|
二维码（扫一下试试看！） |
| Study on Multi-scale Classification of Eight Types of Pulp Fibers Based on ResNeXt50 Model |
| Received:December 11, 2025 Revised:February 06, 2026 |
| DOI：10.11980/j.issn.0254-508X.2026.06.013 |
| Key Words:pulp fiber classification ResNeXt50 CBAM |
| Fund Project:国家特殊海外高层次人才计划项目；制浆造纸国家工程实验室基金项目（ZZY2025JBGS04）。 |
| Author Name | Affiliation | Postcode | | YU Jiabao* | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | GAO Jie | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | FAN Shujie | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | WANG Zhen | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | CHEN Xuemei | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | YANG Guangzhao | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | LI Lijun | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | DU Xiu | 3Zhongqing Special Fiber Materials Co., Ltd., Langfang, Hebei Province, 065001
4Hebei Advanced Paper-based Functional Materials Technology Innovation Center, Langfang, Hebei Province, 065001 | 065001 | | ZHANG Xue* | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 | | DU Fan* | 1China National Pulp and Paper Research Institute Co., Ltd., Beijing, 100102
2National Engineering Lab for Pulp and Paper, Beijing, 100102 | 100102 |
|
| Hits: 800 |
| Download times: 57 |
| Abstract:To address the issues of low efficiency， high labour costs， and reliance on specialised personnel in traditional pulp fibre identification， this paper proposed an improved ResNeXt50 multi-scale classification model for eight types of typical pulp fibres. Using ResNeXt50 as the backbone network， classification performance was enhanced through three key optimisations： Employing transfer learning to freeze STAGE0 parameters， accelerating fitting and improving model generalisation. Introducing the convolutional block attention module (CBAM) mechanism to focus on key fibre morphological details while suppressing background noise. Designing a parallel multi-scale module to integrate local texture and global morphological features. Based on the constructed dataset of eight types of pulp fibres， model training and validation were completed on the PyTorch platform and compared with mainstream models. The results demonstrated that the improved model performed excellently in the classification of eight types of pulp fibres， achieving a precision of 92.41%， recall of 91.87%， and F1 score of 91.90%， with overall superior performance. |
| View Full Text HTML View/Add Comment Download reader |
|
|
|