An improved hand detection by employing corner detector

This paper proposes a fast hand detection method by employing corner detector to solve the finger fragment problem occurred during hand detection. Weshrink the ROI into a much smaller range while performing corner detection which can reduce the time consumption efficiently. The proposed method needs neither data training nor wrist segmenting. It also does not need to find the contour of hand. These advantages make the system favorably to be implemented in embedded systems. The experimental results show that the proposed method can detect hands faster with lower computational consumption.