Downloads: 118

Survey Paper | Information Technology | India | Volume 3 Issue 11, November 2014

A Survey on Adaptive Content Based Image Retrieval System Using Neural Network

Pratima R. Jadhav | Rashmi S. Phalnikar

Abstract: In several aspects of medical, technology, aerospace, bio-informatics, government organization, large collections of digital photographs are being created. A number of these collections are formed due to the product of digitizing existing collections of analogue images, diagrams, sketches, paintings, and documents. Generally, the only method of exploring these collections was by browsing or indexing of keywords, Digital photograph databases nevertheless, start the way to content-based searching. In this paper, we have present a method that has no previous knowledge about the image within the database, but retrieval is done considering the content information of the images likely to be called as content based image retrieval. Here we are trying to improve the image retrieval system for more accuracy and efficiency by using Radial basis Function neural network. This deals with multilayer feed forward network perception. By using this methodology we can easily find out exact relevant image according to the query provided by the user. The Scale Invariant Feature Transform (SIFT) is one of the most local feature detector and descriptors which is used in most of the vision software. In this paper regarding CBIR system we can utilize SIFT algorithm to extract the local features of the images. Also Back-propagation methodology, which is a managerial system for learning is utilized for training the neural network. We can use back-propagation for computation of errors in backward direction. Also k means clustering is applied to clustering Parameterize Gaussian function application. At the end actual and observed outputs are compared to calculating actual correct output.

Keywords: content based image retrieval, Radial basis Function, SIFT, and k means clustering

Edition: Volume 3 Issue 11, November 2014,

Pages: 2916 - 2919

Similar Articles with Keyword 'content'