Professional Documents
Culture Documents
Motherboard Detector
Motherboard Detector
By:
CHAPTER I........................................................................................................................... 1
CHAPTER II .......................................................................................................................... 4
2.4.4 TensorFlow........................................................................................................ 10
i
3.1.3 Software Application Development ............................................................ 16
REFERENCES ..................................................................................................................... 21
ii
CHAPTER I
INTRODUCTION
Moving and still images have been included in learning materials for
technology aims to detect the target objects using various methods of image
objects. (Doshi, Punktambekar, Kini, & Dhami, 2014). Text recognition is important
develop a computer application with the ability to automatically read the text
1
& Mizan, 2017), They proposed an algorithm for solving character recognitions.
Given an input in the form of images. The algorithm was trained on the training
data that was initially present in the database and done preprocessing like noise
Odobez, & Chen, 2003) study embeds the traditional character segmentation
step followed by an OCR algorithm. Their method is split into two main parts: the
detection of text lines, followed by the recognition of text in these lines. In the
study of (Zhou, McGuiness, & O'Connor, 2018) about Text recognition and
retrieval system for e-Business image management stated that the on-going
networks and MSERs for image indexing and a novel dataset is prepared
for each patient and also on automatic number plate recognition which is
The problem that this study wants to address is the insufficient information
2
create an application to identify the specifications of a motherboard by
manual.
another method of learning the specifications that is not stated or written on the
motherboard to be used for purposes such as for their studies or even purchasing
This study will only focus on recognizing the name of the model that is
written on the motherboard. The application will use a camera for identifying the
motherboards model name and will only be limited to an Android mobile phone.
Only certain specifications such as the Socket type, form factor, and memory
support will be covered in this study. This study will only identify specific
inductors, and the backside of the motherboard. The model name of the
3
CHAPTER II
REVIEW OF RELATED LITERATURE
2.1 Motherboard
The computer motherboard (Figure 2.1) is one of the most important part of
the Personal Computer. It is the hardware of the computer installed inside the
system unit. This Hardware act as the main board or core assembly of the system
unit. Inside the CPU case there are a lot of other hardware connected to the
motherboard, namely: Hard Disk Drive, CD/DVD ROM, Processor, RAM, Lan Card,
4
Below is the specific motherboards that is currently available and is in the
market that will be used in this study for the purpose of testing.
1. Gigabyte GA-A320M-DS2
2. Gigabyte GA-H61M-S1
3. ASUS EX-H310-V3
be identified. Other specifications will not be identified for the reason that it is
already physically visible such as how many Memory slots, PCIe slots, etc.
Specification Description
Memory Support The type of memory the motherboard supports and what
each of which has a particular location and value. (Woods & Gonzalez, 1992).
5
preprocessing to enable object detection, classification, and tracking. (Wang,
2001)
through an optical mechanism. (Jana, Chodhury, & Islam, 2014). Human eyes
naturally recognize various patterns, fonts or styles. For computers, it is hard work
the image of paper documents into a text file. Then, it becomes possible to
used for different purposes. They can be scanned in search of patterns and vital
data, used to generate reports and draw up charts, distributed into spreadsheets,
computational vision. Text Recognition used in official task in which the large
data must type like post offices, banks, colleges, etc. In Real life applications
where we want to collect some information from text written image. (Karmakar,
and discussion channel for the mass media. (Oksman, 2010). With more than a
billion mobile phones being carried around by consumers of all ages, the mobile
6
phone has become the most pervasive pocket-carried device. Today’s
different ways. (Iftode, Borcea, Ravi, Kang, & Zhou Peng, 2004).
Much like the Linux or Windows operating system controls your desktop or laptop
other programs can run on mobile devices. The operating system is responsible
for determining the functions and features available on your device, such as
7
messaging and more. The mobile OS will also determine which third-party
As smart phones and tablets become more popular, the operating systems
for those devices become more important. Android is such an operating system
for low powered devices that run on battery and are full of hardware like GPS
receivers, cameras, light and orientation sensor, Wi-Fi Connectivity, and touch
screen. Like all operating systems, Android enables applications to make use of
2.4 Programming
whose quality is tested by the quality of the resulting program or software. A well
2.4.1 Python
language. Python combines remarkable power with very clear syntax. Python is
8
widely considered as the preferred language for teaching and learning machine
explicitly programming it. In today’s world every system that does well has a
2.4.2 Firebase
2014). It is a backend platform for building web, android, and iOS applications. It
offers real time database, different APIs, multiple authentication types, hosting
platform and much more. Firebase can power your app’s backend, including
data storage, user authentication, static hosting and more. The database in
store and fetch data. Database is highly reliable and superfast means that data
is updated and synchronized in no time and data is maintained even the user
2.4.3 ML Kit
Android and iOS apps in a powerful yet easy-to-use package. ML Kit provides
convenient APIs that help you use your custom TensorFlow Lite models in your
APIs. Cloud APIs process data on the Google Cloud Platform, therefore object
recognition is performed more precisely. However, cloud models are larger than
on-device models. On-device models require less space, they can function
9
offline, and data processing is faster, but these models are not so accurate.
(Kuprenko, 2019).
2.4.4 TensorFlow
ready to use with TensorFlow. It handles downloading and preparing the data.
(Cristianini, 2018). TensorFlow was designed for use both in research and
development and in production systems. It can run on single CPU systems, GPUs
within the Android operating system, Android Studio uses a Gradle-based build
Android Studio has one or more modalities with source code and resource files.
These modalities include Android app modules, Library modules, and Google
10
CHAPTER III
METHODOLOGY
prioritizes rapid prototyping and quick feedback over long drawn out testing
cycles and can make multiple iterations and updates to a software rapidly
programming, software testing, and evaluation, to easily keep track of the step
by step process. The first phase, initial requirements, we gather the required
software and data for the database that will be used for this study.
11
In the programming phase the developer will integrate software
development kits to be used for the application and also prepare the database
interface of the application will be planned and also discuss how the application
will work using a use case diagram. Software testing phase then test the
accuracy on how the application recognizes the text correctly and will then be
analyzed.
During the gathering of the initial requirements, the proponents will create
the initial concept and identify the requirements that is needed to develop the
application. Afterwards, the developer will then be able to generate a plan and
development kit to be used is the Firebase and ML Kit, respectively. The mobile
least 3 GB of RAM, and a quad-core processor for the reason that most mobile
In this phase, the proponent has identified all the requirements and
information as well as creating the plan and schedule for developing the
12
application. The proponents will then design the user interface of the application
and design how the application will work using a use case diagram and a
and detailing how the components are going to interact with each other.
The user interface will consist of buttons users click on for capturing the
image, the text field where the specifications will be displayed. Designers focus
on building interfaces that users will find highly usable and efficient. The user
manipulates the application via an input and then the expected results are
displayed via the output. The motherboard’s model name should be within the
13
mobile application’s targeted box in order for it to recognized so that only the
model name and not other characters within the motherboard to avoid it being
In Figure 3.3, the User starts the application and chooses to capture the
image of the name of the motherboard’s model. The User will automatically see
recognized and fetched from the database. The admin of the application has
14
3.1.2.3 Application Flowchart
Figure 3.4 shows the flowchart of the application. The process of how the
application starts, then the user must capture the image of a motherboard, the
application will then recognize the text on the image. If the recognized text
matches one on the database it will then fetch of the specifications of that
15
3.1.2.4 Database Design
specifically the motherboard model and its specifications which will be fetched
and displayed after the application has captured, recognized, and matched
In this phase, after all the initial requirements have been gathered the
developer will now prepare the motherboard database that will be used for the
android application. The developer will then Integrate the Firebase library and
the Software Development Kits (SDK) into the program, specifically the ML Kit.
These SDKs are required for developing platform-specific applications and also,
In this phase, the proponent will conduct an activity to check whether the
actual results match the expected or desired results to ensure that the software is
effective and efficient. The following are the steps that will be performed during
testing.
3. Wait until the application detects and recognizes the model name text.
4. Proponents will now record the data on how many characters are correct
16
3.1.4.1 Environment for Testing
In this study, the preferred illumination of the environment for testing the
levels for everyday workspaces such as libraries, classrooms, and homes, also
shadows and other obstructions caused by objects that can affect the result
not be more than a 30cm for the reason that this study is only after to capture
the image where the motherboard’s model name is located and not the entire
motherboard. The angle for testing is that the camera should be perpendicular
This section shows the data that is required that will be gathered through
testing the application and will be written on the tables below. There will be
three (3) motherboards that will be tested for which will have fifty (50) tests in
each table for a total of one hundred fifty (150) tests. The proponent shall record
17
MOTHERBOARD #1 TEST OF THE PROPOSED APPLICATION
TESTS Test Entirely Correct Incorrect
Characters Y =1 | N = 0 Y =1 | N = 0
1
2
3
…
50
MEAN
STANDARD DEVIATION
Table 3.1 Calculations of the Mean and Standard Deviation with a motherboard
Table 3.2 Data and Calculations for the Mean and Standard Deviation with a
18
MOTHERBOARD #3 TEST OF THE PROPOSED APPLICATION
TESTS Test Entirely Correct Incorrect
Characters Y =1 | N = 0 Y =1 | N = 0
1
2
3
…
50
MEAN
STANDARD DEVIATION
Table 3.3 Data and Calculations for the Mean and Standard Deviation with a
The study will use a one sample Z-Test to see if the proposed application’s
data that will be used will be based on the completed proposed applications
testing and analysis which will have a total of fifty (50) tests.
significant difference or character error between the original text and the
recognized text.
19
To determine the value of the Z-Score, this formula will be used:
(3.6)
Where:
α = 0.05
σ = Standard Deviation
The variables in the equation are α, x̄, σ, n, and µ, which will be used for
the calculation of the Z – Test. x̄ is the mean of the entirely correct recognized
The proponent will then conduct the z – test after calculating the Z value.
The proponent shall use a significance level of 0.05. The null Hypothesis will be
20
REFERENCES
Arjun, P., & Mirnalinee, T. (2016). Machine parts recognition and defect
detection in automated assembly systems using computer vision
techniques. Revista Técnica de la Facultad de Ingeniería
Universidad del Zulia.
Bourlard, H., Odobez, J.-M., & Chen, D. (2003). Text detection and
recognition in images and video frames. Pattern Recognition, 597-
606.
Doshi, P., Punktambekar, S., Kini, N., & Dhami, S. S. (2014). Object
Detection and its Application using Convulutional Neural Network.
NBNSSOE.
21
Gutierrez, K. (2014). Studies Confirm Power of Visuals in eLearning. SHIFT
eLearning.
Iftode, L., Borcea, C., Ravi, N., Kang, P., & Zhou Peng. (2004). Smart Phone:
An Embedded System for Universal Interactions. Rutgers, 1-3.
Jalled, F., & Voronkov, I. (2016). Object Detection Using Image Processing.
ArXiv 2016, 1.
Jana, R., Chodhury, A., & Islam, M. (2014). Optical Character Recognition
from Text Image. International Journal of Computer Applications
Technology and Research, 1-3.
Karmakar, S., Chakraborty, T., & Mizan, C. (2017). Text Recognition using
Image Processing. IJARCS, 765-767.
Le, J. (2018). Snapchat’s Filters: How computer vision recognizes your face.
Medium, 1.
22
Mehta, B. M., Madhani, N., & Patwardhan, R. (2017). Firebase: A Platform
for your Web and Mobile Applicatios. International Journal of
Advance Research in Science and Engineering, 45-47.
Muresan, H., & Oltean, M. (2017). Fruit Recognition from Images using
Deep Learning. ArXiv 2017.
Quero, J., Cruciani, F., Seidenari, L., Espinilla, M., & Nugent, C. (2019).
Straightforward Recognition of Daily Objects in Smart Environments
from Wearable Vision Sensor. ARDUOUS'19.
23
Tamplin, J. (2014). Firebase is Joining Google! The Firebase Blog, 1.
Tian, Z., Tong, H., He , P., & Huang , W. (2016). Detecting Text in Natural
Image with Connectionist Text Proposal Network. Computer Vision -
ECCV 2016, 1-14.
Verschae, R., & Ruiz-del Solar, J. (2015). Object Detection: Current and
Future Directions. Researchgate, 1-3.
Zhao, Z.-Q., Zheng, P., Xu, S.-t., & Wu, X. (2019). Object Detection with
Deep Learning: A Review. arXiv.
Zhou, J., McGuiness, K., & O'Connor, N. (2018). A Text Recognitio and
Retrieval System for e-business Image Management. MMM2018, 1-
11.
24
PRICE LIST
25