JPH0473191B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of the Invention) The present invention relates to a method for calculating a non-sharp mask signal used when performing non-sharp mask processing on an image signal, and an apparatus for implementing the method.

(Technical Background of the Invention and Prior Art) Certain types of phosphors are exposed to radiation (X-rays, α-rays, β-rays,
A part of the energy of this radiation is accumulated in the phosphor, and when the phosphor is then irradiated with excitation light such as visible light, the accumulated energy is Accordingly, the phosphor exhibits stimulated luminescence. A phosphor exhibiting such properties is called a stimulable phosphor (stimulable phosphor).

Using this stimulable phosphor, radiation image information of a subject such as a human body is temporarily recorded on a stimulable phosphor sheet (hereinafter referred to as a stimulable phosphor sheet),
The sheet is scanned with excitation light to cause stimulated luminescence, and the stimulated luminescence is read photoelectrically to obtain an image signal,
A radiographic image information recording and reproducing system that processes this image signal to obtain a radiographic image of a subject with good diagnostic suitability has been proposed (for example, Japanese Patent Application Laid-open No. 12429/1983,
No. 55-116340, No. 55-163472, No. 56-11395
No. 56-104645, etc.).

According to this system, radiation image information stored and recorded on a stimulable phosphor sheet is photoelectrically read, the electrical signal is subjected to appropriate signal processing, and this electrical signal is used to produce recording materials such as photographic materials.
By outputting a radiographic image as a visible image on a display device such as a CRT, a very large effect can be obtained in that a radiographic image with excellent observation and interpretation suitability (diagnosis suitability) can be obtained.

One such signal processing method is, for example,
There is a non-sharp mask processing method as described in Japanese Patent No. 55-163472.

Briefly, this method photoelectrically reads radiation image information stored and recorded on the stimulable phosphor sheet by scanning with excitation light to obtain an original image signal (electrical signal) Sorg at each scanning point. Obtain the unsharp mask signal Sus corresponding to the ultra-low spatial frequency at each scanning point of the stimulable phosphor sheet, set the emphasis coefficient to β, and calculate Sorg at each scanning point above.
The signal is converted by the calculation S′=Sorg+β(Sorg−Sus) to emphasize the frequency component (frequency band) above the ultra-low spatial frequency, and this signal S′ is used for the visible image formation. It is characterized by being used for this purpose.

This non-sharp mask processing method increases the contrast of diagnostic objects (pulmonary blood vessels, epiphyses, tumors, soft tissue, etc.), increases the efficiency of interpretation and diagnosis, and is extremely useful in reducing the rate of missed lesions. This is a processing method.

However, the optimum values of the emphasis frequency band and the emphasis coefficient in this unsharp mask processing vary within the same image depending on the diagnostic purpose of the doctor. For example, if we take a fracture in a limb as an example, it is desirable to emphasize relatively high frequency bands to clearly depict the outline of the bone in order to properly diagnose the fracture line, etc. In order to diagnose soft tissue inflammation, it is desirable to emphasize relatively low frequency bands.

In other words, when interpreting and diagnosing images,
There are cases where it is desired to perform a plurality of unsharp mask processes on one piece of image information under different conditions (different emphasis frequency bands and different emphasis coefficients), and perform a diagnosis based on these multiple processed images.

As one of the unsharp mask processing image display methods suitable for such cases, the above-mentioned unsharp mask signal Sus
is sequentially changed, unsharp mask processing is performed based on each changed Sus, and each unsharp mask processed image is sequentially and continuously processed.
Display on a display device such as CRT. In other words, a method has been considered in which an image is displayed on a display device as a moving image that changes from moment to moment, and the image changes can be stopped arbitrarily.

According to this method, a large number of unsharp mask-processed images with different emphasis frequency bands are continuously processed from moment to moment.
Since the images are displayed as moving images on the display device, doctors can obtain and observe all of the information and optimal images necessary for various diagnostic purposes in a short time, dramatically improving diagnostic performance and improving diagnosis. Work can be streamlined (time saved).

However, in order to implement the above method,
It is necessary to quickly perform unsharp mask processing calculations based on each changed Sus, and to do so, it is necessary to perform unsharp mask processing calculations based on each of the different unsharp mask signals Sus.
It is necessary to find them sequentially and quickly.

(Object of the Invention) In view of the above-mentioned circumstances, the object of the present invention is to use the present invention particularly preferably when sequentially changing the unsharp mask signal Sus as described above and continuously performing unsharp mask processing based on each Sus. An object of the present invention is to provide an unsharp mask signal calculation method that can quickly calculate different unsharp mask signals Sus, and an apparatus for implementing the method.

(Structure of the Invention) In order to achieve the above object, the method according to the present invention is a method of sequentially calculating different unsharp mask signals Sus, and includes microphone size N×M (N and M are the number of scanning points, When a predetermined scanning point is located at the center of a rectangular basic unsharp mask (an integer other than 1), the original image signal of all scanning points included in the mask
Calculate the basic unsharp mask signal Sus at a given scanning point by simply averaging Sorg, calculate this basic unsharp mask signal Sus for each scanning point, and store the result (Sus). , Subsequent calculations of different Sus at the above-mentioned predetermined scanning points are performed using _α 1 × α ₂ pieces included in the unsharp mask of mask size α ₁ N × α ₂ M (α ₁ and α ₂ are integers excluding 1). Calculated by simply averaging the stored basic unsharp mask signal Sus for the basic _unsharp mask _of It is characterized by doing.

Further, in order to achieve the above object, the apparatus according to the present invention provides a method for detecting a predetermined image at the center of a rectangular basic unsharp mask of mask size N×M (N and M are the number of scanning points, and are arbitrary integers except 1). A first step of calculating the basic unsharp mask signal Sus at the predetermined scanning point by simply averaging the original image signals Sorg of all the scanning points included in this mask when the scanning point is located.
an arithmetic unit, a memory for storing the unsharp mask signal Sus for each scanning point calculated by the above arithmetic unit, and a mask size α ₁ N×α ₂ M (α ₁ and α ₂ are integers excluding 1). The basic unsharp mask signals Sus for the α ₁ × α ₂ basic unsharp masks in the unsharp mask are read from the memory, and they are simply averaged to obtain different unsharp mask signals at the predetermined scanning points.
a second arithmetic unit that calculates Sus; and a control device that outputs a command to the second arithmetic unit to find the unsharp mask signal Sus at each α ₁ and α ₂ while sequentially changing α ₁ and α ₂ . It is characterized by comprising the following.

(Effects of the Invention) As described above, the unsharp mask signal Sus calculation method and device according to the present invention have a mask size of N×M.
The basic unsharp mask signal Sus for the basic unsharp mask (N, M are arbitrary integers except 1) is obtained for each scanning point, and each basic unsharp mask signal Sus is memorized. α ₁ N×α ₂ M (α ₁ and α ₂ are integers excluding 1, such as 2, 3, 4, 5, 6, 7, 8,
9,...) for the non-sharp mask, and in obtaining them, α ₁ × included in the above α ₁ N × α ₂ M size mask.
α It is obtained by simply averaging _{the two} basic unsharp mask signals Sus.

Therefore, when calculating using only conventional Sorg, the signal for a mask of size α ₁ N × α ₂ M
In order to calculate Sus, it is necessary to add α ₁ N × α ₂ M Sorgs and divide by α ₁ N × α ₂ M, and in the end, it is necessary to perform α ₁ α ₂ MN + 1 operation, and However, according to the present invention, it is sufficient to add α ₁ × α ₂ basic unsharp mask signals Sus and divide by α ₁ N × α ₂ M,
In the end, only α ₁ α ₂ +1 calculations are required, so the calculation time is shortened, and as a result, different Sus can be calculated sequentially and quickly.

(Embodiments) Hereinafter, embodiments of the present invention will be described with reference to the drawings.

The unsharp mask signal calculation method and device according to the present invention scans a recording medium to obtain an original image signal Sorg for each scanning point, and calculates S'=Sorg+β(Sorg−Sus) for each original image signal Sorg. ) However, Sus: unsharp mask signal corresponding to ultra-low spatial frequency at each scanning point β: performs unsharp mask processing by performing an emphasis coefficient calculation, and sequentially performs unsharp mask processing based on different signals Sus Since this is used when the process is performed several times in succession, this point will be explained first.

The recording medium may be, for example, the stimulable phosphor sheet or an X-ray photographic film.

In the case of a stimulable phosphor sheet, Sorg scans the stimulable phosphor sheet onto which the radiation incident on the subject is incident with excitation light such as a laser beam, and by this scanning, the stimulable phosphor sheet emitted from the sheet is It is determined by photoelectrically reading the emitted light using a photomultiplier or other photoelectric reading means.

In the case of X-ray photographic film, Sorg is, for example, by optically scanning the X-ray photographic film on which an image has been developed and photoelectrically reading the light transmitted through or reflected by the film. It is required.

The unsharp mask signal Sus is obtained by preparing a rectangular unsharp mask of an arbitrary size, and preparing the mask size (the mask size is expressed by the length of two adjacent sides of the mask, and the length of each side is the number of scanning points. A method is used in which the Sorg of a predetermined scanning point and the Sorg of scanning points surrounding the scanning point are read out from the memory in accordance with the following: In other words, when determining the Sus of a predetermined scanning point, prepare a rectangular non-sharp mask of a predetermined size, and when the predetermined scanning point is located in the center of the mask, the Sus of all scanning points located within the mask is calculated. It is obtained by simply averaging Sorg.

The unsharp mask processing described above performs the calculation S' = Sorg + β (Sorg - Sus) for each scanning point in one image, and such unsharp mask processing is performed successively based on different signals Sus. In order to perform this multiple times, it is necessary to calculate a different signal Sus for each scanning point extremely quickly. signal
If the Sus calculation speed is slow, the unsharp mask processing speed will also be slow, and as a result, for example, when trying to display a large number of processed images in succession as if they were moving images, it becomes difficult to display such moving images. Because it will be.

Different Sus means that the unsharp mask sizes used when calculating Sus are different. Therefore, sequentially calculating different Sus at predetermined scanning points means sequentially changing the mask size at an appropriate step size and within an appropriate range, and calculating Sus for each non-sharp mask of each size. .

Next, we will discuss a method and apparatus according to the present invention used when performing image processing as described above, that is, a method and apparatus for obtaining different Sus based on non-sharp masks of various sizes that are changed in appropriate steps. An embodiment will be described.

FIG. 1 is a block diagram showing an example of an image display device including an embodiment of the device according to the present invention.

The illustrated device performs unsharp mask processing multiple times while sequentially changing not only the unsharp mask signal Sus but also the emphasis signal β as necessary, and continuously switches and displays these unsharp mask-processed images. This is an image display device configured to be able to scan a recording medium, and includes an original image signal memory 10 that stores an original image signal Sorg for each scanning point obtained by scanning a recording medium;
A non-sharp mask processing device 20 that performs non-sharp mask processing on Sorg, and each Sus that is changed by sequentially changing at least one of the Sus and β.
or a control device 30 that operates the unsharp mask processing device 20 to perform the unsharp mask processing under each value of β; a CRT device 40 that continuously displays a plurality of processed images based on the image signal S' obtained by the above processing as a moving image;
It consists of:

It is necessary to input various control information to the control device 30, such as the mode (whether to change Sus or β), change range and step size of the parameter (Sus or β), start or stop of motion, etc. There is. These control information may be input directly to the control device 30 via a human or a host computer, but in this device, the control information input device 5
0, and the input device 50 is configured to receive the various control information described above from a human or a host computer, convert the control information into an appropriate signal, and output the signal to the control device 30.

The CRT device 40 includes a first frame memory 401 and a second frame memory 402 that store a processed image signal S' subjected to unsharp mask processing in the unsharp mask processing device 20 for one image; , 402;
It is equipped with a CRT 46 that outputs and displays a visible image based on the signal output from the A converter.

The non-sharp mask processing device 20 has a mask size N×M (N, M are integers excluding 1,
A first arithmetic unit 201 that calculates a basic unsharp mask signal Sus for a basic unsharp mask (preferably N is an odd number and N=3), a memory 202 that stores each basic unsharp mask signal Sus, and a control device. 3
A second computing unit 203 that calculates non-sharp mask signals Sus of different mask sizes by suitably adding and averaging the respective non-sharp mask signals Sus according to a command from 0;
A subtracter 204 performs the operation Sorg−Sus between Sus obtained in this way and the original image signal Sorg, and β is added to the output from the subtracter 204.
A multiplier 205 performs an operation β(Sorg−Sus) by multiplying
and an adder 206 that performs the calculation Sorg+β(Sorg−Sus).

The first arithmetic unit 201, the memory 202, the second arithmetic unit 203, and the control device 30 constitute an embodiment of the Sus calculation device according to the present invention.

In this embodiment, a square mask with a mask size of 3×3 is set as the basic unsharp mask, and this mask allows the basic unsharp mask signal for each scanning point to be
Sus is calculated by the first arithmetic unit 201.

Figure 2 shows a stimulable phosphor sheet 6 which is a recording medium.
0, and each square in the diagram represents one scanning point. For example, the basic unsharp mask signal Sus at a predetermined scanning point A in FIG.
is the basic unsharp mask Ma with mask size 3×3
It is obtained by simply averaging the Sorg of a total of nine scanning points included in the mask Ma when the scanning point A is located at the center of the mask Ma. The Sorg of each scanning point is read from the original image signal memory 10 and used. In this way, the basic unsharp mask signal Sus at each scanning point is calculated by the first arithmetic unit 201, each calculated Sus is stored in the memory 202, and the subtracter 204 uses the Sus to calculate Sorg-Sus. (The second arithmetic unit 203 is used to calculate a different Sus after this), and the basic unsharp mask signal Sus is sent to the CRT device 40 via a multiplier 205 and an adder 206. An image that has undergone unsharp mask processing based on this is output.

Next, the case of calculating different Sus at the predetermined scanning point A will be explained. Calculating different Sus is based on different sized unsharp masks.
The purpose of this invention is to calculate α ₁ N × α ₂ M (α ₁ ,
α ₂ is a variable consisting of an integer other than 1) The unsharp mask signal Sus for the unsharp mask of size is sequentially calculated.

In this embodiment, all masks are assumed to be square, and the mask size is set to the above basic 3×3 (α ₁ =α ₂ ).
Based on 9×9, 15×15, 21×21, 27×
27,..., 111×111 and 19 different mask sizes are changed in odd multiples of 3, and the Sus for each mask size is calculated. The mask size 3×3 is stored in the memory 202 as described above, and the mask size 9
×9 Sus is 9 mask size 3×3 Sus,
For example, in Fig. 8, when finding the Sus of the mask size 9 x 9 for the scanning point A, a total of 9 Sus for each mask M _A to M _I of the mask size 3 x 3 is calculated.
It is obtained by simply averaging Sus in the second arithmetic unit 203. Mask size 15×15, 21×
Similarly, when calculating each Sus of 21,..., memory 20
Sus of mask size 3×3 stored in 2
25, 49, etc. are obtained by simple addition and averaging.

The non-sharp mask processed image based on each Sus obtained in this way is stored in the first frame memory 401.
and the second frame memory 402, and the D/A converter 4 is alternately stored from both frame memories.
4 and displayed as a visible image on the CRT 46. For example, while a processed image based on Sus with a mask size of 3×3 is stored in the first frame memory 401 and displayed on the CRT 46,
A processed image signal based on the next Sus with a mask size of 9×9 is written to the second frame memory 402,
When that process is finished, this second frame memory 4
02 is connected to the D/A converter 44, the images displayed on the CRT 46 are switched, and this switching is performed one after another in a short period of time, and the images are displayed as if they were moving images.

Sus calculations for different mask sizes in the second arithmetic unit 203 are performed based on instructions from the control device 30. A specific example of the role of the control device 30 will be explained below.

First, in the case of a Sus change, information indicating that it is a Sus change is received from the control information input device 50, and mode control is performed. In this case, since it is a Sus change, β is fixed. Next, based on the conditions given to the control information input device 50 (for example, in what step size and within what range Sus should be changed), the required number of processing repetitions and the mask size for each process (for example, 3 ×3, 9×9, 21×21,…57×57)
Calculate etc. Next, upon receiving a motion start signal from the control information input device 50, the entire apparatus is started through the synchronization line. When the calculation and image display with the first mask size are completed, the next mask size is applied to the necessary parts of the entire device and the next calculation and image display are started. When this process is repeated a predetermined number of times or when a motion stop signal is received from the control information input device 50, the process is stopped and stopped at the CRT display state at that time.

In the case of β change, there is no need for complicated calculations as in the case of Sus change, and the control device 30 simply uses information such as the change range received from the control information input device 50
It is sufficient to simply change the multiplier β in the multiplier 205.

In the above embodiment, N=3, but as long as N is an integer other than 1, it may be a numerical value other than 3, such as 2 or 5. When N and M are 1, α ₁ , α ₂
In the case where a different Sus is obtained by changing Sorg, the result is equivalent to calculation using the original image signal Sorg, and the speed-up of calculation, which is an effect of the present invention, cannot be achieved, so this method is excluded.

Further, in the above embodiment, α ₁ =α ₂ and α ₁ and α ₂ were changed in order as 3, 5, 7, 9, etc.
These α ₁ and α ₂ may also be integers other than 1, and they may be changed in any manner. for example,
It is also possible to sequentially change every other odd number such as 3, 7, 11, 15, etc. When α ₁ and α ₂ are 1, the size is the same as that of the basic unsharp mask, and it is excluded. Of course, the maximum and minimum values of α ₁ and α ₂ can be set arbitrarily.

In the present invention, Sus is calculated as described above. For example, when calculating a mask of size α ₁ N × α ₂ M, α ₁ × α ₂ basic unsharp mask signals Sus are added together. Just divide by α ₁ N × α ₂ M,
A total of α ₁ α ₂ +1 calculations is required, and different Sus can be calculated sequentially and quickly.

In addition, when calculating Sus in the present invention, the step width of change in Sus (width of change in mask size) is limited to some extent.For example, in the above embodiment, calculation can only be performed for a mask that is an odd multiple of 3×3. For example, the above-mentioned Sus is sequentially changed to perform unsharp mask processing, and each processed image is continuously switched and displayed on a CRT like a moving image.
When making a medical diagnosis using the moving images, a step size of this level causes no practical problems and can be used for practical purposes.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an example of an image display device including an embodiment of the unsharp mask signal calculation device according to the present invention, and FIG. 2 is a diagram showing a recording medium and scanning points thereon. 10...Original image signal memory, 30...
Control device, 40... CRT device, 60... Recording medium, 201... First computing unit, 202... Memory,
203...Second arithmetic unit.

Claims (1)

[Claims] 1. Scan the recording medium to obtain an original image signal Sorg for each scanning point, and for each original image signal Sorg, S′=Sorg+β(Sorg−Sus) where Sus: each scanning Unsharp mask signal β corresponding to the ultra-low spatial frequency at a point: An emphasis coefficient is calculated to perform unsharp mask processing, and the unsharp mask processing is sequentially performed based on different unsharp mask signals Sus. A method for calculating the above-mentioned different unsharp mask signals Sus used when performing multiple scans, which is a basic unsharp mask signal of a rectangular shape with a mask size of N×M (N and M are the number of scanning points and are arbitrary integers except 1). When a mask is set and a predetermined scanning point is located in the center of this basic unsharp mask, the original image signal Sorg of all the scanning points included in this mask is simply added and averaged to obtain the basic image signal at the above-mentioned predetermined scanning point. Unsharp mask signal
Sus is calculated, the basic unsharp mask signal Sus is calculated for each scanning point, each calculated basic unsharp mask signal Sus is memorized, and a different unsharp mask signal is generated at the predetermined scanning point thereafter. The calculation of Sus is based on the mask size α ₁ N × α ₂ M
α ₁ in the unsharp mask of (α ₁ , α ₂ are integers excluding 1)
×α This is performed by simply averaging the stored basic unsharp mask signals Sus for the _two basic unsharp masks, and the calculation of the unsharp mask signal Sus by this method is performed using different α ₁ , α A non-sharp mask signal calculation method characterized by performing the calculation multiple times under _binary values. 2 Scan the recording medium to obtain the original image signal Sorg for each scanning point, and for each original image signal Sorg, S' = Sorg + β (Sorg - Sus) where Sus: ultra-low space at each scanning point Unsharp mask signal β corresponding to frequency: Used when performing an unsharp mask process by performing a calculation called an emphasis coefficient, and when performing the unsharp mask process multiple times consecutively based on different unsharp mask signals Sus. The above-mentioned device for calculating the different unsharp mask signals Sus scans the center of a rectangular basic unsharp mask having a mask size N×M (N and M are the number of scanning points and are arbitrary integers except 1). a first arithmetic unit that calculates a basic unsharp mask signal Sus at the predetermined scanning point by simply adding and averaging the original image signals Sorg of all scanning points included in this mask when the point is located; A memory for storing an unsharp mask signal Sus for each scanning point calculated by the first arithmetic unit, and an unsharp mask signal Sus of mask size α ₁ N × α ₂ M (α ₁ and α ₂ are integers except 1). The basic unsharp mask signals Sus for the α ₁ × α _two basic unsharp masks in the mask are read from the memory, and they are simply averaged to obtain different unsharp mask signals at the predetermined scanning points.
a second arithmetic unit that calculates Sus, and a control that outputs a command to the second arithmetic unit to obtain the unsharp mask signal Sus at each α ₁ and α ₂ while sequentially changing α ₁ and α ₂ respectively. 1. A non-sharp mask signal calculation device comprising: