SYSTEMS AND METHODS FOR IMAGE RECOGNITION USING GRAPH-BASED PATTERN MATCHING

• US 20090324107A1
• Filed: 06/25/2009
• Published: 12/31/2009
• Est. Priority Date: 06/25/2008
• Status: Active Grant

• Current Assignee: Gannon Technologies Group, LLC
• Original Assignee: Gannon Technologies Group, LLC
• Inventors: Walch, Mark A.
• Granted Patent: US 8,452,108 B2
• Time in Patent Office: Days
• US Class Current: 382/224
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/18181 Graphical representation, e.g. directed attributed graph
• CPC Class Codes: G06V 30/2272 with lexical matching

Abstract

A method for creating a modeling structure for classifying objects in an image comprises converting an image into digital image data; using a processor, simplifying the digital image data; using the processor, isolating objects in the simplified digital image data; using the processor, creating graphs of the isolated objects, the graphs comprising vertices and edges; using the processor, converting the graphs into representative graph data structures, the graph data structures comprising a database key based on the vertices and edges.

Claims

1. A method for creating a modeling structure for classifying objects in an image, comprising:

converting an image into digital image data;

using a processor, simplifying the digital image data;

using the processor, isolating objects in the simplified digital image data;

using the processor, creating graphs of the isolated objects, the graphs comprising vertices and edges;

using the processor, converting the graphs into representative graph data structures, the graph data structures comprising a database key based on the vertices and edges.

2. The method of claim 1, wherein each graph data structure includes an object classification, and wherein each of the object classifications is associated with the vertices and edges.

3. The method of claim 1, further including using the database key to extract the graphs of the objects from a database.

4. The method of claim 2, wherein the vertices and edges define a topological relationship and concomitant feature data.

5. The method of claim 1, wherein the graph data structure is based on a regression tree classifier model.

6. The method of claim 1, wherein the graph data structure is based on a neural network classifier model.

7. The method of claim 1, wherein the graph data structure is based on a discriminant analysis model.

8. The method of claim 1, further comprising storing the graphs as templates in a database for identifying the same objects in new images.

9. The method of claim 8, further comprising:

receiving a new image;

using the processor, converting the new image into digital image data;

using the processor, simplifying the digital image data;

using the processor, isolating objects in the simplified digital image data;

using the processor, creating graphs of the isolated objects, the graphs comprising vertices and edges; and

using the processor, comparing the graphs to the templates to determine whether one of the isolated objects is in the new image.

10. The method of claim 1, wherein simplifying the digital image data comprises reducing the image complexity and segmenting one or more foreground objects from background images.

11. The method of claim 10, wherein reducing the image complexity comprises digital signal processing techniques including at last one of the following:

Wiener filtering, automatic brightness adjustment, and contrast stretching.

12. The method of claim 10, wherein segmenting includes at least one of the following methods:

adaptive thresholding, morphological reconstructions, and ridge detection.

13. The method of claim 10, wherein reducing the image complexity comprises accentuating sharp straight edges and suppressing amorphous objects with in the image.

14. The method of claim 1, wherein isolating objects in the simplified image data further comprises isolating objects embedded in a larger more complex image.

15. The method of claim 1, wherein the graph data structure includes intrinsic and extrinsic characteristics of the isolated objects.

16. A system for creating a modeling structure for classifying objects in an image, the system comprising:

a database;

an image input interface configured to provide digital image data; and

a processor coupled with the database and the image input interface, the processor configured to;

simplify the digital image data,

isolate objects in the simplified digital image data,

create graphs of the isolated objects, the graphs comprising vertices and edges, and

convert the graphs into representative graph data structures, the graph data structures comprising a database key based on the vertices and edges.

17. The system of claim 16, wherein each graph data structure includes an object classification, and wherein each of the object classifications is associated with the vertices and edges.

18. The system of claim 16, wherein the processor is further configured to use the database key to extract the graphs of the objects from the database.

19. The system of claim 17, wherein the vertices and edges define a topological relationship and concomitant feature data.

20. The system of claim 16, wherein the processor is further configured to create the graph data structure based on a regression tree classifier model.

21. The system of claim 16, wherein the processor is further configured to create the graph data structure based on a neural network classifier model.

22. The system of claim 16, wherein the processor is further configured to create the graph data structure based on a discriminant analysis model.

23. The system of claim 16, wherein the processor is further configured to store the graphs as templates in the database for identifying the same objects in new images.

24. The system of claim 23, wherein the processor is further configured to:

receive a new digital image data from the image input interface;

simplify the new digital image data;

isolate objects in the simplified new digital image data;

create graphs of the isolated objects, the graphs comprising vertices and edges; and

compare the graphs to the templates to determine whether one of the isolated objects is in the new image.

25. The system of claim 16, wherein the processor is configured to simplify the digital image data by reducing the image complexity and segmenting one or more foregrounds object from background images.

26. The system of claim 25, wherein reducing the image complexity comprises at last one of the following:

Wiener filtering, automatic brightness adjustment, and contrast stretching.

27. The system of claim 25, wherein segmenting includes at least one of the following:

adaptive thresholding, morphological reconstructions, and ridge detection.

28. The system of claim 25, wherein reducing the image complexity comprises accentuating sharp straight edges and suppressing amorphous objects with in the image.

29. The system of claim 24, wherein isolating objects in the simplified image data further comprises isolating objects embedded in a larger more complex image.

30. The system of claim 16, wherein the graph data structure includes intrinsic and extrinsic characteristics of the isolated objects.