JPH06274614A - Image processing method - Google Patents

Abstract

PURPOSE:To provide naturally reproduced images corresponding to visual impressions without emphasizing noise by performing image processing such as emphasis to signals in one frequency band at least among signals in plural frequency bands provided by performing wavelet transformation to image signals. CONSTITUTION:Since wavelet transformation 2 is performed to an image signal 1 expressing a radiograph and because the image signal 1 is decomposed into signals 3 in plural frequency bands, false vibrations are not generated in this signal like Fourier transformation. Therefore, prescribed image processing 4 such as blur mask processing, gradation processing or emphasis processing is performed to the signal 3 in one frequency band at least among these signals 3 in plural frequency bands, and inverse wavellite transformation 5 is performed to this signal, to which the image processing 4 is performed, and the other signals. Thus, the image in the desired frequency band can be naturally emphasized or the like corresponding to visual impressions without emphasizing noise.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method for reproducing a visible image based on an image signal representing a radiation image.

[0002]

2. Description of the Related Art There are various fields in which a recorded image is read to obtain an image signal, the image signal is subjected to appropriate image processing, and then the image is reproduced and displayed. For example, an X-ray image is recorded using a film with a low gamma value designed to be suitable for subsequent image processing, the X-ray image is read from the film on which the X-ray image is recorded, and an electric signal (image signal) is read. ), The image signal is subjected to image processing, and then reproduced as a visible image on a copy photograph or the like to obtain a reproduced image with good image quality performance such as contrast, sharpness, and graininess. There
-5193).

In addition, the applicant of the present invention has conducted radiation (X-ray, α
Radiation, β rays, γ rays, electron rays, ultraviolet rays, etc.), some of this radiation energy is accumulated, and when excitation light such as visible light is subsequently irradiated, accumulation that shows stimulated emission according to the accumulated energy Using fluorescent phosphor (stimulable phosphor),
Radiation image information of a subject such as a human body is once recorded on a sheet-shaped stimulable phosphor, and the stimulable phosphor sheet is scanned with excitation light such as laser light to generate stimulated emission light, and the obtained luminescence is obtained. A radiation image recording / reproducing system has already been proposed which photoelectrically reads out the emitted light and obtains an image signal, and outputs a radiation image of a subject on the basis of the image signal as a visible image on a recording material such as a photographic photosensitive material or a CRT display device. (JP-A-55-12429, JP-A-56-11395, JP-A-55-1)
63472, 56-104645, 55-116340, etc.).

This system has the practical advantage of being able to record images over a very wide radiation exposure area as compared to conventional radiographic systems using silver halide photography. That is, in the stimulable phosphor, it has been recognized that the amount of emitted light stimulated by excitation after storage is proportional to the radiation exposure dose over a very wide range, and therefore the radiation exposure dose varies depending on various imaging conditions. Even when the value fluctuates considerably, the amount of stimulated emission light emitted from the stimulable phosphor sheet is set to an appropriate value for the reading gain, read by the photoelectric conversion means, and converted into an electrical signal (image signal). By using this image signal to output a radiation image as a visible image on a recording material such as a photographic light-sensitive material or a display device such as a CRT display device, a radiation image that is not affected by fluctuations in radiation exposure amount can be obtained.

In the various systems described above, it is general to perform various image processes on the image signal obtained in order to obtain a visible image suitable for observation and diagnosis as described above. As this image processing, a gradation processing for adjusting the density of a radiation image reproduced according to the level of an image signal is known.

In order to improve the diagnostic performance of radiographic images, the applicant of the present invention has proposed a method of subjecting image signals to frequency enhancement processing such as blur mask processing (Japanese Patent Laid-Open No. 55-163472). JP-A-55-87953). In this frequency processing, the read image signal Sorg is added with a value obtained by subtracting the blur mask signal Sus from the read image signal Sorg and multiplying it by the emphasis degree β, whereby a predetermined spatial frequency component in the image is obtained. Is emphasized. This can be expressed by the following formula.

S = Sorg + β (Sorg−Sus) (1) (S: frequency-processed signal, Sorg: read image signal,
Sus: blurring mask signal, β: enhancement degree) As another method of performing frequency processing on an image signal, a method using frequency analysis by Fourier transform is known. This method Fourier transforms the image signal,
The image signal is decomposed into signals in a plurality of frequency bands, and a predetermined image processing such as enhancement is applied to a signal in a desired frequency band among the decomposed signals. The frequency analysis method by this Fourier transform, for the image signal,

[0008]

[Equation 1]

Is a method of analyzing which frequency band component is included in this image signal by the following equation:
It is a method used for analysis in various fields such as vibration analysis and spectrum analysis.

[0010]

However, in the Fourier transform, the basis function when the image signal is decomposed into a plurality of signals for each frequency band is a trigonometric function, so that the characteristics such as the amplitude of the signal are different for each part of the signal. When it changes to, there is a problem of false vibration that vibration is generated in a portion of the image signal that is not originally vibrated. For example, when a frequency signal obtained by performing a Fourier transform on a signal Sorg as shown in FIG. 7 and performing an inverse Fourier transform for each frequency band is used, the original signal Sorg is obtained.
Of the frequency band 7
It can be seen that false vibration is generated in the corresponding portion A'of. This occurs because the non-vibrating part is affected by other vibrating parts because the Fourier transform tries to force the original signal Sorg to match the sine wave.

As described above, since false vibration occurs in the Fourier transform, if frequency processing is performed by using the Fourier transform, when the image signal subjected to this frequency processing is reproduced, the reproduced image is visually There is a problem that the image does not match the impression and the image is unnaturally emphasized.

Further, since the above-mentioned blur mask processing emphasizes all the components higher than a predetermined frequency, in the method of performing the frequency enhancement processing using this blur mask processing, only the desired frequency component is obtained. In some cases, noise included in the image signal may be emphasized, and it is not possible to obtain a reproduced image having excellent observation and interpretation suitability.

In view of the above circumstances, it is an object of the present invention to provide an image processing method for an image signal representing a radiation image, which is excellent in observation and interpretation suitability and can perform natural enhancement or the like according to a visual impression. To do.

[0014]

According to a first image processing method of the present invention, when reproducing a visible image of a radiation image from an image signal representing the radiation image, the image processing is performed on the image signal. In the method, the image signal is decomposed into signals in a plurality of frequency bands by applying a wavelet transform to the image signal, and at least one of the plurality of frequency bands is decomposed.
Perform predetermined image processing on signals in one frequency band,
It is characterized in that an inverse wavelet transform is applied to the image-processed signal and other signals.

A second image processing method according to the present invention is the above-described first image processing method according to the present invention, wherein the predetermined image processing multiplies a signal in the at least one frequency band by a predetermined number. It is characterized by being processing.

Now, the wavelet transform will be described.

The wavelet transform has been recently developed as a method of frequency analysis, and has been applied to stereo pattern matching, data compression, etc. (OLIVIER RIOUL and MARTIN VETTERLI; Wavelets a
nd Signal Processing, IEEESP MAGAZINE, P.14-38, OCTOB
ER 1991, Stephane Mallat; Zero-Crossings of a Wavel
et Transform, IEEE TRANSACTIONS ON INFORMATION THEO
RY, VOL.37, NO.4, P.1019-1033, JULY 1991).

This wavelet transform uses a function h as shown in FIG. 8 as a basis function.

[0019]

[Equation 2]

In the equation, the signal is decomposed into signals in a plurality of frequency bands, so that the problem of false oscillation unlike the Fourier transform does not occur. That is, if the filtering process is performed by changing the period and reduction ratio of the function h and moving the original signal, it is possible to create a signal suitable for a desired frequency from a fine frequency to a coarse frequency. For example, as shown in FIG. 9, when the signal Sorg, which is the same as the signal shown in FIG. 7 described above, is wavelet-transformed and inverse-wavelet-transformed for each frequency band, the signal Sorg corresponds to the oscillation of the original signal Sorg. A signal in the specified frequency band can be obtained. That is, in the portion B'of the frequency band W7 corresponding to the portion B of the original signal Sorg, vibration is not generated as in the original signal.

[0021]

In the image processing method according to the present invention, the image signal representing the radiation image is subjected to the wavelet transform so that the image signal is decomposed into signals in a plurality of frequency bands. As in the Fourier transform shown in FIG. 7, no false vibration is generated. Therefore, at least one of the signals in the plurality of frequency bands is
Noise is emphasized by applying predetermined image processing such as blur mask processing, gradation processing, and emphasis processing to signals in one frequency band, and applying inverse wavelet transform to this image processed signal and other signals. Without doing so, it is possible to perform natural enhancement or the like that has a visual impression on an image in a desired frequency band.

It is mathematically guaranteed that the original image can be completely restored when the wavelet transformed image signal is subjected to the inverse wavelet transform without being subjected to the image processing. Therefore, according to the image processing method of the present invention, , The information of the original radiation image is not lost.

[0023]

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

FIG. 1 is a diagram showing the basic concept of the image processing method according to the present invention. As shown in FIG. 1, the image processing method according to the present invention applies a wavelet transform 2 to an image signal 1 representing a radiation image to decompose the image signal 1 into signals 3 in a plurality of frequency bands. Next, a predetermined image processing 4 is applied to at least one frequency band signal 3 of the signals 3, and an inverse wavelet transform 5 is applied to the signal subjected to the image processing 4 and other signals. It is a thing.

An embodiment of the image processing method according to the present invention will be described in detail below. FIG. 2 is a diagram illustrating an example of the radiation image reading apparatus.

An image is taken by a radiation imaging device (not shown), and the stimulable phosphor sheet 14 on which the radiation image is recorded is set at a predetermined position of the reading device 20.

When the stimulable phosphor sheet 14 is set at a predetermined position of the reading device 20, the sheet 14 is conveyed (sub-scanned) in the arrow Y direction by the endless belt 22 driven by the motor 21. On the other hand, a light beam 24 emitted from a laser light source 23 is reflected and deflected by a rotating polygon mirror 26 driven by a motor 25 and rotating at a high speed in the direction of an arrow, and passes through a focusing lens 27 such as an fθ lens, and then an optical path by a mirror 28. Instead, the light is incident on the sheet 14 and the sub-scanning direction (arrow Y
Direction) and main scanning is performed in the direction of arrow X, which is substantially perpendicular to the (direction). From the location where the excitation light 24 of the sheet 14 is irradiated, the stimulated emission light 29 of a light amount corresponding to the stored and recorded radiation image information is diverged, and this stimulated emission light 29 is guided by the light guide 30. It is detected photoelectrically by a photomultiplier (photomultiplier tube) 31. The light guide 30 is made by molding a light guide material such as an acrylic plate, and is arranged so that the linear incident end face 30a extends along the main scanning line on the stimulable phosphor sheet 14. The injection end face 30b formed in a ring shape
Is connected to the light receiving surface of the photomultiplier 31.
The stimulated emission light 29 that has entered the light guide 30 from the incident end face 30a.
Travels through the light guide 30 by repeating total reflection, is emitted from the emission end face 30b and is received by the photomultiplier 31, and the stimulated emission light 29 representing the radiation image is converted into an electric signal by the photomultiplier 31. To be done.

The analog output signal S0 output from the photomultiplier 31 is logarithmically amplified by the logarithmic amplifier 32 and digitized by the A / D converter 33, whereby the image signal Sorg is obtained and the image processing device 40 is obtained. Entered in.
The image processing device 40 includes a CRT display 41 for reproducing and displaying a visible image, a main body 42 having a CPU, an internal memory, an interface and the like built therein, and a floppy disk drive unit 4 in which a floppy disk is loaded and driven.
3, and a keyboard 44 for inputting information necessary for this X-ray image reading apparatus.

The image processing described above is performed in the image processing device 40.

FIG. 3 is a diagram showing details of the wavelet transform for the image signal Sorg.

In this embodiment, the orthogonal wavelet transform in which each coefficient of the wavelet transform is orthogonal is performed.

As shown in FIG. 3, the function g obtained from the basic wavelet function in the main scanning direction of the image signal Sorg.
_The filtering process is performed by ₀ and the function h ₀ . That is, the filtering processing for each column of pixels arranged in the main scanning direction by such functions g ₀ and h _{0 is} performed while shifting each pixel in the sub scanning direction to obtain the wavelet transform coefficient signal Wg0 of the image signal Sorg in the main scanning direction. , Wh0.

Here, the functions g ₀ and h ₀ are uniquely obtained from the basic wavelet function. For example, in the case of the basic wavelet function shown in FIG. 8, they are as shown in Table 1 below. Further, when the functions g ₀ and h ₀ are shown in the figure,
The function g ₀ is shown in FIG. 4 (a), and the function h ₀ is shown in FIG. 5 (a).

[0034]

[Table 1]

In this way, when the wavelet transform coefficient signals Wg0 and Wh0 are obtained, the functions g ₀ and h in the sub-scanning direction of the wavelet transform coefficient signals Wg0 and Wh0, respectively.
Performs filtering processing by _0, obtain wavelet transform factor signals WW _0, WV _0, VW ₀ and VV _0. Then, filtering processing is performed in the main scanning direction of the wavelet transform coefficient signal VV _{0 by} the functions g ₁ and h ₁ .

That is, as in the case of g ₀ and h ₀ described above, the filtering process for each column of pixels arranged in the main scanning direction by the functions g ₁ and h _{1 is} performed while shifting each pixel in the sub scanning direction, and the wavelet Wavelet transform coefficient signals Wg1 and Wh1 of the transform coefficient signal VV _{0 in} the main scanning direction
Is to seek.

Here, the functions g ₁ and h ₁ are the functions g ₀ and h
₀ in which was generated by inserting a 0 1 between each sampling point. Further, when the functions g ₁ and h ₁ are shown in the figure, the function g ₁ is shown in FIG. 4 (b), and the function h ₁ is shown in FIG. 5 (b).

In this way, when the wavelet transform coefficient signals Wg1 and Wh1 are obtained, the functions g ₁ and h in the sub-scanning direction of the wavelet transform coefficient signals Wg 1 and Wh ₁ , respectively.
Performs filtering processing by ₁ to obtain the wavelet transform factor signals WW _1, WV _1, VW ₁ and VV _1.

Thereafter, in the same manner as described above, the functions g ₂ , h in the main scanning direction of the wavelet transform coefficient signal VV ₁ are obtained.
_The filtering process by ₂ is performed, and the obtained wavelet transform coefficient signal is processed by the function g in the sub-scanning direction.
₂ and h _{2 are} filtered to obtain wavelet transform coefficient signals WW ₂ , WV ₂ , VW ₂ and VV ₂ .

Here, the functions g ₂ and h ₂ are the functions g ₀ and h _0.
It is created by inserting two 0s between each sampling point of, and the function g ₂ is the function h ₂ in FIG. 4 (c).
Are shown in FIG. 5 (c), respectively.

By repeating such wavelet transform N times, wavelet transform coefficient signals WW _{0 to} W
_{W N, WV 0 ~WV N,} VW 0 ~VW N, and VV _N
To get Here, the functions g _N and h _N used in the N-th wavelet transform are created by inserting 2 ^N-1 zeros between the sampling points of the functions g ₀ and h ₀ . Therefore, the function g _N, h _N becomes a large function of the period as N is large.

Here, the wavelet transform coefficient signal WW
_i (i = 0 to N, the same applies hereinafter) represents a change in the frequency of the image signal Sorg in both the main and sub directions, and the larger i is, the lower the frequency signal becomes. The wavelet transform coefficient signal WV _i represents the change in the frequency of the image signal Sorg in the main scanning direction, and the larger i is, the lower the frequency signal becomes. Further, the wavelet transform coefficient signal VW _i is the image signal Sor
It represents a change in frequency of g in the sub-scanning direction, and the larger i is, the lower the frequency signal becomes.

Then, a table β preset for the wavelet transform coefficient signal thus obtained is set.
_The desired frequency band is emphasized by multiplying _WW (i), β _WV (i), and β _VW (i). That is, for the i-th wavelet transform coefficient signal WW _i , WV _i , VW _i , WW _i ′ = β _WW (i) × WW _i (4) WV _i ′ = β _WV (i) × WV _i (5) The calculation of VW _i ′ = β _VW (i) × VW _i (6) is performed.

Here, when it is desired to emphasize the high frequency component of the radiation image, β _WW (i), β _WV (i), and β _VW (i)
The smaller i is, the larger the value should be set. Further, when the low frequency component is desired to be emphasized, the larger i is, the larger the value should be set. Furthermore, if we want to emphasize only the main-scanning direction component of the radiation image, for β _WV (i), β _WW (i)
, Β _VW (i).

The wavelet transform coefficient signal W multiplied by β _WW (i), β _WV (i) and β _VW (i) as described above.
Once W _i ′, WV _i ′ and VW _i ′ have been determined, these coefficient signals WW _i ′, WV _i ′, VW _i ′ and VV _i
Inverse wavelet transform is performed using _i ′.

FIG. 6 is a diagram showing details of the inverse wavelet transform.

[0047] As shown in FIG. 6, the function h _N that first the aforementioned wavelet transform factor signals VV _N in the sub-scanning direction, performs a filtering process by a function g _N described above wavelet transform factor signals VW _N in the sub-scanning direction . That is, the wavelet transform coefficient signals VV _N and VW _N ′ by the functions g _N and h _N are subjected to a filtering process for each pixel in a row arranged in the sub-scanning direction by shifting pixel by pixel in the main scanning direction to obtain the wavelet transform coefficient signal VV. _N , VW
_An inverse wavelet transform coefficient signal WhN 'is obtained by obtaining N'inverse wavelet transform coefficient signals and adding them.

On the other hand, in parallel with this, the wavelet transform coefficient signal WV _N ′ is filtered in the sub-scanning direction by the function h _N , and the wavelet transform coefficient signal WW _N ′ is filtered in the sub-scanning direction by the function g _N. transform factor signals WV _N ', WW _N' give the inverse wavelet transform factor signals, obtaining the inverse wavelet transform factor signals WGN 'by adding this.

Next, the inverse wavelet transform coefficient signal W
The wavelet transform coefficient signal Wh is obtained by filtering the hN ′ in the main scan direction by the function h _N ′ and the inverse wavelet transform coefficient signal WgN ′ in the main scan direction by the function g _N.
N ', WgN' inverse wavelet transform coefficient signals are obtained, and by adding them, an inverse wavelet transform coefficient signal VVN _-1 'is obtained.

Then, the inverse wavelet transform coefficient signal VVN _-1 'is applied in the sub-scanning direction by the above-mentioned function hN _-1 .
The wavelet transform coefficient signal VW _N-1 is filtered in the sub _- scanning direction by the above-mentioned function g _N-1 . That is, the filtering processing for each row of pixels of the wavelet transform coefficient signals VVN _-1 , VWN _-1 'by the functions g _N-1 , h _N-1 arranged in the sub-scanning direction is shifted by one pixel in the main scanning direction. While performing the wavelet transform coefficient signal VV
The inverse wavelet transform coefficient signals of _N-1 and VW _N-1 'are obtained, and the inverse wavelet transform coefficient signal WhN-1' is obtained by adding them.

On the other hand, in parallel with this, the wavelet transform coefficient signal WVN _-1 'in the sub-scanning direction is given by the function hN _-1 , and the wavelet transform coefficient signal WWN _-1 ' is given in the sub-scanning direction by the function _gN-. performs filtering processing by _1, the wavelet transform factor signals _{WV N-1 ', WW N} -1' give the inverse wavelet transform factor signals of, obtaining the inverse wavelet transform factor signals WGN-1 'by adding this.

Next, the inverse wavelet transform coefficient signal W
The function h _N -1 ′ in the main scanning direction is used for hN-1 ′ and the function g _N-1 is used for the inverse wavelet transform coefficient signal WgN-1 ′ in the main scanning direction.
Filtering process by the wavelet transform factor signals WhN-1 ', WgN-1 ' give the inverse wavelet transform factor signals of, obtaining the inverse wavelet transform factor signal VV _N-2 'by adding this.

Thereafter, the inverse wavelet transform coefficient signal VV _i ′ (i = −1 to N) is sequentially created, and finally the inverse wavelet transform coefficient signal VV ₋₁ ′ is obtained. This final inverse wavelet transform coefficient signal VV ₋₁ ′ is the image signal Sorg.
The image signal in which the desired frequency band is emphasized.

The wavelet transform coefficient signal VV _-1 ′ thus obtained is sent to an image reproducing device (not shown) for reproduction of a radiation image.

This reproducing apparatus may be a display means such as a CRT or a recording apparatus for performing optical scanning recording on a photosensitive film, or for that purpose, an image signal is once stored in an image file such as an optical disk or a magnetic disk. It may be replaced by a storage device.

In this way, by emphasizing the signal obtained by performing the wavelet transform, it is possible to obtain a more visual impression than when emphasizing the signal obtained by the Fourier transform. You can get the image. For example, when there is an edge (shade of heart, bone, etc.) whose density changes suddenly in the radiation image, the image signal representing this radiation image is Fourier-transformed to emphasize the edge portion, and the false contour is found at the edge. Is noticeable. This is because the signal of the edge portion is forcibly matched with the trigonometric function, and the vibration is generated even in the portion of the image signal where the vibration is not generated. On the other hand, when the image signal representing the radiation image is subjected to the wavelet transform, a signal without false vibration is obtained, so that the edge portion is relatively naturally emphasized and a natural image without false contour can be obtained.

Further, since only the signal in the desired frequency band can be emphasized, the image emphasizing process can be carried out without emphasizing the high frequency component which becomes the noise component unlike the blur mask process shown in the above-mentioned formula (1). It can be performed.

[0058] In the embodiment described above, the wavelet transform factor signals to a table β _WWi, β _WVi, β
The signal in the desired frequency band of the image Sorg is multiplied by _VWi to perform the enhancement process, but the invention is not limited to this, and the wavelet transform coefficient signal WW is used.
Different blur mask processing may be applied to _i , WV _i and VW _i . For example, the image enhancement of the main subject portion can be performed by performing the blur mask processing shown in the above-mentioned formula (1) only on the signal in the frequency band expressing the main subject in the radiation image. Further, gradation processing may be performed on the wavelet transform coefficient signals WW _i , WV _i and VW _i .

Further, in the above-mentioned embodiment, the function shown in FIG. 8 is used as the basic wavelet function, but the function is not limited to this, and any function suitable for wavelet transform may be used. You may use.

Further, in the above-mentioned embodiment, the example in which the orthogonal wavelet transform is used as the wavelet transform has been shown, but the present invention is not limited to this, and the coefficient signal divided into frequency bands by the non-orthogonal wavelet transform is used. A two-dimensional fundamental wavelet may be used, or a wavelet transform may be used in which convolution is performed by enlarging, reducing, moving, or rotating the two-dimensional fundamental wavelet.

[0061]

As described above in detail, in the image processing method according to the present invention, at least one frequency band of a plurality of frequency band signals obtained by subjecting an image signal to wavelet transform is used. Since image processing such as emphasis is applied to the signal, when an image signal obtained by performing inverse wavelet transform on the image signal subjected to this image processing is reproduced, without emphasizing noise,
It is possible to obtain a natural reproduced image that matches the visual impression.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic concept of an image processing method according to the present invention.

FIG. 2 is a diagram showing a radiation image reading device.

FIG. 3 is a diagram showing details of a wavelet transform.

FIG. 4 is a diagram showing a function g ₀ .

FIG. 5 is a diagram showing a function h ₀ .

FIG. 6 is a diagram showing details of inverse wavelet transform.

FIG. 7 is a diagram for explaining a Fourier transform.

FIG. 8 is a diagram showing a basic wavelet function used for wavelet transform.

FIG. 9 is a diagram for explaining a wavelet transform.

[Explanation of symbols]

 14 stimulable phosphor sheet 23 laser light source 24 laser light 29 stimulated emission light 40 image processing device 44 keyboard

Claims (2)

[Claims] 1. An image processing method for subjecting a visible image of a radiation image to an image signal representing the radiation image, wherein the image signal is subjected to image processing by wavelet transforming the image signal. The image signal is decomposed into signals of a plurality of frequency bands, predetermined image processing is performed on the signal of at least one frequency band of the plurality of frequency bands, and the image processed signal and other signals are inversed. An image processing method characterized by applying a wavelet transform. 2. The image processing method according to claim 1, wherein the predetermined image processing is processing for multiplying a signal in the at least one frequency band by a predetermined number.