Research on Gesture Recognition Algorithm of Steering Wheel Hands-off Detection Based on Threshold Self-Learning
Legally,the ability of hands-off detection has become an indispensable part of the current autonomous vehicle.Targeting at the contradiction between the universality and gesture recognition performance of the current hands-off detection system,a gesture recog-nition algorithm of capacitive hands-off detection system with self-learning ability is proposed.Through hand gesture vector method,the mathematical model of hand gesture is established and the signal characteristics of classical gestures in people are analyzed,so the hand gesture recognition algorithm with self-learning ability is built.This method can identify the driver's hand information and provide a more appropriate gesture judgment threshold for the driver combined with the requirement of minimum gesture recognition.According to the general gesture recognition performance evaluation method,the accuracy of gesture recognition results of the proposed algorithm is compared with that of the common algorithm without self-learning ability.It is found that under the same input,the overall accuracy rate of the proposed algorithm is higher than that of the general algorithm without self-learning ability under the condition of guaranteeing the basic performance of hands-off detection.
万方数据
维普
