Design and Performance Test of Gesture Recognition Control System Based on Machine Learning in Artificial Intelligence
Download as PDF
Under the background of artificial intelligence, with the arrival of big data era, the scale of data to be processed is becoming larger and larger, and machine learning has gradually penetrated into all aspects of people's life and work. With people's increasing requirements for accuracy and ease of use of gesture recovery and gesture recognition, key problems and challenges such as lack of prior knowledge representation, compromise between accuracy and delay, tedious calibration process and insufficient labeled data are superimposed alternately. In order to realize fast and accurate static gesture recognition, a new classification method based on support vector machine is proposed in this paper. By selecting proper function subset, the recognition rate of discriminant function is optimized, and a learning machine with generalization ability and optimal classification ability is obtained. The gesture recognition test is carried out by Kinect, and the results show that the gesture recognition algorithm based on improved SVM vector machine has better accuracy and accuracy.
Artificial Intelligence, Machine Learning, SVM, Gesture Recognition