Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks
We present a new approach of classification decision of tomato (Solanum lycopersicum) leaf disease based on the summarization of the output from a series of parallel convolutional neural networks with different configurations. The use of Swish, LeakyReLU-Swish, ReLU-Swish, Elu-Swish, and ClippedReLU...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2022-12-01
|
| Series: | Smart Agricultural Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772375522000193 |
| _version_ | 1818232229823774720 |
|---|---|
| author | M.P. Islam K. Hatou T. Aihara S. Seno S. Kirino S. Okamoto |
| author_facet | M.P. Islam K. Hatou T. Aihara S. Seno S. Kirino S. Okamoto |
| author_sort | M.P. Islam |
| collection | DOAJ |
| description | We present a new approach of classification decision of tomato (Solanum lycopersicum) leaf disease based on the summarization of the output from a series of parallel convolutional neural networks with different configurations. The use of Swish, LeakyReLU-Swish, ReLU-Swish, Elu-Swish, and ClippedReLU-Swish activation layers, as well as the Batch Normalization-Instance Normalization layer significantly improves the network performance, achieving classification accuracy over 99.0% with training, 97.5% with validation and 98.0% with testing dataset. Despite various performance metrics observed, none of the proposed networks overfitted on the validation dataset. Furthermore, we use various techniques to visualize the network performance. This demonstrates how the networks (Network 1, Network 2, Network 3, Network 4, Network 5) learn from the training dataset and can show diseased areas of leaves with high confident scores under real conditions. Network 1 shows the best performance in terms of network stability and visualization of the disease location. By computing, summarizing, and scoring the output of a series of parallel convolutional neural networks, the weakness of Network 4 and Network 5 in predicting the Healthy class can be overcome. This research will inspire and encourage further use of deep learning techniques to automatically detect and classify plant diseases under real conditions and improve financial condition of the farmers worldwide. |
| first_indexed | 2024-12-12T11:02:58Z |
| format | Article |
| id | doaj.art-613daa39faaf45e480aa93719c674e22 |
| institution | Directory Open Access Journal |
| issn | 2772-3755 |
| language | English |
| last_indexed | 2024-12-12T11:02:58Z |
| publishDate | 2022-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Smart Agricultural Technology |
| spelling | doaj.art-613daa39faaf45e480aa93719c674e222022-12-22T00:26:28ZengElsevierSmart Agricultural Technology2772-37552022-12-012100054Performance prediction of tomato leaf disease by a series of parallel convolutional neural networksM.P. Islam0K. Hatou1T. Aihara2S. Seno3S. Kirino4S. Okamoto5Corresponding author.; Department of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanDepartment of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanDepartment of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanDepartment of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanDepartment of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanDepartment of Biomechanical Systems, Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790-8566, JapanWe present a new approach of classification decision of tomato (Solanum lycopersicum) leaf disease based on the summarization of the output from a series of parallel convolutional neural networks with different configurations. The use of Swish, LeakyReLU-Swish, ReLU-Swish, Elu-Swish, and ClippedReLU-Swish activation layers, as well as the Batch Normalization-Instance Normalization layer significantly improves the network performance, achieving classification accuracy over 99.0% with training, 97.5% with validation and 98.0% with testing dataset. Despite various performance metrics observed, none of the proposed networks overfitted on the validation dataset. Furthermore, we use various techniques to visualize the network performance. This demonstrates how the networks (Network 1, Network 2, Network 3, Network 4, Network 5) learn from the training dataset and can show diseased areas of leaves with high confident scores under real conditions. Network 1 shows the best performance in terms of network stability and visualization of the disease location. By computing, summarizing, and scoring the output of a series of parallel convolutional neural networks, the weakness of Network 4 and Network 5 in predicting the Healthy class can be overcome. This research will inspire and encourage further use of deep learning techniques to automatically detect and classify plant diseases under real conditions and improve financial condition of the farmers worldwide.http://www.sciencedirect.com/science/article/pii/S2772375522000193Plant diseaseClassificationDeep learningCNN architecture |
| spellingShingle | M.P. Islam K. Hatou T. Aihara S. Seno S. Kirino S. Okamoto Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks Smart Agricultural Technology Plant disease Classification Deep learning CNN architecture |
| title | Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| title_full | Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| title_fullStr | Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| title_full_unstemmed | Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| title_short | Performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| title_sort | performance prediction of tomato leaf disease by a series of parallel convolutional neural networks |
| topic | Plant disease Classification Deep learning CNN architecture |
| url | http://www.sciencedirect.com/science/article/pii/S2772375522000193 |
| work_keys_str_mv | AT mpislam performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks AT khatou performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks AT taihara performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks AT sseno performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks AT skirino performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks AT sokamoto performancepredictionoftomatoleafdiseasebyaseriesofparallelconvolutionalneuralnetworks |