DenoDet V2: Phase-Amplitude Cross Denoising for SAR Object Detection

TL;DR

DenoDet V2 introduces a novel transform domain approach with phase-amplitude cross denoising for SAR object detection, significantly improving accuracy and reducing model complexity over previous methods.

Contribution

It presents a new attention-based architecture that leverages phase and amplitude spectra for enhanced SAR object detection, outperforming prior models.

Findings

Achieves 0.8% higher accuracy on SARDet-100K dataset.

Reduces model complexity by 50%.

Demonstrates state-of-the-art performance across multiple SAR datasets.

Abstract

One of the primary challenges in Synthetic Aperture Radar (SAR) object detection lies in the pervasive influence of coherent noise. As a common practice, most existing methods, whether handcrafted approaches or deep learning-based methods, employ the analysis or enhancement of object spatial-domain characteristics to achieve implicit denoising. In this paper, we propose DenoDet V2, which explores a completely novel and different perspective to deconstruct and modulate the features in the transform domain via a carefully designed attention architecture. Compared to DenoDet V1, DenoDet V2 is a major advancement that exploits the complementary nature of amplitude and phase information through a band-wise mutual modulation mechanism, which enables a reciprocal enhancement between phase and amplitude spectra. Extensive experiments on various SAR datasets demonstrate the state-of-the-art performance of DenoDet V2. Notably, DenoDet V2 achieves a significant 0.8\% improvement on SARDet-100K dataset compared to DenoDet V1, while reducing the model complexity by half. The code is available at https://github.com/GrokCV/GrokSAR.