Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/38562
Title: Neural network based intelligent image analyzer and retrieval
Authors: Khoo, Jia Jun.
Keywords: DRNTU::Engineering::Computer science and engineering::Information systems::Information storage and retrieval
Issue Date: 2010
Abstract: There is an increasing number of digital video and images being captured from image capturing devices such as camera phones and digital camera and stored. It has hence become increasingly challenging if not impossible to cope with this huge growth of visual data. With the advent of the high speed internet, searching through an online image database will have to be more efficient than before as the traditional practice of tagging thousands of images and searching manually becomes highly inefficient. Numerous researches had been done in the field of Content Based Image Retrieval (CBIR). CBIR could be the key to removing the need for the tremendous amount of manual labour of annotating the images. The main goal of a CBIR system is to identify closely matched images and return them to the user accurately in the shortest possible time. In this proposed system, the use of the Haar wavelet transform with combined RGB and RgYb colour channels in neural networks had proven to be effective in retrieving images from the COREL 1k dataset. This scheme was developed and implemented in an image search system worked well when tested on images taken outside the COREL 1k dataset. Experiments were carried out to find out the performance of the proposed system.
URI: http://hdl.handle.net/10356/38562
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
SCE0188.pdf
  Restricted Access
1.75 MBAdobe PDFView/Open

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.