The main objective of our work is to make the limited user log play the fullest role.

This probabilistic scheme is to make images with higher similarity to most positive examples more likely similar to the query image.

Content Based Image Retrieval

INTRODUCTION

DUE to the rapidly growing amount of digital image data on the Internet and in digital libraries, there is a great need for large image database management and effective image retrieval tools. Content-based image retrieval (CBIR) is the set of techniques for searching for similar images from an image database using automatically extracted image features.

Most current content-based image retrieval systems are still incapable of providing users with their desired results. The major difficulty lies in the gap between lowlevel image features and high-level image semantics. To address the problem, this study reports a framework for effective image retrieval by employing a novel idea of memory learning. It forms a knowledge memory model to store the semantic information by simply accumulating user-provided interactions.

DRAWBACKS

Incapability of capturing semantics. Most RF techniques in CBIR absolutely copy ideas from textural information retrieval. Scarcity and imbalance of feedback examples. Very few users are willing to go through endless iterations of feedback with the hopes of getting the best results. Lack of the memory mechanism.

A feedback knowledge memory model is presented to gather the users' feedback information during the process of image search and feedback. It is efficient and can be simply implemented. A learning strategy based on the memorized information is a proposed. It can estimate the hidden semantic relationships among images. Consequently, this technique could address the problem of user log sparsity in a certain extent.

During the interactive process, a seamless combination of normal RF (low-level feature based) and the memory learning (semantics based) is proposed to improve the retrieval performance. A semantics-based image annotation propagation scheme is proposed using both memorized and learned semantics. In contrast with existing algorithms of propagating annotation by visual similarity, its precision is much better.

we briefly review the related work. we present the feedback knowledge memory model. The learning strategy to estimate the hidden semantics. The image retrieval framework by memory learning. The experimental results are shown in.

1. Feedback Knowledge Memory Model

2. Semantic Correlation Analysis by a Learning Strategy

Image Semantic Clustering
Image Authoritative Rank Hidden Semantic Correlation Between Two Images

3. Image Retrieval Framework by the Memory Learning

Image Similarity Measure
Relevance Feedback Integrating SVM With Memory Learning Image Annotation and Annotation Learning

Propagation

 Review of Related Work Feedback Knowledge Memory Model

Semantic Correlation Analysis by a Learning Strategy Image Retrieval Framework by the Memory Learning Experimental Results

The idea of long-term learning in CBIR is borrowed from the work of collaborative filtering and link structure analysis in the web information retrieval. Unlike the collaborative filtering, many web search engines search for web pages by the link structure analysis.
Pages

that are co-cited by a certain page are likely to relate to the same topic, and pages that are often visited in succession by a certain user are possibly similar.

In order to supply effective image retrieval to users, this paper has presented a new memory learning framework in which low-level featurebased RF and semantics-based memory learning are combined to help each other to achieve better retrieval performance. There are two novel characteristics that distinguish the memory learning framework from the existing RF techniques. The proposed framework is easy to implement and can be efficiently incorporated into an image retrieval system.

The proposed framework is easy to implement and can be efficiently incorporated into an image retrieval system. Experimental evaluations on a large-scale image database have already shown very promising results. Our future work will investigate the possibility to develop more sophisticated and theoretical learning schemes.

M. Flickner, H. Sawhney, and W. Niblack, "Query by image and video content: The QBIC system,"

IEEE Computer,

A. P. Penland, R. W. Picard, and S. Sclaroff, "Photobook: Content-based manipulation of image databases, H. Zhang and Z. Su, "Relevance feedback in CBIR," presented at the Int. Workshop on Visual Databases. X. S. Zhou and T. S. Huang, "Relevance feedback in image retrieval: A comprehensive review,"