Multi-Plant Disease Identification Based on Lightweight ResNet18 Model
Deep-learning-based methods for plant disease recognition pose challenges due to their high number of network parameters, extensive computational requirements, and overall complexity. To address this issue, we propose an improved residual-network-based multi-plant disease recognition method that com...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/13/11/2702 |
| _version_ | 1797460567443636224 |
|---|---|
| author | Li Ma Yuanhui Hu Yao Meng Zhiyi Li Guifen Chen |
| author_facet | Li Ma Yuanhui Hu Yao Meng Zhiyi Li Guifen Chen |
| author_sort | Li Ma |
| collection | DOAJ |
| description | Deep-learning-based methods for plant disease recognition pose challenges due to their high number of network parameters, extensive computational requirements, and overall complexity. To address this issue, we propose an improved residual-network-based multi-plant disease recognition method that combines the characteristics of plant diseases. Our approach introduces a lightweight technique called maximum grouping convolution to the ResNet18 model. We made three enhancements to adapt this method to the characteristics of plant diseases and ultimately reduced the convolution kernel requirements, resulting in the final model, Model_Lite. The experimental dataset comprises 20 types of plant diseases, including 13 selected from the publicly available Plant Village dataset and seven self-constructed images of apple leaves with complex backgrounds containing disease symptoms. The experimental results demonstrated that our improved network model, Model_Lite, contains only about 1/344th of the parameters and requires 1/35th of the computational effort compared to the original ResNet18 model, with a marginal decrease in the average accuracy of only 0.34%. Comparing Model_Lite with MobileNet, ShuffleNet, SqueezeNet, and GhostNet, our proposed Model_Lite model achieved a superior average recognition accuracy while maintaining a much smaller number of parameters and computational requirements than the above models. Thus, the Model_Lite model holds significant potential for widespread application in plant disease recognition and can serve as a valuable reference for future research on lightweight network model design. |
| first_indexed | 2024-03-09T17:06:53Z |
| format | Article |
| id | doaj.art-e5224e6ba2e84fb0b5cdcf67753ce000 |
| institution | Directory Open Access Journal |
| issn | 2073-4395 |
| language | English |
| last_indexed | 2024-03-09T17:06:53Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj.art-e5224e6ba2e84fb0b5cdcf67753ce0002023-11-24T14:23:39ZengMDPI AGAgronomy2073-43952023-10-011311270210.3390/agronomy13112702Multi-Plant Disease Identification Based on Lightweight ResNet18 ModelLi Ma0Yuanhui Hu1Yao Meng2Zhiyi Li3Guifen Chen4College of Information and Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information and Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Information and Technology, Jilin Agricultural University, Changchun 130118, ChinaCollege of Instrumentation & Electrical Engineering, Jilin University, Changchun 130012, ChinaChangchun Humanities and Sciences College, Changchun 130118, ChinaDeep-learning-based methods for plant disease recognition pose challenges due to their high number of network parameters, extensive computational requirements, and overall complexity. To address this issue, we propose an improved residual-network-based multi-plant disease recognition method that combines the characteristics of plant diseases. Our approach introduces a lightweight technique called maximum grouping convolution to the ResNet18 model. We made three enhancements to adapt this method to the characteristics of plant diseases and ultimately reduced the convolution kernel requirements, resulting in the final model, Model_Lite. The experimental dataset comprises 20 types of plant diseases, including 13 selected from the publicly available Plant Village dataset and seven self-constructed images of apple leaves with complex backgrounds containing disease symptoms. The experimental results demonstrated that our improved network model, Model_Lite, contains only about 1/344th of the parameters and requires 1/35th of the computational effort compared to the original ResNet18 model, with a marginal decrease in the average accuracy of only 0.34%. Comparing Model_Lite with MobileNet, ShuffleNet, SqueezeNet, and GhostNet, our proposed Model_Lite model achieved a superior average recognition accuracy while maintaining a much smaller number of parameters and computational requirements than the above models. Thus, the Model_Lite model holds significant potential for widespread application in plant disease recognition and can serve as a valuable reference for future research on lightweight network model design.https://www.mdpi.com/2073-4395/13/11/2702computer visiondeep learningimage processingdisease identificationconvolutional neural networks |
| spellingShingle | Li Ma Yuanhui Hu Yao Meng Zhiyi Li Guifen Chen Multi-Plant Disease Identification Based on Lightweight ResNet18 Model Agronomy computer vision deep learning image processing disease identification convolutional neural networks |
| title | Multi-Plant Disease Identification Based on Lightweight ResNet18 Model |
| title_full | Multi-Plant Disease Identification Based on Lightweight ResNet18 Model |
| title_fullStr | Multi-Plant Disease Identification Based on Lightweight ResNet18 Model |
| title_full_unstemmed | Multi-Plant Disease Identification Based on Lightweight ResNet18 Model |
| title_short | Multi-Plant Disease Identification Based on Lightweight ResNet18 Model |
| title_sort | multi plant disease identification based on lightweight resnet18 model |
| topic | computer vision deep learning image processing disease identification convolutional neural networks |
| url | https://www.mdpi.com/2073-4395/13/11/2702 |
| work_keys_str_mv | AT lima multiplantdiseaseidentificationbasedonlightweightresnet18model AT yuanhuihu multiplantdiseaseidentificationbasedonlightweightresnet18model AT yaomeng multiplantdiseaseidentificationbasedonlightweightresnet18model AT zhiyili multiplantdiseaseidentificationbasedonlightweightresnet18model AT guifenchen multiplantdiseaseidentificationbasedonlightweightresnet18model |