Content based image retrieval: A past, present and new feature descriptor

Abstract

A Technique used for automatic retrieval of images in a large database that perfectly matches the query image is called as Content Based Image Retrieval (CBIR) system. Image querying refers to the predicament of finding objects that are pertinent to a user inquiry within image databases. The paper discusses the fundamental concept of image retrieval on the basis of features like color, texture, and shape. The system retrieves an image when the similarity between the query image and database images is observed. In the concluding section, we propose a new approach based on texture classification which helps to improve the accuracy and efficiency of CBIR system. The CBIR technology has been used in various applications such as military, industrial area, crime prevention, medical diagnosis, education system etc.