Searching and filtering content streams using contour transformations

• US 6,459,809 B1
• Filed: 07/12/1999
• Issued: 10/01/2002
• Est. Priority Date: 07/12/1999
• Status: Expired due to Term

• Current Assignee: Apple Inc.
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Carter, Stephen R., Jensen, Del
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): Mariam, Daniel G.
• Application Number:
  US09/352,421
• Time in Patent Office: 1,177 Days
• Field of Search: 382/159, 382/170, 382/176, 382/180, 382/181, 382/190, 382/195, 382/199, 382/200, 382/203, 382/206, 382/209, 382/219, 382/224, 382/225, 382/227, 382/241, 382/242, 382/276, 382/305, 358/403, 707/1, 707/2, 707/3, 707/6, 375/240.08, 375/240.09
• US Class Current: 382/203
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06V 10/757 Matching configurations of points or features
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

Methods and systems are provided for locating objects within a content stream, by transforming prospective objects and evaluating the results to identify meaningful semantic values. Transformation is accomplished using various contour transformations, possibly in combination with other tools and techniques. The semantic values produced by contour transformation can be efficiently searched and classified against a dictionary of archetypes to identify objects and object features in the content stream. Contour transformations may be scale-invariant and/or rotationally invariant or otherwise symmetric, so that distinctions between content objects based on their scale or orientation are avoided.

Claims

1. A method for analyzing digital data, comprising the computer-implemented steps of:

choosing a plurality of data-independent contours;

placing the chosen contours on the digital data, each contour defining a connected, closed, bounded area within the digital data, the contours being nested in that one of the plurality of contours surrounds the other contours; and

transforming the chosen contours of data using a contour similarity class transformation to produce at least one semantic value.

2. The method of claim 1, wherein:

the choosing step chooses three contours of data; and

the placing step places each contour on the digital data to define a connected, closed, bounded area within the digital data, the three contours being nested in that one of the three contours surrounds the other two contours, and the transforming step transforms the three contours of data using a contour similarity class transformation to produce at least one semantic value.

3. The method of claim 1, wherein the contour similarity class transformation uses contour ratios to produce the semantic value.

4. The method of claim 1, wherein the choosing step comprises using an entity finder to choose an area of interest within the digital data.

5. The method of claim 1, further comprising the step of storing the semantic value within a dictionary of archetypes.

6. The method of claim 5, wherein the dictionary of archetypes contains individual archetypes each of which comprises at least one semantic value and a corresponding textual label, and the semantic value produced by the transforming step is stored based on an ordering imposed on the semantic values of the archetypes.

7. The method of claim 1, wherein the transforming step produces a region having a plurality of semantic values.

8. The method of claim 1, wherein each chosen contour is spherically symmetric.

9. The method of claim 1, wherein each chosen contour is elliptically symmetric.

10. The method of claim 1, wherein each chosen contour is circularly symmetric.

11. A computer program storage medium having a configuration that represents data and instructions which will cause at least a portion of a computer system to perform method steps for analyzing non-textual digital data, the method steps comprising the steps of:

choosing at least two data-independent contours;

placing the chosen contours on the non-textual digital data to define areas of data within the non-textual digital data, each area being a connected, closed, bounded area within the non-textual digital data, the areas being nested; and

transforming the areas using a contour similarity class transformation to produce at least one semantic value.

12. The configured storage medium of claim 11, wherein the contour similarity class transformation is scale-invariant.

13. The configured storage medium of claim 11, wherein the contour similarity class transformation is rotationally-invariant.

14. The configured storage medium of claim 11, further comprising the step of accessing a dictionary of archetypes.

15. The configured storage medium of claim 11, wherein the digital data is two-dimensional.

16. The configured storage medium of claim 11, wherein the digital data is multi-dimensional.