Paddy Diseases Multi-Class Classification using CNN Variants
Agriculture is critical to ensuring and securing food supply for human consumption. Paddy crops are an agricultural practice that produces one of the highest food yields in agriculture. Paddy diseases must now be detected and diagnosed as soon as possible rather than later. Farmers currently use the...
| id |
Aziz D.I.A.B.A.; Yusoff M.; Ibrahim N.; Alazaidah R. |
|---|---|
| spelling |
Aziz D.I.A.B.A.; Yusoff M.; Ibrahim N.; Alazaidah R. 2-s2.0-85189140711 Paddy Diseases Multi-Class Classification using CNN Variants 2023 2023 24th International Arab Conference on Information Technology, ACIT 2023 10.1109/ACIT58888.2023.10453746 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189140711&doi=10.1109%2fACIT58888.2023.10453746&partnerID=40&md5=3725e53b861633ac06919fab83374d2e Agriculture is critical to ensuring and securing food supply for human consumption. Paddy crops are an agricultural practice that produces one of the highest food yields in agriculture. Paddy diseases must now be detected and diagnosed as soon as possible rather than later. Farmers currently use the naked eye to observe suspected infected paddies. The suspected infected paddy will then be sampled and sent to the laboratory for further examination. These methods are highly inefficient, time-consuming, and error prone. Most paddy farmers lack the knowledge to make an informed and accurate diagnosis of which paddies are infected by which diseases. This paper explores using deep learning algorithms to detect and classify various paddy diseases automatically. This method will take an image of paddy leaves and classify it based on the disease category indicated by the idea. The PaddyDoctor dataset from the Kaggle website was used to train and test the deep-learning models. Convolutional Neural Network (CNN) and Transfer Learning models, including MobileNet, MobileNetv2, and InceptionV3, have been constructed. The results of an experiment involving different dataset split ratios, learning rates, and batch sizes. CNN with MobileNetV2 outperformed CNN with MobileNetV1 and CNN with InceptionV3 by 98.05 percent, 80.08 percent, and 83.98 percent, respectively. © 2023 IEEE. Institute of Electrical and Electronics Engineers Inc. English Conference paper |
| author |
2-s2.0-85189140711 |
| spellingShingle |
2-s2.0-85189140711 Paddy Diseases Multi-Class Classification using CNN Variants |
| author_facet |
2-s2.0-85189140711 |
| author_sort |
2-s2.0-85189140711 |
| title |
Paddy Diseases Multi-Class Classification using CNN Variants |
| title_short |
Paddy Diseases Multi-Class Classification using CNN Variants |
| title_full |
Paddy Diseases Multi-Class Classification using CNN Variants |
| title_fullStr |
Paddy Diseases Multi-Class Classification using CNN Variants |
| title_full_unstemmed |
Paddy Diseases Multi-Class Classification using CNN Variants |
| title_sort |
Paddy Diseases Multi-Class Classification using CNN Variants |
| publishDate |
2023 |
| container_title |
2023 24th International Arab Conference on Information Technology, ACIT 2023 |
| container_volume |
|
| container_issue |
|
| doi_str_mv |
10.1109/ACIT58888.2023.10453746 |
| url |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85189140711&doi=10.1109%2fACIT58888.2023.10453746&partnerID=40&md5=3725e53b861633ac06919fab83374d2e |
| description |
Agriculture is critical to ensuring and securing food supply for human consumption. Paddy crops are an agricultural practice that produces one of the highest food yields in agriculture. Paddy diseases must now be detected and diagnosed as soon as possible rather than later. Farmers currently use the naked eye to observe suspected infected paddies. The suspected infected paddy will then be sampled and sent to the laboratory for further examination. These methods are highly inefficient, time-consuming, and error prone. Most paddy farmers lack the knowledge to make an informed and accurate diagnosis of which paddies are infected by which diseases. This paper explores using deep learning algorithms to detect and classify various paddy diseases automatically. This method will take an image of paddy leaves and classify it based on the disease category indicated by the idea. The PaddyDoctor dataset from the Kaggle website was used to train and test the deep-learning models. Convolutional Neural Network (CNN) and Transfer Learning models, including MobileNet, MobileNetv2, and InceptionV3, have been constructed. The results of an experiment involving different dataset split ratios, learning rates, and batch sizes. CNN with MobileNetV2 outperformed CNN with MobileNetV1 and CNN with InceptionV3 by 98.05 percent, 80.08 percent, and 83.98 percent, respectively. © 2023 IEEE. |
| publisher |
Institute of Electrical and Electronics Engineers Inc. |
| issn |
|
| language |
English |
| format |
Conference paper |
| accesstype |
|
| record_format |
scopus |
| collection |
Scopus |
| _version_ |
1828987866325188608 |