FEATURE SELECTION FOR IMAGE RETRIEVAL BASED ON EVOLUTIONARY COMPUTATION

Abstract

In this paper we implemented feature selection for content based image retrieval using evolutionary computation. In this system, we used feature extraction techniques for color, texture and shape. The three techniques are used for feature extraction such as color moment, Gabor filter, and Edge histogram descriptor. To reduce the dimensionality and find best optimal features from feature set using feature selection based on two evolutionary computations i.e. Genetic algorithm, and Binary Bat Algorithm. These subset features are divided into similar image classes using k-means clustering algorithm for fast retrieval and improve the computational time. We compared these two algorithms with different parameters i.e. precision, recall and computational time of image retrieval. The experimental result shows feature selection using BBA reduces the time for retrieval and also increases the retrieval precision, thus it gives better and faster results as compared to feature selection using GA. In this method selects different combinations of features which user retrieves more similar images. The CBIR system is more efficient and better performs using feature selection based on BBA.