Claims
- 1. An electronic content-based image retrieval (CBIR) system, comprising a front-end including a pipeline of one or more dynamically-constructed filters for measuring perceptual similarities between a query image and one or more candidate images retrieved from a back-end comprised of a knowledge base accessed by an inference engine.
- 2. The system of claim 1, wherein the images include at least one color set having a set of properties including a number of pixels each having at least one color, a culture color associated with the color set, a mean and variance of the color set, a moment invariant, and a centroid.
- 3. The system of claim 2, wherein the filters analyze and compare the set of properties of the query image to the set of properties of the candidate images.
- 4. The system of claim 2, wherein the filters are selected from a group comprising:
a Color Mask filter that identifies identical culture colors in the images, a Color Histogram filter that identifies a distribution of colors in the images, a Color Average filter that performs a similarity comparison on the average of the color sets of the images, a Color Variance filter that performs a similarity comparison on the variances of the color sets of the images, a Spread filter that identifies a spatial concentration of a color in the images, an Elongation filter that identifies a shape of a color in the images, and a Spatial Relationship filter that identifies a spatial relationship between the color sets in the images.
- 5. The system of claim 1, wherein one or more of the filters are selected and weighted to support personalizable queries.
- 6. The system of claim 1, wherein one or more of the filters include distance functions to measure the perceptual similarities of colors, shapes, and spatial distributions in the images.
- 7. The system of claim 6, wherein the distance functions are selected from a group comprising Hamming, Gaussian and Triangle functions.
- 8. A method for query processing in an electronic content-based image retrieval (CBIR) system, comprising
retrieving one or more candidate images from a back-end comprised of a knowledge base accessed by an inference engine; and measuring perceptual similarities between a query image and the candidate images using a pipeline of one or more dynamically-constructed filters in a front-end.
- 9. The method of claim 8, wherein the images include at least one color set having a set of properties including a number of pixels each having at least one color, a culture color associated with the color set, a mean and variance of the color set, a moment invariant, and a centroid.
- 10. The method of claim 9, wherein the filters analyze and compare the set of properties of the query image to the set of properties of the candidate images.
- 11. The method of claim 9, wherein the filters are selected from a group comprising:
a Color Mask filter that identifies identical culture colors in the images, a Color Histogram filter that identifies a distribution of colors in the images, a Color Average filter that performs a similarity comparison on the average of the color sets of the images, a Color Variance filter that performs a similarity comparison on the variances of the color sets of the images, a Spread filter that identifies a spatial concentration of a color in the images, an Elongation filter that identifies a shape of a color in the images, and a Spatial Relationship filter that identifies a spatial relationship between the color sets in the images.
- 12. The method of claim 8, wherein one or more of the filters are selected and weighted to support personalizable queries.
- 13. The method of claim 8, wherein one or more of the filters include distance functions to measure the perceptual similarities of colors, shapes, and spatial distributions in the images.
- 14. The method of claim 13, wherein the distance functions are selected from a group comprising Hamming, Gaussian and Triangle functions.
- 15. An article of manufacture embodying logic for query processing in an electronic content-based image retrieval (CBIR) system, comprising
retrieving one or more candidate images from a back-end comprised of a knowledge base accessed by an inference engine; and measuring perceptual similarities between a query image and the candidate images using a pipeline of one or more dynamically-constructed filters in a front-end.
- 16. The article of manufacture of claim 15, wherein the images include at least one color set having a set of properties including a number of pixels each having at least one color, a culture color associated with the color set, a mean and variance of the color set, a moment invariant, and a centroid.
- 17. The article of manufacture of claim 16, wherein the filters analyze and compare the set of properties of the query image to the set of properties of the candidate images.
- 18. The article of manufacture of claim 16, wherein the filters are selected from a group comprising:
a Color Mask filter that identifies identical culture colors in the images, a Color Histogram filter that identifies a distribution of colors in the images, a Color Average filter that performs a similarity comparison on the average of the color sets of the images, a Color Variance filter that performs a similarity comparison on the variances of the color sets of the images, a Spread filter that identifies a spatial concentration of a color in the images, an Elongation filter that identifies a shape of a color in the images, and a Spatial Relationship filter that identifies a spatial relationship between the color sets in the images.
- 19. The article of manufacture of claim 15, wherein one or more of the filters are selected and weighted to support personalizable queries.
- 20. The article of manufacture of claim 15, wherein one or more of the filters include distance functions to measure the perceptual similarities of colors, shapes, and spatial distributions in the images.
- 21. The article of manufacture of claim 20, wherein the distance functions are selected from a group comprising Hamming, Gaussian and Triangle functions.
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority under 35 U.S.C. §119(e) to U.S. Provisional Patent Application No. 60/190,141, filed Mar. 16, 2000, by Edward Y. Chang, entitled “TOWARDS PERCEPTION-BASED IMAGE RETRIEVAL,” which application is incorporated by reference herein.
Provisional Applications (1)
|
Number |
Date |
Country |
|
60190141 |
Mar 2000 |
US |