Tracking-by-Detection Algorithm for Underwater Target Based on Improved Multi-Kernel Correlation Filter

Wenrong Yue,Feng Xu,Juan Yang

DOI: https://doi.org/10.3390/rs16020323

IF: 5

2024-01-13

Remote Sensing

Abstract:Joint detection and tracking of weak underwater targets are challenging problems whose complexity is intensified when the target is disturbed by reverberation. In the low signal-to-reverberation ratio (SRR) environment, the traditional detection and tracking methods perform poorly in tracking robustness because they only consider the target motion characteristics. Recently, the kernel correlation filter (KCF) based on target features has received lots of attention and gained great success in visual tracking. We propose an improved multi-kernel correlation filter (IMKCF) tracking-by-detection algorithm by introducing the KCF into the field of underwater weak target detection and tracking. It is composed of the tracking-by-detection, the adaptive reliability check, and the re-detection modules. Specifically, the tracking-by-detection part is built on the multi-kernel correlation filter (MKCF), and it uses multi-frame data weighted averaging to update. The reliability check helps keep the tracker from corruption. The re-detection module, integrated with a Kalman filter, identifies target positions when the tracking is unreliable. Finally, the experimental data processing and analysis show that the proposed method outperforms the single-kernel methods and some traditional tracking methods.

environmental sciences,imaging science & photographic technology,remote sensing,geosciences, multidisciplinary

What problem does this paper attempt to address?

This paper mainly discusses how to solve the problem of weak target detection and tracking in underwater low signal-to-noise ratio (SRR) environment. Traditional detection and tracking methods perform poorly under high false alarm rate, and existing methods including random finite set (RFS) based methods and deep learning methods have their limitations. The paper proposes an improved Multi-Kernel Correlation Filter (IMKCF) tracking detection algorithm, which introduces the KCF algorithm into the field of underwater weak target detection and tracking.

The IMKCF algorithm consists of three parts: detection-based tracking, adaptive reliability check, and re-detection module. Among them, the detection-based tracking part adopts Multi-Kernel Correlation Filter (MKCF) and updates by weighted averaging of multiple frame data. The adaptive reliability check module is used to evaluate the reliability of tracking results to prevent error propagation. The re-detection module combines the Kalman filter to reposition the target location when tracking is unreliable.

Experimental data processing and analysis show that the proposed IMKCF algorithm is superior to single-kernel methods and some traditional tracking methods, improving the robustness and accuracy of long-term tracking processes. The main contribution of the paper lies in the application of KCF to the detection and tracking of weak targets with multiple motion types underwater, adapting to multiple nonlinear kernel parameters through the weighted information of historical samples, enhancing the utilization of multiple complementary features, and realizing real-time evaluation of the reliability of target tracking results and proposing an effective re-detection strategy.

In conclusion, this research provides a new solution for underwater weak target detection and tracking, especially for improved performance in complex environments.