Method for partitioning area, and inspection device

• US 9,672,628 B2
• Filed: 02/25/2014
• Issued: 06/06/2017
• Est. Priority Date: 03/29/2013
• Status: Active Grant

• Current Assignee: Omron Corporation
• Original Assignee: Omron Corporation
• Inventors: Ikeda, Yasuyuki, Minato, Yoshihisa, Kiuchi, Yutaka, Yanagawa, Yukiko, Kato, Yutaka
• Primary Examiner(s): YEH, EUENG NAN
• Application Number:
  US14/781,075
• Publication Number:
  US 20160300358A1
• Time in Patent Office: 1,197 Days
• CPC Class Codes: G06T 2200/24 involving graphical user interfaces [GUIs]
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/20081 Training; Learning
• CPC Class Codes: G06T 2207/20101 Interactive definition of point of interest, landmark or seed
• CPC Class Codes: G06T 2207/20104 Interactive definition of region of interest [ROI]
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 2201/06 Recognition of objects for industrial automation

Abstract

An area partitioning method for partitioning an image into a foreground and a background includes a foreground designation step of causing a user to designate a part of pixels in an area that should be the foreground in the image as a foreground designating pixel, an estimation step of estimating a foreground color distribution and a background color distribution based on a color of the foreground designating pixel designated by the user, and an area partition step of partitioning the image into a foreground area and a background area with the estimated foreground color distribution and the estimated background color distribution as a condition. The estimation step includes dividing the color distribution of the image into a plurality of clusters, and selecting at least one cluster having a large relationship with the foreground designating pixel designated by the user in the plurality of clusters as the foreground color distribution.

Claims

1. An area partitioning method for partitioning an image into a foreground and a background, the method comprising:

a foreground designation step comprising causing a user to designate a part of pixels in an area in the image as a foreground designating pixel;

an estimation step comprising estimating a foreground color distribution and a background color distribution based on a color of the foreground designating pixel designated by the user; and

an area partition step comprising partitioning the image into a foreground area and a background area with the estimated foreground color distribution and the estimated background color distribution as a condition,

wherein the estimation step comprises;

dividing the color distribution of the image into a plurality of clusters;

selecting at least one cluster having a large relationship with the foreground designating pixel designated by the user in the plurality of clusters as the foreground color distribution; and

selecting at least one cluster having a small relationship with the foreground designating pixel as the background color distribution, and

wherein, in the area partition step, an optimum solution of the foreground area is obtained from a plurality of candidate areas that are of candidate solutions of the foreground area by evaluating identity of the foreground of a color of each pixel inside any of the plurality of candidate areas for the foreground color distribution and identity of the background of a color of each pixel outside the candidate area for the background color distribution with respect to the plurality of candidate areas.

2. The area partitioning method according to claim 1, wherein, in the area partition step, the plurality of candidate areas are set to include the foreground designating pixel designated by the user.

3. The area partitioning method according to claim 2, wherein, in the estimation step, magnitude of relationship with the foreground designating pixel is evaluated with respect to each of the plurality of cluster based on a count result of the foreground designating pixel included in the cluster.

4. The area partitioning method according to claim 2, wherein, in the estimation step, the cluster in which the relationship with the foreground designating pixel is greater than or equal to a threshold is selected as the foreground color distribution, or a predetermined number of clusters having the large relationship with the foreground designating pixel is selected in descending order as the foreground color distribution.

5. The area partitioning method according to claim 2, wherein, in the estimation step, the duster having no foreground designating pixels or the cluster in which the relationship with the foreground designating pixel is less than a threshold is selected as the background color distribution, or a predetermined number of clusters having the small relationship with the foreground designating pixel is selected in ascending order as the background color distribution.

6. The area partitioning method according to claim 1, wherein, in the estimation step, magnitude of relationship with the foreground designating pixel is evaluated with respect to each of the plurality of cluster based on a count result of the foreground designating pixel included in the cluster.

7. The area partitioning method according to claim 1, wherein, in the estimation step, the cluster in which the relationship with the foreground designating pixel is greater than or equal to a threshold is selected as the foreground color distribution, or a predetermined number of clusters having the large relationship with the foreground designating pixel is selected in descending order as the foreground color distribution.

8. The area partitioning method according to claim 1, wherein, in the estimation step, the cluster having no foreground designating pixels or the cluster in which the relationship with the foreground designating pixel is less than a threshold is selected as the background color distribution, or a predetermined number of clusters having the small relationship with the foreground designating pixel is selected in ascending order as the background color distribution.

9. The area partitioning method according to claim 1, wherein, in the estimation step, magnitude of relationship with the foreground designating pixel is evaluated with respect to each of the plurality of cluster based on a count result of the foreground designating pixel included in the cluster.

10. The area partitioning method according to claim 9, wherein, in the estimation step, the foreground designating pixel included in the cluster is counted while a weight is provided according to a position of the foreground designating pixel.

11. The area partitioning method according to claim 10, wherein, in the estimation step, the cluster in which the relationship with the foreground designating pixel is greater than or equal to a threshold is selected as the foreground color distribution, or a predetermined number of clusters having the large relationship with the foreground designating pixel is selected in descending order as the foreground color distribution.

12. The area partitioning method according to claim 10, wherein,

in the foreground designation step, a plurality of pixels is designated as the foreground designating pixel by a manipulation to draw a graphic on the image displayed on a display device, and

in the estimation step, a weight of a pixel located in a central portion of the drawn graphic out of the plurality of pixels is set larger than a weight of other pixel in the drawn graphic.

13. The area partitioning method according to claim 10, wherein,

a plurality of foreground designating pixels are included in one cluster, and

in the estimation step, a weight of each pixel of the plurality of foreground designating pixels is set according to the position of the pixel in the cluster.

14. The area partitioning method according to claim 13, wherein,

in the estimation step, a weight of a pixel located in a central portion of the cluster is set larger than a weight of other pixel in the cluster.

15. The area partitioning method according to claim 9, wherein, in the estimation step, the cluster in which the relationship with the foreground designating pixel is greater than or equal to a threshold is selected as the foreground color distribution, or a predetermined number of clusters having the large relationship with the foreground designating pixel is selected in descending order as the foreground color distribution.

16. The area partitioning method according to claim 9, wherein, in the estimation step, the cluster having no foreground designating pixels or the cluster in which the relationship with the foreground designating pixel is less than a threshold is selected as the background color distribution, or a predetermined number of clusters having the small relationship with the foreground designating pixel is selected in ascending order as the background color distribution.

17. The area partitioning method according to claim 1, wherein, in the estimation step, the cluster in which the relationship with the foreground designating pixel is greater than or equal to a threshold is selected as the foreground color distribution, or a predetermined number of clusters having the large relationship with the foreground designating pixel is selected in descending order as the foreground color distribution.

18. The area partitioning method according to claim 1, wherein, in the estimation step, the cluster having no foreground designating pixels or the cluster in which the relationship with the foreground designating pixel is less than a threshold is selected as the background color distribution, or a predetermined number of clusters having the small relationship with the foreground designating pixel is selected in ascending order as the background color distribution.

19. The area partitioning method according to claim 1, where, in the foreground step, the foreground designating pixel is designated by a manipulation to draw a graphic on the image displayed on a display device.

20. The area partitioning method according to claim 19, wherein the graphic is a straight line, a curve, a circle, an ellipse, a polygon, or a closed graphic.

21. A program stored on a non-transitory computer-readable medium that causes a computer to perform an area partitioning method for partitioning an image into a foreground and a background, comprising:

a foreground designation step comprising causing a user to designate a part of pixels in an area in the image as a foreground designating pixel;

an estimation step comprising estimating a foreground color distribution and a background color distribution based on a color of the foreground designating pixel designated by the user; and

an area partition step comprising partitioning the image into a foreground area and a background area with the estimated foreground color distribution and the estimated background color distribution as a condition,

wherein the estimation step comprises;

dividing the color distribution of the image into a plurality of clusters;

selecting at least one cluster having a large relationship with the foreground designating pixel designated by the user in the plurality of clusters as the foreground color distribution, and

selecting at least one cluster having a small relationship with the foreground designating pixel as the background color distribution,

wherein in the area partition step, an optimum solution of the foreground area is obtained from a plurality of candidate areas that are of candidate solutions of the foreground area by evaluating identity of the foreground of a color of each pixel inside any of the plurality of candidate areas for the foreground color distribution and identity of the background of a color of each pixel outside the candidate area for the background color distribution with respect to the plurality of candidate areas.

22. The program according to claim 21, wherein,

in the estimation step, magnitude of relationship with the foreground designating pixel is evaluated with respect to each of the plurality of cluster based on a count result of the foreground designating pixel included in the cluster.

23. The program according to claim 22, wherein,

in the estimation step, the foreground designating pixel included in the cluster is counted while a weight is provided according to a position of the foreground designating pixel.

24. The program according to claim 23, wherein,

in the foreground designation step, a plurality of pixels is designated as the foreground designating pixel by a manipulation to draw a graphic on the image displayed on a display device, and

in the estimation step, a weight of a pixel located in a central portion of the drawn graphic out of the plurality of pixels is set larger than a weight of other pixel in the drawn graphic.

25. The program according to claim 23, wherein,

a plurality of foreground designating pixels are included in one cluster, and

in the estimation step, a weight of each pixel of the plurality of foreground designating pixels is set according to the position of the pixel in the cluster.

26. The program according to claim 25, wherein,

in the estimation step, a weight of a pixel located in a central portion of the cluster is set larger than a weight of other pixel in the cluster.