Hand Detecting and Positioning Based on Depth Image of Kinect Sensor
Yu Zhu · 2014
gesture features also shows great results. Zhang (5) tested Kinect with a hidden Markov model (HMM) in order to achieve gesture trajectory recognition and Wu (6) tested the skeleton point cloud. This study is about the use of image processing library of Pykinect SDK and of OpenCV to process Kinect depth image and get accurate hand segmentation. According to the relationship between the gray level of the depth image and the actual distance, we can detect the hand contour portion, and then calculate the center of gravity to find out the position of the hand. This method of segmentation process is much simpler than the traditional visual inspection methods. In the palm of source it creates a hand coordinate system, and a demonstration of a simple gesture trajectory tracking instance.