Land use classification through fusion of remote sensing images and multi-source data
In the land classification problem, although the application of land data and remote sensing technology can provide a lot of data, the difference of data quality, data format, and data sources lead to the difficulty of land classification. Therefore, a land use classification method based on remote...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2025-06-01
|
| Series: | Open Geosciences |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/geo-2025-0820 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849686700876365824 |
|---|---|
| author | Guo Zhiqian Ren Yushui Li Xin Ma Kang Qian Shujun |
| author_facet | Guo Zhiqian Ren Yushui Li Xin Ma Kang Qian Shujun |
| author_sort | Guo Zhiqian |
| collection | DOAJ |
| description | In the land classification problem, although the application of land data and remote sensing technology can provide a lot of data, the difference of data quality, data format, and data sources lead to the difficulty of land classification. Therefore, a land use classification method based on remote sensing image and multi-source data was proposed. The multi-structure element binary morphology is used to carry out the corrosion operation on the mutation pixels in the remote sensing image to complete the denoising. Based on this, the chaotic leapfrog algorithm is used to enhance the denoised remote sensing image. Through the fusion of multi-source feature data, the spatial information of remote sensing image is combined with the attribute information of other data sources to extract spectral and shape features and complete the classification of land use. The experimental results show that the R
2 value of the proposed method is 0.97, the MAE value is 0.09, and the Kappa coefficient remains above 0.9. This indicates that the method can effectively enhance the features of land remote sensing images through remote sensing and multi-source data fusion, and has ideal accuracy for multi-class land use classification, which can achieve accurate classification of land use. |
| format | Article |
| id | doaj-art-64d8ca3d59c842fe9a1c419bcc1c6dc4 |
| institution | DOAJ |
| issn | 2391-5447 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Open Geosciences |
| spelling | doaj-art-64d8ca3d59c842fe9a1c419bcc1c6dc42025-08-20T03:22:37ZengDe GruyterOpen Geosciences2391-54472025-06-0117111810.1515/geo-2025-0820Land use classification through fusion of remote sensing images and multi-source dataGuo Zhiqian0Ren Yushui1Li Xin2Ma Kang3Qian Shujun4Shandong Provincial NO.4 Institute of Geological and Mineral Survey, Weifang, Shandong, 261021, ChinaWeifang Marine Development Research Institute, Weifang, Shandong, 261035, ChinaWeifang Marine Development Research Institute, Weifang, Shandong, 261035, ChinaWeifang Marine Development Research Institute, Weifang, Shandong, 261035, ChinaShandong Provincial NO.4 Institute of Geological and Mineral Survey, Weifang, Shandong, 261021, ChinaIn the land classification problem, although the application of land data and remote sensing technology can provide a lot of data, the difference of data quality, data format, and data sources lead to the difficulty of land classification. Therefore, a land use classification method based on remote sensing image and multi-source data was proposed. The multi-structure element binary morphology is used to carry out the corrosion operation on the mutation pixels in the remote sensing image to complete the denoising. Based on this, the chaotic leapfrog algorithm is used to enhance the denoised remote sensing image. Through the fusion of multi-source feature data, the spatial information of remote sensing image is combined with the attribute information of other data sources to extract spectral and shape features and complete the classification of land use. The experimental results show that the R 2 value of the proposed method is 0.97, the MAE value is 0.09, and the Kappa coefficient remains above 0.9. This indicates that the method can effectively enhance the features of land remote sensing images through remote sensing and multi-source data fusion, and has ideal accuracy for multi-class land use classification, which can achieve accurate classification of land use.https://doi.org/10.1515/geo-2025-0820remote sensing imagemulti-source dataland use classificationbinary morphology of multiple structural elementschaotic leapfrog algorithm |
| spellingShingle | Guo Zhiqian Ren Yushui Li Xin Ma Kang Qian Shujun Land use classification through fusion of remote sensing images and multi-source data Open Geosciences remote sensing image multi-source data land use classification binary morphology of multiple structural elements chaotic leapfrog algorithm |
| title | Land use classification through fusion of remote sensing images and multi-source data |
| title_full | Land use classification through fusion of remote sensing images and multi-source data |
| title_fullStr | Land use classification through fusion of remote sensing images and multi-source data |
| title_full_unstemmed | Land use classification through fusion of remote sensing images and multi-source data |
| title_short | Land use classification through fusion of remote sensing images and multi-source data |
| title_sort | land use classification through fusion of remote sensing images and multi source data |
| topic | remote sensing image multi-source data land use classification binary morphology of multiple structural elements chaotic leapfrog algorithm |
| url | https://doi.org/10.1515/geo-2025-0820 |
| work_keys_str_mv | AT guozhiqian landuseclassificationthroughfusionofremotesensingimagesandmultisourcedata AT renyushui landuseclassificationthroughfusionofremotesensingimagesandmultisourcedata AT lixin landuseclassificationthroughfusionofremotesensingimagesandmultisourcedata AT makang landuseclassificationthroughfusionofremotesensingimagesandmultisourcedata AT qianshujun landuseclassificationthroughfusionofremotesensingimagesandmultisourcedata |