Designing a classifier for automatic detection of fungal diseases in wheat plant: By pattern recognition techniques

The most important factor for reduction in quality and quantity of wheat crop, is wheat plant disease. The purpose of this paper is designing a classifier for fungal diseases detection in wheat plants by pattern recognition techniques. Unhealthy regions were segmented by thresholding method and morphological operators and their texture, color and shape features were extracted. To reduce dimensionality of the features space, significant features are selected by minimal-redundancy-maximal-relevance criterion (mRMR). A radial basis function (RBF) neural network was employed to classify wheat diseases. According to the results, the proposed method could effectively detect and classify wheat diseases to an accuracy of 98.3%.