Detection of Diseases in Sugarcane Using Image Processing Techniques
K. Thilagavathi
Abstract
India is one of the largest producers of sugarcane and ranks second in the world. The cropping season and duration of sugarcane depends on the varieties and sowing time. The time taken for its maturity is between 12 and 18 months. With high sensitivity to the environment, it easily gets affected by numerous diseases and pests. If the affected plant is not identified and taken adequate measures at the right time, it will affect the whole yield. The present study focuses on detecting various diseases in sugarcane leaves using the image processing techniques and developing a web application for the farmers to detect the major diseases of sugarcane as well. The system collects the images of the leaves and processes by means of Adaptive Histogram Equalization (AHE) superseded by segmentation using k means clustering algorithm. Using Gray Level Co-occurrence Matrix (GLCM) and Principal Component Analysis (PCA), statistical characteristics such as variance, skewness, standard deviation, mean, and covariance are extracted. Finally, the detection and classification are implemented using Support Vector Machine (SVM) that results the average accuracy value of 95%. If any variety disease is identified, the required control measures are also suggested.