Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2000-07-03
2004-09-21
Pardo, Thuy N. (Department: 2175)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000, C707S793000, C382S128000, C382S305000
Reexamination Certificate
active
06795818
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method of searching multimedia data, and more particularly to a method of searching multimedia data more accurately by setting initial weights of features in images through a feedback algorithm.
2. Background of the Related Art
Recently, the digital image signal processing technology has been developing rapidly and has been applied in many fields. Some of these fields include a search system for automatically editing only a face of a specific character in a moving picture file of a movie or drama; a security system for permitting access only to those registered in the system; and a search system for searching a particular data from an image or video detected by a detecting system. In any application field, the performance of a system essentially depends on the accuracy and speed of detecting or searching a desired object. Accordingly, various image searching methods have been proposed in the related art.
One image search system which detects a degree of similarity with an image to be searched using features such as color, texture, or shape is disclosed in U.S. Pat. No. 5,579,471 entitled “An Image Query System and Method.” In a given search, the importance of a feature may vary depending upon a reference image to be searched and within one particular feature such as the color, the importance of a feature element such as the red or green color may also vary. However, this search system does not take into consideration the different importance of features or feature elements of an image to be searched.
In another searching method entitled “Virage image search engine” (www.virage.com), a user directly inputs the level of importance for features such as a color, texture and shape by assigning weight values. Although an image may be searched according to an importance of a feature using this method, it may be difficult for a user to determine the weights of features.
Therefore, Yong Rui in “Relevance feedback techniques in interactive” SPIE Vol.3312, discloses a method in which images similar to a reference image are found and the importance of features or weights for features are automatically obtained by calculating the similarities among the found images. However, the weight importance information is not maintained after a search for a specific image is finished and must be calculated for each image search, even for a same image.
Finally, “Using relevance feedback in content based image metasearch,” IEEE Internet Computing, pp. 59~69, July~August 1998 discloses a technology in which an image is automatically fed back when searching the image to learn the weight for features in the image. Thereafter, the learned weights and the image are tabled. In the above system, although images learned by feedback may be effectively searched, an image that has not been learned cannot be effectively searched even if weights are used.
Accordingly, “Using relevance feedback in content based image metasearch” also discloses a technology in which images are grouped and a reference image that has not been learned is searched using a learned weight of another image if the learned weight belongs to the same group as that of the reference image. However, there is a limitation to an effective search of an image using the above system as the weight of features in even the same group depends on each image.
Moreover, learning the weights after a few number of feedback would deteriorate the performance of the search system and the reliability of the search results. To obtain accurate weights, more than a given number of times of feedback should be used.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to solve at least the problems and disadvantages of the related art.
An object of the present invention is to provide a more effective method of searching multimedia data.
Another object of the present invention is to provide a method of searching multimedia data, in which weights of features in a specific image are automatically learned by grouping all images stored in a search system, giving initial weights to the grouped images to search and classify the images, determining errors from the classified results, and re-sorting the images using automatic feedback.
A further object of the present invention is to provide a multimedia data structure for an effective multimedia data search.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and advantages of the invention may be realized and attained as particularly pointed out in the appended claims.
To achieve the objects and in accordance with the purposes of the invention, as embodied and broadly described herein, a method of searching multimedia data comprises grouping the multimedia data in a database of a search system; searching and classifying the grouped multimedia data using initial weights of features in the grouped multimedia data; receiving at least one feedback reference multimedia data depending on a degree of error of the classified multimedia data; and updating weights of features in the fed back multimedia data.
In another embodiment, a method of searching multimedia data comprises searching for a reference image using initial weights; receiving user feedback on similar or dissimilar images; setting group information for a corresponding class using fed back information; updating the weights using fed back data; re-searching for the reference image using the updated weights; automatically determining at least one data using a degree of error for the searched data and the set group information; feeding the determined data back; and updating the weights using the fed back data.
Also, a multimedia data structure for use in a multimedia data search according to the present invention comprises a global information which represents specific multimedia data and spatial information which represents a feature in the multimedia data, wherein weights of the features include a type weight descriptor which represents importance of the feature element, an element weight descriptor which represents importance of elements depicted in one feature, and a position weight descriptor which represents important information from the spatial information.
REFERENCES:
patent: 5579471 (1996-11-01), Barber et al.
patent: 5642502 (1997-06-01), Driscoll
patent: 5872859 (1999-02-01), Gur et al.
patent: 5999664 (1999-12-01), Mahoney et al.
patent: 6078914 (2000-06-01), Redfern
patent: 6269361 (2001-07-01), Davis et al.
patent: 6278989 (2001-08-01), Chaudhuri et al.
patent: 6408293 (2002-06-01), Aggarwal et al.
patent: 6463431 (2002-10-01), Schmitt
patent: 6529901 (2003-03-01), Chaudhuri et al.
Yong Rui et al. “Relevance Feedback Techniques in Interactive Content-Based Image Retrieval” Beckman Institute and Dept. of Computer Science, University of Illinois, pp. 1-12.
Yong Rui et al. “Relevance Feedback: A Power Tool for Interactive Content-Based Retrieval” IEEE Transactions on Circuits and System for Video Technology, vol. 8, No. 5, Sep. 1998, pp. 644-655.
Ana B. Benitez et al. “Using Relevance Feedback In Content-Based Image Metasearch” IEEE Internet Computing, Jul.-Aug., 1998, pp. 59-69.
Fleshner & Kim LLP
LG Electronics Inc.
Pardo Thuy N.
LandOfFree
Method of searching multimedia data does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of searching multimedia data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of searching multimedia data will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3229798