JERAJournal of Electronic Research and Application2208-35022208-3510Bio-Byword Scientific Publishing Pty. Ltd.10.26689/jera.v10i2.14386ArticleGRIN Lens End Face Classification Detection Based on Deep Learninghttps://artdesignp.com/journal/JERA/10/2/10.26689/jera.v10i2.14386ZhangJianqiang,FuYong'an,ZhangFoxiang20261022026-04-03With the rapid development of optical communication technology, the demand for gradient-index (GRIN) lenses has increased significantly, making quality inspection of lens end faces an increasingly critical issue. In particular, accurate detection of defects on both ends of GRIN lenses remains a challenging task. To address this problem, this study employs a transfer learning-based parameter fine-tuning approach to evaluate the classification performance of four deep learning models on a defect dataset. Among the evaluated models, ResNet50 and DenseNet-169 demonstrated superior performance and were selected for further optimization. Attention mechanisms, including squeeze-and-excitation (SE) and convolutional block attention module (CBAM), were incorporated into these models to enhance feature representation. Experimental results show that, after integrating the SE module, the classification accuracy of ResNet50 and DenseNet-169 increased by 0.0243 and 0.0272, respectively. With the addition of the CBAM module, the accuracy improvements reached 0.0437 for ResNet50 and 0.0506 for DenseNet-169. These results indicate that the proposed improvements significantly enhance the defect detection capability of the models. All evaluation metrics show consistent improvement over the baseline models, demonstrating that the integration of attention mechanisms effectively increases the classification accuracy and overall performance of the original network architectures.

Xing J, Jiao M, Liu Y, 2015, Application of Self Focusing Lens in all Solid-State Lasers. Journal of Xi’an University of Technology, 31(2): 127–131+124.

Wu S, Zhang M, Zhang B, 2016, Hierarchical Structure Information and Its Expression for Image Features Extraction and Processing of GRIN Lens End, Science And Engineering Research Center, Proceedings of 2016 International Conference on Electrical Engineering and Automation (ICEEA 2016), 643–646.

Wu S, Zhang B, Zhang M, 2017, An Improved Median Filtering Method and its Applications in Features Extraction of GRIN Lens EndImage”.

Chu H, 2011, Research on Surface Defect Detection Technology of Optical Components in High Power Laser Devices Based on Machine Vision, thesis, Chongqing University.

Sowers I, 1999, Optical Cleanliness Specifications and Cleanliness Verification, Proceedings of the 44th Annual Meeting of the International Symposium on Optical Science, Engineering, and Instrumentation, 525–530.

Shi W, 2000, Research on High-Precision Cleanliness Detection Methods, thesis, Sichuan University, 9–15.

Liu G, Wang J, Ning R, et al., 2021, Self Focusing Lens End Face Image Processing and Defect Feature Extraction. Electronic Production, 2021(5): 74–76.

Wu J, 2020, Research and Implementation of Fine-Grained Emotion Classification Method Based on Attention Mechanism, thesis, Heilongjiang University.

Bas A, Huber P, Smith W, et al., 2017, 3D Morphable Models as Spatial Transformer Networks, 2017 IEEE International Conference on Computer Vision Workshop (ICCVW), 895–903.

Woo S, Park J, Lee J, et al., 2018, Cbam: Convolutional Block Attention Module. 2018 European Conference on Computer Vision (ECCV), 3–19.

Jie H, Li S, Samuel A, et al., 2020, Squeeze-and-Excitation Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(8): 2011–2023.

Woo S, Park J, Lee J, et al., 2018, CBAM: Convolutional Block Attention Module, The European Conference on Computer Vision, 3–19.