JP5288919B2 - Image shooting device - Google Patents

Description

  The present invention relates to an image capturing apparatus, and more particularly to an image capturing apparatus that acquires image data indicating an image detected on an image receiving surface by sequentially reading each line in one direction constituting the image one by one.

  2. Description of the Related Art Conventionally, in a radiographic imaging apparatus that captures a radiographic image, radiation that indicates a subject by irradiating the radiation image detector (so-called imaging plate) having a photoelectric conversion layer sensitive to the radiation with radiation that has passed through the subject. The image was recorded, and the radiation image detector on which the radiation image was recorded was scanned by the reading device while irradiating light from the line light source, and accumulated in the radiation image detector for each line according to the radiation dose. It is known that a digital radiation image is obtained by reading the electric charge of each pixel as an electric signal and converting the read electric signal into digital data.

  In recent years, a radiographic imaging apparatus using an FPD (flat panel detector) that can arrange an X-ray sensitive layer on a TFT (Thin film transistor) active matrix substrate and convert X-ray information directly into digital data has been put into practical use. ing. This FPD is provided with a sensor portion corresponding to each intersection where a plurality of scanning wirings and a plurality of signal wirings are arranged so as to intersect each other. Is output as an electrical signal, and the read electrical signal is converted into digital data, thereby obtaining a digital radiation image.

  By the way, in such a radiographic imaging apparatus that reads out a radiographic image line by line, an impact is applied to the reading apparatus from the outside during readout of the radiographic image, or a power supply device, power wiring, etc. of the radiographic imaging apparatus As a result of slight voltage fluctuations, noise may enter the read image signal and line noise may occur in the radiation image.

As a technique for removing this line noise, Patent Document 1 provides an unexposed area that is not irradiated with radiation for each line, and obtains a profile of the unexposed area and subtracts it from the image signal. Techniques for removal are disclosed.
JP 2000-174982 A

  However, although the technique described in Patent Document 1 can correct correctly if the noise of each line has the same density throughout the line, for example, if the density is different between one end and the other end of the line, a correction residual occurs. Noise may remain.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide an image photographing apparatus capable of removing line noise.

In order to achieve the above object, the image capturing apparatus of the present invention configures image data indicating image data indicating an image detected on an image receiving surface, a part of which is a non-detection area along a direction intersecting one direction. while, extends along the one direction, and obtaining means for obtaining by reading the image signals of the respective lines are arrayed one by predetermined lines along the intersecting direction, the image data acquired by the acquisition unit generating means for generating profile data which indicates the image of the non-detection region corresponding to a region of the image of the distribution for the cross direction of the noise amount included in each line of the image represented by, generated by the generating means the indicated by the profile data based on the noise amount of each line by changing the processing intensity for each line in the image data Image processing means for performing image processing of removing line noise to the respective line of the image shown Ri, and a.

The image capturing apparatus of the present invention, the acquisition means, together with the partial image data representing the image detected by the image-receiving surface which is a non-detection region along the cross direction with respect to one direction constitutes the image, wherein It is obtained by sequentially reading out the image signals of each line extending along one direction and arranged in plural along the intersecting direction .

In the present invention, the generation means, distribution for the cross direction of the noise amount included the image of the region corresponding to the non-detection region of the image represented by the image data acquired by the acquisition means to said each line of the image profile data indicating is generated by the image processing unit by changing the processing intensity for each line based the indicated by the profile data generated by the generating means to the noise amount of each line of the image represented by the image data the image processing of removing line noise for each line is performed.

As described above, in the present invention, profile data indicating the distribution of the noise amount included in each line of the image with respect to the intersecting direction is generated from the image of the region corresponding to the non-detection region of the image indicated by the image data, Since line processing is performed on each line of the image indicated by the image data by changing the processing intensity for each line based on the noise amount of each line indicated by the profile data, line noise Can be removed.

Note that the image processing means, wherein each line spread the by expanding by a predetermined amount in one said profile data by the both or either the amount of noise peak noise amount becomes and longitudinally of large the line indicated It obtains the envelope of the peak of removing line noise with respect to each line of the image by changing the processing intensity for each line in accordance with the noise amount of each line indicated by the envelope indicated by the image data Image processing may be performed.

Also, the generation means performs a filtering process on an image of a region corresponding to the non-detection region includes an image of a region corresponding to the non-detection region after the filtering process on the respective line of the image It is preferable to generate profile data indicating a distribution of the noise amount with respect to the intersecting direction .

Further, the image processing means determines the magnitude of the line noise by comparing the noise amount of each line with one or more predetermined threshold values, and increases the processing intensity as the line noise increases. the line noise to each line of the image the image processing may be performed to remove indicated by the data.

  Moreover, you may further provide the warning means which alerts, when the said noise amount is larger than a predetermined threshold value.

Further, the image processing means, a frequency enhancement processing for each line of the image represented by the image data by changing the intensity of the frequency enhancement processing in accordance with the noise amount of each line indicated by the profile data You may go.

Furthermore, the image receiving surface is made a non-detection region by covering the non-detection region with a blocking unit that blocks radiation, and the acquisition unit is a radiation that is detected on the image receiving surface by being irradiated with radiation that has passed through the subject. the image data representing the image, may alternatively be obtained by reading the one direction of the respective lines constituting the radiation image in order by a predetermined line.

As described above, according to the present invention, the profile data indicating the distribution of the noise amount included in each line of the image with respect to the intersecting direction is generated from the image of the region corresponding to the non-detection region of the image indicated by the image data. Since image processing is performed to remove line noise for each line of the image indicated by the image data by changing the processing intensity for each line based on the noise amount of each line indicated by the generated profile data. It has an excellent effect that line noise can be removed.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following, the case where the present invention is applied to a radiographic image capturing apparatus 10 that captures a radiographic image using X-rays as an image capturing apparatus will be described. [First Embodiment]
FIG. 1 shows a schematic configuration of a radiation image capturing apparatus 10 according to the present embodiment.

  As shown in the figure, the radiographic imaging device 10 according to the present exemplary embodiment includes a radiation generation unit 12 that generates radiation such as X-rays (X-rays), and radiation provided at a distance from the radiation generation unit 12. A detection unit 14, an operation button 15 on which various operation instructions are input, an operation panel 15 provided with a display unit for displaying various messages and the like, and a control unit 80 that controls the operation of the entire radiographic imaging apparatus 10. And.

  In addition, the X-ray irradiation surface side of a part of the radiation generation unit 12 is configured by a member mainly composed of lead that absorbs X-rays, and blocks X-rays irradiated to a part of the image receiving surface. A portion 18 is provided.

  Between the radiation generation unit 12 and the radiation detection unit 14 is an imaging position where the subject 16 is positioned during imaging. X-rays radiated from the radiation generation unit 12 pass through the subject 16 located at the imaging position and reach the radiation detection unit 14. Thereby, the radiation carrying the image information indicating the subject 16 is irradiated to the radiation detection unit 14.

  The radiation detection unit 14 includes a radiation image detector (details will be described later) on which a radiation image is recorded, and an image reading unit (details will be described later) that reads out the radiation image recorded on the radiation image detector. . The radiation image detector according to the present embodiment includes an electrostatic recording unit including a photoconductive layer that exhibits conductivity when irradiated with radiation, and receives radiation that carries image information. Image information is recorded in a recording unit, and an image signal representing the recorded image information is output. Other examples of the radiation image detector include an optical reading type radiation image detector that reads image information recorded on the electrostatic recording unit by using a semiconductor material that generates an electric charge by light irradiation, and radiation irradiation. There is a radiation image detector of a type that accumulates charges generated by the above-described method and reads the accumulated charges by turning on and off switching elements such as thin film transistors (TFTs) one by one. In the following, the configuration of the optical reading type radiographic image detector will be described as an example.

  FIG. 2 shows a schematic configuration of the radiation image detector 20 according to the present embodiment.

  As shown in the figure, the radiation image detector 20 includes a first electrode layer 22 having transparency to radiation from the radiation generator 12 (referred to as recording light to distinguish reading light described later), A recording photoconductive layer 28 that generates conductivity when irradiated with recording light transmitted through one electrode layer 22, and a reading that generates conductivity when charged with reading light. The photoconductive layer 32 for use, the second electrode layer 38 having the transparent linear electrode 38B, and the substrate 40 having transparency to the reading light are sequentially provided. In addition, at the interface between the recording photoconductive layer 28 and the reading photoconductive layer 32, a two-dimensionally distributed power storage unit that accumulates latent image charges carrying a radiographic image generated in the recording photoconductive layer 28. 30 is formed.

  Further, an image reading unit 68 for reading out a radiographic image recorded on the radiographic image detector 20 is provided on the substrate 40 side of the radiographic image detector 20. The image reading unit 68 includes a line light source 54 in which a large number of LEDs and the like are arranged along the arrangement direction (main scanning direction) of the transparent linear electrodes 38B in the radiation image detector 20.

  The blocking unit 18 described above is disposed at the end of the image receiving surface along a direction orthogonal to the main scanning direction (sub-scanning direction), and blocks X-rays irradiated to a part of the image receiving surface.

In the radiographic imaging apparatus 10 according to the present exemplary embodiment, a radiographic image is recorded in the power storage unit 30 by performing radiographic imaging under predetermined imaging conditions. However, as illustrated in FIG. A region corresponding to a part of the image receiving surface where X-rays are blocked by the blocking unit 18 is an unexposed region where no X-rays are irradiated.

  At the time of reading image information from the radiation image detector 20, a number of LEDs of the line light source 54 are turned on by a drive circuit (not shown) that is a part of the image reading unit 68, and the substrate 40 of the radiation image detector 20. The side surface is irradiated with line-shaped reading light. Further, the line light source 54 emits radiation along the extending direction of the transparent linear electrode 38B (sub-scanning direction (reading direction): arrow A direction in FIG. 2) by a moving mechanism (not shown) that is a part of the image reading unit 68. The image detector 20 is supported so as to be movable on the surface of the substrate 40 side. When the image data is read from the radiation image detector 20, it is moved (sub-scanned) in the sub-scanning direction at a constant moving speed by the moving mechanism described above. Thereby, the line-shaped reading light is sequentially irradiated on the entire surface of the radiation image detector 20 on the substrate 40 side.

  In addition, the image reading unit 68 includes a large number of charge amplifiers 56 respectively connected to different transparent linear electrodes 38B of the radiation image detector 20 and individual transparent linear shapes when the radiation image detector 20 is irradiated with radiation. A high voltage power supply 58 for applying a high voltage between the electrode 38B and the first electrode layer 22, and an electric signal input from one of the charge amplifiers 56 connected to the output terminals of a large number of charge amplifiers 56 are selected. And an A / D converter 62 that is connected to the output terminal of the multiplexer 60 and converts an electric signal input through the multiplexer 60 into digital data and outputs the digital data.

  In the radiation image detector 20, when the line-shaped reading light emitted from the line light source 54 is irradiated, of the image information recorded in the radiation image detector 20 as a latent image charge accumulated in the power storage unit 30. The image information for one line recorded in the portion irradiated with the reading light is output as an electric signal having a level corresponding to the amount of the latent image charge for each pixel through each transparent linear electrode 38B. Is done. The multiplexer 60 outputs the electric signals to the A / D converter 62 so that the electric signals output through the individual transparent linear electrodes 38B and amplified by the charge amplifier 56 are sequentially output to the A / D converter 62. Cycle through the signals. As a result, image data for one line is sequentially output from the A / D converter 62. The above processing is repeated until the line-shaped reading light emitted from the line light source 54 is irradiated on the entire surface of the radiation image detector 20 on the substrate 40 side, and is recorded in the radiation image detector 20. All image information for one image is read out as image data. The read image data is output to the control unit 80.

  FIG. 4 shows the configuration of the control unit 80 according to the present embodiment.

  As shown in the figure, the control unit 80 includes a CPU (central processing unit) 82 that controls the operation of the entire radiographic imaging apparatus 10 and a RAM (Random Access) used as a work area when the CPU 82 executes various processing programs. Memory) 84, various control programs, various image processing programs such as line noise elimination processing described later, ROM (Read Only Memory) 86 in which various parameters are stored in advance, and HDD (hard disk drive) that stores various information Drive) 88, an image detection control unit 90 that controls radiographic image capturing operation and radiographic image reading operation by the radiation detection unit 14, and power supply to the radiation generation unit 12. A source control unit 92 for controlling the X-ray emission and a panel control unit for detecting an operation state of the operation panel 15 It is provided with a 4, a.

  The CPU 82, RAM 84, ROM 86, HDD 88, image detection control unit 90, radiation source control unit 92, and panel control unit 94 are connected to each other via the system bus BUS.

  Therefore, the CPU 82 accesses the RAM 84, ROM 86, and HDD 88, controls the radiation image capturing operation and radiation image reading operation by the radiation detection unit 14 via the image detection control unit 90, and the radiation source control unit 92. The X-ray emission from the radiation generating unit 12 can be controlled. Further, the CPU 82 can grasp the operation state of the user with respect to the operation buttons provided on the operation panel 15 via the panel control unit 94. Further, the CPU 82 can control the display of a message on a display unit provided on the operation panel 15 via the panel control unit 94.

  Next, the operation of the radiographic image capturing apparatus 10 according to the present embodiment will be described.

When performing radiographic imaging, measurement engineer, a subject 16 is disposed between the radiation generating unit 12 and the radiation detecting section 14, performs a predetermined instruction operation for instructing shooting to the operation panel 15.

  The radiographic image capturing apparatus 10 controls the radiation generating unit 12 via the radiation source control unit 92 to emit X-rays from the radiation generating unit 12 when a predetermined instruction operation for instructing the operation panel 15 to perform imaging is performed. .

  X-rays emitted from the radiation generation unit 12 pass through the subject 16 and reach the radiation detection unit 14.

  As a result, a charge corresponding to the dose of the irradiated X-ray is accumulated in the power storage unit 30 of the radiation image detector 20.

  When reading out a radiation image, the CPU 82 controls the image reading unit 68 via the image detection control unit 90 to turn on a large number of LEDs of the line light source 54 and move the image reading unit 68 in the sub-scanning direction. Excitation light emitted from a large number of LEDs is made incident on the radiation image detector 20. Thereby, all the image data for one image recorded in the radiation image detector 20 is read out.

  When the image data read from the radiation image detector 20 by the image reading unit 68 is input, the CPU 82 performs predetermined image processing such as offset correction and shading correction on the input image data, and each image The processed image data is temporarily stored in the RAM 84.

  By the way, in the radiographic imaging device 10 as in the present exemplary embodiment, for example, an impact is applied to the image reading unit 68 from the outside during reading of the radiographic image, or the power supply device of the radiographic imaging device 10 or the like. When a slight voltage fluctuation occurs in the power wiring or the like, noise may enter the read image signal, and line noise may occur in the radiation image along the main scanning direction. Further, when an impact is applied to the image reading unit 68 from the outside, line noise may occur over a plurality of lines with an inclination with respect to the main scanning direction with respect to the radiation image.

  Therefore, the CPU 82 performs image processing for removing such line noise from the radiation image indicated by the image data stored in the RAM 84.

  FIG. 5 is a flowchart showing a detailed flow of the line noise removal process executed by the CPU 82.

In step 100, the unirradiated region image corresponding to a portion of the image receiving surface of a radiation image represented by the image data stored in the RAM 84, for example, for each line in the main scanning direction, for each pixel of the line By calculating the average value of the pixel values or the cumulative value of the pixel values of each pixel of the line and determining the average value or the distribution of the cumulative values in the sub-scanning direction, as shown in FIG. Profile data indicating the amount of noise included in the line is generated.

  In the next step 102, the average value of the noise amount of each line indicated by the generated profile data is obtained, and the average value is subtracted from the noise amount for each line, as shown in FIG. Is converted to zero. Note that the low frequency component of the profile may be extracted by filtering or the like, and the center of the profile may be converted to zero by subtracting the low frequency component from the profile.

  In the next step 104, as shown in FIG. 7, the noise amount peak of each line indicated by the profile data is expanded by a predetermined amount in the direction in which the noise amount increases, and further expanded by several pixels before and after the peak of each line. Are obtained on the plus side and the minus side, respectively.

  In the next step 106, the absolute value of the noise amount indicated by the plus and minus envelopes obtained in step 104 is obtained for each line.

  In the next step 108, line noise is removed for each line of the radiographic image indicated by the image data by changing the processing intensity in accordance with the larger absolute value of the noise amount indicated by the envelope for each line. Perform image processing. In this embodiment, the positive and negative envelopes are obtained, and the processing intensity is changed according to the larger absolute value of the noise amount indicated by the positive and negative envelopes. When the image processing handles the plus side and the minus side independently, the processing intensity may be changed between the plus side and the minus side using both the plus side and the minus side envelope.

  As image processing for removing line noise, for example, the following method can be used.

  Hereinafter, a case will be described in which image processing for removing line noise (hereinafter, also referred to as “line noise removal processing”) is performed after pre-processing is performed on image data to moderate a sudden change in pixel value.

  FIG. 11 shows an image 120E including line noise 121E indicated by the image data.

  12A to 12D are views showing a state in which preprocessing is performed on image data. FIG. 12A is a line in the sub-scanning direction (hh ′ line) of the image 120E shown in FIG. FIG. 12B is a graph showing a difference value between pixel values of adjacent pixels on one line (h-h ′ line) in the sub-scanning direction, and FIG. c) is a graph showing a low-frequency difference value obtained by moving and averaging the difference values, and FIG. 12D shows the difference corresponding to a portion where the low-frequency difference value is a value equal to or greater than a predetermined threshold value. It is a graph which shows the change of the difference value replaced by making the value 0.

  FIG. 13A is a diagram showing an image 120E ′ represented by image data obtained by accumulating the difference values shown in the graph of FIG. 12D, and FIG. It is a graph which shows the change of the pixel value of each pixel calculated | required by accumulating the difference value shown by the graph of (d).

  In the line noise removal processing according to the present embodiment, first, as preprocessing, image data Se indicating a difference value between pixel values of pixels adjacent to each other in the sub-scanning orthogonal direction in the original image 120E is obtained. Specifically, for example, as shown in FIGS. 12A and 12B, for pixels De1 and De2 adjacent to each other in the image data De, a value obtained by subtracting the pixel value of the pixel De1 from the pixel value of the pixel De2. The difference value U1 is obtained, and this difference value U1 is set as the value of the pixel Se1 corresponding to the pixel De1 in the original image 120E. Further, for adjacent pixels De2 and De3 in the image data De, a difference value U2 that is a value obtained by subtracting the pixel value of the pixel De2 from the pixel value of the pixel De3 is obtained, and this difference value U2 is obtained as a pixel in the original image 120E. The value of the pixel Se2 corresponding to De2. In this way, image data Se indicating the entire difference value is obtained.

  Next, in the line noise removal processing according to the present embodiment, the image data Se indicating the difference value is subjected to moving average in the sub-scanning orthogonal direction to obtain moving averaged image data He. Specifically, for example, a value V2 obtained by averaging the values of three pixels Se1, Se2, Se3 is obtained, and this value V2 is set as the value of the pixel He2 corresponding to the pixel Se2 in the original image 120E. In this way, image data He indicating the entire moving average is obtained.

  Subsequently, in the image processing for removing line noise according to the present embodiment, the image data Se indicating the difference value has a value exceeding a predetermined threshold ± Kh in the moving averaged image data He. Image data Te is obtained by performing processing for reducing the absolute value of the pixel value in the image data Se corresponding to the pixel. Specifically, for example, the values of the pixel Te2 and the pixel Te3 in the image data Se corresponding to the pixel He2 and the pixel He3 having a value exceeding the positive threshold + Kh are set to 0, for example. In this way, the image data Te is obtained. That is, the image data Te is obtained by setting the difference value of the pixels corresponding to the regions J1 and J2 whose values change drastically in the sub-scanning orthogonal direction in the image data Se to 0.

  Note that the threshold value ± Kh is the same as the image data indicating the areas J1 and J2 that have large ringing when the process of extracting the uneven stripe component, which is the area where the value of the image data De in the image 120E changes rapidly, is performed. It is determined so that it can be separated from the image data indicating the area. Further, the streak component and the linear image information component are also separated by setting the threshold value.

  Next, in the image processing for removing line noise according to the present embodiment, the image data Te ′ is obtained by accumulating the image data Te in the sub-scanning orthogonal direction. Specifically, for example, the value of the pixel De1 ′ obtained by sequentially adding (cumulatively adding) the image data Te from one side (h side in the h-h ′ line) in the sub-scanning orthogonal direction. The value of the pixel De1 is obtained by adding the value of the pixel Te1 to the pixel De2 ', and the operation of obtaining the value of the pixel De3' by adding the value of the pixel Te2 to the value of the pixel De2 'is repeated. The image data De ′ shown is obtained.

  The image data De ′ obtained here is obtained by gradual change of the value of the image data De in the sub-scanning orthogonal direction in the original image 120E. FIG. 13A shows an image 120E ′ represented by the image data De ′. Further, areas J1 and J2 in the value of the image data De ′ shown in FIG. 13B are areas in which the rapid change in the value of the image data De is moderated.

  Even when the moving average process is not performed in the preprocessing, the rapid change in the value of the image data De in the sub-scanning orthogonal direction in the original image 120E can be moderated. That is, as shown in FIG. 12B, with respect to the image data Se indicating the difference value, the absolute value of the value of the image data having a value exceeding a predetermined threshold value ± Ks among the image data is set. Similar to the above, the image data obtained by performing the reduction process is cumulatively added in the sub-scanning orthogonal direction to create image data that moderates sudden changes in the image data value in the sub-scanning orthogonal direction in the image. You may make it do.

  The threshold value is a separation that separates a region having a large ringing from another region in the process of extracting the uneven stripe component when the threshold value is set for the image data Se and when the threshold value is set for the image data He. The performance is different. That is, the components represented by the image data generally have more low-frequency components than non-uniform components (high-frequency components). Since the moving average attenuates the high frequency component, the threshold for separating the uneven stripe component can be reduced, and the separation performance for separating the uneven stripe component from the image data is improved.

  Next, in the image processing for removing the line noise according to the present embodiment, the image data De ′ is subjected to filter processing (for example, low-pass filter processing) in the sub-scanning orthogonal direction to show the line noise 121E ′. Extract data. FIG. 14 is a diagram showing a state in which image data is filtered to extract a stripe uneven component, and FIG. 14A-1 is a diagram showing an image represented by preprocessed image data. 14 (a-2) is a diagram showing the value of the preprocessed image data, and FIG. 14 (b-1) is the image data obtained by filtering the preprocessed image data. 14 (b-2) is a diagram showing values of image data obtained by applying filter processing to the preprocessed image data, and FIG. 14 (c-). FIG. 14C is a diagram showing the extracted uneven stripe component, and FIG. 14C-2 is a diagram showing image data values representing the extracted uneven stripe component. The figure showing the value of each image data shows the value of the image data located on the hh ′ line extending in the sub-scanning orthogonal direction in each image.

  In the image processing for removing the line noise according to the present embodiment, the image data De ′ indicating the image 120E ′ is subjected to the low-pass filter processing in the sub-scanning orthogonal direction (arrow X direction in the figure) as the filter processing. Image data Db indicating an image 120B composed of a low frequency component is obtained, and image data Dc representing an image 120C composed of a high frequency component is obtained by subtracting the image data Db from the image data De ′.

  At this time, the processing intensity can be changed by changing the threshold value Kh according to the larger absolute value of the noise amount indicated by the envelope.

  This image 120C is an image showing the line noise 121E '. In the process of extracting the image data indicating the non-uniformity component, the low-pass filter process is performed on the image data De ′ and the low-pass filter process is further performed in the sub-scanning direction to obtain an image composed of low-frequency components. You may make it acquire the image data Db 'which shows 120B'.

  As described above, the processing for extracting the stripe unevenness component is performed on the image data De ′ in which the change in the value of the image data in the sub-scanning orthogonal direction is moderated in the preprocessing, and the image data indicating the stripe unevenness component is extracted. By doing so, it is possible to extract image data showing a streaky component while suppressing the occurrence of ringing as compared with the comparative example in which the filter processing is performed on the image data De that is not subjected to the preprocessing.

  Next, in the image processing for removing the line noise according to the present embodiment, the uneven stripe component is removed from the image. FIG. 15 is a diagram illustrating a state in which a stripe unevenness component is removed from an image. FIG. 15A-1 is a diagram illustrating the original image. FIG. 15A-2 is a diagram illustrating image data representing the original image. FIG. 15B-1 is a diagram illustrating the extracted stripe unevenness component, and FIG. 15B-2 is a diagram illustrating the value of the image data representing the extracted stripe unevenness component. 15 (c-1) is a diagram showing an image obtained by removing the stripe unevenness component from the original image, and FIG. 15 (c-2) shows a value of the image data obtained by removing the stripe unevenness component from the image data showing the original image. FIG. The figure showing the value of each image data shows the value of the image data located on the hh ′ line extending in the sub-scanning orthogonal direction in each image.

  In the image processing for removing the line noise according to the present embodiment, the original data 120E is subtracted from the image data De indicating the original image 120E, that is, the image data Dc indicating the image 120C indicating the high frequency component in the original image 120E. Image data Dd indicating an image 120D obtained by removing the uneven stripe component 121E from the image 120E is obtained. Here, the preprocessing can suppress the disappearance of the linear image information component from the image due to the removal of the uneven stripe component.

  As described above, according to the present embodiment, by changing the processing intensity of image processing for removing line noise in accordance with the envelope for each line, the line has a strong intensity with respect to a line with large line noise. Since image processing for removing noise is performed, line noise can be removed even when there is a density change in the stripe direction. Further, since the processing intensity is expanded in the sub-scanning direction by expanding the envelope by a predetermined amount, even when line noise occurs over a plurality of lines with an inclination with respect to the main scanning direction. Line noise can be removed.

  In this embodiment, the case where the processing intensity of image processing for removing line noise is changed in accordance with the envelope of the noise amount profile indicated by the profile data has been described. However, the present invention is not limited to this. Instead, for example, in the noise amount profile indicated by the profile data, a difference in noise amount between adjacent lines (for example, between the lines on one line) is obtained, and the above-described noise amount difference profile is used to determine the above-described noise amount difference profile. Such an envelope may be obtained, and the processing intensity of image processing for removing line noise may be changed in accordance with the envelope of the noise amount difference profile.

[Second Embodiment]
Next, another form of image processing for removing line noise will be described.

  Since the configuration of the radiographic image capturing apparatus 10 according to the second embodiment is the same as that of the first embodiment (see FIGS. 1, 2, and 4), description thereof is omitted here.

  FIG. 8 is a flowchart showing a detailed flow of the line noise removal processing according to the second embodiment. Note that the description of the same parts in FIG. 8 as those in FIG. 5 is omitted.

In step 204 (in this embodiment, + threshold, - threshold) threshold for determining switching the intensity of a predetermined processing amount of noise for each line indicated by the profile data or exceeds the threshold value as compared to the Whether or not the processing intensity of the image processing is “high” or “small” is determined based on whether or not the processing intensity of the lines before and after the line where the processing intensity becomes “high” exceeding the threshold as shown in FIG. The shape is expanded to “large”.

  In the next step 206, the processing intensity of the actual image processing is determined for each line from the processing intensity on the plus side and the minus side determined in step 204, and either on the plus side or the minus side. A radiation image indicated by image data with a processing intensity of “high” when the processing intensity is “high” and with a processing intensity of “low” when the processing intensity is “small” on both the plus side and the minus side. Image processing for removing line noise is performed on each line. The positive threshold value and the negative threshold value have the same absolute value except for the sign, but may be determined separately. Further, when the processing intensity is determined to be three or more levels, a plurality of threshold values may be determined and compared with each threshold value. This threshold value may be determined by experiment or simulation according to the processing intensity required to remove noise.

  In the next step 208, image processing for removing line noise is performed on each line of the radiation image indicated by the image data with the processing intensity obtained in step 206 for each line.

  As described above, according to the present embodiment, the processing intensity of image processing for removing line noise is controlled in stages by determining the processing intensity by comparing the noise amount of each line with a predetermined threshold. .

  In each of the above-described embodiments, the case where the light reading type radiation image detector 20 is used for the radiation detection unit 14 has been described. However, the present invention is not limited to this, for example, on the image receiving surface. A radiographic image detector 20 such as an FPD that directly converts received X-rays into digital data may be used. The radiation image detector 20 is provided with a plurality of two-dimensional sensor units having sensitivity to X-rays on the image receiving surface, and captures a radiation image received on the image receiving surface.

  FIG. 10 shows a schematic configuration of a radiation image detector 20 in which a plurality of sensor units are provided in a two-dimensional manner.

  As shown in the figure, the radiation image detector 20 has sensitivity to X-rays, and stores a sensor unit 70 that accumulates charges according to the dose of irradiated X-rays, and the sensor unit 70 stores the charges. A plurality of pixels each including a TFT (Thin film transistor) switch 71 for reading out charges is provided in a two-dimensional manner.

  In the radiation image detector 20, a plurality of scanning wirings 72 for turning on / off the TFT switch 71 and a plurality of signal wirings 73 for reading out charges accumulated in the sensor unit 70 are mutually connected. It is provided crossing.

  An electric signal corresponding to the amount of charge accumulated in the sensor unit 70 flows through each signal line 73 when any TFT switch 71 connected to the signal line 73 is turned on. Each signal wiring 73 is connected to a signal detection circuit 74 that detects an electrical signal flowing out to each signal wiring 73, and each scanning wiring 72 is used to turn on / off the TFT switch 71 in each scanning wiring 72. A scan signal control circuit 75 for outputting the control signal is connected.

  The signal detection circuit 74 includes an amplification circuit for amplifying the input electric signal for each signal wiring 73. In the signal detection circuit 74, the electric signal input from each signal wiring 73 is amplified and detected by the amplification circuit, and is stored in each sensor unit 70 as information (pixel value) of each pixel constituting the image. Detect the amount of charge.

  The signal detection circuit 74 and the scan signal control circuit 75 perform predetermined processing on the electrical signal detected by the signal detection circuit 74 and output a control signal indicating the signal detection timing to the signal detection circuit 74. A signal processing circuit 76 that outputs a control signal indicating the output timing of the scan signal is connected to the scan signal control circuit 75.

  In the radiographic imaging apparatus 10 including such a radiographic image detector 20, when reading out a radiographic image, the CPU 82 controls the signal processing circuit 76 to sequentially scan each line 72 from the scan signal control circuit 75 line by line. The ON signal (+10 to 20 V) is output to each TFT, and each TFT switch 71 connected to each scanning wiring 72 is sequentially turned on line by line, so that each signal wiring 73 is stored in each sensor unit 70 line by line. An electric signal corresponding to the amount of electric charge flows out, and the signal detection circuit 74 detects the amount of electric charge accumulated in each sensor unit 70 as a pixel value of each pixel constituting the image based on the electric signal flowing out to the signal wiring 73. By doing so, it is possible to obtain image data indicating a radiographic image indicated by X-rays irradiated to the radiographic image detector 20.

  Even in such a radiographic imaging apparatus 10, an impact is applied from the outside to the wiring or the like of the radiographic imaging apparatus 10 during reading of the radiographic image, or a slight amount is applied to the power supply device or power wiring of the radiographic imaging apparatus 10. When voltage fluctuation occurs, noise may enter the read image signal, and line noise may occur in the radiation image along the main scanning direction.

  However, the line noise can be removed by performing image processing for removing the line noise as in each of the above-described embodiments on the radiation image indicated by the read image data.

  Here, in an apparatus for detecting an electrical signal indicating a radiographic image using a plurality of substrates, if the cause of the electric noise generated in the electrical signal is ground noise, depending on how the noise is superimposed, streaks may occur on the plurality of substrates. The phase may be reversed. For this reason, when electrical noise is not removed by image processing that removes line noise, when generating profile data, the electrical noise is canceled by taking out and averaging the signals of the same number of pixels from each substrate. Can do. On the other hand, when electric noise is removed by image processing that removes line noise, the profile data is generated from each substrate and the processing intensity is obtained when generating the profile data, and then the absolute value of each line. By selecting the larger one as the processing intensity and applying the image processing for removing the line noise, the electric noise can be reliably removed.

  In each of the above embodiments, the following processing may be performed together with image processing for removing line noise.

Noise and point defects, to remove the effects of line defects, after the filtering by the median filter in the stripe direction of the profile data to the unirradiated region of the radiographic image (in FIG. 6 the main scanning direction), the profile data You may ask for. Alternatively, the median value of the data in the stripe direction may be obtained and used as profile data. In order to further reduce the influence of noise, filter processing for removing high frequency components from the profile data may be performed.

  In addition, the optical reading type radiological image detector 20 can extract only noise caused by electric as an electric signal by performing reading without irradiating the reading light. Therefore, by providing a region in the radiation image detector 20 where the reading light is not irradiated, extracting a streak signal from the profile data of the region and subtracting it from the image data, it is possible to remove the streak caused by the electric field.

  Also, image processing that facilitates diagnosis, such as frequency enhancement processing, may be applied to the image data. Here, when it is determined that a large unevenness has entered using the threshold value as described above, the intensity of image processing such as frequency enhancement processing may be reduced for the image. Thereby, since the emphasis by the image processing of the streak portion is weakened, it is possible to make the streak that cannot be removed by the streak removal processing inconspicuous.

  Furthermore, when it is determined that a large uneven stripe as described above has entered, a warning may be given to the display unit provided on the operation panel 15 to notify the user. Alternatively, the fact that a large stripe is entered may be left as log data of the apparatus so that the cause can be analyzed later.

  Further, in each of the above embodiments, the case where a part of the image receiving surface is set as a non-detection area by providing the blocking portion 18 on a part of the image receiving surface has been described, but the present invention is not limited to this. For example, in a radiographic image detector 20 such as an FPD that converts X-rays received on the image-receiving surface into digital data, the sensitivity to X-rays is eliminated by a part of the sensor unit 70 on the image-receiving surface. Good. In the radiation image detector 20 such as an FPD, the sensor unit 70 is made sensitive to X-rays by, for example, forming amorphous silicon or the like, but amorphous silicon is not formed on a part of the image receiving surface. Thus, a part of the image receiving surface can be set as a non-detection area.

  Further, in each of the above embodiments, the case where the present invention is applied to the radiographic imaging apparatus 10 that detects a radiographic image by X-rays as radiation has been described. However, the present invention is not limited to this, for example, The radiation may be gamma rays or the like, and of course can be applied to an apparatus that captures image data including visible light.

  In addition, the configuration (see FIGS. 1, 2, 4, and 10) of the radiation image capturing apparatus 10 described in each of the above embodiments is an example, and may be changed as appropriate without departing from the gist of the present invention. It goes without saying that it is possible.

  Further, the flow of line noise removal processing (see FIGS. 5 and 8) described in each of the above embodiments is also an example, and it goes without saying that it can be changed as appropriate without departing from the gist of the present invention. .

It is a block diagram which shows schematic structure of the radiographic imaging apparatus which concerns on embodiment. It is a block diagram which shows schematic structure of the radiographic image detector which concerns on embodiment. It is a figure which shows the non-exposed area | region and effective image area | region of a radiographic image which concern on embodiment. It is a block diagram which shows the structure of the control part which concerns on embodiment. It is a flowchart showing a detailed flow of the engagement Ru line noise removal processing in the first embodiment. It is the schematic diagram which showed typically the profile of the radiation image which concerns on embodiment, and the unexposed area | region of the said radiation image. It is the schematic diagram which showed typically the process with respect to the profile which concerns on 1st Embodiment. It is a flowchart showing a detailed flow of the engagement Ru line noise removal processing to the second embodiment. It is the schematic diagram which showed typically the process with respect to the profile which concerns on 2nd Embodiment. It is a figure which shows schematic structure of the radiographic image detector which concerns on 2nd Embodiment. It is a figure which shows the image containing the line noise which concerns on embodiment. It is a figure which shows a mode that pre-processing is performed to the image data which concerns on embodiment. It is a figure which shows the image data represented by the image data which carried out the cumulative addition which concerns on embodiment, and this image data. It is a figure which shows a mode that a filter process is performed to the image data in which the pre-processing which concerns on embodiment was performed, and a stripe unevenness component is extracted. It is a figure which shows a mode that a stripe unevenness component is removed from the original image which concerns on embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Radiation imaging device 14 Radiation detection part 15 Operation panel (warning means)
16 Subject 18 Blocking part (blocking means)
20 Radiation image detector (acquisition means)
82 CPU (generating means, image processing means)

Claims (7)

The image data representing the image part to be detected by the image-receiving surface which is a non-detection region along the cross direction the other hand with respect to the direction, as well as constituting the image, extends along the one direction, and in the cross direction Acquisition means for acquiring by sequentially reading out the image signal of each line arranged in a row along a predetermined line;
Generating for generating profile data which indicates the distribution for the cross direction of the noise amount included the image of the region corresponding to the non-detection region of the image represented by the image data acquired by the acquisition unit to the respective line of the image Means,
Image reject line noise on the respective lines of the image represented by the image data by changing the processing intensity for each line on the basis of the noise amount of each line indicated by the profile data generated by the generating means Image processing means for performing processing;
An image photographing apparatus comprising: Wherein the image processing means, the profile data by the peak of the line spread the by expanding by a predetermined amount said and in one longitudinal direction of both or either direction noise amount a peak of the noise amount of each line is increased as shown Determination envelope, the image to remove the line noise with respect to each line of the image represented the by the image data by changing the processing intensity for each line in accordance with the noise amount of each line indicated by the envelope The image photographing device according to claim 1 which performs processing. Said generating means performs a filtering process on an image of a region corresponding to the non-detection region, the amount of noise included the image of the region corresponding to the non-detection region after the filtering process on the respective line of the image The image photographing device according to claim 1 , wherein profile data indicating a distribution of the crossing direction is generated. The image processing means discriminates the magnitude of the line noise by comparing the noise amount of each line with a predetermined threshold value or a plurality of threshold values. The image photographing device according to any one of claims 1 to 3, wherein image processing for removing line noise is performed on each line of the displayed image. The image capturing apparatus according to claim 1, further comprising a warning unit that issues a warning when the amount of noise is larger than a predetermined threshold. Wherein the image processing means, according to the frequency enhancement processing on each line of the image represented by the image data by changing the intensity of the frequency enhancement processing in accordance with the noise amount of each line indicated by the profile data The image photographing device according to any one of claims 1 to 5. The image receiving surface is a non-detection area by being covered with a blocking means for blocking the radiation.
The acquisition unit acquires by reading the image data representing a radiation image radiation transmitted through the subject is detected by the image-receiving surface is irradiated in the forward one direction of the respective lines constituting the radiation image by a predetermined line The image capturing device according to any one of claims 1 to 6.

Images