SIMILAR SEMICIRCLE DETECTING APPARATUS AND COMPUTER-READABLE STORAGE MEDIUM STORING SIMILAR SEMICIRCLE DETECTING PROGRAM

• US 20090245656A1
• Filed: 03/18/2009
• Published: 10/01/2009
• Est. Priority Date: 03/25/2008
• Status: Active Grant

• Current Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Original Assignee: Fujifilm Corporation (Fujifilm Holdings Corporation)
• Inventors: HU, Xuebin
• Granted Patent: US 8,363,945 B2
• Time in Patent Office: Days
• US Class Current: 382/203
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/473 using gradient analysis

Abstract

The present invention performs a process of integrating plural pixel evaluation values corresponding to plural evaluation pixels on a circular arc overlapping a circle using a predetermined pixel in interest as its center and calculating an integrated evaluation value to evaluate whether the circular arc is a circular arc corresponding to a similar semicircular shape, for each of plural circular arcs that overlap the circle in overlapping phases which are different from each other, compares plural integrated evaluation values corresponding to the plural circular arcs, and extracts the circular arc with which the integrated evaluation value indicating the most approximation to the similar semicircular shape is associated, from the plural circular arcs.

Claims

1. A similar semicircle detecting apparatus that detects a similar semicircular shape similar to a semicircular shape from an image represented by pixels which are two-dimensionally arranged, the similar semicircle detecting apparatus comprising:

a first evaluation value calculating section that calculates a pixel evaluation value to evaluate whether a current evaluation pixel is a pixel on a circular arc in which the predetermined pixel in interest is centered, for each of a plurality of evaluation pixels which are arranged along a circle in which a predetermined pixel in interest is centered and surround the predetermined pixel in interest;

a second evaluation value calculating section that performs a process of integrating plural pixel evaluation values corresponding to the plurality of evaluation pixels on the circular arc overlapping the circle so as to calculate an integrated evaluation value to evaluate whether the circular arc is a circular arc corresponding to a similar semicircular shape, for each of a plurality of circular arcs which overlap the circle in overlapping phases which are different from each other; and

a circular arc extracting section that compares a plurality of integrated evaluation values corresponding to the plural circular arcs, and extracts the circular arc with which the integrated evaluation value indicating the most approximation to the similar semicircular shape is associated, from the plural circular arcs.

2. The similar semicircle detecting apparatus according to claim 1,

wherein the first evaluation value calculating section includes:

a gradient direction calculating section that calculates a gradient direction where a characteristic amount on the image varies, for each of the plurality of evaluation pixels; and

a pixel evaluation value calculating section that calculates the pixel evaluation value based on an angle which is formed by a direction in which the current evaluation pixel and the pixel in interest and the gradient direction calculated by the gradient direction calculating section, for each of the plurality of evaluation pixels.

3. The similar semicircle detecting apparatus according to claim 2,

wherein the gradient direction calculating section includes:

a gradient value calculating section that calculates each gradient value on the basis of a difference between a characteristic amount of the current evaluation pixel and a characteristic amount of each of a plurality of gradient calculation pixels which circumferentially surround the current evaluation pixel;

a smoothing process section that generates a smoothing gradient value group by applying a smoothing process to a gradient value group in which the gradient values each corresponding to each of the plurality of gradient calculation pixels are arranged in an order in which the plurality of gradient calculation pixels circumferentially surround the current evaluation pixel; and

a gradient direction determining section that determines as the gradient direction a direction in which the current evaluation pixel and the gradient calculation pixel corresponding to the maximum smoothing gradient value in the smoothing gradient value group among the plurality of gradient calculation pixels which circumferentially surround the current evaluation pixel.

4. The similar semicircle detecting apparatus according to claim 3,

wherein the smoothing process section includes:

a high value suppressing process section that replaces the gradient value exceeding a predetermined threshold value among the individual gradient values constituting the gradient value group with the predetermined threshold value.

5. The similar semicircle detecting apparatus according to claim 3,

wherein the smoothing process section applies a moving average process of suppressing a noise to the gradient value group.

6. The similar semicircle detecting apparatus according to claim 3,

wherein the smoothing process section includes:

a quantizing process section that replaces each of the gradient values constituting the gradient value group by any one of the plural gradient values that have values approximated to the gradient values and are discretely set, and generates a new gradient value group.

7. The similar semicircle detecting apparatus according to claim 3,

wherein the smoothing process section applies a moving average process of developing a gradient value having a unique maximum value in the gradient value group.

8. The similar semicircle detecting apparatus according to claim 2,

wherein the pixel evaluation value calculating section includes:

a table storage section that stores a table common to a plurality of pixels in interest and a plurality of evaluation pixels in which table both of a first value which indicates a direction of the evaluation pixel when viewed from the pixel in interest and a second value which indicates the gradient direction of the evaluation pixel and the pixel evaluation value of the current evaluation pixel are associated with each other; and

a table referring section that calculates a pixel evaluation value of the current evaluation pixel from the table based on the specific first value which indicates the direction of the evaluation pixel when the specific evaluation pixel is viewed from the specific pixel in interest and the specific second value that indicates the gradient direction of the current evaluation pixel.

9. A computer-readable storage medium that stores a similar semicircle detecting program that is executed in a computer and causes the computer to operate as a similar semicircle detecting apparatus which detects a similar semicircular shape similar to a semicircular shape from an image represented by pixels arranged two-dimensionally, the similar semicircle detecting apparatus including:

a first evaluation value calculating section that calculates a pixel evaluation value to evaluate whether a current evaluation pixel is a pixel on a circular arc in which the predetermined pixel in interest is centered, for each of a plurality of evaluation pixels which are arranged along a circle in which a predetermined pixel in interest is centered and surround the predetermined pixel in interest;

a second evaluation value calculating section that performs a process of integrating plural pixel evaluation values corresponding to the plurality of evaluation pixels on the circular arc overlapping the circle so as to calculate an integrated evaluation value to evaluate whether the circular arc is a circular arc corresponding to a similar semicircular shape, for each of a plurality of circular arcs which overlap the circle in overlapping phases which are different from each other; and

a circular arc extracting section that compares a plurality of integrated evaluation values corresponding to the plural circular arcs, and extracts the circular arc with which the integrated evaluation value indicating the most approximation to the similar semicircular shape is associated, from the plural circular arcs.