A robust method for multi object tracking in autonomous ship navigation systems
Надёжный метод многoобъектового сопровождения в системах автономной навигации судов
2024-07-31
SCID: 54.1/82yrcjvv
Discuss with AI
Jari Maritime Tracking DatasetYOLOv7autonomous ship navigationcamera motion compensationmaritime multi-object tracking
Figures from the paper
Abstract (AI)
A novel Maritime Multi-Object Tracking method is proposed, combining a deep learning-based object detector with target association algorithms to achieve robust sea-surface multi-object tracking. Specifically, the Multi-Object Tracking employs You Only Look Once version 7 detector for object detection. In the data association part, a module for onboard camera motion compensation is developed, a maritime dynamic spatial information-based intersection-over-union is presented as a similarity metric, and a progressive refinement cascade matching strategy is designed to enhance the tracker's sea-surface multi-object tracking capabilities. The Jari Maritime Tracking Dataset is utilised to validate the effectiveness performance of the proposed method. Experimental results demonstrate that compared to the earlier process, the proposed method exhibits a significant enhancement in multiple object tracking accuracy, with an increase of 27.8% and an achieved score of 81.3. In particular, it reduces the number of identifications switching and missed targets, achieving holistically preferable performance. Meanwhile, the speed of the proposed Multi-Object Tracking fulfils the engineering application requirements for an autonomous ship navigation system.
Key Findings
1
A progressive refinement cascade matching strategy improves tracking performance in maritime multi-object scenarios.
2
A robust maritime multi-object tracking method combines YOLOv7 detection with camera-motion compensation and target-association algorithms.
3
On the Jari Maritime Tracking Dataset, tracking accuracy increased by 27.8%, reaching a score of 81.3 compared with the earlier approach.
4
The method introduces maritime dynamic spatial information-based intersection-over-union as a similarity metric for sea-surface target association.
5
The proposed tracker reduces identity switches and missed targets while maintaining speed sufficient for autonomous ship navigation engineering requirements.
Research Object
sea-surface maritime objects observed by autonomous ship navigation systems
Research Subject
robust multi-object tracking accuracy, identity-switch and missed-target reduction, and real-time tracking performance under onboard camera motion
Publication Details
Publication Date
2024-07-31
Journal
Publisher
ISSN
Cited by
35
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest