Augmented Ship Tracking Under Occlusion Conditions From Maritime Surveillance Videos

Xinqiang Chen,Xueqian Xu,Huafeng Wu,Yongsheng Yang,Jiansen Zhao,Jinjun Tang

doi:10.1109/access.2020.2978054

Abstract

Ship tracking provides crucial on-site microscopic kinematic traffic information which benefits maritime traffic flow analysis, ship safety enhancement, traffic control, etc., and thus has attracted considerable research attentions in the maritime surveillance community. Conventional ship tracking methods yield satisfied results by exploring distinct visual ship features in maritime images, which may fail when the target ship is partially or fully sheltered by obstacles (e.g., ships, waves, etc.) in maritime videos. To overcome the difficulty, we propose an augmented ship tracking framework via the kernelized correlation filter (KCF) and curve fitting algorithm. First, the KCF model is introduced to track ships in the consecutive maritime images and obtain raw ship trajectory dataset. Second, the data anomaly detection and rectification procedure are implemented to rectify the contaminated ship positions. For the purpose of performance evaluation, we implement the proposed framework and another three popular ship tracking models on the four typical ship occlusion videos. The experimental results show that our proposed framework successfully tracks ships in maritime video clips with high accuracy (i.e., the average root mean square error (RMSE), root mean square percentage error (RMSPE), mean absolute deviation (MAD) and mean absolute percentage error (MAPE) are less than 10), which significantly outperforms the other popular ship trackers.

Highlights

IntroductionSmart ship will revolutionize the maritime shipping industry in the decade due to the advantages of reducing ship crew risk at sea (as less crew will be deployed on-board), enhancing maritime traffic efficiency, etc
Smart ship will revolutionize the maritime shipping industry in the decade due to the advantages of reducing ship crew risk at sea, enhancing maritime traffic efficiency, etc
After carefully checking the ship tracking results, we found that both the mean-shift and SAMF models tracked the same obstacle-ship when the target ship was occluded in the maritime image sequences

Summary

Introduction

Smart ship will revolutionize the maritime shipping industry in the decade due to the advantages of reducing ship crew risk at sea (as less crew will be deployed on-board), enhancing maritime traffic efficiency, etc. Visual ship tracking task provides fundamental information for helping smart ship make intelligent sailing decisions, and many studies have been conducted for the purpose of tackling ship tracking challenges via maritime video data. Previous academic studies suggest that visual ship tracking workflow consists of generative and discriminant models [1]. The generative based model works in a similar logic as that of the template. The generative relevant models determine ship positions by considering the region in the image which is quite similar to the input training samples. The generative models may fail to fully exploit background information in maritime images, and the model performance may be severely degraded under heavy background interference situations (caused by neighboring ship occlusion, water waves, etc.)

Methods

Results

Conclusion