A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation
Abstract Vehicle detection and tracking from a UAV perspective often encounters omission and misdetection due to the small targets, complex scenes and target occlusion, which finally influences hugely on detection accuracy and target tracking stability. Additionally, the number of parameters of curr...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01687-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571174694944768 |
|---|---|
| author | Quan Wang Guangfei Ye Qidong Chen Songyang Zhang Fengqing Wang |
| author_facet | Quan Wang Guangfei Ye Qidong Chen Songyang Zhang Fengqing Wang |
| author_sort | Quan Wang |
| collection | DOAJ |
| description | Abstract Vehicle detection and tracking from a UAV perspective often encounters omission and misdetection due to the small targets, complex scenes and target occlusion, which finally influences hugely on detection accuracy and target tracking stability. Additionally, the number of parameters of current model is large that makes it is hard to be deployed on mobile devices. Therefore, this paper proposes a YOLO-LMP and NGCTrack-based target detection and tracking algorithm to address these issues. Firstly, the performance of detecting small targets in occluded scenes is enhanced by adding a MODConv to the small-target detection head and increasing its size; In addition, excessive deletion of prediction boxes is prevented by utilizing LSKAttention mechanism to adaptively adjust the target sensing field at the downsampling stage and combining it with the Soft-NMS strategy. Furthermore, the C2f module is replaced by the FPW to reduce the pointless computation and memory utilization of the model. At the target tracking stage, the so-called NGCTrack in our algorithm replaces IOU with GIOU and employs a modified NSA Kalman filter to adjust the state-space aspect ratio for width prediction. Finally, the camera adjustment mechanism was introduced to improve the precision and consistency of tracking. The experimental results show that, compared to YOLOv8, the YOLO-LMP model improves map50 and map50:95 metrics by 10.3 and 12.2%, respectively and the number of parameters is decreased by 47.7%. After combined it with the improved NGCTrack, the number of IDSW reduced by 73.6% compared to the ByteTrack method, while the MOTA and IDF1 increase by 5.2 and 9.8%, respectively. |
| format | Article |
| id | doaj-art-778166230b1e4633b535dd338cd1496b |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-778166230b1e4633b535dd338cd1496b2025-02-02T12:49:52ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-12-0111111810.1007/s40747-024-01687-7A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportationQuan Wang0Guangfei Ye1Qidong Chen2Songyang Zhang3Fengqing Wang4College of Internet of Things Engineering, Wuxi UniversityCollege of Computer Science, Nanjing University of Information EngineeringCollege of Internet of Things Engineering, Wuxi UniversityCollege of Electronic and Information Engineering, Suzhou University of Science and TechnologyCollege of Internet of Things Engineering, Wuxi UniversityAbstract Vehicle detection and tracking from a UAV perspective often encounters omission and misdetection due to the small targets, complex scenes and target occlusion, which finally influences hugely on detection accuracy and target tracking stability. Additionally, the number of parameters of current model is large that makes it is hard to be deployed on mobile devices. Therefore, this paper proposes a YOLO-LMP and NGCTrack-based target detection and tracking algorithm to address these issues. Firstly, the performance of detecting small targets in occluded scenes is enhanced by adding a MODConv to the small-target detection head and increasing its size; In addition, excessive deletion of prediction boxes is prevented by utilizing LSKAttention mechanism to adaptively adjust the target sensing field at the downsampling stage and combining it with the Soft-NMS strategy. Furthermore, the C2f module is replaced by the FPW to reduce the pointless computation and memory utilization of the model. At the target tracking stage, the so-called NGCTrack in our algorithm replaces IOU with GIOU and employs a modified NSA Kalman filter to adjust the state-space aspect ratio for width prediction. Finally, the camera adjustment mechanism was introduced to improve the precision and consistency of tracking. The experimental results show that, compared to YOLOv8, the YOLO-LMP model improves map50 and map50:95 metrics by 10.3 and 12.2%, respectively and the number of parameters is decreased by 47.7%. After combined it with the improved NGCTrack, the number of IDSW reduced by 73.6% compared to the ByteTrack method, while the MOTA and IDF1 increase by 5.2 and 9.8%, respectively.https://doi.org/10.1007/s40747-024-01687-7Target detectionMulti-target trackingUrban transportation vehiclesLightweightUAV detection and tracking |
| spellingShingle | Quan Wang Guangfei Ye Qidong Chen Songyang Zhang Fengqing Wang A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation Complex & Intelligent Systems Target detection Multi-target tracking Urban transportation vehicles Lightweight UAV detection and tracking |
| title | A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| title_full | A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| title_fullStr | A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| title_full_unstemmed | A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| title_short | A UAV perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| title_sort | uav perspective based lightweight target detection and tracking algorithm for intelligent transportation |
| topic | Target detection Multi-target tracking Urban transportation vehicles Lightweight UAV detection and tracking |
| url | https://doi.org/10.1007/s40747-024-01687-7 |
| work_keys_str_mv | AT quanwang auavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT guangfeiye auavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT qidongchen auavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT songyangzhang auavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT fengqingwang auavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT quanwang uavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT guangfeiye uavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT qidongchen uavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT songyangzhang uavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation AT fengqingwang uavperspectivebasedlightweighttargetdetectionandtrackingalgorithmforintelligenttransportation |