Image analytics using artificial intelligence (fire & smoke detection using artificial intelligence)

• Main Author: Jackson Pallattutharayil Varghese
• Other Authors: Yap Kim Hui
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/177087
• Institution: Nanyang Technological University

This project explores on leveraging deep learning techniques within the domain of artificial intelligence, with a keen focus on the evolution and critical significance of image analysis methodologies. Here, we focus on developing a fire and smoke detection system using artificial intelligence and visual data analytics, utilising the YOLO v8 model, the system achieves real-time detection of fire and smoke instances in video streams, with high precision and recall values. Motivated by the necessity of fire safety, especially in industrial environments, this research delves into the exploration of deep learning's potential in enhancing fire and smoke detection capabilities. By developing a robust fire detection system, we intend to overcome the limitations of traditional methods, offering improved accuracy, real-time monitoring, and reduced false alarms. Additionally, an email notification system is integrated to enable remote monitoring and prompt notification of detected fire incidents. Through the integration of advanced technologies and methodologies, the developed system has the capacity to transform fire safety practices by providing an automated and effective solution for hazard detection. The project underscores the significance of visual data analytics in enhancing safety measures and highlights the transformative impact of innovative technologies in addressing critical societal challenges.