JPH0786926B2 - X-ray image processing method and apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Use) The present invention records an X-ray image when copying an image obtained by irradiating an object with X-rays (hereinafter referred to as “X-ray image”). After reading the density of the X-ray image recorded in the original photo (hereinafter referred to as "original image density") from the original plate (hereinafter referred to as "original photo"), the X-ray image applied to the signal representing this original image density The present invention relates to a processing method and an apparatus for carrying out this method.

(Prior Art) A photographic film for recording an X-ray image must have sufficient sensitivity and a wide exposure range for photographing, and must also have high contrast, sharpness and fine graininess necessary for observation and interpretation. However, these conditions often conflict with each other, and it is difficult to obtain an X-ray image satisfying all of these conditions by photographing X-rays directly on a photographic film. The reality is that the film is designed at the expense of little by little.

Therefore, an X-ray image is recorded by using a photographic film having a low gamma value designed to meet the later image processing.
After reading the X-ray image from the original photo on which the line image was recorded and converting it into an electrical signal, and subjecting this to image processing,
Various methods for improving contrast, sharpness, and graininess by reproducing a visible image on a copy photograph or the like have been studied.

(Problems to be Solved by the Invention) As an image processing method to be applied to the electric signal, for example, an original X-ray photograph proposed by the applicant in Japanese Patent Publication No. 62-62374 is scanned and recorded. After reproducing the X-ray image information from the X-ray image information and converting it to an electric signal, when reproducing it on a copy photo, etc., the density D _us of the non-sharp mask corresponding to the ultra-low spatial frequency at each scanning point was calculated, and the density of the original photo was calculated. _Is D _org , the emphasis coefficient is β, and the density reproduced on a copy photograph or the like is D ′, there is a method of performing the following operation: D ′ = D _org + β (D _org −D _us ). By using this image processing method, it is possible to obtain a reproduced image with good image quality performance such as contrast and sharpness by emphasizing the spatial frequency region important for observation.

However, due to X-ray quantum noise at the time of shooting,
The spatial frequency of particle noise that makes a line image rough and grainy is also overlapped with the spatial frequency that affects other image quality performance such as contrast and sharpness. The particle noise is also emphasized at the same time as the image quality is improved. Especially, when the X-ray dose applied to the subject at the time of shooting is reduced, the influence of X-ray quantum noise on the X-ray image becomes large, resulting in a rough surface. It was a reproduced image of the impression.

Further, as one of the methods of positively attenuating the particle noise, the method proposed by the applicant in Japanese Patent Application No. 62-265018 can be used. This method includes that β is set to β <0 in the above formula (1), and the spatial frequency component of the particle noise is positively suppressed to make the particle noise inconspicuous, and contrast, sharpness and other Although it is possible to minimize the deterioration of image quality performance, as mentioned above, the spatial frequency of particle noise overlaps with other spatial frequencies that affect image quality performance such as contrast and sharpness, and it is necessary to reduce particle noise. Therefore, the contrast, sharpness, etc. are also deteriorated to some extent, so that it can be used only when the particle noise is too conspicuous.

In view of the above circumstances, the present invention suppresses particle noise in an X-ray image and improves sharpness, contrast, and the like to improve the image quality performance of the entire image, and further allows a sufficient calculation time without complicating the apparatus. It is an object of the present invention to provide an X-ray image processing method within the range, and an apparatus capable of implementing this method.

(Means for Solving the Problems) The X-ray image processing method of the present invention scans an original photograph on which X-ray image information is recorded, which is obtained by irradiating a photographic film with X-rays, and scans the original of each scanning point. After reading the image density, when reproducing an X-ray image as a visible image on a copy photograph, etc., the density of the blurred mask obtained by averaging the original image density within a predetermined surrounding area corresponding to each scanning point. To
D _us , the original image density is D _org , the coefficient corresponding to the blur mask is β, and the image density after the calculation process is D ′, the coefficient β is the value of the X-ray irradiated to each point of the photographic film. Β <0 (β> -1) to β> 0 as the dose increases
Is calculated by using the coefficient β ₁ and the calculation is performed according to the equation of D ′ = D _org + β (D _org −D _us ) ... (2), and the dose is low in one X-ray image. In the region, the spatial frequency component higher than the spatial frequency component of the blur mask density D _us is attenuated, and in the region where the dose is high, the spatial frequency component higher than the spatial frequency component of the blur mask density D _us is emphasized. It is what

The X-ray image processing method of the present invention may further include other calculation in addition to the calculation corresponding to the equation (2). That is, according to another X-ray image processing method of the present invention, an original photograph recorded with X-ray image information obtained by irradiating a photographic film with X-rays is scanned and the original image density at each scanning point is read. After that, when reproducing an X-ray image as a visible image on a copy photograph or the like, the original image density within a predetermined peripheral range corresponding to each scanning point or an image density obtained by performing an intermediate process on a signal representing the original image density. 1 obtained by averaging
Or the density of a plurality of blur masks obtained by changing the predetermined range is D _us.k (k = 1, 2, ..., N; n is an integer indicating the number of blur masks), the original image density or this original image density The image densities after the intermediate processing of the signal representing the are expressed as D _b1 , D _b2 , and the one or more coefficients respectively corresponding to the one or more blur masks are β _k (k = 1,
2, ..., N), where D ′ is the image density after the arithmetic processing, at least one coefficient β _l (l is 1 to n among the coefficients β _k (k = 1, 2, ..., N) Is an integer within the range of β _l <0 as the dose of X-rays applied to each point of the photographic film increases.
It is a function that shifts from (β ₁ > −1) to β ₁ > 0, and by using this coefficient β _1. In one X-ray image, the spatial frequency component higher than the spatial frequency component having the density D _{us.l of} the blur mask corresponding to the coefficient β ₁ is attenuated in one X-ray image, and ,
In the region where the dose is high, the spatial frequency component higher than the spatial frequency component having the density D _{us.l of} the blur mask corresponding to the coefficient β ₁ is emphasized.

Further, one of the X-ray image processing apparatuses of the present invention for carrying out the above X-ray image processing method is to scan an original photograph having X-ray image information recorded, which is obtained by irradiating a photographic film with X-rays. Then, after reading the original image density at each scanning point, the signal representing the original image density is processed by the arithmetic unit, and based on the signal representing the processed image density,
In an X-ray image processing apparatus that reproduces an X-ray image as a visible image on a copy photograph or the like, the calculation unit obtains it by averaging original image densities within a predetermined range around each scanning point. The density of the blurred mask is D _us , the original image density is D _org ,
When the coefficient corresponding to the blur mask is β and the image density after the arithmetic processing is D ′, the coefficient β is the X radiated to each point of the photographic film.
It is a function that shifts from β <0 (β> -1) to β> 0 as the dose of the radiation increases, and using this coefficient β _l , D ′ = D _org + β (D _org −D _us ) (4 ) Is performed according to the equation (1).

The X of the present invention for carrying out the X-ray image processing method of the present invention
The line image processing apparatus may have a function of performing another calculation in addition to the calculation corresponding to the above equation (4).

That is, the other X-ray image processing apparatus of the present invention scans an original photograph in which X-ray image information is recorded, which is obtained by irradiating a photographic film with X-rays, and reads the original image density at each scanning point. Then, in the X-ray image processing apparatus, the signal representing the original image density is processed by the arithmetic unit, and the X-ray image is reproduced as a visible image on a copy photograph or the like based on the signal representing the processed image density. The calculation unit averages one of the original image densities in the surrounding predetermined range or an image density obtained by subjecting the signal representing the original image density to intermediate processing, corresponding to each scanning point, or one of the predetermined ranges. the concentration of the plurality of unsharp mask obtained by changing the _{D us.k (k = 1,2, ...} , n; n is an integer indicating the number of unsharp mask), the original image density or signal representing the original image density The image density after subjected to during processing D _b1, D _b2, one or a plurality of coefficients respectively corresponding to the one or more unsharp mask _{β k (k = 1,2, ...} , n), The image density after calculation processing is D ′
, At least one coefficient β _l (l is an integer within 1 to n) of the coefficient β _k (k = 1, 2, ..., N) is applied to each point of the photographic film X Β _l <0 as the radiation dose increases
It is a function that shifts from (β ₁ > −1) to β ₁ > 0, and by using this coefficient β _1. It is characterized in that the calculation is performed in accordance with the equation.

Here, in each of the above inventions, “shift from β <0 (β> -1) to β> 0” (“β _l <0 (β _l > -1) to β _l
“> 0” ”includes, for example, those having a region of β = 0 (β ₁ = 0) on the way as shown in FIG. 1C described later.

Further, in performing the calculation according to the above equations (2) to (5), as a signal representing the original image density D _org , a signal (D _org = k · E) proportional to the light quantity E of the transmitted light or the reflected light of the original photograph. , K; constant), and the corresponding signals such as D _us , β, D _us.k , β _k , D _b1 , D _b2 may be used to perform the calculation. Original image density proportional to the logarithmic value of the light amount E
A signal representing D _org is used (D _org = k ′ · logE, k ′; constant), and each corresponding signal D _us , β, D _us.k ,
The calculation may be performed using β _k , D _b1 , and D _b2 .

(Operation) Usually, in one X-ray image, an area having a large dose of the irradiated X-rays and an area having a small dose of the irradiated X-rays are mixed due to the distribution of various tissues forming the object, the difference in the thickness of the object, and the like. ing.
When a signal representing the image density obtained by reading this X-ray image is subjected to a calculation in a uniform manner, for example, by using the method shown in Japanese Patent Publication No. 62-62374, the contrast and the sharpness are emphasized. The image quality performance is improved because the large amount of the particle noise is originally small, but the particle noise is emphasized in the region with a large amount of particle noise and the small X-ray dose, which gives a rough and rough impression as described above. On the other hand, in order to suppress the particle noise in the area with a small X-ray dose, the particle noise is positively reduced by using the method proposed in Japanese Patent Application No. 62-265018 described above. It will result in lowering. Therefore, whichever method is used, the image processing is performed while observing the balance of the total image quality performance of the entire image, and the image processing is inevitable.

In the X-ray image processing method of the present invention, attention is paid to the fact that a region with a large X-ray dose and a region with a small X-ray dose coexist in one X-ray image as described above, and image processing is performed according to the above equation (2). , The coefficient β becomes β as the X-ray dose increases.
It is adapted to shift from <0 (β> -1) to β> 0.

By doing so, the particle noise is positively reduced in an area in the X-ray image where the X-ray dose is small and the particle noise is large, and the X-ray image in the area where the X-ray image is large and the original particle noise is small. In this case, the image quality performance such as sharpness and contrast is positively improved, and therefore, the image quality performance of the reproduced image can be improved much more than in the case where uniform image processing is performed on the whole.

The X-ray dose of each area of the X-ray image can be known by scanning a signal representing the original image density read by scanning the original image photograph.

Further, in order to perform the optimum image processing for each area as described above, for example, the above Japanese Patent Publication No. 62-62374 and Japanese Patent Application No. 62-25601.
By combining the methods shown in No. 8, the original image density is D _org , the densities of the two blur masks that have been appropriately frequency-processed are D _us ′, D _us ″, the function of the original image density (function of X-ray dose). ), Two coefficients β ′ and β ″ (β ′, β ″> 0) each having an appropriate function form, and D ′ being the image density after processing, D ′ = D _org −β ′ (D _org −D _us ′) + β ″ (D _org −D _us ″)… (6) There is also a method of performing the operation according to the equation. However, if this method is used, at least the second term β ′ (D _org −D _us ′) And the third term β ″ (D _org −D _us ″) must be calculated for each scanning point of the X-ray image.

On the other hand, when the most basic expression (2) of the present invention is used, only one term of β (D _org −D _us ) needs to be calculated, and the calculation time is about half that of the above expression (6). And when the device is configured to perform this operation in hardware, the structure of the device becomes much simpler.

The X-ray image processing method of the present invention may include, in addition to the calculation of the above formula (2), other calculation as shown in the above formula (3). When the above equation (3) is transformed, Becomes

If both the first term D _b1 and the second term D _b2 of the equation (7) are set as the original image density D _org , the first term and the second term are obtained.
The combination of terms D _b1 + β ₁ (D _b2 −D _us.l ) is the same as the above equation (2). That is, one X-ray image may be subjected to various image processes other than the present invention, such as various noise reduction processes other than the present invention, and window processes for extracting only necessary spatial frequency components. When using the X-ray image processing method of No. 3, it is not always necessary to use the original image density D _org obtained by reading the X-ray image as it is, and the signal representing the original image density D _org is subjected to the above-described intermediate processing, for example. The image density after the above may be used, and in this case as well, it is possible to obtain an effect substantially equivalent to that of the expression (2), and it is possible to effectively combine it with other arithmetic processing. Further, the image density after the intermediate processing may be the image density generated in the process of carrying out the present invention.

Next, the third and fourth terms of the above equation (7) will be described.
The spatial frequency component of particle noise and the spatial frequency component that carries image quality performance such as sharpness and contrast are spread over a wide range. Therefore, when it is necessary to perform more detailed image processing than in the case of performing image processing only with the combination of the first term and the second term of the equation (7), the spatial frequency band is changed from that of the second term. , The same operation as the second term can be performed by the third or fourth term.

In addition, in the third term and the fourth term, the coefficient β _m (m ≠ l) is set to β _m
> 0, in order to complement the operations performed in the first term and the second term, for example, an operation proposed by the applicant in Japanese Patent Publication No. 62-62374, which emphasizes a specific spatial frequency component, is applied to the entire image. Alternatively, β _n (n ≠ l) may be changed to β _n
When <0, the calculation for reducing the particle noise proposed by the present applicant in Japanese Patent Application No. 62-265018 may be performed on the entire image.

As described above, the calculation time of the first term and the second term of the formula (7) (corresponding to the calculation of the formula (2)) is significantly shorter than the calculation of the formula (6). Therefore, it is possible to perform more detailed image processing by performing the above-described calculation of the third and fourth terms by utilizing this margin time.

In the X-ray image processing apparatus of the present invention, since the arithmetic unit has a function of performing the arithmetic processing, as described above, when this function is realized by software, the execution time of the software can be shortened. It can be achieved, and when realized by hardware, the structure of the device can be simplified.

(Examples) Examples of the present invention will be described below with reference to the accompanying drawings.

FIG. 3 is a perspective view showing an example of a radiation image processing apparatus using the radiation image processing method of the present invention.

The original photograph 1 on which the X-ray image information of the subject is recorded is conveyed (sub-scanned) in the arrow Y direction by the photograph conveying means 3 driven by the motor 2. On the other hand, the reading light 5 emitted from the laser light source 4 is reflected and deflected by the rotary polygon mirror 6 which is driven by the motor 13 and rotates at a high speed in the arrow direction.
After passing through a focusing lens 7 such as an fθ lens, the optical path is changed by a mirror 8 to enter the original photograph 1 and main scanning is performed in an arrow X direction substantially perpendicular to the sub-scanning direction (arrow Y direction). This read light 5 is modulated by the density of the X-ray image recorded on the original photograph 1 (original image density) and transmitted, and the read light thus transmitted is condensed by the condenser 10 to serve as a photodetector. It is detected photoelectrically by a photomultiplier (photomultiplier tube) 11. The light collector 10 is made by molding a light guide material such as an acrylic plate, and the linear incident end face 10a is arranged so as to extend along the main scanning line on the original photograph 1. The photomultiplier 1 is formed on the emitting end face 10b formed in an annular shape.
1 light receiving surface is bonded. The readout light 5 that has entered the light collector 10 from the incident end face 10a travels through the inside of the light collector 10 by repeating total reflection, is emitted from the emission end face 10b, and is received by the photomultiplier 11, and the X The light quantity of the reading light 5 carrying the line image information is detected by the photomultiplier 11.

The analog output signal D output from the photomultiplier 11 is amplified by the amplifier 16 and digitized by the A / D converter 17 with a predetermined recording scale factor.

The digitized original image density D _org thus obtained is the X-ray that was irradiated on the photographic film when the photographic film was irradiated with X-rays to record a radiation image.
It is roughly proportional to the line dose. A signal representing the original image density D _org is input to the calculation unit 18, and the calculation unit 18
Then, the density D _us of the blur mask is obtained by averaging the image densities in the surrounding predetermined range corresponding to each scanning point, and β <0 when the original image density D _org is low corresponding to this blur mask. (beta> -1), as the original image density D _org is the beta> 0 when dark, using the coefficient beta prepared in advance as a function of the original image density _{D org, D '= D org} + β (D _org −D _us ) ... The image density D ′ after the arithmetic processing is obtained according to the equation (9).

In this calculation process, β <0 is set in the region where the X-ray irradiation dose in the X-ray image is small and the particle noise is large, so that the spatial frequency component higher than the spatial frequency component of the density D _{us of} the blur mask is attenuated. Become. By appropriately selecting the spatial frequency component to be attenuated and the degree of attenuation, the particle noise of the image is suppressed and the deterioration of other image quality performance such as sharpness is suppressed to the minimum. Further, since β> 0 in a region where the X-ray irradiation dose is large and particle noise is small, spatial frequency components higher than the spatial frequency component of the density D _{us of} the blur mask are emphasized, and image quality performance such as sharpness is positive. To be improved.

Further, the calculation unit 18 averages one of the original image densities in a predetermined surrounding range corresponding to each scanning point or the image densities obtained by subjecting the signals representing the original image densities to intermediate processing, or The densities of a plurality of blur masks obtained by changing the above predetermined range are D _us.k (k = 1,2, ..., n; n
Is an integer indicating the number of blur masks), the original image density or the image density after intermediate processing of the signal representing the original image density is D _b1 , D _b2 , 1 corresponding to the one or more blur masks, respectively. When one or more coefficients are β _k (k = 1,2, ... n) and the image density after the calculation process is D ′, among the coefficients β _k (k = 1,2, ..., n) At least one coefficient β ₁ (where l is an integer within 1 to n) changes from β ₁ <0 (β ₁ > -1) to β as the dose of X-rays applied to each point of the photographic film increases. is a function that turns to _l > 0, and using this coefficient β _l The calculation may be performed in accordance with the equation.

The image density D'after the above-mentioned calculation by the calculation unit 18 is stored in the memory 19, and the X-ray image based on the signal representing the image density D'is reproduced and displayed on the image display device 20 as required. .

Figures 1A, 1B, and 1C show X irradiated on photographic film.
It is the graph which showed the example of the function of coefficient (beta) which made the dose of a line into the variable. As mentioned earlier, here is the original image density
Since D _org is read so as to be proportional to the X-ray dose, the dose on the horizontal axis may be replaced with the original image density D _org . Further, in actuality, from the viewpoint of data amount compression, the logarithm of the light amount E of the reading light transmitted through the original photograph 1 is used as a signal representing the original image density D _org (D _org = k ′ · lo).
g E, k ′; constant), but the original image density D _org increases as the X-ray dose increases (positive correlation).

FIG. 1A shows that the particle noise is suppressed by setting the coefficient β = α (<0) in the area C where the image density where the particle noise is noticeable is small, and the structure is clear in detail in the area E where the particle noise is relatively small and the image density is high. This means that the coefficient β is set to β> 0 so that the image quality performance such as contrast and sharpness is positively improved so that the intermediate region D is increased as the image density is increased.

FIG. 1B is an example in which β is monotonically increased as the image density increases.

FIG. 1C shows an example in which β = 0 in the intermediate area F because the image quality performance is balanced.

It should be noted that FIGS. 1A, 1B, and 1C above are merely examples, and an appropriate functional form is determined according to the type of subject, etc., but in an area with a low image density (area with a small X-ray dose), β <0 (β
> -1), and β> 0 is common in a region where the image density is high (a region where the X-ray dose is large). According to an experiment conducted by the inventor of the present invention, the functional form of β varies depending on the type of subject and the like, but β ≦ −1 hardly occurs and −1 <β
In this range, sufficiently good image processing can be performed.

FIG. 2A is a block diagram showing a configuration example of the calculation unit 18 shown in FIG.

A signal representing the original image density D _org is input to the storage means 21 from the left side of the drawing and temporarily stored. The signal representing the original image density D _org temporarily stored in the storage means 21 is
It is input to the blur mask calculation means 22, the coefficient generation means 23, the addition term calculation means 24, and the addition means 25.

The blur mask calculating means 22 averages the image densities of the surrounding N × N scanning points corresponding to each scanning point based on the input signal representing the original image density D _org to obtain the blur mask density D _us. Desired. The signal representing the obtained density D _{us of} the blur mask is the addition term calculation means 2 described later.
Entered in 4.

The coefficient generating means 23 stores a correspondence table between the signal representing the original image density D _org and the coefficient β, and corresponds to each scanning point corresponding to the input original image density D _org of each scanning point. The coefficient β is generated, and the addition term calculation means 24
Sent to.

In the addition term calculation means 24, the input original image density
Based on the signal representing D _org , the signal representing the density D _us of the blur mask, and the coefficient β, β (D _org −D
_us ) is calculated and output.

The output calculation result β (D _org −D _us ) is added by the adding means 25.
Is input to the adding means 25, and D '= D _org + β (D _org -D _us ) (11) is calculated to obtain the image density D'after the arithmetic processing.

FIG. 2B is a block diagram showing another configuration example of the arithmetic unit 18 shown in FIG.

A signal representing the original image density D _org is input from the left side of the drawing to the storage means 21 'and temporarily stored. Storage means 21 '
The signal representative of the original image density D _org temporarily stored in is the n number of blur mask calculation means 22a ′, 22b ′, ..., 22.
, 23n ', n coefficient generation means 23a', 23b ', ..., 23n', n addition term calculation means 24a ', 24b', ..., 24n ', and addition means 25'.

In the n blur mask calculating means 22a ', 22b', ..., 22n 'such as the first blur mask calculating means 22a', N ₁ × N ₁ peripheral portions, N ₂ × N ₂ corresponding to each scanning point, respectively. , ..., N _n × N _n image densities of the scanning points are averaged to obtain the blur mask densities D _us.1 , D _us.2 , ..., D _us.n. The signals representing the densities D _us.1 , D _us.2 , ..., D _us.n of these blur _masks are n addition term calculation means 24a ', 24b', such as the first addition term calculation means 24a '. …, Input to 24n ′.

N coefficient generating means 23 such as the first coefficient generating means 23a '
a ′, 23b ′, ..., 23n ′ contains the original image density D
_org and n blur mask densities D _us.1 , D _us.2 , ..., D _us.n
Factor beta ₁ respectively corresponding to, beta _2, ..., beta _n has the correspondence table is stored in the coefficient beta ₁ corresponding to correspond to the original image density of each scanning point input to each scanning point, β
₂ , ..., β _n are generated respectively, and first addition term calculation means
N adder calculation means 24a ', 24b', ..., 24n 'such as 24a'
Sent to each of the.

N pieces of addition term calculation means 2 such as the first addition term calculation means 24a '
In 4a ′, 24b ′, ..., 24n ′, the signal representing the input original image density D _org , the density D of the corresponding blur mask,
_us.1 , D _us.2 , ..., D _us.n signal, corresponding coefficient β ₁ ,
β ₁ (D _org −D) for each scanning point based on β ₂ , ..., β _n
us.1 ), β ₂ (D _org −D _us.2 ),…, β _n (D _org −D)
_us.n ) Each is calculated and output.

This output calculation result β ₁ (D _org −D _us.1 ), β
₂ (D _org −D _us.2 ), ..., β _n (D _org −D _us.n ) are input to the adding means 25 ′, and in the adding means 25 ′, Is calculated, and the image density D ′ after the arithmetic processing is obtained.

FIG. 2C is a block diagram showing a further different configuration example of the arithmetic unit 18 shown in FIG.

A signal representing the original image density D _org is input from the left side of the drawing to the storage means 21 ″ and temporarily stored. The storage means 21 ″.
The signal representative of the original image density D _org temporarily stored in is input to the blur mask calculation means 22 ″, the coefficient generation means 23 ″, the addition term calculation means 24 ″, and the addition means 25 ″.

The blur mask calculation means 22 ″ averages the image densities of the surrounding N ₁ × N ₁ scanning points corresponding to each scanning point based on the input signal representing the original image density D _org , and then the density of the blur mask is calculated. D _us.1 is obtained, and the signal representing the obtained density D _us.1 of the _{blur mask} is the addition term calculation means 2
Typed in 4 ″.

The coefficient generating means 23 ″ includes image density and coefficients β ₁ , β ₂ , ..., β
A correspondence table with _n is stored. When a signal representing the original image density D _org is input to the coefficient generating means 23 ″, the correspondence table of the input original image density D _org and the coefficient β ₁ is referred to, and the input original image density of each scanning point. A coefficient β ₁ corresponding to each scanning point is generated corresponding to D _org and is sent to the addition term calculation means 24 ″.

In the addition term calculation means 24 ″, β ₁ (D is obtained for each scanning point based on the input signal representing the original image density D _org , the signal representing the density D _us.1 of the _{blur mask} , and the coefficient β _1.
org- D _us.1 ) is calculated and output.

The output calculation result β ₁ (D _org −D _us.1 ) is input to the adding means 25 ″, and in the adding means 25 ″, the image density D ₁ = after the intermediate processing of the original image density D _org is performed. D _org + β
₁ (D _org −D _us.1 ) is calculated.

The signal representing the image density D ₁ of this calculation result is stored in the storage means 2
It is returned to 1 ″ and stored instead of the signal representing the original image density D _org stored in the storage means 21 ″. The signal representing the image density D ₁ is the blur mask calculation means 22 ″,
It is sent to the coefficient generation means 23 ″ and the addition term calculation means 24 ″. In the blur mask calculating means 22 ″, the image densities of the surrounding N ₂ × N ₂ scan points are averaged based on the image density D ₁ to obtain the blur mask density D _us.2. Further, the coefficient generating means 23 ″ refers to the correspondence table of the image density D ₁ and the coefficient β ₂ to find the coefficient β ₂ corresponding to each scanning point corresponding to the input image density D ₁ of each scanning point. It is generated and sent to the addition term calculation means 24 ″.
At 24 ″, β ₂ (D ₁ −D _us.2 ) is calculated this time. The calculation result is sent to the adding means 25 ″, and the image density D _{1 is obtained.}
The image density D ₂ = D ₁ + β ₂ (D ₁ −D _us.2 ) after the second intermediate processing is further applied to the signal representing

By repeating the above loop n times, the density D ′ after the final calculation processing is D ′ = D _n−1 −β _n (D _n−1 −D _us.n ) (13 ) Is required.

Thus, the image density D _1, D ₂ which has been subjected to intermediate treatment, ..., D _n-1
By calculating the density of the _{blurred mask} D _us.1 , D _us.2 , ..., D _{us.n and} the calculation represented by the equation (13) by using This particle noise can be effectively attenuated in a conspicuous region, and image performance such as sharpness and contrast can be positively improved in a region where the X-ray dose is large and the particle noise is originally small.

In the above equation (13), the same image densities D _n-1 are used as the image densities D _b1 and D _b2 as compared with the above equation (10). For example, the addition term calculating means shown in FIG. 2C is used. The signal representing the original image density D _org inputted at 24 ″ is stored, and the image density D ₁ , D ₂ , ..., D _n-1 after the intermediate processing is used in the calculation in the addition term calculating means 24 ″. Without using the original image density D _org , calculate β ₁ (D _org −D _us.1 ) β ₂ (D _org −D _us.2 ) ……………………………… The image densities D _b1 and D _b2 may be different, for example, by finally performing D ′ = D _n−1 −β _n (D _org −D _us.n ) (14).

(Effects of the Invention) In the X-ray image processing method of the present invention, an original photograph recorded with X-ray image information obtained by irradiating a photographic film with X-rays is scanned, and the original image density at each scanning point is read. Then, using a coefficient β that shifts from β <0 (β> -1) to β> 0 as the dose of X-rays applied to each point of the photographic film increases, D ′ = D _org + β (D _org -D _us ) ... Since the calculation is performed according to the equation (2), particle noise can be reduced in the region where the X-ray dose in the X-ray image is small and the particle noise is large, and the X-ray image can be reduced. In the region where the dose is high and the original particle noise is low, the image quality performance such as sharpness and contrast is positively improved, and the image quality performance of the entire X-ray image is improved.

Moreover, since the calculation of the formula (2) requires a short calculation time, the spare time is used to calculate the formula (3) described above. By performing the calculation according to, finer image processing can be performed.

Further, the apparatus for implementing the above method is not particularly complicated, and the calculation time can be set within a sufficiently allowable range.

[Brief description of drawings]

FIGS. 1A, 1B, and 1C are graphs showing examples of the functional form of the coefficient β, and FIGS. 2A, 2B, and 2C are different configuration examples of the calculation unit 18 shown in FIG. FIG. 3 is a block diagram shown in FIG. 3, and FIG. 3 is a perspective view showing an example of an X-ray image processing apparatus for carrying out the X-ray image processing method of the present invention. 1 …… Original photograph 2,13 …… Motor, 3 …… Photo transport means 4 …… Laser, 6 …… Rotating polygon mirror 10 …… Concentrator 11 …… Photomultiplier 16 …… Amplifier, 17 …… A / D converter 18 …… Calculator, 19 …… Memory 20 …… Image display device

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location 9163-4C A61B 6/00 350 M

Claims (4)

[Claims] 1. An original photograph on which X-ray image information is recorded, obtained by irradiating a photographic film with X-rays, is scanned, the original image density at each scanning point is read, and then an X-ray image is formed on a copy photograph or the like. When reproducing as a visible image, the density of the blur mask obtained by averaging the original image densities in the surrounding predetermined range corresponding to each scanning point
D _us , the original image density is D _org , the coefficient corresponding to the blur mask is β, and the image density after the arithmetic processing is D ′, the coefficient β is the irradiation of each point on the photographic film. Β <0 (β> − as the X-ray dose increases
1) to β> 0, the coefficient β ₁ is used to perform the operation according to the equation D ′ = D _org + β (D _org −D _us ), and in one of the X-ray images, In the low dose area, the spatial frequency component higher than the spatial frequency component of the density D _{us of the} blur mask is attenuated, and in the high dose area, the spatial frequency component higher than the spatial frequency component of the density D _{us of the} blur mask is changed. An X-ray image processing method characterized by enhancing. 2. An original photograph on which X-ray image information is recorded, obtained by irradiating a photographic film with X-rays, is scanned, the original image density at each scanning point is read, and then an X-ray image is formed on a copy photograph or the like. Was reproduced as a visible image by averaging the original image densities within a predetermined peripheral range corresponding to each scanning point or the image densities obtained by intermediately processing the signals representing the original image densities 1
Or the density of a plurality of blur masks obtained by changing the predetermined range is D _us.k (k = 1, 2, ..., N; n is an integer indicating the number of blur masks), the original image density or this original image The image densities after intermediate processing of the signals representing the densities are D _b1 , D _b2 , and one or a plurality of coefficients respectively corresponding to the one or a plurality of blur masks are β _k (k
= 1,2, ..., n), where D'is the image density after calculation, at least one coefficient β _l (l) of the coefficients β _k (k = 1,2, ..., n) Is an integer in the range of 1 to n) is a function that changes from β ₁ <0 (β ₁ > -1) to β ₁ > 0 as the dose of the X-rays applied to each point of the photographic film increases. _Yes , using this coefficient β _l The calculation is performed in accordance with the equation to reduce the spatial frequency component higher than the spatial frequency component having the density D _{us.l of} the blur mask corresponding to the coefficient β ₁ in the low dose region in one X-ray image. Further, in a region where the dose is high, a spatial frequency component higher than a spatial frequency component included in the density D _{us.l of} the blur mask corresponding to the coefficient β ₁ is emphasized. 3. An original photograph on which X-ray image information is recorded, obtained by irradiating a photographic film with X-rays, is scanned, and the original image density at each scanning point is read, and then the original image density is displayed. The signal is processed by an arithmetic unit, and an X-ray is made on a copy photograph or the like based on the signal representing the processed image density.
In an X-ray image processing apparatus that reproduces a line image as a visible image, the calculation unit averages the original image densities within a predetermined range around each scanning point to obtain a density of a blur mask D _us , the original image density is D _org ,
When the coefficient corresponding to the blur mask is β and the image density after the arithmetic processing is D ′, the coefficient β becomes β <as the dose of the X-rays irradiated to each point of the photographic film increases. 0 (β>-
X-ray image processing characterized in that it is a function that shifts from 1) to β> 0, and that this coefficient β is used to perform an operation according to the equation D ′ = D _org + β (D _org −D _us ). apparatus. 4. An original photograph recorded with X-ray image information, which is obtained by irradiating a photographic film with X-rays, is scanned, and the original image density at each scanning point is read. In an X-ray image processing device for reproducing an X-ray image as a visible image on a copy photograph or the like based on a signal representing the image density after processing, wherein the operation unit corresponds to each scanning point. Then, one image density obtained by averaging the original image density within a predetermined surrounding range or an image density obtained by subjecting a signal representing the original image density to intermediate processing, or a plurality of blur masks obtained by changing the predetermined range , D _us.k (k = 1, 2, ..., N; n is an integer indicating the number of blur masks), the original image density or an image density obtained by performing an intermediate process on a signal representing the original image density is D _b1 and D _b2 , one or a plurality of coefficients respectively corresponding to the one or a plurality of blur masks, β _k (k = 1, 2, ..., N), and the image density after the arithmetic processing is D ′.
Then, at least one coefficient β _l (l is an integer within 1 to n) of the coefficients β _k (k = 1, 2, ..., N) is applied to each point of the photographic film. It is a function that changes from β ₁ <0 (β ₁ > -1) to β ₁ > 0 as the dose of the X-ray increases, and by using this coefficient β _1. An X-ray image processing apparatus, characterized in that the calculation is performed in accordance with