FSDN-DETR: Enhancing Fuzzy Systems Adapter with DeNoising Anchor Boxes for Transfer Learning in Small Object Detection

Zhijie Li, Jiahui Zhang, Yingjie Zhang, Dawei Yan, Xing Zhang, Marcin Woźniak () and Wei Dong ()
Additional contact information
Zhijie Li: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Jiahui Zhang: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Yingjie Zhang: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Dawei Yan: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Xing Zhang: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Marcin Woźniak: Institute of Mathematics, Silesian University of Technology, Kaszubska 23, 44-100 Gliwice, Poland
Wei Dong: College of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China

Mathematics, 2025, vol. 13, issue 2, 1-25

Abstract: The advancement of Transformer models in computer vision has rapidly spurred numerous Transformer-based object detection approaches, such as DEtection TRansformer. Although DETR’s self-attention mechanism effectively captures the global context, it struggles with fine-grained detail detection, limiting its efficacy in small object detection where noise can easily obscure or confuse small targets. To address these issues, we propose F uzzy S ystem DN N- DETR involving two key modules: Fuzzy Adapter Transformer Encoder and Fuzzy Denoising Transformer Decoder. The fuzzy Adapter Transformer Encoder utilizes adaptive fuzzy membership functions and rule-based smoothing to preserve critical details, such as edges and textures, while mitigating the loss of fine details in global feature processing. Meanwhile, the Fuzzy Denoising Transformer Decoder effectively reduces noise interference and enhances fine-grained feature capture, eliminating redundant computations in irrelevant regions. This approach achieves a balance between computational efficiency for medium-resolution images and the accuracy required for small object detection. Our architecture also employs adapter modules to reduce re-training costs, and a two-stage fine-tuning strategy adapts fuzzy modules to specific domains before harmonizing the model with task-specific adjustments. Experiments on the COCO and AI-TOD-V2 datasets show that FSDN-DETR achieves an approximately 20% improvement in average precision for very small objects, surpassing state-of-the-art models and demonstrating robustness and reliability for small object detection in complex environments.

Keywords: object detection; transformer; transfer learning; DEtection TRansformer; fuzzy system; adapter (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/13/2/287/pdf (application/pdf)
https://www.mdpi.com/2227-7390/13/2/287/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:13:y:2025:i:2:p:287-:d:1569313

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().