An Adaptive Snake Based Shadow Segmentation for Robust Driver Fatigue Detection: A 3D Facial Feature based Photometric Stereo Perspective

Fatigue detection has many applications, one of them is intelligent transportation for accident prevention. It challenging to detect driver fatigue adopting photometric stereo and 3D fatigue-related facial action unit identification. During reconstruction shadows are encountered, causing error, therefore, accurate shadow removal crucial. This study presents a novel snake based approach handling in stereo. Seven light sources were employed region visibility the source each interest (ROI) was deduced. Surface normal estimation enhancement achieved via model labeling. Discrepancies measured among shadowed regions ROIs recorded. Light having high intersection with map excluded from process at pixel level. results an automatic related employing uncalibrated Lambertian surface under conditions. The developed method tested compared other established methods higher performance terms accuracy ( i.e. 97.85%) achieved. proves be well suited