JP2001223901A - Image storage method and device, and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image storage device that can output a high resolution image including a grid image or an image with a standard resolution not causing moire at a high-speed depending on the purpose of application. SOLUTION: A wavelet conversion section 32 uses a low pas filter having a grid suppression function to apply wavelet conversion to a high resolution (HQ) image in the case of filtering processing for at least a first stage to obtain a wavelet conversion coefficient signal. The coding means 34 encodes the wavelet conversion coefficient signal to obtain coded data DA. By decoding the coded data DA to have a resolution level of 1, almost no moire stripe due to a still grid 4 appeasers in a signal LL1 carrying an image with a standard resolution (SQ). A coding means 31 encodes the HQ image to obtain coded data DB. A file server 62 stores both the coded data DA, DB in cross-reference with each other.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image storing method and apparatus, and more particularly, to storing an image including a periodic pattern such as an image obtained by performing photographing using a stationary grid. And a recording medium.

[0002]

2. Description of the Related Art Heretofore, when radiation such as X-rays is irradiated, a part of this radiation energy is accumulated, and then when stimulated light such as visible light is irradiated, photostimulated light having a light quantity corresponding to the accumulated energy is emitted. A radiation image of a subject such as a human body is once photographed and recorded on a sheet-like stimulable phosphor using a stimulable phosphor (stimulable phosphor) that emits light, and the stimulable phosphor sheet is excited by laser light or the like. Scanning with light to generate stimulated emission light, the obtained stimulated emission light is photoelectrically read by a reading unit such as a photomultiplier to obtain an image signal,
A radiation recording / reproducing system for outputting a radiation image of a subject as a visible image on a recording material such as a photographic material or a CRT based on the image signal has been proposed (Japanese Patent Laid-Open Publication No. Sho.
55-12429, 56-11395, 55-163472, 56-164
No. 645, No. 55-116340).

When a radiation image of a subject is captured and recorded on the above-described stimulable phosphor sheet or the like, radiation is not transmitted at a fine pitch of about 4 lines / mm so that radiation scattered by the subject is not irradiated onto the sheet. In some cases, a static grid in which lead or the like and aluminum or wood that is easily permeable is alternately arranged between a subject and a sheet to perform photographing. When imaging is performed using this stationary grid, the radiation scattered by the subject is less likely to be irradiated on the sheet, and therefore the contrast of the radiation image of the subject can be improved. A patterned grid image is recorded.

For this reason, the image data is Fourier-transformed,
After removing the frequency data corresponding to the grid stripes, an inverse Fourier transform is performed, or a filtering process is performed to remove the spatial frequency components of the grid stripes, thereby obtaining an image that is easy to observe with reduced stripes. A grid suppression method has been proposed (JP-A-3-12785).
No. 3-114039). In this grid suppression method, for example, when the grid pitch is 4 lines / mm, a stripe pattern appears in a spatial frequency band around 4.0 cycles / mm, and thus a filter for removing or reducing the response in this frequency band. Filtering by
Stripes are removed.

On the other hand, in recent years, in the field of image processing, multi-resolution conversion such as wavelet conversion and Laplacian pyramid expansion has been proposed as a new method for converting image resolution (pixel density), and a converted signal obtained by the multi-resolution conversion has been proposed. Encoding and compression for image transmission,
Alternatively, a technique has been proposed in which the information is recorded and stored in a desired medium and provided. Using the multi-resolution conversion,
A relatively high-resolution image (HQ image read at high density)
High Quality) and a lower resolution (eg, 1/2)
Resolution) standard resolution image (SQ image; Standard Qua
lity) can be managed efficiently. Also, for example, depending on the purpose of use, an arbitrary resolution level such as a configuration in which an HQ image is obtained when decoding is performed to the highest level of resolution and an SQ image is obtained when decoding is performed to a half resolution level is possible. , That is, the image can be arbitrarily enlarged or reduced.

However, when pixel density conversion is performed on an image including a periodic pattern such as a static grid image (hereinafter referred to as a periodic pattern), the spatial frequency corresponding to the periodic pattern is sufficiently high, and the image having the highest resolution level has Even when the periodic pattern is not so noticeable, aliasing repeatedly occurs as the image becomes lower in resolution level, and moire (striped pattern), which is a spatial frequency component lower than the spatial frequency corresponding to the periodic pattern, is extremely low. Stand out,
There is a problem that the image becomes difficult to see.

[0007]

On the other hand, as a mode of reading an image at a high density and storing the image, conventionally,
A method of storing an HQ image with a high resolution level read at a high density and an image with a lower resolution level than the HQ image instead of storing the HQ image (in order to reduce storage cost and improve data handling performance) Image with small number of pixels and small file size; eg SQ
Image) is used.

Here, when a method of storing an HQ image acquired using a still grid is used, an image stored in a file server or the like is read out at a transfer destination or the like, and a high-density image having a high resolution level is again read. Can be printed on a film or the like, but in the subsequent handling of images, low-resolution images (for example, SQ images) with reduced resolution are used in order to reduce the amount of data and improve handling. Done. in this case,
As described above, when the resolution is lowered by converting the pixel density, moiré becomes conspicuous, so that moiré removal processing is required each time, and an image cannot be reproduced and output at high speed.

On the other hand, when a method of saving a low-resolution image (for example, an SQ image) to which moiré removal processing has been performed in advance is used, an SQ image with less noticeable moiré at a transfer destination can be reproduced and output at high speed. However, even if the pixel density (number of pixels) can be made the same as that of the HQ image by pixel interpolation or the like,
Since the reproduction frequency range of the obtained image cannot be made the same as that of the original HQ image, it is not possible to output an image with high sharpness to a film.

Therefore, when handling one digital image (for example, an HQ image) having a periodic pattern such as moire caused by a stationary grid, an image obtained by compressing the HQ image using an encoding technique is used. When storing and decoding the image to the original resolution level when necessary to display the HQ image, or when lowering the resolution of the image to display the SQ image or a lower resolution image, the resolution level is lower than the HQ image. An image from which the periodic pattern such as moiré is removed is prepared in advance so that a periodic pattern such as moiré does not occur, and an image in which the periodic pattern such as a grid image (including moiré) is not conspicuous at any resolution level can be obtained at high speed. It is desirable to be able to reproduce and output.

SUMMARY OF THE INVENTION The present invention has been made in view of such a demand, and when saving an image including a periodic pattern such as an image acquired using a still grid, the image according to the purpose is used at high speed (for example, display). Output) and arbitrarily enlarge / enlarge the image by pixel density conversion for the stored image.
Image storage method and apparatus capable of obtaining a reproduced image in which a periodic pattern is inconspicuous even when reduced (resolution level is changed), and a computer-readable recording medium storing a program for causing a computer to execute the image storage method The purpose is to provide.

[0012]

According to the image storing method of the present invention, an original image including a periodic pattern image is subjected to a periodic pattern suppressing process for suppressing a spatial frequency component corresponding to the periodic pattern image, thereby suppressing the suppressed image. And storing the suppressed image data carrying the suppressed image in association with the original image data carrying the original image.

Here, the “periodic pattern image” is an image having a periodic pattern, and is included in, for example, a radiation image obtained by photographing using a stationary grid.
An image (grid image) representing the stationary grid.

"Suppressing the spatial frequency component corresponding to the periodic pattern image" means not only suppressing the spatial frequency component of the periodic pattern image itself, but also sampling at a sampling period equal to or less than the Nyquist frequency or reducing the sampling frequency. By processing, it is meant to include suppressing a moiré component generated due to the periodic pattern image.

"Saving the suppressed image data carrying the suppressed image in association with the original image data carrying the original image" means that the original image including the periodic pattern image and the original image including the periodic pattern image are stored in accordance with the purpose of use by the user. This means that any one of the suppressed images subjected to the periodic pattern suppression processing is stored in a predetermined storage device so that it can be immediately used (for example, reproduced and output).

In the image storing method according to the present invention, a low-resolution suppressed image having a lower resolution than that of the original image and in which the spatial frequency component corresponding to the periodic pattern image is suppressed is obtained. It is desirable that low-resolution image data to be carried be stored instead of suppressed image data.

Here, to obtain a low-resolution suppressed image having a lower resolution than the resolution of the original image and in which the spatial frequency component corresponding to the periodic pattern image is suppressed, the periodic pattern suppressing process is performed to perform the suppressed image. After obtaining the low-resolution image, the low-resolution processing may be performed on the suppressed image to obtain a low-resolution suppressed image, or the low-resolution processing may be performed on the original image to obtain a low-resolution image. After that, the low resolution image may be subjected to the periodic pattern suppression processing to obtain a low resolution suppressed image. Alternatively, a method of simultaneously performing the periodic pattern suppression process and the resolution reduction process on the original image may be used (see below).

As a resolution reduction process for obtaining an image having a resolution lower than the resolution of the original image (a low-resolution image or a low-resolution suppressed image), for example, a method of performing a pixel density conversion process can be used.

In the image storing method according to the present invention, a difference image is obtained by taking a difference between the original image and the suppressed image, and the difference image data carrying the difference image is stored instead of the original image data. Is also good. When a low-resolution suppressed image is obtained as described above, the low-resolution suppressed image is treated as a “suppressed image”, and the difference between the low-resolution suppressed image and the original image is calculated.

Further, in the image storing method of the present invention,
In order to reduce the data amount of the image information to be stored, the original image data carrying the original image is encoded to obtain the original image encoded data, and the image data carrying the suppressed image is encoded to suppress the encoding. It is desirable that the data be obtained and the above-mentioned storage be performed by using the suppressed encoded data instead of the suppressed image data and using the original image encoded data instead of the original image data. When a low-resolution suppressed image is obtained as described above, the low-resolution suppressed image is handled as a “suppressed image”. Further, when the difference image is to be obtained as described above, it goes without saying that this difference image is encoded instead of the original image. In this case, the original image encoded data is converted into the difference image encoded data. Think about it.

When encoding two images (an original image or a difference image and a suppressed image), the original image encoded data (or the difference image It is preferable that at least one of the encoding for obtaining the encoded data) and the encoding for obtaining the suppressed encoded data is irreversible compression encoding.

When encoding two images (same as above), it is preferable to store each encoded data in one bit stream.

It is to be noted that only one of the original image (or the difference image) including the periodic pattern image and the suppressed image is encoded, and encoded data carrying this one image and the other unencoded image are encoded. Although a method of storing the data carrying the images in association with each other is also conceivable, from the viewpoint of reducing the amount of information to be stored, it is needless to say that it is preferable to encode both images as described above. .

In the image storing method according to the present invention,
The periodic pattern suppressing process is to convert the original image into a multi-resolution space by repeatedly performing a filtering process using a predetermined filter on the original image, and at least a periodic filter is used as a predetermined filter in the first-stage filtering process. It is preferable to use one having a function of removing a pattern image. This method is an aspect of the above-described method of simultaneously performing the periodic pattern suppression processing and the resolution reduction processing on the original image. Here, as a method of converting into a multi-resolution space, for example, there is a method using wavelet transform or Laplacian pyramid expansion.

The present invention is not limited to this. For example, after performing a periodic pattern suppressing process using a moiré removing filter having predetermined characteristics, and performing a pixel density converting process using a known method, the resolution level can be reduced. A different image that does not include the periodic pattern image may be obtained.

The image storage device according to the present invention performs a periodic pattern suppressing process for performing a periodic pattern suppressing process for suppressing a spatial frequency component corresponding to a periodic pattern image on an original image including the periodic pattern image to obtain a suppressed image. Means for storing the suppressed image data carrying the suppressed image in association with the original image data carrying the original image.

In the image storage device according to the present invention, the low-resolution suppressed image obtaining means for obtaining a low-resolution suppressed image having a lower resolution than the resolution of the original image and in which the spatial frequency component corresponding to the periodic pattern image is suppressed. It is desirable that the low-resolution image data carrying the low-resolution suppressed image be stored instead of the suppressed image data. It should be noted that the low-resolution suppressed image obtaining means may obtain a low-resolution suppressed image by performing a pixel density conversion process.

The image storage device according to the present invention further comprises subtraction means for obtaining a difference image by taking the difference between the original image and the suppressed image, and the difference image data carrying the difference image is converted to the original image. It is desirable to save it instead of data.

Further, in the image storage device according to the present invention, the first encoding means for encoding the original image data carrying the original image to obtain the encoded original image data, and the image data carrying the suppressed image, Second encoding means for encoding to obtain suppressed encoded data, wherein the suppressed encoded data is used instead of the suppressed image data, and the original image code is substituted for the original image data. It is preferable that the above-mentioned storage be performed using the conversion data.

Further, in the image storage device according to the present invention, it is preferable that at least one of the first and second encoding means performs irreversible compression encoding.

Further, in the image storage device according to the present invention, it is preferable that each encoded data is stored in a single bit stream.

Further, in the image storage device according to the present invention, the periodic pattern suppressing processing means converts the original image into a multi-resolution space by repeatedly performing a filtering process using a predetermined filter on the original image. In addition, it is preferable that a filter having a function of removing a periodic pattern image is used as the predetermined filter in at least the first-stage filtering process.

Note that a program for causing a computer to execute the image storage method according to the present invention may be provided by being recorded on a computer-readable recording medium.

[0034]

According to the image storing method and apparatus of the present invention, the original image including the periodic pattern image is subjected to the periodic pattern suppressing process to obtain a suppressed image, and the suppressed image carrying the suppressed image is provided. The data is saved in association with the original image data carrying the original image, so that the image can be used at high speed, such as outputting the image according to the purpose in a short time, and the saved image can be saved. By arbitrarily enlarging or reducing the suppressed image carried by the processed image data by pixel density conversion, it is possible to obtain an enlarged image or a reduced image in which the periodic pattern is inconspicuous.

If a low-resolution suppressed image is obtained by using a method such as a pixel density conversion process, and low-resolution image data carrying the low-resolution suppressed image is stored, the amount of stored information is reduced. be able to.

Further, if a difference image is obtained by taking the difference between the original image and the suppressed image, and the difference image data carrying the difference image is stored instead of the original image data, the amount of stored information can be increased. Can be reduced.

Further, the amount of stored information can be further reduced by encoding and storing the original image or the difference image and the suppressed image, and the amount of stored information can be further reduced by using irreversible compression encoding for at least one of the images. Can be further reduced.

Further, if a multi-resolution conversion is used as the periodic pattern suppression processing and a filter having a function of removing the periodic pattern image is used in at least the first-stage filtering processing, enlargement processing and reduction processing can be performed. This is convenient, and is also convenient for transferring images via a network, for example.

Further, by forming each encoded data into one bit stream, a plurality of images can be managed as one data file, and database management becomes easy.

[0040]

Embodiments of the present invention will be described below in detail with reference to the drawings. In the following description, as described in JP-A-55-12429 and JP-A-56-11395, a radiation image information recording / reproducing system using a stimulable phosphor sheet as a recording sheet is described. In the following, an example will be described in which a radiation image of a human body recorded on the sheet is read as a digital image signal by laser beam scanning. The grid pitch is 3.4 lines /
In both the case of mm and the case of 4 lines / mm, a description will be given on the assumption that a stripe pattern caused by a static grid is not generated at least at a resolution level lower than the SQ image.

FIG. 1 is a diagram schematically showing a radiation image photographing apparatus. As shown in FIG. 1, radiation 2 emitted from a radiation source 1 passes through a subject 3 and further reaches a grid 4. The grid 4 is made of lead 4a that absorbs radiation 2.
And the aluminum 4b that transmits the radiation 2 is slightly inclined depending on the position so that the radiation 2 emitted from the radiation source 1 enters the sheet 11 straight through the aluminum 4b (see FIG. 1). ), Are alternately arranged at a pitch of 3.4 lines / mm. For this reason, the radiation 2 emitted from the radiation source 1 and transmitted straight through the subject 3 is absorbed by the lead 4a and blocked, while passing through the aluminum 4b and irradiating the sheet 11, and the sheet 11 is placed on the sheet 11 together with the subject image. A grid image having a stripe pattern of book / mm is stored and recorded. On the other hand, since the scattered radiation 2a scattered in the subject 3 is incident obliquely to the inclination of the grid 4, the incident radiation on the aluminum 4b is also absorbed by the lead 4a inside the grid 4 or the surface of the grid 4 Therefore, the sheet 11 is not irradiated, and therefore, a clear radiation image with little irradiation of the scattered radiation 2a is accumulated and recorded on the sheet 11.

FIG. 2 is a diagram showing a subject image (hatched portion in the figure) 5 and a striped grid image 6 which are stored and recorded on the sheet 11 by performing photographing using the grid 4. In this manner, a radiation image on which the subject image 5 and the grid image 6 are superimposed is recorded on the sheet 11.

FIG. 3 is a perspective view of a radiation image reading apparatus for acquiring an image including a static grid image.

The sheet 11 on which the radiation image set at the predetermined position of the reading section 10 is recorded is scanned by an endless belt or other sheet conveying means 15 driven by a driving means (not shown) at a scanning pitch of 10 lines / mm. The sheet is conveyed (sub-scan) in the Y direction. On the other hand, a light beam 17 emitted from a laser light source 16 is reflected and deflected by a rotating polygon mirror 18 driven by a motor 24 and rotating at high speed in the direction of the arrow.
After passing through a converging lens 19 such as an fθ lens, the optical path is changed by a mirror 20 to be incident on the sheet 11, and main scanning is performed in an arrow X direction substantially perpendicular to the sub-scanning direction (arrow Y direction).
From the portion of the sheet 11 irradiated with the light beam 17, the stimulated emission light 21 in a quantity corresponding to the radiation image information stored and recorded is diverged, and the stimulated emission light 21 is
2 is incident on the light incident end face 22a of the light guide 2 and travels through the inside of the guide 22 by repeating total reflection. The light is emitted from the light emitting end face 22b and received by the photomultiplier (photomultiplier tube) 23 to represent a radiation image. The photostimulable emission light 21 is
3 is photoelectrically detected and converted into an electric signal Sa. The image signal S has a highest spatial frequency (Nyquist frequency described later) fn = 5.0 cycle / mm in a desired range of spatial frequency bands required for reproducing and outputting a good radiation visible image. It also includes low spatial frequency band information of 3.4 cycles / mm, which is information on the grid image 6 shown in FIG. The information on the grid image 6 is one of the causes of making it difficult to see the visible image when observing the visible image, and is information to be removed.

The analog output signal Sa is logarithmically amplified by the log amplifier 26, sampled by the A / D converter 28 at a sampling interval corresponding to the spatial frequency fs = 10.0 cycle / mm, digitized, and read density. , A digital image signal S1 carrying a high-density image (HQ image at the highest resolution level) is obtained.

As shown in FIG. 4, the image signal S is obtained by moving the sheet 11 in the sub-scanning direction (vertical direction) while scanning the sheet 11 with a laser beam in the main scanning direction (horizontal direction). 11 represents two-dimensionally scanned image information. It should be noted that the image signal S1 obtained in this manner carries information of the Nyquist frequency fn or less, and therefore also includes the information (3.4 cycle / mm) of the grid image 6 shown in FIG. In this embodiment, the information of the grid image 6 (3.4 cycle / mm)
Since digitization is performed at a sampling interval corresponding to twice or more the spatial frequency fs, moire of the grid image 6 due to aliasing does not occur. The image signal S1 is temporarily stored in the storage unit 29, and then input to the image signal processing device 30, where a predetermined process is performed.

FIG. 5 is a schematic block diagram showing an image processing device 30 of the first embodiment including an image storage device according to the present invention, and FIG. 6 is a schematic diagram showing an image processing system using an image storing method according to the present invention. It is a block diagram. As shown in FIG. 6, the image processing system includes image processing apparatuses 30 and 8.
0, a file server 62, a network 63, a CRT monitor 71, and a film output device 72.

As shown in FIG. 5, the image processing device 30
An image signal S1 representing a high-density image (HQ image) obtained by high-density reading in the radiation image reading apparatus shown in FIG. 3 is read from the storage unit 29, and is converted into an image signal S1 represented in the read time domain. On the other hand, encoding means 31 for encoding according to a predetermined encoding rule and multi-resolution decomposition processing using wavelet transformation processing as a method of pixel density conversion are applied to image signal S1 to be handled in the frequency domain. Transform unit (multi-resolution decomposition processing means) for obtaining a band-limited image signal that can be processed
32, and among the band-limited image signals decomposed into a plurality of frequency bands in the wavelet transform unit 32, a signal up to at least a 1/2 resolution level, which is a resolution level one step lower than the resolution of the HQ image, is quantized. Encoding means 34 for encoding according to a predetermined encoding rule
And an image processing unit 35 for performing a desired image processing on the HQ image or the image with the reduced resolution level to obtain a processed image signal S2, and using the processed image signal S2.
An output format forming unit 36 for reproducing and outputting an image in conformity with a desired output format, and a restoration unit 40 for restoring the image stored in the file server 62 are provided.

The encoding rules in the encoding means 31 and 34 may be different from each other. Here, the encoding means 31 uses JPEG (Joint Photographic).
Experts Group) system or a lossless encoding system such as an entropy encoding system widely used in JPEG / LS, etc., and the encoding means 34 increases the compression rate (for example, about 1/5 to 1/20). , An irreversible encoding method employing various known techniques.

The decoding section 41 corresponding to the encoding section 31 and the decoding section 34 corresponding to the encoding section 34 are provided in the restoration section 40 so that the image stored in the file server 62 can be read out, reproduced and output. The wavelet transform unit 3 connected to the decoding unit 42 and the decoding unit 42
2 is provided.

The image processing device 30 and the network 6
An image processing device 80 and a restoration unit 82 having the same configurations as the image processing unit 35 and the restoration unit 40 provided in the image processing device 30 are provided in the image processing device 80 connected via the third unit 3. Although not shown in FIG. 6, it goes without saying that a CRT monitor and a film output device are connected to the image processing device 80.

In the image processing system having the above configuration, the image signal S1 representing the HQ image read from the storage unit 29
Is input to the image processing device 30, the image processing means 35
Is subjected to a predetermined image processing and processed image signal S2
And the processed image signal S2 is input to the output format means 36. As a result, the HQ image is displayed on the CRT
The data is reproduced and output on the monitor 71 or output to the film by the film output device 72 and used for diagnosis.

Next, an image storage method according to the present invention will be described. FIG. 7 is a functional block diagram for explaining a wavelet transform process as a multi-resolution decomposition process performed in the wavelet transform unit 32, and FIG. 8 is a block diagram showing details of each wavelet transform unit 32a. As shown in FIG. 7, the wavelet transform unit 32 is provided with wavelet transform means 32a for the number of stages corresponding to the resolution (pixel density) level. It should be noted that the inverse wavelet transform unit 44 described later has at least the wavelet transform means 32a so that the SQ image can be restored.
The inverse wavelet transform means 34a is provided for one stage less than the total number of stages. In the present embodiment, it is assumed that two-dimensional orthogonal wavelet transform in which each coefficient of the wavelet transform is orthogonal.

As shown in FIG. 7 and FIG.
When the image signal S1 representing the HQ image read out from the image signal is input to the wavelet transform unit 32, the image signal S1 is used as an original image signal Sorg, and the original image signal Sorg is subjected to wavelet transform. That is, the original image signal S
org (equivalent to the signal LL0 representing the HQ image) in the main scanning direction is subjected to filtering processing by the wavelet functions H1 and G1, and pixels in the main scanning direction are thinned out every other pixel (represented by ↓ 2 in the figure), and the main scanning is performed. The number of pixels in the direction is 1 /
Make 2 Here, the function H1 is a high-pass filter, and the function G1 is a low-pass filter. Further, filtering processing is performed on each of the signals obtained by thinning out the pixels in the sub-scanning direction using the functions H1 and G1, and the pixels in the sub-scanning direction are thinned out every other pixel to reduce the number of pixels in the sub-scanning direction. The wavelet transform coefficient signals (hereinafter, also simply referred to as signals) HH1, HL1, LH
1, LL1 is obtained. Here, the signal LL1 represents a ４ reduced image obtained by reducing the vertical and horizontal dimensions of the original image to 各 々, respectively. In each １ ／ reduced image of the HQ image, the signal LH1 is a high-frequency component in the sub-scanning direction (vertical direction). The signal HL1 is an image representing a high-frequency component (vertical edge) in the main scanning direction (horizontal direction), the signal HL1 is an image representing a high-frequency component (diagonal edge) in the diagonal direction, and the signal LL1 is HQ. SQ which is a low-frequency component image of 1/2 resolution with respect to the image
It will represent an image. Frequency that is the reference for band division,
That is, the frequency at the boundary between the low-frequency component and each high-frequency component is determined by the filter characteristics of the functions H1 and G1, and in the first stage of the filtering process using the functions H1 and G1, the function G1 as a low-pass filter is Corresponding to the grid pitch of the stationary grid 4, the transfer characteristic (response) of a spatial frequency of 3.4 cycles / mm or more has a characteristic of being substantially zero, and the function H1 as a high-pass filter is obtained by reducing the low-pass characteristic of the function G1. It has high-pass characteristics to compensate. That is, the filter used at the first stage of the wavelet transform has a function of removing a moire component as a periodic pattern image (moire removal function).
As a low-pass filter having such a moiré removing function, for example, a filter having the same characteristics as the moiré removing filter described in Japanese Patent Application No. 10-164737 filed by the present applicant, that is, a spatial frequency corresponding to the grid pitch of the stationary grid 4 A component that reduces the response of the spatial frequency component of 97% or more to the component to 5% or less is used. For example, (17, 7) taps shown in Table 1 and (1) shown in Table 2
It can be realized by a wavelet transform filter using filter coefficients such as (3, 7) taps and (15, 5) taps shown in Table 3. FIG. 9 shows the frequency response characteristics of the low-pass filter in each wavelet transform filter.
Shown in As shown in FIG. 9, both filters G1 are:
The response at a spatial frequency of 3.3 cycles / mm or more is reduced to 5% or less.

[0055]

[Table 1]

[Table 2]

[Table 3]

Thus, a grid pitch of 3.4 lines / mm or less (reading density per mm is larger than 3.4 lines / mm)
, The grid component appears in the signal LH1 when the horizontal grid is used, the grid component appears in the signal HL1 when the vertical grid is used, and the signal HH when the cross grid is used.
1, the response at a spatial frequency of 3.3 cycles / mm or more is reduced to 5% or less in the signal LL1 representing the SQ image regardless of the grid direction, and a stripe pattern (moire) caused by the stationary grid 4 is removed. Hardly appear.

Further, in the wavelet transform means 32a, using the basic wavelet functions H0 and G0,
The signal LL1 is subjected to wavelet transform to obtain signals HH2, HL2, LH2, and LL2. Here, the signal LL2 represents a 1/16 reduced image obtained by reducing the length and width of the original image to １ ／, respectively. In each 1/16 reduced image of the original image, the signals HL2, LH2, and HH2 are
In the same manner as described above, the image represents a vertical edge, a horizontal edge, or an oblique edge component image. Note that, as described above, S
Since the grid component hardly appears in the signal LL1 representing the Q image, the wavelet function H used in the second stage is used.
Unlike the wavelet functions H1 and G1 used in the first stage, 0 and G0 do not need to be set corresponding to the grid pitch of the stationary grid 4. For example, Daubechies shown in Table 4 is used. (9,7)
A wavelet transform filter using tap filter coefficients may be used. FIG. 9 shows the frequency response characteristics of the low-pass filter G0 in the (9, 7) tap wavelet transform filter.

[0058]

[Table 4]

Thereafter, in the same manner as in the second stage, the wavelet transform coefficient signal LL obtained in each frequency band
By repeating the wavelet transform for k n times, the wavelet transform coefficient signals HH1 to HHn, HL1
To HLn, LH1 to LHn and LL1 to LLn. Here, wavelet transform coefficient signals HHn, HLn, LHn, LL obtained by the n-th wavelet transform
n represents a (1) ²ⁿ reduced image in which the number of pixels in each of the main and sub directions is (1) ⁿ compared to the original image signal Sorg, and each wavelet transform coefficient signal HHn, HLn, LHn,
LLn has a lower frequency band as n increases. Thus, the wavelet transform coefficient signals HHk, HLk, L
Hk, LLk (k is a resolution level, k = 1 to n
) Are band-limited image signals carrying a predetermined range of frequency components in the frequency range of the original image signal Sorg. The signal HHk represents a change in the frequency of the original image signal Sorg in both the main and sub directions, and the larger the k is, the lower the frequency of the signal becomes. The signal HLk represents a change in the frequency of the original image signal Sorg in the main scanning direction. Further, the signal LHk is the original image signal Sor
This represents a change in the frequency of g in the sub-scanning direction. The larger the value of k, the lower the frequency of the signal.

FIG. 10 is a diagram showing a result of decomposing an original image signal into components by performing a wavelet transform. Note that FIG. 10 shows a state in which the first two-dimensional wavelet transform is performed (FIG. 10A), and then a state in which the second two-dimensional wavelet transform is performed (FIG. 10B).

As can be seen from the description of the operation of the wavelet transform unit 32, the wavelet transform unit 32a at the first stage in the wavelet transform unit 32 includes the periodic pattern suppression processing unit, the low-resolution suppressed image acquisition unit, and the low-resolution suppressed image acquisition unit according to the present invention. It functions as a pixel density conversion unit. By inputting a signal LLk representing an image of each resolution level below the SQ image to the image processing means 35, (1/1 /
2) ^k resolution, no moiré (1/2) ^2k
The reduced image can be reproduced and output on the CRT monitor 71 or the like.

In this manner, the wavelet transform coefficient signals HH1 to HHn, HL1 to HL of a number of resolution levels
After n, LH1 to LHn and LL1 to LLn are obtained,
1 step lower than the resolution of the HQ image of each signal
/ 2 resolution level signal, that is, a signal that can be restored to an SQ image, is input to the encoding means 34,
Encoded data D that has been subjected to quantization and lossy encoding
Converted to A. As is clear from the above description, the encoded data DA is data that hardly contains moire components. On the other hand, the image signal S representing the HQ image read from the storage unit 29 is also input to the encoding unit 31, where the image signal S is subjected to a lossless encoding process and converted into encoded data DB. FIG. 11 schematically shows the above-described processing process by focusing on an image.

Then, each of the converted encoded data DA,
The DBs are stored in the file server 62 in association with each other. At this time, each encoded data DA,
The DB is stored (stored) in a single bit stream (file). Also, if the encoded data DA is SQ
Appropriate identification data is added as attached data (for example, header data) of each encoded data so that the encoded data DB indicates an HQ image. If the user desires a resolution image at the original image level, that is, an HQ image (especially for the purpose of image interpretation), the coded data DB is decoded by the restoration unit 40,
Alternatively, decoding is performed after image transfer (data transfer) to another device (the image processing device 80 in this example) via the network 63. On the other hand, if the resolution is not more than the 1/2 reduced image level (particularly for reference purposes), the encoded data DA representing the SQ image is decoded.

FIG. 12 shows a bit stream (file).
FIG. 3 is a diagram showing an example of the data structure of FIG. After obtaining the SQ image of 1/2 resolution based on the HQ image and obtaining the encoded data DB of the HQ image and the encoded data DA of the SQ image, the encoded data DA of the SQ image becomes the leading side and the HQ image The encoded data DB is combined into one encoded data so that the encoded data DB is located on the rear side. Next, header information is added to the head of the collected encoded data. The header information includes information about each of the HQ image and the SQ image, for example, the number of pixels, the number of lines, the number of bits, the amount of encoded data, the Huffman table, and the like. In addition, each encoded data DA, DB
Before combining them into one, header information may be added to each piece of coded data of each image, and then combined into one piece of data. Each encoded data DA and DB are separately stored (stored)
When the transfer is performed, two files must be managed, and the database management becomes complicated. In this way, each of the encoded data DA and DB is stored in the same bit stream (file).
By doing so, the two images can be conveniently managed as one data file.

When the image is transferred after storing the image after being quantized and coded, when restoring the image, for example, MPEG-4 (Moving Picture Experts Groove) is used.
SNR scalability or spatial scalability used in up-4) may be used. Where SN
R scalability is a method of hierarchically quantizing a wavelet transform coefficient signal. As shown in the conceptual diagram of FIG. 13A, the wavelet transform coefficient signal is first quantized by a coarse quantization step. The encoding is performed, and the quantization error of the wavelet transform coefficient signal is sequentially quantized in a dense quantization step and encoded. Then, the image processing apparatus at the transfer destination can reproduce an image containing a little distortion (noise) by decoding only the first part of the received coded data, and can successively reduce the quantization error. The S / N ratio of the image can be gradually increased by decoding the encoded data that has been quantized. On the other hand, the spatial scalability is a method of sequentially and stepwise quantizing the wavelet transform coefficient signal having the lowest resolution level (the lowest frequency component). As shown in the conceptual diagram of FIG. , Quantizes and encodes the wavelet transform coefficient signal of the low frequency component, and sequentially encodes higher frequency components. Then, the image processing device at the transfer destination can reproduce the image of the low-frequency component by decoding only the first part of the received encoded data, and performs encoding corresponding to the sequentially received high-frequency component. By decoding the data, the spatial resolution can be gradually increased.

Next, a method for restoring an original image based on the encoded data stored in the file server 62 will be described.

The coded data DA and DB are read from the file server 62 into the restoration unit 40, and the coded data DA is read by the decoding unit 4 by referring to the attached data for identification.
2, the encoded data DB is input to the decoding means 41. Then, in each of the decoding units 41 and 42, a decoding process and an inverse quantization process corresponding to the encoding process are performed, and the substantially original signal is decoded.

Thereafter, the encoded data DA representing the SQ image
Are decoded by the decoding means 42, and the wavelet transform coefficient signals LLn, HLk, LHk, HHk (k =
1 to n), the inverse wavelet transform unit 44 sequentially performs inverse wavelet transform from the lowest resolution level n to level 1 (the resolution level of the SQ image).

FIG. 14 is a schematic block diagram showing the configuration of the inverse wavelet transform unit 44, and FIG. 15 is a functional block diagram for explaining the inverse wavelet transform process performed by each inverse wavelet transform unit 34a. FIG.
As shown in FIG. 4, first, signals HHn,
HLn, LHn, and LLn are subjected to inverse wavelet transform by inverse wavelet transform means 44a to obtain signal LLn-1.

In the case of the inverse wavelet transform, FIG.
As shown in FIG. 1, first, the signal LLn and the signal LHn (L
Hk) performs a process of spacing one pixel between pixels in the sub-scanning direction (represented by # 2 in the figure), and an inverse wavelet transform function corresponding to the functions G0 and H0 used in the wavelet transform A filtering process is performed in the sub-scanning direction by G0 ′ and H0 ′, these are added, and a signal obtained by the addition (referred to as a first addition signal) is equivalent to one pixel between pixels in the main scanning direction. A first signal is obtained by performing a process of providing an interval and performing a filtering process in the main scanning direction using the function G0 ′. On the other hand, in the sub-scanning direction of the signal HLn (HLk) and the signal HHn (HHk), a process is performed to provide an interval of one pixel between pixels, and a filtering process is performed in the sub-scanning direction by the functions G0 ′ and H0 ′. These signals are added to each other, and a signal obtained by the addition (hereinafter referred to as a second added signal) is subjected to a process of providing an interval of one pixel between pixels in the main scanning direction, and is subjected to a filtering process using a function H0 ' In the main scanning direction to obtain a second signal. Then, the first and second signals are added to obtain a signal LLn-1 (LLk-1).

Next, the signals HHn-1, HLn-1, LH
The inverse wavelet transform unit 44a performs inverse wavelet transform on n-1 and LLn-1 in the same manner as described above to obtain a processed signal LLn-2. Then, the signal LL1 representing the SQ image is obtained by repeating the inverse wavelet transform up to the resolution level 1 in the same manner as described above, and an image substantially equivalent to the original SQ image (since the lossy encoding method is used; Here, the restored image is also referred to as an SQ image).

On the other hand, by using a signal obtained by decoding the coded data DB representing the HQ image by the decoding means 41, a complete HQ image can be restored (because the lossless coding method is used. ).

The HQ image and the SQ image thus restored are subjected to predetermined image processing in the image processing means 35 and then reproduced and output on the CRT monitor 71 or the like via the output format forming means 36. . here,
The HQ image is an image including a grid image, but as described above, the grid frequency is not so noticeable because the sampling frequency fs is sufficiently high. On the other hand, as described above, H
In the signal LL1 of the resolution level 1 representing the SQ image obtained by the wavelet transform on the Q image, the response at the spatial frequency of 3.3 cycles / mm or more is reduced to 5% or less, and the static grid 4 of 3.4 lines / mm is used. Even when photographing is performed using the stationary grid 4 of 4 lines / mm, the striped pattern caused by the stationary grid 4 hardly appears and the grid component is suppressed. Also, a signal LLk of level 1 or later obtained by performing a wavelet transform on the signal LL1 in which the grid component is suppressed is obtained.
Since no moiré component is included in the image, no moiré due to a static grid is generated in all low-resolution images (reduced images). Further, even if the image is enlarged or reduced at an arbitrary magnification using an SQ image (of course, an image of a lower resolution level), artifacts due to moiré do not occur and a high-quality image that is easy to view can be provided. . Further, since the wavelet transform is used, it is convenient for performing enlargement / reduction processing, and is also convenient for transferring an image via the network 63.

In the restoration unit 82 of the image processing apparatus 80 connected to the network, based on the transferred encoded data DA and DB, similarly to the above description, an HQ image in which a grid image is inconspicuous, A low-resolution image (for example, an SQ image) that does not generate the image can be restored. Furthermore, even if the image is enlarged or reduced at an arbitrary magnification using an SQ image (an image having a resolution level lower than the SQ image), artifacts due to moire do not occur and a high-quality image that is easy to view can be provided. .

As described above, in the image processing system having the above configuration to which the present invention is applied, the image including the grid image (HQ image in the above example) and the image in which the periodic pattern caused by the grid image is suppressed (the above example). (SQ image) and save it, so that the image according to the purpose can be used at high speed, and the stored SQ image can be arbitrarily enlarged or reduced by pixel density conversion. However, it is possible to reproduce an enlarged image or a reduced image in which the periodic pattern is inconspicuous.

Next, a second embodiment according to the present invention will be described. FIG. 16 is a schematic block diagram showing an image processing device 30 according to the second embodiment including an image storage device and an image restoration device according to the present invention, and corresponds to FIG. 5 described above. The difference from the first embodiment is that the SQ image is restored based on the encoded data DA, and then interpolated and enlarged to obtain an HQ ′ image having the same resolution level (number of pixels) as the HQ image. The difference from the image is taken, and the image (DQ image; Differential Quan) obtained by this difference
lity) encoded data DD
Is stored in the file server 62 in association with the coded data DA instead of the coded data DB representing the HQ image.

As shown in FIG. 16, the image processing apparatus 30 according to the second embodiment is connected to the decoding means 42 for restoring the SQ image, and has the same resolution level as that of the HQ image.
Interpolation magnifying means 45 for obtaining a Q 'image, HQ' image and HQ
And a subtraction unit 37 for calculating a difference from the image. The difference image DQ represented by the output signal of the subtraction unit 37 is input to the encoding unit 34 instead of the HQ image. Encoding means 3
4 and the file server 62
There is provided a switching means 47 for selectively inputting any one of the encoded data DA read out from the decoding means 42 to the decoding means 42. Further, as means for restoring the HQ image, in addition to the decoding means 41, the difference image DQ represented by the difference signal decoded by the decoding means 41 and the HQ 'image obtained by the interpolation enlarging means 45 are added. Addition means 5
4 are provided.

When storing an image in the image processing apparatus 30 according to the second embodiment, first, as in the first embodiment, encoded data DA for an SQ image is obtained. Next, the switching means 47 is set to the input terminal a side, the encoded data DA is directly input to the decoding means 42, and the decoding means 42 and the inverse wavelet transform means 44 use the SQ image (correctly, a substantially SQ image). (Equal images).

Next, the interpolation enlargement means 45 interpolates one pixel having a value of 0 between each pixel of the restored SQ image one by one, and then performs a predetermined filtering process to obtain an HQ 'image which is an enlarged image. The HQ ′ image is an image having the same resolution level (the number of pixels) as the HQ image, but is a low-quality image with lower sharpness than the HQ image. However, as described above, the HQ ′ image is an image obtained by interpolating and expanding the SQ image containing no moiré component, and is an image containing no moiré component.

Next, the subtraction means 37 obtains a difference between the HQ 'image obtained by the interpolation enlargement means 45 and the HQ image read from the storage unit 29 to obtain a difference image DQ. Then, the difference image DQ is input to the encoding unit 31 and subjected to a lossless encoding process to obtain encoded data DD.

Then, the encoded data DA and DD are stored in the file server 62 in association with each other. Therefore, the image processing device 30 according to the second embodiment can also save the encoded data that can restore the SQ image and the HQ image. Further, since the encoded data DD representing the difference image DQ is obtained instead of the encoded data DB representing the HQ image, the encoding efficiency is increased, and as a result, the total information amount of the stored image can be reduced. . FIG. 17 schematically illustrates the above-described processing process by focusing on an image.

On the other hand, when restoring the original image based on the stored and transferred encoded data, first, the switching means 4
7 is switched to the input terminal b side, and based on the encoded data DA read from the file server 62, the SQ image (correctly, approximately) is decoded by the decoding unit 42 and the inverse wavelet transform unit 44 in the same manner as in the first embodiment. (Equivalent to the SQ image). Further, after decoding the encoded data DD representing the difference image DQ by the decoding means 41 to restore the difference image DQ, the adding means 54 adds the difference image DQ to the restored SQ image, so that the complete H
Restore the Q image. The HQ image and the SQ image thus restored are subjected to predetermined image processing in the image processing means 35 and then reproduced and output on the CRT monitor 71 or the like via the output format forming means 36. Also, in the image processing device 80 connected to the network, the HQ image and the SQ image can be restored by configuring the restoration unit 82 to have the same configuration as the restoration unit 40 according to the second embodiment.

Although the preferred embodiments of the image storing method and apparatus according to the present invention have been described above, the present invention is not necessarily limited to the above-described embodiments.

For example, in the above description, the pixel density conversion is performed by the wavelet conversion using the filter having the moiré removing function. However, the present invention is not limited to this. A method of using a filter as a grid removal filter for removing a grid image.
As proposed in No. 4645, a method of performing pixel density conversion by Laplacian pyramid expansion using a moiré removal filter may be used. In addition, Japanese Patent Application No.
After suppressing a moiré component using a moiré removal filter described in JP-A-10-164737, a pixel density conversion process using a known method may be performed.

Also, In the above description, the HQ image including the grid image and the low-resolution image in which the periodic pattern caused by the grid image is suppressed (in the above example, the SQ image having a half resolution level of the HQ image) are stored in association with each other. But
The resolution level of the image in which the periodic pattern is suppressed is not necessarily H
It is not necessary that the image has a lower resolution than the Q image. For example, the image may be an image having the same resolution level (pixel density) as the HQ image, in which the grid component of the HQ image in which the grid image is inconspicuous is removed by a moire removing filter. Such an image is obtained by, for example, performing a Fourier transform on an HQ image and removing frequency data corresponding to a grid stripe pattern as proposed in Japanese Patent Application Laid-Open Nos. 3-12785 and 3-114039. , An inverse Fourier transform, or a filtering process for removing the spatial frequency components of the grid pattern. Further, as understood from the above description, the HQ ′ obtained by interpolation and enlargement of the SQ image is obtained.
It may be an image. When Laplacian pyramid expansion is used, an image obtained by performing inverse expansion to the resolution level of the HQ image may be used. In short, by storing the image including the grid component that has not been subjected to the moiré removal processing and the image that has been subjected to the moiré removal processing in association with each other, an image in which a periodic pattern such as moiré is inconspicuous at any resolution level Can be reproduced and output at high speed according to the purpose of use.

In the above description, as an example of an image including a periodic pattern image, a radiographic image including a static grid image obtained by photographing using a static grid has been described. The image is not necessarily limited to a static grid image as long as the image is composed of a pattern.

Further, in the above description, in order to reduce the amount of data stored in the file server, both an image including a grid image (HQ image) and an image in which no moiré component is generated (SQ image) are encoded. Although it was made to save, even if only one of them is encoded, the data carrying one encoded image and the data carrying the other unencoded image are stored in association with each other. Good. Further, both of the images may be stored in association with each other without being encoded.

Further, the above-described image storing method of the present invention may be executed by a computer, and a program for causing the computer to execute the method may be recorded on a computer-readable recording medium and provided.

[Brief description of the drawings]

FIG. 1 is a diagram schematically illustrating a radiation image capturing apparatus.

FIG. 2 is a diagram showing a radiographic image obtained by grid imaging;

FIG. 3 is a perspective view showing an example of a radiation image reading apparatus.

FIG. 4 is a diagram showing a relationship between a scanning direction and a read image.

FIG. 5 is a block diagram schematically illustrating an image processing apparatus according to a first embodiment including the image storage apparatus of the present invention.

FIG. 6 is a schematic block diagram showing an image processing system using an image storage method according to the present invention.

FIG. 7 is a functional block diagram for explaining a wavelet transform process performed in a wavelet transform unit;

FIG. 8 is a block diagram showing details of a wavelet transform unit;

FIG. 9 is a diagram showing a frequency response characteristic of a low-pass filter in a wavelet transform filter.

FIG. 10 is a diagram illustrating a result of decomposing an original image signal into components by performing a wavelet transform.

FIG. 11 is a diagram schematically illustrating a processing process of the image processing apparatus according to the first embodiment, focusing on an image.

FIG. 12 is a diagram illustrating an example of a data structure of a bit stream file.

FIG. 13A is a diagram illustrating SNR scalability in image transfer, and FIG. 13B is a diagram illustrating spatial scalability in image transfer.

FIG. 14 is a schematic block diagram illustrating a configuration of an inverse wavelet transform unit.

FIG. 15 is a functional block diagram for explaining an inverse wavelet transform process performed by an inverse wavelet transform unit;

FIG. 16 is a block diagram schematically illustrating an image processing apparatus according to a second embodiment including the image storage apparatus of the present invention.

FIG. 17 is a diagram schematically illustrating a processing process of the image processing apparatus according to the second embodiment, focusing on an image.

[Explanation of symbols]

REFERENCE SIGNS LIST 4 static grid 5 subject image 6 grid image 11 stimulable phosphor sheet 30 image processing device including image storage device 31 encoding means 32 wavelet transform unit 34 acting as periodic pattern suppression processing means 34 encoding means 35 image processing means 36 Output format forming unit 37 Subtraction unit 40 Restoring unit 41 Decoding unit 42 Decoding unit 44 Inverse wavelet transform unit 45 Interpolation enlarging unit 46 Switching unit 47 Adding unit 80 Image processing device 81 Image processing unit 82 Restoring unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2H013 AC06 4C093 AA01 AA28 CA13 CA21 EB05 EB24 FD01 FD02 FD04 FD05 FD20 FF09 FF13 FF34 FH02 5B057 AA08 CD05 CE06 CG02 CG05 5C078 AA04 BA53 CA23 DB03 DA02 DA03 DA02 DA02 DD09 EE02 EE04 EE05 HH24 HH27 JJ09 JJ27 KK42 KK60

Claims (24)

[Claims] 1. An original image including a periodic pattern image is subjected to a periodic pattern suppression process for suppressing a spatial frequency component corresponding to the periodic pattern image to obtain a suppressed image, and the suppression of carrying the suppressed image is performed. And storing the finished image data in association with the original image data carrying the original image. 2. A low-resolution image having a lower resolution than the resolution of the original image and in which a spatial frequency component corresponding to the periodic pattern image is suppressed, and a low-resolution image carrying the low-resolution suppressed image. 2. The image storing method according to claim 1, wherein data is stored instead of the suppressed image data. 3. The low-resolution suppressed image is obtained by performing a pixel density conversion process.
The image storage method described. 4. The method according to claim 1, wherein a difference image is obtained by taking a difference between the original image and the suppressed image, and difference image data carrying the difference image is stored instead of the original image data. Item 4. The image storing method according to any one of Items 1 to 3. 5. Encoding original image data carrying the original image to obtain encoded original image data; encoding image data carrying the suppressed image to obtain suppressed encoded data; 5. The storage according to claim 1, wherein the storage is performed by using the suppressed coded data instead of the image data, and by using the coded original image data instead of the original image data. The image storage method described. 6. The method according to claim 1, wherein at least one of the encoding for obtaining the original image encoded data and the encoding for obtaining the suppressed encoded data is lossy compression encoding. The image storage method according to claim 5. 7. The method according to claim 5, wherein each of the encoded data is stored in a single bit stream.
Or the image storage method according to 6. 8. The periodic pattern suppressing process, wherein the original image is converted into a multi-resolution space by repeatedly performing a filtering process using a predetermined filter on the original image, and at least the first stage of the filtering is performed. 8. The image storing method according to claim 1, wherein a filter having a function of removing the periodic pattern image is used as the predetermined filter in the processing. 9. A periodic pattern suppression processing means for performing, on an original image including the periodic pattern image, a periodic pattern suppression processing for suppressing a spatial frequency component corresponding to the periodic pattern image to obtain a suppressed image. An image storage device for storing suppressed image data carrying a suppressed image in association with original image data carrying said original image. 10. A low-resolution suppressed image obtaining means for obtaining a low-resolution suppressed image having a lower resolution than the resolution of the original image and in which a spatial frequency component corresponding to the periodic pattern image is suppressed. 10. The image storage device according to claim 9, wherein the low-resolution image data carrying the suppressed image is stored instead of the suppressed image data. 11. The low-resolution suppressed image acquiring unit obtains the low-resolution suppressed image by performing a pixel density conversion process.
An image storage device as described in the above. 12. A subtraction means for obtaining a difference image by taking a difference between the original image and the suppressed image, wherein the difference image data carrying the difference image is stored instead of the original image data. The image storage device according to any one of claims 9 to 11, wherein: 13. A first encoding unit that encodes original image data carrying the original image to obtain encoded original image data, and encodes image data carrying the suppressed image to suppress the encoded image. A second encoding unit that obtains data, using the suppressed encoded data in place of the suppressed image data, and using the original image encoded data instead of the original image data, 13. The image storage device according to claim 9, wherein the image storage device performs storage. 14. The image storage device according to claim 13, wherein at least one of said first and second encoding means performs lossy compression encoding. 15. The image storage device according to claim 13, wherein each of the encoded data is stored by being included in one bit stream. 16. The periodic pattern suppression processing means for converting the original image into a multi-resolution space by repeatedly performing a filtering process with a predetermined filter on the original image, and at least the first stage As the predetermined filter in the filtering process,
The image storage device according to any one of claims 9 to 15, wherein a device having a function of removing the periodic pattern image is used. 17. An original image including a periodic pattern image,
A step of performing a periodic pattern suppression process for suppressing a spatial frequency component corresponding to the periodic pattern image to obtain a suppressed image; and suppressing the suppressed image data carrying the suppressed image with original image data carrying the original image. And a procedure for storing the program in association with the program. 18. The method according to claim 18, further comprising: obtaining a low-resolution suppressed image having a resolution lower than that of the original image and in which a spatial frequency component corresponding to the periodic pattern image is suppressed. 18. The recording medium according to claim 17, wherein low-resolution image data carrying a resolution-suppressed image is stored instead of said suppressed image data. 19. The method according to claim 18, wherein the step of obtaining the low-resolution suppressed image includes obtaining the low-resolution suppressed image by performing a pixel density conversion process.
The recording medium according to the above. 20. A step of obtaining a difference image by taking a difference between the original image and the suppressed image, wherein the step of storing the difference image data includes the difference image data carrying the difference image. 20. The method according to claim 17, wherein the information is stored instead.
Recording medium according to the item. 21. A procedure for encoding original image data carrying the original image to obtain encoded original image data, and a procedure for encoding image data carrying the suppressed image to obtain encoded encoded data. The step of storing includes performing the saving by using the suppressed encoded data instead of the suppressed image data and using the original image encoded data instead of the original image data. 21. The recording medium according to claim 17, wherein the recording medium is a medium. 22. At least one of encoding for obtaining the original image encoded data and encoding for obtaining the suppressed encoded data is lossy compression encoding. The recording medium according to claim 21. 23. The recording medium according to claim 21, wherein the storing step includes storing the encoded data in one bit stream. 24. The step of obtaining the suppressed image, wherein the original image is converted to a multi-resolution space by repeatedly performing a filtering process using a predetermined filter on the original image, and at least the first stage As the predetermined filter in the filtering process,
24. The recording medium according to claim 17, wherein a recording medium having a function of removing the periodic pattern image is used.