Page 1
IT4010 Research Project 4th Year, 1st Semester Final Report IT21166556 Nadeeshan I.U.N B.Sc. (Hons) Degree in Information Technology Sri Lanka Institute of Information Technology April 2025 1
Contenu du document
Page 2
DECLARATION Researchers affirm that this proposal is entirely this original work and does not contain any material previously submitted for a degree or diploma at any other university or institute of higher learning. Furthermore, researchers confirm that to the best of researcher’s knowledge, it does not include any previously published or written material by another person, except where appropriately cited in the text. Name Student ID Signature Nadeeshan I.U.N IT21166556 The above candidate is carrying out research for the undergraduate Dissertation under our supervision. Signature of the Supervisor Date 11/04/2025 ……………………………. ……………………………. 2
Page 3
ABSTRACT The increasing number of road accidents in urban environments has highlighted the limitations of traditional traffic monitoring systems that rely on manual reporting or costly sensor-based mechanisms. This dissertation proposes a deep learning-based accident detection system that operates in real-time using existing CCTV infrastructure. The primary objective is to develop a cost-effective, scalable, and automated solution that enhances emergency response capabilities and improves public safety in smart cities. The methodology is centered around convolutional neural networks (CNNs), explicitly utilizing the MobileNetV2 architecture due to its proven performance in balancing accuracy and computational efficiency. A labeled CCTV image dataset was curated and divided into training, validation, and testing sets with balanced class distribution. Several deep learning models were trained and evaluated, with MobileNetV2 achieving the highest test accuracy of approximately 92% and low inference latency of 250 milliseconds. The trained model was integrated into a real-time processing pipeline using OpenCV and Flask, with accident alerts delivered via WhatsApp API. The frontend dashboard, developed using Next.js and Firebase, allows for visualization, log history, and real-time interaction with the system. Testing confirmed the system’s ability to detect and classify accidents with high accuracy across diverse lighting and environmental conditions. The model was also converted to TensorFlow and deployed on a Raspberry Pi, demonstrating edge-level feasibility. The conclusions indicate that the system significantly enhances emergency response capabilities and reduces manual workload. Future work includes incorporating temporal models like LSTM or 3D-CNN, geolocation-based alerting, and multilingual notification support to increase adaptability and deployment readiness across urban scenarios. Keywords: Accident Detection, Deep Learning, CCTV Surveillance, Smart Cities, Real-Time Monitoring 3
Page 4
TABLE OF CONTENTS DECLARATION..........................................................................................................2 ABSTRACT..................................................................................................................3 LIST OF FIGURES .....................................................................................................6 LIST OF ABBREVIATIONS......................................................................................7 1.INTRODUCTION.....................................................................................................8 1.1 Background & Literature survey .....................................................................9 1.2 Research Gap....................................................................................................11 1.3 Research Problem ............................................................................................13 2. OBJECTIVES ........................................................................................................14 2.1 Main Objectives................................................................................................14 2.2 Specific Objectives ...........................................................................................14 3. METHODOLOGY.................................................................................................16 3.1 Research Area...................................................................................................16 3.2 System Diagram ...............................................................................................17 3.3 Individual System Diagram.............................................................................19 3.4 Data Collection.................................................................................................21 3.4.1 Dataset Source and Relevance .................................................................21 3.4.2 Dataset Structure and Labeling...............................................................21 3.4.3 Image Format and Preprocessing Requirements...................................22 3.4.4 Data Augmentation Strategy....................................................................23 3.4.5 Justification and Impact...........................................................................24 3.5 Data Preprocessing...........................................................................................24 3.5.1 Image Resizing and Format Standardization.........................................24 3.5.2 Dataset Directory Structuring .................................................................25 3.5.3 Image Augmentation (Training Only).....................................................25 3.5.4 Preprocessing Functions for Pretrained Models....................................25 3.5.5 Batch Management and Memory Optimization ....................................26 3.6 Commercialization aspects of the product.....................................................26 3.6.1 Market Demand and Relevance...............................................................26 3.6.2 Core Value Proposition.............................................................................27 3.6.3 Revenue Models.........................................................................................27 4
Page 5
3.6.4 Competitive Advantage.............................................................................28 3.6.5 Scalability and Expansion Opportunities ...............................................28 3.6.5 Target Audience.........................................................................................28 3.6.6 Commercial Versions................................................................................29 3.6.7 Entrepreneurial Opportunities................................................................30 3.7 Testing and Implementation............................................................................30 3.7.1 System Implementation Overview...........................................................30 3.7.2 Model Training and Performance Testing..............................................31 3.7.3 Real-Time Inference and Integration Testing.........................................33 3.7.4 Notification Delivery and Communication Testing................................34 3.7.5 Dashboard Testing and Usability Evaluation.........................................35 3.7.6 Deployment Testing (TFLite for Edge Devices)......................................35 4. RESULTS & DISCUSSION..................................................................................36 4.1 Results ...............................................................................................................36 4.1.1 Model Evaluation Metrics........................................................................36 4.1.2 Confusion Matrix Analysis.......................................................................37 4.1.3 Visual Result Validation............................................................................37 4.1.4 Real-Time Video Evaluation ....................................................................37 4.1.5 Alerting and Notification Performance...................................................38 4.1.6 Frontend Interface Synchronization .......................................................38 4.2 Research Findings............................................................................................38 4.3 Discussion..........................................................................................................40 5. PROJECT REQUIREMENT ...............................................................................42 5.1 Functional Requirements ................................................................................42 5.2 Non-Functional Requirements........................................................................43 5.3 Technical Functional Requirements...............................................................44 5.4 Work Breakdown Chart..................................................................................45 6.FUTURE ENHANCEMENT .................................................................................46 7. CONCLUSION.......................................................................................................48 8. REFERENCES.......................................................................................................51 9. APPENDICES ........................................................................................................53 5
Affichage des 5 premières pages. Téléchargez le document pour le voir en entier.
Commentaires (0)
Aucun commentaire pour le moment. Soyez le premier !