Content Based Image Retrieval using Multi Channel Decoded Local Binary Patterns with Relevant Feedback
Mumthaz Muhammed M, Amitha Mathew · Zenodo (CERN European Organization for Nuclear Research) · 2018
The tremendous growth in the multimedia applications has resulted in exponential growth in the size of the image database. The proper management, retrieval and indexing of relevant images based on a context may affect the proper functioning of the systems in various image related applications. Content based image processing is an essential and efficient tool which rectifies the issue by putting more emphasis on content of the image rather than texts, tags or annotations. The content of the image can be extracted by using various image features including color, texture and shape. The selection of proper features in various applications is important as it may affect the efficiency of the entire system. The proposed system is to develop a content based retrieval mechanism which will provide semantically close query results quickly and efficiently with the help of relevant feedback. The relevant feedback can increase the level of perception and quality of query results during the successive levels. The comparison of the images are done based on the distance measures such as Euclidean distance, squared Euclidean distance, L1 distance, chi-square distance, earth movers distance and cosine distance.