Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching
Registering light detection and ranging (LiDAR) data with optical camera images enhances spatial awareness in autonomous driving, robotics, and geographic information systems. The current challenges in this field involve aligning 2D-3D data acquired from sources with distinct coordinate systems, ori...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/3/357 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850199787308056576 |
|---|---|
| author | Jennifer Leahy Shabnam Jabari Derek Lichti Abbas Salehitangrizi |
| author_facet | Jennifer Leahy Shabnam Jabari Derek Lichti Abbas Salehitangrizi |
| author_sort | Jennifer Leahy |
| collection | DOAJ |
| description | Registering light detection and ranging (LiDAR) data with optical camera images enhances spatial awareness in autonomous driving, robotics, and geographic information systems. The current challenges in this field involve aligning 2D-3D data acquired from sources with distinct coordinate systems, orientations, and resolutions. This paper introduces a new pipeline for camera–LiDAR post-registration to produce colorized point clouds. Utilizing deep learning-based matching between 2D spherical projection LiDAR feature layers and camera images, we can map 3D LiDAR coordinates to image grey values. Various LiDAR feature layers, including intensity, bearing angle, depth, and different weighted combinations, are used to find correspondence with camera images utilizing state-of-the-art deep learning matching algorithms, i.e., SuperGlue and LoFTR. Registration is achieved using collinearity equations and RANSAC to remove false matches. The pipeline’s accuracy is tested using survey-grade terrestrial datasets from the TX5 scanner, as well as datasets from a custom-made, low-cost mobile mapping system (MMS) named Simultaneous Localization And Mapping Multi-sensor roBOT (SLAMM-BOT) across diverse scenes, in which both outperformed their baseline solutions. SuperGlue performed best in high-feature scenes, whereas LoFTR performed best in low-feature or sparse data scenes. The LiDAR intensity layer had the strongest matches, but combining feature layers improved matching and reduced errors. |
| format | Article |
| id | doaj-art-d19b22c2ec784785b6887cebf44ae579 |
| institution | OA Journals |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-d19b22c2ec784785b6887cebf44ae5792025-08-20T02:12:33ZengMDPI AGRemote Sensing2072-42922025-01-0117335710.3390/rs17030357Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based MatchingJennifer Leahy0Shabnam Jabari1Derek Lichti2Abbas Salehitangrizi3Advanced Spatial Intelligence Lab, Department of Geodesy & Geomatics Engineering, University of New Brunswick, 3 Bailey Dr., Fredericton, NB E3B 5A3, CanadaAdvanced Spatial Intelligence Lab, Department of Geodesy & Geomatics Engineering, University of New Brunswick, 3 Bailey Dr., Fredericton, NB E3B 5A3, CanadaDepartment of Geomatics Engineering, University of Calgary, 2500 University Dr. NW, Calgary, AB T2N 1N4, CanadaAdvanced Spatial Intelligence Lab, Department of Geodesy & Geomatics Engineering, University of New Brunswick, 3 Bailey Dr., Fredericton, NB E3B 5A3, CanadaRegistering light detection and ranging (LiDAR) data with optical camera images enhances spatial awareness in autonomous driving, robotics, and geographic information systems. The current challenges in this field involve aligning 2D-3D data acquired from sources with distinct coordinate systems, orientations, and resolutions. This paper introduces a new pipeline for camera–LiDAR post-registration to produce colorized point clouds. Utilizing deep learning-based matching between 2D spherical projection LiDAR feature layers and camera images, we can map 3D LiDAR coordinates to image grey values. Various LiDAR feature layers, including intensity, bearing angle, depth, and different weighted combinations, are used to find correspondence with camera images utilizing state-of-the-art deep learning matching algorithms, i.e., SuperGlue and LoFTR. Registration is achieved using collinearity equations and RANSAC to remove false matches. The pipeline’s accuracy is tested using survey-grade terrestrial datasets from the TX5 scanner, as well as datasets from a custom-made, low-cost mobile mapping system (MMS) named Simultaneous Localization And Mapping Multi-sensor roBOT (SLAMM-BOT) across diverse scenes, in which both outperformed their baseline solutions. SuperGlue performed best in high-feature scenes, whereas LoFTR performed best in low-feature or sparse data scenes. The LiDAR intensity layer had the strongest matches, but combining feature layers improved matching and reduced errors.https://www.mdpi.com/2072-4292/17/3/357camera–LiDAR registrationLiDAR feature layersSuperGlueLoFTRmatchingcollinearity equations |
| spellingShingle | Jennifer Leahy Shabnam Jabari Derek Lichti Abbas Salehitangrizi Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching Remote Sensing camera–LiDAR registration LiDAR feature layers SuperGlue LoFTR matching collinearity equations |
| title | Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching |
| title_full | Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching |
| title_fullStr | Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching |
| title_full_unstemmed | Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching |
| title_short | Enhancing Cross-Modal Camera Image and LiDAR Data Registration Using Feature-Based Matching |
| title_sort | enhancing cross modal camera image and lidar data registration using feature based matching |
| topic | camera–LiDAR registration LiDAR feature layers SuperGlue LoFTR matching collinearity equations |
| url | https://www.mdpi.com/2072-4292/17/3/357 |
| work_keys_str_mv | AT jenniferleahy enhancingcrossmodalcameraimageandlidardataregistrationusingfeaturebasedmatching AT shabnamjabari enhancingcrossmodalcameraimageandlidardataregistrationusingfeaturebasedmatching AT dereklichti enhancingcrossmodalcameraimageandlidardataregistrationusingfeaturebasedmatching AT abbassalehitangrizi enhancingcrossmodalcameraimageandlidardataregistrationusingfeaturebasedmatching |