JPH01130287A - Gray level histogram arithmetic unit - Google Patents

Abstract

PURPOSE:To obtain a gray level histogram in a real time by deriving the concentration histogram by utilizing an output of a window conversion. CONSTITUTION:A window processor 1 brings an original image stored in an image memory to window conversion under an optimum window converting condition, and this result is stored in a frame memory 3. An image data stored in the frame memory 3 is supplied to a CRT display 9 through a D/A converter 8. A gray code outputted from the window processor 1 is written as an address in a histogram memory 13. An MPU 5 reads a data of each address of the histogram memory 13 and obtains a gray level histogram.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a density histogram calculation device for obtaining density histograms of medical images and the like.

(Prior Art) For example, when displaying a medical image such as a CT image transmitted by an X-ray CT device on a display for diagnosis, it is necessary to calculate the density of each pixel of a large number of pixels that make up the image. A histogram of density CCT values as shown in FIG. 9 is created. In Figure 9, the vertical axis is frequency;
The horizontal axis indicates the density, and by distributing the determined pixel densities on the same plane, it is possible to know the maximum density, minimum density, average density, etc. of the image. Diagnosis can be made based on such results.

Conventionally, in order to obtain such a density histogram, a processor was prepared, which read the density of each pixel, divided into preset density groups, and counted the frequency. ing. However, this calculation method of calculating the density of each pixel using the J processor takes a long time, and in practice, it involves setting an ROI (region of interest) at an arbitrary position in the image and moving this ROI arbitrarily to obtain the density histogram of the ROI. In many cases, the processing method used in this paper cannot achieve a calculation speed sufficient for practical use.

(Problems to be Solved by the Invention) As described above, since the processing speed of the conventional density histogram calculation device is slow, there is a problem that it is impossible to obtain a density histogram in real time.

The present invention has been made in response to the above-mentioned problems.
It is an object of the present invention to provide a density histogram calculation device that can obtain a density histogram in real time.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, the present invention converts the frequency of data of a large number of pixels into a Gray code by making good use of window transformation, and converts the frequency into a Gray code. The code is written as an address in the histogram memory, the data for each address of this histogram is read out and added to an incrementer, and the output of this incrementer is re-embedded at the same address in the histogram memory.

(Operation) By using the Gray code output from the window transformation, this is written into the histogram memory as an address. The data at each address in the histogram memory is added to the incrementer, and the data at each address in the incrementer is added to the incrementer.
is rewritten to the same corresponding address in the histogram memory. As a result, the frequency of each pixel density corresponding to the Gray code is counted by the histogram memory, so that a density histogram is obtained. Since such a Wffi operation is performed using the output of the window transform, it is performed at a speed proportional to the window transform speed. Therefore, calculation processing can be performed at high speed, and a density histogram can be obtained in real time.

(Embodiment) FIG. 1 is a block diagram showing an embodiment of the leanness histogram calculation device of the present invention, in which a wind processor 1 converts an original image (for example, a CT image) stored in an image memory 4 under optimal window conversion conditions. This is for performing window conversion and storing the result in the frame memory 3. The image memory 4 stores CT value data in, for example, 12 bits, and this CT value data is window-converted and compressed into, for example, 8-bit data, which is then stored in the frame memory 3. Wind table 2 contains window widths (W, W) that determine the range of displayed CT values in accordance with the window conversion conditions.
) and the wind level (W, L) that determines the CT value at the center of the range.

Image data is stored in frame memory 3 as a display image.
This image data is D/A converted by a D/A converter 8 and then displayed as an image on a CRT display 9. In other words, the CT of the subject measured by an X-ray CT device, i.e., the portion of ifi, is converted into shading using window conversion conditions set so that the area to be observed is displayed with the optimum contrast. The image will be displayed on the display 9.

The bit plane 7 has a bitmap display function, and graphics controller 6 performs graphic display at high speed.

M P LJ 5 consists of a dedicated processor,
Performs control operations for the entire device. ROI input means 12 is 71
When an image is displayed on the display 9, ROI can be placed at any position.
When set, an ROI of arbitrary shape such as a quadrilateral, a circle, an ellipse, etc. based on the setting is displayed on the image at high speed by the function of the graphic controller 6. Further, this ROI is configured to be movable to any position and also configured to be enlarged and reduced. 10. ,．
11 is a data bus.

The wind processor 1 performs CT according to the window conversion conditions determined from two parameters W, W and W, L.
Determines the gray code for display from the value and outputs it.

Figure 2 shows this kind of window conversion.
The vertical axis represents the Gray code, which is made up of, for example, 2B=256 steps, and the horizontal axis is the CT value, which is made up of, for example, 2+2=4096 steps.

The 12-bit CT value is compressed to correspond to the 8-bit Gray code, and for this reason, it is divided into multiple groups, and each group is represented by one CT value D1, D2, D3, etc. is taken. For example, CT value D
The group represented by 1 includes a range of CT values, all of which correspond to Gray code 1. Such window conversion conditions can be stored in the window table 2. ' The Gray code output from the window processor 1 is written into the histogram memory 13 as an address. As shown in FIG.
It has 256 levels of addresses up to 5, and each address corresponds to a Gray code. For example, Gray code 2 output from the window processor 1 is written as address 2 in the histogram memory 13. The contents of the histogram memory 13 are all reset to O in the initial state. The data at each address in the histogram memory 13 is added to the incrementer 14, and the incrementer 14 increments the data (adds 1) and rewrites the data to the same address in the histogram memory 13.

Next, the operation of this embodiment will be explained.

If, for example, Gray code 2 is written from the window processor 1 to the histogram memory 13 as address 2 while the contents of the histogram memory 13 are all retted to O, then the data O(O
) is added to the incrementer 14. The incrementer 14 increments this data O (that is, adds 1) and rewrites the resulting 1 to address 2 of the histogram memory 13, which is the same address. As a result, data 1 will be stored at address 2. FIG. 4(a) shows this situation. Next, suppose that, for example, Gray code 4 is written as address 4 in the histogram memory 13 from the window processor 1, the same principle can be used as shown in FIG. 4(b).
Data 1 will be stored at address 4 as shown in FIG.

Next, if Gray code 2 is written again to address 2 of the histogram memory 13, the already stored data 1 is added to the incrementer 14 from this address 2. Then, the incrementer 14 rewrites the incremented data 2 to address 2. As a result, data 2 is stored at address 2 as shown in FIG. 4(C).

After writing the Gray code to all addresses in this way, the MPU 5 reads the data from address 1 to address 255 in the histogram memory 13, thereby writing each Gray code corresponding to each address. The number (frequency) can be counted. Roughly based on the relationship shown in FIG. 2, a density histogram as shown in FIG. 5 can be obtained by replacing the CT values Di, D2, D3, . . . corresponding to each gray code and indicating the frequency. These operations can be easily performed by using the functions of the window processor 1 and the σ graphics controller 6. Since window conversion processing can normally be performed at a processing speed of 10 to 30 times/second, the density histogram calculation process can be performed in the reciprocal of 1710 to 1730 seconds. Therefore, since the calculation can be performed at high speed, the density histogram can be obtained in real time.

Further, when an image is displayed on the display 9, by setting the ROI 15 at an arbitrary position on the image using the ROI input means 12, an ROI 115 is placed within the ROI 15 on the image as shown in FIG. Only the concentration histogram can be displayed. Further, this density histogram can also be displayed outside the ROI 15 as shown in FIG.

By roughly moving, enlarging, or reducing the ROI 15, the density histogram can be displayed each time. Furthermore, as shown in FIG. 8, it is also possible to display the density histogram only for an arbitrary specific area 16 on the image.

Set the ROI only at the required position on the image and use this ROI.
Since a method for determining the concentration histogram within the range is in high demand in practice, it is extremely effective to be able to perform such processing at a high calculation speed.

The image compression method shown in the main embodiment is merely an example, and its contents can be arbitrarily changed as necessary.

[Effects of the Invention] As described above, according to the present invention, since the density histogram is obtained using the output of the window transformation,
It enables high-speed calculation processing and allows concentration histograms to be obtained in real time.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the concentration hiss 1 to gram numbing device of the present invention, FIG. 2 is an explanatory diagram of a window conversion method by the device of this embodiment, and FIG. 3 is a histogram memory of the device of this embodiment. 4(a) to 8(C) are explanatory diagrams of the principle of density histogram calculation according to the present invention. FIG. The figure is a display example of a density histogram according to another embodiment of the present invention, and FIG. 9 is an explanatory diagram of the density histogram. 1... Wind processor, 2... Wind table, 3... Frame memory,
4... Image memory, 12... ROI input means, 13
...Histogram memory, 14...Incrementer. Agent Patent Attorney Rules
Takeshi Kondo Figure 3 (C) Funeral Figure 8 Figure C9

Claims (2)

[Claims] (1) In a density histogram calculation device that obtains a density histogram of an image by determining the density of each pixel of a large number of pixels constituting the image, a window conversion means that converts data of a large number of pixels into a gray code, and a window conversion means. a histogram calculation means that inputs the output gray code as an address and counts the frequency of each address; and a histogram calculation means that reads data corresponding to the address from the histogram calculation means, increments this data, and counts it to the same address of the histogram calculation means. 1. A density histogram calculation device comprising: an incrementer that rewrites data as data to be processed. (2) The density histogram calculation device according to claim 1, further comprising ROI input means for setting an ROI at an arbitrary position in an image, and in which only this ROI is subjected to histogram calculation.