A Method for Localizing and Grasping Objects in a Picking Robot System Using Kinect Camera

This paper proposes a method for localizing and grasping objects in a picking robot system using Kinect stereo camera sensor. To do this task, the followings are done. Firstly, an image processing system including Kinect camera sensor is described. Secondly, RGB color map and new depth map for image inpainting are obtained using Kinect SDK mapping function to align RGB image with depth image. Thirdly, a new rectangle representation algorithm based on the Kinect camera is proposed to localize the grasping point of the object. Finally, the effectiveness and the applicability of the proposed algorithms is verified by using experiment. The experimental results show that the proposed algorithm successfully detects an object and finds its grasping points with accuracy 100% for no transparent objects and with 95% for transparent objects. The proposed algorithm also has higher accuracy to successfully find the grasping points than the 2D planar method.