REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING
Remote sensing image scene classification has gained remarkable attention, due to its versatile use in different applications like geospatial object detection, ground object information extraction, environment monitoring and etc. The scene not only contains the information of the ground objects, but...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2019-09-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-2-W16/279/2019/isprs-archives-XLII-2-W16-279-2019.pdf |
| _version_ | 1818681722098679808 |
|---|---|
| author | Y. Yao Y. Yao H. Zhao H. Zhao D. Huang D. Huang Q. Tan Q. Tan |
| author_facet | Y. Yao Y. Yao H. Zhao H. Zhao D. Huang D. Huang Q. Tan Q. Tan |
| author_sort | Y. Yao |
| collection | DOAJ |
| description | Remote sensing image scene classification has gained remarkable attention, due to its versatile use in different applications like geospatial object detection, ground object information extraction, environment monitoring and etc. The scene not only contains the information of the ground objects, but also includes the spatial relationship between the ground objects and the environment. With rapid growth of the amount of remote sensing image data, the need for automatic annotation methods for image scenes is more urgent. This paper proposes a new framework for high resolution remote sensing images scene classification based on convolutional neural network. To eliminate the requirement of fixed-size input image, multiple pyramid pooling strategy is equipped between convolutional layers and fully connected layers. Then, the fixed-size features generated by multiple pyramid pooling layer was extended to one-dimension fixed-length vector and fed into fully connected layers. Our method could generate a fixed-length representation regardless of image size, at the same time get higher classification accuracy. On UC-Merced and NWPU-RESISC45 datasets, our framework achieved satisfying accuracies, which is 93.24% and 88.62% respectively. |
| first_indexed | 2024-12-17T10:07:28Z |
| format | Article |
| id | doaj.art-c5dd65f32ae14ba896ce83c654ac7af9 |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-12-17T10:07:28Z |
| publishDate | 2019-09-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-c5dd65f32ae14ba896ce83c654ac7af92022-12-21T21:53:08ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342019-09-01XLII-2-W1627928410.5194/isprs-archives-XLII-2-W16-279-2019REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLINGY. Yao0Y. Yao1H. Zhao2H. Zhao3D. Huang4D. Huang5Q. Tan6Q. Tan7Department of Civil Engineering, Tsinghua University, Beijing 10084, China3S Center, Tsinghua University, Beijing 10084, ChinaDepartment of Civil Engineering, Tsinghua University, Beijing 10084, China3S Center, Tsinghua University, Beijing 10084, ChinaDepartment of Civil Engineering, Tsinghua University, Beijing 10084, China3S Center, Tsinghua University, Beijing 10084, ChinaDepartment of Civil Engineering, Tsinghua University, Beijing 10084, China3S Center, Tsinghua University, Beijing 10084, ChinaRemote sensing image scene classification has gained remarkable attention, due to its versatile use in different applications like geospatial object detection, ground object information extraction, environment monitoring and etc. The scene not only contains the information of the ground objects, but also includes the spatial relationship between the ground objects and the environment. With rapid growth of the amount of remote sensing image data, the need for automatic annotation methods for image scenes is more urgent. This paper proposes a new framework for high resolution remote sensing images scene classification based on convolutional neural network. To eliminate the requirement of fixed-size input image, multiple pyramid pooling strategy is equipped between convolutional layers and fully connected layers. Then, the fixed-size features generated by multiple pyramid pooling layer was extended to one-dimension fixed-length vector and fed into fully connected layers. Our method could generate a fixed-length representation regardless of image size, at the same time get higher classification accuracy. On UC-Merced and NWPU-RESISC45 datasets, our framework achieved satisfying accuracies, which is 93.24% and 88.62% respectively.https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-2-W16/279/2019/isprs-archives-XLII-2-W16-279-2019.pdf |
| spellingShingle | Y. Yao Y. Yao H. Zhao H. Zhao D. Huang D. Huang Q. Tan Q. Tan REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING |
| title_full | REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING |
| title_fullStr | REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING |
| title_full_unstemmed | REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING |
| title_short | REMOTE SENSING SCENE CLASSIFICATION USING MULTIPLE PYRAMID POOLING |
| title_sort | remote sensing scene classification using multiple pyramid pooling |
| url | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLII-2-W16/279/2019/isprs-archives-XLII-2-W16-279-2019.pdf |
| work_keys_str_mv | AT yyao remotesensingsceneclassificationusingmultiplepyramidpooling AT yyao remotesensingsceneclassificationusingmultiplepyramidpooling AT hzhao remotesensingsceneclassificationusingmultiplepyramidpooling AT hzhao remotesensingsceneclassificationusingmultiplepyramidpooling AT dhuang remotesensingsceneclassificationusingmultiplepyramidpooling AT dhuang remotesensingsceneclassificationusingmultiplepyramidpooling AT qtan remotesensingsceneclassificationusingmultiplepyramidpooling AT qtan remotesensingsceneclassificationusingmultiplepyramidpooling |