Identification of Disease Types on Tea – Plant Varieties Based Image Processing With K-Nearest Neighbor Method

Abstract

Tea is a plant typed Camelia sinesis, frequently consumed by many people. It can cure various diseases and can be enjoyed to relax. Within the traded tea leaves, the quality of tea leaves determine the tea beverage. Many factors influence quality of tea plants. One of them is biotic disturbance caused by macro-organism and micro-organism. Disease problems could be noticed when the farmers plant them on the field. Generally, the detections of disease and pest on the plants can be done in various ways but the method used in this case was HSV (Hue, Saturation, Value) to detect the colors. Classification method in this system used KNN (K-Nearest Neighbor) by classifying the system based on the closes data with the subject. The disease classification on this observation consisted of 3: tea leaf chicken pox, caterpillar roller, and tea mites. To compare which one is diseased plant to the healthy one, this research had classified the leaf diseases by using microscope with accuracy level of normal eye 100%, dealing with the chicken pox disease of the leaf 71.4%, and 28.5% for empoasca sp with successful level 85.7%, and error level 14.2%. Dealing with caterpillar roller pest, the successful level was 92.8% with error level 7.2%.