String shape recognition using enhanced matching method from 3D point cloud data

Abstract

The deformable object such as string, cloth and paper, is soft and can change its shape easily. It is difficult for a robot to manipulate deformable object because it needs to deal with various shape. For the operation of deformable object, it is important to recognize the shape of them. In our previous research, point chain model and the string shape recognition method were proposed. Point chain model is a structure to describe a string shape by a series of connections of points. And the string shape recognition method are algorithms to recognize a string shape from 3D point cloud data and output point chain model. However, the previous method sometimes occurred misrecognition of segments. Therefore, enhanced matching method is proposed to improve recognition performance.