Classification of Intact Cashew Grading System with Fuzzy Logic

This paper proposed an intelligent fuzzy logic based computer vision system for classification of whole cashew kernels by assigning the grades. Proposed approach is divided into various phases. Image acquisition phase captures images of the cashew kernel under investigation. Pre-processing phase smoothes acquired image to handle distortion. The cashew part (region of interest) is differentiated from the background part of the image in segmentation phase. In this system color along with other morphological features such as length, width and thickness are considered as essential features, and extracted in the feature extraction phase. Obtained quantitative information is fuzzified and taken as input to the fuzzy inference system (FIS) during the fuzzy classification phase. Final grade of the cashew kernel is decided based on the result of the classification.