Plant Leaf Disease Detection Using Adaptive Neuro-Fuzzy Classification

The paper deals with classification of different types of diseases of tomato and brinjal/eggplant. The patterns of the diseases are considered as a feature. It may be possible that the diseases are recognized by its texture patterns. A method that uses the texture patterns of the diseases in pure grayscale is applied for feature extraction purpose. A dedicated GLCM matrix is used to compute the features. The ANFIS based classification model is used for disease recognition by classification. The pattern based features with ANFIS recognition gives accuracy of 90.7% and 98.0% for TPDS 1.0 and BPDS 1.0 datasets respectively.