基于毫米波雷达和摄像头的农用车辆障碍物检测

Obstacle Detection of Agricultural Vehicles Based on Millimeter Wave Radar and Camera

本文针对单个传感器仅能获得局部环境特征信息,为进一步提高障碍物检测的准确性,提出一种融合毫米波雷达和视觉传感器的障碍物检测方法。通过毫米波雷达得到障碍物的数量与坐标点,同时对摄像头采集到的场景图像进行预处理,得到待分割图像。然后以毫米波雷达得到的坐标相对应的像素坐标为种子点,采用区域生长法进行图像分割。至此,即可检测到障碍物的位置和轮廓信息。

For a single sensor,only local environment feature information can be obtained,in order to further improve the accuracy of obstacle detection,an obstacle detection method combining millimeter wave radar and vision sensor is proposed.Obtain the number of obstacles and coordinate points through the millimeter wave radar,and preprocess the scene image collected by the camera to obtain the image to be segmented.Then the pixel coordinates corresponding to the coordinates obtained by millimeter waveradar are taken as seed points and the image is segmented by region growing method.So far,the position and contour information of the obstacle can be detected.