Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data
Urbanization is a substantial contributor to anthropogenic environmental change, and often occurs at a rapid pace that demands frequent and accurate monitoring. Time series of satellite imagery collected at fine spatial resolution using stable spectral bands over decades are most desirable for this...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2072-4292/10/3/471 |
| _version_ | 1818321897368059904 |
|---|---|
| author | Haobo Lyu Hui Lu Lichao Mou Wenyu Li Jonathon Wright Xuecao Li Xinlu Li Xiao Xiang Zhu Jie Wang Le Yu Peng Gong |
| author_facet | Haobo Lyu Hui Lu Lichao Mou Wenyu Li Jonathon Wright Xuecao Li Xinlu Li Xiao Xiang Zhu Jie Wang Le Yu Peng Gong |
| author_sort | Haobo Lyu |
| collection | DOAJ |
| description | Urbanization is a substantial contributor to anthropogenic environmental change, and often occurs at a rapid pace that demands frequent and accurate monitoring. Time series of satellite imagery collected at fine spatial resolution using stable spectral bands over decades are most desirable for this purpose. In practice, however, temporal spectral variance arising from variations in atmospheric conditions, sensor calibration, cloud cover, and other factors complicates extraction of consistent information on changes in urban land cover. Moreover, the construction and application of effective training samples is time-consuming, especially at continental and global scales. Here, we propose a new framework for satellite-based mapping of urban areas based on transfer learning and deep learning techniques. We apply this method to Landsat observations collected during 1984–2016 and extract annual records of urban areas in four cities in the temperate zone (Beijing, New York, Melbourne, and Munich). The method is trained using observations of Beijing collected in 1999, and then used to map urban areas in all target cities for the entire 1984–2016 period. The method addresses two central challenges in long term detection of urban change: temporal spectral variance and a scarcity of training samples. First, we use a recurrent neural network to minimize seasonal urban spectral variance. Second, we introduce an automated transfer strategy to maximize information gain from limited training samples when applied to new target cities in similar climate zones. Compared with other state-of-the-art methods, our method achieved comparable or even better accuracy: the average change detection accuracy during 1984–2016 is 89% for Beijing, 94% for New York, 93% for Melbourne, and 89% for Munich, and the overall accuracy of single-year urban maps is approximately 96 ± 3% among the four target cities. The results demonstrate the practical potential and suitability of the proposed framework. The method is a promising tool for detecting urban change in massive remote sensing data sets with limited training data. |
| first_indexed | 2024-12-13T10:48:12Z |
| format | Article |
| id | doaj.art-e5c6003711ea4a01912c20c0b588b8f4 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-12-13T10:48:12Z |
| publishDate | 2018-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-e5c6003711ea4a01912c20c0b588b8f42022-12-21T23:50:02ZengMDPI AGRemote Sensing2072-42922018-03-0110347110.3390/rs10030471rs10030471Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat DataHaobo Lyu0Hui Lu1Lichao Mou2Wenyu Li3Jonathon Wright4Xuecao Li5Xinlu Li6Xiao Xiang Zhu7Jie Wang8Le Yu9Peng Gong10Ministry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaRemote Sensing Technology Institute (IMF), German Aerospace Center (DLR), Wessling 82234, GermanyMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaDepartment of Geological & Atmospheric Science, Iowa State University, Ames, IA 50014, USAMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaRemote Sensing Technology Institute (IMF), German Aerospace Center (DLR), Wessling 82234, GermanyState Key Lab of Remote Sensing Science, Jointly Sponsored by Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing 100101, ChinaMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaMinistry of Education Key Laboratory for Earth System Modeling, Department of Earth System Science, Tsinghua University, Beijing 100084, ChinaUrbanization is a substantial contributor to anthropogenic environmental change, and often occurs at a rapid pace that demands frequent and accurate monitoring. Time series of satellite imagery collected at fine spatial resolution using stable spectral bands over decades are most desirable for this purpose. In practice, however, temporal spectral variance arising from variations in atmospheric conditions, sensor calibration, cloud cover, and other factors complicates extraction of consistent information on changes in urban land cover. Moreover, the construction and application of effective training samples is time-consuming, especially at continental and global scales. Here, we propose a new framework for satellite-based mapping of urban areas based on transfer learning and deep learning techniques. We apply this method to Landsat observations collected during 1984–2016 and extract annual records of urban areas in four cities in the temperate zone (Beijing, New York, Melbourne, and Munich). The method is trained using observations of Beijing collected in 1999, and then used to map urban areas in all target cities for the entire 1984–2016 period. The method addresses two central challenges in long term detection of urban change: temporal spectral variance and a scarcity of training samples. First, we use a recurrent neural network to minimize seasonal urban spectral variance. Second, we introduce an automated transfer strategy to maximize information gain from limited training samples when applied to new target cities in similar climate zones. Compared with other state-of-the-art methods, our method achieved comparable or even better accuracy: the average change detection accuracy during 1984–2016 is 89% for Beijing, 94% for New York, 93% for Melbourne, and 89% for Munich, and the overall accuracy of single-year urban maps is approximately 96 ± 3% among the four target cities. The results demonstrate the practical potential and suitability of the proposed framework. The method is a promising tool for detecting urban change in massive remote sensing data sets with limited training data.http://www.mdpi.com/2072-4292/10/3/471urban mappingdeep learningrecurrent neural networktransfer learninglong time series |
| spellingShingle | Haobo Lyu Hui Lu Lichao Mou Wenyu Li Jonathon Wright Xuecao Li Xinlu Li Xiao Xiang Zhu Jie Wang Le Yu Peng Gong Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data Remote Sensing urban mapping deep learning recurrent neural network transfer learning long time series |
| title | Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data |
| title_full | Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data |
| title_fullStr | Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data |
| title_full_unstemmed | Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data |
| title_short | Long-Term Annual Mapping of Four Cities on Different Continents by Applying a Deep Information Learning Method to Landsat Data |
| title_sort | long term annual mapping of four cities on different continents by applying a deep information learning method to landsat data |
| topic | urban mapping deep learning recurrent neural network transfer learning long time series |
| url | http://www.mdpi.com/2072-4292/10/3/471 |
| work_keys_str_mv | AT haobolyu longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT huilu longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT lichaomou longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT wenyuli longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT jonathonwright longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT xuecaoli longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT xinluli longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT xiaoxiangzhu longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT jiewang longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT leyu longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata AT penggong longtermannualmappingoffourcitiesondifferentcontinentsbyapplyingadeepinformationlearningmethodtolandsatdata |