Call of Papers for Current Volume **************** OnLine Submission of Paper

Volume & Issue no: Volume 3, Issue 8, August 2014


Author Name:
Harjeet Kaur, Jagbir Singh Gill
ABSTRACT Widespread cardinality of images and paintings has made traditional keyword based search, an inefficient method for retrieval of required input image data. When we input an image into the database, then Content-Based Image Retrieval (CBIR) system gets the similar images from a large database for that query image. Implementation of CBIR can be done using features like color, texture and shape, these features are called low level features. In this thesis, using a feed-forward back propagation neural network, classification of images in CBIR system is proposed. At first, the neural network is trained about the features of images in the database. The image features that are used in this CBIR system are color histogram, GRB Pattern and Side loop level normalized r2. There are two steps of BPNN 1. Training 2. Testing the training is carried out using BPNN. On the bases of training, resultant image comes out. So training is the best and crucial step in CBIR system for better recognition of image from database when query image is inputted. The proposed method shows the promising results in terms of precision and recall of image retrieval. The proposed method is applied on the African dataset. The whole result simulation is taken place in MATLAB environment. KEYWORDS:- CBIR, Neural Network, image
Cite this article:
Harjeet Kaur, Jagbir Singh Gill , " CONTENT BASED IMAGE RETRIEVAL SYSTEM BASED ON NEURAL NETWORKS" , International Journal of Application or Innovation in Engineering & Management (IJAIEM) , Volume 3, Issue 8, August 2014 , pp. 171-175 , ISSN 2319 - 4847.
Full Text [PDF]                           Back to Current Issue