Image analysis – Image transformation or preprocessing – Image storage or retrieval
Reexamination Certificate
2000-08-04
2004-11-09
Mehta, Bhavesh M. (Department: 2625)
Image analysis
Image transformation or preprocessing
Image storage or retrieval
C382S282000, C382S278000, C382S209000, C382S171000, C358S538000, C358S520000, C358S463000
Reexamination Certificate
active
06816633
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to an image retrieval apparatus and method, e.g., an apparatus and method for retrieving similar images.
BACKGROUND OF THE INVENTION
In a method of retrieving similar images using an image database storing a plurality of images, first a number of images are prepared, the feature quantities of these images are calculated, and the images and their feature quantities are registered together in the image database. Next, a search or target image for which a search is to be conducted is prepared, the feature quantity of this search image is calculated through a method similar to that used when the feature quantities were registered in the image database, an image the feature quantity whereof is nearest that of the calculated feature quantity of the search image is retrieved from the image database and this image is adopted as the similar image.
The technique described above, however, has certain shortcomings.
First, in a case where a search is conducted for images resembling an image that originally had a portion that is unnecessary, such as a black or white border surrounding the image, feature quantities are calculated with regard to the unnecessary border portion as well. As a consequence, the feature quantities of images having the same borders are very similar to the feature quantity of the target image and, hence, even images that are not that similar to the target image except for the presence of the surrounding border are retrieved as similar images.
Second, feature quantities which originally include unnecessary portions as well are calculated and registered in the image database in a manner similar to the case where images are registered in the image database, as a result of which images that actually resemble the target image are not retrieved.
In order to avoid the problems mentioned above, a method that has been proposed involves creating anew an image from which a border unnecessary for feature extraction has been removed from the start, and calculating the feature quantity from the newly created image. However, manipulating the image actually obtained is laborious, the original image must be saved separately and managing these images is extremely troublesome.
SUMMARY OF THE INVENTION
Accordingly, an object of the present invention is to make it possible to easily calculate a feature quantity from which unnecessary portions in an image have been eliminated, thereby making possible the more effective retrieval of similar images.
According to the present invention, the foregoing object is attained by providing an image retrieval method for retrieving images that resemble a specified image from a plurality of images, comprising: a decision step of deciding an unnecessary area in the specified image; a calculation step of calculating a feature quantity upon removing the unnecessary area, which has been decided at the decision step, from the specified image; and a retrieval step of retrieving resembling images from the plurality of images using the feature quantity calculated at the calculation step.
Further, according to the present invention, the foregoing object is attained by providing an image retrieval apparatus for retrieving images that resemble a specified image from a plurality of images, comprising: decision means for deciding an unnecessary area in the specified image; calculation means for calculating a feature quantity upon removing the unnecessary area, which has been decided by the decision means, from the specified image; and retrieval means for retrieving resembling images from the plurality of images using the feature quantity calculated by the calculation means.
Other features and advantages of the present invention will be apparent from the following description taken in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the figures thereof.
REFERENCES:
patent: 5014330 (1991-05-01), Kobayashi et al.
patent: 5185809 (1993-02-01), Kennedy et al.
patent: 5229589 (1993-07-01), Schneider
patent: 5264838 (1993-11-01), Johnson et al.
patent: 5579471 (1996-11-01), Barber et al.
patent: 5712474 (1998-01-01), Naneda
patent: 5825922 (1998-10-01), Pearson et al.
patent: 5880858 (1999-03-01), Jin
patent: 5893095 (1999-04-01), Jain et al.
patent: 6084978 (2000-07-01), Taylor et al.
patent: 0 287 113 (1988-10-01), None
W. Niblack, “SlideFinder : A Tool For Browsing Presentation Graphics Using Content-Based Retrieval”; Content-Based Access Of Image And Video Libraries, 1999 (Cbaival '99); Proceedings, IEEE Workshop, Fort Collins, CO., USA, Jun. 22, 1999, p. 114-118.
Seyoon Jeong et al., An Effective Method For Combining Multiple Features Of Image Retrieval: Proceedings of IEEE, IEEE Region 10 Conference, Tencon 99, Multimedia Technology For Asia-Pacific Information Infrastructure (CAT. No. 99CH 37030), pp. 982-985 vol. 2,XP010368443 1999 Piscataway, NJ, IEEE, USA, ISBN 0-7803-0573906.
Djeraba C. et al., “Digital Information Retrieval”, Proceeding of the 6th Int'l. Conf. on Information And Knowledge Management, CIKM '97, Las Vegas, Nov. 10-14, 1997, pp. 185-192.
Djeraba C. et al., “Retrieve of Images by Content”, Intelligence And Systems, 1996., IEEE, Int'l. Joint Symposia On Rockville Md. USA, Nov. 4-5, 1996, Los Alamitos, CA, pp. 261-267.
Enokida Miyuki
Kusama Kiyoshi
Matsumoto Kentaro
Yamamoto Kunihiro
Canon Kabushiki Kaisha
Fitzpatrick ,Cella, Harper & Scinto
Kassa Yosef
Mehta Bhavesh M.
LandOfFree
Image retrieval apparatus and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image retrieval apparatus and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image retrieval apparatus and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3319725