IMAGE PROCESSING APPARATUS, IMAGE PROCESSING METHOD, AND COMPUTER-READABLE RECORDING MEDIUM

• US 20120076419A1
• Filed: 08/11/2011
• Published: 03/29/2012
• Est. Priority Date: 09/29/2010
• Status: Active Grant

• Current Assignee: Olympus Corporation
• Original Assignee: Olympus Corporation
• Inventors: KONO, Takashi, KANDA, Yamato, KITAMURA, Makoto, MATSUDA, Takehiro, HIROTA, Masashi
• Granted Patent: US 8,670,622 B2
• Time in Patent Office: Days
• US Class Current: 382/195
• CPC Class Codes: G06T 2207/10068 Endoscopic image
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06T 7/181 involving edge growing; involving edge linking

Abstract

An image processing apparatus includes a gradient magnitude calculating unit that calculates a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged; a candidate-edge detecting unit that detects contour-candidate edge positions on the basis of the gradient magnitude of each pixel; a reference-range setting unit that sets a reference range, which is to be referred to when a contour edge position is selected from among the contour-candidate edge positions, on the basis of the gradient magnitude of each pixel; and a contour-edge selecting unit that selects, as the contour edge position, one of the contour-candidate edge positions in the reference range.

Claims

1. An image processing apparatus comprising:

a gradient-magnitude calculating unit that calculates a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged;

a candidate-edge detecting unit that detects contour-candidate edge positions on the basis of the gradient magnitude of each pixel;

a reference-range setting unit that sets a reference range, which is to be referred to when a contour edge position is selected from among the contour-candidate edge positions, on the basis of the gradient magnitude of each pixel; and

a contour-edge selecting unit that selects, as the contour edge position, one of the contour-candidate edge positions in the reference range.

2. The image processing apparatus according to claim 1, wherein the gradient-magnitude calculating unit includes a noise removing unit that removes noise from the target image.

3. The image processing apparatus according to claim 1, wherein the gradient-magnitude calculating unit includes a structure-information acquiring unit that acquires shape change information on a surface of the target object on the basis of pixel values of the target image.

4. The image processing apparatus according to claim 1, wherein the gradient-magnitude calculating unit includes a differential processing unit that performs a differential filter process on the target image to calculate the gradient magnitude of each pixel.

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

the candidate-edge detecting unit includes

a gradient-change-direction detecting unit that detects a gradient change direction of each pixel on the basis of the gradient magnitude of each pixel;

and

a convex-position detecting unit that detects, as a convex position, a pixel at which a change in a gradient magnitude is greatest in an area in a same gradient direction, on the basis of the gradient change direction of each pixel, wherein

the convex position is set as the contour-candidate edge position.

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

the candidate-edge detecting unit includes

a candidate-edge-line detecting unit that detects an edge line on the basis of the gradient magnitude of each pixel;

and

a thresholding unit that performs thresholding on gradient magnitudes of pixels on the edge line, and detects a pixel of which value of a gradient magnitude is equal to or greater than a predetermined threshold from the pixels on the edge line, wherein

the pixel of which value of the gradient magnitude is equal to or greater than the predetermined threshold on the edge line is set as the contour-candidate edge position.

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

the candidate-edge-line detecting unit includes a thinning unit that detects a thinned edge line on the basis of the gradient magnitude of each pixel.

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

the thresholding unit performs hysteresis thresholding to detect a pixel of which value of the gradient magnitude is equal to or greater than the predetermined threshold from the pixels on the edge line.

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

the reference-range setting unit includes

a reference-direction specifying unit that specifies a reference direction by which the reference range is set;

and

a reference-range acquiring unit that acquires, as the reference range, a range that contains the contour-candidate edge position and that has a continuous upward gradient or a continuous downward gradient in the reference direction.

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

the reference-direction specifying unit specifies a plurality of reference directions;

and

the reference-range acquiring unit acquires a reference range for each of the reference directions.

11. The image processing apparatus according to claim 10, wherein the reference directions are two or more directions including a lateral direction and a longitudinal direction of the target image.

12. The image processing apparatus according to claim 9, wherein the reference-direction specifying unit includes a reference-direction calculating unit that calculates a normal direction of the contour-candidate edge position as the reference direction.

13. The image processing apparatus according to claim 9, wherein the reference-range acquiring unit includes a gradient-magnitude thresholding unit that performs thresholding on a value of the gradient magnitude of a pixel in the reference direction, and acquires an upward gradient range or a downward gradient range along the reference direction.

14. The image processing apparatus according to claim 1, wherein the candidate-edge detecting unit detects the contour-candidate edge position in the reference range that is set by the reference-range setting unit.

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

the contour-edge selecting unit includes

an edge-characteristic acquiring unit that acquires a maximum value of the gradient magnitude of the contour-candidate edge position for each contour-candidate edge line on which the contour-candidate edge positions are continuously arranged, and acquires, as an edge characteristic, a contour-candidate edge position on a contour-candidate edge line corresponding to the maximum value of the gradient magnitude;

and

an edge-characteristic comparing unit that compares edge characteristics of contour-candidate edge positions contained in the reference range, and selects the contour edge position on the basis of a result of comparison.

16. The image processing apparatus according claim 15, wherein

the gradient-magnitude calculating unit calculates directional components of the gradient magnitude of each pixel, the directional components corresponding to a plurality of directions,

the candidate-edge detecting unit detects the contour-candidate edge position on the basis of the directional components of the gradient magnitude of each pixel,

the reference-range setting unit handles the directions as reference directions and sets the reference range for each of the reference directions,

the edge-characteristic comparing unit includes

a directional-component contour-edge selecting unit that selects a directional-component contour edge position from each of the reference ranges corresponding to the respective reference directions;

and

a directional-component contour-edge integrating unit that integrates, as the contour edge, reliable directional-component contour edge positions from among the directional-component contour edge positions in each of the reference directions.

17. An image processing method comprising:

calculating a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged;

detecting contour-candidate edge positions on the basis of the gradient magnitude of each pixel;

setting a reference range, which is to be referred to when a contour edge position is selected from among the contour-candidate edge positions, on the basis of the gradient magnitude of each pixel; and

selecting, as the contour edge position, one of the contour-candidate edge positions in the reference range.

18. An image processing method comprising:

calculating a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged;

setting a reference range, which is to be referred to when a contour edge position is selected from among contour-candidate edge positions, on the basis of the gradient magnitude of each pixel;

detecting contour-candidate edge positions in the reference range on the basis of gradient magnitudes of pixels contained in the reference range; and

selecting, as the contour edge position, one of the contour-candidate edge positions in the reference range.

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

calculating a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged;

detecting contour-candidate edge positions on the basis of the gradient magnitude of each pixel;

setting a reference range, which is to be referred to when a contour edge position is selected from among the contour-candidate edge positions, on the basis of the gradient magnitude of each pixel; and

selecting, as the contour edge position, one of the contour-candidate edge positions in the reference range.

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

calculating a gradient magnitude of each pixel on the basis of pixel values of a target image in which a predetermined target object is imaged;

setting a reference range, which is to be referred to when a contour edge position is selected from among contour-candidate edge positions, on the basis of the gradient magnitude of each pixel;

detecting contour-candidate edge positions in the reference range on the basis of gradient magnitudes of pixels contained in the reference range; and

selecting, as the contour edge position, one of the contour-candidate edge positions in the reference range.