SWAN: Synergistic Wavelet-Attention Network for Infrared Small Target Detection

TL;DR

SWAN is a novel infrared small target detection framework that combines wavelet-based frequency analysis and attention mechanisms to improve accuracy in complex backgrounds.

Contribution

It introduces a synergistic wavelet-attention network with novel modules for cross-domain fusion and long-range dependency modeling, advancing IRSTD performance.

Findings

Outperforms existing methods in benchmark tests

Achieves higher detection accuracy in complex scenes

Enhances robustness against background clutter

Abstract

Infrared small target detection (IRSTD) is thus critical in both civilian and military applications. This study addresses the challenge of precisely IRSTD in complex backgrounds. Recent methods focus fundamental reliance on conventional convolution operations, which primarily capture local spatial patterns and struggle to distinguish the unique frequency-domain characteristics of small targets from intricate background clutter. To overcome these limitations, we proposed the Synergistic Wavelet-Attention Network (SWAN), a novel framework designed to perceive targets from both spatial and frequency domains. SWAN leverages a Haar Wavelet Convolution (HWConv) for a deep, cross-domain fusion of the frequency energy and spatial details of small target. Furthermore, a Shifted Spatial Attention (SSA) mechanism efficiently models long-range spatial dependencies with linear computational complexity, enhancing contextual awareness. Finally, a Residual Dual-Channel Attention (RDCA) module adaptively calibrates channel-wise feature responses to suppress background interference while amplifying target-pertinent signals. Extensive experiments on benchmark datasets demonstrate that SWAN surpasses existing state-of-the-art methods, showing significant improvements in detection accuracy and robustness, particularly in complex challenging scenarios.