Analysis of Content-Based Retrieval Techniques

CS580W08 RESEARCH PAPER TITLE & OUTLINE Bali, Deepak

RESEARCH PAPER TITLE: Analysis of Content-Based Retrieval Techniques

THESIS STATEMENT: This paper will present and do comparative analysis of three approaches to texture retrieval: content-based retrieval using heuristic search; texture retrieval based on a non-parametric measure for multivariate distributions; and content-based retrieval in gene expression databases.

RESEARCH PAPER OUTLINE

I. Introduction

A. What is content-based retrieval?

1. Definition

B. What is the problem with texture representation?

1. Robust Texture Representation

2. High-True Dimensionality

II. Comparative Analysis of Three Studies

A. Heuristic Search

B. Non-parametric Measure for Multivariate Distributions

C. Gene Expression Databases

III. Results and Conclusion

Analysis of Content-Based Retrieval Techniques

Introduction

Research in the field of content-based retrieval has primarily been focused on image, video and audio information [3]. An efficient strategy for retrieving texture images from large texture databases must be introduced and studied within a distributional framework. The exponential growth in the field of multimedia imaging and video databases has consequently resulted in extensive research for efficient and reliable retrieval methods [1]. Content-based image retrieval (CBIR), which is also termed as query by image content (QBIC) and content-based visual information retrieval (CBVIR) is the function of the computer visualization to the image retrieval problem, that is, the difficulty of searching for digital images in extremely large databases. "Content-based" literally means that the search should analyze the actual contents of the image. The word 'content' in this situation might refer to colors, shapes, textures, or any other information that can be derived from the image. Without the capability to examine...