Deep Learning Model for the Inspection of Coffee Bean Defects
Source abstract
Abstract
The detection of coffee bean defects is the most crucial step prior to bean roasting. Existing defect detection methods used in the specialty coffee bean industry entail manual screening and sorting, require substantial human resources, and are not standardized. To solve these problems, this study developed a deep learning algorithm to detect defects in coffee beans. The results reveal that when the pooling layer was used to enhance features and reduce neural dimensionality, some of the coffee been features were lost or misclassified. Therefore, a novel dimensionality reduction method was adopted to increase the ability of feature extraction. The developed model also overcame the drawbacks of padding causing blurred image boundaries and the dead neurons causing impeding feature propagation. Images of eight types of coffee beans were used to train and test the proposed detection model. The proposed method was verified to reduce the bias when classifying defects in coffee beans. The detection accuracy rate of the proposed model was 95.2%. When the model was only used to detect the presence of defects, the accuracy rate increased to 100%. Thus, the proposed model is highly accurate in coffee bean defect detection in the classification of eight types of coffee beans.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Food Supply Chain Traceability · Smart Agriculture and AI
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mechanical Engineering, National Yunlin University of Science and Technology, Douliou 64002, Taiwan
Funding
Ministry of Science and Technology, Taiwan
References (28)
Show source-provided references
1. Automatic detection and classification of sewer defects via hierarchical deep learning10.1109/TASE.2019.2900170 ↗
2. Deep-structured machine learning model for the recognition of mixed-defect patterns in semiconductor fabrication processes10.1109/TSM.2018.2825482 ↗
3. Image-based deep learning automated sorting of date fruit10.1016/j.postharvbio.2019.04.003 ↗
4. Computer vision based detection of external defects on tomatoes using deep learning10.1016/j.biosystemseng.2019.12.003 ↗
5. Casaño, C.D.L.C., Sánchez, M.C., Chavez, F.R., and Ramos, W.V. (2020, January 21–23). Defect detection on andean potatoes using deep learning and adaptive learning. Proceedings of the IEEE Engineering International Research Conference (EIRCON), Lima, Peru.
6. Chouhan, S.S., Kaul, A., and Singh, U.P. (2019, January 17–19). A deep learning approach for the classification of diseased plant leaf images. Proceedings of the International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India.10.1109/ICCES45898.2019.9002201 ↗
7. Marzougui, F., Elleuch, M., and Kherallah, M. (2020, January 28–30). A Deep CNN Approach for plant disease detection. Proceedings of the 2020 21st International Arab Conference on Information Technology (ACIT), Giza, Egypt.10.1109/ACIT50332.2020.9300072 ↗
8. An innovative multivariate strategy for HSI-NIR images to automatically detect defects in green coffee10.1016/j.talanta.2019.02.049 ↗
9. Birhanu, H., and Tesfa, T. (2015, January 8–9). Ethiopian roasted coffee classification using imaging techniques. Proceedings of the 3rd International Conference on the Advancement of Science and Technology, Bahirdar, Ethiopia.
10. Coffee bean grade determination based on image parameter10.12928/telkomnika.v9i3.747 ↗
11. Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
12. Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., and Rabinovich, A. (2015, January 7–12). Going deeper with convolutions. Proceedings of the the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA.10.1109/CVPR.2015.7298594 ↗
13. He, K., Zhang, X., Ren, S., and Sun, J. (July, January 26). Deep residual learning for image recognition. Proceedings of the the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.
14. Zhou, H., Zhuang, Z., Liu, Y., Liu, Y., and Zhang, X. (2020). Defect classification of green plums based on deep learning. Sensor, 20.10.3390/s20236993 ↗
15. Zhu, J., and Song, J. (2020). An intelligent classification model for surface defects on cement concrete bridges. Appl. Sci., 10.10.3390/app10030972 ↗
16. ImageNet classification with deep convolutional neural networks10.1145/3065386 ↗
17. Chatfield, K., Simonyan, K., Vedaldi, A., and Zisserman, A. (2014, January 1–5). Return of the devil in the details: Delving deep into convolutional nets. Proceedings of the British Machine Vision Conference, Nottingham, UK.10.5244/C.28.6 ↗
18. Stursa, D., and Dolezel, P. (2019, January 11–14). Comparison of ReLU and linear saturated activation functions in neural network for universal approximation. Proceedings of the 22nd International Conference on Process Control (PC19), Štrbské Pleso, Slovakia.10.1109/PC.2019.8815057 ↗
19. Hongyo, R., Egashira, Y., and Yamaguchi, K. (2018, January 6–9). Deep neural network based predistorter with relu activation for Doherty power amplifiers. Proceedings of the Asia-Pacific Microwave Conference (APMC), Kyoto, Japan.10.23919/APMC.2018.8617612 ↗
20. Lee, K., Sung, S.H., Kim, D.H., and Park, S.H. (2019, January 9–10). Verification of normalization effects through comparison of CNN models. Proceedings of the International Conference on Multimedia Analysis and Pattern Recognition (MAPR), Ho Chi Minh City, Vietnam.10.1109/MAPR.2019.8743531 ↗
21. Tian, Y., Chen, F., Wang, H., and Zhang, S. (2020, January 16–18). Real-time semantic segmentation network based on lite reduced atrous spatial pyramid pooling module group. Proceedings of the 5th International Conference on Control, Robotics and Cybernetics (CRC), Wuhan, China.10.1109/CRC51253.2020.9253492 ↗
22. Universal pooling—A new pooling method for convolutional neural networks10.1016/j.eswa.2021.115084 ↗
23. de Buy Wenniger, G.M., Schomaker, L., and Way, A. (2019, January 20–25). No padding please: Efficient neural handwriting recognition. Proceedings of the International Conference on Document Analysis and Recognition (ICDAR), Sydney, Australia.
24. Douglas, S.C., and Yu, J. (2018, January 28–31). Why RELU units sometimes die: Analysis of single-unit error backpropagation in neural networks. Proceedings of the 52nd Asilomar Conference on Signals, Systems, and Computers (ACSSC), Pacific Grove, CA, USA.10.1109/ACSSC.2018.8645556 ↗
25. He, K., Zhang, X., Ren, S., and Sun, J. (1995, January 20–23). Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. Proceedings of the IEEE international conference on computer vision, Cambridge, MA, USA.
26. Design and calibration of a dual-frame force and torque sensor10.1109/JSEN.2020.2999156 ↗
27. The role of neural network activation functions10.1109/LSP.2020.3027517 ↗
28. Xu, J., Li, Z., Du, B., Zhang, M., and Liu, J. (2020, January 7–10). Reluplex made more practical: Leaky ReLU. Proceedings of the IEEE Symposium on Computers and Communications (ISCC), Rennes, France.10.1109/ISCC50000.2020.9219587 ↗