JPH0664019B2 - Image display processor - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to an image display processing device connected to a measuring device having a low resolution such as an ultrasonic flaw detector, a radiation CT device, and a radar imaging device, and more particularly to a measuring device. The present invention relates to an image display processing device suitable for obtaining a high-resolution video despite its low resolution.

[Background of the Invention]

FIG. 25 is an explanatory diagram of an ultrasonic flaw detector for detecting the shape of a rectangular protrusion 50 of height “1” existing on a plane of height “0”. The ultrasonic beam 52 emitted from the ultrasonic flaw detection sensor 51 has a certain spread. When the protrusion 50 is imaged while scanning the sensor 51 in the X direction, the information obtained from the reflected wave that the ultrasonic beam 52 irradiates the plane position of the height “0” is “height = 0”, and the protrusion 50 The information obtained when irradiating the top of the is “height = 1”. However, the information obtained by irradiating the rising portion and the falling portion of the protrusion 50 is the average value of the reflected wave from the plane of height = 0 and the reflected wave from the top of height = 1. Therefore, the shape of the protrusion 50 obtained by scanning from the sensor 51 in the X direction is as shown in the lower part of FIG.
Compared to the true shape (the shape in the middle of Fig. 25), the shape becomes more blunt (blunt). In other words, the ultrasonic flaw detector has a low spatial resolution due to the spread of the ultrasonic beam, which causes a problem that only an image with low resolution can be obtained.

In order to improve the spatial resolution and the resolution, it is conceivable that the ultrasonic beam is focused and measured. But height =
When the ultrasonic beam is focused so that the plane portion of 0 is focused, the ultrasonic beam spreads at the top of the protrusion 50. It is preferable to be able to adjust the focus position of the ultrasonic beam in the height direction of the protrusion 50, but this is difficult. Even if the focus position can be adjusted, it is impossible to perform measurement while adjusting the focus position by following the shape of the measurement target whose shape is unknown.

[Object of the Invention]

An object of the present invention is to provide an image display processing device capable of obtaining a high-resolution image from a measurement signal obtained by using a low-resolution measuring device.

[Summary of the Invention] An object of the above is to provide an image display processing device that removes a shape blur in a captured image of the test body based on the spatial resolution of the measurement device and displays the image when the test device captures the image and displays the image as an image. In the above, first means for Fourier-transforming a measurement signal obtained by imaging a reference object having a known shape with a measuring device to obtain a first Fourier transform function, and Fourier-transforming a signal showing the known shape for a second Means for obtaining the Fourier transform function of the third Fourier transform function, and third means for obtaining the third Fourier transform function indicating the spatial resolution of the measuring device by dividing the first Fourier transform function by the second Fourier transform function. And a fourth means for Fourier-transforming a measurement signal obtained by imaging the test object to be measured with the measurement device to obtain a fourth Fourier-transform function, and the fourth Fourier-transform function as the fourth means. Fifth means for obtaining a fifth Fourier transform function in which the shape blur of the test body is removed by dividing by the Fourier transform function of No. 3, and the fifth Fourier transform function is subjected to inverse Fourier transform to obtain a power spectrum. And a sixth means for displaying the power spectrum as an image.

In the present invention, it is possible to obtain a high-resolution image by knowing the spatial resolution of the measuring device and correcting the measurement signal of the measuring device based on the spatial resolution. Therefore, in the present invention, a signal having a known shape is Fourier-transformed to obtain a second Fourier transform function, and separately from this, a measurement signal obtained when a reference object having a known shape is imaged by a measuring device is Fourier-transformed. Then, the first Fourier transform function is obtained, and the (first Fourier transform function) / (second Fourier transform function) = spatial resolution (third Fourier transform function) of the measuring device is obtained. On the other hand, the Fourier transform function of the true shape of the test body = (fourth Fourier transform function obtained by Fourier transforming the measurement signal obtained by imaging the test body with the measuring device)
/ (Spatial resolution of the measuring device), the fourth Fourier transform function is divided by the third Fourier transform function, and the inverse Fourier transform is performed, so that the true shape of the test object, that is, the shape without blurring is converted into an image. It becomes possible to display.

〔Example〕

 An embodiment of the present invention will be described below with reference to the drawings.

First, the principle of the present invention will be described. Non-destructive inspection devices that display the shape of internal defects or internal structures in images, such as ultrasonic flaw detectors, radiation CT, and microwave radar probing devices, are used for spatial measurement of ultrasonic waves, radiation, or microwaves. The image is blurred due to the spread and the size of the openings of those sensor receiving portions. This blur can be regarded as the spatial resolution peculiar to the measuring device. The spatial resolution is the spatial spread of a point object displayed as an image. The essence of the present invention is to grasp the spatial resolution peculiar to the measuring device, correct the blur of the image caused by the spatial resolution, and display the image close to the true shape. Therefore, first, assuming that the spatial resolution peculiar to the measuring device is known, the principle of deriving an image without blur from an image with blur will be described.

First, for simplicity, consider a mathematical process for correcting one-dimensional blur. Although the actual image is two-dimensional, the mathematical process for correcting the one-dimensional blur described below can be applied to the two-dimensional case in exactly the same manner. This is because a measuring device such as an ultrasonic flaw detector is configured to output a one-dimensional image signal and obtain a two-dimensional image by scanning in the main scanning direction and the sub-scanning direction.

Now, the image signal from the measuring device will be referred to as "spectrum" for convenience. The horizontal axis of the spectrum is the spatial position, and the vertical axis is the intensity. A spectrum (having a blur) measured by the measuring device is called a spectrum O and is shown in FIG. On the other hand, the spatial resolution that represents the blur of the measuring device
, And is shown in FIG. The spectrum O shown in FIG. 2 and the spectrum R shown in FIG.
If the spectrum is represented by (X), and the spectrum without blur is represented by the spectrum I and the function I (X), there is a relation of the following expression (1) between these functions.

Here, in order to derive the unknown function I (X), it is derived using the relationship described below.

First, the left side when Fourier transform is performed on both sides of Expression (1), It is defined as Here, j is an imaginary number and ω is a frequency. Similarly, the Fourier transform function of I (X) is The Fourier transform function of R (X) And defined by the following equations (3) and (4).

When the right side of the equation (1) is Fourier transformed, it becomes as follows.

By the way, the Fourier transform of the function f (X−x ₀ ) is obtained as follows. That is, X-x ₀ is converted into a variable by t.

here, Is the Fourier transform function of f (X). From the relationship of the expression (6), the part in [] in the expression (5) is as follows.

Therefore, equation (5) becomes Can be transformed into

When the relationship of the equation (1) is established, the following equation (9) represented by the product of Fourier transform functions is obtained as is apparent from the equation (8).
Get a relationship.

Since O (X) is the measured spectrum and R (X) is a spectrum assumed to be known, the true shape spectrum I (X) is derived in the form of the Fourier transform function of the following equation (10). it can.

Now, the Fourier transform function To return to the actual spatial spectrum I (X),
It is possible with the procedure of the inverse Fourier transform shown in 1).

From the expressions (10) and (11), I (X) is calculated by the following expression.

Next, how to know the spatial resolution of the measuring device, which has been assumed to be known, will be described. That is, in equation (12) Is obtained by Fourier transforming the measured spectrum O (X), The next problem is how to obtain the spectrum R (X) which is the original function of.

If the spectrum R (X) is the spatial resolution of the measuring device,
An image of a point-like object can be imaged by a measuring device, and the spatial spread of the image can be regarded as a spectrum R (X).
However, in reality, a point-like object cannot exist, and the spatial resolution may change depending on the position. Therefore,
The spatial resolution of the measuring device is derived by imaging the object of known shape with the measuring device and comparing it with the image of the true shape. The true image of the object may be derived from the formula (12) using the resolution thus derived. Therefore, in order to derive the resolution, that is, the spectrum R of the measuring device, if the spectrum Is that represents the true shape and the spectrum that represents the image of the known object are Os, then from the relationship of equation (9), To get here Of the spectrum R, Is the spectrum of Os, and Are Fourier transform spectra of the spectrum Is, respectively. According to this, the spatial resolution of the measuring device The defocused image obtained by the measurement Can be corrected by the procedure of Expression (12) to calculate the spectrum I (X) representing the true image.

FIG. 1 is a block diagram of an image display processing apparatus according to an embodiment of the present invention. First, the resolution or spectrum of the measuring device The operation of the embodiment shown in FIG.

Measuring device When obtaining the spectrum, the week force of the switching controller 3 is set to "1". As a result, the terminals a- of the switches 5, 6, 7
Conduction is established between c. Next, when the output of the selection instruction circuit 24 is set to "1", the output of the AND circuit 25 is set to "1", and the terminals ac of the switch 4 are brought into conduction. Now the spectrum of the measuring device It becomes possible to ask. It should be noted that the switches 5, 6, and 7 are brought into conduction between the terminals bc when "0" is output from the switching controller 3. Also, the switch 4 is an AND circuit 25.
Only when the "1" is output from the terminal, the terminals a and c are brought into conduction, and in other cases, the terminals b and c are brought into conduction.

Since the terminals a to c of the switch 4 are now in the introduction state, the contents stored in the reference shape memory 1 (the spatial position of the known shape is associated with the address and the strength is the stored content). , The Fourier transform function 9 is input to the Fourier transformer 9 via the switch 4, and the Fourier transform function Is calculated. Fourier transform function Is stored in the real and imaginary parts of the reference spectrum memory 10 via the switch 5.

Next, the output of the selection indicator 24 is set to “0”, the spectrum obtained by measuring the sample of the reference shape is input from the measuring device, and stored in the measured spectrum memory 2. The contents of the measured spectrum memory 2 are input to the Fourier transformer 9 via the switch 4, where the spectrum according to equation (4) Is calculated and divided into real part and imaginary part It is stored in the spectrum memory 11.

The contents of the spectrum memory 11 are input to the port A of the vector divider 14, and the contents of the reference spectrum memory 10 are input to the port B of the vector divider 14 via the switch 6. The vector divider 14 uses the values Ar (real part) and Ai (imaginary part) input to port A and the values Br (real part) and Bi (imaginary part) input to port B as ) (1
Calculate the real value Cr and imaginary value Ci according to 5).

Via Cr, Ci calculated by the above formula and switch 7, It is stored in the real and imaginary parts of the spectrum memory 12. This gives a spectrum for the spatial resolution of the measuring device. Is obtained.

Next, the output of the switching controller 3 is set to "0". And
The result of measuring the unknown sample is input from the measuring device to the Fourier transformer 9 via the switch 4, and according to the equation (2),
Spectrum Is calculated and divided into a real number part and an imaginary number part, and through the switch 5, It is recorded in the spectrum memory 11.

The contents of the spectrum memory 11 are transferred to the port A of the vector divider 14 via the switch 6, The contents of the spectrum memory 12 are divided by the switch 6
It is input to the port B of 14 and divided by the formula (14), (15)
The calculation is performed according to Via switch 7 It is recorded in the real number part and the imaginary number part of the spectrum memory 13, respectively.

The content of the spectrum memory 13 is input to the inverse Fourier transformer 15, where the spectrum I is calculated according to the equation (11), and the result is divided into the real part and the imaginary part in the I spectrum memory 16. Is stored. The power calculator 17 is I
Using the value Ir of the real part and the value Ii of the imaginary part of the spectrum memory 16, Ip is calculated according to the following equation (16).

Ip = (Ir ² + Ii ² ) ^1/2 (16) This value is recorded in the power spectrum memory 18.

The address converter 19 shifts the address of the contents of the memory 18,
The data is rearranged in the shift spectrum memory 20 and recorded. The address shift is performed as follows. There are addresses up to 2N,
When the content of the address k of the power spectrum memory 18 is represented by Ip (k), the content of the memory 20 of the address m according to the following equation (17).
Isp (m) determines. As described above, in this embodiment, the power spectrum is converted into a spectrum in which the coordinates are folded around the spectrum width. The reason for this will be described later.

The power converter 21 has the contents Isp (m) of the memory 20 at the address m.
The voltage value V _Y proportional, address conversion unit 22 outputs the vertical deflection and horizontal deflection signals of the address m proportional to the voltage value V _X of the memory 20 respectively image display device 23, thereby, the image display device 23 Displays a non-blurred image.

Next, an example will be described in which the thickness of an object is measured by a radiation transmission method, and the spectrum obtained thereby is processed by using the image display processing device shown in FIG. FIG. 4 is a diagram showing an example of a test body thickness measuring apparatus by a radiation transmission method. The radiation 101 emitted from the radiation source 100 is a collimator.
It is collimated by 103 and detected by the detector 104. Source 10
The detector 104 including 0 and the collimator 103 is translationally scanned in the X-axis direction. Then, the test body 102 is arranged between the radiation source 100 and the detector 104. The thickness measuring device 105 records the thickness of the test body 102 in the Y-axis direction at the scanning line X. Scan position X
Let φ (X) be the detection intensity of the radiation 101 by the detector 104. Letting φ _{0 be} the detected intensity in a place where the test body 102 is not present and d (X) be the thickness in the Y-axis direction of the test body 102, the detected intensity φ (X) can be expressed by the following equation (18).

φ (X) = φ ₀ exp {−μd (X)} (18) where μ is the linear absorption coefficient of the test body 102 for the radiation 101.

When the relation of equation (18) is transformed, it becomes like the following equation (19).

φ (X) / φ ₀ = exp {−μd (X)} (19) When the logarithm of both sides of equation (19) is taken, To get The thickness measuring device uses the relationship of the equation (20) and the thickness distribution d based on the detected intensity φ (x) of the transmitted radiation.
(X) is measured.

FIG. 5A is a diagram showing a thickness distribution obtained by measuring the test body 102, in which the vertical axis d represents the thickness and the horizontal axis x represents the position. FIG. 5 (b) is a diagram showing the progression thickness distribution of the test body 102. The thickness distribution obtained by the measurement shown in FIG. 5 (a) is a smooth distribution, that is, a blurred distribution, as compared with the true thickness distribution shown in FIG. 5 (b).

The resolution of the measuring device shown in FIG. 4 becomes the distribution shown in FIG. 6 (a), and this distribution is used as the spectrum R to display the thickness distribution without blur as an image display using the image display processing device shown in FIG. Then, the spectral distribution shown in FIG. 6 (b) is obtained. Comparing the true thickness distribution shown in FIG. 5 (b) and the spectrum shown in FIG. 6 (b), it can be seen that almost the same thickness distribution is obtained.

The image display processing apparatus shown in FIG. 1 sometimes emphasizes the variation when the measured value has a large variation in the spectrum measured by the measuring apparatus. Therefore, it is necessary to reduce the variation due to the measurement error in the measurement spectrum and derive the clear spectrum I. When the measured spectrum includes many variations due to measurement errors, the image of the spectrum I obtained by the image display processing device of FIG. 1 becomes like the spectrum shown in FIG. As is apparent from FIG. 7, the specific frequency component is generated in a ripple shape. To reduce this noise, the spectrum I may be smoothed to remove high spatial frequency components. That is, the smoothing process as shown in the following equation (21) is performed on the spectrum I shown in FIG. That is, for each point Xi, the average value of the spectral intensities Isp (Xi ₊ k) of the surrounding points Xi ₊ k is set as the new spectral intensity Isp '(Xi).

FIG. 8 shows a new spectrum obtained by smoothing the spectrum of FIG. 7 by the equation (21). Spectrum I shown in FIG.
To subtract the spectrum shown in FIG. 8 to obtain the spectrum shown in FIG. The spectrum of FIG. 9 is the ripple noise included in the spectrum of FIG. Therefore, the characteristic frequency of the ripple-like noise can be extracted by Fourier-transforming the spectrum shown in FIG.

FIG. 10 is a diagram showing an example of a spectrum obtained by Fourier transforming the spectrum shown in FIG. The power spectrum obtained by Fourier-transforming the spectrum of FIG. 9 has coordinate axes, the left end of the axis is the origin, the spatial frequency is ω = 0, and the center of the axis is ω =
ωmax (the cycle corresponds to twice the sampling interval in FIG. 9), and ω = 0 at the right end of the axis. That is, the spectrum becomes bilaterally symmetric at the center of the axis, and the center of the axis indicates a region where the spatial frequency is higher. This is because the Fourier transform is expanded by the cosine function [cos (−ωx)]. Therefore,
Since the spatial frequency of the ripple noise shown in FIG. 9 is high, the peak formed at the center of the ω axis in the spectrum shown in FIG. 10 is the noise caused by the ripple noise. Therefore, among the spectra shown in FIG. 10, the spectra within the range of the central width W on the ω axis (shown in FIG. 11).
Only extracted and shown in Figure 1 Spectrum stored in spectrum memory 13 By subtracting the spectrum in Fig. 11 from In particular, the frequency component forming the ripple noise can be removed.

The contents according to the above processing are shown in the figure as follows.
The image of FIG. 7 is Fourier transformed to obtain the spectrum shown in FIG. From this spectrum, subtract the spectrum of Fig. 11 which is a high frequency region component showing ripple noise,
The spectrum shown in FIG. 23 is obtained. This spectrum is
It is a component in the low frequency region of the spectrum in Fig. 22. When the spectrum of FIG. 23 is inverse Fourier transformed, an image having a shape in which ripple-like noise is suppressed as shown in FIG. 8 is obtained.

Here, the coordinate transformation by the inverse Fourier transform will be described.
Inverse Fourier transform of the spectrum of FIG. 23 yields the spectrum of FIG. The coordinate axis x'of this spatial spectrum has a spatial position x '=-xmax / 2 at the left end of the axis, x' = 0 at the center of the axis, and x '= xmax / 2 at the right end of the axis. The left side of this axis (range from x '=-xmax / 2 to x' = 0) is
This corresponds to the area of the real space position x = xmax / 2 to x = xmax. Therefore, at the center of the axis of x '(x' = 0), the left end position x '=-xmax / 2 on the left side (range from x' =-xmax / 2 to x '= 0) is set to the right side (x' = 0 to x '= xmax
By folding back to the right end position x '= xmax / 2 (in the range up to / 2) and connecting them, an image along the real space position shown in FIG. 8 is obtained. After the inverse Fourier transform, it is necessary to fold the spatial position because the inverse Fourier transform is a sine function [sin (−ω
x)], and the spatial coordinates in the conversion process shift.

FIG. 12 is a block diagram of a ripple removing device that executes a series of processes (processes relating to FIGS. 7 to 11) for removing the above-mentioned ripple, and this ripple removing device is the image display process of FIG. Used by adding to the device. First
In FIG. 12, it is the equipment within the dotted line frame that executes the series of processes for ripple removal described above, and the equipment outside the dotted line frame is the same as the equipment shown in FIG. As shown, A switch 201 is inserted between the spectrum memory 13 and the inverse Fourier transformer 15, and a switch 202 is inserted between the power spectrum memory 18 and the address transformer 19. When the output of the ripple removal indicator 200 is set to "1", the terminals c and a of the switches 201 and 202 are brought into conduction, and when the output of the ripple removal indicator 200 is set to "0", the switches 201 and 2 are switched.
The terminal c and the terminal b of 02 become conductive. First, the output level of the ripple removal indicator 200 is set to "1".

The contents of the spectrum memory 13 are input to the inverse Fourier transformer 15 via the switch 201, and as described below with reference to the image display processing device of FIG. The spectrum with is stored. This spectrum is input to the smoothing calculator 203 via the switch 202, processed according to equation (21),
It is converted into the smoothed spectrum shown in FIG. 8 and stored in the smoothed spectrum memory 204.

Next, the subtractor 205 determines the contents of the power spectrum memory 18,
That is, the contents of the smoothed spectrum memory 204, that is, the smoothed spectrum is subtracted from the spectrum having ripples, and the ripple spectrum as shown in FIG. The Fourier transform 207 Fourier transforms the content of the subtraction spectrum memory 206, divides the ripple frequency spectrum shown in FIG. 10 into a real number part and an imaginary number part, and stores them in the noise spectrum memory 208. The contents of the noise spectrum memory 208 are as shown in FIG.
Only the spectrum having the width W in the central portion of the spectrum of FIG. 0 is stored, and the spectrum other than the width W is converted into a spectrum having 0, and the spectrum is stored in the ripple spectrum memory 210 by dividing into a real part and an imaginary part.

The subtractor 211, as shown in the figure, By subtracting the contents of the ripple spectrum memory 210 from the contents of the spectrum memory 13, the ripple removal spectrum memory 2
The real part and the imaginary part are stored separately in 12. Next, when the output of the ripple removal indicator 200 is set to "0", the contents of the ripple removal spectrum memory 212 are input to the inverse Fourier transformer 15 via the switch 201, and the image display processing device of FIG. As described above, the spectrum is stored up to the power spectrum memory 18, and the stored content flows to the address converter 19 via the switch 202 to display the image from which the ripple is removed.

Next, a specific example in which the blur and ripple in the measurement signal of the ultrasonic flaw detector are corrected by using the image display processing device in which the ripple removing device shown in FIG. 12 is added will be described.

FIG. 13 is a diagram showing an example of the ultrasonic flaw detector, which aims to visualize the irregularities on the surface of the test body. In FIG. 13, the probe 30
While scanning 0 in the X-axis direction by the scanning device 301, the ultrasonic beam 307 is emitted and the ultrasonic wave reflected from the surface of the test body 304 installed in the water 303 of the water tank 302 is received. Flaw detector 305
Then, the time interval t (X) from the ultrasonic transmission to the reception at the scanning position is measured, and the thickness d of the sample is calculated by the following equation (22).
(X) is measured.

d (X) = d ₀ −v · t (X) / 2 (22) In equation (22), d ₀ is the distance from the probe 300 to the bottom of the water tank 302, and v is the speed of sound in the water 303. Is. Ultrasonic beam 3
Since 07 spreads in water, the spatial resolution of the measurement system at this time is close to the spatial intensity distribution of the ultrasonic beam 307,
The spectrum R shown in FIG.

Due to this spatial resolution, when measuring the shape of the test body 304 having the thickness distribution as shown in FIG. 14 (a), the flaw detector 305 obtains the spectrum shown in FIG. 14 (c). When the spectrum shown in FIG. 14 (c) is used and only the image display processing device shown in FIG. 1 is used for visualization, a thickness distribution with many ripples may be obtained as shown in FIG. 15 (a). . However, if the ripple removing device shown in FIG. 12 is added to this image display processing device, as shown in FIG. 15 (b), a true thickness distribution without ripples can be obtained.

In this way, when there are many variations in the spectrum output from the measurement device, by adding the ripple removal device shown in Fig. 12, a clear ripple-free spectrum as shown in Fig. 15 (b) can be obtained. Can be derived.

As a method of reducing ripples and deriving a clear spectrum, another processing method different from that of the ripple removing apparatus of FIG. 12 can be adopted. Next, another processing method will be described.

First, the measured spectrum shown in FIG. 16 (a) has cos (2π
・ The spectrum shown in FIG. 16 (b) is obtained by multiplying (X / ΔX). In addition, the spectrum shown in FIG.
Multiply by (2π · X / ΔX) to obtain the spectrum shown in FIG. 16 (c). Next, when the spectrum of the real part of FIG. 16 (b) is taken and the Fourier transform of FIG. 16 (c) is taken as the spectrum of the imaginary part, the spatial frequency component of the period (1 / ΔX) is emphasized and the relative noise is relatively increased. It is possible to reduce the intensity of the harmonic component due to.

FIG. 17 is a diagram showing an example of a ripple removing apparatus that executes this processing. The ripple removing device in the inspection frame of FIG. 17 is a part to be added to the image display processing device of FIG.
The ripple removing device shown in FIG. 17 is inserted between the measured spectrum memory 2 and the switch 4 in FIG.

Next, the operation contents of the ripple removing device in FIG. 17 will be described.
In FIG. 17, the cos generator 400 uses the coordinate value x proportional to the address of the measured spectrum memory 2 to calculate the function cos (2
π · x / Δx) is generated. Similarly, the sin generator 401 also uses the coordinate value x to generate the function sin (2π · x / Δx).
The multiplier 402 multiplies the value of the cos generator 400 by the content of the measured spectrum memory 2 and records the result in the real part of the carrier spectrum memory 405. On the other hand, the multiplier 403 uses sin
The value of the generator 401 is multiplied by the content of the measured spectrum memory 2, and the result is recorded in the imaginary part of the carrier spectrum memory 405. The contents of the carrier spectrum memory 405 are input to the image display processing device of FIG.

The image obtained by adding the ripple removing device shown in FIG. 17 to the image display processing device shown in FIG. 1 may have some ripples as compared with the case where the ripple removing device is added. The lip shown in Fig. 15 (a) has the effect of reliably reducing the lip.

If both the ripple removing devices shown in FIGS. 12 and 17 are added to the image display processing device shown in FIG. 1, the effect will be further enhanced.

As described above, the image display processing apparatus according to the embodiment obtains the spatial resolution peculiar to the measuring apparatus, and the measured spectrum in which blurring occurs due to the resolution is a spectrum without blurring, that is, a measuring apparatus having higher resolution. The spectrum is converted into a spectrum equivalent to the spectrum measured in.

The operation of the image display processing apparatus shown in FIG. 1 is briefly shown in the flowchart of FIG. When the ripple removing device of FIG. 17 is added to the image display processing device of FIG. 1, instead of the process of step S1 in the flowchart of FIG. 18, instead of the flowchart of FIG. 19 and the process of step S2. 20th
It becomes the figure which inserted the flow chart of a figure.

The flow chart of the operation when the ripple removing device of FIG. 12 is added to the image display processing device of FIG. 1 is the flow chart shown in FIG. 18, and the flow chart of FIG. 21 is inserted between the processing of steps S3 and S4. It becomes the shape.

〔The invention's effect〕

According to the present invention, a measurement spectrum deteriorated by the spatial resolution peculiar to the measurement device can be converted into a spectrum with significantly improved resolution without any modification of the measurement device, and a clear image without blur is obtained. It becomes possible.

Furthermore, it is possible to easily reduce the ripple noise due to the influence of the variation caused by the measurement error included in the measurement spectrum by adding a ripple removing device, and it is possible to obtain a clear image with few errors. .

Furthermore, according to the present invention, since the characteristic of the measuring device and the spatial resolution are measured, the present invention can be applied to a measuring device having any resolution. Therefore, the ultrasonic flaw detection method, which was generally said to have poor resolution,
It becomes possible to display a high-resolution image from a measurement signal obtained by a radar imaging method, a radiation imaging method, or the like.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the image display processing device of the present invention, FIG. 2 is a diagram showing an example of a spectrum obtained by measurement with a measuring device having a low resolution, and FIG. 3 is a diagram showing the measuring device. FIG. 4 is a diagram showing an example of spatial resolution, FIG. 4 is a diagram showing an example of an apparatus for measuring the thickness distribution of a test body by a radiation transmission method, and FIG.
FIG. 5 (a) is a diagram showing an example of the thickness distribution of the test body measured by the apparatus shown in FIG. 4, FIG. 5 (b) is a diagram showing the true thickness distribution, and FIG. 6 (a) is shown in FIG. 6 (b) is a diagram showing the spatial resolution of the device shown in FIG. 6, FIG. 6 (b) is a diagram showing the thickness distribution of the test body derived from the measurement spectrum shown in FIG. 5 (a) using the image display processing device shown in FIG. FIG. 7 is a diagram showing an output spectrum obtained by the image display processing device of FIG. 1 when the spectrum obtained by measurement has a variation due to an error,
FIG. 8 is a diagram showing a spectrum obtained by smoothing the spectrum shown in FIG. 7, FIG. 9 is a diagram showing a spectrum of ripple noise obtained by subtracting the spectrum of FIG. 8 from the spectrum of FIG. 7, and FIG. FIG. 11 is a diagram showing a spectrum obtained by performing a Fourier transform on the spectrum of FIG. 9, FIG. 11 is a diagram showing a spectrum obtained by extracting a portion having a width W in the central portion of the spectrum of FIG.
The figure is a block diagram of the ripple removing device added to the image display processing device of FIG. 1, FIG. 13 is a diagram showing an example of the device for measuring the sample thickness by ultrasonic waves, and FIG. 14 (a) is the diagram of FIG. Fig. 14 is a diagram showing the true thickness distribution of the sample measured under the conditions, Fig. 14 (b) is a diagram showing the spatial resolution of the measuring device of Fig. 13,
FIG. 15C is a diagram showing the measured thickness distribution of the sample, and FIG. 15A is a diagram showing the true thickness distribution of the sample derived using the image display processing device shown in FIG. b) is a diagram showing the true thickness distribution of the sample derived by using the ripple removing device of FIG. 12, FIG. 16 (a) is a diagram showing an example of a measurement spectrum with many variations, and FIG. 16 (b) is The figure which shows the spectrum which multiplied the cos wave to the spectrum shown in FIG.
FIG. 6 (c) is a diagram showing a spectrum obtained by multiplying the spectrum shown in FIG. 16 (a) by a sin wave, FIG. 17 is a block diagram of a ripple removing device different from FIG. 12, and FIG. 18 is FIG. FIG. 19 is a flowchart for explaining the operation of the image display processing device shown in FIG.
Flowchart newly added to the flow chart of FIG. 18 by adding the ripple removal device of the figure to the image display processing device,
The figure also shows the flow chart of the newly added operation as in FIG. 19, and FIG. 21 shows the flow chart of the newly added operation in FIG. 18 when the ripple removal device of FIG. 12 is added to the image display processing device. FIG. 22 is a diagram showing a spectrum obtained by Fourier transforming the spectrum of FIG. 7, FIG. 23 is a diagram showing a spectrum obtained by subtracting the ripple-like noise spectrum of FIG. 11 from the spectrum of FIG. FIG. 24 is a diagram showing a spectrum obtained by performing an inverse Fourier transform on the spectrum of FIG. 23, and FIG. 25 is a diagram for explaining blurring caused by the ultrasonic flaw detector. 1 ... Reference shape memory, 2 ... Measurement spectrum memory, 3
...... Switching controller, 4,5,6,7 ...... Switch, 9 ...... Fourier transformer, 10 ...... Reference spectrum memory, 11 ...... Spectrum memory, 12 …… Spectrum memory, 13 …… Spectrum memory, 14 ... Vector divider, 15 ... Inverse Fourier transformer, 16 ... I spectrum memory, 17 ... Power calculator, 18 ... Power spectrum memory, 19 ... Address converter, 20 ... Shift Spectral memory, 21 ... Power converter, 22 ... Address converter, 23 ... Image display, 103 ...
… Collimator, 104 …… Detector, 105 …… Thickness measuring device,
200 …… Ripple removal indicator, 201,202 …… Switch, 20
3 ... Smoothing calculator, 204 ... Smoothing spectrum memory, 205
…… Subtractor, 206 …… Subtractive spectrum memory, 207 …… Fourier transform, 208 …… Noise spectrum memory, 209…
… Frequency gate, 210 …… Ripple spectrum memory, 2
11 ... Subtractor, 212 ... Ripple-removed spectrum memory, 300 ... Probe, 301 ... Scanning device, 305 ... Flaw detector, 400 ... Cos generator, 401 ... Sin generator, 402, 403 ...
… Multiplier, 405… Carrier spectrum memory.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Izumi Yamada 1168 Moriyama-cho, Hitachi City, Hitachi, Ibaraki Pref., Institute of Energy Research, Ltd. (72) Kazunori Koga 1168 Moriyama-cho, Hitachi City, Ibaraki Hitachi, Ltd. Energy Research Laboratory (72) Inventor Tadashi Sonobe 1168 Moriyama-cho, Hitachi City, Ibaraki Pref., Energy Research Laboratory, Hiritsu Seisakusho Co., Ltd. (56) Reference JP-A-50-137184 (JP, A)

Claims (4)

[Claims] 1. An image display processing apparatus for removing a shape blur in a captured image of the test body based on the spatial resolution of the measuring apparatus and displaying the image when the test apparatus captures the test body and displays it as an image. Means for obtaining a first Fourier transform function by Fourier transforming a measurement signal obtained by capturing an image of the reference object with a measuring device, and a second Fourier transform function by Fourier transforming a signal having the known shape. And a third means for obtaining a third Fourier transform function indicating the spatial resolution of the measuring device by dividing the first Fourier transform function by the second Fourier transform function. Means for obtaining a fourth Fourier transform function by Fourier transforming the measurement signal obtained by imaging the test object of 1. with the measurement device, and the fourth Fourier transform function for the third Fourier transform function. Fifth means for obtaining a fifth Fourier transform function by removing the shape blur of the test object by dividing it by a function, and inverse Fourier transform of the fifth Fourier transform function to obtain a power spectrum and obtain the power spectrum as an image. And a sixth means for displaying as. 2. The method according to claim 6, wherein
An image display processing device, wherein the means converts the power spectrum obtained by the inverse Fourier transform into a power spectrum in which coordinates are folded back around the spectrum width and displays the power spectrum as an image. 3. The method according to claim 1, wherein
The means smoothes the power spectrum, subtracts the smoothed power spectrum from the unsmoothed power spectrum to obtain a ripple spectrum, and Fourier transforms the ripple spectrum to obtain a ripple spectrum, within a predetermined range from the center of the spectrum width. Is extracted and the extracted spectrum is subtracted from the power spectrum obtained by the inverse Fourier transform to be displayed as an image. 4. The method according to claim 5, wherein:
An image display processing device, wherein the means divides a value obtained by multiplying the fourth Fourier transform function by a cosine function and a sine function of a predetermined period by the third Fourier transform function. .