Fine-scale classification of horticultural crops using Sentinel-2 time-series images in Linyi country, China

作者：Riqiang Chen, Shuping Xiong, Na Zhang, Zehua Fan, Ning Qi, Yiguang Fan, Haikuan Feng, Xinming Ma, Hao Yang, Guijun Yang, Jinpeng Cheng · 发表于：Computers and Electronics in Agriculture · 年份：2025 · DOI：10.1016/j.compag.2025.110425 · 被引用次数：9 · 研究领域：Remote Sensing in Agriculture、Smart Agriculture and AI、Remote Sensing and Land Use

Satellite imagery holds great potential for crop mapping. However, the high degree of similarity of remote sensing features and fragmented plots of horticultural crops challenges their fine-scale classification. In this study, we mapped horticultural crops by fusing Sentinel-2 time-series images, object-based method and machine learning. Features including original reflectance, vegetation index (VI), and texture were first constructed from Sentinel-2 images. ReliefF feature selection techniques were then implemented to score the features according to their importance for the classification purposes. Finally, the classification performance of pixel-based and object-based methods was evaluated by combining them with Classification and Regression Tree (CART), Random Forest (RF), and Support Vector Machine (SVM) algorithms. The results indicate that both the segmentation methods yield favorable outcomes, with the pixel-based method achieving a slightly higher accuracy (OA = 83.82 %, Kappa = 0.76) compared with the object-based method (OA = 79.99 %, Kappa = 0.70), but the object-based method is able to delineate the boundaries of certain orchards in detail as well as to ensure the consistency of classification results within the orchard. In addition, the findings reveal that training a random forest model using all features leads to exceptional accuracy, with apple, peach, and persimmon exhibiting the most effective classification performance. The Producer Accuracy (PA) and user a...