Four-wheel positioning homography matrix optimization algorithm based on minimum re-projection error

Abstract

A fast and accurate homography matrix method for four-wheel positioning detection was presented in the paper. Fewer sensors were required with simpler operation and faster detection. Firstly, eight feature points were extracted by using the target detection algorithm based on the fitting method. Secondly, six feature points were obtained by line fitting-based selection. Thirdly, from the selected six feature points, five points were randomly chosen to minimize the re-projection error. Finally, four points were randomly selected from these five feature points to find the homography matrix, and the other point was back to the homography matrix for verification. The experimental results show that the mean re-projection error is reduced by about 3.41%-4.57% compared with the modified RANSAC (Random sample consensus) algorithm. With the optimized algorithm, the error is reduced by about 12.81%-13.86% compared with the improved RANSAC algorithm. Compared with traditional targets, the average calibration time is reduced by about 26.95%-27.88%. The results indicated that the combination of target algorithm and optimization algorithm could ensure the accuracy and reliability of four-wheel positioning.

Keywords

computer vision / four-wheel alignment / line fitting / re-projection error / homography matrix

References

[1] SUN F. Research on uncertainty and calibration technology of automobile four-wheel positioning parameters. Zhenjiang: Jiangsu University, 2019.

[2] SHAO C H, ZHANG Z Y, XU S S, et al. Calibration point distribution study of a four-wheel alignment optimization device based on a blanket technology algorithm. Review of Scientific Instruments, 2020, 91(4): 044102.

[3] YANG W W, HE J L, LIU Q P, et al. Study on testing mechanism and experimental research of four-wheel alignment verification device. Metrology Science and Technology, 2022, 66(6): 54-59.

[4] PAN M Y, CHEN S W, LÜ X Y, et al. Error analysis on parameters measurement model of wheel-mounted four-wheel alignment//2020 11th International Conference on Prognostics and System Health Management, October 23-25, 2020, Ji’nan, China. New York: IEEE, 2020: 592-598.

[5] CHEN H, WANG D Q, CHEN Y Q. Research on the influence of calibration image on reprojection error//2021 International Conference on Big Data Engineering and Education, August 12-14, 2021, Guiyang, China. New York: IEEE, 2021: 60-66.

[6] ZHANG Z, ZHAO R J, LIU E H, et al. A single-image linear calibration method for camera. Measurement, 2018, 130: 298-305.

[7] XU G, HE W, CHEN F, et al. One-dimension orientation method of caster and kingpin inclination of vehicle wheel alignment. Measurement, 2022, 198: 111371.

[8] LI D D, CHEN G Z, LI C J, et al. 3D feature points calibration method for depth-camera//2021 5th Asian Conference on Artificial Intelligence Technology, October 29-31, 2021, Haikou, China. New York: IEEE, 2021: 735-741.

[9] PEREZ A J, PEREZ-CORTES J C, GUARDIOLA J L. Simple and precise multi-view camera calibration for 3D reconstruction. Computers in Industry, 2020, 123: 103256.

[10] CHENG W, ZHU Z F, YAO Y, et al. An improved RANSAC algorithm for 3D wheel alignment. Journal of Measurement Science and Instrumentation, 2022, 13(4): 407-417.

[11] KANG C H, HONG L, REN J W, et al. High-precision camera calibration method based on sub-pixel edge detection and circularity correction compensation. Laser & Optoelectronics Progress, 2024,61(8): 146-154.

[12] REN Y Z, HU F. Camera calibration with pose guidance//ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing, June 6-11, 2021, Toronto, ON, Canada. New York: IEEE, 2021: 2180-2184.

[13] YAO Y N, HUANG X Y, LÜ J L. A Space Joint calibration method for lidar and camera on self-driving car and its experimental verification//2021 6th International Symposium on Computer and Information Processing Technology, June 11-13, 2021, Changsha, China. New York: IEEE, 2021: 388-394.

[14] SUN W, HUO J, LI Y, et al. Combined calibration method for large field of view multi-camera system//2022 China Automation Congress, November 25-27, 2022, Xiamen, China. New York: IEEE, 2022: 4165-4170.

[15] HUANG C W, PAN X, CHENG J C, et al. Deep image registration with depth-aware homography estimation. IEEE Signal Processing Letters, 2023, 30: 6-10.

[16] JIANG P P. Research on fast calibration method of vehicle multi-camera vision system. Chongqing: Chongqing University of Posts and Telecommunications, 2020.

[17] XIN R, WU S H, LI A J. High-precision camera calibration method based on circular mode plane target. Journal of Yantai University: Natural Science and Engineering, 2020, 33(4): 472-478.

[18] SU F, WANG Z J. Error analysis and correction of a photoelastic method based on a pixelated polarization camera. Optics and Lasers in Engineering, 2023, 161: 107374.

[19] LI Z X, WANG K Q, JIA W Y, et al. Multiview stereo and silhouette fusion via minimizing generalized reprojection error. Image and Vision Computing, 2015, 33: 1-14.

[20] CHEN T, GUO J F, XIE X L, et al. High-precision image mosaic algorithm based on adaptive homography transform//2021 40th Chinese Control Conference, July 26-28, 2021, Shanghai, China. New York: IEEE, 2021: 3030-3035.

[21] ZENG D H. Research on image stitching technology based on partition detection and improved RANSAC algorithm. Chongqing: Southwest University, 2021.

[22] LIAN Z W, LIU Y, ZHANG Q. Research on algorithm of center positioning for automatic refractometer target ring. Journal of North University of China (Natural Science Edition), 2022, 43(6): 548-553.