基于Kinect的实时障碍物检测

Real-time obstacle detection based on Kinect

传统的传感器在移动机器人障碍物检测领域都有其各自的局限性。文章提出基于Kinect的障碍物检测方法:利用Kinect传感器获取环境深度图像;通过Kinect标定配准之后获取校准参数;通过该参数获得图像像素点与空间三维坐标的对应关系;通过空间三维坐标确定地平面与障碍物区域,并将障碍物区域作为感兴趣区域;通过三维坐标在x轴和z轴的连续性对感兴趣区域进行处理,分割出各个障碍物。实验结果表明,文中算法可以有效且实时地检测到障碍物信息。

Traditional sensors have their own limitations in the field of obstacle detection for mobile robots. In this paper,an obstacle detection method based on Kinect is proposed. The Kinect sensor is used to obtain the depth image of the environment; The calibration parameters are obtained after Kinect calibration and registration; By these parameters,the correspondence between the image pixel points and the spatial threedimensional coordinates is obtained; The ground plane and the obstacle area are determined by the spatial three-dimensional coordinates,and the obstacle area is regarded as the region of interest; The region of interest is processed by the continuity of the x and z axes in three-dimensional coordinates,and each obstacle is segmented. Finally,the experimental results show that the proposed algorithm can detect the obstacle information efficiently and in real time.