Gesture Recognition System Based On Kinect Depth Information

In the continuous advance of science and technology, people pay more attention to humane and simplified human-computer interaction. The gesture recognition is an important mean of human-computer interaction. Conventional gesture recognition method has a high demanding of human and environmental background. To solve this problem, using Microsoft launched the Kinect sensor can effectively get the depth of field images. It has better interactivity depth information based on gesture recognition. In this paper, the Kinect depth information is designed for static and dynamic real-time human gesture recognition algorithm.The localization of the joint points of the hand is realized by using the Kinect skeleton.The depth image is acquired by the depth sensor, and the joint points of the hand are tracked continuously; After locating the position of the hand, the region of interest is intercepted; The segmentation image is processed by morphology, and the circular rate, filling rate, perimeter rate, convex hull, convex defect, Hu moments of the hand contour are 9 kinds of features; Six kinds(0 to 5) of gesture are recognized using SVM method. Recognition rate and robustness of gesture recognition experiments are conducted in static and dynamic environment respectively. The experimental results show that the proposed algorithm can achieve better recognition result in a variety of environments.Dynamic gesture recognition based on depth information uses the good characteristics of Kinect skeleton acquisition and tracking. The trajectory of the body’s 10 skeleton points are set as the input feature vectors. Taking the center and normalization method to process the acquired feature vectors. Multiplayer dynamic gesture data are collected to establish the gesture library in training phase. Using template matching of dynamic time warping method to recognize the left-draw, right-draw, push hands, clapping, hands in the air five kinds of dynamic gesture. Recognition rate and robustness of gesture recognition experiments are conducted in static and dynamic environment respectively. Using the proposed system to achieve the control of appropriate software and the experimental results show that the proposed algorithm can achieve better recognition result in a variety of environments.