EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Surface defect detection method for shawo radishes based on improved RT-DETR

Liying Ma, Xufeng Hua, Na Zhang, Xiaoxue Li, Chunyang Qian, Cunkun Chen, Chenghu Dong, Zhiqiang Zhu, Lihong Hao

Frontiers in Nutrition · 03.08.2026 · 13

DOI: 10.3389/fnut.2026.1890294 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Nutrition, online erschienen am 03.08.2026. Verfasst von Liying Ma und 8 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Information Engineering, Tianjin Agricultural University, Tianjin, China. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Surface defect detection of shawo radishes is challenged by complex defect characteristics, large variations in defect scale, and insufficient feature representation, which can limit automated quality inspection and grading performance.

Methods

To address these challenges, this study proposes an improved RT-DETR-r18 framework based on a frequency–spatial collaborative modeling strategy. The proposed model incorporates a Wavelet Transform Block (WT_Block), an Attention-based Intra-scale Feature Interaction with High-Low Attention module (AIFI-HiLo), and a VoVGSCSP-PConv Cross-Scale Feature Fusion Module (VP_CCFM) to enhance feature extraction, feature interaction, and multi-scale representation.

Results

Experiments conducted on a self-constructed shawo radish surface defect dataset showed that the proposed method achieved a Precision of 94.5%, a Recall of 81.2%, and a mean Average Precision at an Intersection over Union threshold of 0.5 (mAP@50) of 85.3%, representing improvements of 2.9, 2.8, and 1.9 percentage points over the baseline RT-DETR-r18 model, respectively. Meanwhile, the parameter count and computational complexity were reduced by 7 and 12.7%, respectively. Comparative experiments further showed that the proposed model outperformed several mainstream object detection methods in overall detection performance.

Discussion

These results indicate that the proposed frequency–spatial collaborative framework can improve the accuracy, robustness, and efficiency of shawo radish surface defect detection, providing technical support for automated quality inspection and intelligent grading of root vegetable products.

Schlagwörter

Stichwörter der Autor:innen

RT-DETRcomputer visiondeep learningshawo radishsurface defectswavelet transform

Autor:innen und Institutionen

  • College of Computer and Information Engineering, Tianjin Agricultural University, Tianjin, China.
  • State Key Laboratory of Vegetable Biobreeding, Key Laboratory of Storage and Preservation of Agricultural Products, Tianjin Key Laboratory of Postharvest Physiology and Storage and Preservation of Agricultural Products,…
  • Huangnan Tibetan Autonomous Prefecture Agriculture and Animal Husbandry Comprehensive Service Center, Tongren, Qinghai, China.
  • Tianjin Xiaowo Village Guli Fruit and Vegetable Planting Farmers' Professional Cooperative, Tianjin, China.

Förderung

  • National Key Research and Development Program of China

Zitieren

APA

Ma, L., Hua, X., Zhang, N., Li, X., Qian, C., Chen, C., Dong, C., Zhu, Z., & Hao, L. (2026). Surface defect detection method for shawo radishes based on improved RT-DETR. Frontiers in Nutrition, 13. https://doi.org/10.3389/fnut.2026.1890294

Vancouver

Ma L, Hua X, Zhang N, Li X, Qian C, Chen C, et al. Surface defect detection method for shawo radishes based on improved RT-DETR. Frontiers in Nutrition. 2026;13. doi:10.3389/fnut.2026.1890294

BibTeX
@article{Ma2026,
  author = {Ma, Liying and Hua, Xufeng and Zhang, Na and Li, Xiaoxue and Qian, Chunyang and Chen, Cunkun and Dong, Chenghu and Zhu, Zhiqiang and Hao, Lihong},
  title = {Surface defect detection method for shawo radishes based on improved RT-DETR},
  journal = {Frontiers in Nutrition},
  year = {2026},
  volume = {13},
  doi = {10.3389/fnut.2026.1890294}
}
Kostenlos starten →