Super-Resolution Reconstruction of LIDAR Point Cloud Image on Mobile Platform
Download as PDF
DOI: 10.23977/meimie.2019.43081
Author(s)
Hanwen Liu, Zili Huang, Zhengzhong Huang
Corresponding Author
Hanwen Liu
ABSTRACT
Target recognition technology based on 3D LIDAR scanning technology provides important significance for vehicle and missile-borne systems. In view of the low resolution of existing 3D LIDAR the combination of LIDAR scanning technology and image processing technology is proposed. ICP algorithm is mainly used to reconstruct dense point cloud images through sparse laser point cloud image sequence, and to improve low resolution three-dimensional point cloud images. Super-resolution reconstruction is beneficial to target detection and recognition and other subsequent processing. Firstly, we focus on the application of ICP algorithm in three-dimensional laser point cloud registration, and use ICP algorithm to reconstruct three-dimensional point cloud image with super-resolution. Secondly, a simplified model is proposed to quantitatively evaluate the performance of point cloud reconstruction algorithm.
KEYWORDS
Super-resolution reconstruction, target recognition, 3-Dpointcloudregistration, ICPalgorithm