Denoising-Enhanced YOLO for Robust SAR Ship Detection

2026-02-27

AI Summary

This work proposes CPN-YOLO, a novel detection framework addressing the challenges of frequent missed detections and false alarms in SAR imagery caused by clutter and speckle noise. Built upon YOLOv8, the method incorporates a learnable large-kernel denoising module to suppress noise, integrates the PPA attention mechanism to enhance multi-scale feature representation, and introduces a Gaussian similarity loss based on the normalized Wasserstein distance to improve sensitivity to small targets and model generalization. Evaluated on the SSDD dataset, CPN-YOLO achieves 97.0% precision, 95.1% recall, and 98.9% mAP, significantly outperforming the YOLOv8 baseline and other state-of-the-art detectors.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Learning on the Edge & Model Compression
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

With the rapid advancement of deep learning, synthetic aperture radar (SAR) imagery has become a key modality for ship detection. However, robust performance remains challenging in complex scenes, where clutter and speckle noise can induce false alarms and small targets are easily missed. To address these issues, we propose CPN-YOLO, a high-precision ship detection framework built upon YOLOv8 with three targeted improvements. First, we introduce a learnable large-kernel denoising module for input pre-processing, producing cleaner representations and more discriminative features across diverse ship types. Second, we design a feature extraction enhancement strategy based on the PPA attention mechanism to strengthen multi-scale modeling and improve sensitivity to small ships. Third, we incorporate a Gaussian similarity loss derived from the normalized Wasserstein distance (NWD) to better measure similarity under complex bounding-box distributions and improve generalization. Extensive experiments on HRSID and SSDD demonstrate the effectiveness of our method. On SSDD, CPN-YOLO surpasses the YOLOv8 baseline, achieving 97.0% precision, 95.1% recall, and 98.9% mAP, and consistently outperforms other representative deep-learning detectors in overall performance.

Problem

• SAR ship detection
• speckle noise
• false alarms
• small target detection
• robustness

Innovation

• denoising module
• PPA attention
• Gaussian similarity loss
• normalized Wasserstein distance
• small ship detection