文献类型： 外文期刊

作者： Luo Hong-xia ¹ ; Dai Sheng-pei ¹ ; Li Mao-fen ¹ ; Liu En-ping ¹ ; Zheng Qian ¹ ; Hu Ying-Ying ¹ ; Yi Xiao-ping ² ;

作者机构： 1.Chinese Acad Trop Agr Sci, Inst Sci & Tech Informat, Key Lab Pract Trop Crops Informat Technol Hainan, Haikou 571000, Hainan, Peoples R China

2.Minist Nat Resources China, Land Use Key Lab, Chinese Land Survey & Planning Inst, Beijing 100101, Peoples R China

3.Minist Agr & Rural Affairs, Key Lab Agr Remote Sensing, Beijing 100100, Peoples R China

关键词： mango plantations; GLCM texture; SVM; RF; GF-1

期刊名称：JOURNAL OF INTEGRATIVE AGRICULTURE （ 影响因子：2.848； 五年影响因子：2.979 ）

ISSN： 2095-3119

年卷期： 2020 年 19 卷 11 期

页码：

收录情况： SCI

摘要： Mango is a commercial crop on Hainan Island, China, that is cultivated to develop the tropical rural economy. The development of accurate and up-to-date maps of the spatial distribution of mango plantations is necessary for agricultural monitoring and decision management by the local government. Pixel-based and object-oriented image analysis methods for mapping mango plantations were compared using two machine learning algorithms (support vector machine (SVM) and Random Forest (RF)) based on Chinese high-resolution Gaofen-1 (GF-1) imagery in parts of Hainan Island. To assess the importance of different features on classification accuracy, a combined layer of four original bands, 32 gray-level co-occurrence (GLCM) texture indices, and 10 vegetation indices were used as input features. Then five different sets of variables (5, 10, 20, and 30 input variables and all 46 variables) were classified with the two machine learning algorithms at object-based level. Results of the feature optimization suggested that homogeneity and variance were very important variables for distinguishing mango plantations patches. The object-based classifiers could significantly improve overall accuracy between 2-7% when compared to pixel-based classifiers. When there were 5 and 10 input variables, SVM showed higher classification accuracy than RF, and when the input variables exceeded 20, RF showed better performances. After the accuracy achieved saturation points, there were only slightly classification accuracy improvements along with the numbers of feature increases for both of SVM and RF classifiers. The results indicated that GF-1 imagery can be successfully applied to mango plantation mapping in tropical regions, which would provide a useful framework for accurate tropical agriculture land management.