IST-YOLO: Infrared Small Target Detector based on Improved YOLOv8

Ruoyu Wang, Bicao Li, Bei Wang, Danting Niu, Yongzhao Wang
Proceedings of the 16th Asian Conference on Machine Learning, PMLR 260:399-414, 2025.

Abstract

Compared with natural images, the target of a single-frame infrared small target image occupies fewer pixels, has fuzzy imaging, less shape and texture information, and a more complex background. This leads to lower detection accuracy and makes it difficult to achieve accurate target localization. Therefore, in this paper, an infrared small target detection algorithm, IST-YOLO, is proposed based on yolov8. First, our algorithm improves the structure of standard model by adding an upsampling layer and a higher resolution detection head, which has a better ability to detect small targets. Second, we designed the Adaptive Residual Module (ARM) by combining the residual structure with the frequency adaptive dilated convolution to enhance the capacity of extracting deep small target position information while retaining the rich semantic information in the shallow layers. Finally, the Local and Globa Fusion (LGFusion) module is designed to enhance the information interaction between local and global features of the model. Experiments show that the accuracy of IST-YOLO outperforms both standard and popular algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v260-wang25c, title = {{IST-YOLO}: {I}nfrared Small Target Detector based on Improved YOLOv8}, author = {Wang, Ruoyu and Li, Bicao and Wang, Bei and Niu, Danting and Wang, Yongzhao}, booktitle = {Proceedings of the 16th Asian Conference on Machine Learning}, pages = {399--414}, year = {2025}, editor = {Nguyen, Vu and Lin, Hsuan-Tien}, volume = {260}, series = {Proceedings of Machine Learning Research}, month = {05--08 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v260/main/assets/wang25c/wang25c.pdf}, url = {https://proceedings.mlr.press/v260/wang25c.html}, abstract = {Compared with natural images, the target of a single-frame infrared small target image occupies fewer pixels, has fuzzy imaging, less shape and texture information, and a more complex background. This leads to lower detection accuracy and makes it difficult to achieve accurate target localization. Therefore, in this paper, an infrared small target detection algorithm, IST-YOLO, is proposed based on yolov8. First, our algorithm improves the structure of standard model by adding an upsampling layer and a higher resolution detection head, which has a better ability to detect small targets. Second, we designed the Adaptive Residual Module (ARM) by combining the residual structure with the frequency adaptive dilated convolution to enhance the capacity of extracting deep small target position information while retaining the rich semantic information in the shallow layers. Finally, the Local and Globa Fusion (LGFusion) module is designed to enhance the information interaction between local and global features of the model. Experiments show that the accuracy of IST-YOLO outperforms both standard and popular algorithms.} }
Endnote
%0 Conference Paper %T IST-YOLO: Infrared Small Target Detector based on Improved YOLOv8 %A Ruoyu Wang %A Bicao Li %A Bei Wang %A Danting Niu %A Yongzhao Wang %B Proceedings of the 16th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Vu Nguyen %E Hsuan-Tien Lin %F pmlr-v260-wang25c %I PMLR %P 399--414 %U https://proceedings.mlr.press/v260/wang25c.html %V 260 %X Compared with natural images, the target of a single-frame infrared small target image occupies fewer pixels, has fuzzy imaging, less shape and texture information, and a more complex background. This leads to lower detection accuracy and makes it difficult to achieve accurate target localization. Therefore, in this paper, an infrared small target detection algorithm, IST-YOLO, is proposed based on yolov8. First, our algorithm improves the structure of standard model by adding an upsampling layer and a higher resolution detection head, which has a better ability to detect small targets. Second, we designed the Adaptive Residual Module (ARM) by combining the residual structure with the frequency adaptive dilated convolution to enhance the capacity of extracting deep small target position information while retaining the rich semantic information in the shallow layers. Finally, the Local and Globa Fusion (LGFusion) module is designed to enhance the information interaction between local and global features of the model. Experiments show that the accuracy of IST-YOLO outperforms both standard and popular algorithms.
APA
Wang, R., Li, B., Wang, B., Niu, D. & Wang, Y.. (2025). IST-YOLO: Infrared Small Target Detector based on Improved YOLOv8. Proceedings of the 16th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 260:399-414 Available from https://proceedings.mlr.press/v260/wang25c.html.

Related Material