Content Based Image Retrieval (CBIR)

ABSTRACT

Content Based Image Retrieval CBIR is becoming very popular because of the high demand for searching image databases of ever-growing size. Since speed and precision are important, we need to develop a system for retrieving images that is both efficient and effective.


The emergence of multimedia technology and the rapidly expanding image and video collections on the internet have attracted significant research efforts in providing tools for effective retrieval and management of visual data. Content Based Image retrieval CBIR is based on the availability of a representation scheme of image content. Thus CBIR can be used as a powerful tool for retrieving images from the database by utilizing the visual cues alone. Image descriptors may be visual features such as color, texture and shape.



The motivation of our approach is to design and implement an effective and efficient framework of image retrieval techniques, using a variety of visual features such as color and texture. When a query image is given to our system, these features are extracted from it and compared with those in the database based on similarity measure. Finally twenty images which are most relevant to the query image are retrieved. We implemented Histogram Quadratic Distance Measure as color similarity which is most efficient Thus we have implemented both color and texture giving efficiency to our system.





INTRODUCTION

Content-based image retrieval is one of the techniques for automatic retrieval of images from a database by color, texture etc. The features used for retrieval can be either primitive or semantic but the abstraction process must be predominantly automatic.
The goal of Content-Based Image Retrieval (CBIR) systems is to operate on collections of images and, in response to visual queries, extract relevant image. The application potential of CBIR for fast and effective image retrieval is enormous, expanding the use of computer technology to a management tool.

Existing Systems
IBM’s QBIC system is the first commercial CBIR system and probably the best known of all CBIR systems. QBIC supports users to retrieval images by color, shape and texture. QBIC provides several query methods: Simple, Multi-feature and Multi-pass. In the simple method, a query is processed using only one feature.


OUR APPROACH

Query image is taken as the input for processing and is normalized to 320*480 sizes. When our system receives the query message it passes the image and weights of the features to the feature extraction mechanism. After the features are extracted, the feature extraction mechanism sends the feature information of the query image to the similarity measure mechanism. According to the feature information, the similarity measure mechanism measures the similarity of the features information between the query image and the database images. Finally, our system gives the most relevant images as the output along with the query image.




Hardware Requirements:
Pentium 4 Processor
1 GB RAM
80 GB Hard Disk Space

Software Requirements:
Microsoft Windows Xp Professional.
Sun JDK 1.6