An Efficient Color Method for Retrieving Color-based Images

Abstract

Image retrieval plays a vital role in many of the areas. CBIR is content-based image retrieval for browsing images from a large dataset. Image retrieval is used in many of the applications such as image processing, pattern recognition, military, medical fields and forensic fields. In the proposed image retrieval method, color method is used to extract the features and retrieve similar images based on similarity of features. In the color method, HSV color histogram method is used to extract the color features. After extracting, the features are compared with the database images and distance is calculated using the Euclidian distance method. Finally similar images are retrieved according to the user satisfaction. Performance of the system is calculated using the processing time, execution time, precision and recall rate metrics.

Keywords: CBIR (content based image retrieval), color-HSV, processing time,
execution time, recall, precision