Pattern matching method selection

• US 7,397,956 B2
• Filed: 06/15/2007
• Issued: 07/08/2008
• Est. Priority Date: 03/18/2002
• Status: Expired due to Term

• Current Assignee: National Instruments Corporation (Emerson Electric Company)
• Original Assignee: National Instruments Corporation (Emerson Electric Company)
• Inventors: Nair, Dinesh, Vazquez, Nicolas, Wenzel, Lothar, Schultz, Kevin L.
• Primary Examiner(s): Desire; Gregory M
• Application Number:
  US11/763,902
• Publication Number:
  US 20070237396A1
• Time in Patent Office: 389 Days
• Field of Search: 382/103, 382/141, 382/203, 382/209, 382/217, 382/218
• US Class Current: 382/209
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches

Abstract

A system and method for performing pattern matching to locate zero or more instances of a template image in a target image. An image is received by a computer from an image source, e.g., a camera. First pattern matching is performed on the image using a first pattern matching technique to determine a plurality of candidate areas. Second pattern matching is performed on each of the candidate areas using a second different pattern matching technique to generate final pattern match results. An output is generated indicating the final pattern match results. The second pattern matching may determine a second plurality of candidate areas which may be analyzed to determine the final pattern match results. The first pattern matching may use a plurality of pattern matching techniques, the results of which may be used to select a best technique from the plurality of techniques to use for the second pattern match.

Claims

1. A method for detecting patterns in an image, the method comprising:

receiving a first image of a first type;

performing first pattern matching on the first image using a plurality of different pattern matching techniques, wherein said performing pattern matching determines one or more candidate areas for each of said plurality of different pattern matching techniques;

analyzing the one or more candidate areas determined by each of said plurality of different pattern matching techniques to select one of the pattern matching techniques;

receiving a plurality of images of the first type;

performing pattern matching on each of the plurality of images of the first type using said selected one of the pattern matching techniques, thereby generating pattern match results; and

generating output indicating said pattern match results.

2. The method of claim 1, wherein the plurality of pattern matching techniques comprises two or more of:

a geometric pattern matching technique;

a correlation-based pattern matching technique;

a color matching technique;

a hue plane pattern matching technique;

or

a luminance pattern matching technique.

3. The method of claim 1, wherein the plurality of pattern matching techniques comprises one or more of:

a coarse pattern matching technique;

or

a fine pattern matching technique.

4. The method of claim 1, wherein at least one of the plurality of pattern matching techniques uses a hill-climbing technique.

5. The method of claim 1, wherein at least one of the plurality of pattern matching techniques uses a sub-sampling technique.

6. The method of claim 1, wherein said analyzing the one or more candidate areas determined by each of said plurality of different pattern matching techniques to select one of the pattern matching techniques comprises selecting one of the pattern matching techniques based on a desired characteristic of a pattern matching task, the desired characteristic comprising one or more of:

accuracy;

or

speed.

7. The method of claim 1, further comprising:

a Graphical User Interface (GUI) receiving user input specifying a desired characteristic of a pattern matching task prior to said analyzing, wherein said selecting one of the pattern matching techniques is based on said desired characteristic.

8. The method of claim 7, wherein the user specifying said desired characteristic comprises the user specifying said characteristic using a GUI control.

9. The method of claim 8,

wherein said Graphical User Interface (GUI) control comprises a slider control;

and

wherein said slider control is operable by the user to select a value indicating a degree of said desired characteristic.

10. The method of claim 1,

wherein said performing first pattern matching comprises generating a first pattern matching score for each of the plurality of candidate areas, and wherein the one or more candidate areas are determined based on said first pattern matching scores.

11. The method of claim 1,

wherein said performing first pattern matching comprises generating a first pattern matching score for each of the plurality of candidate areas, and wherein said analyzing the one or more candidate areas includes analyzing the first pattern matching scores.

12. The method of claim 11, wherein said analyzing the first pattern matching scores comprises:

comparing the first pattern matching scores to pre-determined matching scores for known matching patterns in the first image.

13. The method of claim 1, wherein two or more of the plurality of pattern matching techniques are performed concurrently.

14. The method of claim 13,

wherein the two or more of the plurality of pattern matching techniques are performed concurrently using a multiprocessor computer system.

15. The method of claim 13,

wherein the two or more of the plurality of pattern matching techniques are performed concurrently using a plurality of programmable hardware elements.

16. The method of claim 13,

wherein the two or more of the plurality of pattern matching techniques are performed concurrently using one or more processors and one or more programmable hardware elements.

17. A system for detecting patterns in an image, the system comprising:

an input which is operable to receive images from one or more image sources;

a processor; and

a memory medium coupled to the processor, wherein the memory medium stores one or more pattern matching programs for detecting patterns in the image;

wherein said input is operable to perform;

receiving a first image of a first type

wherein the processor is operable to execute the one or more pattern matching programs to implement;

performing first pattern matching on the first image using a plurality of different pattern matching techniques, wherein said performing pattern matching determines one or more candidate areas for each of said plurality of different pattern matching techniques;

analyzing the one or more candidate areas produced by each of said plurality of different pattern matching techniques to select one of the pattern matching techniques;

wherein said input is further operable to perform;

receiving a plurality of images of the first type; and

wherein the processor is further operable to execute the one or more pattern matching programs to implement;

performing pattern matching on each of the plurality of images of the first type using said selected one of the pattern matching techniques, thereby generating pattern match results; and

generating output indicating said pattern match results.

18. A system for detecting patterns in an image, the system comprising:

means for receiving a first image of a first type;

means for performing first pattern matching on the first image using a plurality of different pattern matching techniques to determine a plurality of candidate areas for each of said plurality of different pattern matching techniques;

means for analyzing the plurality of candidate areas determined by each of said plurality of different pattern matching techniques to select one of the plurality of pattern matching techniques;

means for receiving a plurality of images of the first type;

means for performing pattern matching on each of the plurality of images of the first type using said selected one of the pattern matching techniques, thereby generating pattern match results; and

means for outputting said pattern match results.

19. A computer-accessible memory medium comprising program instructions for detecting a pattern in an image, wherein the program instructions are executable by one or more processors to implement:

receiving a first image of a first type;

performing pattern matching on the first image using a plurality of different pattern matching techniques, wherein said performing pattern matching determines one or more candidate areas for each of said plurality of different pattern matching techniques;

analyzing the one or more candidate areas produced by each of said plurality of different pattern matching techniques;

selecting one of the pattern matching techniques in response to said analyzing;

receiving a plurality of images of the first type;

performing pattern matching on each of the plurality of images of the first type using said selected one of the pattern matching techniques, thereby generating pattern match results; and

generating output indicating said pattern match results.