A high-precision image classification network model based on a voting mechanism
With the development of satellite remote sensing technology, image classification task, as the basis of remote sensing data interpretation, has received wide attention to improving accuracy and robustness. At the same time, in-depth learning technology has been widely used in remote sensing and has...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2022-12-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1080/17538947.2022.2142306 |
| _version_ | 1797678473502785536 |
|---|---|
| author | Jianghong Zhao Xin Wang Xintong Dou Yingxue Zhao Zexin Fu Ming Guo Ruiju Zhang |
| author_facet | Jianghong Zhao Xin Wang Xintong Dou Yingxue Zhao Zexin Fu Ming Guo Ruiju Zhang |
| author_sort | Jianghong Zhao |
| collection | DOAJ |
| description | With the development of satellite remote sensing technology, image classification task, as the basis of remote sensing data interpretation, has received wide attention to improving accuracy and robustness. At the same time, in-depth learning technology has been widely used in remote sensing and has a far-reaching impact. Since the existing image classification methods ignore the feature that the general image semantics are the same as the semantics of a single pixel, this paper presents an algorithm that uses the semantics of an image to achieve high-precision image classification. Based on the idea of partial substitution for global, this algorithm designs a split result voting mechanism and builds a Vgg-Vote network model. This mechanism votes on the semantically segmented result of an image and uses the maximum filtering function to select the category containing the most significant number of pixels as the prediction category of the image. Experiments on UC Merced Land-User complete datasets and five types of incomplete datasets with varying degrees of interference, including noise, data occlusion and loss, show that the Vote mechanism dramatically improves the classification accuracy, robustness and anti-jamming capability of Vgg-Vote. |
| first_indexed | 2024-03-11T23:00:16Z |
| format | Article |
| id | doaj.art-28ad4c6526bf401180009ffcd3a50aaf |
| institution | Directory Open Access Journal |
| issn | 1753-8947 1753-8955 |
| language | English |
| last_indexed | 2024-03-11T23:00:16Z |
| publishDate | 2022-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj.art-28ad4c6526bf401180009ffcd3a50aaf2023-09-21T14:57:11ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552022-12-011512168218310.1080/17538947.2022.21423062142306A high-precision image classification network model based on a voting mechanismJianghong Zhao0Xin Wang1Xintong Dou2Yingxue Zhao3Zexin Fu4Ming Guo5Ruiju Zhang6School of Geomatics and Urban Spatial InformaticsSchool of Geomatics and Urban Spatial InformaticsSchool of Geomatics and Urban Spatial InformaticsGuangzhou Panyu PolytechnicSchool of Geomatics and Urban Spatial InformaticsSchool of Geomatics and Urban Spatial InformaticsSchool of Geomatics and Urban Spatial InformaticsWith the development of satellite remote sensing technology, image classification task, as the basis of remote sensing data interpretation, has received wide attention to improving accuracy and robustness. At the same time, in-depth learning technology has been widely used in remote sensing and has a far-reaching impact. Since the existing image classification methods ignore the feature that the general image semantics are the same as the semantics of a single pixel, this paper presents an algorithm that uses the semantics of an image to achieve high-precision image classification. Based on the idea of partial substitution for global, this algorithm designs a split result voting mechanism and builds a Vgg-Vote network model. This mechanism votes on the semantically segmented result of an image and uses the maximum filtering function to select the category containing the most significant number of pixels as the prediction category of the image. Experiments on UC Merced Land-User complete datasets and five types of incomplete datasets with varying degrees of interference, including noise, data occlusion and loss, show that the Vote mechanism dramatically improves the classification accuracy, robustness and anti-jamming capability of Vgg-Vote.http://dx.doi.org/10.1080/17538947.2022.2142306deep learningimage classificationrobustnessremote sensing imagevote mechanism |
| spellingShingle | Jianghong Zhao Xin Wang Xintong Dou Yingxue Zhao Zexin Fu Ming Guo Ruiju Zhang A high-precision image classification network model based on a voting mechanism International Journal of Digital Earth deep learning image classification robustness remote sensing image vote mechanism |
| title | A high-precision image classification network model based on a voting mechanism |
| title_full | A high-precision image classification network model based on a voting mechanism |
| title_fullStr | A high-precision image classification network model based on a voting mechanism |
| title_full_unstemmed | A high-precision image classification network model based on a voting mechanism |
| title_short | A high-precision image classification network model based on a voting mechanism |
| title_sort | high precision image classification network model based on a voting mechanism |
| topic | deep learning image classification robustness remote sensing image vote mechanism |
| url | http://dx.doi.org/10.1080/17538947.2022.2142306 |
| work_keys_str_mv | AT jianghongzhao ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT xinwang ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT xintongdou ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT yingxuezhao ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT zexinfu ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT mingguo ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT ruijuzhang ahighprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT jianghongzhao highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT xinwang highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT xintongdou highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT yingxuezhao highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT zexinfu highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT mingguo highprecisionimageclassificationnetworkmodelbasedonavotingmechanism AT ruijuzhang highprecisionimageclassificationnetworkmodelbasedonavotingmechanism |