GESTURE SEGMENTATION USING AN ADAPTIVE THRESHOLD ALGORITHM
Mei Wang, Jzau‐Sheng Lin, Zhou Xing Fu, Guo Qing Meng · 2014
Hand gesture segmentation is a key step for gesture recognition. Based on the construction of a new color space of skin model, a new dynamic-thresholding segmentation approach named Adaptive Threshold Segmentation Algorithm (ATSA) was further developed and segmentation effect evaluation was conducted. Some images of hand gesture were processed by using ATSA and the Fixed Threshold Segmentation (FTS) algorithm as well as the Similarity algorithm of Skin Color (SSC). Comparing with FTS and SSC algorithms, The ATSA is experimentally demonstrated that, the segmentation results have a less brightness impact, a lower redundancy rate, a lower rate of false alarm and missing, and a higher integrity rate.