Infrared Small Target Detection based on Adjustable Sensitivity Strategy and Multi-Scale Fusion

2024-07-29

arXiv.org

AI Summary

To address the challenges of feature sparsity and poor cross-resolution and cross-device generalization in infrared small target detection, this paper proposes a Multi-Scale Direction-Aware Network (MSDA-Net). Methodologically: (1) an Edge-Enhanced Difficulty Mining (EEDM) loss is designed to strengthen learning from hard samples and edge regions; (2) a tunable-sensitivity adaptive thresholding post-processing strategy is introduced to improve localization robustness; and (3) a multi-scale direction-aware fusion module is developed to jointly model shape, edge, and texture features at multiple granularities. Evaluated on the PRCV 2024 Wide-Domain Infrared Small Target Detection Challenge, MSDA-Net ranked first, achieving significant improvements in detection recall and segmentation accuracy. Crucially, it demonstrates strong generalization across diverse resolutions and hardware platforms, exhibiting device-agnostic robustness without requiring domain-specific fine-tuning.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

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

Abstract

Recently, deep learning-based single-frame infrared small target (SIRST) detection technology has made significant progress. However, existing infrared small target detection methods are often optimized for a fixed image resolution, a single wavelength, or a specific imaging system, limiting their breadth and flexibility in practical applications. Therefore, we propose a refined infrared small target detection scheme based on an adjustable sensitivity (AS) strategy and multi-scale fusion. Specifically, a multi-scale model fusion framework based on multi-scale direction-aware network (MSDA-Net) is constructed, which uses input images of multiple scales to train multiple models and fuses them. Multi-scale fusion helps characterize the shape, edge, and texture features of the target from different scales, making the model more accurate and reliable in locating the target. At the same time, we fully consider the characteristics of the infrared small target detection task and construct an edge enhancement difficulty mining (EEDM) loss. The EEDM loss helps alleviate the problem of category imbalance and guides the network to pay more attention to difficult target areas and edge features during training. In addition, we propose an adjustable sensitivity strategy for post-processing. This strategy significantly improves the detection rate of infrared small targets while ensuring segmentation accuracy. Extensive experimental results show that the proposed scheme achieves the best performance. Notably, this scheme won the first prize in the PRCV 2024 wide-area infrared small target detection competition.

Problem

• Enhancing infrared small target detection robustness
• Improving multi-scale feature perception and edge focus
• Regulating target confidence for complex scenarios

Innovation

• Multi-scale fusion enhances target feature perception
• Edge enhancement loss focuses on challenging regions
• Adjustable sensitivity strategy improves detection adaptability