Object-Oriented Crop Classification Using Time Series Sentinel Images from Google Earth Engine

• Published in: Remote sensing (Basel, Switzerland) Vol. 15; no. 5; p. 1353
• Main Authors: Xue, Hanyu, Xu, Xingang, Zhu, Qingzhen, Yang, Guijun, Long, Huiling, Li, Heli, Yang, Xiaodong, Zhang, Jianmin, Yang, Yongan, Xu, Sizhe, Yang, Min, Li, Yafeng
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.03.2023
• Subjects: Accuracy; Algorithms; China; Classification; Cloud computing; Clustering; Crop identification; Crops; Fertility; Google Earth Engine; Image classification; Image segmentation; Industrial parks; Internet; Land use; Land use classification; object-oriented; Pixels; Precipitation; Radar; random forest classification; Remote sensing; Rice; Satellite imagery; Satellites; Sentinel images; SNIC algorithm; support vector machine classification; Support vector machines; Time series; time series analysis; Wheat; Mongolia; China
• DOI: 10.3390/rs15051353
• ISSN: 2072-4292

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.