Coffee Disease Visualization and Classification
Deep learning architectures are widely used in state-of-the-art image classification tasks. Deep learning has enhanced the ability to automatically detect and classify plant diseases. However, in practice, disease classification problems are treated as black-box methods. Thus, it is difficult to tru...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-06-01
|
| Series: | Plants |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2223-7747/10/6/1257 |
| _version_ | 1797412018517442560 |
|---|---|
| author | Milkisa Yebasse Birhanu Shimelis Henok Warku Jaepil Ko Kyung Joo Cheoi |
| author_facet | Milkisa Yebasse Birhanu Shimelis Henok Warku Jaepil Ko Kyung Joo Cheoi |
| author_sort | Milkisa Yebasse |
| collection | DOAJ |
| description | Deep learning architectures are widely used in state-of-the-art image classification tasks. Deep learning has enhanced the ability to automatically detect and classify plant diseases. However, in practice, disease classification problems are treated as black-box methods. Thus, it is difficult to trust the model that it truly identifies the region of the disease in the image; it may simply use unrelated surroundings for classification. Visualization techniques can help determine important areas for the model by highlighting the region responsible for the classification. In this study, we present a methodology for visualizing coffee diseases using different visualization approaches. Our goal is to visualize aspects of a coffee disease to obtain insight into what the model “sees” as it learns to classify healthy and non-healthy images. In addition, visualization helped us identify misclassifications and led us to propose a guided approach for coffee disease classification. The guided approach achieved a classification accuracy of 98% compared to the 77% of naïve approach on the Robusta coffee leaf image dataset. The visualization methods considered in this study were Grad-CAM, Grad-CAM++, and Score-CAM. We also provided a visual comparison of the visualization methods. |
| first_indexed | 2024-03-09T04:54:54Z |
| format | Article |
| id | doaj.art-2b46335f97ef4110b2dcb8e06e59023c |
| institution | Directory Open Access Journal |
| issn | 2223-7747 |
| language | English |
| last_indexed | 2024-03-09T04:54:54Z |
| publishDate | 2021-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Plants |
| spelling | doaj.art-2b46335f97ef4110b2dcb8e06e59023c2023-12-03T13:06:24ZengMDPI AGPlants2223-77472021-06-01106125710.3390/plants10061257Coffee Disease Visualization and ClassificationMilkisa Yebasse0Birhanu Shimelis1Henok Warku2Jaepil Ko3Kyung Joo Cheoi4Department of Computer Engineering, Kumoh National Institute of Technology, Gumi 39177, KoreaArtificial Intelligence Center (AIC), Addis Ababa 2Q92+88, EthiopiaDepartment of IT-Bio Convergence System, Electronics Engineering, Graduate School, Chosun University, Gwangju 61452, KoreaDepartment of Computer Engineering, Kumoh National Institute of Technology, Gumi 39177, KoreaDepartment of Computer Science, Chungbuk National University, Cheongju 28644, KoreaDeep learning architectures are widely used in state-of-the-art image classification tasks. Deep learning has enhanced the ability to automatically detect and classify plant diseases. However, in practice, disease classification problems are treated as black-box methods. Thus, it is difficult to trust the model that it truly identifies the region of the disease in the image; it may simply use unrelated surroundings for classification. Visualization techniques can help determine important areas for the model by highlighting the region responsible for the classification. In this study, we present a methodology for visualizing coffee diseases using different visualization approaches. Our goal is to visualize aspects of a coffee disease to obtain insight into what the model “sees” as it learns to classify healthy and non-healthy images. In addition, visualization helped us identify misclassifications and led us to propose a guided approach for coffee disease classification. The guided approach achieved a classification accuracy of 98% compared to the 77% of naïve approach on the Robusta coffee leaf image dataset. The visualization methods considered in this study were Grad-CAM, Grad-CAM++, and Score-CAM. We also provided a visual comparison of the visualization methods.https://www.mdpi.com/2223-7747/10/6/1257coffee disease classificationcoffee disease visualizationdeep learningGrad-CAMScore-CAM |
| spellingShingle | Milkisa Yebasse Birhanu Shimelis Henok Warku Jaepil Ko Kyung Joo Cheoi Coffee Disease Visualization and Classification Plants coffee disease classification coffee disease visualization deep learning Grad-CAM Score-CAM |
| title | Coffee Disease Visualization and Classification |
| title_full | Coffee Disease Visualization and Classification |
| title_fullStr | Coffee Disease Visualization and Classification |
| title_full_unstemmed | Coffee Disease Visualization and Classification |
| title_short | Coffee Disease Visualization and Classification |
| title_sort | coffee disease visualization and classification |
| topic | coffee disease classification coffee disease visualization deep learning Grad-CAM Score-CAM |
| url | https://www.mdpi.com/2223-7747/10/6/1257 |
| work_keys_str_mv | AT milkisayebasse coffeediseasevisualizationandclassification AT birhanushimelis coffeediseasevisualizationandclassification AT henokwarku coffeediseasevisualizationandclassification AT jaepilko coffeediseasevisualizationandclassification AT kyungjoocheoi coffeediseasevisualizationandclassification |