Disease Classification of Paddy leaves using HSI Feature Extraction and SVM Technique

The value of paddy is strongly related to the quality, types and sizes of paddy without any damage or disease of that leaves. Hence detecting the disease or damage area of the leaf is very important to improve the utilization rate. Though the crop production is well grown there is still lagging in visual inspection of diseases. Although it is done manually, it is not accurate in all the times. So, there is a need for technique to detect the diseases. The system proposed is based on this method which can detect the diseases using artificial neural network and support vector machine classification. The main contribution of this approach is the support vector machine classification. This system involves image acquisition, converting the RGB images into HSI image where morphological process is used for removing noise. The block level feature extraction is used for extracting the features like mean and standard deviation. Finally, it is classified and compared both ANN and SVM approach for more accuracy and less execution time. For more accuracy, the stem cells samples can be used for detecting the disease.