复杂场景下基于C-SHOT特征的3D物体识别与位姿估计

3D Object Recognition and 6Do F Pose Estimation in Scenes with Occlusions and Clutter Based on C-SHOT 3D Descriptor

为了准确地同时识别复杂点云中的多个目标,提出一种基于法矢改进点云特征C-SHOT的3D物体识别方法.首先,在估计RGB-D数据的点云法矢时将邻域点距离信息考虑在内,计算带距离权重的协方差矩阵得到更精确的点云法矢;其次根据特征点处法矢与邻域法矢的夹角余弦构造点云形状直方图,同时统计点云纹理直方图并与形状直方图融合成C-SHOT描述符;最后对场景与模板分别提取C-SHOT特征,利用Kd树快速求得对应对,引入3D霍夫投票机制,并结合点云局部坐标系克服噪声遮挡问题完成多目标初识别.基于LM-ICP实现精确定位及位姿估计,画出目标包围盒,采用基准数据库CVLab以及采集实验室真实数据进行实验,结果验证了该方法的有效性与精确性.

A3D object detection and recognition method is proposed in this paper.The method achieves pose estimations of multiple object instances in3D scenes with some occlusions and clutter.First,the normal vector of point is estimated by computing the distance between the neighboring points and the feature one within the local spherical support domain.The longer the distance is,the smaller the weight is.Next,we encode the3D descriptor called color signatures of histogram of orientations(C-SHOT)based on improved normal vector.Then we match3D feature correspondences between scenes and models to prove the existence of the objects being sought on3D hough voting space.Finally,we reject wrong feature correspondences and get rough transformation using random sample consensus(RANSAC).Once reliable feature correspondences have been selected,a final transformation matrix based on levenberg marquardt iterative closest point(LM-ICP),can be performed to further refine pose estimations.A thorough experimental evaluations is carried on CVLab3D datasets and real lab3D datasets for object recognition.Experimental results demonstrate the recognition accuracy and robust performance of the proposed method.