J. He, Z.P. Zhang

Pages: 23-34

Abstract
In order to improve the obstacle detection ability of driverless vehicle moving image, an obstacle detection method of driverless vehicle based on visual filter operator is proposed. On the basis of constructing the objective function of driverless vehicle travel filter detection, matching the driving characteristics of unmanned vehicles. Through the visual filtering operator method, the obstacle of the unmanned vehicle is detected and processed, and the data to be acquired is preprocessed to reduce the error The processed obstacle data points are connected into an obstacle contour, and the obstacle detection and recognition are completed according to the obstacle classification information to improve the output signal-to-noise ratio of the image. The simulation results show that the obstacle detection signal-to-noise ratio of this method reaches 59.34dB, the detection accuracy rate reaches 97%, and the detection time is only 7s, and the average accuracy rate of monitoring different obstacles also reaches 96.8%. So, the output quality of the obstacle in the image of the unmanned vehicle traveling by this method is better.
Keywords: visual filtering; driverless vehicle; obstacle filtering detection; feature matching