Dynamic Hand Segmentation
Chandana N. Aithal, P Ishwarya, Soily Ghosh Sneha, C. N. Yashvardhan, Dayanand Kumar, Kaggere V. Suresh · 2021 IEEE 8th Uttar Pradesh Section International Conference on Electrical, Electronics and Computer Engineering (UPCON) · 2021
Real-time segmentation of moving areas in picture sequences is required for many vision systems, including automated visual surveillance, human-computer interactions, and very low-bandwidth telecommunications. Hand gesture segmentation faces numerous obstacles, including fluctuating illumination, shadows, and varied backgrounds. Despite the fact that there are a various techniques for detecting moving hand areas, they have a tendency to produce false positives. This paper proposes a robust hand gesture segmentation approach based on adaptive background removal with a skin colour based threshold. Under a variety of lighting conditions and complicated backgrounds, the suggested approach attempts to automatically extract hand gestures from a video. The proposed approach gives better accuracy when compared with the traditional techniques as per the findings of the experiments. Python and OpenCV are used to program the system.