Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine
The resulting maps of land use classification obtained by pixel-based methods often have salt-and-pepper noise, which usually shows a certain degree of cluttered distribution of classification image elements within the region. This paper carries out a study on crop classification and identification...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/5/1353 |
| _version_ | 1797614471987855360 |
|---|---|
| author | Hanyu Xue Xingang Xu Qingzhen Zhu Guijun Yang Huiling Long Heli Li Xiaodong Yang Jianmin Zhang Yongan Yang Sizhe Xu Min Yang Yafeng Li |
| author_facet | Hanyu Xue Xingang Xu Qingzhen Zhu Guijun Yang Huiling Long Heli Li Xiaodong Yang Jianmin Zhang Yongan Yang Sizhe Xu Min Yang Yafeng Li |
| author_sort | Hanyu Xue |
| collection | DOAJ |
| description | The resulting maps of land use classification obtained by pixel-based methods often have salt-and-pepper noise, which usually shows a certain degree of cluttered distribution of classification image elements within the region. This paper carries out a study on crop classification and identification based on time series Sentinel images and object-oriented methods and takes the crop recognition and classification of the National Modern Agricultural Industrial Park in Jalaid Banner, Inner Mongolia, as the research object. It uses the Google Earth Engine (GEE) cloud platform to extract time series Sentinel satellite radar and optical remote sensing images combined with simple noniterative clustering (SNIC) multiscale segmentation with random forest (RF) and support vector machine (SVM) classification algorithms to classify and identify major regional crops based on radar and spectral features. Compared with the pixel-based method, the combination of SNIC multiscale segmentation and random forest classification based on time series radar and optical remote sensing images can effectively reduce the salt-and-pepper phenomenon in classification and improve crop classification accuracy with the highest accuracy of 98.66 and a kappa coefficient of 0.9823. This study provides a reference for large-scale crop identification and classification work. |
| first_indexed | 2024-03-11T07:11:53Z |
| format | Article |
| id | doaj.art-c8a92bf8965147b996517b7ec4b93083 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T07:11:53Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-c8a92bf8965147b996517b7ec4b930832023-11-17T08:31:55ZengMDPI AGRemote Sensing2072-42922023-02-01155135310.3390/rs15051353Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth EngineHanyu Xue0Xingang Xu1Qingzhen Zhu2Guijun Yang3Huiling Long4Heli Li5Xiaodong Yang6Jianmin Zhang7Yongan Yang8Sizhe Xu9Min Yang10Yafeng Li11Key Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaSchool of Agricultural Engineering, Jiangsu University, Zhenjiang 212013, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaTianjin Development and Demonstration Center for High-Quality Agricultural Products, Tianjin 301508, ChinaTianjin Development and Demonstration Center for High-Quality Agricultural Products, Tianjin 301508, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaKey Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, ChinaThe resulting maps of land use classification obtained by pixel-based methods often have salt-and-pepper noise, which usually shows a certain degree of cluttered distribution of classification image elements within the region. This paper carries out a study on crop classification and identification based on time series Sentinel images and object-oriented methods and takes the crop recognition and classification of the National Modern Agricultural Industrial Park in Jalaid Banner, Inner Mongolia, as the research object. It uses the Google Earth Engine (GEE) cloud platform to extract time series Sentinel satellite radar and optical remote sensing images combined with simple noniterative clustering (SNIC) multiscale segmentation with random forest (RF) and support vector machine (SVM) classification algorithms to classify and identify major regional crops based on radar and spectral features. Compared with the pixel-based method, the combination of SNIC multiscale segmentation and random forest classification based on time series radar and optical remote sensing images can effectively reduce the salt-and-pepper phenomenon in classification and improve crop classification accuracy with the highest accuracy of 98.66 and a kappa coefficient of 0.9823. This study provides a reference for large-scale crop identification and classification work.https://www.mdpi.com/2072-4292/15/5/1353Sentinel imagesobject-orientedSNIC algorithmrandom forest classificationsupport vector machine classificationGoogle Earth Engine |
| spellingShingle | Hanyu Xue Xingang Xu Qingzhen Zhu Guijun Yang Huiling Long Heli Li Xiaodong Yang Jianmin Zhang Yongan Yang Sizhe Xu Min Yang Yafeng Li Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine Remote Sensing Sentinel images object-oriented SNIC algorithm random forest classification support vector machine classification Google Earth Engine |
| title | Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine |
| title_full | Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine |
| title_fullStr | Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine |
| title_full_unstemmed | Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine |
| title_short | Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine |
| title_sort | object oriented crop classification using time series sentinel images from google earth engine |
| topic | Sentinel images object-oriented SNIC algorithm random forest classification support vector machine classification Google Earth Engine |
| url | https://www.mdpi.com/2072-4292/15/5/1353 |
| work_keys_str_mv | AT hanyuxue objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT xingangxu objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT qingzhenzhu objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT guijunyang objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT huilinglong objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT helili objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT xiaodongyang objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT jianminzhang objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT yonganyang objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT sizhexu objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT minyang objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine AT yafengli objectorientedcropclassificationusingtimeseriessentinelimagesfromgoogleearthengine |