VISVESVARAYA TECHNOLOGICAL UNIVERSITY
JNANA SANGAMA, BELAGAVI -590 014
DEPARTMENT OF
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
A PROJECT
REPORT ON
“AERIAL ACTIVITY RECOGNITION SYSTEM”
Submitted in partial fulfillment of the requirement for the award of
BACHELOR OF ENGINEERING
IN
ARTIFICIAL INTELLIGNCE & MACHINE LEARNING
Submitted By
MADHURA C D (4EG21AI019)
RITU CHAKRABORTY (4EG21AI035)
SURYA S (4EG21AI043)
UDAY KIRAN M S (4EG21AI049)
Project Guide Project Coordinator
Soumya M S M.Tech Manjunath S M.Tech
Asst. Prof. Dept. of AI&ML Asst. Prof. Dept. of CSE
GEC CHALLAKERE GEC CHALLAKERE
Head of the Department
SAHANA V M.Tech., Ph.d
Asst. Prof. Dept. of AI&ML
GEC CHALLAKERE
2024-25
GOVERNMENT ENGINEERING COLLEGE, CHALLAKERE
Ballary road, Challakere, Chitradurga- 577522
(Affiliated to Visvesvaraya Technological University, Belagavi, Recognized by AICTE, New Delhi and Approved by Government
of Karnataka)
GOVERNMENT ENGINEERING COLLEGE
CHALLAKERE-577522
(Affiliated to Visvesvaraya Technological University, Belagavi, Recognized by AICTE, New
Delhi and Approved by Government of Karnataka)
Ballary road, Challakere, Chitradurga- 577522
DEPARTMENT OF
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
CERTIFICATE
This is to certify that the project work entitled “ARTIFICIAL INTELLIGNCE
& MACHINE LEARNING is a bonfired work carried out by MADHURA C D
(4EG21AI019), RITU CHAKRABORTY (4EG21AI035), SURYA S (4EG21AI043),
UDAY KIRAN M S (4EG21AI049) in partial fulfillment for the 7th semester of
Bachelor of Engineering in Artificial Intelligence and Machine Learning of the
Visvesvaraya Technological University, Belagavi during the academic year:
2024- 2025. It is certified that all corrections /suggestions indicated for the
Internal Assessment have been approved as it is satisfying the academic
requirements in respect of project work prescribed for the Bachelor of
Engineering Degree
.…………………… …………….…… …..……..…………
Signature of the Guide Signature of the HOD Signature of the Principal
Soumya M S M.Tech.. Sahana V M.Tech., Ph.d. Dr. M M Benal M.E., Ph.d.
Asst.Prof., Dept. of AI&ML, Asst. Prof., Dept. of AI&ML, Principal.
GEC CHALLAKERE GEC CHALLAKERE GEC CHALLAKERE
External Viva
Name of the Examiners Signature with Date
1. …………………………. ……………………
2. …………………………. ………..…………...
GOVERNMENT ENGINEERING COLLEGE
CHALLAKERE-577522
(Affiliated to Visvesvaraya Technological University, Belagavi, Recognized by AICTE, New Delhi
and Approved by Government of Karnataka)
Ballary road, Challakere, Chitradurga- 577522
DEPARTMENT OF
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
DECLARATION
We, MADHURA C D (4EG21AI019), RITU CHAKRABORTY (4EG21AI035),
SURYA S (4EG21AI043), UDAY KIRAN M S (4EG21AI049) are the students pursuing in
Bachelor of Engineering, Department of Artificial Intelligence And Machine Learning,
Government Engineering College Challakere, hereby declare that the project work entitled
“AERIAL ACTIVITY RECOGNITION SYSTEM” has been carried out by us under the
supervision and guidance of department faculty, submitted in partial fulfillment for the award of
Bachelor of Engineering degree in Artificial Intelligence And Machine Learning from
Visvesvaraya Technology University, Belgaum during academic year 2024-25. We also declare
that, this work has not been submitted previously for the award of any degree or diploma, by us, to
any institution.
Place: Challakere Signature:
Date:
ACKNOWLEDGEMENT
The satisfaction and euphoria that accompany the completion of any task would be
incomplete without the mention of the people who made it possible, whose constant guidance
and encouragement ground our efforts with success.
We consider it as a privilege to express my gratitude and respect to all those who
guided us in completion of this Project work.
We will remain indebted forever to all the Management Authorities of Government
Engineering College, Challakere for support to carrying out this Project work successfully.
We extend our sincere and heartful thanks to our Dr. M M Benal M.E,Ph.d., Principal, and
Asst. Prof, Sahana VM.Tech.,Ph.d HOD, Department of AI&ML, for providing us the right ambience,
constant inspiration and support for carrying out this Project work successfully.
We are profoundly indebted to our Project work guide, Asst. Prof. SOUMYA M S
M.Tech., for innumerable acts of timely advice, encouragement and we sincerely express our
gratitude to her.
We express our thanks to the Project Coordinator Asst. Prof. Manjunath S M.Tech., of
Department of Computer Science & Engineering, for his guidance, invaluable help, advice
and suggestions.
We express our enormous pleasure and thankfulness to all the teaching and non-
teaching staff of the Department of Artificial Intelligence and Machine Learning.
Cordially
MADHURA C D (4EG21AI019)
RITU CHAKRABORTY (4EG21AI035)
SURYA S (4EG21AI043)
UDAY KIRAN M S (4EG21AI049)
ABSTRACT
Aerial activity recognition is a crucial task in surveillance, disaster response, and
security applications. This project aims to develop a robust Aerial Activity Recognition System
that utilizes deep learning and computer vision techniques to analyze aerial footage and
accurately classify human activities. The system leverages drone-captured videos, satellite
imagery, or UAV-mounted cameras to detect and recognize activities such as walking, running,
fighting, or vehicle-related movements. By integrating convolutional neural networks (CNNs),
recurrent neural networks (RNNs), and transformer-based architectures, the system enhances
recognition accuracy under varying environmental conditions. The proposed model undergoes
extensive training on aerial datasets, employing techniques like data augmentation, transfer
learning, and real-time object tracking to improve robustness. The implementation of this
system can be instrumental in applications such as military surveillance, crowd monitoring,
disaster relief, and border security, ensuring efficient and automated aerial activity analysis.
Aerial activity recognition plays a vital role in security, surveillance, and disaster
management by enabling automated detection and classification of human activities from aerial
footage. This project aims to develop an intelligent Aerial Activity Recognition System using
deep learning and computer vision techniques to analyze video and image data captured from
drones, UAVs, or satellites. The system is designed to recognize various activities such as
walking, running, fighting, vehicle movements, group interactions, and emergency situations.
i
CONTENTS
Page No.
Abstract i
Contents ii
List of Figures iv
List of Table v
Chapter 1 Introduction
1.1 Project Overview 1
1.2 Objectives 2
1.3 Problem Statement 2
1.4 Scope 3
Chapter 2 Literature Survey
2.1 Existing System 4
2.2 Proposed System 5
2.3 Overview of Literature Survey 10
Chapter 3 System Study & Analysis
3.1 Hardware & Software Requirements 12
3.2 Feasibility Study 14
Chapter 4 System Design
4.1 System Architecture 15
4.2 ER Diagram 15
4.3 System Perspective 16
4.4 Dataflow / Content Diagram 17
ii
Chapter 5 Implementation
5.1 Application User Interface 18
5.2 Coding 18
Chapter 6 Testing
6.1 Software Testing 21
21
6.1.1 Unit Testing
21
6.1.2 Integration Testing
21
6.1.3 Functional Testing 21
6.1.4. Functional Testing 22
6.1.5. Stress Testing 22
6.1.6. Usability Testing 22
6.1.7. Regression Testing 22
6.1.8. End-to-End (E2E) Testing 22
6.1.9. Security Testing 23
6.1.10. Simulations/Field Testing 23
6.1.11. Model Validation Testing 23
6.1.12. Data Quality Testing 23
6.1.13. Compatibility Testing 23
Chapter 7 Results & Analysis 24
Chapter 8 Conclusion 26
Bibliography
iii
LIST OF FIGURES
Figure Description Page No.
2.2.1 The proposed model for human action recognition in
aerial videos 5
2.2.2 CNN-LSTM model architecture 6
2.2.3 Illustration of an LSTM memory cell 6
2.2.4 Illustration of WGAN-GP model 8
4.2.1 Architecture diagram for Geographic info system 13
4.3.1 Architecture Diagram 14
4.4.1 Level 0 DFD for Geographic info layering with
embedded transfer system 14
4.4.2 Level 1 DFD of User 15
iv
LIST OF TABLES
Page No.
Table 2.1 Overview of Literature survey 10
