Detection and Prevention of Tea Leaf Diseases by using K-NN Algorithm

  • Ms. Payal Mhaiske, Prof. K.V.Warkar


All around the world tea is popular beverage and in India the cultivation of tea plays a vital role. The proper growth of tea leaves leading to its reduction of hindering for the production of tea because many diseases affect on the growth of tea leaves. But, if identified the disease at the early age then it would solve all the problems through the pruning of the disease leaves to prevent the spread further of the disease. An image processing is the best option to solve this problem to detect and diagnose the disease. The main goal of this research to developed an image processing system for identify and classify the six most widespread tea leaves diseases from healthy tea leaf. Disease identification is the first step in this step there are many methods are used for leaf disease identification. In this paper, K-NN classifier (K-NN) is used for disease recognition. During the classification analyzed the thirteen features. Thus features are used to find the most suitable match for the disease. An image is every time uploaded into a K-NN database.  When the new picture is uploaded into a system the most suitable match is found and then recognized the disease. In this approach the number of features compared by the K-NN classifier, which retains an accuracy of more than 96%. This also speed up the identification process, with each tea leaf image taking time is 200ms less processing time compared to the previous research by using K-NN, thus ensuring in given time frame the greater number of leaves can be processed. Thus proposed system increase the efficiency of the detection, identification, classification and prevention process will enable the tea industry in India to become globally more competitive for reducing the losses which are suffered due to the leaf diseases and thus increasing the overall rate of tea production.