0% found this document useful (0 votes)
15 views7 pages

Human Counting and Detection Project Report

This project aims to develop a human counting and detection system using computer vision techniques. The system will use Single Shot Detector (SSD) MobileNet model for object detection and centroid tracking for counting. It analyzes existing people counting methods and proposes a new system. The system is implemented in Python using OpenCV and TensorFlow libraries. Testing shows the model achieves 96.64% accuracy.

Uploaded by

Meghana S
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views7 pages

Human Counting and Detection Project Report

This project aims to develop a human counting and detection system using computer vision techniques. The system will use Single Shot Detector (SSD) MobileNet model for object detection and centroid tracking for counting. It analyzes existing people counting methods and proposes a new system. The system is implemented in Python using OpenCV and TensorFlow libraries. Testing shows the model achieves 96.64% accuracy.

Uploaded by

Meghana S
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

A Project Report on

“HUMAN COUNTING AND


DETECTION”
Submitted in Partial Fulfilment of the Requirement
For the Award of the Degree
MASTER OF COMPUTER APPLICATIONS
of

BANGALORE UNIVERSITY
By

MEGHANA.S

P03DB22S126028

Under the guidance of


Mr. K. MOHAN
ASSISTANT PROFESSOR

Department of Computer Applications


(2023-2024)

G.T. INSTITUTE OF MANAGEMENT STUDIES AND


RESEARCH
(Affiliated to Bangalore University, Recognized by Govt. of Karnataka and Approved by AICTE)
10, 15, Magadi Main Rd, Vishwaneedam Post, Muthuraya Swamy Extension, Sunkadakatte,
Bengaluru, Karnataka 560091
G.T. INSTITUTE OF MANAGEMENT STUDIES
AND RESEARCH
(Affiliated to Bangalore University, Recognized by Govt. of Karnataka and Approved by AICTE)Bengaluru,

Karnataka 560091.

DEPARTMENT OF COMPUTER APPLICATIONS


CERTIFICATE
This is to certify that the project entitled
“HUMAN COUNTING AND
DETECTION”
Submitted in partial fulfilment of the requirement of
Master of Computer Applications

Submitted by

MEGHANA.S

P03DB22S126028

During the academic year 2023-2024

Internal Guide: ........................................ HOD: ....................................................


[Link] Dr. R. Gogulan
Assistant professor HOD

Examiners 1. …………………………………….………………………

2. …………………………………….……………………….…….…
G.T. INSTITUTE OF MANAGEMENT STUDIES
AND RESEARCH
(Affiliated to Bangalore University, Recognized by Govt. of Karnataka and Approved by AICTE)Bengaluru,

Karnataka 560091.

DECLARATION

We are the student of III Semester MCA at G.T. Institute of Management Studies
and Research, Bangalore, hereby declare that, this project entitled Human Counting
And Detection is an independent and original work carried out by us and under
the guidance of
[Link] , Assistant Professor , Department of Computer Applications, G.T.
Institute of Management Studies and Research, Bangalore submitted to Bangalore
University, Bangalore in the partial fulfilment of requirement for the award of the
Degree of Master in Computer Applications during the academic year 2023-2024.

We, also declare that this report is Bonafide work carried out by us and has not been
submitted to any other University/Institution in the partial fulfilment of any other
Degree.

Place : Bengaluru

Date :

REGISTER NO. STUDENTS NAME SIGNATURE

P03DB22S126028 MEGHANA.S …………………………..


ACKNOWLEDGEMENT

The development and evolution of this project has been a never-ending source of both
challenges and joy to us. The euphoria and satisfaction that accompanies the success
of our project would be complete only with the mention of the people and the
organization that made it possible, whose support rewarded our effort with success.

First and the foremost, we would like to thank with immense gratitude to our beloved
Chief Administrator Officer Dr. B.G. Satyaprasad , Director Dr M. Deepika, and
Principal Dr. S. Manjunath for their continuous encouragement and moral support
which has been the key for the successful completion of the project.

We express our gratitude to Head of the Department Dr. R. Gogulan, Associate


Professor, Department of Computer Applications for his encouragement and support
in successful completion of the project.

We would like to express our heart-felt gratitude to our guide


…………………………………………… for his/her valuable suggestions and
excellent guidance rendered throughout this project.

Finally, we thank our parents and all my friends for helping and encouraging us
during the period of our project.

Place : Bangalore
Date :

REGISTER NO. STUDENTS NAME

P03DB22S126028 MEGHANA.S
ABSTRACT

Nowadays population growth increased. The linear growth of population has resulted
in a large number of people comes to public places. This system will thus provide the
count of the person in particular area malls, supermarket, etc. some business is only
depending upon on the customer schedule and its timing.

Hence , our work satisfies the problem and provides a solution by designing a
People counting system . So , the single shot detector (SSD) mobile net is proposed
along with a centroid tracker . This model replaces the VGG16 base network with a
Mobile Net network for better extracting features , and after the base network , it
connects with six convolutional layers for classification . The centroid tracking
algorithm obtains bounding box coordinates from an object detector SSD and applies
them to calculate the centre of a bounding box. After the centroid is calculated, it will
assign an id to each person and this model operates the dataset with training and
testing the data.

The maximum true positive rate (TPR) is 95.03% and FPR (false positive rate) is
0.08%. A total of 2416 positive images samples and 1218 negative samples. After that,
the accuracy of the model is obtained and also 96.64% accuracy rate is obtained by
testing the given data set. Sometimes method do not produce high accuracy due to the
high people counting, clothes of person, shadow issue etc.

This project investigates and reports benchmarks for detecting and enumerating humans
through real time images , videos and camera . This is very useful in various image
processing and performing computer vision [Link] schemes have been implemented
in Python programming language , and using various tech-stacks like OpenCv and
Tensorflow etc.

Keywords : Computer Vision, Human Detection, Enumeration.


CONTENT

[Link] PARTICULARS PAGE NO.

1 INTRODUCTION 1
[Link] of Project
[Link] Description

2 LITERATURE SURVEY 2-3

3 PROJECT ANALYSIS
3.1. Existing Analysis
3.2. Proposed Analysis
3.3. System Architecture 4-8
3.4. Flow Diagram
3.5. System specification

4 PROJECT DESIGN
[Link] Components 9-14
[Link] Components
[Link] Development
[Link] Flow Diagram (DFD)
[Link] Diagram
[Link] Diagram
[Link] case Diagram

5 PROJECT CODE 15-24

6 PROJECT TESTING 25-28


[Link] Testing
[Link] of Testing
[Link] cases

7 PROJECT IMPLEMENTATION
[Link] Detection
[Link] 29-40
[Link]
[Link]

8 PROJECT MAINTENANCE 41-45

9 CONCLUSION 46

10 FUTURE ENHANCEMENT 47

11 REFERENCES 48

12 BIBLIOGRAPHY 49
LIST OF FIGURES

Figure No Name of Figures Page No

3.3 System Architecture 6

3.4 Flow Diagram 7

4.4 Data Flow Diagram 10

4.5 Class Diagram 11

4.6 Sequence Diagram 12

4.7 Use Case Diagram 13-14

LIST OF TABLES

Table No Name of Table Page No

6.3.1 Unit Test Case-1 27

6.3.2 Unit Test Case-2 27

6.3.3 Unit Test Case-3 28

6.3.4 Unit Test Case-4 28

You might also like