Classification and Grading the Level of Paddy Leaf Diseases Using Multi Classifiers
Agriculture is the main backbone for most of the developing/developed countries; Agriculture production itself
is the main feed for ever growing populations and it is the major source of income for the rural people/farmers especially in
India. In India farmers are called “the backbone of India”. The main aim of the proposed system is to detect and classify the
diseases in paddy leafs. Paddy Diseases Classification comprises of two steps: first one is Detection, Extraction and
Segmentation of diseases part by using Weiner, Adaptive histogram techniques as a pre-processing techniques, Two-
Threshold Binary Decomposition by using Otsu algorithms for thresholding. Secondly, Feature extraction, Classification
and Grade the level of disease by using boundary detection technique, Support Vector Machine (SVM) and Fuzzy logic
classifiers respectively. In proposed system we are classifying three paddy leaf diseases by using above steps, the diseases
we considered for classification are leaf blast, brown spot and Sheath blight. The proposed system has been experimentally
tested for our own dataset and results achieved are encouraging.
Keywords— Otsu, SVM, Fuzzy Logic, Leaf Blast, Brown Spot, Sheath Blight.