Degraded Infrared Small Object Detection via Degradation-Adapted Physics-Guided Restoration

2026-08-10

AI Summary

This work addresses the challenge of infrared small target detection under severe contrast degradation caused by factors such as fog and non-uniformity, where existing methods exhibit limited generalization. To overcome this, we propose the DAISOD framework, which first identifies the type and severity of degradation to adaptively activate dedicated processing branches. It then incorporates a physics-guided restoration mechanism that explicitly estimates degradation parameters and performs controllable restoration based on the imaging model, thereby preventing target loss. Finally, multi-branch results are fused to achieve robust detection. We also introduce the first comprehensive infrared small target dataset encompassing diverse degradation types and intensities. Extensive experiments demonstrate that DAISOD significantly outperforms state-of-the-art methods across various degraded scenarios, confirming its superior performance and generalization capability.

Abstract

Infrared small object detection has made significant progress in recent years. However, degradations such as fog and nonuniformity can suppress target-background contrast, substantially increasing detection difficulty. Existing methods mainly rely on image restoration as preprocessing, but they are typically designed for specific degradation types and fail to generalize to varying degradations. To alleviate this, we propose DAISOD, a degradation-adapted infrared small object detection framework for robust detection under different degradations. DAISOD first identifies the type and severity of degradations, then adapts the processing via dedicated branches, and finally fuses the results for subsequent detection. Moreover, a physics-guided restoration mechanism is incorporated to explicitly estimate degradation parameters and remove degradation effects through physical models, avoiding excessive restoration that may erase small targets. Moreover, we construct a degraded infrared small object detection dataset covering diverse degradation types and levels. Extensive experiments show that DAISOD outperforms state-of-the-art methods under various degradation conditions.

Problem

• infrared small object detection
• image degradation
• degradation generalization
• target-background contrast
• nonuniformity

Innovation

• degradation-adapted
• physics-guided restoration
• infrared small object detection
• adaptive processing
• degraded dataset