Recognition and Classification of Similar Looking Food Grain Images using Artificial Neural Networks

Abstract

This paper presents an recognition and classification of similar looking food grain images using artificial neural networks. Schemes for visual classification usually proceed in two stages. First, features are extracted which represents the image and Second, a classifier is applied to the extracted features to reach a decision regarding the represented type of images. We have considered four pairs of eight different types of similar looking commonly available Indian food grain images namely Jira, Badesoup, Mongdaal Woduddal. Ragi, Mustard, Soya, and Alasandi. The algorithms are developed to extract 18 color and 27 texture features. A Back Propagation Neural Network (BPNN) is used to classify and recognize the Food grain image samples using three different types of feature sets, viz, color, texture, combination of both color and texture features. The study reveals that the combination of color and texture features are out performed the individual color and texture features in recognition and classification of different similar looking food grain images samples.

Keywords

• Similar Looking Food Grain Images
• Feature Extraction
• Artificial Neural Networks