Rina Project

A PROJECT REPORT
ON
ENHANCING NIGHTTIME PEDESTRIAN DETECTION

WITH VISUAL AND MILLIMETER-WAVE RADAR FUSION
Submitted in partial fulfillment of requirements
for the award of the degree of
MASTER OF COMPUTER APPLICATIONS

By
TUPAKULA RINA
(22091F0042)
Under the esteemed guidance of

G.RAJASEKHAR REDDY M.tech(Ph.D)
Assistant Professor, Dept. of CSE

RAJEEV GANDHI MEMORIAL COLLEGE OF ENGINEERING & TECHNOLOGY
(AUTONOMOUS)
AFFILIATED TO J.N.T UNIVERSITY ANANTAPUR. ACCREDITED BY NBA (TIER-1) &
NAAC OF UGC. NEW DELHI, WITH A+ GRADE
RECOGNIZED UGC-DDU KAUSHAL KENDRA
NANDYAL-518501, (Estd-1995)
YEAR: 2023-2024
Rajeev Gandhi Memorial College of Engineering &Technology
(AUTONOMOUS)
AFFILIATED TO J.N.T UNIVERSITY ANANTAPUR. ACCREDITED BY NBA (TIER-1) &
NAAC OF UGC. NEW DELHI, WITH A+ GRADE
RECOGNIZED UGC-DDU KAUSHAL KENDRA
NANDYAL-518501, (Estd-1995)
(ESTD – 1995)
CERTIFICATE
This is to certify that TUPAKULA RINA (22091F0042), of MCA III- semester, has carried out
the mini-project work entitled “ENHANCING NIGHTTIME PEDESTRIAN DETECTION
WITH VISUAL AND MILLIMETER-WAVE RADAR FUSION” under the supervision and
guidance of G.RAJASEKHAR REDDY M.tech.(Ph.D), Professor, CSE Department, in partial
fulfillment of the requirements for the award of Degree of Master of Computer Applications
from Rajeev Gandhi Memorial College of Engineering & Technology (Autonomous),
Nandyal is a bonafied record of the work done by him during 2023-2024.
Project Guide Head of the department
G.Rajasekhar Reddy M.tech(Ph.d). Dr. K. SUBBA REDDY, M. Tech, Ph.D.

AssistantProfessor, Professor& Head, Dept. of CSE
Dept. of CSE
Place: Nandyal
Date: External Examiner
CONFIRMATION LETTER
05-03-24
To
The HOD,
Department of Computer Applications,
Rajeev Gandhi Memorial College of Engineering and Technology,Nandyal,
Andhra Pradesh.
Respected Sir/Madam,
Sub – Academic Project Confirmation Letter
This is to certify Ms.TUPAKULA RINA, bearing Reg. No. 22091F0042 student of Rajeev
Gandhi Memorial College of Engineering and Technology, Nandyal has been admitted to do a project
titled “ENHANCING NIGHTTIME PEDESTRIAN DEFECTION WITH VISUAL AND
MILLIMETER WAVE-RADAR FUSION” on Python Technology at “Tru Research Support, Hyderabad”
The project duration is from 1th Feb, 2024 to 1st June, 2024.
The project will be done under the guidance of our technical Team. The project Team is guided by P CARMEL
VANDHANA, Project Developer
From
TRU RESEARCH SUPPORT
P CARMEL VANDHANA
AUTHORISED SIGNATURE
ACKNOWLEDGEMENT
The satisfaction that accomplishes completion of any task would be incomplete without
the mention of people who made it possible and whose encouragement and guidance has been
a source of inspiration throughout the course of the project I want to express my sincere
gratitude to my project guide G.RAJASEKHAR REDDY, M.tech,Ph.D. Professor in CSE
Department Under his guidance, I have learned a lot about different aspects in research,
including finding a good research problem.
I express my profound sense of gratitude to Dr. K. SUBBA REDDY, Head of the

Department of CSE for inspiring and guiding us throughout the project.
I express my profound sense of gratitude to Dr. T. JAYA CHANDRA PRASAD,
Principal, and Rajeev Gandhi Memorial College of Engineering and Technology

(autonomous).
TUPAKULA RINA
(22091F0042)
DECLARATION
I hereby declare that the project report entitled “ENHANCING NIGHTTIME

PEDESTRIAN DETECTION WITH VISUAL AND MILLIMETER WAVE-RADAR
FUSION”,under the guidance of G.RAJASEKHAR REDDY,M.Tech,Ph.D. is submitted for
the award of degree of bachelor of technology in MASTER OF COMPUTER
APPLICATIONS.
This is a record of bonafide work carried out by us and the results embodied in this
project have not been reproduced or copied from any source. The results embodied in this
project have not been submitted to any other university or institute for the award of any other
Degree.
TUPAKULA RINA
(22091F0030)
CONTENTS
CHAPTER PAGE NO.
1. INTRODUCTION 2
2. LITERATURE REVIEW 3
2.1 Existing System 3
2.2 Disadvantages 4
2.3 Proposed System 4
2.4 Advantages 5
3. SYSTEM MODEL / DESIGN 6
3.1. Algorithms 6
3.2. Software Devolopment Life Cycle 8
3.2.1 Spiral Model 8
3.3 Feasibility Study 9
3.3.1 Economical Feasibility 10
3.3.2 Technical Feasibility 10
3.3.3 Social Feasibility
3.3.4 Feasibility Analysis 10
3.3.5 Request Approval 12

3.4 System Design and Development 12
3.4.1 Input Design 12
3.4.2 Output Design 13
3.5 UML Diagram 14
3.5.1 Class Diagram 15
3.5.2 Sequence Diagram 16
3.5.3 Data Flow Diagram 17
3.5.4 Activity Diagram 18
3.5.5 Collaborative Diagram 19
3.5.6 Component Diagram 19
3.5.7 Deployment Diagram 20
3.6 Modules 21
3.7 Hardware and Software Requirements 22
3.7.1 Hardware Requirements 22
3.7.2 Software Requirements 22
4. Implementation 23
4.1 Python 23
4.1.1 History Of Python 23
4.1.2 Features of Python 23
4.2 Technology Discription 25
5 Testing 27
5.1 Structural Testing 28

5.2 Behavioral Testing 29
6 Test Cases 30
7 Future Enhancement 31
8 Conclusion 32
Appendix: A. Project Screenshots 33
Appendix: B. Project References 37

LIST OF FIGURES
FIG NO. NAME OF THE FIGURE PAGE NO.
Fig: 1 Architecture Diagram 4
Fig: 2 Spiral Model 8
Fig: 3 Use case 13
Fig: 4 Class Diagram 14
Fig: 5 Sequence Diagram 15
Fig: 6 Data Flow Diagram 16
Fig: 7 Remote User 17
Fig: 8 Service Provider 18

OPTIMIZED YOLOV5 FOR SOLAR CELL SURFACE
DEFECT DETECTION
Abstract
A target detection algorithm based on the fusion of vision sensors and millimeter wave radar data
can effectively improve the safety of self-driving vehicles. However, a single sensor cannot obtain
comprehensive category status information on the target in the nighttime environment. To improve
pedestrian detection in nighttime traffic scenarios, this paper proposes a nighttime pedestrian
detection method based on the fusion of infrared vision information and millimeter wave (MMW)
radar data. The lateral localization and category features of the target are obtained using the
improved YOLOv5 deep learning algorithm, the distance and velocity information of the target is
obtained by preprocessing the MMW radar acquisition data, the pedestrian target is tracked by
using the extended Kalman filter for MMW radar detection, the projection of the valid radar target
point on the Infrared Radiatio (IR) image is completed according to the spatiotemporal fusion, and
then the correlation gate method is used to correlate the data, and the visual information is inherited
to the radar points to get the target multimodal information, and the success associated valid target
sequences are weighted to obtain the accurate target position. Finally, a decision-level fusion
algorithm framework is proposed to complete the output of pedestrian multimodal information in
nighttime traffic scenes. Theoretical analysis and experimental results show that the accuracy and
robustness of this method for nighttime pedestrian recognition are better than those of a single
sensor. In propose work author has modified YoloV5 but not experimented with advance YOLO
family other version as like Yolo6, 7 or 8. So as extension we have trained same dataset with
Yolov6 and it’s giving better prediction accuracy compare to Optimized Yolov5.
Dept. Of MCA, RGMCET, Nandyal 1

DEFECT DETECTION
CHAPTER-1
1. INTRODUCTION
At present, the research related to automotive self-driving technology is getting more and more in-
depth, and the perception of the surrounding environment in complex traffic scenes is an important
part of self-driving cars. The environment perception system is equipped with sensors such as
LIDAR, MMW radar, and cameras to obtain information on the category, location, and speed of
surrounding traffic targets, which can effectively reduce the risk of collision and improve the safety
performance of self-driving cars. Target detection is the process of extracting object classes and
locations based on the observation information from different sensors. As a key step in an
autonomous driving environment sensing system, target detection techniques based on deep
learning have received extensive attention and research. Among them, single-stage target detection
algorithms include YOLO [1], [2], [3], [4], [5], SSD [6], and RetinaNet [7]. The two-stage target
detection algorithms are mainly based on R-CNN [8] and Faster R-CNN [9] as the main research.
In recent years, the YOLO series algorithms have good detection accuracy and detection speed in
target detection. However, the current target detection task mainly focuses on visible images, but
visible images are greatly affected by the environment, especially in extremely bad weather and
nighttime conditions, and visible images lose some details, leading to degradation of target
detection performance.

DEFECT DETECTION
CHAPTER-2
2. LITERATURE REVIEW
In complex scenes, it is a huge challenge to accurately detect motion-blurred, tiny, and dense
objects in the thermal infrared images. To solve this problem, robust thermal infrared vehicle and
pedestrian detection method is proposed in this paper. An important weight parameter β is first
proposed to reconstruct the loss function of the feature selective anchor-free (FSAF) module in its
online feature selection process, and the FSAF module is optimized to enhance the detection
performance of motion-blurred objects. The proposal of parameter β provides an effective solution
to the challenge of motion-blurred object detection. Then, the optimized anchor-free branches of
the FSAF module are plugged into the YOLOv3 single-shot detector and work jointly with the
anchor-based branches of the YOLOv3 detector in both training and inference, which efficiently
improves the detection precision of the detector for tiny and dense objects. Experimental results
show that the method proposed is superior to other typical thermal infrared vehicle and pedestrian
detection algorithms due to 72.2% mean average precision (mAP).
2.1 Existing System

Current research on target detection and recognition based on MMW radar and vision fusion has
the following problems and shortcomings: for night and low light scenes, the visible camera
detection effect will be greatly affected, and the vision sensor cannot provide valid information for
sensor fusion. If the region of interest is defined using the target location and state information
detected by radar, the valid target may be affected by redundant targets. If different sensor
information is fused directly, the fusion detection performance is weak and biased.
2.2 Disadvantages
1. However, a single sensor cannot obtain comprehensive category status information on the
target in the nighttime environment.
2. Sensors were dependent on similarity distance to identify objects but this technique prove
to be insufficient.

DEFECT DETECTION
3. So sensors alone are not sufficient for accurate detection on pedestrians at night or day
time.
2.3 Proposed System

In this project, we propose a nighttime pedestrian target detection method based on the fusion of an
infrared camera and MMW radar sensors, taking full advantage of the performance of the infrared
camera and MMW radar. The method uses infrared cameras and MMW radar to simultaneously
acquire forward target information. isual information processing is improved based on the YOLOv5
model, the CVC infrared dataset is trained using a deep migration learning approach, and pre-
training weights are used to perform pedestrian detection on the acquired nighttime infrared images.
The MMW radar acquisition data is preprocessed to filter out valid radar targets. For the target loss
problem caused by radar jitter and target occlusion, the valid radar target is tracked using extended
Kalman filtering, and the filtered radar target information is projected onto the IR image to generate
a radar detection target frame centered on the projected point. Finally, a decision-level fusion of the
two sensor detection targets is performed to achieve nighttime pedestrian detection.
The author of the proposed work has altered YoloV5, but hasn't played around with other advanced
Yolo family versions like Yolo6, 7, or 8. As a result, we trained Yolov6 on the same dataset, and
its prediction accuracy is higher than that of Optimized Yolov5.
2.4 Advantages
1. improving the network structure of the YOLOv5 algorithm model, and verifying the good
performance of this improved algorithm in infrared image target detection through target
detection algorithm comparison experiments.
2. A data association method based on visual information is proposed for multimodal

information fusion of visual information and radar detection points based on the
spatiotemporal calibration of images.
3. A decision-level fusion strategy is designed for target detection in the nighttime environment,
and it is experimentally verified that the strategy reduces the missed false detection rate of a
single sensor and improves the accuracy of target detection.

DEFECT DETECTION
CHAPTER-3
3. SYSTEM DESIGN
Fig1: Architecture Diagram

DEFECT DETECTION
3.1 Algorithms
Faster R-CNN (FRCNN): Faster R-CNN is a two-stage object detection algorithm. It employs a
region proposal network to generate candidate regions and a fast RCNN for object detection.
Region proposals and features are simultaneously learned, enhancing accuracy and speed in
detecting objects within an image.
Improved YOLOv5: Improved YOLOv5, an evolution of the YOLO (You Only Look Once)
architecture, introduces advanced backbone networks, efficient anchor generation, and optimized
training strategies. This enhances real-time object detection performance, achieving higher
accuracy and speed for a wide range of applications.
Extension YOLOv6: YOLOv6 extends the YOLO (You Only Look Once) series by incorporating
innovative features like dynamic model scaling, advanced augmentation, and improved network
architectures. This extension aims to enhance object detection performance, providing a flexible
and efficient solution for various computer vision tasks with improved accuracy.

DEFECT DETECTION
3.2 Software Development Life Cycle
PROCESS MODEL USED WITH JUSTIFICATION
3.2.1 SPIRAL MODEL
A SPIRAL MODEL is used in this project. Here are some details regarding the project: The
Spiral Model is an iterative approach to software development that blends aspects of waterfall and
prototyping approaches The Spiral Model could be used in this project in the following ways:
• Identification of Goals and Requirements (Planning): Specify the goals, limitations, and
specifications of the project for the AI-powered e-vaccination system.
• Risk Analysis and Evaluation (Risk Assessment): Determine possible dangers connected to
user acceptability, data security, and AI integration.
Analyze the consequences and probability of these hazards.
• Engineering Prototype Development: Create a simple e-vaccination system prototype to
highlight important aspects.This might be a condensed version of the system with an emphasis on
key features like AI Chabot interaction and appointment scheduling.
• Build and Test (Engineering and Testing): Create the essential e-vaccination system
components, such as the database configuration and AI integration.
Test everything thoroughly to find and address any problems or bugs.
• Risk Analysis and Evaluation (Risk Assessment - Iteration 1): Reassess the risks that have
been identified, taking into account comments and insights from the testing and prototyping stages.
• Risk Analysis and Evaluation (Risk Assessment - Iteration 2): Reevaluate risks in light of
developments and new information.
• Complete System Development (Engineering - Iteration 2): Create the user interfaces, data
security protocols, and all anticipated features of the whole e-vaccination system.
• Comprehensive Testing (Testing - Iteration 2): Test the completely developed system
thoroughly to make sure it satisfies all requirements and is error-free.
The fully completed e-vaccination system should be implemented in a controlled setting, like a
pilot program for a particular community.

DEFECT DETECTION
Fig2: Spiral Model

• Evaluation (Review and Evaluation): Get input from stakeholders and users.
Evaluate the security, usability, and performance of the system.
• Maintenance and Enhancement (Maintenance and Support): - Give the system continuous
maintenance and assistance.
- Make ongoing changes and improvements by utilizing feedback and assessments.
• Risk Assessment - Iteration 3: Risk Analysis and Evaluation- As long as the system is being
actively used, keep evaluating and mitigating risks.
Throughout the course of a project, the Spiral Model facilitates iterative development by
integrating risk management and feedback.
3.3 Feasibility Study
In this stage, the viability of the project is examined, and a business proposal with a basic project
plan and some cost estimates is presented. During system analysis, the viability of the suggested
system needs to be examined. This is to ensure that the recommended approach won't put undue
strain on the company. A feasibility study requires a basic understanding of the main
requirements of the system.
Economic feasibility, technical feasibility, and social feasibility are the three main factors taken
into account in the feasibility analysis.

DEFECT DETECTION
3.3.1 ECONOMICAL FEASIBILITY
The purpose of this study is to evaluate the system's potential financial impact on the company.
The amount of money the corporation can devote to system research and development is limited.
The costs have to make sense. Given the financial limits, the intended system could also be
implemented because the majority of the technologies were publicly available. Just the items that
can be customized need to be bought.
3.3.2 TECHNICAL FEASIBILITY
The purpose of this study is to evaluate the system's technological requirements, or technical
feasibility. Any system's development shouldn't put an undue burden on the available technical
resources. The availability of technological resources will consequently be in high demand. As a
result, the client will need to follow strict guidelines. Because implementing the intended system
will only require minimal or nonexistent adjustments, it must have the fewest requirements
possible.
3.3.3 SOCIAL FEASIBILITY
One of the goals of the study is to assess the system's level of user acceptability. This entails
instructing the user on the proper usage of the technology. The user needs to accept the system as
a necessity rather than a threat. The methods employed to acquaint and instruct users about the
system will dictate the degree of acceptance among them. Being the final user of the system, he
needs more confidence to provide some insightful criticism, which is much welcomed.
3.3.4 FEASIBILITY ANALYSIS
Finding that the system request is doable is a significant result of the preliminary investigation.
Only if it is doable with the time and resources available will this be viable. The many
feasibilities that require analysis include
 Operational Feasibility
 Economic Feasibility
 Technical Feasibility

DEFECT DETECTION
Operational Feasibility
Operational feasibility is the analysis of the system's potential that has to be developed.
Operationally, this approach relieves the administrator of all his stress and assists him in efficiently
monitoring the project's advancement. There is no doubt that this type of automation will save the
time and energy that were previously used for manual labor. The analysis demonstrates the
operational viability of the system.
Economic Feasibility
An evaluation of the financial case for a computer-based project is known as economic feasibility
or cost-benefit analysis. The hardware project had a cheap cost because hardware was installed
from the start and served many tasks. Any number of employees linked to the organization's local
area network (LAN) can utilize this tool at any time because it is network-based. The organization's
current resources will be used to construct the virtual private network. Therefore, the proposal is
financially viable.
Technical Feasibility
Technical feasibility, according to Roger S. Pressman, is the evaluation of the organization's
technical resources. The company requires IBM compatible computers that are linked to the
Intranet and Internet using a graphical web browser. The system is designed to operate in a
platform-independent setting. The system is developed using Java Server Pages, JavaScript,
HTML, SQL Server, and WebLogic Server. The technical viability analysis has been completed.
The system can be created using the current facility and is technically possible.

DEFECT DETECTION
3.3.5 REQUEST APPROVAL
Not every project that is requested is worthwhile or practical. Only a small percentage of the
project requests that client users submit to an organization are actually taken up. On the other
hand, undertakings that are both desirable and doable ought to be scheduled. Following approval,
a project request's projected cost, priority, completion time, and staffing needs are utilized to
decide where on the project list to place it. Indeed, once the aforementioned conditions are met,
development activity can begin.
3.4 SYSTEM DESIGN AND DEVELOPMENT
3.4.1 INPUT DESIGN

Input design is critical to the software development life cycle and requires developers to give it
their fullest attention. Providing accurate data to the application is the aim of the input design. In
order to minimize errors that could occur during feeding, it is intended that inputs be efficiently
designed. Software Engineering Concepts states that the input forms or screens are designed to
provide users with control over the validation of the input limit, range, and other pertinent
validations.
Input screens are present in nearly every module of this system. Error messages are designed to
warn users of mistakes they make and point them in the correct direction to prevent the entry of
erroneous data. Let's examine this in more detail under module design.
Input design is the process of converting user-created data into a format that computers can
understand. The goal of the input design is to provide logical and error-free data entry. Errors in
the input are managed by the input design. User-friendliness has been the primary focus of the
application's design. Because of the way the forms are designed, while processing the pointer is
automatically relocated to the required entry area. In certain cases, the user is further provided with
the option to select a suitable input from a list of options related to the field.
Each entry of data needs to be verified. The user can go on to the following pages if they have
completed all of the entries on the current page. Every time a user enters inaccurate data.

DEFECT DETECTION
3.4.2 OUTPUT DESIGN

The primary purpose of the computer's output is to create a reliable channel of communication
within the company, particularly between the project manager and his team, or between the
administrator and the clients. With the help of the system the VPN generates, the project manager
may manage his clients by adding new ones, assigning them new projects, monitoring the validity
of the projects, and allowing user-level access to folders based on the projects that are assigned to
each client. After completing one job, the client could be assigned another. User authentication
protocols are respected from the outset. New users can be created by the administrator, or a new
user, however the administrator is the only one in charge of assigning tasks and confirming new
users.
An application launches when it is used for the first time. Once the server is started, the browser
used is Internet Explorer. The other connected computers can operate as clients and the server
computer as the administrator because the project will be carried out over a local area network.

DEFECT DETECTION
3.5 UML Diagrams

Fig3: Use Case

DEFECT DETECTION
3.5.1 Class Diagram
Fig4: Class Diagram

DEFECT DETECTION
3.5.2 Sequence Diagram
Fig 5: Sequence Diagram

DEFECT DETECTION
3.5.3 Data Flow Diagram
Import libraries
Yes NO NO PROCESS
VERIFY
Importing the dataset
Data processing
Feature Selection
Splitting the data into train & test
Building the model- FasterRCNN, YoloV5 with CA Attention, Extension

YoloV6 with VGG16
Training the model
Signup & sign in
User input
Final outcome
End process
Fig 6 : Data Flow Diagram

DEFECT DETECTION
3.5.4 ACTIVITY Diagrams
Fig 7: Actrivity Diagram

DEFECT DETECTION
3.5.5 Collaboration diagram:
Fig 8: Service Provider
3.5.6 Component diagram:

DEFECT DETECTION
3.5.7 Deployment diagram:
User System

DEFECT DETECTION
3.6 Modules
Data loading: using this module we are going to import the dataset.
Data Processing: Using the module we will explore the data.
Splitting data into train & test: using this module data will be divided into train & test
Model generation: Model building - FRCNN - Improved YoloV5 - Extension YoloV6.

Algorithms accuracy calculated
User signup & login: Using this module will get registration and login
User input: Using this module will give input for prediction
Prediction: final predicted displayed

DEFECT DETECTION
3.7 Hardware and Software Requirements
3.7.1 Hardware Requirements
Processor : i5 and above
RAM : 8 GB
Hard Disk : 25 GB
Key Board : Standard Windows Keyboard
Mouse : Two or Three Button Mouse
Monitor : SVGA
3.7.2 Software Requirements

Operating system : Windows 11.
Coding Language : Python.
Front-End : Flask
Back-End : Jupiter-Notebook
Designing : Html, css, java script,Bootstrap4
Data Base : SQLlite3.

DEFECT DETECTION
CHAPTER-4
4. Implementation
4.1 Python
Python is an object-oriented, interpreted, high-level programming language. The design of Python

emphasizes readability. It contains fewer syntactical structures than other languages and usually
uses English terms in contrast to other languages that use punctuation.
Python is Interpreted: The interpreter handles Python at runtime. Your software doesn't need to
be assembled in order to execute. This reminds me of PHP and PERL.
Interactive: Python enables you to sit down at a prompt and work directly with the interpreter to
create programs.
Object-oriented programming, which encapsulates code into objects, is supported by Python.
Python is an Excellent Language for Novice Programmers: If you're just starting out with
programming, Python is an excellent language. It makes a wide range of program creation simple,
including basic text processing, Web browsers, and games.
4.1.1 History of Python
At the Netherlands' National Research Institute for Mathematics and Computer Science in the late
eighties and early nineties, Guido van Rossum created the programming language Python.
Numerous additional languages, such as ABC, Modula-3, C, C++, Algol-68, Small Talk, Unix
shell, and other scripting languages, are developed from Python. Copyright protects Python. The
GNU General Public License (GPL) is now used to license Python source code, just like it does
Perl. Although a core development team at the institute currently manages Python, Guido van
Rossum continues to play a crucial role in steering its direction.
4.1.2 Features of Python
Python's features include:
 Easy-to-learn: Python has few keywords, simple structure, and a clearly defined syntax.

DEFECT DETECTION
 Easy-to-read: Python code is more clearly defined and visible to the eyes.
 Easy-to-maintain: Python's source code is fairly easy-to-maintain.
 A broad standard library: Python's bulk of the library is very portable and cross-platform
compatible on UNIX, Windows, and Macintosh.
 Interactive Mode: Python has support for an interactive mode which allows interactive
testing and debugging of snippets of code.
 Portable: Python can run on a wide variety of hardware platforms and has the same
interface on all platforms.
 Extendable: You can add low-level modules to the Python interpreter. These modules
enable programmers to add to or customize their tools to be more efficient.
 Databases: Python provides interfaces to all major commercial databases.
 GUI Programming: Python supports GUI applications that can be created and ported to
many system calls, libraries and windows systems, such as Windows MFC, Macintosh,
and the X Window system of Unix.
 Scalable: Python provides a better structure and support for large programs than shell
scripting.

DEFECT DETECTION
Python has a big list of good features:
• It can be used as a scripting language or to compile large programs into byte-code; it supports
both functional and structured programming approaches in addition to object-oriented
programming.
• It provides very high-level dynamic data types and enables dynamic type checking. It integrates
well with C, C++, COM, Java, ActiveX, CORBA, and C. It makes garbage collection automatic.
4.2 TECHNOLOGY DESCRIPTION
Python is an object-oriented, high-level, general-purpose, interpreted programming language.
Python is an interpreted language with a syntax that lets programmers express ideas in fewer lines
of code than they might in languages like C++ or Java. Its design philosophy emphasizes code
readability, most notably by using whitespace indentation to delimit code blocks rather than curly
brackets or keywords. It offers building blocks that facilitate understandable programming at both
small and large sizes. There are interpreters for Python on a wide range of operating systems. The
non-profit Python Software Foundation oversees Python, the reference implementation of Python.
Python has automated memory management and a dynamic typing system. It features a sizable
and extensive standard library and supports a variety of programming paradigms, such as
imperative, functional, procedural, and object-oriented programming.
What is Python?
Python has automated memory management and a dynamic typing system. It features a sizable
and extensive standard library and supports a variety of programming paradigms, such as
imperative, functional, procedural, and object-oriented programming.
One well-liked programming language is Python. Guido van Rossum was the creator, and it was
published in 1991.
It is employed in server-side web development, software development, mathematics, and system

programming.
What is Python capable of?

Web applications can be created on a server using Python.

DEFECT DETECTION
Workflows can be created with Python in conjunction with other programs.
Database systems can be connected to using Python. It has the ability to read and edit files.
Complex mathematics and handling of large data can be accomplished with Python.
Why Use Python?

Python operates on various platforms, including Windows, Mac, Linux, Raspberry Pi, and others.
Python's syntax is straightforward and resembles that of the English language. Compared to some
other programming languages, Python's syntax enables programmers to construct programs with
less lines.

DEFECT DETECTION
CHAPTER-5
5. Testing
System testing, also referred to as system-level tests or system-integration testing, is the process
in which a quality assurance (QA) team evaluates how the various components of an application
interact together in the full, integrated system or application. System testing verifies that an
application performs tasks as designed. This step, a kind of black box testing, focuses on the
functionality of an application. System testing, for example, might check that every kind of user
input produces the intended output across the application.
Phases of system testing:
A video tutorial about this test level. System testing examines every component of an application
to make sure that they work as a complete and unified whole. A QA team typically conducts system
testing after it checks individual modules with functional or user-story testing and then each
component through integration testing.
If a software build achieves the desired results in system testing, it gets a final check via acceptance
testing before it goes to production, where users consume the software. An app-dev team logs all
defects, and establishes what kinds and amount of defects are tolerable.
Software Testing Strategies:
Optimization of the approach to testing in software engineering is the best way to make it effective. A
software testing strategy defines what, when, and how to do whatever is necessary to make an end-product
of high quality. Usually, the following software testing strategies and their combinations are used to achieve
this major objective:
5.1Static Testing:
The early-stage testing strategy is static testing: it is performed without actually running the developing
product. Basically, such desk-checking is required to detect bugs and issues that are present in the code itself.
Such a check-up is important at the pre-deployment stage as it helps avoid problems caused byerrors in the
code and software structure deficits.

DEFECT DETECTION
5.2 Structural Testing:

It is not possible to effectively test software without running it. Structural testing, also known as
white-box testing, is required to detect and fix bugs and errors emerging during the pre-production
stage of the software development process. At this stage, unit testing based on the software
structure is performed using regression testing. In most cases, it is an automated process working
within the test automation framework to speed up the development process at this stage.
Developers and QA engineers have full access to the software’s structure and data flows (data
flows testing), so they could track any changes (mutation testing) in the system’s behavior by

DEFECT DETECTION
comparing the tests’ outcomes with the results of previous iterations (control flow testing).
5.3 Behavioral Testing:

The final stage of testing focuses on the software’s reactions to various activities rather than on
the mechanisms behind these reactions. In other words, behavioral testing, also known as black-
box testing, presupposes running numerous tests, mostly manual, to see the product from the
user’s point of view. QA engineers usually have some specific information about a business or
other purposes of the software (‘the black box’) to run usability tests, for example, and react to
bugs as regular users of the product will do. Behavioral testing also may include automation
(regression tests) to eliminate human error if repetitive activities are required. For example, you
may need to fill 100 registration forms on the website to see how the product copes with such an
activity, so the automation of this test is preferable.

DEFECT DETECTION
CHAPTER-6
6.TEST CASES:
S.NO INPUT If available If not available
1 User signup User get registered into There is no process

the application
2 User signin User get login into the There is no process

application
3 Enter input for prediction Prediction result There is no process

displayed

DEFECT DETECTION
7. Future Enhancement
Despite solving the issue with the previous work, it has several drawbacks. The likelihood of the
estimate will be zero in the event that there are no class labels. Adding more machine learning
classification models to the proposed study would improve overall performance and boost the
accuracy of crime prediction in the future. By considering the income data for areas, it helps to
provide a better analysis for future improvements by predicting whether there is a relationship, if
any, between a neighborhood's crime rate and its residents' income levels.

DEFECT DETECTION
CHAPTER-8
8. CONCLUSION
To solve the problem of low accuracy of pedestrian target detection by a single sensor at night
when the road environment is complex and the contrast of the surrounding environment is low,
this paper proposes a nighttime pedestrian detection algorithm based on the fusion of visual sensors
and MMW radar. The YOLOv5-ours algorithm is used to detect infrared images of pedestrians at
night, which improves the detection accuracy by 3.7% compared to the original YOLOv5
algorithm while increasing the detection speed from 29 FPS to 42 FPS. MMW radar can detect
target location and state information, and the EKF tracking algorithm is used to continuously
estimate and correct filter noise online for MMW radar observation data. statistical features to
detect and track the radar detection targets to improve the reliability and stability of radar data.
Then a decision-level fusion detection algorithm is proposed to compensate for the false detection
misses of infrared cameras and millimeter-wave radar. This paper designs target detection
experiments under different nighttime traffic road scenarios, and the experimental results show
that the accuracy of the fusion algorithm for nighttime pedestrian detection can reach 95.57%
under nighttime traffic scenarios such as small targets at a long distance, the presence of pedestrian
occlusion and foggy weather, which effectively reduces the false detection rate of single sensor
target detection and to a certain extent solves the fusion algorithm target detection performance
The problem of weak target detection performance of the fusion algorithm is solved to a certain
extent, and the accuracy and robustness of nighttime pedestrian detection are improved. The
current study cannot be directly applied to self-driving cars. In the next step, we will integrate the
sensors into the controller of self-driving cars through embedded devices to achieve dynamic real-
time target detection of pedestrians in front of self-driving cars in night scenarios, enriching the
application scenarios of this study.

DEFECT DETECTION
APPENDIX-A:SCREEN SHOTS
1. User Login
2. Login Details

DEFECT DETECTION
2. User Admin
3. Upload File

DEFECT DETECTION
4. View Details
5. Upload Details

DEFECT DETECTION
6. Coding

DEFECT DETECTION
APENDIX-B: REFERENCES
[1] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, ‘‘You only look once: Unified, real-time
object detection,’’ in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Jun. 2016, pp.
779–788.
[2] J. Redmon and A. Farhadi, ‘‘YOLO9000: Better, faster, stronger,’’ in Proc. IEEE Conf.
Comput. Vis. Pattern Recognit. (CVPR), Jul. 2017, pp. 6517–6525.
[3] J. Redmon and A. Farhadi, ‘‘YOLOv3: An incremental improvement,’’ 2018,

arXiv:1804.02767.
[4] A. Bochkovskiy, C.-Y. Wang, and H.-Y. Mark Liao, ‘‘YOLOv4: Optimal speed and accuracy
of object detection,’’ 2020, arXiv:2004.10934.
[5] P. Jiang, D. Ergu, F. Liu, Y. Cai, and B. Ma, ‘‘A review of YOLO algorithm developments,’’
Proc. Comput. Sci., vol. 199, pp. 1066–1073, 2022.
[6] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, Alexander, and C. Berg,
‘‘SSD: Single shot MultiBox detector,’’ in Proc. 14th Eur. Conf. Comput. Vis. (ECCV), 2016,
pp. 21–37.
[7] T. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, ‘‘Focal loss for dense object detection,’’
in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), Oct. 2017, pp. 2999–3007.
[8] R. Girshick, J. Donahue, T. Darrell, and J. Malik, ‘‘Rich feature hierarchies for accurate object
detection and semantic segmentation,’’ in Proc. IEEE Conf. Comput. Vis. Pattern Recognit.
(CVPR), Jun. 2014, pp. 580–587.
[9] R. Girshick, ‘‘Fast R-CNN,’’ in Proc. IEEE Int. Conf. Comput. Vis. (ICCV), Dec. 2015, pp.
1440–1448.
[10] S. M. Patole, M. Torlak, D. Wang, and M. Ali, ‘‘Automotive radars: A review of signal
DEFECT DETECTION
processing techniques,’’ IEEE Signal Process. Mag., vol. 34, no. 2, pp. 22–35, Mar. 2017.
[11] X. Liu, T. Yang, and J. Li, ‘‘Real-time ground vehicle detection in aerial infrared imagery
based on convolutional neural network,’’ Electronics, vol. 7, no. 6, p. 78, May 2018.
[12] M. T. Mahmood, S. R. A. Ahmed, and M. R. A. Ahmed, ‘‘Detection of vehicle with infrared

images in road traffic using YOLO computational mechanism,’’ in Proc. IOP Conf. Ser., Mater.
Sci. Eng., 2020, pp. 1–9.
[13] Y. Liu, H. Su, C. Zeng, and X. Li, ‘‘A robust thermal infrared vehicle and pedestrian
detection method in complex scenes,’’ Sensors, vol. 21, no. 4, p. 1240, Feb. 2021.
[14] Z. J. Zhu et al., ‘‘A parallel fusion network-based detection method for aerial infrared
vehicles with small targets,’’ J. Photon., vol. 51, no. 2, pp. 182–194, 2022.
[15] T. Wu, T. Wang, and Y. Liu, ‘‘Real-time vehicle and distance detection based on improved
YOLO v5 network,’’ in Proc. 3rd World Symp. Artif. Intell. (WSAI), Jun. 2021, pp. 24–28.
[16] H.-K. Jung and G.-S. Choi, ‘‘Improved YOLOv5: Efficient object detection using drone
images under various conditions,’’ Appl. Sci., vol. 12, no. 14, p. 7255, Jul. 2022.
[17] X. Zhu, S. Lyu, X. Wang, and Q. Zhao, ‘‘TPH-YOLOv5: Improved YOLOv5 based on
transformer prediction head for object detection on drone-captured scenarios,’’ in Proc.
IEEE/CVF Int. Conf. Comput. Vis. Workshops (ICCVW), Oct. 2021, pp. 2778–2788.
[18] X. Jin, Z. Li, and H. Yang, ‘‘Pedestrian detection with YOLOv5 in autonomous driving
scenario,’’ in Proc. 5th CAA Int. Conf. Veh. Control Intell. (CVCI), Oct. 2021, pp. 1–5.
[19] M. Kasper-Eulaers, N. Hahn, S. Berger, T. Sebulonsen, Ø. Myrland, and P. E. Kummervold,
‘‘Short communication: Detecting heavy goods vehicles in rest areas.

Rina Project

Uploaded by

Copyright:

Available Formats

Rina Project

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rina Project

Uploaded by

Copyright:

Available Formats

A PROJECT REPORT

ENHANCING NIGHTTIME PEDESTRIAN DETECTION

MASTER OF COMPUTER APPLICATIONS

Under the esteemed guidance of

MASTER OF COMPUTER APPLICATIONS

MASTER OF COMPUTER APPLICATIONS

Project Guide Head of the department

G.Rajasekhar Reddy M.tech(Ph.d). Dr. K. SUBBA REDDY, M. Tech, Ph.D.

Department of Computer Applications,

Rajeev Gandhi Memorial College of Engineering and Technology,Nandyal,

Sub – Academic Project Confirmation Letter

TRU RESEARCH SUPPORT

I express my profound sense of gratitude to Dr. K. SUBBA REDDY, Head of the

I express my profound sense of gratitude to Dr. T. JAYA CHANDRA PRASAD,

Principal, and Rajeev Gandhi Memorial College of Engineering and Technology

I hereby declare that the project report entitled “ENHANCING NIGHTTIME

CHAPTER PAGE NO.

2.1 Existing System 3

2.3 Proposed System 4

3. SYSTEM MODEL / DESIGN 6

3.2. Software Devolopment Life Cycle 8

3.2.1 Spiral Model 8

3.3 Feasibility Study 9

3.3.1 Economical Feasibility 10

3.3.2 Technical Feasibility 10

3.3.3 Social Feasibility

3.3.4 Feasibility Analysis 10

3.3.5 Request Approval 12

3.4.1 Input Design 12

3.4.2 Output Design 13

3.5 UML Diagram 14

3.5.1 Class Diagram 15

3.5.2 Sequence Diagram 16

3.5.3 Data Flow Diagram 17

3.5.4 Activity Diagram 18

3.5.5 Collaborative Diagram 19

3.5.6 Component Diagram 19

3.5.7 Deployment Diagram 20

3.7 Hardware and Software Requirements 22

3.7.1 Hardware Requirements 22

3.7.2 Software Requirements 22

4.1.1 History Of Python 23

4.1.2 Features of Python 23

4.2 Technology Discription 25

5.1 Structural Testing 28

Appendix: A. Project Screenshots 33

Appendix: B. Project References 37

FIG NO. NAME OF THE FIGURE PAGE NO.

Fig: 1 Architecture Diagram 4

Fig: 2 Spiral Model 8

Fig: 3 Use case 13

Fig: 4 Class Diagram 14

Fig: 5 Sequence Diagram 15

Fig: 6 Data Flow Diagram 16

Fig: 7 Remote User 17

Fig: 8 Service Provider 18

Dept. Of MCA, RGMCET, Nandyal 1

Dept. Of MCA, RGMCET, Nandyal 2

2.1 Existing System

Dept. Of MCA, RGMCET, Nandyal 3