Abstract:Detection and tracking using video synthetic aperture radar (ViSAR) have attracted a great deal of attention in recent years due to their ability to produce high-resolution videos for regions of interest. In this work, we have chosen the background-aware correlation filter (BACF) as a key algorithm due to its superior performance in real-time tracking scenarios. Dim and small shadows with weak visual features, time-varying shadows, and complex environments pose serious challenges to the target tracking problem using ViSAR videos. These factors lead to a multipeak problem in the BACF and many other correlation filter-based algorithms. As a result, incorrect target detection, template model contamination, and tracker drift or failure during long-term tracking can occur. In this work, we propose a novel appearance-distance information-assisted probabilistic data association (AD-PDA) algorithm to tackle the multipeak problem. Based on the correlation outputs of the BACF, the AD-PDA algorithm selects multiple peak locations as validated measurements. By exploiting the appearance-distance probability distribution functions, the AD-PDA algorithm calculates the mixed appearance-distance association weights and estimates target states accurately. Furthermore, we propose an efficient AD-PDA-BACF algorithm that can track targets accurately by combining the AD-PDA and BACF algorithms. This study conducts experiments using two public ViSAR video datasets released by the Sandia National Laboratories, Albuquerque, NM, USA. Our results demonstrate that the proposed algorithm outperforms several state-of-the-art algorithms in tracking dim and small targets in terms of tracking accuracy, success rate, and tracking speed.

An Enhanced V-BM3D Algorithm for VideoSAR Denoising Combined with Temporal Information

A TLRTV Dual-band SAR Image Denoise-Fusion Strategy and Its Preliminary Experiment Analysis in Multi-band Airborne Radar System

Shadow-based Moving Target Detection for Video SAR

Shadow-Background-Noise 3D Spatial Decomposition Using Sparse Low-Rank Gaussian Properties for Video-SAR Moving Target Shadow Enhancement

A Robust Moving Target Shadow Detection and Tracking Method for VideoSAR

Moving Target Shadow Detection Method Based on Improved ViBe in VideoSAR Images

Implementation and Evaluation of BM3D Video Denoising Algorithms

A New Method of Video SAR Ground Moving Target Detection and Tracking Based on the Interframe Amplitude Temporal Curve

Time-Varying Sar Interference Suppression Based On Delay-Doppler Iterative Decomposition Algorithm

Vision-Based Scattering Key-Frame Extraction for VideoSAR Summarization

Imaging and Detection of Moving Targetsbased on Spaceborne Video SAR Mode

Video SAR Moving Target Shadow Detection Based on Intensity Information and Neighborhood Similarity

A Trilateral Filter for Video SAR Speckle Noise Reduction

Non-Local SAR Image Despeckling Based on Sparse Representation

Enhancing RABASAR for Multi-Temporal SAR Image Despeckling through Directional Filtering and Wavelet Transform

Moving Target Detection Based on Circular Video SAR

Robust Shadow Tracking for Video SAR

Video SAR Moving Target Indication Using Deep Neural Network

Detecting Moving Target on Ground Based on Its Shadow by Using VideoSAR

A Novel AD-PDA-BACF Algorithm for Real-Time Moving Target Shadow Tracking Using ViSAR Imagery

Tracking of Moving Target Based on SiamMask for Video SAR System