Grape leaf disease detection and classification using multi-class support vector machine

In the era of technology burst and usage of software as an alternative for the manual involvement for decision making, every field is trying to find its own comfort and cost cutting solutions in replacing software methods for best possible expert opinion. SVM, is initially proposed for binary classification technique, with simple manipulation can be used for a multiple class case. This project tries to attempt for improvement in classifying the leaf diseases. Most of the work until now involves extracting statistical features of RGB signal converted into LAB form[1]. HSI image has a reputation that the hue does not change even when the background light over the image changes. Hence few of the properties of HSI image are added to the database[2]. SVM is applied for classification for a larger space points. (properties)