JP2507659B2 - Radiation image reader - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Object of the Invention (Industrial field of application) The present invention provides a method for scanning radiation image information by scanning a laser beam on an image recording medium on which radiation image information is recorded. The present invention relates to a radiation image reading device that reads radiation image information from an image recording medium.

(Prior Art) As a method for diagnosing an image stored on an image recording medium, for example, a film, particularly a radiation image by an X-ray film, conventionally, the X-ray film was placed on a Schaukasten or the like and directly diagnosed visually. . On the other hand, in recent years, after narrowly focused laser light is scanned with respect to the image information stored in the film and converted into an electric signal, various image processing such as spatial frequency processing and gradation processing is performed. A method using a film image reading apparatus, in which information effective for medical diagnosis is emphasized, reproduced, and then diagnosed, has become more commonly used.

In the method using this film image reading apparatus, much diagnostic information can be obtained from the X-ray imaging of the first floor, and the diagnostic performance is improved. Further, it is advantageous in terms of efficiency of storage and retrieval of X-ray image information, and future development of this device is expected.

FIG. 4 is a specific configuration diagram of a system using such a film image reading apparatus, the details of which will be described below.

The image recorded on the film 1, which is an image recording medium, is read by scanning the film 1 with a laser beam by the film image reading device 2. This read information is sent to the data processing device 3 after being converted into a digital signal. The data processing device 3 performs data processing such as frequency emphasis and edge emphasis on the sent image information. An image having excellent diagnostic suitability is obtained by this data processing, and the obtained image data is displayed and output on the display device 4.

5 (a) and 5 (b) show the film image reading device 2 described above.
It is a conceptual diagram of. Reference numeral 5 is a laser oscillator for generating a laser beam, and 6 is a beam expander 6 for expanding the diameter of the laser beam incident from the laser oscillator 5 to an arbitrary size to reduce the divergence angle of the laser beam. is there. For example, when the beam diameter of the laser light is expanded five times with the beam expander 6, as a result, the spread angle of the laser light is reduced to 1/5.

Reference numeral 7 denotes an optical deflector 7 for reflecting the laser light incident through the beam expander 6 with a constant angular velocity in the main scanning direction, and a galvanometer or a polygon is generally applied. Reference numeral 8 denotes an fθ lens which, when the laser light having a constant angular velocity reflected by the optical deflector 7 is incident, serves to focus the laser light on the same plane while keeping the linear velocity constant.

9a and 9b are a pair of film feed rollers, which hold the film 10 on which image information is recorded,
Is run at a predetermined speed in a direction (sub-scanning direction) perpendicular to the main scanning direction. By the sub-scanning by the film feed rollers 9a and 9b and the main scanning by the optical deflector 7, the entire surface of the film 10 is scanned with the laser light.

Reference numeral 11 denotes a condenser for guiding the laser light that has passed through the film 10 and is incident to a detector 12 that converts it into an electric signal, and an optical system such as an optical fiber or a lens in a bundle form the incident light. It is composed of a transparent acrylic resin or the like whose output end is processed so as to be efficiently transmitted to the detector 12.

An electronic circuit, which will be described later, is connected to the subsequent stage of the detector 12, and the electronic circuit converts the light transmitted through the film 10 into time series digital signals in association with the positional information of each pixel.

The principle of film density measurement is as follows. Let the reference light amount incident on the detector 12 be I _0, and let the light amounts incident on the detector 12 with and without the film be I ₁ and I ₂ , respectively, and let the densities at that time be D ₁ and D _2. , The relationship between the density and the light amount is expressed by the following equation: D ₁ = −log (I ₁ / I ₀ ), (1) D ₂ = −log (I ₂ / I ₀ ), (2).

At this time, when the relational expression of the density D _{3 which} is the difference between the density D ₂ with the film and the density D ₁ without the film is calculated, D ₃ = D ₂ −D ₁ = log (I ₂ / I ₁ ) (3) which indicates the density of the film.

That is, when the density D _{1 in the} absence of a film is measured and stored in advance, the density D ₂ when the film is placed is measured and the difference from the previously stored density D ₁ is calculated,
The value indicates the film density D.

FIG. 6 shows the configuration of the electronic circuit of the film image reading device 2. Since the concentration has the dimension of “log” as described above, the signal converted into the electric signal by the detector 12 is first log-converted by the log amplifier 13.

The sample / hold circuit 14 holds the output signal of the log amplifier 13 at the preceding stage in synchronization with a clock signal input from a controller (not shown) that controls the entire electronic circuit. The A / D converter 15 converts the signal held by the sample / hold circuit 14 in the preceding stage into a digital signal.

The above controller divides continuous image information into pixels by generating a clock signal corresponding to the speed of main scanning. The switch 16 sends the signal output from the A / D converter 15 to either the calibration buffer 17 or the line buffer 18 according to the signal instructed by the controller. That is, the switch 16 sends the density information when the film 10 is not placed in FIG. 5 to the calibration buffer 17, and sends the density information when the film 10 is placed to the line buffer 18. The switching operation is performed as described above.

The calibration buffer 17 and the line buffer 18 are circuits each having a function of temporarily storing a digital signal of the number of pixels for one line, and the position of the pixel is updated while sequentially updating the address by the clock signal input from the controller. Information and stored information are associated with each other.

Reference numeral 19 denotes a difference calculation circuit 19, which corresponds to the difference between the density information of the film 10 stored in the line buffer 18 in the preceding stage and the density information stored in the calibration buffer 17 by the signal input from the controller. It has a function of calculating for each pixel, and this difference becomes the density of the film.

20 is an interface (I / F), the difference calculation circuit 1
It serves to send the output signal of 9 to the data processing device 3.

In the above configuration, first, the density of one line when the film 10 is not present is measured and stored in the calibration buffer 17. Next, the film feed rollers 9a and 9b move the film 10 in the sub-scanning direction to the main scanning line by the optical deflector 7, whereby the density of one line with the film is measured, and the measured value is the line It is stored in the buffer 18.

After that, the contents of the calibration buffer 17 and the line buffer 18 according to the signal from the controller,
It is sent to the difference calculation circuit 19 one after another. The difference calculation circuit 19 calculates the difference between the two contents, that is, the density of the film, and sends the calculation result to the data processing device 3 via the interface 20.

After the data for one line is transferred to the data processing device 3, the film feed rollers 9a and 9b are moved by a predetermined distance,
That is, the film 10 is moved by the width of one line, and then the measurement is performed at a new line position. The density of the entire surface of the film 10 is measured by sequentially repeating such operations.

By the way, one of the main causes of degrading the image quality of the image stored in the film 10 is scattered radiation generated by the interaction between the subject and the X-ray. This scattered radiation is imaged on the film and reduces the contrast of the image. Therefore, in order to reduce the amount of scattered radiation reaching the film, grit for removing scattered radiation is widely used.

Normal grit has a density of about 28 to 40 lines / cm, and when this grit is left stationary during X-ray exposure, streaks visible to the naked eye appear on the film image. Affect doctor's diagnosis. For this reason, a buki device is used to reciprocate the grit during X-ray irradiation so that the grit is not transferred onto the film.

However, recently, in order to increase the sharpness of an image, a high-density grit having a grit density of about 60 lines / cm has begun to be used. Since the resolution of the human eye is about 50 lines / cm, the fine grit fringes due to the high-density grit imprinted on this film cannot be seen. Therefore, with a high-density grit, even if the image is taken in a stationary state, the doctor's diagnosis will not be hindered. Also, static grit has the advantage of reducing motion blur. For this reason, when a high-density grit is used, it tends to be imprinted on the film.

(Problems to be Solved by the Invention) However, when a film on which grit is imprinted is measured by a conventional film image reading device, a false image is generated in the read image. Since this false image is displayed overlapping the captured X-ray image, it adversely affects the doctor's diagnosis and must be removed.

Here, generation of a false image due to grit will be described.

When the grit frequency is higher than the Nyquist frequency, a false image occurs at the position where the grit frequency is folded back at the Nyquist frequency. That is, assuming that the sampling frequency is Fs and the grid frequency is Fg, the false image frequency Fa is Fa = | (1/2) Fs− (Fg− (1/2) Fs) | = | Fs−Fg | (4) Becomes

For example, as shown in FIG. 7, when a 5 Hz original image is sampled at 4 Hz, a 1 Hz false image occurs. The sampling pitch of the radiation image reading device is generally 100 to 200 μm.
This is a sampling density of 50-100 lines / cm.
Is equivalent to Therefore, for a high-density grit of 60 lines / cm, the Nyquist frequency is exceeded and a false image occurs.

Also, for example, a film on which a grit with a grit density Dg of 60 pieces / cm is imprinted has a sampling pitch Ps = 200.
The grit frequency Fg [MHz] when read in μm is
If the sampling frequency Fs at this time is 1.5MHz, it will be 1.8MHz from the formula of Fg = Ps x Dg x Fs (5), but the sampling frequency is 1.5MHz.
Therefore, in this case, a false image having a frequency of 0.3 [MHz] (= 1.8-1.5) is generated.

From the above equation, it can be seen that the grid frequency changes with the sampling pitch. Sampling pitch is 10
The grid frequencies for 0 μm and 150 μm are 0.9 MHz each
And 1.3MHz.

The present invention has been made in view of the above circumstances,
An object of the invention is to provide a radiation image reading apparatus capable of obtaining a radiation image from which a false image caused by a high-density grit projected on an image recording medium is completely removed.

[Structure of the Invention] (Means for Solving the Problems) The present invention is directed to scanning an image recording medium on which radiation image information is recorded with a laser beam and sampling the radiation image information so that the radiation image information is recorded from the image recording medium. In a radiation image reading apparatus for reading radiation image information, a low-pass filter having a variable cutoff frequency is provided between a log amplifier and a sample / hold circuit, and a preliminary image of an image recording medium is provided in reverse to an actual image reading operation. If you read the image information of one line at an arbitrary position where is captured, apply FFT (Fast Fourier Transform) to calculate the frequency spectrum, and if the spectrum due to grit is found, , The relationship between the discovered spectrum and the cutoff frequency for blocking the spectrum is stored. Referring to Table, by determining automatically the cutoff frequency of the low pass filter is obtained by so as to pass through the necessary image information to remove only the component of the grit.

(Operation) As described above, according to the configuration of the present invention, a low-pass filter having a variable cutoff frequency is provided between the log amplifier and the sample / hold circuit, and a part of the image data is preliminarily read and stored in the low-pass filter. When a grit is detected, the cutoff frequency for shutting off the grit is analyzed and automatically set in the low pass filter. Can be completely removed, necessary image information can be passed, and diagnosis can be performed without affecting.

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

FIG. 1 shows an electronic circuit configuration of a film image reading apparatus according to an embodiment of the present invention. Since the basic configuration is the same as that shown in FIG. 6, the same portions are the same. Reference numerals are given and detailed description thereof is omitted.

Then, log amp 13 and sample / hold circuit 14
A low-pass filter 21 is provided between the low-pass filter 21 and the low-pass filter 21 so that all log-converted image information output from the log amplifier 13 passes through the low-pass filter 21 and reaches the sample / hold circuit 14. This low-pass filter 21 has an amplitude of −3d.
A frequency at which the frequency becomes B, that is, a cutoff frequency can be variably set is used, and the setting operation is controlled by the microcomputer 22. The microcomputer 22 is also connected to the sample / hold circuit 14 and the A / D converter 15, and has an FFT algorithm function for performing a frequency analysis of the image information obtained here by an operation program, and It also has a table showing the relationship between the frequency of the grit detected by the FFT algorithm function and the cutoff frequency of the low-pass filter 21 that is optimal for blocking the frequency.

The cut-off frequency of the low-pass filter 21 is such that the frequency component of grit is cut off, but the frequency component derived from the biological information of the subject must be passed.

It is said that the frequency characteristics of human body images are within 2.5 lines / mm, and most of them are below 0.5 lines / mm. 2.5 / 0.75MHz at sampling pitch of 200μm
Equivalent to, 0.37M when sampling pitch is 100μm
Equivalent to Hz.

On the other hand, when the grid density is 60 lines / cm, the grid frequencies at the sampling pitches of 100 μm and 150 μm are 0.9 MHz and 1.3 MHz, respectively, as described above, which is sufficiently far from the human body image frequency band. There is. Therefore,
It is possible to set a cutoff frequency that allows image information to pass through the low-pass filter 21 and cuts off the grid frequency.

In order to pass the image information up to the highest possible frequency, it is desirable that the attenuation gradient of the low-pass filter 21 be as steep as possible. When the attenuation gradient of the low-pass filter 21 is set to −48 dB / oct and the grid frequency 0.9 MHz at the sampling pitch of 100 μm needs to be attenuated to −30 dB, the cut-off frequency of the low-pass filter 21 is as shown in FIG. It becomes 0.62MHz. At this time, the image frequency of 0.37 MHz corresponding to the frequency characteristic of 2.5 lines / mm of the human body image can pass through this filter with the attenuation of 0.

When the attenuation gradient and the grit frequency attenuation are determined, the cutoff frequency is naturally determined. In the case of FIG. 2, when the sampling pitch is 150 μm and 200 μm, the cutoff frequencies are 0.9 MHz and 1.2 MHz, respectively. However, in this case, since the sampling frequency is set to 1.5 MHz, the Nyquist frequency is 0.75 MHz. If the frequency of the image exceeds the Nyquist frequency, a correct image cannot be reproduced by the sampling theorem, so the upper limit of the cutoff frequency of the low pass filter 21 is the sampling frequency.
It becomes 0.75MHz which is equivalent to 1/2.

On the other hand, the microcomputer 22 performs so-called "spectral analysis" in which the spectrum is obtained from the image information obtained by the sample / hold circuit 14 and the A / D converter 15 and the frequency of the signal is analyzed.

Generally, the spectrum is obtained by Fourier-transforming the obtained signal, and by converting the image information into the spectrum form, the property of the waveform of the image can be easily expressed.

In order to perform the Fourier transform of the digitized discrete signal sequence by the microcomputer 22, it is necessary to rewrite the definition formula of the Fourier transform in a discrete form and to perform an enormous number of complex number operations. How microcomputer
However, since it takes a lot of time to calculate, which is 22. Here, the FFT algorithm that can dramatically reduce the calculation time is used. This FFT algorithm program can be built in the microcomputer 22 easily because it can be constructed in a very short program of about 30 steps even if it is built in BASIC, for example.

The spectral distribution of an image can be known by spectrally analyzing one line of image data with this FFT. Third
The image shows a grit with a grit density of 60 lines / cm and a sampling pitch of 100 μm and a sampling frequency of 1.5.
It shows the spectral distribution when measured in MHz. Since a steep waveform is obtained for the spectrum of one grid, it can be easily known by the microcomputer 22 that the spectrum exists at 0.9 MHz in this example.

The table built in the microcomputer 22 will be described below.

As described above, if the attenuation gradient of the low-pass filter 21 and the attenuation degree of the grid frequency are determined, the cutoff frequency can be determined naturally. Further, for example, when the sampling pitch is 100 μm and the grit spectrum obtained at that time is 0.9 MHz, it can be easily known that 60 grit / cm grit is imaged.

Therefore, the table of the microcomputer 22 shows that the cutoff frequency is 0.62 for a grid frequency of 0.9 MHz.
MHz, the cutoff frequency is 0.75MHz for the grid frequency of 1.2MHz or 1.8MHz.
The frequency of the grit spectrum obtained in
The cutoff frequency for attenuation to dB is stored in association.

 Next, the operation of the above embodiment will be described below.

Normally, it is meaningless to set the cutoff frequency of the low-pass filter 21 to a frequency equal to or higher than the Nyquist frequency according to the above-mentioned sampling theorem, so the cutoff frequency is fixedly set to the Nyquist frequency.

Prior to the actual image reading operation, the microcomputer 22, which is a controller, preliminarily moves the film 10 to an appropriate degree, first reads one line of data, and spectrally analyzes it by FFT. Therefore, the cutoff frequency of the low-pass filter 21 is variably set according to whether or not the spectrum due to the grid is found.

That is, when the frequency component due to the grit is found, a table contained in the microcomputer 22 is referred to search for a cutoff frequency that attenuates the spectrum to a predetermined attenuation degree, and the cutoff frequency is low-pass filtered. Set to 21. After that, the microcomputer 22 searches the film 10 back and starts the actual reading operation from the leading end of the film 10.

The frequency component of the grit in the image information thus obtained is attenuated to −30 dB by the low-pass filter 21, so the amplitude is compressed to about 3%. That is, the concentration of grit projected on the film 10 is generally 0.8.
The density of the grit that has passed through the low-pass filter 21 is about 0.024, although it is about the same. Since the density difference that can be recognized by human eyes is about 0.03, the signal due to this grit has disappeared, and the image frequency of the human body can pass up to the upper limit. Note that such a low-pass filter 21
By using, it is possible to remove noise with a high frequency component that is mixed into the detector 12 from the outside.

If the frequency component due to the grid is not found, the cutoff frequency of the low pass filter 21 is left in the original state (Nyquist frequency). After that, the microcomputer 22 searches the film back and starts the reading operation from the leading edge of the film. The operation in this case is similar to that in the case of FIG.

It is needless to say that the present invention is not limited to the above embodiment and various modifications can be made.

For example, in the above embodiments, the image recording medium is a film. However, the present invention can be applied even when a medium other than the film, for example, an imaging plate containing a stimulable phosphor is used.

Further, in the above embodiment, the case where a high density 60 lines / cm grit was described was explained, but a low-resolution grit having a lower resolution than this, for example, a low pass filter even when 28 to 40 lines / cm grit was printed. The same effect can be obtained by setting the cutoff frequencies of 21 correspondingly.

[Effects of the Invention] As described in detail above, according to the present invention, a low-pass filter having a variable cut-off frequency is provided, and an image on an image recording medium is preliminarily captured before an actual image reading operation. Read the pixel information of one line at an arbitrary position, calculate the frequency spectrum by applying the read image information to the FFT, and if a spectrum due to grit is found, cut it to cut off the found spectrum and that spectrum. Since the cutoff frequency of the low-pass filter is automatically determined by referring to the table in which the relationship of the off-frequency is stored, only the grit component is removed and necessary image information is passed. , Even if a high-density grit that cannot be seen is imaged on the image recording medium, only the grit component is It is possible to resist the radiation image reading apparatus that can completely remove and pass necessary image information to eliminate deterioration of image quality and to perform diagnosis without causing misdiagnosis.

[Brief description of drawings]

FIG. 1 is a block diagram showing a circuit configuration of a film image reading apparatus according to an embodiment of the present invention, FIG. 2 is a diagram illustrating frequency-amplitude characteristics of the low-pass filter of FIG. 1, and FIG. FIG. 4 is a diagram illustrating the spectral distribution of a captured image, FIG. 4 is a block diagram showing a specific configuration of the entire system using a film image reading device, FIG. 5 is a conceptual diagram of the film image reading device, and FIG. FIG. 7 is a block diagram showing a circuit configuration of a conventional film image reading device, and FIG. 7 is a diagram illustrating generation of a false image. 1,10 …… film, 2 …… film image reading device, 3…
... Data processing device, 4 ... Display device, 5 ... Laser oscillator, 6 ... Beam expander, 7 ... Optical deflector 7,8
…… fθ lens, 9a, 9b …… film feed roller, 11…
… Concentrator, 12 …… detector, 13 …… log amplifier, 14 …… sample / hold circuit, 15 …… A / D converter, 16 …… switch, 17 …… calibration buffer, 18 …… Line buffer, 19 ... Difference calculation circuit, 20 ... Interface (I / F), 21 ... Low-pass filter, 22 ... Microcomputer.

Claims (3)

(57) [Claims] 1. A radiation image reading apparatus for reading radiation image information from the image recording medium by scanning the image on the image recording medium on which the radiation image information is recorded with light and sampling the radiation image information. Means for specifying the grid frequency of the radiographic image imprinted on, and a cutoff frequency corresponding to the grid frequency obtained by the specifying means is set, the filter for removing the frequency component due to the grid in the radiographic image information A radiographic image reading device comprising: 2. The radiographic image reading apparatus according to claim 1, further comprising a microcomputer having a memory for storing the cutoff frequency of the filter in a table format. 3. The radiation image reading apparatus according to claim 1, wherein the upper limit of the cutoff frequency of the filter is set to 1/2 of the sampling frequency.