Content based Image Retrieval A Hybrid Model Approach

Abstract

With the rapid development of internet and mobile devices, people can easily obtain audio and newlinevideo information everywhere. Because of the exponential growth of multimedia information, newlinehow to efficiently retrieve and manage multimedia information from huge databases becomes an newlineimportant issue. Efficient image searching, browsing and retrieval tools are required by users newlinefrom various domains, including remote sensing, fashion, crime prevention, publishing, newlinemedicine, architecture, etc. content-based image retrieval have been the long-standing research newlinetopics in the field of computer vision, largely because they can help to solve a wide range of newlinevisual recognition and understanding problems. CBIR is an active research domain for more newlinethan 20 years. CBIR systems are complex retrieval platforms which combine multiple areas of newlineexpertise from computer vision and machine learning to information retrieval. The objective of newlinecontent based image retrieval is to develop techniques to automatically extract and retrieve newlinerelevant similar images from the huge database. newlineThe purpose of content-based image retrieval is to retrieve images based on their visual newlinecontents. Recently, focus is being given to Content-Based Image Retrieval (CBIR) due to the newlineremarkable growth in terms of the number and size of video collections and digital image found newlineon web. Content Based Image Retrieval is a process to retrieve a stored image from a database newlineby supplying an image as a query instead of text. As a result, developing optimized feature newlinerepresentations and similarity measures is important to the retrieval performance of a CBIR newlinesystem. A necessity for developing a successful image retrieval system is the extraction of newlinerepresentative features to describe the query image and the ones in the database. Due to the huge newlinenumber of images, it is important to develop effective and efficient methods for searching, newlineretrieving, and recognizing candidates within the expanding collections. newline