JPS61194576A - Image processor - Google Patents

Abstract

PURPOSE:To set window width and a window level by constituting an image processor capable of setting automatically window level and window width in window processing of image data of small scale hardware. CONSTITUTION:The window level calculation means 5 calculates the average image concentration based on the image data and furnishes same to the window processing means 3 as the window level. The window width calculation means 6 calculates the output of the window level calculation means 5 and from the image data calculates the dispersion value of the image concentration and outputs this as the window width. To the output of this window width calculation means 6, a multiplier 7 for multiplying a coeff. and a coeff. variable section 8 to modify the coeff. to adjust the output of the window width calculation means 6 is provided to the window width adjusting means 9 outputting to the window processing means 3. By so doing, with small scale hardware, automatic setting of window level and window width in window processing is made possible.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an image processing device that performs image processing of image data to be displayed.

[Technical background of the invention and its problems 1 So-called image diagnostic equipment usually has a data acquisition section.

It is composed of a data processing section and an image display section, of which the image display section is located at the final output section of the image diagnostic equipment and plays an important role as a point of contact with the doctor's eyes.

However, the image data output from the data processing section has a very wide dynamic range, and the signal values are often concentrated in a certain range. If the dynamic range of the image data output from the data processing is made to correspond one-to-one to the dynamic range of the image display section, the display will be extremely lacking in contrast, or the entire image will be too dark or too bright. This makes the image unsuitable for diagnosis.

Therefore, a method of so-called look-up table conversion (particularly called window conversion in medical equipment) has been adopted in which a certain specific range is selected and made to correspond to the display range of the image display unit. This is as shown in Figure 6.
This is done by inputting the density value of the image data from the image memory 4 into the address line of the look-up table memory 25, and displaying the selected memory contents on the display section M2. The lookup table memory 25 stores a function as shown in FIG. 7, for example, and has two parameters (1, window level WL, window mWWf7). The address on the horizontal axis in the figure is the address of the lookup table memory 25, which corresponds to the output density of the image memory 4, and the data on the vertical axis means the output density of the lookup table 25. do. The image diagnostic apparatus is usually provided with a function key as shown in FIG. 8(a'), and by operating this function key, the window level WL is adjusted.
, window width WW, respectively, as shown in FIG. 8(b).

It can be changed as shown by the broken line in (C).

In this way, by adjusting the window level and window width, it is possible to set the brightness (11 degrees) and contrast of the displayed image to the optimum state, respectively, and it is possible to display images with excellent medical diagnostic performance. .

However, since the window level and window width differ depending on the type of image data, for example, a comprehensive image diagnosis system (PA'C8: Picture A
archive and communicati
On System), etc., the window level and window width must be adjusted according to various modalities (X-ray, ultrasound, 0-cell, etc.), and when it comes to the above-mentioned dynamic adjustment, it is extremely difficult. It becomes troublesome.

Therefore, methods of automatically setting the window level and window width have been tried in recent years. For example, a method using a density histogram can be considered. That is, the ninth
In the case of the concentration probability distribution as shown in Figure 9(a), after integrating this in the concentration direction as shown in Figure 9(b), it corresponds to, for example, 10% and 90% of the concentration probability distribution. The window level WL and window width WW are set based on the detected density.

Here, the above-mentioned lr! In order to create the probability distribution of I, we first need to find the concentration probability density function.

FIG. 10 is a block diagram showing the configuration of hardware for creating this concentration probability density function. The image density sequentially read out from the image memory 4 is transferred to the address line of the histogram memory 26, thereby accessing the address of the histogram memory 26. The address of this histogram memory 26 corresponds to the concentration axis in the concentration probability density function, and the memory contents correspond to the function value.

Then, the data read from the histogram memory 26 is incremented by 1 in an adder 27, and is written to the same address in the histogram memory 26 again. In this way, the concentration probability density function is obtained.

However, in the configuration shown in FIG. 10, assuming that the number of bits in the image density direction is 12 bits, the memory size of the histogram memory 26 is 2 + 2 = 4.
However, the number of bits in one word of this histogram memory 26 is 18 pits and 10 bits in the image matrix size 512 x 512.
For 24x1024, 20 bits are required, and the histogram memory 26 alone becomes large-scale in terms of hardware.

[Object of the Invention] The present invention has been made in view of the above circumstances, and its object is to provide an image processing device that can automatically set the window level and window width in window processing of image data. The goal is to provide this on small-scale hardware.

[Summary of the Invention] The outline of the present invention for achieving the above object is to calculate an average value of image density from image data and output this as a window level to a window processing means. a window level calculation means; a window width sieving means for calculating a variance value of image density from the output of the window level locking means and the image data and outputting this as a window width; and an output of the window width calculation means. It is characterized by comprising a multiplier for multiplying by a coefficient and a coefficient variable part for varying the coefficient, and a window width modification means for modifying the output of the window width calculation means and outputting it to the window processing means, This allows automatic setting of window level and window width in window processing using small-scale hardware.

[Embodiments of the invention] Hereinafter, the present invention will be specifically explained with reference to Examples.

FIG. 1 is a block diagram showing the configuration of an image processing apparatus according to an embodiment of the present invention. As shown in FIG. 1, the device of this embodiment includes a window processing means 3 that performs window processing of image data to be displayed on a display device 2, and an image density based on image data transmitted from an image memory 4. A window level calculation means 5 calculates the average of the values and outputs this as a window level to the window processing means 3, and calculates the image density from the output of the window level calculation means 5 and the image data transmitted from the image memory 4. A window width calculation means 6 that calculates a variance value and outputs it as a window width, a multiplication section 7 that weights the output of the window width calculation means 6 by multiplying it by a coefficient, and a multiplication section 7 that is used for multiplication. The window width modification section 8 comprises a coefficient variable section 8 that can vary the coefficients to be calculated, and a window width modification section 9 that modifies the output of the window width calculation section 6 and outputs it to the window processing section 3.

The window level calculation means 5 includes the image memory 4
The adder 10 adds the output of a delay device 11 (described later) to the image data transmitted by the adder 10, and the delay device 11 delays the output of the adder 10. As a result of the execution of the calculation, the image density is "(x)" indicates the image data transmitted from the image memory 4, and N indicates the total number of pixels. We assume that it can be considered as a one-dimensional shape.

Also, for convenience of explanation, a block that performs the calculation 1/N is omitted.

The window width calculation means 6 includes a subtracter 12 that subtracts the output of the window level calculation means 5 from the image data transmitted from the image memory 4, and a subtractor 13 that squares the output of the subtractor 12. and an adder 14 that adds the output of a delay device 15, which will be described later, to the output of this multiplier 13;
A delay device 15 is provided to delay the output of the adder 14, and by executing the calculation shown in equation (2), an image*m degree separation, 2, is output. - (2) For convenience of explanation, the block that performs the calculation 1/N2 is omitted.

Next, details of the coefficient variable section 8 in the into width adjusting means 9 will be explained based on FIG. 2.

This coefficient variable section 8 has first coefficients having different center frequencies.
．． Second. a third bandpass filter 16°17.18; Second. Third bandpass filter 16,1
7.18 Calculating the variance value of the image based on each output 1st. Second. a third dispersion section 19.20.21; Second. Maximum value detection unit 2 that compares the outputs of the third dispersion unit 19, 20, and 21 and detects the one showing the largest value.
2, and a table memory 2 in which coefficients as a visual sensitivity correction function are stored in advance, and the stored contents are read out in accordance with the output of the maximum value detection section 22.
3.

Here, the above 1. Second. Third bandpass filter 1
The filter characteristics of the filters 6, 17, and 18 are as shown by a, b, and c in the characteristic diagram of FIG. Only the middle and low frequency components are present, and the second
Only the middle frequency components pass through the third band pass filter 17, and only the high frequency components pass through the third band pass filter 18.

In addition, the above-mentioned No. 1. Second. The configuration of each of the third dispersion sections 19, 20 and 21 is basically that of the window width calculation means 6.
It is similar to

Next, the operation of the embodiment device configured as described above will be explained.

The image data transmitted from the image memory 4 is input to each of the window processing means 3, the window level calculation means 5, and the window width calculation means 6.

The window level calculation means 5 executes the calculation according to the above equation (1) on the input image data, and outputs the calculation result to the window processing means 3 as a window level. Now, the image data is shown in Figure 4 f(x)
Then, the output of the window level calculating means 5 will be as shown by AV (average value) in the figure.

Further, the window width calculation means 6 executes the calculation according to the above equation (2) based on the input image and the output of the window level calculation means 5, and the calculation result is sent to the multiplication unit in the window width modification means 9. 7, and is subjected to multiplication with the coefficient output from the coefficient variable section 8. Here, the output of the window width calculation means 6 is shown in FIG. 4 BR (variance).
It will be shown as follows.

Further, as will be described in detail later, the multiplication in the multiplication section 7 is for modifying the window width according to the visual sensitivity, and the multiplication result is sent to the window processing means 3 as the modified window width. Output.

Here, the details of the operation of the window width adjusting means 9 will be explained.

1st. Second. Third bandpass filter 16°17.1
The specific frequency components that passed through the first . Second
．． The third dispersion unit 19, 20.21 calculates the dispersion value. The calculation result is then sent to the maximum value detection section 22
is input, and the frequency range with the largest dispersion value is detected. In accordance with this detection result, a coefficient as a visual sensitivity correction function is read from the table memory 23, and this read coefficient is output to the multiplier 7 and multiplied by the output of the window processing means 6. Served.

The reason why the weight of the window width is changed in accordance with the frequency distribution state of image data is as follows.

Various studies have been conducted on the psychological perception of the physical brightness and brightness of images, and the visual characteristics of still images are as shown in FIG.

FIG. 5 is a visual frequency characteristic diagram, in which the horizontal axis represents spatial frequency and the vertical axis represents visual sensitivity.

Further, the parameter is the average brightness of the image. If the image is bright enough, for example, if you look at the curve for average brightness 9,
It can be seen that the way you feel varies by about 10 times depending on the frequency of the image. In other words, a smooth image (both edges with low frequency) and a fine image (image' with high frequency) have the same amplitude, but the perception is about 10 times different.

Therefore, in consideration of such visual sensitivity, coefficients as correction functions are stored in advance in the table memory 23, and the highest frequency range of the frequency distribution state of the image data to be displayed is set to the first. Second. Third bandpass filter 16.1
7.18 and 1st ° 2nd. Third dispersion section 19, 20.2
1, and the window width, that is, the contrast in the displayed image is corrected using the coefficients read from the table memory 23 in accordance with the detection results. Specifically, for image data to be displayed that has many components in frequency ranges with high visual sensitivity, the window width is widened to lower the contrast, and for image data with low visual sensitivity, For those having many components in the frequency range, the window width is narrowed to increase the contrast.

Window processing according to the window level and window width automatically set in this manner is performed by the window processing means 3, and the processing results are displayed on the display device 2. Since the image display on the display device 2 has been subjected to window processing, the display has excellent medical diagnostic performance.

As explained above, in the present embodiment device, the average value of image density is calculated from the captured image data, the variance value of the image density is calculated from this calculation result and the image data, and the calculated average value is calculated from the image data. Since this method automatically sets the window level and window width in window processing using the value 1 variance value, and does not use the histogram memory 26 as explained based on FIG. 10, it is extremely advantageous in terms of hardware configuration. Become.

Further, the coefficient variable section 8 in the window width modification means 8 is configured to include a table memory 23 in which coefficients as a correction function of visual sensitivity are stored in advance, and moreover, Accordingly, the table memory 23 read out the coefficient, and this coefficient is applied to the multiplier 7.
Since the contrast of the displayed image is corrected according to the visual sensitivity by applying the multiplier to the multiplier, visually optimal brightness adjustment can be quickly performed.

Although one embodiment of the present invention has been described above, it goes without saying that the present invention is not limited to the above-mentioned embodiment, and can be modified as appropriate within the scope of the gist of the present invention.

For example, by increasing the number of bandpass filters and dispersion sections shown in FIG. 2, it is possible to improve the accuracy of correction according to the visual sensitivity of contrast compared to the above embodiment.

The function of the coefficient variable section 8 in the above embodiment is to find out in which frequency range the frequency components of the image are most present, and the peak frequency is detected by a relatively simple method. , the average frequency ω of the image
It may also be possible to detect. This average frequency 7 is m=(1/N)ΣVar [f (7) ] number f ω
It can be expressed as the dispersion of the band image yar: rω where ω: frequency fω: center wave number ω, and the calculation of equation (3) can be performed.

Note that the peak frequency and the average frequency do not necessarily match.

Further, the coefficient variable section 8 in FIG. 1 may be configured with a coefficient adjuster, such as a variable resistor, which can manually vary the coefficients used for multiplication by the multiplication section 7. With this configuration, it is impossible to automatically adjust the contrast of the front thread image according to visual sensitivity, but the hardware configuration is smaller than in the above embodiment. It goes without saying that even in this case, it is possible to automatically set the window level.

[Effects of the Invention] As detailed above, according to the invention, the window level in window processing of image data can be improved.

In addition, an image processing device that can automatically set the in-toe width can be provided using small-scale hardware. By adding the image processing device according to the present invention to, for example, a comprehensive image management system, it can greatly contribute to improving medical diagnostic performance.

[Brief explanation of drawings]

FIG. 1 is a block diagram of an image processing device that is an embodiment of the present invention, FIG. 2 is a block diagram showing details of the coefficient variable section in FIG. 1, and FIG. 3 is a block diagram of the bandpass filter in FIG. 2. 4 is a waveform diagram for explaining the operation of the apparatus of this embodiment, and □FIG. 5 is a visual frequency characteristic diagram, FIGS. 6 and 7. Figures 8, (a), (b), and (c) are block diagrams and explanatory diagrams for explaining the conventional window processing method, respectively, and Figures 9 (a) and (b) are density histograms. FIG. 10 is a block diagram showing the configuration of hardware for creating a concentration probability density function. 1... Image processing device, 3... Window processing means, 5... Window level calculation means, 6...
. . . window width calculation means, 7 . . . multiplication unit,
8...Coefficient variable section, 9...Window width adjustment means, 23...
・Table memory. Figure 3 n 1/2 @ 1F skin included C/deg Figure 8 (C)

Claims (3)

[Claims] (1) In an image processing device comprising window processing means for performing window processing of image data according to a set window level and window width, an average value of image density is calculated from the image data, and this is window level calculation means for outputting a window level to the window processing means; window width calculation means for calculating a variance value of image density from the output of the window level calculation means and the image data and outputting this as a window width; , a multiplication section that multiplies the output of the window width calculation means by a coefficient, and a coefficient variable section that changes the coefficient, and a window width modification means that modifies the output of the window width calculation means and outputs it to the window processing means. An image processing device comprising: (2) The coefficient variable section in the window width modification means is configured to include a table memory that stores in advance coefficients as a correction function of visual sensitivity, and the coefficient variable section in the window width modification section is configured to have a table memory that stores coefficients as a correction function of visual sensitivity in advance, and adjusts the coefficients in the table memory according to the frequency distribution state of the image data. 2. The image processing apparatus according to claim 1, wherein the coefficients used for multiplication by the multiplication unit are varied by reading out coefficients from the multiplication unit. (3) According to claim 1, the coefficient variable section in the window width modifying means includes a coefficient adjuster that manually varies the coefficients used for multiplication in the multiplication section. The image processing device described.