JPH0828844B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to an X-ray CT apparatus, a magnetic resonance imaging apparatus,
Nuclear medicine diagnostic device or DSA (Digiatl Subtraction Argiogr
aphy) and the like.

(Prior Art) Generally, when performing image diagnosis by CRT, either high contrast resolution (hereinafter abbreviated as high-con) or low-contrast resolution (hereinafter abbreviated as low-con) is emphasized depending on the target region and diagnostic purpose. In this case, the noise level is almost constant depending on the image. Therefore, it is usual to create an image according to the purpose by changing the shape of the image filter or the like.

That is, when it is desired to attach importance to the high contrast, as shown in FIG. 9 (A), the CT value of the target portion is yc for the noise level 6c.
Since the contrast difference is sufficiently large, the diagnosis can be sufficiently performed as it is, and further, in order to emphasize the contrast difference, edge enhancement may be performed by emphasizing a high frequency region.

On the other hand, if you want to attach importance to the low-profile converter, you can use the function shown in Fig. 9 (C).
As shown in, the contrast difference with the surroundings of the target is not so large compared to the noise level (the CT value of the target region is, for example, 1/3 of that in the case of emphasizing high-con). Therefore, an image filter (high-cut filter) that reduces the high frequency region is operated to relatively reduce the noise level as shown in FIG. 9 (E).

The above explanation is on the CT value scale, but consider below on the gray level. Normally, CRT diagnosis has a window function, and by controlling the window width (hereinafter abbreviated as WW), the difference in the dynamic range of the target CT value is controlled. In other words, if high-concentration is important, take a large WW,
If low-concentration is important, WW should be small. Then,
The images after windowing the high-concentration-oriented and low-concentration-oriented cases shown in FIGS.
As shown in FIGS. 9B and 9D, the difference in gray level of the target region is adjusted to the same degree and displayed. Further, as a result, the noise level increases in inverse proportion as WW is made smaller (in the examples of FIGS. (B) and (D), the noise level with emphasis on low-conversion is 3 compared to the case with emphasis on high-conversion. Therefore, in the case of emphasizing low contrast, the high cut filter described above is used to cut the high frequency region as shown in FIG. 9 (F) to reduce the noise level on the gray level. It is decreasing.

In this way, if a high-cut filter is used so that the gray level of noise is almost the same as in the case of emphasizing high-con in the case of emphasizing low-con, then simply thinking, both high-con and low-con / N can be about the same.

(Problems to be Solved by the Invention) In general, it is often decided whether to attach importance to high-con or low-con, depending on the target. For example, brain integrity emphasizes low-con, and bones and lung fields emphasize high-con. ing.

By the way, in the case of the chest, there are cases where the lung field emphasizes high-con, the vertical section emphasizes low-con, and even in the same lung field, the high-con emphasizes more in pneumoconiosis and the low-con emphasize in lung cancer. We need to create an image with several different filters.

However, making an image by changing several kinds of filters is complicated in practical use, and it is difficult for the user to determine the optimum filter. It was the current situation to match.

Therefore, an object of the present invention is to automatically determine an appropriate filter for reducing the S / N on the gray level according to which of high contrast resolution or low contrast resolution is important and its degree. Another object of the present invention is to provide an image processing device that can execute filtering by using the above method.

[Structure of the Invention] (Means for Solving Problems) The present invention has a window operation unit and a display unit having a window processing function, and an image displayed on the display unit or the display unit. In an image processing device for processing an image formed on an image forming surface such as a screen or film,
By inputting the window width information set in the window operation unit, the cutoff frequency of the high frequency cut filter is increased when the window width is large, and the cutoff frequency of the high frequency cut filter is decreased when the window width is small. An image processing apparatus is configured by providing a filter generator that generates a characteristic filter, and a filtering unit that filters an image displayed on the display unit or an image combined on the image plane from the display unit based on the filter. are doing.

(Working) The degree of emphasis on high contrast resolution or low contrast resolution is almost correlated with the window width WW. That is, WW is large in the case of emphasizing high-con, and WW is small in the case of emphasizing low-con. Therefore, paying attention to the above facts, by generating a high-frequency cut filter with different cutoff frequencies based on the WW information, the S / N ratio on the gray level in the case of emphasizing high-conversion and low-conversion can be made approximately the same. Filtering is performed so that
As the filtering means, in addition to filtering the image displayed on the display unit digitally or in an analog manner, the image formed on the image plane of the screen, film or the like from the display unit by the imager, such as imager photographing, can be processed in an analog manner. May be filtered to.

(Example) Hereinafter, the present invention will be specifically described based on illustrated examples.

FIG. 1 is a block diagram of an apparatus according to an embodiment of the present invention. In the figure, reference numeral 1 denotes a window operation unit, which is composed of, for example, a window width knob (WW knob) and a window level knob (WL knob), and has a desired WW and WL.
The setting value corresponding to is input. A window controller 2 controls the setting of the WW and WL in the window processing section in the display section 5 on the basis of the setting value input by the window operation section 1. The window controller 2 outputs WW information to the filter generator 6 described later. A data storage unit 3 stores the original image to be displayed on the display unit 5. The original image stored in the data storage unit 3 is input to the display unit 5 via the selector 4.

The filter generator 6 determines an appropriate filter shape according to the high-concentration-oriented or low-concentration-oriented, based on the WW information from the window controller 2. Then, in this filter generator 6, when the WW is large, the cutoff frequency f _c of the filter is increased,
When the WW is small, the filter cutoff frequency f
A filter with a low _c is generated in real time.

FIG. 2 is a diagram showing an example of a method of controlling the cutoff frequency of the filter by WW, and the shape of the filter is shown as a gain with respect to the frequency in the radial direction as a rotation target. In this case, the cutoff frequency f _c is continuously controlled by WW as shown by the filter A in the figure, and WW is divided into several stages,
The filter B, which is tabulated for each stage, may be switched and used. When the gradual filter B is used, it is preferable that the number of stages and the like can be set arbitrarily. Here, it means that the lower the cut-off frequency f _c of the high-pass filter is f _c3 → f _c2 → f _c1 , the wider the cut frequency range is toward the lower range.

The filtering calculator 7 is based on the filter generated by the filter generator 6, and is based on the data storage unit 3
The original image from is digitally filtered and output.

The operation of the apparatus configured as above will be described.

First, the selector 4 is switched to the data storage unit 3 side, and the original image of the data storage unit 3 is input to the display unit 5 via the selector 4 and displayed. The operator operates the window operation unit 1 while viewing the image displayed on the display unit 5 to set the window. The window controller 2 determines WW and WL based on the information from the window operation unit 1 and outputs them to the window processing unit of the display unit 5, while outputting the WW information to the filter generator 6. In this filter generator 6, when the WW is large, the cutoff frequency of the filter is increased and
When is smaller, the cutoff frequency of the filter is lowered and the filter shape is determined by either the filter A or the filter B shown in FIG. Then, the filtering arithmetic unit 7 performs the filtering by digitally performing the above-mentioned filter on the original image from the data storage unit 3. This processed image is input to the display unit 5 via the selector 4 switched in advance, and is displayed after being subjected to window processing based on the WW and WL in the window processing unit.

This display image is filtered by a high-frequency cut filter with a lower cutoff frequency as the emphasis on the low contrast becomes higher, so the S / N ratio on the gray level in the case of emphasizing the high contrast and low contrast is almost the same. Can be Moreover, the above filtering process can be executed in real time following the operation on the window operation unit 1.

Here, when diagnosing the chest or vertebral body as the same object, the degree of emphasis on high-containers and low-containers is different as shown in the table below.

Even in the above case, in the present invention,
Just by setting the WW according to the degree of emphasis on low-con, the appropriate filter can be automatically determined and filtering can be performed.

The present invention is not limited to the above embodiment,
Various modifications can be made within the scope of the present invention.

For example, the filter generator 6 may have the configuration shown in FIG. The filter generator 6 shown in the figure includes a first storage unit 6A for storing the high-frequency cut filter f ₁ , a second storage unit 6B for storing the high-frequency enhancement filter f ₂ , and
A coefficient generator 6C for generating a coefficient α according to the window width WW, and a first multiplier 6D for executing a multiplication of α · f ₁ .
A second multiplier 6E for executing the multiplication of (1-α) · f ₂ and α
· F ₁ + is composed of a (1-α) · f ₂ adder 6F to perform an addition. At this time, the coefficient α is W.
Either a method in which it is set to a value that decreases continuously as W increases, or a coefficient α with different values corresponding to multiple types of window width areas is stored, and a coefficient α that does not depend on the window width Any method selected and generated may be used. As described above, since the two types of filters f ₁ and f ₂ are weighted by the coefficient α, a desired filter f ₃ according to WW can be obtained, and only the two types of filters f ₁ and f ₂ are stored. It is possible to set a filter with a high degree of freedom simply by performing. If a desired filter is obtained by weighting two or more types of filters, the degree of freedom is further increased. Moreover, since a high-frequency enhancement filter is adopted as one of them, a desired frequency band can be emphasized.

The filtering means is not limited to the filtering calculator 7 based on the digital calculation described above. In digital filtering operations, especially when the number of image matrices is large (256 × 256 to 1024 × 102
4) Real-time calculation is often difficult.

In such a case, processing equivalent to real-time filtering can be performed by an analog method.

FIG. 6 is a diagram showing an example of performing optical filtering when a display image is film-captured by an imager. In the figure, 5A is a display surface of the display unit 5, and 10 is an image forming surface (film surface) of the imager. It should be noted that when screen projection is carried out, 10 is exactly the same as the screen surface. Reference numeral 11 is a lens for converging an optical image on the image plane. Here, since applying the high-cut filter is equivalent to blurring the image formed on the image forming surface 10, for example, by moving the lens 11 in the plus or minus direction, the focus position is shifted to As shown in the figure, as a result, filtering in which the cutoff frequency of the high-frequency cut filter is variable can be realized. In particular, in medical diagnosis, it is often the case that the diagnosis is finally performed on the film, so it is meaningful to secure real-time filtering by an optical method and improve the S / N on the film as in this embodiment. is there.

FIG. 8 shows the CRT5 as the display unit using the method shown in FIG.
It is the figure which showed the example implement | achieved by. In the figure, 5B is an electron gun, 5C is a beam deflection and focusing coil, 5D is a power supply for supplying a current to this coil to generate a magnetic field, and 5F
Indicate the fluorescent screens, respectively. According to the same principle as in FIG. 6, a high cut filter can be realized by defocusing the electron beam. Therefore, by slightly changing the current passed through the coil 5C, the beam converged diameter d on the phosphor screen 5F can be varied to perform a process equivalent to filtering by a high-cut filter.

[Effects of the Invention] As described above, according to the present invention, a high-frequency cut filter having a different cutoff frequency is automatically determined according to the window width, and filtering is performed based on this filter, whereby a high contrast is obtained. It is possible to reduce the S / N on the gray level depending on whether the emphasis is on the resolution or the low contrast resolution and the degree thereof.

[Brief description of drawings]

FIG. 1 is a block diagram of an apparatus according to an embodiment of the present invention, FIG. 2 is a characteristic diagram showing a relation between a window width and a cutoff frequency of a high cut filter, and FIG. 3 is a relation between a cutoff frequency and a gain. 4 is a block diagram showing a modified example of the filter generator, FIG. 5 is a characteristic diagram showing the relationship between the coefficient α in the coefficient generator shown in FIG. 4 and the window width,
FIG. 6 is a schematic explanatory diagram showing an example of performing filtering by an optical method, FIG. 7 is a characteristic diagram showing the relationship between the lens position and the cutoff frequency, and FIG. 8 is that the convergent diameter of the electron beam is variable. 9A to 9F are schematic explanatory diagrams showing an example in which filtering is performed in FIG. 9, and FIGS. 9A to 9F are characteristics showing CT value levels and gray levels in each case of applying high cut filter to high-concentration-oriented, low-concentration-oriented and low-concentration-oriented images It is a figure. 1 ... Window operation part, 5 ... Display part, 6 ... Filter generator, 7 ... Filtering means.

Claims (9)

[Claims] 1. A window operation unit and a display unit having a window processing function, and an image displayed on the display unit or an image formed on the image plane of a screen, a film or the like by the display unit. In an image processing device for processing an image,
By inputting the window width information set in the window operation unit, the cutoff frequency of the high frequency cut filter is increased when the window width is large, and the cutoff frequency of the high frequency cut filter is decreased when the window width is small. An image comprising a filter generator for generating a characteristic filter and filtering means for filtering an image displayed on the display unit or an image formed on the image forming surface based on the filter. Processing equipment. 2. The image processing apparatus according to claim 1, wherein the filter generator generates a filter whose cutoff frequency is continuously variable according to the window width. 3. A filter generator stores high-frequency cut filters having different cutoff frequencies corresponding to a plurality of types of window width regions, and selects any one of the filters according to the window width to generate. Claims No. 1
The image processing device according to the item. 4. The filter generator is a high cut filter f ₁
First storage unit for storing a high-frequency enhancement filter
a _second storage unit for storing f ₂ , a coefficient generation unit for generating a coefficient α according to the window width, a _first multiplier for multiplying α · f ₁ , and (1-α) · The image processing apparatus according to claim 1, comprising a _second multiplier that executes multiplication of f _{2 and} an adder that executes addition of α · f ₁ + (1−α) · f _2. . 5. The image processing apparatus according to claim 4, wherein the coefficient generating unit generates a coefficient α that is continuously variable according to the window width. 6. The coefficient generating unit stores coefficient α having different values corresponding to a plurality of types of window width areas, and selects and generates any one coefficient α according to the window width. The image processing apparatus according to item 4. 7. The image processing apparatus according to claim 1, wherein the filtering means digitally performs a filtering operation on the image input to the display section. 8. The image processing apparatus according to claim 1, wherein the filtering means controls the converging diameter of the beam converged on the phosphor screen of the display unit to execute the filtering. 9. The filtering means executes filtering by varying a distance between a lens for forming an optical image on the display unit on the image forming surface and the image forming surface. The image processing apparatus according to item 1.