IMAGE PROCESSING

• US 20090310861A1
• Filed: 09/29/2006
• Published: 12/17/2009
• Est. Priority Date: 10/31/2005
• Status: Active Grant

• Current Assignee: Sony Europe BV (Sony Group Corp.)
• Original Assignee: Sony United Kingdom Limited (Sony Group Corp.)
• Inventors: Williams, Michael John, Thorpe, Jonathan Richard, Prayle, Paul Edward, Beresford, Ratna, Porter, Robert Mark Stefan, Lang, Mikael Carl
• Granted Patent: US 8,385,646 B2
• Time in Patent Office: Days
• US Class Current: 382/173
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/7335 Graphical querying, e.g. query-by-region, query-by-sketch, query-by-trajectory, GUIs for designating a person/face/object as a query predicate end-user interface involving hot spots associated with the video H04N21/4725; end-user interface for selecting a Region of Interest H04N21/4728
• CPC Class Codes: G06F 16/7847 using low-level visual features of the video content
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06F 3/048 Interaction techniques based on graphical user interfaces [GUI]
• CPC Class Codes: G06Q 30/0621 Item configuration or customization
• CPC Class Codes: G06Q 30/0631 Item recommendations
• CPC Class Codes: G06Q 30/0633 Lists, e.g. purchase orders, compilation or processing
• CPC Class Codes: G06Q 30/0643 Graphical representation of items or shoppers
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/10 Segmentation; Edge detection motion-based segmentation G06T7/215
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/143 involving probabilistic approaches, e.g. Markov random field [MRF] modelling
• CPC Class Codes: G06T 7/45 using co-occurrence matrix computation
• CPC Class Codes: G06V 10/457 by analysing connectivity, e.g. edge linking, connected component analysis or slices
• CPC Class Codes: G06V 10/56 relating to colour

Abstract

An image processing method includes partitioning an image under test to form a plurality of contiguous image segments having similar image properties, deriving feature data from a subset including one or more of the image segments, and comparing the feature data from the subset of image segments with feature data derived from respective image segments of one or more other images so as to detect a similarity between the image under test and the one or more other images.

Claims

1. An image processing method comprising the steps of:

partitioning an image under test to form a plurality of image segments, each segment representing a set of pixels having similar image properties, at least some segments being contiguous;

deriving feature data from a subset comprising one or more of the image segments;

comparing the feature data from the subset of image segments with feature data indicative of respective reference image segments so as to detect a degree of similarity between the image segments under test and the reference image segments.

2. A method according to claim 1, in which:

the reference image segments are respective images segments of one or more other images; and

the comparing step is operable to detect a degree of similarity between the image under test and the one or more other images.

3. A method according to claim 1, in which the partitioning step comprises:

(i) deriving image property data from pixels of the image under test;

(ii) grouping pixels into image segments in accordance with their respective image property data.

4. A method according to claim 3, in which the partitioning step comprises:

(iii) detecting the separation of image segments in terms of image property data differences; and

(iv) merging image segments which have less than a threshold separation.

5. A method according to claim 4, in which step (iii) is carried out recursively.

6. A method according to claim 3, further comprising the step of:

(v) spatially filtering the image segments, the spatial filtering tending to reassign pixels from small image segments into larger image segments.

7. A method according to claim 3, further comprising the steps of:

detecting whether the image under test has a dark border;

removing such a border before the partitioning step; and

reapplying such a border after the partitioning step.

8. A method according to claim 1, in which the feature data is indicative of at least colour properties.

9. A method according to claim 1, in which the feature data is indicative of at least image texture properties.

10. A method according to claim 1, further comprising the steps of:

deriving respective feature data from each image segment; and

where the subset of image segments, in respect of the image under test, comprises more than one image segment, performing the comparing step using a combination of the feature data from all of the image segments in the subset.

11. A method according to claim 1, in which each segment is contiguous.

12. A method according to claim 1, further comprising the steps of:

generating feature data indicative of a plurality of reference image segments, each reference image segment having an associated data item;

detecting user selection of an image segment derived from the image under test;

accessing the data item associated with the reference image data detected to be most similar to the image segment selected by the user.

13. A method according to claim 12, in which:

the associated data items are hyperlinks; and

the step of accessing the associated data item comprises accessing the hyperlink target associated with that hyperlink.

14. A method according to claim 12, in which the step of accessing the associated data item comprises providing to the user data indicative of a purchase price of an article or service associated with that reference image segment.

15. A method according to claim 14, further comprising the step of:

detecting user operation of a purchase control indicating that the user wishes to purchase the article or service.

16. A method according to claim 12, in which the step of detecting user selection comprises:

displaying at least a part of the image under test; and

detecting user operation of a pointing device directed at that image segment.

17. Computer software comprising program code which, when executed on a computer, executes a method according to any one of the preceding claims.

18. A medium by which computer software according to claim 17 is provided.

19. A medium according to claim 18, the medium being a storage medium.

20. A medium according to claim 18, the medium being a transmission medium.

21. Image processing apparatus comprising:

means for partitioning an image under test to form a plurality of image segments, each segment representing a set of pixels having similar image properties, at least some segments being contiguous;

means for deriving feature data from a subset comprising one or more of the image segments; and

means for comparing the feature data from the subset of image segments with feature data indicative of respective reference image segments so as to detect a degree of similarity between the image segments under test and the reference image segments.

22. Image capture apparatus comprising image processing apparatus according to claim 21.

23. An image processing method comprising the steps of:

partitioning an image under test to form a plurality of image segments, each segment representing a set of pixels having similar image properties, at least some being contiguous, the partitioning step comprising;

(i) deriving image property data from pixels of the image under test;

(ii) grouping pixels into image segments in accordance with their respective image property data.

24. Image processing apparatus comprising:

means for partitioning an image under test to form a plurality of image segments, each segment representing a set of pixels having similar image properties, at least some being contiguous, the partitioning means comprising;

means for deriving image property data from pixels of the image under test;

means for grouping pixels into image segments in accordance with their respective image property data.