Image processing apparatus, image processing method, and computer-readable recording medium

• US 8,811,698 B2
• Filed: 04/27/2011
• Issued: 08/19/2014
• Est. Priority Date: 04/30/2010
• Status: Active Grant

• Current Assignee: Olympus Corporation
• Original Assignee: Olympus Corporation
• Inventors: Kono, Takashi, Kanda, Yamato, Kitamura, Makoto, Matsuda, Takehiro, Hirota, Masashi
• Primary Examiner(s): Koziol, Stephen R
• Assistant Examiner(s): Azima, Shaghayegh
• Application Number:
  US13/095,157
• Publication Number:
  US 20110274321A1
• Time in Patent Office: 1,210 Days
• Field of Search: 382/128, 382/103, 382/160, 382/167, 382/131, 382/199, 382/165
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10068 Endoscopic image
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20116 Active contour; Active surface; Snakes
• CPC Class Codes: G06T 2207/30028 Colon; Small intestine
• CPC Class Codes: G06T 2207/30092 Stomach; Gastric
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/149 involving deformable models, e.g. active contour models
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling

Abstract

An image processing apparatus includes: a gradient information calculating unit that calculates gradient information of each of pixels, based on pixel values of an intraluminal image; a closed region creating unit that, based on the gradient information, creates a closed region satisfying a condition where the closed region does not include, on the inside thereof, any pixel of which the gradient strength is equal to or higher than a predetermined value, and also, the boundary of the closed region does not curve toward the interior of the closed region, with a curvature equal to or larger than a predetermined value; and an abnormal part detecting unit that detects an abnormal part from the inside of the closed region.

Claims

1. An image processing apparatus that detects an abnormal part from an intraluminal image, the apparatus comprising:

a processor; and

a memory storing computer-readable instructions that, when executed by the processor, implement;

a gradient information calculating unit that calculates gradient information of each of pixels, based on pixel values of the intraluminal image;

a closed region creating unit that, based on the gradient information, creates a closed region so that the closed region has a boundary with pixels with positions satisfying a condition where the closed region does not include, on an inside thereof, any pixel of which a gradient strength is equal to or higher than a predetermined value, and also, a boundary of the closed region does not curve toward an interior of the closed region, with a curvature equal to or larger than a predetermined value; and

an abnormal part detecting unit that detects the abnormal part from the inside of the closed region.

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

the intraluminal image includes a plurality of wavelength components,

the gradient information calculating unit includes:

a specific wavelength image extracting unit that extracts, from the intraluminal image, an image corresponding to a specific wavelength component specified depending on a degree of absorption or dispersion within a living body; and

a gradient strength calculating unit that calculates gradient strengths of the pixel values of the pixels, based on the image corresponding to the specific wavelength component, and

the gradient strengths are the gradient information.

3. The image processing apparatus according to claim 1, wherein the closed region creating unit includes:

a dividing unit that divides the intraluminal image into a plurality of regions, based on the gradient information; and

an integrating unit that, based on feature data of each of the regions, integrates two or more of the regions under a condition where a boundary of a region resulting from the integration does not curve toward an interior of the region, with a curvature equal to or larger than a predetermined value, and the region resulting from the integration performed by the integrating unit is the closed region.

4. The image processing apparatus according to claim 3, wherein the dividing unit includes:

a line edge detecting unit that detects a line edge, based on the gradient information;

a base point detecting unit that detects an end point of the line edge and a curving point at which the line edge curves with a curvature equal to or larger than a predetermined value and uses the end point and the curving point as base points of a dividing line; and

a dividing line creating unit that draws dividing lines from each of the base points into two mutually-different directions, until each of the dividing lines meets a line edge or a rim of an image, and

each of regions defined by the line edges and the dividing lines is a different one of the plurality of regions.

5. The image processing apparatus according to claim 3, wherein the integrating unit includes:

a region information calculating unit that calculates, as the feature data of each of the regions, region information including a shape feature data of the region and a length of a boundary between the region and another region positioned adjacent thereto;

a region integrating unit that selects the two or more regions being integration targets based on the region information and integrates the selected regions; and

an integration controller that controls a repetition of processes performed by the region information calculating unit and the region integrating unit and ends the repetition when there are no longer any regions being the integration targets.

6. The image processing apparatus according to claim 1, wherein the closed region creating unit includes:

an initial closed region creating unit that creates an initial shape of the closed region, based on the gradient information;

a first energy calculating unit that calculates, as a first energy, a weighted sum of at least two selected out of;

an internal energy exhibiting a value where the smoother the boundary of the closed region is, the smaller is the value;

an image energy exhibiting a value where the higher the gradient strength at the boundary of the closed region is, the smaller is the value; and

an external energy exhibiting a value where the larger a size of the closed region is, the smaller is the value; and

a closed region updating unit that, while using the initial shape of the closed region as a reference, updates a shape of the closed region so as to reduce the first energy.

7. The image processing apparatus according to claim 6, wherein the initial closed region creating unit creates the initial shape of the closed region in a position where the gradient strength is relatively lower than a surrounding thereof.

8. The image processing apparatus according to claim 7, wherein the initial closed region creating unit creates the initial shape of the closed region so that the closed region does not overlap any other closed regions.

9. The image processing apparatus according to claim 7, wherein the initial closed region creating unit creates the initial shape of the closed region so that the closed region partially overlaps another closed region and, as a result of calculating the first energy with respect to the created closed region, if the calculated first energy exceeds a predetermined value, the initial closed region creating unit corrects the initial shape of the closed region so that the created closed region does not overlap said another closed region.

10. The image processing apparatus according to claim 6, wherein the closed region updating unit includes:

an updating unit that updates the shape of the closed region;

a second energy calculating unit that calculates, as a second energy, a weighted sum of at least two selected out of the following, with respect to the updated closed region;

an internal energy exhibiting a value where the smoother a boundary of the updated closed region is, the smaller is the value;

an image energy exhibiting a value where the higher the gradient strength at the boundary of the updated closed region is, the smaller is the value; and

an external energy exhibiting a value where the larger a size of the updated closed region is, the smaller is the value;

a correcting unit that compares the first energy with the second energy, corrects the shape of the updated closed region back to the shape of the closed region prior to the update if the first energy is smaller, and updates the first energy with the second energy if the second energy is smaller; and

a closed region update controller that controls a repetition of processes performed by the updating unit, the second energy calculating unit, and the correcting unit and ends the repetition when the first energy no longer exhibits any change.

11. The image processing apparatus according to claim 1, wherein the abnormal part detecting unit includes:

a reference color estimating unit that estimates a reference color indicating a normal biological tissue, for each of the closed regions; and

an abnormal color part detecting unit that detects, for each of the closed regions, a region having color feature data deviating from the reference color by a predetermined amount or larger, as an abnormal color part, and

the abnormal color part is the abnormal part.

12. The image processing apparatus according to claim 11, wherein the reference color estimating unit estimates the reference color by using a morphology filter.

13. The image processing apparatus according to claim 9, wherein the abnormal color part detecting unit detects the abnormal color part, based on a difference value between a pixel value on the inside of the closed region and the reference color.

14. An image processing method for detecting an abnormal part from an intraluminal image, the method comprising:

calculating gradient information of each of pixels, based on pixel values of the intraluminal image;

creating, based on the gradient information, a closed region so that the closed region has a boundary with pixels with positions satisfying a condition where the closed region does not include, on an inside thereof, any pixel of which a gradient strength is equal to or higher than a predetermined value, and also, a boundary of the closed region does not curve toward an interior of the closed region with a curvature equal to or larger than a predetermined value; and

detecting the abnormal part from the inside of the closed region.

15. A non-transitory computer-readable recording medium with an executable program stored thereon, wherein the program instructs a processor to perform:

calculating gradient information of each of pixels, based on pixel values of the intraluminal image;

creating, based on the gradient information, a closed region so that the closed region has a boundary with pixels with positions satisfying a condition where the closed region does not include, on an inside thereof, any pixel of which a gradient strength is equal to or higher than a predetermined value, and also, a boundary of the closed region does not curve toward an interior of the closed region with a curvature equal to or larger than a predetermined value; and

detecting the abnormal part from the inside of the closed region.

16. An image processing apparatus that detects an abnormal part from an intraluminal image, the apparatus comprising:

a gradient information calculating unit that calculates gradient information of each of pixels, based on pixel values of the intraluminal image;

a closed region creating unit that, based on the gradient information, creates a closed region so that the closed region has a boundary with pixels with positions satisfying a condition where the closed region does not include, on an inside thereof, any pixel of which a gradient strength is equal to or higher than a predetermined value, and also, a boundary of the closed region does not curve toward an interior of the closed region, with a curvature equal to or larger than a predetermined value; and

an abnormal part detecting unit that detects the abnormal part from the inside of the closed region.