Crop Disease Classification on Inadequate Low-Resolution Target Images
Currently, various agricultural image classification tasks are carried out on high-resolution images. However, in some cases, we cannot get enough high-resolution images for classification, which significantly affects classification performance. In this paper, we design a crop disease classification...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/16/4601 |
| _version_ | 1797557833505439744 |
|---|---|
| author | Juan Wen Yangjing Shi Xiaoshi Zhou Yiming Xue |
| author_facet | Juan Wen Yangjing Shi Xiaoshi Zhou Yiming Xue |
| author_sort | Juan Wen |
| collection | DOAJ |
| description | Currently, various agricultural image classification tasks are carried out on high-resolution images. However, in some cases, we cannot get enough high-resolution images for classification, which significantly affects classification performance. In this paper, we design a crop disease classification network based on Enhanced Super-Resolution Generative adversarial networks (ESRGAN) when only an insufficient number of low-resolution target images are available. First, ESRGAN is used to recover super-resolution crop images from low-resolution images. Transfer learning is applied in model training to compensate for the lack of training samples. Then, we test the performance of the generated super-resolution images in crop disease classification task. Extensive experiments show that using the fine-tuned ESRGAN model can recover realistic crop information and improve the accuracy of crop disease classification, compared with the other four image super-resolution methods. |
| first_indexed | 2024-03-10T17:21:52Z |
| format | Article |
| id | doaj.art-cff736d8fe244502852cc1b082f48cb4 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T17:21:52Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-cff736d8fe244502852cc1b082f48cb42023-11-20T10:19:49ZengMDPI AGSensors1424-82202020-08-012016460110.3390/s20164601Crop Disease Classification on Inadequate Low-Resolution Target ImagesJuan Wen0Yangjing Shi1Xiaoshi Zhou2Yiming Xue3College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, ChinaCollege of Information and Electrical Engineering, China Agricultural University, Beijing 100083, ChinaCollege of Information and Electrical Engineering, China Agricultural University, Beijing 100083, ChinaCollege of Information and Electrical Engineering, China Agricultural University, Beijing 100083, ChinaCurrently, various agricultural image classification tasks are carried out on high-resolution images. However, in some cases, we cannot get enough high-resolution images for classification, which significantly affects classification performance. In this paper, we design a crop disease classification network based on Enhanced Super-Resolution Generative adversarial networks (ESRGAN) when only an insufficient number of low-resolution target images are available. First, ESRGAN is used to recover super-resolution crop images from low-resolution images. Transfer learning is applied in model training to compensate for the lack of training samples. Then, we test the performance of the generated super-resolution images in crop disease classification task. Extensive experiments show that using the fine-tuned ESRGAN model can recover realistic crop information and improve the accuracy of crop disease classification, compared with the other four image super-resolution methods.https://www.mdpi.com/1424-8220/20/16/4601super-resolutionGenerative Adversarial NetworksConvolutional Neural Networksdisease classification |
| spellingShingle | Juan Wen Yangjing Shi Xiaoshi Zhou Yiming Xue Crop Disease Classification on Inadequate Low-Resolution Target Images Sensors super-resolution Generative Adversarial Networks Convolutional Neural Networks disease classification |
| title | Crop Disease Classification on Inadequate Low-Resolution Target Images |
| title_full | Crop Disease Classification on Inadequate Low-Resolution Target Images |
| title_fullStr | Crop Disease Classification on Inadequate Low-Resolution Target Images |
| title_full_unstemmed | Crop Disease Classification on Inadequate Low-Resolution Target Images |
| title_short | Crop Disease Classification on Inadequate Low-Resolution Target Images |
| title_sort | crop disease classification on inadequate low resolution target images |
| topic | super-resolution Generative Adversarial Networks Convolutional Neural Networks disease classification |
| url | https://www.mdpi.com/1424-8220/20/16/4601 |
| work_keys_str_mv | AT juanwen cropdiseaseclassificationoninadequatelowresolutiontargetimages AT yangjingshi cropdiseaseclassificationoninadequatelowresolutiontargetimages AT xiaoshizhou cropdiseaseclassificationoninadequatelowresolutiontargetimages AT yimingxue cropdiseaseclassificationoninadequatelowresolutiontargetimages |