Detection Of Disease On Plant Leaves Using Novel Structure Algorithm

Abstract

The production of Indian crops is affected with pests in large, which enters at the root level and on leaves as well. Plant diseases cause huge damage to crops in turn creates economic losses for the country. Through early diagnosis, identification of disease helps the production to improve remarkably. Large crops are damaged and left over every year due to the rapid infestation of insects. Performing early diagnosis is bit difficult on infected plant leaf and it is due to the symptoms of several resemblance diseases. New novel technique is proposed here to identify the types of diseases in plant leaves due to pests. Images of leaves affected by certain diseases are taken as samples for preprocessing based on the structured algorithm. The image is detected based on the looking edges and further it has been enhanced. Images detected by the edges will be taken in to advanced fuzzy k-means clustering for segmentation. Subsequently, the color features are extracted, then the processing of correlation, entropy, texture features such as energy, contrast, edges, etc are also performed. Then, the image features are compared with the ordinary leaf image. Finally, the exact disease detection and medical related diagnosis will be finalized based on the novel algorithm. The way of detection of disease on plant leaf is purely based on the advanced technique which is adopted here, when compared with previous techniques. The algorithm is framed and simulated in MATLAB.