JP4823737B2 - Periodic pattern suppression processing method and apparatus - Google Patents

Description

  The present invention relates to a periodic pattern suppression processing method and apparatus for suppressing a spatial frequency component corresponding to a periodic pattern in an image signal.

  Conventionally, a part of radiation energy is accumulated by irradiation with radiation (X-ray, α-ray, β-ray, γ-ray, electron beam, ultraviolet ray, etc.), and excitation light such as visible light or laser light is accumulated in the accumulation region. , A stimulable phosphor (stimulable phosphor) that emits stimulated emission light in accordance with the accumulated radiation energy is widely used in radiographic image readers and the like. Specifically, for example, the stimulable phosphor sheet is disposed such that the stimulable phosphor sheet in which the stimulable phosphor is laminated on the support is irradiated with radiation that has passed through a subject such as a human body. The stimulable phosphor sheet temporarily accumulates and records radiation image information formed by radiation transmitted through the subject. Then, the stimulable phosphor sheet is irradiated with excitation light to generate stimulated emission light. When this stimulated emission light is photoelectrically converted, an image signal corresponding to the radiation image information can be obtained.

  When photographing and recording a radiographic image of a subject on the above-described stimulable phosphor sheet, lead or the like that does not transmit radiation, and aluminum or wood that is easily transmitted so that the scattered phosphor sheet is not irradiated with radiation scattered by the subject. In some cases, photographing is performed by placing a stationary grid in which the etc. are alternately arranged at a pitch of about 4 lines / mm between the subject and the stimulable phosphor sheet. When photographing is performed using the stationary grid, the radiation scattered by the subject is less likely to be applied to the stimulable phosphor sheet, so that the contrast of the radiation image of the subject can be improved. However, a striped periodic pattern (hereinafter referred to as “grid image”) corresponding to a stationary grid is also recorded together with a radiographic image of the subject.

Therefore, Patent Document 1 discloses a method for removing a grid image from an original image mixed with the grid image. Specifically, a one-dimensional high-pass filter process is performed in the direction of the grid pattern stripe pattern, and a one-dimensional low-pass filter process is performed in a direction parallel to the grid image stripe pattern to obtain a spatial frequency component corresponding to the grid image. Extract from the original image and subtract the extracted spatial frequency component from the original image. Thereby, the grid image mixed in the original image can be removed.
JP 2003-150954 A

  However, the method disclosed in Patent Document 1 has the following problems. When the above-described one-dimensional high-pass filter processing and one-dimensional low-pass filter processing are performed on the original image, a grid image appears in a portion where the grid image does not originally exist. The portion where the grid image does not originally exist mainly corresponds to a portion where the grid image is crushed in the original image (that is, a high density portion of the original image). Therefore, as a result of performing the process of removing the grid image from the original image, a stripe pattern that does not originally exist is mixed into the high density portion, which is a problem.

  If the cut-off frequency of the one-dimensional low-pass filter is set low, the occurrence of a striped pattern can be suppressed, but high-frequency components are also cut off, so that the contrast of the processed image is lowered.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a periodic pattern suppression processing method and apparatus for suppressing a periodic pattern of an original image without degrading image quality.

  In order to solve the above problems, the periodic pattern suppression processing method of the present invention extracts a spatial frequency component of the periodic pattern from the original image signal of the original image including the periodic pattern by filtering, In the periodic pattern suppression processing method for suppressing the periodic pattern from the original image by subtracting a frequency component from the original image signal, the original image signal is filtered in the arrangement direction of the periodic pattern. A first step, a second step of filtering the image signal obtained by the first step in a direction parallel to the periodic pattern, and an image signal obtained by the first step. And the image obtained by the first step as the contrast difference between the image signals obtained by the second step is larger. The third step of increasing the weight of the spatial frequency component of the signal and adding it to the spatial frequency component of the image signal obtained by the second step, and the first step as the density of the original image signal is higher Including at least one of the fourth step of increasing the weight of the spatial frequency component of the image signal obtained by the above and adding it to the spatial frequency component of the image signal obtained by the second step, And / or a subtracting step of subtracting the added spatial frequency component obtained by the fourth step from the original image signal.

  Here, the “periodic pattern” is, for example, a striped grid image corresponding to the stationary grid included in the original image when radiation imaging is performed through the stationary grid, or normal imaging using a digital camera or the like. A periodic pattern or the like included in a photographed image when photographed through a screen door or a blind. Further, for example, not only a grid image by a spatial frequency component of the stationary grid itself but also a striped moire generated due to a numerical value of a sampling frequency used when reading a radiographic image, a reduction process, or the like.

  Further, the “alignment direction of the periodic pattern” indicates, for example, the alignment direction of the stripe pattern of the grid image, and the “direction parallel to the periodic pattern” indicates, for example, a direction parallel to the stripe pattern of the grid image, That is, it indicates a direction orthogonal to the arrangement direction of the striped pattern of the grid image.

  Further, “the greater the contrast difference of the image signal, the greater the weight of the spatial frequency component of the image signal obtained in the first step” is obtained in the first step in proportion to the contrast difference. The degree of weighting of the spatial frequency component of the image signal may be increased, or when the contrast difference is within a predetermined range, the degree of weighting of the spatial frequency component is constant and the contrast difference is within the predetermined range. When the value exceeds the value, the degree of weighting of the spatial frequency component may be increased in proportion to the contrast difference.

  Further, a fifth filtering process is performed on the image signal obtained by the first step in a direction parallel to the periodic pattern, using a cutoff frequency higher than the cutoff frequency in the filtering process of the second step. The third step is obtained by the fifth step as the contrast difference between the image signal obtained by the fifth step and the image signal obtained by the second step is larger. It is also possible to increase the weight of the spatial frequency component of the image signal and add it to the spatial frequency component of the image signal obtained by the second step.

  Further, the greater the contrast difference between the image signal obtained in the first step and the image signal obtained in the fifth step, the greater the weight of the spatial frequency component of the image signal obtained in the first step. A third step, wherein the third step increases the contrast difference between the image signal obtained by the fifth step and the image signal obtained by the second step. The spatial frequency component obtained by the sixth step is further added to the spatial frequency component obtained by increasing the weight of the spatial frequency component of the image signal obtained by the step, and the added spatial frequency component is added to the second step. It is good also as what is added to the spatial frequency component of the image signal obtained by.

  Further, the periodic pattern suppression processing apparatus according to the present invention extracts a spatial frequency component of the periodic pattern from the original image signal of the original image including the periodic pattern by filtering processing, and the spatial frequency component is extracted from the original image signal. In a periodic pattern suppression processing apparatus that suppresses the periodic pattern from the original image by subtracting from the first image, the first means for performing a filtering process on the original image signal in the arrangement direction of the periodic pattern, A second means for performing a filtering process on the image signal obtained by the first means in a direction parallel to the periodic pattern; an image signal obtained by the first means; and the second means As the contrast difference of the obtained image signal is larger, the weight of the spatial frequency component of the image signal obtained by the first means is increased. Third means for adding to the spatial frequency component of the image signal obtained by the second means, and the higher the density of the original image signal, the higher the weight of the spatial frequency component of the image signal obtained by the first means. And at least one of fourth means for adding to the spatial frequency component of the image signal obtained in the second means, and obtained in the third means and / or the fourth means. Subtracting means for subtracting the added spatial frequency component from the original image signal.

  Further, a fifth filter processing is performed on the image signal obtained by the first means in a direction parallel to the periodic pattern, using a cutoff frequency higher than the cutoff frequency in the filter processing of the second means. The third means is obtained in the fifth means as the contrast difference between the image signal obtained in the fifth means and the image signal obtained in the second means is larger. Alternatively, the weight of the spatial frequency component of the image signal may be increased and added to the spatial frequency component of the image signal obtained by the second means.

  Further, the greater the contrast difference between the image signal obtained by the first means and the image signal obtained by the fifth means, the greater the weight of the spatial frequency component of the image signal obtained by the first means. The third means further includes a fifth means for increasing the contrast difference between the image signal obtained by the fifth means and the image signal obtained by the second means. The spatial frequency component obtained by the sixth means is further added to the spatial frequency component obtained by increasing the weight of the spatial frequency component of the image signal obtained in step 6 and the added spatial frequency component is added to the second means. It is good also as what is added to the spatial frequency component of the image signal obtained in (1).

  The greater the contrast difference between the image signal obtained in the first step and the image signal obtained in the second step, the greater the weight of the spatial frequency component of the image signal obtained in the first step, Adding to the spatial frequency component of the image signal obtained by the step and / or increasing the weight of the spatial frequency component of the image signal obtained by the first step as the density of the original image signal is higher. By adding to the spatial frequency component of the image signal obtained by the step and subtracting the added spatial frequency component obtained by the third step and / or the fourth step from the original image signal, thereby obtaining the image signal , Suppresses the periodic pattern mixed in the original image without degrading the image quality of the original image, and further, the portion where the grid image does not originally exist (for example, If, it is also possible to suppress the occurrence of the periodic pattern in the high density portion) in the original image.

  Embodiments of the present invention will be described below with reference to the drawings. In the following embodiments, the case where the periodic pattern suppression processing apparatus according to the present invention is used in a radiographic image reading apparatus will be described, but images are taken through a screen door, blinds, etc. in normal imaging using a digital camera or the like. In some cases, the image processing apparatus may be used for suppressing a periodic pattern included in a captured image. The radiographic image reading apparatus reads a radiographic image of a human body recorded on a stimulable phosphor sheet as a digital image signal by laser beam scanning.

  FIG. 1 is a schematic diagram of a radiographic image capturing apparatus. The radiation 2 emitted from the radiation source 1 passes through the subject 3 and reaches a stationary grid (hereinafter simply referred to as “grid”) 4. In the grid 4, lead 4 a that absorbs the radiation 2 and aluminum 4 b that transmits the radiation 2 are alternately arranged at a pitch of about 4 lines / mm, for example. Further, the lead 4a is disposed with a slight change in inclination according to the position so that the radiation 2 passes through the aluminum 4b and enters the stimulable phosphor sheet 11.

  Therefore, the radiation 2 that has passed through the subject 3 is absorbed by the lead 4 a and blocked from reaching the stimulable phosphor sheet 11, while passing through the aluminum 4 b and reaching the stimulable phosphor sheet 11. On the stimulable phosphor sheet 11, a striped grid image of 4 lines / mm is recorded together with the subject image. On the other hand, the scattered radiation 2a scattered in the subject 3 is absorbed by the lead 4a installed with an inclination according to the position, or is reflected by the surface of the grid 4, and therefore reaches the stimulable phosphor sheet 11. do not do. Therefore, the stimulable phosphor sheet 11 can record a clear radiation image with little irradiation of the scattered radiation 2a. FIG. 2 is an example of a radiographic image of the subject image 5 and the grid image 6 accumulated and recorded on the stimulable phosphor sheet 11 when the radiographic image capturing apparatus shown in FIG. 1 is used.

  FIG. 3 is a combination of a perspective view and a functional block diagram of the radiation image reading apparatus. The stimulable phosphor sheet 11 set at a predetermined position of the reading unit 10 is, for example, in the arrow Y direction at a scanning pitch of 10 lines / mm by a sheet conveying unit 15 such as an endless belt driven by a driving unit (not shown). Carried (sub-scanned). On the other hand, the light beam 17 emitted from the laser light source 16 is reflected toward a focusing lens 19 such as an fθ lens by a rotating polygon mirror 18 that is driven by a motor 24 and rotates at high speed in the direction of the arrow. After passing through the focusing lens 19, the light beam 17 is reflected by the mirror 20 toward the stimulable phosphor sheet 11. The stimulable phosphor sheet 11 is main-scanned by the light beam 17 in the arrow X direction substantially perpendicular to the sub-scanning direction (arrow Y direction).

  From the location where the light beam 17 is irradiated on the stimulable phosphor sheet 11, the amount of stimulated emission light 21 corresponding to the accumulated radiographic image information is emitted. The stimulated emission light 21 enters from the incident end face 22a of the light guide 22, and repeats total internal reflection within the light guide 22 and exits from the exit end face 22b. The emitted stimulated emission light 22 is received by the photomultiplier 23, subjected to photoelectric conversion, and converted into an analog image signal Sa.

  The analog image signal Sa is amplified logarithmically by the log amplifier 26, and then sampled and digitized by the A / D converter 28 at a sampling interval corresponding to the spatial frequency fs = 10 cycles / mm. , Simply expressed as “image signal Sd”). As shown in FIG. 4, the image signal Sd moves the stimulable phosphor sheet 11 in the sub-scanning direction (longitudinal direction) while scanning the light beam 17 in the main scanning direction (lateral direction) relative to the stimulable phosphor sheet 11. ) Shows the radiation image information obtained by two-dimensional scanning of the stimulable phosphor sheet 11. Note that the image signal Sd obtained in this way includes information about the grid image 6 corresponding to the grid 4 in addition to the information about the radiation image 5 corresponding to the subject 3.

  The image signal Sd is temporarily stored in the storage unit 29 and then input to the image signal processing unit 30. The image signal processing unit 30 includes a periodic pattern suppression processing unit (corresponding to a periodic pattern suppression processing device) 40 for executing the periodic pattern suppression processing method of the present invention. Hereinafter, the periodic pattern suppression processing unit 40 will be described in detail using an embodiment.

Example 1
FIG. 5A is a flowchart for explaining the configuration of a periodic pattern suppression processing unit 40a which is an example of a periodic pattern suppression processing unit. The image signal Sd read from the storage unit 29 is subjected to HPF processing by a horizontal high-pass filter (hereinafter abbreviated as “horizontal HPF”) 41 (first step). The signal after HPF processing is referred to as signal Sh. Here, the “horizontal HPF” means a means for performing HPF processing in the direction of the arrow shown in the “main scanning direction (horizontal direction)” of FIG. 4, that is, the direction of the striped pattern of the grid image. The signal Sh is further subjected to LPF processing by a vertical low-pass filter (hereinafter abbreviated as “vertical LPF”) 42 (second step). The signal after the LPF processing is defined as a signal Shl. Here, the “vertical LPF” means means for performing the LPF processing in the direction of the arrow shown in the “sub-scanning direction (vertical direction)” in FIG. 4, that is, in the direction parallel to the striped pattern of the grid image. The subtractor 43 receives the signal Sh and the signal Sh1, performs a subtraction process, and outputs a signal S1. Here, the signal S1 = (signal Sh−signal Shl). This signal S1 is a signal indicating the contrast difference (light / dark difference) between the signal Sh and the signal Sh1, and in the first embodiment, the periodic pattern (by changing the spatial frequency component subtracted from the image signal Sd in accordance with the contrast difference) A periodic pattern processing unit to which a method of suppressing (grid image) is applied is described.

  Next, the signal S <b> 1 is subjected to function processing by the first function processing unit 44. Here, a graph of a function executed in the first function processing unit 44 is shown in FIG. In the graph shown in the figure, the horizontal axis is input (that is, signal S1 = Sh−Sl), and the vertical axis is output (that is, fc (Sh−Sl)). Here, a detailed description will be given using the function fc indicated by a solid line. The function fc is a graph in which output is 0 when −TH ≦ signal S1 ≦ TH, and output is increased or decreased in proportion to the signal S1 when signal S1> TH and signal S1 <−TH. That is, the range of TH ≦ signal S1 ≦ TH is output = 0, but when the signal S1 exceeds TH, the greater the value of the signal S1, that is, the greater the contrast difference between the signal Sh and the signal Sh1, the greater the output. Becomes larger. Note that the value of TH may be changed according to the imaging region, imaging conditions, and the like. That is, it may be possible to select from a plurality of functions (for example, a function fc, a function fc1, a function fc2, etc.) according to the imaging region, imaging conditions, and the like.

  The signal after the function processing by the first function processing unit 44 is set as a signal S2 = fc (Sh−Shl). The adder 45 receives the signal S2 and the signal Shl, performs addition processing, and outputs a signal S3 = S2 + Shl = fc (Sh−Shl) + Shl (third step). Further, the subtractor 46 receives the signal S3 and the image signal Sd, performs a subtraction process (subtraction step), and outputs an image signal Sp = Sd−S3 = Sd− {fc (Sh−Shl) + Shl}. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer.

  Here, a conventional periodic pattern suppression processing method will be described. FIG. 7A shows an example of an image corresponding to the image signal Sd. A grid image G is included in the radiation image X corresponding to the subject. The region Xa in the radiation image X indicates a high density portion. The image obtained by performing the horizontal HPF process on the radiographic image shown in FIG. 7A is the image shown in FIG. 7B, and the image obtained by further applying the vertical LPF to the image of FIG. 7B. Is the image shown in FIG. That is, the image of FIG. 7C is an image corresponding to the signal Shl in the flowchart of FIG. Here, in the image shown in FIG. 7C, a grid image also exists in the region corresponding to the region Xa in FIG. An image obtained by subtracting the image of FIG. 7C from the image of FIG. 7A is the image shown in FIG. Here, it is desirable that only the radiation image X is extracted from the image of FIG. 7D, but the image after the vertical LPF (that is, the image shown in FIG. 7C) originally has a grid image. Since the grid image also exists in the region corresponding to the region Xa not to be processed, the grid image has been mixed into the region Xa as shown in FIG.

  Therefore, in the first embodiment of the present invention, the image after the horizontal HPF process (for example, the image shown in FIG. 7B) and the image after the vertical LPF process (for example, the image shown in FIG. 7C) are processed. A contrast difference is calculated (corresponding to the subtractor 43 in FIG. 5A), and in a region where the contrast difference is large, the spatial frequency component corresponding to the image after the horizontal HPF processing is increased (FIG. 5A). And equivalent to the adder 45 in FIG. 5A). Thereby, while suppressing the grid image mixed in the entire image, it is also possible to suppress the appearance of the grid image in a high density portion (for example, the region Xa in FIG. 7D) where the grid image originally does not exist. it can.

  Expressions corresponding to the flowchart shown in FIG. 5A are as follows.

Sp = Sd− {fc (Sh−Shl) + Shl} (1)
The function fc shown in FIG. 5B in the present embodiment is shown as increasing the weight of the spatial frequency component of the signal Sh as the density of the signal Sh is higher than the density of the signal Sh1, but the periodic pattern When gradation inversion processing or the like is performed on the image signal Sd before being input to the suppression processing unit 40a, the method of weighting the spatial frequency component of the signal Sh is also inverted. That is, when gradation inversion processing is performed on the image signal Sd, the weight of the spatial frequency component of the signal Sh is increased as the density of the signal Sh is lower than the density of the signal Sh1.

  Further, the function used in the first function processing unit 44 may be a function corresponding to a graph as shown in FIG. The function shown in FIG. 6B is assumed to be fc1. The function fc1 increases or decreases in proportion to the magnitude of the signal S1 in the range of −TH ≦ signal S1 ≦ + TH, and the output is constant in the range of −TH> signal S1 and + TH <signal S1. . FIG. 6A is a flowchart for explaining the configuration of a periodic pattern suppression processing unit 40b which is an example of a periodic pattern suppression processing unit. The image signal Sd read from the storage unit 29 is subjected to HPF processing by the horizontal HPF 41. The signal after HPF processing is referred to as signal Sh. The signal Sh is further subjected to LPF processing by the vertical LPF 42. The signal after the LPF processing is defined as a signal Shl. The subtractor 43 receives the signal Sh and the signal Sh1, performs a subtraction process, and outputs a signal S1. Next, the signal S <b> 1 is subjected to function processing by the third function processing unit 54. The graph of the function executed in the first function processing unit 54 is the graph shown in FIG.

The signal after the function processing by the third function processing unit 54 is set as signal S18 = fc1 (Sh-Shl). Further, the subtractor 53 receives the signals S1 and S18, performs subtraction processing, and outputs a signal S19. The adder 45 receives the signal S19 and the signal Shl, performs addition processing, and outputs a signal S3 = S19 + Shl = {(Sh-Shl) -fc1 (Sh-Shl)} + Shl = Sh-fc1 (Sh-Shl). . Further, the subtractor 46 receives the signal S3 and the image signal Sd, performs a subtraction process, and outputs an image signal Sp = Sd−S3 = Sd− {Sh−fc1 (Sh−Shl)}. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer. When this function fc1 is used, the image signal Sp is Sp = Sd− {Sh−fc1 (Sh−Shl)} (2)
It will be expressed by the following formula.

(Example 2)
In the first embodiment, in the region where the contrast difference between the images corresponding to the signals Sh and Sh1 is large, the weight of the spatial frequency of the signal Sh is increased and added to the signal Sh1, and the added signal is subtracted from the image signal Sd. The method for suppressing the grid image has been described. In the second embodiment, a method of suppressing the grid image by increasing the spatial frequency weight of the signal Sh in the high density portion of the image signal Sd and adding it to the signal Sh1, and subtracting the added signal from the image signal Sd. explain.

  FIG. 8A is a flowchart for explaining the configuration of a periodic pattern suppression processing unit 40c, which is an example of a periodic pattern suppression processing unit. The image signal Sd read from the storage unit 29 is subjected to HPF processing by the horizontal HPF 41 (signal Sh). The signal Sh is further subjected to LPF processing by the vertical LPF 42 (signal Sh1). The subtractor 43 receives the signal Sh and the signal Sh1, performs a subtraction process, and outputs a signal S1.

  On the other hand, the image signal Sd is subjected to function processing by the second function processing unit 47. Here, a graph of the function executed in the second function processing 47 is shown in FIG. In the graph shown in the figure, the horizontal axis is input (that is, the image signal Sd), and the vertical axis is output. Moreover, the one closer to the vertical axis on the horizontal axis indicates a higher concentration, and the lower the distance from the vertical axis, the lower the concentration. Here, the function fd indicated by a solid line will be described in detail. The function fd is output = 1 when 0 ≦ image signal Sd ≦ TH1, and output = 0 when TH2 ≦ image signal Sd. The range of TH1 <image signal Sd <TH2 is a graph in which the output value increases as the density of the image signal Sd increases, and the output value decreases as the density decreases. Note that the values of TH1 and TH2 may be changed according to the imaging region, imaging conditions, and the like. That is, it may be possible to select from a plurality of functions (for example, a function fd, a function fd1, a function fd2, etc.) according to the imaging region, imaging conditions, and the like.

  A signal after function processing by the second function processing unit 46 is set as signal S6 = fd (Sd). The multiplier 48 receives the signal S6 and the signal S1, performs multiplication, and outputs a signal S4 = fd (Sd) × S1 = fd (Sd) × (Sh−Shl). The signal S4 and the signal Shl are subjected to addition processing by the adder 45 (fourth step), and a signal S5 = S4 + Shl = fd (Sd) × (Sh−Shl) + Shl is output. Here, the second function processing unit 46 outputs a value closer to 1 as the image signal Sd has a higher density value, and outputs a value closer to 0 as the image signal Sd has a lower density value. . That is, if the image signal Sd is a value indicating a high density, the ratio of the signal Sh included in the signal S5 is increased (the weight of the signal Sh is increased), and if the image signal Sd is a value indicating a low density, the signal The ratio of the signal Sh included in S5 is reduced.

  Further, the subtractor 46 receives the signal S5 and the image signal Sd, performs a subtraction process, and outputs an image signal Sp = Sd−S5 = Sd− {fd (Sd) × (Sh−Shl) + Shl}. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer. That is, the equation corresponding to the flowchart shown in FIG.

Sp = Sd− {fd (Sd) × (Sh−Shl) + Shl} (3)
As described above, the weight of the spatial frequency component of the signal Sh in the high density portion of the image signal Sd is increased and added to the signal Sh1, and the image signal Sp is obtained by subtracting the added signal from the image signal Sd. This suppresses the grid image mixed in the entire image without degrading the image quality of the image corresponding to the image signal Sd, and further suppresses the appearance of the grid image in the high density portion where the grid image does not originally exist. Can do.

  Note that the function fd shown in FIG. 8B in the present embodiment is shown as increasing the weight of the spatial frequency component of the signal Sh as the image density indicated by the image signal Sd is higher. When gradation inversion processing or the like is performed on the image signal Sd before being input to the processing unit 40a, the method of weighting the spatial frequency component of the signal Sh is also inverted. That is, when gradation inversion processing is performed on the image signal Sd, the weight of the spatial frequency component of the signal Sh is increased as the image density indicated by the image signal Sd is lower.

(Example 3)
In the third embodiment, a method in which the periodic pattern suppression processing methods described in the first and second embodiments are combined will be described. FIG. 9 is a flowchart for explaining a configuration of a periodic pattern suppression processing unit 40d which is an example of a periodic pattern suppression processing unit. The image signal Sd read from the storage unit 29 is subjected to HPF processing by the horizontal HPF 41 (signal Sh). The signal Sh is further subjected to LPF processing by the vertical LPF 42 (signal Sh1). The subtractor 43 receives the signal Sh and the signal Sh1, performs a subtraction process and outputs a signal S1, and the signal S1 is input to the first function processing unit 44 to be subjected to function processing. The function process executed in the first function processing unit 44 is the same as the function process described in the first embodiment. The signal after function processing is defined as signal S2.

  On the other hand, the image signal Sd is subjected to function processing by the second function processing unit 47. Here, the function processing executed in the second function processing unit 47 is the same as the function processing described in the second embodiment. The signal after function processing is defined as signal S6. The multiplier 48 receives the signal S6 and the signal S2 and performs a multiplication process, and outputs a signal S7 = S6 × S2 = fd (Sd) × fc (Sh−Shl). The adder 45 receives the signal S7 and the signal Shl and performs addition processing, and outputs a signal S8 = S7 + Shl = fd (Sd) × fc (Sh−Shl) + Shl.

  Further, the subtractor 46 receives the signal S8 and the image signal Sd, performs a subtraction process, and outputs an image signal Sp = Sd−S8 = Sd− {fd (Sd) × fc (Sh−Shl) + Shl}. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer. That is, the equation corresponding to the flowchart shown in FIG. 9 is as follows.

Sp = Sd− {fd (Sd) × fc (Sh−Shl) + Shl} (4)
That is, the formula (4) is a combination of the formula (1) described in the first embodiment and the formula (3) described in the second embodiment. In addition, although the flowchart is not shown, it is needless to say that the expressions (2) and (3) may be combined.

  As described above, the weight of the signal Sh in the spatial frequency component to be subtracted from the image signal Sd is changed according to the contrast difference between the signal Sh and the signal Shl and the density of the image signal Sd. It is also possible to suppress the appearance of a grid image in a high density portion where the grid image originally does not exist, while suppressing the grid image that is present. Further, the image corresponding to the image signal Sp in the expression (4) has a smooth boundary portion where the spatial frequency weights of the signal Sh and the signal Shl are switched as compared with the image corresponding to the image signal Sp in the first and second embodiments. It becomes an image expressed in.

Example 4
In the third embodiment, the periodic pattern suppression processing method in which the first embodiment and the second embodiment are combined has been described. In the fourth embodiment, a periodic pattern suppression processing method in the case where an LPF using a cutoff frequency higher than the cutoff frequency used in the vertical LPF 42 is further used to reduce image deterioration such as contrast reduction will be described. .

  FIG. 10 is a flowchart for explaining the configuration of a periodic pattern suppression processing unit 40e which is an example of a periodic pattern suppression processing unit. The image signal Sd read from the storage unit 29 is subjected to HPF processing by the horizontal HPF 41 (signal Sh). The signal Sh is further subjected to LPF processing by the vertical LPF 42 (signal Sh1). The signal Sh is subjected to LPF processing by the vertical LPF 49 (signal Shll). Here, it is assumed that the cut-off frequency used in the vertical LPF 49 is higher than the cut-off frequency used in the vertical LPF 42. The subtracter 50 receives the signal Shll and the signal Shl, performs a subtraction process, and outputs a signal S9. The signal S9 is input to the first function processing unit 44 and subjected to function processing. The function process executed in the first function processing unit 44 is the same as the function process described in the first embodiment. The signal after function processing is defined as signal S10.

  On the other hand, the image signal Sd is subjected to function processing by the second function processing unit 47. Here, the function processing executed in the second function processing unit 47 is the same as the function processing described in the second embodiment. The signal after function processing is defined as signal S6. The multiplier 48 receives the signal S6 and the signal S10, performs multiplication processing, and outputs a signal S11 = S6 × S10 = fd (Sd) × fc (Shll−Shl). The adder 45 receives the signal S11 and the signal Shl, performs addition processing, and outputs a signal S12 = S11 + Shl = fd (Sd) × fc (Shll−Shl) + Shl. Further, the subtractor 46 receives the signal S12 and the image signal Sd, performs a subtraction process, and outputs an image signal Sp = Sd−S12 = Sd− {fd (Sd) × fc (Shll−Shl) + Shl}. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer. That is, the equation corresponding to the flowchart shown in FIG. 10 is as follows.

Sp = Sd− {fd (Sd) × fc (Shll−Shl) + Shl} (5)
FIG. 11 is a flowchart for explaining the configuration of a periodic pattern suppression processing unit 40f, which is an example of a periodic pattern suppression processing unit, and is a modification of the periodic pattern suppression processing unit 40e. The image signal Sd read from the storage unit 29 is subjected to HPF processing by the horizontal HPF 41 (signal Sh). The signal Sh is further subjected to LPF processing by the vertical LPF 42 (signal Sh1). The signal Sh is subjected to LPF processing by the vertical LPF 49 (signal Shll). The subtractor 50 receives the signal Shll and the signal Shl, performs a subtraction process, and outputs a signal S9. The signal S9 is input to the first function processing unit 44 and subjected to function processing. The signal after function processing is defined as signal S10.

  Further, the subtractor 51 receives the signal Shll and the signal Sh, performs a subtraction process, and outputs a signal S13. The signal S13 is input to the first function processing unit 44 and subjected to function processing. The signal after function processing is defined as signal S14. Further, the adder 52 inputs the signal S10 and the signal S14 and performs addition processing. The signal after the addition process is referred to as signal S15. Here, the signal S15 = fc (Sh-Shll) + fc (Shll-Shl).

  On the other hand, the image signal Sd is subjected to function processing by the second function processing unit 47. The signal after function processing is defined as signal S6. The multiplier 48 receives the signal S6 and the signal S15, performs multiplication processing, and outputs a signal S16 = S6 * S15 = fd (Sd) * {fc (Sh-Shll) + fc (Shll-Shl)}. Then, the adder 45 signal S16 and the signal Shl are input to perform addition processing, and a signal S17 = S16 + Shl = fd (Sd) × {fc (Sh−Shll) + fc (Shll−Shl)} + Shl is output. Further, the subtractor 46 receives the signal S17 and the image signal Sd and performs a subtraction process, and the image signal Sp = Sd−S17 = Sd− {fd (Sd) × {fc (Sh−Shll) + fc (Shll−Shl)}. + Shl} is output. An image corresponding to the image signal Sp is displayed on display means (not shown) such as a monitor, or output to output means (not shown) such as a printer. That is, the equation corresponding to the flowchart shown in FIG. 11 is as follows.

Sp = Sd− {fd (Sd) × {fc (Sh−Shll) + fc (Shll−Shl)} + Shl} (6)
As described above, the weight of the signal Sh in the spatial frequency component to be subtracted from the image signal Sd according to the contrast difference between the signal Shl and the signal Shll, the contrast difference between the signal Shll and the signal Sh, and the density of the image signal Sd. By changing, it is possible to suppress the appearance of the grid image in the high density portion where the grid image originally does not exist, while suppressing the grid image mixed in the entire image. Further, by obtaining the image signal Sp using the vertical LPF 49 using a cutoff frequency higher than the cutoff frequency used for the vertical LPF 42, image degradation such as image blur can be reduced.

Schematic diagram of radiographic imaging equipment Example of radiographic images of subject image and grid image stored and recorded on the stimulable phosphor sheet Combination of perspective view and functional block diagram of radiation image reader The figure which shows the main scanning direction and sub-scanning direction of a stimulable phosphor sheet The flowchart for demonstrating the structure of the periodic pattern suppression process part in Example 1, and the graph of function fc The flowchart for demonstrating the structure of the periodic pattern suppression process part in Example 1, and the graph of function fc The figure for demonstrating the conventional periodic pattern suppression processing method Flowchart and function fd graph for explaining the configuration of the periodic pattern suppression processing unit in the second embodiment The flowchart for demonstrating the structure of the periodic pattern suppression process part in Example 3. FIG. The flowchart for demonstrating the structure of the periodic pattern suppression process part in Example 4. FIG. The flowchart for demonstrating the structure of the periodic pattern suppression process part in Example 4. FIG.

Explanation of symbols

41 Transverse HPF
42, 49 Longitudinal LPF
43, 46, 50, 51, 53 Subtractor 44 First function processing unit 45, 52 Adder 47 Second function processing unit 48 Multiplier 54 Third function processing unit

Claims (12)

A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing method for suppressing the periodic pattern,
A first step of performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal ;
A second step of performing a one-dimensional filtering process for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first step;
The greater the contrast difference between the image signal obtained in the first step and the image signal obtained in the second step, the greater the weight of the spatial frequency component of the image signal obtained in the first step. Subtracting including a third step of adding to the spatial frequency component of the image signal obtained by the second step, and subtracting the added spatial frequency component obtained by the third step from the original image signal Steps,
A periodic pattern suppression processing method comprising: A fifth step of performing a filtering process in a direction parallel to the periodic pattern on the image signal obtained by the first step by using a cutoff frequency higher than the cutoff frequency in the filtering process of the second step. Further including
In the third step, the spatial frequency of the image signal obtained by the fifth step increases as the contrast difference between the image signal obtained by the fifth step and the image signal obtained by the second step increases. 2. The periodic pattern suppression processing method according to claim 1, wherein the weighting of the component is increased and added to the spatial frequency component of the image signal obtained by the second step. The weight of the spatial frequency component of the image signal obtained by the first step is increased as the contrast difference between the image signal obtained by the first step and the image signal obtained by the fifth step is larger. Further comprising 6 steps,
In the third step, the spatial frequency of the image signal obtained by the fifth step increases as the contrast difference between the image signal obtained by the fifth step and the image signal obtained by the second step increases. The spatial frequency component obtained by the sixth step is further added to the spatial frequency component in which the component weight is increased, and the spatial frequency component of the image signal obtained by the second step is added to the added spatial frequency component. The periodic pattern suppression processing method according to claim 2, wherein the periodic pattern suppression processing method is added to a component.   The original image is composed of a radiographic image taken through a stationary grid and a striped grid image corresponding to the stationary grid, and the periodic pattern is the grid image. The periodic pattern suppression processing method according to any one of claims 1 to 3. A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing apparatus for suppressing the periodic pattern,
First means for performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal ;
Second means for performing one-dimensional filter processing for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first means;
As the contrast difference between the image signal obtained by the first means and the image signal obtained by the second means is larger, the weight of the spatial frequency component of the image signal obtained by the first means is increased. Subtracting means for adding to the spatial frequency component of the image signal obtained in the second means, subtracting the added spatial frequency component obtained in the third means from the original image signal Means,
A periodic pattern suppression processing apparatus comprising: Fifth means for performing filter processing in a direction parallel to the periodic pattern on the image signal obtained by the first means, using a cutoff frequency higher than the cutoff frequency in the filter processing of the second means. Further comprising
In the third means, the spatial frequency of the image signal obtained in the fifth means increases as the contrast difference between the image signal obtained in the fifth means and the image signal obtained in the second means increases. 6. The periodic pattern control apparatus according to claim 5, wherein the weight of the component is increased and added to the spatial frequency component of the image signal obtained by the second means. As the contrast difference between the image signal obtained by the first means and the image signal obtained by the fifth means increases, the weight of the spatial frequency component of the image signal obtained by the first means increases. Further comprising six means,
In the third means, the spatial frequency of the image signal obtained in the fifth means increases as the contrast difference between the image signal obtained in the fifth means and the image signal obtained in the second means increases. The spatial frequency component obtained by the sixth means is further added to the spatial frequency component having a larger component weight, and the spatial frequency component of the image signal obtained by the second means is added to the added spatial frequency component. The periodic pattern suppression processing apparatus according to claim 6, wherein the periodic pattern suppression processing apparatus is added to a component.   The original image is composed of a radiographic image taken through a stationary grid and a striped grid image corresponding to the stationary grid, and the periodic pattern is the grid image. The periodic pattern suppression processing apparatus according to any one of claims 5 to 7. A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing method for suppressing the periodic pattern,
  A first step of performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal;
  A second step of performing a one-dimensional filtering process for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first step;
  As the density of the original image signal is higher, the weight of the spatial frequency component of the image signal obtained by the first step is increased and added to the spatial frequency component of the image signal obtained by the second step. A subtracting step of subtracting the added spatial frequency component obtained by the fourth step from the original image signal,
A periodic pattern suppression processing method comprising: A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing method for suppressing the periodic pattern,
  A first step of performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal;
  A second step of performing a one-dimensional filtering process for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first step;
  The greater the contrast difference between the image signal obtained in the first step and the image signal obtained in the second step, the greater the weight of the spatial frequency component of the image signal obtained in the first step. A third step of adding to the spatial frequency component of the image signal obtained by the second step, and the higher the density of the original image signal, the higher the density of the spatial frequency component of the image signal obtained by the first step. Including the fourth step of increasing the weight and adding to the spatial frequency component of the image signal obtained by the second step, the added space obtained by the third step and the fourth step. A subtraction step of subtracting a frequency component from the original image signal;
A periodic pattern suppression processing method comprising: A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing apparatus for suppressing the periodic pattern,
  First means for performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal;
  Second means for performing one-dimensional filter processing for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first means;
  As the density of the original image signal is higher, the weight of the spatial frequency component of the image signal obtained by the first means is increased and added to the spatial frequency component of the image signal obtained by the second means. Subtracting means for subtracting the added spatial frequency component obtained in the fourth means from the original image signal,
A periodic pattern suppression processing apparatus comprising: A spatial frequency component of the periodic pattern is extracted from the original image signal of the original image including the striped periodic pattern by filtering, and the spatial frequency component is subtracted from the original image signal by subtracting the spatial frequency component from the original image signal. In the periodic pattern suppression processing apparatus for suppressing the periodic pattern,
  First means for performing a one-dimensional filter process for extracting a frequency component including a frequency of the periodic pattern in the arrangement direction of the periodic pattern from the original image signal;
  Second means for performing one-dimensional filter processing for extracting a frequency component including a frequency of the periodic pattern in a direction parallel to the periodic pattern from the image signal obtained by the first means;
  As the contrast difference between the image signal obtained by the first means and the image signal obtained by the second means is larger, the weight of the spatial frequency component of the image signal obtained by the first means is increased. Third means for adding to the spatial frequency component of the image signal obtained by the second means, and the higher the density of the original image signal, the higher the density of the spatial frequency component of the image signal obtained by the first means. A fourth means for increasing the weight and adding to the spatial frequency component of the image signal obtained in the second means, and the added space obtained in the third means and the fourth means; Subtracting means for subtracting a frequency component from the original image signal;
A periodic pattern suppression processing apparatus comprising:

Images