OOP SE-203

MACHINE LEARNING APPROACHES IN MALARIA &

PLANT DISEASE DETECTION

Submitted By:

- ANUJ 2K20/SE/21

- ANURAG MUNSHI 2K20/SE/25

ABSTRACT

Malaria is a significant burden on global health, with more than 400,000 deaths per

year. It is a blood disease caused by parasites of the genus Plasmodium. A

conventional tool for malaria is examining the stained blood cell of the patient in a

microscope. It is a time-consuming process. This review paper will discuss an image

processing system for detecting and quantifying plasmodium parasites in blood smear

slides. Later, we will learn a Machine Learning algorithm to detect and determine the

types of infected cells according to their features.

Crop diseases are a substantial risk to sustenance security. The emergence of

accurate techniques has shown impressive results in the field of leaf-based image

classification. Random Forest is used in identifying between healthy and diseased

leaves from the datasets created. This review paper includes various implementation

phases, namely dataset creation, feature extraction, training the classifier and

classification. Overall, using machine learning to train the large data sets available

publicly gives us a clear way to detect the disease present in plants on a colossal

scale.

MALARIA DETECTION

INTRODUCTION

Malaria is a mosquito-borne life-threatening disease caused by the Plasmodium

parasite. A chemical process enables the visual detection and recognition of

Plasmodium in RBC. The staining process somewhat colorizes the RBCs but

highlight Plasmodium, WBCs and platelets. The detection of Plasmodium requires the

detection of stained objects. Several methods exist for malaria detection. Most

methods for detecting malaria are based on two criteria: (i) images acquired under

well-controlled conditions; (ii) the need for proper microscope equipment. Both

criteria are different to accomplish in the endemic area of malaria, where this type of

equipment is scarce or non-existent in health care facilities—the paper deals with the

challenges involved in the automatic detection of malaria parasite tissues. Instead of

finding the number of infected cells, machine

learning will be used to find the type of infection. A machine learning algorithm takes

examples of inputs and outputs associated with a task and produces a program that can

automatically differentiate them.

AIMS AND OBJECTIVES

The objective of the research is to study the mechanism of malaria detection using

image processing and machine learning, which will help the scientists of laboratories

detect the presence of Plasmodium protozoa with much greater accuracy.

METHODOLOGY

The data consists of heterogeneous datasets, we have a lot of variability in the images.

So, it might be easy to develop an algorithm that works for particular image but it is

much different challenge to find an algorithm that works across the board of

heterogeneous datasets. There, are couple of approaches that can be taken to try an

algorithm that is robust to the heterogeneity or we can cache the data into more

homogeneity state.

a. Data Collection

b. RBC’s Image Database: Image Acquisition

c. Machine Learning Algorithms to train the system to detect healthy and malaria

infected RBCs.

PROCEDURE

The first process is image acquisition in all the image processing steps. RBC's

image database is taken where malaria-infected RBC's and non-infected RBCs are

placed. Global Thresholding Segmentation is the following process in which an image

is separated into regions with respect to objects. Regions are segmented by identifying

common properties. The most uncomplicated property that pixels in a region can share

is intensity. So, thresholding can be a natural way to segment such regions, separating

light and dark regions. The following process is image enhancement. Adjusting digital

images so that the result is more suitable than the original one is called

image enhancement. It is helpful for further processing. Morphology is used for

extracting image components helpful in representing and describing region and shape.
Morphological operations help the input image apply a structuring element, creating

an output image of the same size.

Segmentation (watershed segmentation or color-based segmentation) subdivides

an image into its constituent regions. The detailing of the subdivisions is defined by

the problem being solved itself. Accuracy of segmentation becomes very important as

it determines the eventual success or failure of automated analysis procedures. Finally,

the perimeter of the parasite and the mean perimeter of RBC cells is calculated. If

the parasite is greater than the mean value of RBC cells, the number of infected RBCs

will be counted. If the parasite infects at least one RBC, then the person is considered

to be malaria-infected.

CONCLUSION

We have presented an approach and developed an algorithm for detecting malaria,

automated malaria detection and quantification of malaria infection. Also, we

developed a strategy to train with machine learning, adaptable to detection of malaria

with other types of parasites and also to increase the predictive value with results.

PLANT DISEASE DETECTION

INTRODUCTION

Pests and Diseases result in the destruction of crops or part of the plant resulting

in decreased food production leading to food insecurity. Also, knowledge about

pest management or control and diseases are less in various less developed countries.

Toxic pathogens, poor disease control, and drastic climate changes are vital factors

that arise in dwindled food production. Machine learning and deep learning algorithm

are some modern approaches that have been employed to increase the recognition rate

and the accuracy of the results. Much research under machine learning has taken place

for plant disease detection and diagnoses, such as traditional machine learning
approaches such as random forest, artificial neural network, support vector machine

(SVM), fuzzy logic, K-means method, and Convolutional neural networks.

PROCEDURE

To determine whether the leaf is diseased or healthy, certain steps must be followed.

i.e., Pre-processing, Feature extraction, Training of classifier and Classification. Pre-

processing of an image is bringing all the images size to a reduced uniform size.

Then comes extracting features of a pre-processed image which is done with the help

of HOG. HOG is a feature descriptor used for object detection. In this feature

descriptor, the appearance of the object and the outline of the image is described by its

intensity gradients. One of the advantages of HOG feature extraction is that it operates

on the cells created. Any transformations do not affect this. Here, the use of three

feature descriptors is made.

Hu moments: Image moments with important characteristics of the image pixels

helps in describing the objects. Here Hu moments help in describing the outline of a

particular leaf. Hu moments are calculated over a single channel only. The first step

involves converting RGB to Grayscale and, then the Hu moments are calculated. This

step gives an array of shape descriptors. Haralick Texture: Usually, the healthy leaves

and diseased leaves have different textures. Here the Haralick texture feature to

distinguish between the textures of a healthy and diseased leaf is used. It is based on

the adjacency matrix, which stores the position of (I, J). The texture is calculated

based on the frequency of the pixel I occupying the position next to pixel J. To

calculate Haralick texture, it is required that the image is converted to grayscale.

Color Histogram: Color histogram gives the representation of the colors in the image.

RGB is first converted to HSV color space, and the histogram is calculated for the

same. It is needed to convert the RGB image to HSV since the HSV model

aligns closely with how the human eye discerns the images' colors. Histogram plot

describes the number of pixels available in the given color ranges.

METHODOLOGY
To find out whether the leaf is diseased or healthy, certain steps must be followed. i.e.,

Preprocessing, Feature extraction, Training of classifier and Classification.

Preprocessing of image, is bringing all the images size to a reduced uniform size.

Then comes extracting features of a preprocessed image which is done with the help

of HOG. HoG is a feature descriptor used for object detection. In this feature

descriptor the appearance of the object and the outline of the image is described by its

intensity gradients. One of the advantages of HoG feature extraction is that it operates

on the cells created. Any transformations do not affect this

Here use of three feature descriptors was made:

a. Hu moments

b. Haralick Texture

c. Color Histogram

CONCLUSION

The objective of this algorithm is to recognize abnormalities that occur on plants in

their greenhouses or natural environment. The image captured is usually taken with a

plain background to eliminate occlusion.

REFERENCES

[1] WHO, "Fact sheet: World Malaria Report 2016," in World Health Organization, world

Health Organization, 2016. [Online]. Available: http://www.who.int/malaria/media/world-

malaria-report-2016/en. Accessed: Jan. 25, 2017

[2] Bloland PB (2001) Drug resistance in Malaria, WHO/CDS/ CSR/DRS/ 2001.4. World

Health Organization, Switzerland, 2001

[3] D Ghate, C. Jadhav, and N. U. Rani, "AUTOMATIC DETECTION OF MALARIA

PARASITE FROM BLOOD IMAGES,". [Online]. Available:

http://ijact.org/volume4issue1/IJ0410050.pdf. Accessed: Jan. 25, 2017

4] Ahirwar A., Pattnaik S., Acharya B., Advancced Image Analysis Based System for

Automatic Detection and Classification of Malarial Parasite in Blood Images. International

Journal of Information Technology and Knowledge Management. Jan-June 2012, Volume 5,

No 1, pp. 59-64

[5] Amit Kumar, Prof. Arun Choudhary, Prof. P. U. Tembhare, Prof. C. R. Pote. Enhanced

Identification of Malarial Infected Objects using Otsu Algorithm from Thin Smear Digital

Images. International Journal of Latest Research in Science and Technology Vol.1, Issue 2:

Page No. 159-163, July-August (2012).

[6] Nattkemper, T. W., Schubert, W., Hermann, T., & Ritter, H. (2004). A Hybrid System for

Cell Detection in Digital Micrographs. In B. Tilg (Ed.), Biomedical Engineering, Proc. (Vol.

417). Innsbruck, Austria: ACTA Press.

[7] CDC, "Frequently asked questions (FAQs)," CDC, 2016. [Online]. Available:

https://www.cdc.gov/malaria/about/faqs.html. Accessed: Jan. 25, 2017.

[8] L. Rosadoa, J. M. C. da Costab, D. Elias, and J. S. Cardoso, "Automated detection of

malaria parasites on thick blood smears via mobile devices," 2016

[9] N. A. Khan, H. Pervaz, A. Latif, and A. Musharaff, "Unsupervised identification of

malaria parasites using computer vision," 2017

[10] D. Taylor, N. Green and G. Stout, 3rd ed., “Bilogical Science” The Press Syndicate of

the University of Cambridge, 2002.

[11] S. S. Sannakki and V. S. Rajpurohit,” Classification of Pomegranate Diseases Based on

Back Propagation Neural Network,” International Research Journal of Engineering and

Technology (IRJET), Vol2 Issue: 02 | May-2015

[12] P. R. Rothe and R. V. Kshirsagar,” Cotton Leaf Disease Identification using Pattern

Recognition Technique

[13] Aakanksha Rastogi, Ritika Arora and Shanu Sharma,” Leaf Disease Detection and

Grading using Computer Vision Technology &Fuzzy Logic” 2nd International Conference on

Signal Processing and Integrated Networks (SPIN)2015.

[14] Godliver Owomugisha, John A. Quinn, Ernest Mwebaze and James Lwasa,” Automated

Vision-Based Diagnosis of Banana Bacterial Wilt Disease and Black Sigatoka Disease “,

Preceding of the 1’st international conference on the use of mobile ICT in Africa ,2014.

[15] uan Tian, Chunjiang Zhao, Shenglian Lu and Xinyu Guo,” SVM-based Multiple

Classifier System for Recognition of Wheat Leaf Diseases,” Proceedings of 2010 Conference

on Dependable Computing (CDC’2010), November 20-22, 2010.

[16] S. Yun, W. Xianfeng, Z. Shanwen, and Z. Chuanlei, “Pnn based crop disease recognition

with leaf image features and meteorological data,” International Journal of Agricultural and

Biological Engineering, vol. 8, no. 4, p. 60, 2015.

[17] J. G. A. Barbedo, “Digital image processing techniques for detecting, quantifying and

classifying plant diseases,” Springer Plus, vol. 2, no.660, pp. 1–12, 2013

[18] Caglayan, A., Guclu, O., & Can, A. B. (2013, September). “A plant recognition

approach using shape and color features in leaf images.” In International Conference on

Image Analysis and Processing (pp. 161-170). Springer, Berlin, Heidelberg

[19] Zhen, X., Wang, Z., Islam, A., Chan, I., Li, S., 2014d. “Direct estimation of cardiac bi-

ventricular volumes with regression forests.” In: Accepted by Medical Image Com- puting

and Computer-Assisted Intervention– MICCAI 2014.

[20] Wang P., Chen K., Yao L., Hu B., Wu X., Zhang J., et al. (2016).” Multimodal

classification of mild cognitive impairment based on partial least squares”