Automatic Image Retrieval Using Color Features

Abstract

Image retrieval is geared towards the development of methodologies for analyzing, interpreting cataloging and indexing image databases. In addition to their development, efforts are also being made to evaluate the performance of image retrieval systems [1]. The quality of response is heavily dependent on the choice of the method used to generate feature vectors and similarity measure for comparison of features. Most retrievals use low level features for the extraction of image. These low-level features primarily constitute color and texture. By processing the extracted features instead of the entire image, reduces the memory requirements as well as the computational time required to process the image. During the retrieval, features and descriptors of the query are compared to those of the images in the database in order to retrieve the best match image from database according to its distance from the query. An interactive image recommendation system, which firstly uses color histogram feature and GLCM texture feature to express image content finally based on a feature vectors stored in the database the similar images are retrieved. The regional feature is extracted using the GLCM technique in which the neighbor pixels is considered into account.
The image retrieval technology has been used in several day-to-day applications such as fingerprint identification, biodiversity information systems, digital libraries, crime prevention, medicine, historical research, biomedical field, etc.