JPH0636048A - Volume data clipping system - Google Patents

Abstract

PURPOSE:To shorten the total processing time of three-dimensional image processing and securely determine an aimed area as an object of the processing. CONSTITUTION:This system is equipped with a three-dimensional image memory 10 where volume data as the object of the three-dimensional image processing are stored and boxel data are read out, section by section, in mutually orthogonal (x), (y), and (z) axial directions, an operation terminal 20 for inputting an object light and shade value for the boxel data so as to specify the aimed area in the stored volume data, a data conversion part 12 which converts the light and shade value of boxel data matching the object light and shade value among the boxel data read out of the three-dimensional image memory 10 into a specific light and shade value, a histogram measurement part 14 which measures the boxel data converted into the specific light and shade value, and a control part 18 which determines the border of the aimed area in the (x), (y), and (z) axial direction and specifies the aimed area.

Description

Detailed Description of the Invention

[0001]

The present invention relates to medical and EA (Engineer)
ing Automation) The gray value used in the field is transparency,
The present invention relates to a volume data clipping method which has a meaning as an attribute such as density and is used for a process of reading a tomographic image in an arbitrary direction with respect to a three-dimensional image (volume) input as a tomographic image.

[0002]

2. Description of the Related Art Generally, medical and EA (Engineering Auto)
In the field of (mation), for example, an image of a cross section of an object output from an X-ray device or a computer is conceptually stacked in a three-dimensional region (a space represented by x-y-z axes) and a tomographic image is obtained. An apparatus having a memory for storing an image, that is, a three-dimensional image (volume data) is used.

In this type of device, the structure of an object or the like can be easily judged visually based on the volume data stored in the memory, by a process such as reprojection conversion or volume rendering from any direction. It is possible to display an image when viewed, an image when viewed through a watermark, and the like. Since a large amount of data is handled in this process, high speed processing is required. Conventionally, the following method is used to process the input volume data. 1. With respect to the input volume data, a three-dimensional image processing operation is performed on the entire area.

2. Although there is no spatial limitation on the target area of the input volume data, the number of processed data can be increased by increasing the sampling interval of data in each axis direction such as xyz in the three-dimensional image processing operation. To reduce
Increase the overall processing speed.

3. The user arbitrarily designates the target area range (for example, the length in the x-axis direction) of the input volume data to limit the data amount of the three-dimensional image processing calculation, thereby increasing the processing speed as a whole.

[0006]

As described above, according to the conventional technique 1, a three-dimensional image processing operation which requires a long calculation time is performed on all the input volume data, so that a large amount of the entire processing is performed. It takes time.

Further, in the case of the prior art 2, since the speed is increased by increasing the sampling interval of the input data, sufficient attention must be paid to the sampling interval.
There is a problem that the accuracy of the result of the three-dimensional image processing is lowered.

Further, in the case of the conventional technique 3, since the user can arbitrarily specify the range of the target area to reduce the data amount to increase the speed, it is desired to perform processing when the size is limited. There is a problem that areas may be excluded by mistake.

The present invention has been made in view of the above points, and it is possible to shorten the entire processing time of three-dimensional image processing and to accurately and surely target a region of interest. It is intended to provide a simple volume data clipping method.

[0010]

According to the present invention, among volume data for the purpose of performing three-dimensional image processing,
By pre-designating the gray value that is considered to be included in the processing target, it is possible to limit the range in which the three-dimensional image processing is to be performed based on the pre-designated gray value, and to realize high-speed processing. Characterize.

The x-y-z of the volume data
A two-dimensional tomographic plane perpendicular to each axial direction is read in the direction from the periphery to the center of the volume data to obtain a histogram of gray values, and based on this histogram, a region boundary of a range where three-dimensional image processing should be performed is detected. It is characterized by

Further, the present invention is for storing volume data to be subjected to three-dimensional image processing, and reads out voxel data for each cross section which is perpendicular to each of the x-y-z axis directions orthogonal to each other. 3D image storage means, a gray value input means for inputting a target gray value for voxel data for specifying a region of interest in the volume data stored in the 3D image storage means, and the 3D image storage means Of the voxel data read from the grayscale value input means, the grayscale value of the voxel data corresponding to the target grayscale value input by the grayscale value input means is converted to a predetermined grayscale value, and the predetermined grayscale value is generated by the data conversion means. Histogram measuring means for measuring the voxel data converted into values, and based on the measurement result by the histogram measuring means And a control unit that determines the boundary of the attention area in each of the x, y, and z axis directions and specifies the attention area, limits the range in which three-dimensional image processing should be performed, and realizes high-speed processing. It is characterized by doing.

[0013]

According to such a configuration, the user sets a gray value of interest as a processing target of three-dimensional pixels (voxels) in the input volume, and x of the input volume is set.
-Y-z Limits the target area depending on whether or not the gray value voxel designated by the user exists in the cross section perpendicular to each axial direction. For this reason, it is possible to eliminate the problems of precision deterioration due to sampling of voxel values at coarse intervals and inadvertent exclusion of a region of interest due to size limitation by a method in which the range of the target region is arbitrarily designated by the user.

Further, the reading of the cross section perpendicular to each axial direction is performed by
It is performed in the direction from the periphery of the volume data to the center. In general, the region of interest is often distributed from near the center to the periphery of all the input volume data. Therefore, according to the cross-section reading method in such a direction, the gray value of the cross-section that does not include the specified gray-value is first examined. However, unless the region of interest is extremely small compared to the entire volume data, such cross-section search can be less and the processing time can be shortened.

Further, in cross-section reading for limiting the range to be processed, memory access is generally fast x-.
Since the process is performed in the direction perpendicular to each of the y-z axes and the cross-section that obliquely intersects these axes is not performed, the processing time can be shortened.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a three-dimensional image processing apparatus according to the embodiment to which the volume data clipping method of the present invention is applied. In FIG. 1, a three-dimensional image memory 10 is for storing a three-dimensional image conceptually stacking a large number of cross-sectional images.

The data converter 12 is a three-dimensional image memory 1
The grayscale value of the predetermined voxel data in the image stored in 0 is changed according to an instruction from the control unit 18 described later, and the data thus obtained is output to the histogram measurement unit 14.

The histogram measuring section 14 calculates a histogram table of gray values existing in the voxel data for one section from the data converting section 12, and outputs the density histogram to the control section 18. Pipeline data bus 16
Is used to transfer image data from the three-dimensional image memory 10 to the data conversion unit 12 and from the data conversion unit 12 to the histogram measurement unit 14 at high speed.

The control unit 18 designates a cross section for the three-dimensional image stored in the three-dimensional image memory 10, and the data conversion unit 1
The density of voxel data for 2 is specified, and the histogram measuring unit 14 is instructed to measure the histogram. The control unit 18 designates the density in the data conversion unit 12 according to the target gray value input via the operation terminal 20.

The operation terminal 20 is provided with a display device and an input device such as a keyboard and a mouse, and is used when the user designates a grayscale value of a three-dimensional pixel which is a target of processing by a user in a graphical interface.

Next, the operation of the embodiment will be described with reference to the flow chart shown in FIG. First, the attention area gray value is set in order to specify the range of the target area from the three-dimensional image stored in the three-dimensional image memory 10. Here, the user inputs the designation of the range of the gray value of the area to be the target of the three-dimensional image processing from the operation terminal 20 (step S1). This information is supplied to the data conversion unit 12 via the control unit 18.

The range of gray value is designated by a scroll bar of a graphical interface as shown in FIG. 3, for example. Specifically, when the range between aa ′ in the scroll bar in FIG. 3 is designated as the target gray value, a
Press the mouse button at the point and release the button at the point a '. The control unit 18 determines the target gray value based on the range designated by the scroll bar. This range specification is
Not only one range but also ab ′ can be specified in addition to ab ′, and the range width can be set arbitrarily.

Next, the control unit 18 designates an image to be read out from the three-dimensional image (volume), that is, a section (step S2). Here, the initial orthogonal coordinate axis of the read section is the x-axis, and the initial value of the x-coordinate is the value on the peripheral surface of the volume (after processing on the x-axis, the y-axis, z
Sequentially move to the process for the axis). The control unit 18 first designates a section to be read (the orthogonal coordinate axis z and the coordinate value on that axis), scans the yz section in raster order from the three-dimensional image memory 10, reads voxel values, and reads the pipeline data bus. The data is supplied to the data conversion unit 12 via 16 (step S3). The state of this cross-section reading is shown in FIG.
It shows in (a)-(c).

The data conversion unit 12 determines that the voxel data matching the gray level value of the attention area designated by the control unit 18 is 3
When the density value is supplied from the three-dimensional image memory 10, the density value is converted into a certain grayscale value D and is supplied to the histogram measuring unit 14 via the pipeline data bus 16 (step S4). Here, the voxel data that does not match the focused gray value is converted into a gray value E other than the gray value D and supplied to the histogram measurement unit 14.

The histogram measuring unit 14 calculates a histogram table of existing gray values (statistics of the number of voxels for each density) for the voxel data for one section supplied via the pipeline data bus 16 (step). S5). The processing of steps S1 to S5 described above is executed with the timing adjusted by the activation command from the control unit 18 to each unit.

The control unit 18 checks the number of voxels with respect to the grayscale value D in the histogram table calculated by the histogram measuring unit 14 and checks whether this is "0" or "non-zero" (step S6). Here, when the number of voxels is “0”, that is, as shown in FIG. 4B, the control unit 18
Gives a command to the three-dimensional image memory 10 to shift the x-coordinate of the read section by one voxel toward the center of the volume (step S7), and repeats the processing from steps S2 to S5. On the other hand, when the number of voxels in the histogram is “non-zero” (state shown in FIG. 4C), the control unit 1
8 stores the x coordinate of the cross section at that time as one end point (boundary) of the x coordinate of the region of interest (step S8). In this way, until a "non-zero" voxel is found, that is, two end points (both boundaries) can be checked, step S2
~ The process of S9 is repeated.

The control unit 18 repeats the processing from steps S2 to S9 with the volume boundary opposite to the x coordinate as the starting point. Further, the same processing is repeated for the cross sections perpendicular to the y axis and the z axis (steps S10 and S1).
1) Detect both boundaries of the attention area on each axis. As a result, the coordinates of each vertex of the rectangular parallelepiped circumscribing the attention area are obtained.

In this manner, by inputting the designation of the gray value to be noticed in the input image and checking whether or not the designated gray value exists in each axis direction, the rectangular parallelepiped including the region of interest (circumscribing) can be detected at high speed. Can be determined. Further, since the method of connecting the data conversion unit 12 and the histogram measurement unit 14 by the pipeline data bus 16 is used to realize such processing, the total processing time of the three-dimensional image processing that requires processing time is reduced. It can be significantly reduced.

In order to reduce the amount of data to be processed, the method of coarsely sampling the voxel values is not used. Therefore, the accuracy is not deteriorated, and the method in which the user arbitrarily specifies the range of the target area is also used. Not because
The user interface can be improved in that the problem of excluding the attention area due to carelessness in size limitation can be eliminated.

[0030]

As described above, according to the present invention, it is possible to shorten the entire processing time of the three-dimensional image processing and to accurately and surely target the attention area. is there.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a three-dimensional image processing apparatus to which a volume data clipping method of the present invention is applied.

FIG. 2 is a flowchart for explaining the operation in one embodiment of the present invention.

FIG. 3 is a diagram showing an example of a graphical interface according to an embodiment of the present invention.

FIG. 4 is a diagram for explaining a cross-section read state in one embodiment of the present invention.

[Explanation of symbols]

10 ... Three-dimensional image memory, 12 ... Data converter, 14 ...
Histogram measurement unit, 16 ... Pipeline data bus,
18 ... Control part, 20 ... Operation terminal.

Claims (3)

[Claims] 1. A range in which three-dimensional image processing is to be performed by previously designating a gray value that is considered to be included in a processing target in volume data intended for three-dimensional image processing. A volume data clipping method which is characterized in that the processing is speeded up by limiting based on a gray value specified in advance. 2. A two-dimensional tomographic plane perpendicular to the x, y, and z axis directions of the volume data is read in the direction from the periphery of the volume data toward the center to obtain a histogram of gray values, and based on this histogram. 2. The volume data clipping method according to claim 1, wherein an area boundary of a range in which three-dimensional image processing is to be performed is detected. 3. A three-dimensional voxel data for storing volume data to be subjected to three-dimensional image processing, wherein voxel data for each cross section perpendicular to each x-y-z axial direction orthogonal to each other is read out. An image storage unit; a grayscale value input unit for inputting a target grayscale value for voxel data for specifying a region of interest in the volume data stored in the three-dimensional image storage unit; and reading from the three-dimensional image storage unit. Data conversion means for converting the gray value of the voxel data of the voxel data corresponding to the target gray value input by the gray value input means into a predetermined gray value, and the predetermined gray value by the data converting means. Histogram measuring means for measuring the voxel data, based on the measurement result by the histogram measuring means,
A control means for determining the boundary of the attention area in each of the x-y-z axis directions and specifying the attention area, and limiting the range in which the three-dimensional image processing should be performed, thereby speeding up the processing. Volume data clipping method characterized by realization.