Object-Based Image Classification of Summer Crops with Machine Learning Methods

Remote Sensing

José M. Pe༚¹  Pedro A. Gutiérrez³  César Hervás-Martínez³  Johan Six⁴  Richard E. Plant² 

[1]Institute for Sustainable Agriculture, IAS-CSIC, P.O. Box 4084, E-14080 Córdoba, Spain

[2] E-Mail:

[3]Department of Plant Sciences, University of California, Davis, CA 95616, USA

[4] E-Mail:

[5]Department of Computer Science and Numerical Analysis, University of Cordoba, Campus de Rabanales, E-14071 Córdoba, Spain

[6] E-Mails:

[7]Department of Environmental Systems Sciences, Swiss Federal Institute of Technology, ETH-Zurich, CH-8092 Zurich, Switzerland

[8] E-mail:

关键词: agriculture;    ASTER satellite images;    object-oriented image analysis;    hierarchical classification;    neural networks;   

DOI  :  10.3390/rs6065019

来源: mdpi

【 摘 要 】

The strategic management of agricultural lands involves crop field monitoring each year. Crop discrimination via remote sensing is a complex task, especially if different crops have a similar spectral response and cropping pattern. In such cases, crop identification could be improved by combining object-based image analysis and advanced machine learning methods. In this investigation, we evaluated the C4.5 decision tree, logistic regression (LR), support vector machine (SVM) and multilayer perceptron (MLP) neural network methods, both as single classifiers and combined in a hierarchical classification, for the mapping of nine major summer crops (both woody and herbaceous) from ASTER satellite images captured in two different dates. Each method was built with different combinations of spectral and textural features obtained after the segmentation of the remote images in an object-based framework. As single classifiers, MLP and SVM obtained maximum overall accuracy of 88%, slightly higher than LR (86%) and notably higher than C4.5 (79%). The SVM+SVM classifier (best method) improved these results to 89%. In most cases, the hierarchical classifiers considerably increased the accuracy of the most poorly classified class (minimum sensitivity). The SVM+SVM method offered a significant improvement in classification accuracy for all of the studied crops compared to the conventional decision tree classifier, ranging between 4% for safflower and 29% for corn, which suggests the application of object-based image analysis and advanced machine learning methods in complex crop classification tasks.

【 授权许可】

CC BY   
© 2014 by the authors; licensee MDPI, Basel, Switzerland