Study and analysis of blink detection in real time human system interactions

Abstract

Blink detection and the estimation of the blink rate occupy a preeminent place in human system interaction and envisage a renewed scope in establishing its capability across applications like visual eye gaze tracker, vision guard and drowsy driving. In a natural communication environment, a human face tends to have different roll, yaw and pitch orientation angles. Thus, maintaining a straight focus with a camera focus point is not always practically possible. Motivated by this challenge, the proposed idea incites to examine a single eye to detect the eye state (open or close) in various 3D orientations rather than both eyes, as blink detection is the rapid closure of both eyes simultaneously. In low light conditions, visible light eye blink identification performance is deteriorated due to the lack of illumination. The efforts owe to the use of thermal imaging that uses the intrinsic emissivity of thermal radiation from the human body. The proposed idea incorporates the advantages of thermal colour images to determine eye status (open or close). Since, the heat-map around people s eye has an average temperature around 97-990 F, the red colour is more prominent in the RGB thermal image. The RGB thermal image is split into individual components and combined with more weightage factor for red followed by green and then blue. To create resurgence in the utility world and bestow a fresh space for applications such as eye gaze movements and blink can be synchronized with cursor movement and click control for disabled people to operate on electronic gadgets. newline