Image Recognition using Texture and Color
Rajivkumar Mente, B.V. Dhandra, Gururaj Mukarambi · 2014
Content–Based Image Retrieval (CBIR) technique uses visual contents to search images in a large scale image databases according to the users ’ choice. The recognition accuracy depends on the training data set, the potentiality of features and the classifier used. The visual contents of an image such as color, shape, texture etc. are used in CBIR to retrieve the image. Texture is one of the important feature used in CBIR system. It is semi-repetitive arrangements of pixels. Entropy is a statistical measure of randomness that can be used to characterize the texture of the input image. In this paper color and texture feature entropy are combined to form a feature vector for kNN classifier. The algorithm is tested on a database of 2732 images from 6 different fruit classes. The higher recognition and retrieval accuracy is 97.85 % for the proposed algorithm. General Terms Your general terms must be any term which can be used for general classification of the submitted material such as Pattern Recognition, Security, Algorithms et. al.