Content-based image retrieval apparatus and method via relevance feedback by using fuzzy integral

• US 6,744,935 B2
• Filed: 12/14/2000
• Issued: 06/01/2004
• Est. Priority Date: 11/02/2000
• Status: Expired due to Term

• Current Assignee: Korea Telecom (KT Corporation)
• Original Assignee: Korea Telecom (KT Corporation)
• Inventors: Choi, Young-Sik, Kim, Jin-han, Yoon, Eun-Il, Lee, Sang-Hong
• Primary Examiner(s): MARIAM, DANIEL G
• Application Number:
  US09/735,920
• Publication Number:
  US 20020051576A1
• Time in Patent Office: 1,265 Days
• Field of Search: 382/173, 382/181, 382/190, 382/209, 382/217, 382/218, 382/220, 382/305, 382/403, 707/1-10
• US Class Current: 382/305
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 18/2178 based on feedback of a supervisor
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 18/40 Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

The invention relates to a content-based image retrieval apparatus and method via relevance feedback by using fuzzy integral and a computer readable record medium storing programs for realizing the apparatus and method. According to the invention, it is provided a content-based image retrieval apparatus and method in which image features are associated by using fuzzy integral and parameters necessary for the fuzzy integral are adjusted in similarity calculations via relevance feedback so as to improve search results. The method comprises the steps of: measuring similarities according to feature between images stored in a database and a new query image; associating the similarities measured according to feature by using the fuzzy integral to measure similarities according to the fuzzy integral; bringing images sequentially in the order of higher similarity according to the finally obtained similarities and transmitting a search result; and recalculating similarities according to feature via relevance feedback about the query result to output converted images.

Claims

1. A content-based image retrieval apparatus via relevance feedback by using fuzzy integral, the apparatus comprising:

image input means for receiving images from a user;

feature extraction means for extracting image features necessary for image similarity operation from the images inputted via said image input means;

first storage means for storing the image features extracted by said feature extraction means;

thumb nail generating means for dividing the images inputted via said image input means to generate thumb nails;

second storage means for storing the thumb nails generated by said thumb nail generating means;

similarity calculation means for calculating similarities between a query image and the images based on the image features stored on said first storage means, to thereby generate individual similarities for each image feature;

similarity measuring means for performing fuzzy integral of the individual similarities calculated via said similarity calculation means, to thereby measure final image similarities;

query processing means for interpreting the query image from the user to measure similarities, and for sequentially calling and transmitting the thumb nails stored on said second storage means according to the calculated similarities;

user interface means for performing relevance feedback about the image data transmitted from said query processing means, and for transmitting selective information according to relevance feedback to said query processing means

wherein the final image similarities are measured based on an equation as;

2. The apparatus as recited in claim 1, wherein the fuzzy integral is operated by studying and initializing the fuzzy measurement about individual image features and then studying special interactions to perform the fuzzy measurement necessary for fuzzy via relevance feedback.

3. The apparatus as recited in claim 2, wherein the fuzzy studying of the individual features is performed by establishing fuzzy rules based on representative similarity differences and by measuring the fuzzy integral,

wherein the fuzzy rules are established based on equations as:

4. The apparatus as recited in claim 2, wherein the studying of the special interactions is performed by obtaining the fuzzy measurement whereby the following equation is minimized, and studying while satisfying monotonicity which is a fuzzy measurement condition, wherein the equation is:

5. A content-based image retrieval method via relevance feedback by using fuzzy integral, comprising:

measuring similarities according to feature between images stored in a database and a new query image;

associating the similarities measured according to feature by using the fuzzy integral to measure similarities according to the fuzzy integral;

bringing images sequentially in the order of higher similarity according to the finally obtained similarities and transmitting a search result; and

recalculating similarities according to feature via relevance feedback about the query result to output converted images

wherein the final image similarities are measured based on an equation as;

6. The method as recited in claim 5, wherein the fuzzy integral is operated by studying and initializing the fuzzy measurement about individual features and then studying special interactions to perform the fuzzy measurement necessary for fuzzy via relevance feedback.

7. The method as recited in claim 6, wherein the fuzzy studying of the individual features is performed by establishing fuzzy rules based on representative similarity differences and by measuring the fuzzy integral,

wherein the fuzzy rules are established based on equations as:

8. A computer readable record medium storing instructions for executing a content-based image retrieval method, comprising:

measuring similarities according to feature between images stored in a database and a new query image;

associating the similarities measured according to feature by using the fuzzy integral to measure similarities according to the fuzzy integral;

bringing images sequentially in the order of higher similarity according to the finally obtained similarities and transmitting a search result; and

recalculating similarities according to feature via relevance feedback about the query result to output converted images

wherein the final image similarities are measured based on an equation as;