PERFORMANCE EVALUATION OF FPGA ACCELERATOR FOR VISION BASED FIRE DETECTION SYSTEM

Abstract

Fire disasters are endangering human lives and pose a great threat to economic infrastructure, and the environment around the world. As a result, there is need for robust early detection of fire outbreaks to minimize loss of human lives, protect properties and the ecosystem from great damages. Conventional fire detection systems are based on electronic sensors. As a result, they suffer from the problems of transport delays, conduction delays, limited detection range, high false alarm, and not suitable for outdoor applications. Image processing and computer vision techniques for fire detection have been proposed in recent years to address the limitations of such sensor-based systems due to technological advances in electronics and computer science. However, most of these computer vision-based fire detection solutions are implemented in software platforms and as result, they have shortcomings of inefficiency, high hardware requirements and high cost. In this research study, a vision fire detection accelerator based on Field Programmable Gate Array (FPGA) was developed. MATLAB R2021a software was used for decoding the image dataset into pixel stream data. The design was captured in very high-speed integrated circuit HDL (VHDL). The design was synthesized with Xilinx Vivado 2021 design suite and simulated with Xilinx ISIMI. It has been shown that the design has achieved better resource utilization and power consumption compared to a similar work (15% Look-up Table (LUT), 1% Digital Signal Processor (DSP), 24% Input-Output (IO), 92.062W Dynamic and 1.029W Device Static).  The hardware accelerator which was developed as an Intellectual Property (IP) core can also be employed to speed up image processing and computer vision algorithms in embedded vision, smart camera and video analytics applications.