Research Article
Research on Image Stitching for Parking Assistance System
@INPROCEEDINGS{10.1007/978-3-031-65123-6_22, author={Sheng Liu and Yiqing Yang and Ting Cao}, title={Research on Image Stitching for Parking Assistance System}, proceedings={Quality, Reliability, Security and Robustness in Heterogeneous Systems. 19th EAI International Conference, QShine 2023, Shenzhen, China, October 8 -- 9, 2023, Proceedings, Part II}, proceedings_a={QSHINE PART 2}, year={2024}, month={8}, keywords={Parking Assistance System Image Stitching SIFT}, doi={10.1007/978-3-031-65123-6_22} }- Sheng Liu
Yiqing Yang
Ting Cao
Year: 2024
Research on Image Stitching for Parking Assistance System
QSHINE PART 2
Springer
DOI: 10.1007/978-3-031-65123-6_22
Abstract
Currently, car parking assistance is limited to the small angle rear view imaging of the reversing radar, which can only provide the driver with a limited range of vision and is prone to safety hazards. It is necessary to stitch the surrounding images captured by the parking monitor by widening the monitoring field of view of the car parking and providing the driver with a complete rear panoramic image through the parking assistance system. In this research, we provide an enhanced SIFT image stitching technique that uses random k-d trees and the k-nearest neighbour matching to optimise SIFT feature matching to improve matching speed, the RANSAC algorithm to reject mismatched pairs to improve matching accuracy, and pre-processing to reduce distortion and eliminate stitching gaps by applying columnar projection and weighted fusion to the stitched images. The experimental findings demonstrate that the suggested approach may increase image stitching speed while also producing stitched and displayed images that are more accurate.
