30-11-2012, 02:40 PM
A FAST CLUSTERING ALGORITHM IN IMAGE SEGMENTATION
USING FUZZY - ALGORITHM
ABSTRACT
This project focuses on the problem of image clustering and its relationship to image database management. Image clustering and categorization is a means for high-level description of image content. The goal is to and a mapping of the archive images into classes (clusters) such that the set of classes provide essentially the same information about the image archive as the entire image-set collection. The generated classes provide a concise summarization and visualization of the image content that can be used for different tasks related to image database management.
EXISTING SYSTEM
In recent years there has been a growing interest in developing effective methods for searching large image databases based on image content. The two main approaches for image database interface, applied to existing systems are: “search-by-query” and “browsing”.
Most approaches to image database management have focused on the
“search-by-query” method. This method typically requires that users provide an example image for the query. The example image can be one of the existing images in the database or can be composed by the user. The database is then searched exhaustively for images which are most similar to the query.
PROPOSED SYSTEM
Image archive clustering is an important step for efficiently handling large image databases; it can be applied both to the “search-by-query” and to the browsing environment. In a “search-by-query” process, the query image can be initially compared with all the cluster centers. The subset of clusters that has the largest similarity to the query image is chosen, followed by comparing the query image with all the images within the subset of clusters. Search efficiency is improved due to the fact that the query image is not compared exhaustively to all the images in the database.
Applications of Image Clustering
Object recognition: recognize objects from image
Optical character recognition (characters, words from document image)
Face recognition
Fingerprint recognition
Medical image processing – diagnosis
e.g. cancerous / healthy cell
Industrial automation
Content based image retrieval
e.g. searching for an object class from a video database
MODULE DESCRIPTION
General Model of Mage Segmentation
Image segmentation means to discriminate different areas that are provided with particular meaning in the image according to gray level, color or geometric properties of the image, in which these areas are mutually disjoint, and each area meet conformity of given areas. Take image of an identical object for instance, it is often necessary to separate pixel dots that belong to this object in the image from those of the other objects in the image, and the out separated segmental area should meanwhile meet the following requirements
Image Segmentation Algorithm
Primarily there’re several popular image segmentation Threshold-selection-based image segmentation algorithm is to ascertain discrepancy in terms of gray level between target and the background image, and discriminate the image into a combination between target area and the background area(s), which both are of different gray level. This algorithm has very strong advantage in discriminating scenery that is of considerably strong contrast between the target area and the background (areas). Calculation of this algorithm is considerably simple. Disjoint or non-overlapping areas could be defined by closed and connected edges. Therefore it is a most effective and practical technology for image segmentation.
Process of fuzzy C- means clustering algorithm
Fuzzy C- means clustering algorithm utilizes bi-level iterative to calculate minimum of objective function. The internal level is used for calculating new clustering centre and updating fuzzy subjection-level matrix, and the external level is for judging if the algorithm has been converged to estimated/scheduled threshold. After finishing the iterative, we can know generic subjection-level of certain pixel to certain clustering centre according to generated fuzzy subjection-level matrix, and determine generic category of the pixel by the size of the matrix
Hardware Specification
Machine Type - P4 or above
Processor - 1 GHZ
Monitor - Color Monitor
RAM - 256 MB
Hard disk - 60 GB or more
Key Board - 1049 Keys
Drive - Floppy Drive
Software Specification
Operating System - Windows XP
Front End Tool - Visual Studio 2005(asp.net)
Data base tool - ADO.net
IIS - 5.1version
Back End Tool - Sql Server 2005
Report - Crystal Data report