论文部分内容阅读
为解决煤矿救援机器人在发生灾害后光线不足、煤尘粉尘严重等造成视觉系统中图像不清晰、画面模糊以致障碍物形状无法识别问题,通过实验模拟煤矿井下光照条件,在光照充足、低照度以及零照度3种不同光照条件下采集机器人视觉图像。采用直方图均衡化方法对图像进行增强处理,使机器人显示的图像更加清晰,通过边缘检测算法以及基于风水岭分割改进的算法对障碍物信息进行提取,最终可以得到机器人前方主障碍物的形状特征信息。
In order to solve the problem that the coal mine rescue robot is lack of light, dust and dust of the coal mine dust caused by the visual system image is not clear, the picture is blurred so that the shape of the obstacle can not be identified, the underground mine lighting conditions are simulated experimentally, Zero illumination three different light conditions to capture the robot vision images. The histogram equalization method is used to enhance the image to make the robot display the image clearer. The edge detection algorithm and the improved algorithm based on feng shui ridge segmentation are used to extract the obstacle information. Finally, the shape features of the main obstacle in front of the robot can be obtained information.