Method and apparatus for searching a digital image

• US 6,389,417 B1
• Filed: 06/29/1999
• Issued: 05/14/2002
• Est. Priority Date: 06/29/1999
• Status: Expired due to Term

• Current Assignee: Samsung Electronics Co. Ltd., Regents of the University of California (University of California)
• Original Assignee: Samsung Electronics Co. Ltd., Regents of the University of California (University of California)
• Inventors: Choi, Yang-lim, Shin, Hyun-doo, Manjunath, B. S.
• Primary Examiner(s): Coby, Frantz
• Application Number:
  US09/342,270
• Time in Patent Office: 1,050 Days
• Field of Search: 707/2, 707/3, 707/4, 707/5, 707/6
• US Class Current: 1/1
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

An apparatus for searching a digital image and a method therefor are provided. The digital image search apparatus includes a database processing unit for segmenting an image from the database or an image which is not stored in the database into homogeneous regions and analyzing and storing the segmented regions and an image search engine unit for searching for an image having a segmented region showing a characteristic similar to the characteristic of the query region. It is possible to perform detailed and efficient searching work since it is possible to perform a region-based search by means of segmented regions and a user can perform searching by selecting the characteristic and the position of the segmented regions.

Claims

1. A digital image search apparatus for searching for an image from a database in which a plurality of images are stored, comprising:

a database processing unit operative to segment an image from the database and an image which is not stored in the database into homogeneous query regions and to analyze and store a characteristic of the segmented regions; and

an image search engine unit operative to search for an image having a segmented region showing a characteristic similar to a characteristic of the query region.

2. The apparatus of claim 1, wherein the image search engine unit searches for a plurality of images having a segmented region showing a characteristic similar to the characteristic of the query region and searches for a region having a position similar to at least one of an absolute position or a relative position of the query region in the query image.

3. The apparatus of claim 1, wherein the characteristic of the region is at least one selected from a characteristic group comprising a color histogram showing the characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing the characteristic of shape, and a relative and absolute position of the region from anywhere within an entire image.

4. The apparatus of claim 2, wherein the characteristic of the region is at least one selected from a characteristic group comprising a color histogram showing the characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing the characteristic of shape, and a relative and absolute position of the region from anywhere within an entire image.

5. The apparatus of claim 1, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing the characteristics of colors, Gabor wavelet coefficients showing the characteristics of texture, a boundary description showing the characteristic of shape, and a relative and absolute position from anywhere within an entire image.

6. The apparatus of claim 2, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing the characteristics of colors, Gabor wavelet coefficients showing the characteristics of texture, a boundary description showing the characteristic of shape, and a relative and absolute position from anywhere within an entire image.

7. A digital image search apparatus for searching for an image in a database in which a plurality of images are stored, comprising:

a database processing unit, comprising;

region segmentation means for receiving an image and segmenting the image into homogeneous regions;

region characteristic analyzing means for analyzing a characteristic of the segmented regions; and

an image and region characteristic storing means for storing the characteristic of each image and region, and

an image search engine, comprising;

an inputting unit for receiving a query image from a user interface, determining whether the query image is in the database, and outputting the image;

a selecting unit responsive to the database processing unit for outputting the segmented region and the respective characteristic of the segmented region, the selecting unit also being connected to the user interface and operative to receive data defining a query region and a characteristic selected by the user;

a searching unit for searching an image having a segmented region showing a characteristic similar to a characteristic of the query region in the query image and in the region characteristic storing means and outputting searched image information; and

an outputting unit for outputting at least one searched image, where one of said at least one searched image is the image in the database corresponding to the searched image information which is input from the searching unit.

8. The apparatus of claim 7, wherein the searching unit searches for an image having a segmented region showing a characteristic similar to the characteristic of the query region, receives a position detection command from the user interface, and searches images with regions having a position similar to one of an absolute position or a relative position as the query region in the image according to a command.

9. The apparatus of claim 7, wherein the characteristic is at least one selected from a characteristic group comprising a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the region from anywhere within an entire image.

10. The apparatus of claim 8, wherein the characteristic is at least one selected from a characteristic group comprising a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the region from anywhere within an entire image.

11. The apparatus of claim 7, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a position from anywhere within an entire image.

12. The apparatus of claim 8, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a position from anywhere within an entire image.

13. A digital image search method for searching for an image in a database in which a plurality of images are stored, comprising:

processing a database in order to generate segmented regions of an image from the database or a search image which is not stored in the database, and analyzing and storing the characteristic of the segmented regions; and

searching for an image having a segmented region showing a characteristic similar to a characteristic of a query segmented region,

wherein the database processing step comprises the steps of;

receiving a database image;

segmenting the received image into homogeneous regions;

analyzing the characteristic of the segmented regions; and

storing the characteristic for each image and region, and

wherein the image searching step comprises the steps of;

receiving a search image;

determining whether the received search image is in the database;

restoring characteristic data in each region when it is determined whether or not the received search image is in the database in the determining step;

segmenting the received search image into homogeneous regions when it is determined that a received search image is not in the database in the determining step and analyzing and storing the characteristic of the segmented regions;

selecting a query region of the search image and a characteristic to be searched; and

searching for an image in the database which includes a region having a characteristic similar to that of the query region.

14. The method of claim 13, wherein the searching step comprises searching for an image having a segmented region showing (1) a characteristic similar to the characteristic of the query region and (2) a position similar to an absolute position or a relative position of the query region in the query image.

15. The method of claim 13, wherein the characteristic is one selected from a characteristic group comprising a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the query region from anywhere within an entire image.

16. The method of claim 14, wherein the characteristic is one selected from a characteristic group comprising a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the query region from anywhere within an entire image.

17. The method of claim 13, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the query image from anywhere within an entire image.

18. The method of claim 14, wherein the characteristic is a combination of a plurality of characteristics including a color histogram showing characteristics of colors, Gabor wavelet coefficients showing characteristics of texture, a boundary description showing a characteristic of shape, and a relative and absolute position of the query image from anywhere within an entire image.

19. A program product comprising a storage media and a computer program operative to perform a digital image search method for searching for an image in a database in which a plurality of images are stored, comprising:

processing a database in order to generate segmented regions of an image from the database or a search image which is not stored in the database, and analyzing and storing a characteristic of the segmented regions; and

searching for an image having a segmented region showing a characteristic similar to a characteristic of a query segmented region,

wherein the database processing step comprises the steps of;

receiving a database image;

segmenting the received image into homogeneous regions;

analyzing a characteristic of the segmented regions; and

storing the characteristic for each image and region, and

wherein the image searching step comprises the steps of;

receiving the search image;

determining whether the received search image is in the database;

restoring characteristic data in each region when it is determined whether or not the received search image is in the database in the determining step;

segmenting the received search image into homogeneous regions when it is determined that the received search image is not in the database in the determining step and analyzing and storing the characteristic of the segmented regions;

selecting a query region of the search image and a characteristic to be searched; and

searching for the image in the database which includes a region having a characteristic similar to that of the query region.

20. The program product of claim 19, wherein the program further comprises a searching step comprises searching for an image having a segmented region showing (1) a characteristic similar to the characteristic of the query region and (2) a position similar to an absolute position or a relative position of the query region in the query image.