Image analysis – Image compression or coding – Transform coding
Reexamination Certificate
1998-09-22
2003-05-20
Wu, Jingge (Department: 2723)
Image analysis
Image compression or coding
Transform coding
C375S240250
Reexamination Certificate
active
06567563
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
This application claims the benefit of Korean Application No. 97-76417, filed Dec. 29, 1997 in the Korean Patent Office, the disclosure of which is incorporated herein by reference.
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a video image searching method and apparatus, and more particularly, to a video image searching method and apparatus in which a dominant image element is extracted by a fractal video image compression method and the video image is partially restored.
2. Description of the Related Art
In a fractal video image compression method that assumes a video image is comprised of fixed points of various equations, a target video image is divided into non-overlapping range blocks. In addition, the target video image is divided into domain blocks which are larger than the range blocks and allowed to overlap, and a search for a transform linear equation corresponding to the fixed points is performed. In this way, rather than compressing the image data itself, the fractal video image compression method compresses the image made up by the transform linear equation. A conventional fractal video image compression circuit and method, as disclosed in U.S. Pat. Nos. 5,384,867 and 5,430,812, includes a controller which receives image data to process the image data into blocks, such as range blocks and domain blocks, and supplies the processed data to a plurality of transform circuits and feed circuits. Each transform circuit receives data from the controller and a feed circuit, and compares blocks through a parallel process to generate a fractal transform value in which the image data is expressed in a compressed form.
It is difficult, using conventional methods, to completely search for the video image using a video image data base, and therefore, in order to determine whether the video image data retrieved from the data base is appropriate for the desired image, it is necessary to use more video image data than the video image for which the search is to be made. As a result, the video image data retrieved from the video image data base must be restored into images, and pictures which are not necessarily required to determine whether the retrieved video image is appropriate for the desired video image are restored, which unnecessarily consumes a significant amount of time. Furthermore, adopting the complete restoration causes a considerable loss of time since it takes time for the retrieved video image data to be restored into a complete video image.
SUMMARY OF THE INVENTION
To solve the above problems, it is an object of the present invention to provide a video image searching method in which a video image is partially restored according to a predetermined video image restoring ratio, to thereby reduce the amount of searching time.
It is a further object of the present invention to provide a video image searching apparatus for performing the video image searching method.
Additional objects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Accordingly, objects of the invention are achieved by a method of searching for a video image from a plurality of image data stored in a data base that includes searching and retrieving from the data base one of the stored image data similar to the image, partially restoring the one retrieved image data, and comparing and evaluating the partially restored one retrieved image data with the image.
In a preferred embodiment, the method includes determining an image restoring ratio of retrieved video image data, and extracting a dominant image element from the video image. The extracting step includes compressing the video image through transform linear equations, determining a frequency of generation of fixed points of the linear equations according to values of decision elements of a matrix in an equation from which the fixed points are generated, and extracting the dominant image element according to a desired probability. In the preferred embodiment, the one retrieved image data corresponds to the extracted dominant image element according to the determined image restoring ratio.
Further objects of the invention are achieved by an apparatus for searching for an image in which an image restoring ratio determination unit determines an image restoring ratio, and an image element extracting unit partially restores the image. An image restoring unit partially restores image data corresponding to the partially restored image, according to the image restoring ratio, and a comparing unit compares the restored image data and the image.
In the preferred embodiment, the image element extracting unit includes a compression unit that compresses the image using a transform linear equation, and a determining unit that determines a frequency of generation of fixed points of the transform linear equation according to values of decision elements of a matrix in an equation from which the fixed points are generated. An extracting unit extracts the partially restored image according to a desired probability such that the frequency of generation of the fixed points corresponds to the probability.
Further objects of the invention are achieved by an apparatus for searching for an image in which an image restoring ratio determination unit determines an image restoring ratio, and a dominant image element extracting unit extracts a dominant image element from the image, using fractal image compression. A search unit retrieves image data from an image data base corresponding to the extracted dominant image element. An image restoring unit partially restores the retrieved image data from the search unit according to the image restoring ratio, and a comparing unit compares the restored image data and the image.
In the preferred embodiment, the dominant image element extracting unit includes a compression unit that compresses the image using a transform linear equation, and a determining unit that determines a frequency of generation of fixed points of the transform linear equation, according to values of decision elements of a matrix in an equation from which the fixed points are generated. An extracting unit extracts the dominant image element according to a desired probability, such that the frequency of generation of the fixed points corresponds to the probability.
Further objects of the invention are achieved by an apparatus for searching an image from a plurality of image data stored in a data base that includes a search unit to search and retrieve from the data base one of the stored image data similar to the image. A restoring unit partially restores the one retrieved image data, and a comparator compares the partially restored one retrieved image data with the image.
REFERENCES:
patent: 4941193 (1990-07-01), Barnsley et al.
patent: 5165103 (1992-11-01), Takeda et al.
patent: 5347600 (1994-09-01), Barnsley et al.
patent: 5553277 (1996-09-01), Hirano et al.
patent: 5761655 (1998-06-01), Hoffman
patent: 5862263 (1999-01-01), Kim et al.
patent: 6005679 (1999-12-01), Haneda
patent: 6266451 (2001-07-01), Charrier et al.
patent: 63-286974 (1988-11-01), None
patent: 7-322255 (1995-12-01), None
Menon “Content based image display from image daabases using spatio-temporal transfers and fractal analysis methods” Image Processing, Sep. 1996. Proc., International Conference on, vol. 3, pp. 863-866.*
Julie et al. “Digital Image indexing and retrieval by content using teh fractal transform for multimedia databases” Proc. IEEE international Forum on research and Technology Advances in Digital Libraries, May 1997 p. 2-12.*
Marie-Julie, JM&ESSAFI,H “Image Database Indexing and Retrieval Using the Fractal Transform” Lecture Notes in Computer Science, vol. 1242, pp. 169-182, 1997 (Hei 9-6-10, JICST).
Vrscay, Edward R., “A Hitchhiker's Guide to “Fractal-Based” Function Approximation and Image Compression”, Department of Applied Mathematics, Univers
Samsung Electronics Co,. Ltd.
Wu Jingge
LandOfFree
Video image searching method and apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Video image searching method and apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video image searching method and apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3014854