Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China
Large-scale land-cover classification using a supervised algorithm is a challenging task. Enormous efforts have been made to manually process and check the production of national land-cover maps. This has led to complex pre- and post-processing and even the production of inaccurate mapping products...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/21/4194 |
| _version_ | 1797511948090212352 |
|---|---|
| author | Xuemei Zhao Danfeng Hong Lianru Gao Bing Zhang Jocelyn Chanussot |
| author_facet | Xuemei Zhao Danfeng Hong Lianru Gao Bing Zhang Jocelyn Chanussot |
| author_sort | Xuemei Zhao |
| collection | DOAJ |
| description | Large-scale land-cover classification using a supervised algorithm is a challenging task. Enormous efforts have been made to manually process and check the production of national land-cover maps. This has led to complex pre- and post-processing and even the production of inaccurate mapping products from large-scale remote sensing images. Inspired by the recent success of deep learning techniques, in this study we provided a feasible automatic solution for improving the quality of national land-cover maps. However, the application of deep learning to national land-cover mapping remains limited because only small-scale noisy labels are available. To this end, a mutual transfer network MTNet was developed. MTNet is capable of learning better feature representations by mutually transferring pre-trained models from time-series of data and fine-tuning current data. An interactive training strategy such as this can effectively alleviate the effects of inaccurate or noisy labels and unbalanced sample distributions, thus yielding a relatively stable classification system. Extensive experiments were conducted by focusing on several representative regions to evaluate the classification results of our proposed method. Quantitative results showed that the proposed MTNet outperformed its baseline model about 1%, and the accuracy can be improved up to 6.45% compared with the model trained by the training set of another year. We also visualized the national classification maps generated by MTNet for two different time periods to quantitatively analyze the performance gain. It was concluded that the proposed MTNet provides an efficient method for large-scale land cover mapping. |
| first_indexed | 2024-03-10T05:54:12Z |
| format | Article |
| id | doaj.art-c09579ddb984493782dd2cff7140ba9d |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T05:54:12Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c09579ddb984493782dd2cff7140ba9d2023-11-22T21:29:50ZengMDPI AGRemote Sensing2072-42922021-10-011321419410.3390/rs13214194Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of ChinaXuemei Zhao0Danfeng Hong1Lianru Gao2Bing Zhang3Jocelyn Chanussot4School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaLarge-scale land-cover classification using a supervised algorithm is a challenging task. Enormous efforts have been made to manually process and check the production of national land-cover maps. This has led to complex pre- and post-processing and even the production of inaccurate mapping products from large-scale remote sensing images. Inspired by the recent success of deep learning techniques, in this study we provided a feasible automatic solution for improving the quality of national land-cover maps. However, the application of deep learning to national land-cover mapping remains limited because only small-scale noisy labels are available. To this end, a mutual transfer network MTNet was developed. MTNet is capable of learning better feature representations by mutually transferring pre-trained models from time-series of data and fine-tuning current data. An interactive training strategy such as this can effectively alleviate the effects of inaccurate or noisy labels and unbalanced sample distributions, thus yielding a relatively stable classification system. Extensive experiments were conducted by focusing on several representative regions to evaluate the classification results of our proposed method. Quantitative results showed that the proposed MTNet outperformed its baseline model about 1%, and the accuracy can be improved up to 6.45% compared with the model trained by the training set of another year. We also visualized the national classification maps generated by MTNet for two different time periods to quantitatively analyze the performance gain. It was concluded that the proposed MTNet provides an efficient method for large-scale land cover mapping.https://www.mdpi.com/2072-4292/13/21/4194classificationdeep learningLandsatmultispectralnational land-cover mappingtransfer learning |
| spellingShingle | Xuemei Zhao Danfeng Hong Lianru Gao Bing Zhang Jocelyn Chanussot Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China Remote Sensing classification deep learning Landsat multispectral national land-cover mapping transfer learning |
| title | Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China |
| title_full | Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China |
| title_fullStr | Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China |
| title_full_unstemmed | Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China |
| title_short | Transferable Deep Learning from Time Series of Landsat Data for National Land-Cover Mapping with Noisy Labels: A Case Study of China |
| title_sort | transferable deep learning from time series of landsat data for national land cover mapping with noisy labels a case study of china |
| topic | classification deep learning Landsat multispectral national land-cover mapping transfer learning |
| url | https://www.mdpi.com/2072-4292/13/21/4194 |
| work_keys_str_mv | AT xuemeizhao transferabledeeplearningfromtimeseriesoflandsatdatafornationallandcovermappingwithnoisylabelsacasestudyofchina AT danfenghong transferabledeeplearningfromtimeseriesoflandsatdatafornationallandcovermappingwithnoisylabelsacasestudyofchina AT lianrugao transferabledeeplearningfromtimeseriesoflandsatdatafornationallandcovermappingwithnoisylabelsacasestudyofchina AT bingzhang transferabledeeplearningfromtimeseriesoflandsatdatafornationallandcovermappingwithnoisylabelsacasestudyofchina AT jocelynchanussot transferabledeeplearningfromtimeseriesoflandsatdatafornationallandcovermappingwithnoisylabelsacasestudyofchina |