Image processing apparatus, computer readable recording medium storing therein image processing program, and image processing method

• US 8,768,017 B2
• Filed: 11/06/2009
• Issued: 07/01/2014
• Est. Priority Date: 11/07/2008
• Status: Active Grant

• Current Assignee: Olympus Corporation
• Original Assignee: Olympus Corporation
• Inventors: Kanda, Yamato
• Primary Examiner(s): Morgan, Robert
• Assistant Examiner(s): Tomaszewski, Michael
• Application Number:
  US12/613,963
• Publication Number:
  US 20100119110A1
• Time in Patent Office: 1,698 Days
• Field of Search: None
• US Class Current: 382/128
• CPC Class Codes: A61B 1/000094 extracting biological structures
• CPC Class Codes: A61B 1/00022 removable
• CPC Class Codes: A61B 1/041 Capsule endoscopes for imaging
• CPC Class Codes: G06F 18/21 Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 10/70 using pattern recognition or machine learning optical pattern recognition or electronic computations therefor G06V10/88
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images

Abstract

An image processing apparatus includes an area dividing unit that divides an image obtained by capturing inside of a body lumen into one or more areas by using a value of a specific wavelength component that is specified in accordance with a degree of absorption or scattering in vivo from a plurality of wavelength components included in the image or wavelength components obtained by conversion of the plurality of wavelength components; and a target-of-interest site specifying unit that specifies a target-of-interest site in the area by using a discriminant criterion in accordance with an area obtained by the division.

Claims

1. An image processing apparatus comprising:

an area dividing unit that divides an image obtained by capturing inside of a body lumen into one or more areas by using a value of a specific wavelength component that is specified in accordance with a degree of absorption or scattering in vivo from a plurality of wavelength components included in the image or from a plurality of wavelength components obtained by conversion of the plurality of wavelength components;

a unit that sets a discriminant criterion for each of the one or more areas, each discriminant criterion is independent of the other discriminant criterion; and

a target-of-interest site specifying unit that specifies a target-of-interest site in the one or more areas by using the discriminant criterion in accordance with each of the one or more areas obtained by the division by the area dividing unit.

2. The image processing apparatus according to claim 1, wherein the area dividing unit determines a longest wavelength component among the plurality of wavelength components as the specific wavelength component and divides the image into the one or more areas by using the value of the specific wavelength component.

3. The image processing apparatus according to claim 1, wherein

the area dividing unit includes a specific-wavelength component computing unit that computes the value of the specific wavelength component by using values of the plurality of wavelength components, and

the area dividing unit divides the image into the one or more areas by using the value of the specific wavelength component computed by the specific-wavelength component computing unit.

4. The image processing apparatus according to claim 1, wherein

the area dividing unit includes a pixel-value hierarchy classifying unit that classifies the value of the specific wavelength component of a pixel in the image into a corresponding hierarchy, and

the area is an area made up of pixels each having the value of the specific wavelength component classified as an identical hierarchy by the pixel-value hierarchy classifying unit.

5. The image processing apparatus according to claim 4, wherein

the pixel-value hierarchy classifying unit includes a pixel-value histogram computing unit that computes a histogram of the value of the specific wavelength component, and

the pixel-value hierarchy classifying unit classifies the value of the specific wavelength component into the corresponding hierarchy in accordance with the histogram computed by the pixel-value histogram computing unit.

6. The image processing apparatus according to claim 1, wherein

the target-of-interest site specifying unit includes a pixel feature-data computing unit that computes pixel feature data of each pixel in an area for each of the areas, and

the target-of-interest site specifying unit specifies a target-of-interest site in the area by comparing the pixel feature data computed for each of the areas by the pixel feature-data computing unit with a discriminant criterion for each of the areas.

7. The image processing apparatus according to claim 1, wherein the target-of-interest site specifying unit includes

a pixel feature-data computing unit that computes pixel feature data of each pixel in an area for each of the areas;

and

a pixel feature-data clustering unit that performs clustering on a distribution of the pixel feature data computed for each of the areas by the pixel feature-data computing unit, and

the target-of-interest site specifying unit discriminates a target-of-interest cluster for each of the areas by comparing information on each cluster obtained for each of the areas as a result of clustering by the pixel feature-data clustering unit with a discriminant criterion for each of the areas and specifies a pixel that belongs to a cluster that is discriminated as the target-of-interest cluster as a target-of-interest site in the area.

8. The image processing apparatus according to claim 1, wherein

the area dividing unit includes a small-area dividing unit that divides the image into a plurality of small areas, and

the area dividing unit calculates a within-small-area average value by using the value of the specific wavelength component of a pixel in a small area obtained by the division by the small-area dividing unit and divides the image into the one or more areas by using the within-small-area average value.

9. The image processing apparatus according to claim 8, wherein

the small-area dividing unit includes an edge extracting unit that extracts an edge from the image, and

the small-area dividing unit divides the image into the plurality of small areas by using the edge extracted by the edge extracting unit.

10. The image processing apparatus according to claim 8, wherein

the area dividing unit includes an average-value hierarchy classifying unit that classifies the within-small-area average value of the small area into a corresponding hierarchy, and

the area is an area constituted by a small area that has the within-small-area average value classified as an identical hierarchy by the average-value hierarchy classifying unit.

11. The image processing apparatus according to claim 10, wherein

the average-value hierarchy classifying unit includes an average-value histogram computing unit that computes a histogram of the within-small-area average value, and

the average-value hierarchy classifying unit classifies the within-small-area average value into the corresponding hierarchy in accordance with the histogram computed by the average-value histogram computing unit.

12. The image processing apparatus according to claim 8, wherein

the target-of-interest site specifying unit includes a small-area feature-data computing unit that computes small-area feature data of each small area in an area for each of the areas, and

the target-of-interest site specifying unit specifies a target-of-interest site in the area by comparing the small-area feature data computed for each of the areas by the small-area feature-data computing unit with a discriminant criterion for each of the areas.

13. The image processing apparatus according to claim 8, wherein the target-of-interest site specifying unit includes

a small-area feature-data computing unit that computes small-area feature data of each small area in an area for each of the areas;

and

a small-area feature-data clustering unit that performs clustering on a distribution of the small-area feature data computed for each of the areas by the small-area feature-data computing unit, and

the target-of-interest site specifying unit discriminates a target-of-interest cluster for each of the areas by comparing information on each cluster obtained as a result of clustering by the small-area feature-data clustering unit with a discriminant criterion for each of the areas and specifies a small area that belongs to a cluster that is discriminated as the target-of-interest cluster as a target-of-interest site in the area.

14. The image processing apparatus according to claim 6, wherein the pixel feature-data computing unit computes the pixel feature data by using a value of a wavelength component that corresponds to an absorption band of blood in a pixel in the area.

15. The image processing apparatus according to claim 7, wherein the pixel feature-data computing unit computes the pixel feature data by using a value of a wavelength component that corresponds to an absorption band of blood in a pixel in the area.

16. The image processing apparatus according to claim 12, wherein the small-area feature-data computing unit computes the small-area feature data by using a value of a wavelength component that corresponds to an absorption band of blood in a pixel that constitutes a small area in the area.

17. The image processing apparatus according to claim 13, wherein the small-area feature-data computing unit computes the small-area feature data by using a value of a wavelength component that corresponds to an absorption band of blood in a pixel that constitutes a small area in the area.

18. The image processing apparatus according to claim 1, wherein the image is an RGB image.

19. The image processing apparatus according to claim 18, wherein the specific wavelength component is an R component of the image.

20. The image processing apparatus according to claim 14, wherein

the image is an RGB image, and

the wavelength component that corresponds to the absorption band of blood is a G component or a B component of the image.

21. The image processing apparatus according to claim 15, wherein

the image is an RGB image, and

the wavelength component that corresponds to the absorption band of blood is a G component or a B component of the image.

22. The image processing apparatus according to claim 16, wherein

the image is an RGB image, and

the wavelength component that corresponds to the absorption band of blood is a G component or a B component of the image.

23. The image processing apparatus according to claim 17, wherein

the image is an RGB image, and

the wavelength component that corresponds to the absorption band of blood is a G component or a B component of the image.

24. The image processing apparatus according to claim 1, wherein the target-of-interest site is a site where any of a lesion, blood vessel, and mucous membrane appears.

27. The image processing apparatus according to claim 1, wherein the area dividing unit divides the image by using a value of only the specific wavelength component that is specified in accordance with the degree of absorption or scattering in vivo from the plurality of wavelength components included in the image or from the plurality of wavelength components obtained by conversion of the plurality of wavelength components.

28. The image processing apparatus according to claim 1, wherein the specified target-of-interest site is specified using the discriminant criterion associated with each of the one or more areas, each area including pixels having similar capturing distances within the area.

29. The image processing apparatus according to claim 1, wherein the discriminant criterion is determined from an occurrence probability and a coefficient of probability density function.

25. A computer readable recording device storing therein an image processing program, the image processing program comprising instructions that cause a computer to perform:

dividing an image obtained by capturing inside of a body lumen into one or more areas by using a value of a specific wavelength component that is specified in accordance with a degree of absorption or scattering in vivo from a plurality of wavelength components included in the image or from a plurality of wavelength components obtained by conversion of the plurality of wavelength components;

setting a discriminant criterion for each of the one or more areas, each discriminant criterion is independent of the other discriminant criterion; and

specifying a target-of-interest site in the one or more areas by using the discriminant criterion in accordance with each of the one or more areas obtained by the dividing the image.

26. An image processing method comprising:

dividing an image obtained by capturing inside of a body lumen into one or more areas by using a value of a specific wavelength component that is specified in accordance with a degree of absorption or scattering in vivo from a plurality of wavelength components included in the image or from a plurality of wavelength components obtained by conversion of the plurality of wavelength components;

setting a discriminant criterion for each of the one or more areas, each discriminant criterion is independent of the other discriminant criterion; and

specifying a target-of-interest site in the one or more areas by using the discriminant criterion in accordance with each of the one or more areas obtained by the dividing the image.