Motion multi-object matching and position estimation based on unsynchronized image sequences
Abstract This paper proposes a novel method for motion multi-object matching and position estimation in the absence of salient features based on unsynchronized image sequences. Our proposed method aims to address the issues of traditional feature matching that requires static objects, salient featur...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-03-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-025-92237-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850029857417723904 |
|---|---|
| author | Kai Guo Rui Cao Chenyang Yue Faxin Li Xin Zhou Binbin Wang |
| author_facet | Kai Guo Rui Cao Chenyang Yue Faxin Li Xin Zhou Binbin Wang |
| author_sort | Kai Guo |
| collection | DOAJ |
| description | Abstract This paper proposes a novel method for motion multi-object matching and position estimation in the absence of salient features based on unsynchronized image sequences. Our proposed method aims to address the issues of traditional feature matching that requires static objects, salient features and the need for synchronized images when the epipolar constraint is used. Firstly, unsynchronized image sequences are captured using three calibrated cameras, and for each motion object, three spatial planes are established using multi-images. Each pair of spatial planes determines the candidate trajectory of the very object and calculates the candidate position at a specific height. Subsequently, a candidate position matrix for multiple objects between the first camera and the second camera is obtained, as well as another candidate position matrix between the first camera and the third camera. Then, based on the principle of minimum distances for motion multi-object matching, a flexible search method between the two candidate position matrices is established to calculate the distances and achieve multi-object matching at the minimum distances. According to the matching results, a new method for position estimation based on line-plane constraint is established. Finally, synthetic data and real images are used to experimentally test the performance of our proposed method, and it is compared with the matching algorithm under synchronized images based on epipolar constraint. The experimental results show that our proposed method has better performance in noise sensitivity but is slower in computation speed. |
| format | Article |
| id | doaj-art-33a4ef0167254d8eb1b9152e69aefe32 |
| institution | DOAJ |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-33a4ef0167254d8eb1b9152e69aefe322025-08-20T02:59:24ZengNature PortfolioScientific Reports2045-23222025-03-0115112010.1038/s41598-025-92237-9Motion multi-object matching and position estimation based on unsynchronized image sequencesKai Guo0Rui Cao1Chenyang Yue2Faxin Li3Xin Zhou4Binbin Wang5Northwest Institute of Nuclear TechnologyNorthwest Institute of Nuclear TechnologyNorthwest Institute of Nuclear TechnologyNorthwest Institute of Nuclear TechnologyNorthwest Institute of Nuclear TechnologyNorthwest Institute of Nuclear TechnologyAbstract This paper proposes a novel method for motion multi-object matching and position estimation in the absence of salient features based on unsynchronized image sequences. Our proposed method aims to address the issues of traditional feature matching that requires static objects, salient features and the need for synchronized images when the epipolar constraint is used. Firstly, unsynchronized image sequences are captured using three calibrated cameras, and for each motion object, three spatial planes are established using multi-images. Each pair of spatial planes determines the candidate trajectory of the very object and calculates the candidate position at a specific height. Subsequently, a candidate position matrix for multiple objects between the first camera and the second camera is obtained, as well as another candidate position matrix between the first camera and the third camera. Then, based on the principle of minimum distances for motion multi-object matching, a flexible search method between the two candidate position matrices is established to calculate the distances and achieve multi-object matching at the minimum distances. According to the matching results, a new method for position estimation based on line-plane constraint is established. Finally, synthetic data and real images are used to experimentally test the performance of our proposed method, and it is compared with the matching algorithm under synchronized images based on epipolar constraint. The experimental results show that our proposed method has better performance in noise sensitivity but is slower in computation speed.https://doi.org/10.1038/s41598-025-92237-9 |
| spellingShingle | Kai Guo Rui Cao Chenyang Yue Faxin Li Xin Zhou Binbin Wang Motion multi-object matching and position estimation based on unsynchronized image sequences Scientific Reports |
| title | Motion multi-object matching and position estimation based on unsynchronized image sequences |
| title_full | Motion multi-object matching and position estimation based on unsynchronized image sequences |
| title_fullStr | Motion multi-object matching and position estimation based on unsynchronized image sequences |
| title_full_unstemmed | Motion multi-object matching and position estimation based on unsynchronized image sequences |
| title_short | Motion multi-object matching and position estimation based on unsynchronized image sequences |
| title_sort | motion multi object matching and position estimation based on unsynchronized image sequences |
| url | https://doi.org/10.1038/s41598-025-92237-9 |
| work_keys_str_mv | AT kaiguo motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences AT ruicao motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences AT chenyangyue motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences AT faxinli motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences AT xinzhou motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences AT binbinwang motionmultiobjectmatchingandpositionestimationbasedonunsynchronizedimagesequences |