Degraded Infrared Small Object Detection via Degradation-Adapted Physics-Guided Restoration
2026-08-10 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present DAISOD, a new system to detect tiny objects in infrared images that are affected by issues like fog or uneven lighting. Unlike previous methods made for specific problems, their system first figures out what kind of issue the image has, then uses specialized steps to fix it without removing important details. They also created a dataset with many types of image problems to test their method. Their experiments show DAISOD works better than existing methods in different difficult conditions.
infrared imagingsmall object detectionimage degradationnonuniformityimage restorationphysics-guided restorationtarget-background contrastdegradation adaptationdatasetfog
Authors
Xinkai Lu, Wenjun Chen, Yi Li, Yi Chang, Luxin Yan
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.