|
二维码（扫一下试试看！） |
| Lightweight Paper Defect Detection Model Based on Re-parameterization and Mutile Query Self Attention |
| Received:December 17, 2024 Revised:December 23, 2024 |
| DOI：10.11981/j.issn.1000-6842.2025.04.186 |
| Key Words:paper defect detection;RepLite-YOLO;re-parameterization;mutile query self attention;NMS-Free |
| Fund Project:陕西省科技厅工业领域一般项目（2024GX-YBXM-544）。 |
| Author Name | Affiliation | Postcode | | LIU Xin | School of Electrical and Control Engineering，Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 | 710021 | | ZHOU Qiang* | School of Electrical and Control Engineering，Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 | 710021 | | CHEN Yaxin | School of Electrical and Control Engineering，Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 | 710021 | | ZHANG Xiaohui | Rainbow Group (Shaoyang) Special Glass Co.， Ltd.， Shaoyang， Hu’nan Province， 422000）
（corresponding author：E-mail： zhouqiang@sust.edu.cn | 422000 |
|
| Hits: 2 |
| Download times: 1 |
| Abstract: |
| To address the challenges in existing deep learning algorithms， which struggle to balance detection accuracy and speed in paper defect detection， leading to the limited practicality of machine vision-based systems， this paper proposed a lightweight paper defect detection model named RepLite-YOLO based on re-parameterization and mutile query self attention. First， a lightweight partially depthwise separable convolution was designed to reduce model complexity. On this basis， the RCNXC2f module was constructed to enhance the model’s capability of extracting multi-scale features， with re-parameterization employed to reduce computational costs during inference. Subsequently， the C2PMQA module was integrated into the tail of the backbone network to improve the refinement of global features. Finally， a lightweight detection head module was developed， incorporating a training strategy without Non-Maximum Suppression (NMS) to effectively enhance real-time detection performance. Experimental results on a self-constructed paper defect dataset demonstrated that the proposed RepLite-YOLO achieves a mean Average Precision (mAP) of 99.2%， with a model size of only 1.35 MB. On the NVIDIA 4060 platform， the model achieved a frame rate of 175 FPS for 512×512 resolution paper defect images. RepLite-YOLO model not only ensured the precision of paper defect detection， but also outperformed other paper defect detection models in terms of real-time performance. |
| View Full Text View/Add Comment Download reader HTML |
