Article Details

Title · Authors · Abstract · Keywords · DOI · References

Research on High-precision Pavement Image Stitching Method of Multi-camera Laser System

LU Haizhu1, ZONG Yan1, DONG Jiaqi2, YUAN Zebin2, DAI Ao2, LUO Wenting2, LI Lin2

(1.Jiangsu Expressway Engineering Maintenance Co., Ltd., Nanjing 211100, China; 2.College of Transportation Engineering, Nanjing Tech University, Nanjing 211816, China)

Section：Computer Technology · Vol. 2026, Issue 1 · Pages：101-104 · 2096-4706（2026）01-0101-04 · CLC number：TP391.4；U491.1 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2026.01.020

Abstract

This paper proposes a road detection system fusing multi-camera and laser imaging, aiming to achieve high-efficiency and high-precision three-dimensional pavement data acquisition and two-dimensional image stitching. The system adopts a four-camera and dual-laser design, utilizes incremental encoders to realize synchronous image acquisition, and combines SIFT feature extraction, FLANN matching, and RANSAC homography transformation to complete image alignment. Masking and cropping techniques are adopted to optimize stitching boundaries, effectively eliminating overlap redundancy and geometric distortion. Experiments prove that this method performs excellently in dislocation control and overlap matching, and is suitable for multi-scenario high-precision pavement detection.

Keywords

multi-camera system; laser imaging; image stitching; feature matching; pavement detection

References

  1. [1] Li Xiying, Ye Zhihui, Wei Shikui, etc Overview of Image based 3D Object Detection in Autonomous Driving: Benchmarking, Constraints, and Error Analysis [J]. Chinese Journal of Image and Graphics, 2023, 28 (6): 1709-1740
  2. [2] ZHANG L，YANG F，ZHANG Y D，et al. Road Crack Detection Using Deep Convolutional Neural Network [C]//2016 IEEE International Conference on Image Processing (ICIP).Phoenix：IEEE，2016，69：103-111.
  3. [3] Jin Shuai, Li Xuanpeng, Yang Feng, etc Pseudo laser point cloud enhanced 3D object detection in road scenes [J]. Chinese Journal of Image and Graphics, 2023, 28 (11): 3520-3535
  4. [4] Liu Xiancui Research on Forward Simulation and Application of 3D Ground Penetrating Radar Based on FDTD [D]. Shanghai: Shanghai University of Applied Sciences, 2022
  5. [5] LOWE D G. Distinctive Image Features from Scale-Invariant Keypoints [J].International Journal of Computer Vision，2004，60：91-110.
  6. [6] Wang Ning Research on Algorithm and Maintenance Strategy for Road Surface Defect Detection [D]. Jinan: Shandong Jiaotong University, 2024
  7. [7] Prince Yang Research on 3D Object Detection and Tracking Algorithm Based on Binocular Vision [D]. Guangzhou: South China University of Technology, 2020
  8. [8] Zhuang Hongwei, Wu Dongmin, Zhang Baoshun Binocular pose estimation based on deep semantic model [J]. Optical Technology, 2023, 49 (2): 205-210
  9. [9] Fang Jianzhong, Xu Sixiang, Yang Yu, etc A binocular vision localization and measurement method based on edge fitting [J]. Optical Technology, 2019, 45 (4): 412-417
  10. [10] Yunnan Aerospace Engineering Geophysical Testing Co., Ltd A 3D detection and imaging method for multi vehicle road foundation: 202011322762.5 [P]. 2022-07-15
  11. [11] MUJA M，LOWE D G. Scalable Nearest Neighbor Algorithms for High Dimensional Data [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2014，36（11）：2227-2240.
  12. [12] Chi Dazhao, Xu Zhixian, Liu Haichun, etc Ultrasonic image stitching method based on improved SIFT algorithm [J]. Journal of Welding, 2024, 45 (10): 1-7
  13. [13] Shangqiu Yudong Highway Survey and Design Co., Ltd A road surface detection system based on 3D digitalization of road surface: 202010259421.1 [P]. 2022-08-05

About the authors

Lu Haizhu (September 1981-), male, Han ethnicity, from Nanjing, Jiangsu, senior engineer, doctoral student, research interests include R&D of expressway maintenance information technology and technical management in related fields.