A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images
Maize is a globally important cereal crop, however, maize leaf disease is one of the most common and devastating diseases that afflict it. Artificial intelligence methods face challenges in identifying and classifying maize leaf diseases due to variations in image quality, similarity among diseases,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-04-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844024042956 |
| _version_ | 1797237949639688192 |
|---|---|
| author | Liangliang Liu Shixin Qiao Jing Chang Weiwei Ding Cifu Xu Jiamin Gu Tong Sun Hongbo Qiao |
| author_facet | Liangliang Liu Shixin Qiao Jing Chang Weiwei Ding Cifu Xu Jiamin Gu Tong Sun Hongbo Qiao |
| author_sort | Liangliang Liu |
| collection | DOAJ |
| description | Maize is a globally important cereal crop, however, maize leaf disease is one of the most common and devastating diseases that afflict it. Artificial intelligence methods face challenges in identifying and classifying maize leaf diseases due to variations in image quality, similarity among diseases, disease severity, limited dataset availability, and limited interpretability. To address these challenges, we propose a residual-based multi-scale network (MResNet) for classifying multi-type maize leaf diseases from maize images. MResNet consists of two residual subnets with different scales, enabling the model to detect diseases in maize leaf images at different scales. We further utilize a hybrid feature weight optimization method to optimize and fuse the feature mapping weights of two subnets. We validate MResNet on a maize leaf diseases dataset. MResNet achieves 97.45% accuracy. The performance of MResNet surpasses other state-of-the-art methods. Various experiments and two additional datasets confirm the generalization performance of our model. Furthermore, thermodynamic diagram analysis increases the interpretability of the model. This study provides technical support for the disease classification of agricultural plants. |
| first_indexed | 2024-04-24T17:27:53Z |
| format | Article |
| id | doaj.art-23cdaf745a35499680c5f56bf5b701b9 |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-04-24T17:27:53Z |
| publishDate | 2024-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-23cdaf745a35499680c5f56bf5b701b92024-03-28T06:38:24ZengElsevierHeliyon2405-84402024-04-01107e28264A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf imagesLiangliang Liu0Shixin Qiao1Jing Chang2Weiwei Ding3Cifu Xu4Jiamin Gu5Tong Sun6Hongbo Qiao7College of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Agriculture, Shihezi University, Shihezi, Xinjiang 832061, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR ChinaCollege of Information and Management Science, Henan Agricultural University, Zhengzhou, Henan 450046, PR China; Corresponding author.Maize is a globally important cereal crop, however, maize leaf disease is one of the most common and devastating diseases that afflict it. Artificial intelligence methods face challenges in identifying and classifying maize leaf diseases due to variations in image quality, similarity among diseases, disease severity, limited dataset availability, and limited interpretability. To address these challenges, we propose a residual-based multi-scale network (MResNet) for classifying multi-type maize leaf diseases from maize images. MResNet consists of two residual subnets with different scales, enabling the model to detect diseases in maize leaf images at different scales. We further utilize a hybrid feature weight optimization method to optimize and fuse the feature mapping weights of two subnets. We validate MResNet on a maize leaf diseases dataset. MResNet achieves 97.45% accuracy. The performance of MResNet surpasses other state-of-the-art methods. Various experiments and two additional datasets confirm the generalization performance of our model. Furthermore, thermodynamic diagram analysis increases the interpretability of the model. This study provides technical support for the disease classification of agricultural plants.http://www.sciencedirect.com/science/article/pii/S2405844024042956Multi-scaleResidual networkMulti-classMaize leaf images |
| spellingShingle | Liangliang Liu Shixin Qiao Jing Chang Weiwei Ding Cifu Xu Jiamin Gu Tong Sun Hongbo Qiao A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images Heliyon Multi-scale Residual network Multi-class Maize leaf images |
| title | A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images |
| title_full | A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images |
| title_fullStr | A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images |
| title_full_unstemmed | A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images |
| title_short | A multi-scale feature fusion neural network for multi-class disease classification on the maize leaf images |
| title_sort | multi scale feature fusion neural network for multi class disease classification on the maize leaf images |
| topic | Multi-scale Residual network Multi-class Maize leaf images |
| url | http://www.sciencedirect.com/science/article/pii/S2405844024042956 |
| work_keys_str_mv | AT liangliangliu amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT shixinqiao amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT jingchang amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT weiweiding amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT cifuxu amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT jiamingu amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT tongsun amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT hongboqiao amultiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT liangliangliu multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT shixinqiao multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT jingchang multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT weiweiding multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT cifuxu multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT jiamingu multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT tongsun multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages AT hongboqiao multiscalefeaturefusionneuralnetworkformulticlassdiseaseclassificationonthemaizeleafimages |