: C:\MATLAB\Work, indicating that directory “work” is a subdirectory of
the main directory “MATLAB”; which is installed in drive C. Clicking on the arrow in the
current directory window shows a list of recently used paths. MATLAB uses a search path to
find M-files and other MATLAB related files. Any file run in MATLAB must reside in the
current directory or in a directory that is on search path.

A.2.1.4 Command History Window

The Command History Window contains a record of the commands a user has entered in
the command window, including both current and previous MATLAB sessions. Previously
entered MATLAB commands can be selected and re-executed from the command history
window by right clicking on a command or sequence of commands. This is useful to select
various options in addition to executing the commands and is useful feature when experimenting
with various commands in a work session.
A.2.1.5 Editor Window

The MATLAB editor is both a text editor specialized for creating M-files and a graphical
MATLAB debugger. The editor can appear in a window by itself, or it can be a sub window in
the desktop. In this window one can write, edit, create and save programs in files called M-files.
MATLAB editor window has numerous pull-down menus for tasks such as saving,
viewing, and debugging files. Because it performs some simple checks and also uses color to
differentiate between various elements of code, this text editor is recommended as the tool of
choice for writing and editing M-functions.

A.2.1.6 Graphics or Figure Window

The output of all graphic commands typed in the command window is seen in this
window.

A.2.1.7 Online Help Window

MATLAB provides online help for all it’s built in functions and programming language
constructs. The principal way to get help online is to use the MATLAB help browser, opened as
a separate window either by clicking on the question mark symbol (?) on the desktop toolbar, or
by typing help browser at the prompt in the command window. The help Browser is a web
browser integrated into the MATLAB desktop that displays a Hypertext Markup Language
(HTML) documents. The Help Browser consists of two panes, the help navigator pane, used to
find information, and the display pane, used to view the information. Self-explanatory tabs other
than navigator pane are used to perform a search.

A.3 MATLAB Files

MATLAB has three types of files for storing information. They are: M-files and MAT-
files.

A.3.1 M-Files

These are standard ASCII text file with ‘m’ extension to the file name and creating own
matrices using M-files, which are text files containing MATLAB code. MATLAB editor or
another text editor is used to create a file containing the same statements which are typed at the
MATLAB command line and save the file under a name that ends in .m. There are two types of
M-files:
1. Script Files

It is an M-file with a set of MATLAB commands in it and is executed by typing name of

file on the command line. These files work on global variables currently present in that
environment.

2. Function Files

A function file is also an M-file except that the variables in a function file are all local.
This type of files begins with a function definition line.

A.3.2 MAT-Files

These are binary data files with .mat extension to the file that are created by MATLAB
when the data is saved. The data written in a special format that only MATLAB can read. These
are located into MATLAB with ‘load’ command.

A.4 the MATLAB System:

The MATLAB system consists of five main parts:

A.4.1 Development Environment:

This is the set of tools and facilities that help you use MATLAB functions and files. Many
of these tools are graphical user interfaces. It includes the MATLAB desktop and Command
Window, a command history, an editor and debugger, and browsers for viewing help, the
workspace, files, and the search path.

A.4.2 the MATLAB Mathematical Function:

This is a vast collection of computational algorithms ranging from elementary functions

like sum, sine, cosine, and complex arithmetic, to more sophisticated functions like matrix
inverse, matrix eigen values, Bessel functions, and fast Fourier transforms.
A.4.3 the MATLAB Language:

This is a high-level matrix/array language with control flow statements, functions, data
structures, input/output, and object-oriented programming features. It allows both "programming
in the small" to rapidly create quick and dirty throw-away programs, and "programming in the
large" to create complete large and complex application programs.

A.4.4 Graphics:

MATLAB has extensive facilities for displaying vectors and matrices as graphs, as well as
annotating and printing these graphs. It includes high-level functions for two-dimensional and
three-dimensional data visualization, image processing, animation, and presentation graphics. It
also includes low-level functions that allow you to fully customize the appearance of graphics as
well as to build complete graphical user interfaces on your MATLAB applications.

A.4.5 the MATLAB Application Program Interface (API):

This is a library that allows you to write C and FORTRAN programs that interact with
MATLAB. It includes facilities for calling routines from MATLAB (dynamic linking), calling
MATLAB as a computational engine, and for reading and writing MAT-files.

A.5 SOME BASIC COMMANDS:

pwd prints working directory

Demo demonstrates what is possible in Mat lab

Who lists all of the variables in your Mat lab workspace?

Whose list the variables and describes their matrix size

clear erases variables and functions from memory

clear x erases the matrix 'x' from your workspace

close by itself, closes the current figure window

figure creates an empty figure window

hold on holds the current plot and all axis properties so that subsequent graphing

commands add to the existing graph

hold off sets the next plot property of the current axes to "replace"

find find indices of nonzero elements e.g.:

d = find(x>100) returns the indices of the vector x that are greater than 100

break terminate execution of m-file or WHILE or FOR loop

for repeat statements a specific number of times, the general form of a FOR

statement is:

FOR variable = expr, statement, ..., statement END

for n=1:cc/c;

magn(n,1)=NaNmean(a((n-1)*c+1:n*c,1));

end

diff difference and approximate derivative e.g.:

DIFF(X) for a vector X, is [X(2)-X(1) X(3)-X(2) ... X(n)-X(n-1)].

NaN the arithmetic representation for Not-a-Number, a NaN is obtained as a

result of mathematically undefined operations like 0.0/0.0

INF the arithmetic representation for positive infinity, a infinity is also produced

by operations like dividing by zero, e.g. 1.0/0.0, or from overflow, e.g. exp(1000).

save saves all the matrices defined in the current session into the file,

matlab.mat, located in the current working directory

load loads contents of matlab.mat into current workspace

save filename x y z saves the matrices x, y and z into the file titled filename.mat

save filename x y z /ascii save the matrices x, y and z into the file titled filename.dat

load filename loads the contents of filename into current workspace; the file can

be a binary (.mat) file

load filename.dat loads the contents of filename.dat into the variable filename

xlabel(‘ ’) : Allows you to label x-axis

ylabel(‘ ‘) : Allows you to label y-axis

title(‘ ‘) : Allows you to give title for

plot

subplot() : Allows you to create multiple

plots in the same window

A.6 SOME BASIC PLOT COMMANDS:

Kinds of plots:

plot(x,y) creates a Cartesian plot of the vectors x & y

plot(y) creates a plot of y vs. the numerical values of the elements in the y-vector

semilogx(x,y) plots log(x) vs y

semilogy(x,y) plots x vs log(y)

loglog(x,y) plots log(x) vs log(y)

polar(theta,r) creates a polar plot of the vectors r & theta where theta is in radians

bar(x) creates a bar graph of the vector x. (Note also the command stairs(x))
bar(x, y) creates a bar-graph of the elements of the vector y, locating the bars

according to the vector elements of 'x'

Plot description:

grid creates a grid on the graphics plot

title('text') places a title at top of graphics plot

xlabel('text') writes 'text' beneath the x-axis of a plot

ylabel('text') writes 'text' beside the y-axis of a plot

text(x,y,'text') writes 'text' at the location (x,y)

text(x,y,'text','sc') writes 'text' at point x,y assuming lower left corner is (0,0)

and upper right corner is (1,1)

axis([xmin xmax ymin ymax]) sets scaling for the x- and y-axes on the current plot

A.7 ALGEBRIC OPERATIONS IN MATLAB:

Scalar Calculations:

+ Addition

- Subtraction

* Multiplication

/ Right division (a/b means a ÷ b)

\ left division (a\b means b ÷ a)

^ Exponentiation

For example 3*4 executed in 'matlab' gives ans=12

4/5 gives ans=0.8

 Array products: Recall that addition and subtraction of matrices involved addition or
subtraction of the individual elements of the matrices. Sometimes it is desired to simply multiply
or divide each element of an matrix by the corresponding element of another matrix 'array
operations”.

Array or element-by-element operations are executed when the operator is preceded by a '.'
(Period):

a .* b multiplies each element of a by the respective element of b

a ./ b divides each element of a by the respective element of b

a .\ b divides each element of b by the respective element of a

a .^ b raise each element of a by the respective b element

A.8 MATLAB WORKING ENVIRONMENT:

A.8.1 MATLAB DESKTOP

Matlab Desktop is the main Matlab application window. The desktop contains five sub
windows, the command window, the workspace browser, the current directory window, the
command history window, and one or more figure windows, which are shown only when the
user displays a graphic.

The command window is where the user types MATLAB commands and expressions at
the prompt (>>) and where the output of those commands is displayed. MATLAB defines the
workspace as the set of variables that the user creates in a work session.

The workspace browser shows these variables and some information about them. Double
clicking on a variable in the workspace browser launches the Array Editor, which can be used to
obtain information and income instances edit certain properties of the variable.

The current Directory tab above the workspace tab shows the contents of the current
directory, whose path is shown in the current directory window. For example, in the windows
operating system the path might be as follows: C:\MATLAB\Work, indicating that directory
“work” is a subdirectory of the main directory “MATLAB”; WHICH IS INSTALLED IN
DRIVE C. clicking on the arrow in the current directory window shows a list of recently used
paths. Clicking on the button to the right of the window allows the user to change the current
directory.

MATLAB uses a search path to find M-files and other MATLAB related files, which are
organize in directories in the computer file system. Any file run in MATLAB must reside in the
current directory or in a directory that is on search path. By default, the files supplied with
MATLAB and math works toolboxes are included in the search path. The easiest way to see
which directories are soon the search path, or to add or modify a search path, is to select set path
from the File menu the desktop, and then use the set path dialog box. It is good practice to add
any commonly used directories to the search path to avoid repeatedly having the change the
current directory.

The Command History Window contains a record of the commands a user has entered in
the command window, including both current and previous MATLAB sessions. Previously
entered MATLAB commands can be selected and re-executed from the command history
window by right clicking on a command or sequence of commands.

This action launches a menu from which to select various options in addition to executing
the commands. This is useful to select various options in addition to executing the commands.
This is a useful feature when experimenting with various commands in a work session.

A.8.2 Using the MATLAB Editor to create M-Files:

The MATLAB editor is both a text editor specialized for creating M-files and a graphical
MATLAB debugger. The editor can appear in a window by itself, or it can be a sub window in
the desktop. M-files are denoted by the extension .m, as in pixelup.m.

The MATLAB editor window has numerous pull-down menus for tasks such as saving,
viewing, and debugging files. Because it performs some simple checks and also uses color to
differentiate between various elements of code, this text editor is recommended as the tool of
choice for writing and editing M-functions.
 To open the editor , type edit at the prompt opens the M-file filename.m in an editor
window, ready for editing. As noted earlier, the file must be in the current directory, or in a
directory in the search path.

A.8.3 Getting Help:

The principal way to get help online is to use the MATLAB help browser, opened as a
separate window either by clicking on the question mark symbol (?) on the desktop toolbar, or by
typing help browser at the prompt in the command window. The help Browser is a web browser
integrated into the MATLAB desktop that displays a Hypertext Markup Language(HTML)
documents. The Help Browser consists of two panes, the help navigator pane, used to find
information, and the display pane, used to view the information. Self-explanatory tabs other than

navigator pane are used to perform a search.