Automatic target recognition method for multitemporal remote sensing image
The traditional target recognition method for the remote sensing image is difficult to accurately identify the specified targets from the massive remote sensing image data. Based on the theory of multitemporal recognition, an automatic target recognition method for the remote sensing image is propos...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
De Gruyter
2020-06-01
|
| Series: | Open Physics |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/phys-2020-0015 |
| _version_ | 1819102107062501376 |
|---|---|
| author | Shu Chang Sun Lihui |
| author_facet | Shu Chang Sun Lihui |
| author_sort | Shu Chang |
| collection | DOAJ |
| description | The traditional target recognition method for the remote sensing image is difficult to accurately identify the specified targets from the massive remote sensing image data. Based on the theory of multitemporal recognition, an automatic target recognition method for the remote sensing image is proposed in this article. The proposed recognition method includes four modules: automatic segmentation of multitemporal remote sensing image, automatic target extraction of multitemporal remote sensing image, automatic processing of multitemporal remote sensing image, and automatic recognition of multitemporal remote sensing image. The automatic segmentation of the image target is introduced. The effectiveness of the segmentation technology is verified through the kernel function bandwidth algorithm. Linear feature extraction is used to extract the segmented image. The image extraction processing is described, which includes image profile analysis, image preprocessing, image feature analysis, the region of interest localization, image enhancement processing, recognition processing, and result output. According to the theory of pattern recognition, three different feature recognition images are given, which are partial separable recognition, weakly separable recognition, and fully separable recognition, and then, a new image recognition method is designed. To verify the practical application effect of the recognition method, the proposed method is compared with the traditional recognition method. Experimental results show that the proposed method can accurately identify the specified objects from the massive remote sensing image data and has a high potential for development. This article has an important guiding significance for image recognition. |
| first_indexed | 2024-12-22T01:29:18Z |
| format | Article |
| id | doaj.art-8ac8b11e8b3d46088d01d26d839d83b7 |
| institution | Directory Open Access Journal |
| issn | 2391-5471 |
| language | English |
| last_indexed | 2024-12-22T01:29:18Z |
| publishDate | 2020-06-01 |
| publisher | De Gruyter |
| record_format | Article |
| series | Open Physics |
| spelling | doaj.art-8ac8b11e8b3d46088d01d26d839d83b72022-12-21T18:43:32ZengDe GruyterOpen Physics2391-54712020-06-0118117018110.1515/phys-2020-0015phys-2020-0015Automatic target recognition method for multitemporal remote sensing imageShu Chang0Sun Lihui1Ecological Civilization Research Center, Chinese Research Academy of Environmental Sciences, Beijing, 100012, ChinaEcological Civilization Research Center, Chinese Research Academy of Environmental Sciences, Beijing, 100012, ChinaThe traditional target recognition method for the remote sensing image is difficult to accurately identify the specified targets from the massive remote sensing image data. Based on the theory of multitemporal recognition, an automatic target recognition method for the remote sensing image is proposed in this article. The proposed recognition method includes four modules: automatic segmentation of multitemporal remote sensing image, automatic target extraction of multitemporal remote sensing image, automatic processing of multitemporal remote sensing image, and automatic recognition of multitemporal remote sensing image. The automatic segmentation of the image target is introduced. The effectiveness of the segmentation technology is verified through the kernel function bandwidth algorithm. Linear feature extraction is used to extract the segmented image. The image extraction processing is described, which includes image profile analysis, image preprocessing, image feature analysis, the region of interest localization, image enhancement processing, recognition processing, and result output. According to the theory of pattern recognition, three different feature recognition images are given, which are partial separable recognition, weakly separable recognition, and fully separable recognition, and then, a new image recognition method is designed. To verify the practical application effect of the recognition method, the proposed method is compared with the traditional recognition method. Experimental results show that the proposed method can accurately identify the specified objects from the massive remote sensing image data and has a high potential for development. This article has an important guiding significance for image recognition.https://doi.org/10.1515/phys-2020-0015multi-temporalremote sensing imageimage processingimage segmentationtarget recognitionautomatic recognition |
| spellingShingle | Shu Chang Sun Lihui Automatic target recognition method for multitemporal remote sensing image Open Physics multi-temporal remote sensing image image processing image segmentation target recognition automatic recognition |
| title | Automatic target recognition method for multitemporal remote sensing image |
| title_full | Automatic target recognition method for multitemporal remote sensing image |
| title_fullStr | Automatic target recognition method for multitemporal remote sensing image |
| title_full_unstemmed | Automatic target recognition method for multitemporal remote sensing image |
| title_short | Automatic target recognition method for multitemporal remote sensing image |
| title_sort | automatic target recognition method for multitemporal remote sensing image |
| topic | multi-temporal remote sensing image image processing image segmentation target recognition automatic recognition |
| url | https://doi.org/10.1515/phys-2020-0015 |
| work_keys_str_mv | AT shuchang automatictargetrecognitionmethodformultitemporalremotesensingimage AT sunlihui automatictargetrecognitionmethodformultitemporalremotesensingimage |