An efficient 3D point clouds covariance descriptor for object classification with mismatching correction algorithm. (3rd September 2019)

• Record Type: Journal Article
• Authors: Zhang, Heng; Zhuang, Bin
• Is Part Of: International journal of high performance computing and networking. Volume 14:Number 3(2019)
• Page Start: 356
• Page End: 364
• Publication Date: 2019-09-03
• Subjects: object classification -- point clouds -- covariance descriptor -- mismatching correction; High performance computing -- Periodicals; Computer networks -- Periodicals; High performance computing; Periodicals; 004.05
• Languages: English
• ISSNs: 1740-0562

Abstract

We introduce a new covariance descriptor combining object visual (colour, gradient, depth, etc.) and geometric information (3D coordinates, normal vectors, Gaussian curvature, etc.) for mobile robot with RGB-D camera to deal with point cloud data. The improved mismatching correction algorithm is applied in the feature point mismatching correction of 3D point clouds. This descriptor is able to quickly match the feature points of the point clouds in the surrounding environment and realise the function of object classification. Experimental results show that this descriptor has the advantages of compactness and flexibility compared with the previous descriptor, and greatly reduces the storage space required. At the same time, the instance and category recognition accuracy of the proposed descriptor for objects can respectively reach 94.6% and 86.8%, which are higher than those of the previous methods for object recognition of 3D point clouds.