KR20130090740A - Apparatus and method processing image - Google Patents

Abstract

PURPOSE: An image processing apparatus and a method thereof are provided to supply effective fetal biometric measurements through a data processor. CONSTITUTION: An image processing apparatus includes a data acquiring unit. The data acquiring unit acquires image data about a subject including target bones (S210). Previous image data is obtained through thresholding based on the image data (S220). Through labeling, a plurality of segments within the image data are classified (S230). On the basis of target bone image features, one segment of the plural segments is determined as a target image (S240). The length of the target bone is measured based on the target image (S250). [Reference numerals] (AA) Start; (BB) End; (S210) Acquire image data about a subject including target bones; (S220) Obtain previous image data through thresholding based on the image data; (S230) Classify segments within the image data through labeling; (S240) Determine one segment of the plural segments as a target; (S250) Measure the length of the target bone based on the target image

Description

Image processing apparatus and method {APPARATUS AND METHOD PROCESSING IMAGE}

The present invention relates to an image processing apparatus and method.

The image processing apparatus may be used for a medical imaging apparatus such as an ultrasound imaging apparatus, a computed tomography (CT) apparatus, or a magnetic resonance imaging (MRI) apparatus. For example, the image processing apparatus may be included in the medical imaging apparatus.

The image processing device may be used for fetal biometric measurements. Biometrics can be performed to estimate the gestational age of the fetus, to assess the size of the fetus and to monitor fetal growth. An example of biometrics is measuring the length of a target bone, such as the femur. The measured target bone length is an important factor for predicting fetal malformations.

Therefore, there is a need for an image processing apparatus and an operation method thereof capable of efficiently measuring the length of a target bone.

SUMMARY OF THE INVENTION The present invention has been made in an effort to provide an image processing apparatus and an operation method thereof capable of efficiently measuring the length of a target bone.

According to an aspect of the present invention, there is provided an image processing apparatus including: a data acquirer configured to acquire image data of a subject including a target bone; And obtaining binary image data through thresholding based on the image data, classifying a plurality of segments in the binary image data through labeling, and based on the image characteristics of the target bone. And a data processor to determine one of the segments of the target image as a target image and to measure the length of the target bone based on the target image.

The image data is volume data, and the data processor analyzes a shape of each of the plurality of segments, obtains one or more remaining segments from the plurality of segments based on the analyzed shape, and based on a brightness value. One of the one or more remaining segments may be determined as the target image.

The data processor performs a principal component analysis (PCA) on each of the plurality of segments to analyze the shape of each of the plurality of segments, thereby providing a size of a first principal component for each of the plurality of segments. The size of the second principal component and the size of the third principal component may be obtained, and the one or more remaining segments may be obtained based on the sizes of the first to third principal components.

The data processor obtains a tube-score defined based on the magnitudes of the first to third principal components for each of the plurality of segments, and the tube score of the plurality of segments is greater than a threshold. Segments may be determined as the one or more remaining segments.

로 정의되고,

은 상기 제1 주성분의 크기이고,

는 상기 제2 주성분의 크기이고,

는 상기 제3 주성분의 크기이다. The tube score is

Lt; / RTI >

Is the size of the first principal component,

Is the size of the second principal component,

Is the size of the third principal component.

The data processor may determine a segment having the largest brightness value among the one or more remaining segments as the target image.

The segment determined as the target image may be a segment having the largest sum of brightness values of points belonging to the segment.

The data processor may determine a longitudinal section of the target image and measure a length of the target bone based on the longitudinal section.

The data processor obtains an average point corresponding to an average coordinate of points belonging to the target image, and obtains both end points farthest from a longitudinal section of the volume data passing through the average point among points belonging to the target image. Set spheres around the end points, obtain, for each of the spheres, a first point and a second point corresponding to an average coordinate of points belonging to the sphere while belonging to the target image; A point belonging to the sphere while acquiring a point at which the distance from the first point is equal to the distance from the second point among the points belonging to the image, setting a sphere centered on the point, and belonging to the target image Obtain a third point corresponding to their mean coordinates And it may determine the first point, the second point and a cross-section through the third point in the longitudinal cross-sectional view of the target image.

The data processor may measure the length of the target bone based on a length between the first point and the second point.

The data processor may acquire an adaptive threshold value based on an average and a standard deviation of brightness values based on the image data, and perform the thresholding based on the adaptive threshold value to obtain the binary image data. .

The data processor may perform top-hat conversion and contrast improvement based on the image data before performing the thresholding.

The data processor may perform denoising on the image data before the top-hat conversion.

The image data may be two-dimensional image data, and the data processor may determine, as the target image, a segment having the largest brightness value among the plurality of segments. The segment determined as the target image may be a segment having the largest sum of brightness values of points belonging to the segment.

The data processor may detect both end points in the long axis direction of the target image, and measure the length of the target bone based on the distance between the both end points.

According to another aspect of the present invention, there is provided a method of processing an image, the method including: obtaining image data of a subject including a target bone; Acquiring binary image data through thresholding based on the image data; Classifying a plurality of segments in the binary image data by labeling; Determining one segment of the plurality of segments as a target image based on an image characteristic of the target bone; And measuring a length of the target bone based on the target image.

In a computer-readable recording medium according to another embodiment of the present invention for solving the technical problem, a program for implementing the image processing method is recorded.

According to an embodiment of the present invention, an image processing apparatus and an operation method thereof capable of efficiently measuring the length of a target bone can be provided.

1 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
2 is an example of a 2D image imaged with 2D image data.
3 shows an example of an image preprocessed through denoising.
4 shows an example of a top-hat transformed image.
5 shows an example of a contrast-enhanced image.
6 shows an example of a binary image.
7 shows an example of a labeled binary image.
8 illustrates an example of a binary image in which only a target image is displayed.
9 illustrates an example of a method of measuring the length of the target bone based on the target image.
10 shows an example of volume data.
11 is an example of various types of target images.
12 shows a horizontal cross section of volume data.
13 is a flowchart of an image processing method according to another embodiment of the present invention.

Hereinafter, with reference to the drawings will be described embodiments of the present invention;

1 is a block diagram of an image processing apparatus according to an exemplary embodiment of the present invention.

Referring to FIG. 1, the image processing apparatus 100 may include a data acquisition device 110 and a data processor 120.

The image processing apparatus 100 may be used for a medical imaging apparatus such as an ultrasound imaging apparatus, a computed tomography (CT) apparatus, or a magnetic resonance imaging (MRI) apparatus. For example, the image processing apparatus 100 may be included in the medical imaging apparatus.

The data acquirer 110 may acquire image data of a subject. The subject is an animal including a human body or part of an animal. In addition, the subject includes a target bone, which is in the form of a long thin tube. For example, the subject is a mother and the target bone may be a long bone, such as the femur of the fetus. The image data may be two-dimensional image data capable of imaging a cross section in a subject, or volume data capable of three-dimensional imaging of a three-dimensional space in a subject.

The 2D image data may be a plurality of pixel values, and the volume data may be a plurality of voxel values. The pixel value may be a brightness value of a corresponding pixel, and the voxel value may be a brightness value of a corresponding voxel. Hereinafter, for convenience of description, a point is used as a term meaning a pixel or a voxel.

The data acquirer 110 may acquire image data by scanning a subject using an ultrasonic signal, but is not limited thereto. As another example, the data acquirer 110 may receive scan information obtained by scanning a subject from a scan device external to the image processing apparatus 100 and may acquire image data based on the scan information. As another example, the data acquirer 110 may receive image data from an external device. However, the present invention is not limited thereto, and the image processing apparatus 100 may obtain image data by various methods.

The data processor 120 may process the image data to measure the length of the target bone. The data processor 120 acquires binary image data through thresholding based on the image data, classifies a plurality of segments in the binary image data through labeling, and images of the target bone. One segment of the plurality of segments may be determined as the target image based on the characteristic, and the length of the target bone may be measured based on the target image.

First, when the image data is two-dimensional image data, the image data processing method of the data processor 120 will be described.

2 is an example of a 2D image imaged with 2D image data. The 2D image of FIG. 2 is a B mode ultrasound image.

Referring to FIG. 2, the 2D image includes a target image that is an image of the target bone. However, due to noise and images of other tissues around the target bone, the boundary of the target image in the 2D image is not clear.

Accordingly, the data processor 120 of FIG. 1 may process the 2D image as follows before performing the thresholding.

The data processor 120 may obtain a preprocessed image by preprocessing the 2D image. For example, denosing based on a total variation (TV) algorithm may be performed as a preprocessing.

3 shows an example of an image preprocessed through denoising. Referring to FIG. 3, it can be seen that noise is removed and edges are maintained through denoising.

Denoising based on a TV algorithm is merely an example, and the data processor 120 of FIG. 1 may remove noise from a 2D image and improve image quality through various preprocessing methods. However, the data processor 120 may omit the preprocessing process.

The target image, which is an image of the target bone, is thin and long, and the brightness value of the target image is higher than that of other areas. Accordingly, in order to extract a long region of bright areas from the preprocessed image, the data processor 120 may perform top-hat conversion on the preprocessed image. If the preprocessing process is omitted, the data processor 120 may perform top-hat conversion on the binary image.

The top-hat transform h can be expressed by the following equation.

[Equation 1]

Here, f is an input image, that is, a preprocessed image. b is a structuring element, and ˚ represents an opening operation.

4 shows an example of a top-hat transformed image.

After the top-hat transform, contrast enhancement may be performed on the top-hat transformed image to make the edges of the target image clearer and more clearly distinguish the target image from other areas.

For example, an image obtained by applying contrast enhancement to a top-hat transformed image having 256 gray levels (CEh (p), p represents a point) may be obtained through the following equation.

&Quot; (2) "

Where h (p) is the brightness value of point p in the top-hat transformed image, and max and min are the maximum brightness value and the minimum brightness value in the top-hat transformed image, respectively. In this case, since min is 0 or a value close to 0, an appropriately small value such as min = 20 may be assigned.

5 shows an example of a contrast-enhanced image.

As such, the data processor 120 of FIG. 1 may perform denoising, top-hat conversion, and contrast improvement on the 2D image.

Next, the data processor 120 may acquire a binary image to which an adaptive threshold is applied from the contrast-enhanced image.

For example, the binary image g (p) from the contrast enhanced image CEh (p) may be obtained through the following equation.

&Quot; (3) "

Where T is an adaptive threshold. In other words, the binary image may be an image in which the brightness value of the contrast-enhanced image is greater than the adaptive threshold T, and the remaining points are black.

The adaptive threshold may be obtained based on the mean and standard deviation of the brightness values in the contrast enhanced image. For example, the adaptive threshold T can be obtained through the following equation.

&Quot; (4) "

Where m is the average of the brightness values in the contrast enhanced image, s is the standard deviation of the brightness values, and a is the weight.

6 shows an example of a binary image.

Referring to FIG. 6, the binary image may include a target image, but may include an image of another tissue having a different shape.

Therefore, the data processor 120 of FIG. 1 may process the binary image as follows to extract the target image.

The data processor 120 may distinguish a plurality of segments in the binary image data through labeling. Each of the plurality of segments is an area in which points having a brightness value of 1 are collected. The plurality of segments are candidates of the target image.

7 shows an example of a labeled binary image.

Referring to FIG. 7, each of the plurality of segments in a binary image is labeled with different gray levels. 11 is merely an example of labeling, and a binary image may be labeled in various ways. For example, each of the plurality of segments may be labeled with a number.

Next, the data processor 120 of FIG. 1 may determine one of the plurality of segments as the target image based on the image characteristic of the target bone. The segment having the most image characteristics of the target bone among the plurality of segments is determined as the target image, which is the image of the target bone.

Since the target bone is highly reflective than the surrounding tissue, the image of the target bone is larger than the other areas of the brightness value. Accordingly, the data processor 120 may determine the segment having the largest brightness value among the plurality of segments as the target image. In detail, the data processor 120 may obtain a sum of brightness values of points belonging to a plurality of segments and determine a segment having the largest sum of brightness values as a target image.

The sum S _L of brightness values of each segment may be expressed as in the following equation.

&Quot; (5) "

Where L is the index of the segment and p is the point.

When the target image is determined, the data processor 120 may display all portions except the segment determined as the target image in black.

8 illustrates an example of a binary image in which only a target image is displayed.

The data processor 120 of FIG. 1 may measure the length of the target bone based on the determined target image.

9 illustrates an example of a method of measuring the length of the target bone based on the target image.

Referring to FIG. 9, the data processor 120 of FIG. 1 may obtain a measurement line 11 by skeletonizing a target image. The measurement unit 128 may measure the length of the target bone based on the distance between the first point 12a and the second point 12b, which are intersections of the target image and the measurement line 11.

However, FIG. 9 is merely an example, and the data processor 120 detects both end points 12a and 12b in the long axis direction of the target image in the extracted image in various ways, and thus the distance between the end points 12a and 12b. The length of the target bone can be measured based on.

Up to now, the case where the image data is two-dimensional image data has been described. Hereinafter, the case where the image data is volume data will be described.

10 shows an example of volume data.

Referring to FIG. 10, the volume data 300 includes a target image 310, which is a three-dimensional image of the target bone. For convenience of illustration, although the target image 310 is clearly shown in FIG. 10, the boundary of the target image may not be clear due to noise or the like in the actual volume data.

Pixel values are processed in 2D image data, but voxel values are processed in volume data. Except for this, the two-dimensional image data processing method may be applied to the volume data. Therefore, the same volume data processing method as the two-dimensional image data processing method will be briefly described, and the processing method applied only to the volume data will be described.

The data processor 120 of FIG. 1 may acquire binary volume data from the volume data through thresholding. Before acquiring the binary volume data, the above-described top-hat conversion, contrast improvement, etc. may be performed, and then an adaptive threshold may be applied.

Binary volume data may include a target image, but may include an image of another tissue having a different shape. Therefore, image processing for extracting the target image from the binary volume data may be performed as follows.

The data processor 120 may label the binary volume data. The binary volume data is divided into a plurality of segments through labeling. Each of the plurality of segments is a three-dimensional area in which points having a brightness value of 1 are collected. The plurality of segments are candidates of the target image.

The data processor 120 may determine one segment of the plurality of segments as a target image. Among the plurality of segments, the segment having the most image characteristics of the target bone is determined as the target image.

Image characteristics of the target bone may include shape information and brightness values.

The data processor 120 may analyze the shape of each of the plurality of segments and obtain one or more remaining segments from the plurality of segments based on the analyzed shape. Next, the data processor 120 may determine one of the one or more remaining segments as the target image based on the brightness value.

11 is an example of various types of target images.

Referring to FIG. 11, the target image is in the form of a straight tube or a curved tube.

To analyze the shape of each of the plurality of segments, the data processor 120 of FIG. 1 may perform a principal component analysis (PCA) on each of the plurality of segments. PCA is one of the techniques for analyzing data sets, which is useful for determining the distribution of data. PCA is an analysis technique that finds the direction of maximizing the distribution of data and condenses the data therefrom to express the information of the data more easily. PCA linearly transforms data into a new coordinate system, such that when the data is mapped onto one axis, the axis with the largest variance comes to the first coordinate axis and the second to the second axis.

The data processor 120 may obtain the direction and the magnitude of the first principal component, the direction and the magnitude of the second principal component, and the direction and the magnitude of the third principal component for each of the plurality of segments through the PCA. The data in the form of tubes shows that the size of the first principal component is relatively large and the size of the remaining second and third principal components is relatively small. Accordingly, the data processor 120 may analyze the shape of each of the plurality of segments based on the sizes of the first to third principal components.

Using the sizes of the first to third principal components, a tube-score can be defined as follows.

&Quot; (6) "

은 제1 주성분의 크기이고,

는 제2 주성분의 크기이고,

는 제3 주성분의 크기이다. Where Ts is the tube score,

Is the magnitude of the first principal component,

Is the magnitude of the second principal component,

Is the magnitude of the third principal component.

The data processor 120 may obtain a tube score for each of the plurality of segments, and determine, among the plurality of segments, one or more remaining segments whose tube score is greater than a threshold. That is, segments whose tube score is less than or equal to the threshold are excluded from the target image candidate. The threshold can be set empirically. For example, the threshold may be set to 0.997.

The data processor 120 obtains one or more remaining segments from the plurality of segments based on the analyzed form, and determines one of the one or more remaining segments as the target image based on the brightness value.

The data processor 120 may determine, as the target image, a segment having the largest brightness value among one or more remaining segments. In detail, the data processor 120 may obtain a sum of brightness values of points belonging to one or more remaining segments and determine a segment having the largest sum of brightness values as a target image.

The data processor 120 excludes some of the plurality of segments from the candidate by analyzing the shape of the segments before determining the target image. Through this, it is possible to prevent a relatively large segment from being incorrectly selected as the target image.

The data processor 120 may measure the length of the target bone based on the determined target image.

However, there may be images of a plurality of long bones in the binary volume data. In the manner described above, the data processor 120 may measure the length of the first long bone, that is, the target bone, as a reference. Next, the data processor 120 may select an image of other long bones based on the length of the target bone, and measure the length of the other long bones as well.

Next, a method of measuring the length of the target bone based on the determined target image will be described.

The data processor 120 may determine a longitudinal section of the target image and measure the length of the target bone based on the longitudinal section. In order to accurately measure the length of the target bone, the length of the target bone must be measured at the longitudinal section of the target bone.

At least three points are required to determine a specific plane in three-dimensional space, and these points must not be on a straight line. Thus, to determine the longitudinal section of the target image, three or more points on the longitudinal section must be determined. Next, a method of determining three points on the longitudinal section will be described with reference to FIG. 12.

12 shows a horizontal cross section of volume data.

Referring to FIG. 12, the volume data 300a includes a segment determined as the target image 310a. The data processor 120 of FIG. 1 obtains an average point M1 corresponding to the average coordinates of all the points belonging to the target image 310a. The data processor 120 may acquire the far end points S1 and E1 farthest from the longitudinal section 300b passing through the average point M1 among the points belonging to the target image 310a. The vertical cross section 300b is a vertical cross section of the volume data 300a passing through the average point M1.

The data processor 120 sets spheres 21 and 22 centered at both end points S1 and E1 at both end points S1 and E1. The data processor 120, for each sphere 21, 22, corresponds to the average coordinates of points belonging to the target image 310a and belonging to each sphere 21, 22. E2) is obtained. The radius of each sphere 21, 22 may be set such that the points belonging to the target image 310a and points belonging to each sphere 21, 22 may exist properly. For example, one third of the distance between the end points S1 and E1 may be set as the radius of each sphere 21 and 22.

The data processor 120 obtains a point M2 at which the distance from the first point S2 is equal to the distance from the second point E2 among the points belonging to the target image 310a. That is, the distance between the first point S2 and the point M2 is equal to the distance between the second point E2 and the point M2. Among the points belonging to the target image 310a, there may be a plurality of points at which the distance from the first point S2 is equal to the distance from the second point E2. The data processor 120 may obtain any one of the plurality of points as the point M2.

The data processor 120 sets the sphere 23 centered on the point M2. The data processor 120 obtains a third point M3 corresponding to the average coordinates of the points belonging to the target image 310a and belonging to the sphere 23.

The data processor 120 may determine a cross section passing through the first point S2, the second point E2, and the third point M3 as the longitudinal cross section of the target image 310a. In addition, the data processor 120 may measure the length of the target bone based on the length between the first point S2 and the second point E2.

12 is an example of a method of determining a longitudinal cross section of a target image, and may determine the longitudinal cross section of the target image in various other methods.

13 is a flowchart illustrating an image processing method according to another exemplary embodiment of the present invention.

Referring to FIG. 13, image data of a subject including a target bone is obtained (S210). Binary image data is acquired through thresholding based on the image data (S220). The plurality of segments in the binary image data are distinguished through labeling (S230). Based on the image characteristics of the target bone, one segment of the plurality of segments is determined as the target image (S240). The length of the target bone is measured based on the target image (S250).

The image processing method of FIG. 13 may be performed by the image processing apparatus of FIG. 1. For each step of the image processing method, the contents described with reference to FIGS. 1 to 12 may be applied. Therefore, redundant description will be omitted.

As such, according to an embodiment of the present invention, an image processing apparatus and method capable of efficiently measuring the length of a target bone may be provided.

The image processing apparatus according to the embodiment of the present invention may automatically extract a target image, which is an image of the target bone, based on the image data, and automatically measure the length of the target bone. This is more efficient than when the user views the image and manually measures the length of the target bone. In the case of manual measurement, there is a problem that measurement deviation occurs according to the user's skill and the user's subjective judgment. In addition, in the case of manual measurement, the time required for the measurement may be long. According to an embodiment of the present invention, since the length of the target bone is measured automatically, the accuracy of the measurement can be increased. In addition, the time required for the measurement can be shortened.

In addition, the image processing apparatus according to the embodiment of the present invention may automatically extract the target image from the volume data, and automatically determine the longitudinal section from the target image. If the user has to manually find the end face of the target bone, it is difficult to guarantee accuracy due to the deviation between users. In addition, if the user has to repeat the scan to find the longitudinal section, the repetitive stress injury (RSI) of the user may be increased. Therefore, according to the embodiment of the present invention, the accuracy of the measurement can be increased. In addition, the user's RSI can be reduced.

Meanwhile, the above-described method can be implemented in a general-purpose digital computer that can be created as a program that can be executed by a computer and operates the program using a computer-readable recording medium. In addition, the structure of the data used in the above-described method can be recorded on a computer-readable recording medium through various means. The computer readable recording medium may be a magnetic storage medium such as a ROM, a RAM, a USB, a floppy disk or a hard disk, an optical reading medium such as a CD-ROM or a DVD, ) (E.g., PCI, PCI-express, Wifi, etc.).

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed methods should be considered from an illustrative point of view, not from a restrictive point of view. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

Claims (20)

A data acquirer for obtaining image data of a subject including a target bone; And
Acquire binary image data through thresholding based on the image data,
Labeling distinguishes a plurality of segments in the binary image data,
Determining one segment of the plurality of segments as a target image based on an image characteristic of the target bone,
And a data processor configured to measure a length of the target bone based on the target image. The method of claim 1,
The image data is volume data,
The data processor,
Analyze a shape of each of the plurality of segments, obtain one or more remaining segments from the plurality of segments based on the analyzed shape,
And determining one of the one or more remaining segments as the target image based on a brightness value. The method of claim 2,
The data processor,
In order to analyze the shape of each of the plurality of segments, by performing a principal component analysis (PCA) on each of the plurality of segments, the size of the first principal component, the size of the second principal component and And obtaining the size of a third principal component and acquiring the one or more remaining segments based on the sizes of the first to third principal components. The method of claim 3,
The data processor,
Obtaining a tube-score defined based on the magnitudes of the first to third principal components for each of the plurality of segments, and selecting one or more segments of the plurality of segments having a tube score greater than a threshold value. An image processing apparatus for determining residual segments. 5. The method of claim 4,
The tube score is

Lt; / RTI >

Is the size of the first principal component,

Is the size of the second principal component,

Is the size of the third principal component. The method of claim 5,
The data processor,
And determining the target image having the largest brightness value among the one or more remaining segments as the target image. The method according to claim 6,
And the segment determined as the target image is a segment having the largest sum of brightness values of points belonging to the segment. The method of claim 7, wherein
The data processor,
Determine a longitudinal section of the target image,
An image processing apparatus for measuring the length of the target bone based on the longitudinal section. 9. The method of claim 8,
The data processor,
Obtaining an average point corresponding to an average coordinate of points belonging to the target image,
Obtaining end points farthest from a longitudinal section of the volume data passing through the average point among the points belonging to the target image,
Setting spheres around the end points,
For each of the spheres, obtain a first point and a second point corresponding to an average coordinate of points belonging to the spheres while belonging to the target image,
Obtaining a point at which a distance from the first point is equal to a distance from the second point among points belonging to the target image,
Setting a sphere centered on the point, obtaining a third point belonging to the target image and corresponding to an average coordinate of points belonging to the sphere,
And a cross section passing through the first point, the second point, and the third point as the longitudinal section of the target image. 10. The method of claim 9,
The data processor is configured to measure the length of the target bone based on a length between the first point and the second point. The method of claim 1,
The data processor,
Based on the image data, obtain an adaptive threshold based on the mean and standard deviation of the brightness value,
And the binary image data is obtained by performing the thresholding based on the adaptive threshold value. 12. The method of claim 11,
The data processor,
And a top-hat transform and contrast improvement based on the image data before performing the thresholding. The method of claim 12,
The data processor,
And denoising the image data before the top-hat conversion. The method of claim 13,
The image data is two-dimensional image data,
The data processor,
And a segment having the largest brightness value among the plurality of segments as the target image. 15. The method of claim 14,
And the segment determined as the target image is a segment having the largest sum of brightness values of points belonging to the segment. 16. The method of claim 15,
The data processor,
And detecting both end points in a long axis direction of the target image, and measuring the length of the target bone based on a distance between the both end points. Acquiring image data about a subject including a target bone;
Acquiring binary image data through thresholding based on the image data;
Classifying a plurality of segments in the binary image data by labeling;
Determining one segment of the plurality of segments as a target image based on an image characteristic of the target bone; And
And measuring the length of the target bone based on the target image. 18. The method of claim 17,
The image data is volume data,
Determining one of the plurality of segments as the target image includes:
Analyzing a shape of each of the plurality of segments and obtaining one or more remaining segments from the plurality of segments based on the analyzed shape; And
And determining one of the one or more remaining segments as the target image based on a brightness value. 19. The method of claim 18,
Measuring the length of the target bone is:
Determining a longitudinal section of the target image; And
And measuring the length of the target bone based on the longitudinal section. 20. A computer readable recording medium having recorded thereon a program for implementing the image processing method of any one of claims 17 to 19.

Images