English

Abstract

Tomato crop is primarily infected by various common diseases like Bacterial Canker, bird's-eye fruit spots, Bacterial Spot, Chlorosis, Curly Top, Early Blight, Fusarium Wilt, Gray Leaf, Gray Mold Rot, Leaf Mold, Leaf Roll and Leaf Curl, Powdery mildew, Septoria Leaf Spot, Tobacco Mosaic Virus, Verticillium Wilt. The presented work describes a algorithm for different disease detection based on the infected images of various tomato plants. Images of the infected tomato plants are captured by closed circuit CCD cameras to cover approximately 5 sq. meter area that could acquire good quality images of tomato crop. The acquired images are in jpeg format and are converted to gray scale image. The gray scale image are the enhanced and made noise free. The Otsu algorithm is applied in order to get the thresholded image. The segmentation techniques based in pixel neighborhood are applied to get the segmented leaf and infected part of the leaf. The methods evolved in this system are both image processing and soft computing technique applied on number of diseased tomato plant images. The tomato images are acquired by using a CCD camera of approx. 3 M-Pixel resolution in 24-bits color resolution. The images are then transferred to PC and represented in MATLAB software. The RGB image is then segmented using K-means algorithm for segmentation of fungus in the tomato crop. The segmented fungus part is now analyzed for its percentage presence.