Research on Target Tracking and Gesture Recognition System Combining Artificial Intelligence Algorithm and Computer Vision

Based on the continuous progress of information acquisition and information processing technology, computer vision, that is, how to use computer technology to efficiently and accurately obtain relevant information from environmental images or videos, and then analyze, judge and make decisions on things and phenomena in the objective world, has become a very important research topic. In this paper, a target tracking and gesture recognition model combining artificial intelligence (AI) algorithm and computer vision is proposed. An improved local binary pattern feature operator is used to represent the gesture region, and a gesture classifier is formed after learning and counting the samples through convolutional neural network (CNN). The classifier is used to detect the specified gesture in the video. The simulation results show that when the number of test samples starts to increase, the gesture recognition accuracy of each method shows a downward trend. However, compared with the other two methods, the accuracy of target tracking and gesture recognition based on this method is obviously higher. Compared with the recognition method based on single feature, this method has higher overall recognition rate, and greatly improves the recognition rate of gestures with high local similarity.