To cope with the decline of automatic obstacle avoidance ability of mine inspection robots due to their susceptibility to dust,dim light and other factors,studies an automatic obstacle avoidance strategy based on machine vision,which can effectively avoid obstacles in complex environments.A machine vision algorithm based on improved non-local mean filtering and multi-scale B-spline wavelet transform is proposed,by which higher quality images and the edges of obstacles both can be abtained to ensure that the rescue robot can recognize and avoid obstacles independently.Simulation results show that the proposed algorithm is superior to traditional ones in terms of image noise reduction and edge detection.
关键词
救援机器人/自动避障/非局部均值算法/图像降噪/小波变换/边缘检测
Key words
rescue robot/automatic obstacle/non-local mean algorithm/noise reduction/wavelet transform/edge detection
维普
万方数据