ENHANCING FIRE AND SMOKE DETECTION THROUGH COMPUTER VISION AND OPENCV-BASED ANALYSIS IN HSV COLOR SPACE

Abstract

Advances in computer vision, particularly with tools like OpenCV, are transforming fire and smoke detection by overcoming the limitations of traditional smoke detectors, which are often ineffective in large, open spaces with high ceilings or ventilation. Traditional detectors rely on particle or heat changes, leading to delayed alerts in ventilated areas. Computer vision-based systems, however, enable real-time analysis of video streams to identify early visual indicators of fire and smoke. Fire detection factors include analyzing unique flame characteristics such as color intensity in the RGB or HSV color space, shape irregularities using contour analysis, and flickering patterns through motion detection. Smoke detection involves techniques such as edge blurring, optical flow analysis, and texture variance to identify diffused, moving patterns indicative of smoke. By integrating OpenCV functions like color thresholding, contour detection, and background subtraction, these systems provide faster and more reliable alerts, enhancing safety in residential and commercial environments by mitigating risks associated with delayed responses.