Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Diagnostic Model for Wheat Leaf Conditions Using Image Features and a Support Vector Machine

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  Transactions of the ASABE. 59(5): 1041-1052. (doi: 10.13031/trans.59.11434) @2016
Authors:   Keming Du, Zhongfu Sun, Ying Li, Feixiang Zheng, Jinxiang Chu, Yifeng Su
Keywords:   Diagnostic model, Image processing, Plant disease, Precision agriculture, Support vector machine.

Abstract. This study presents a diagnostic model for wheat leaf conditions (i.e., healthy, powdery mildew, stripe rust, and leaf rust) based on image processing and support vector machine (SVM) technologies. The model follows a process consisting of in-field acquisition, automatic classification, and ongoing detection. This methodology uses various color, texture, and shape features to distinguish between healthy and diseased wheat leaves. First, acquired RGB-colored digital images were preprocessed to segment suspected disease spots. Second, the color, texture, and shape features were extracted from the segmented images. Different combinations of feature vectors were then used as inputs for the SVM, which was coupled with four types of kernel functions. Finally, sample training experiments were performed to identify the optimal SVM classifiers. The results indicate that the SVM classifier with a radial basis kernel function and a selected combina-tion of feature vectors significantly outperformed the other algorithms, achieving a high overall recognition accuracy of 97.73%. This model could be used as an effective tool to identify wheat leaf conditions and has the potential for real-time diagnostic applications under field conditions.

(Download PDF)    (Export to EndNotes)