Image processing method, image processing device, and image processing program

Abstract

An image acquisition device (1) is provided with: a relationship calculation unit (202) that calculates relationship data between pixel values in an image and statistical values of the pixel values; and a parameter setting unit (204) and a filter processing unit (205) that process an image to be processed, the parameter setting unit (204) and the filter processing unit (205) calculating the statistical values of the pixel values on the basis of the pixel values of each pixel in the image and the relationship data. A statistical value is specified for each pixel of an image, a parameter of an edge-saving smoothing filter is set for each pixel of the image on the basis of the specified statistical value, and the image is processed using the edge-saving smoothing filter in which the parameter is set for each pixel of the image.

Description

Technical Field

[0001] This invention relates to image processing methods, image processing apparatus, and image processing programs. Background Technology

[0002] Currently, a bilateral filter, which is a type of edge-preserving smoothing filter, can be used as a filter for removing noise from an image (see, for example, Non-Patent Document 1 and Non-Patent Document 2). The parameters of the bilateral filter are adjusted to obtain a good image.

[0003] Existing technical documents

[0004] Non-patent literature

[0005] Non-Patent Literature 1: Yoza Akio, 4 others, "Parameter Determination of Bilateral Filters in Medical Images", 2019, IEEE Transactions on Electronics, Information and Systems, Vol. 139, No. 9, pp. 1008-1014

[0006] Non-Patent Literature 2: Hayato Hashii, 3 others, "Parameter Adjustment of Bilateral Filter Based on Hellinger Distance", 26th Fuzzy System Symposium, Hiroshima, September 13-15, 2010 Summary of the Invention

[0007] The problem that the invention aims to solve

[0008] Existing noise removal techniques using edge-preserving smoothing filters, as described above, tend to struggle to remove noise from an overall image with varying brightness distributions. Therefore, there is a need to achieve effective noise removal across the entire image.

[0009] This disclosure was developed in view of the above-mentioned technical problems, and its purpose is to provide an image processing method, image processing apparatus and image processing program that can effectively remove noise from the whole image.

[0010] Technical means to solve the problem

[0011] The image processing method according to the first aspect of the implementation includes: a calculation step of calculating the relationship between luminance values ​​and statistical values ​​of luminance values ​​in an image; an acquisition step of acquiring an image of the object to be processed; a setting step of setting parameters for an edge-preserving smoothing filter for each pixel of the image based on the luminance values ​​and relationships of each pixel of the image, and setting parameters for an edge-preserving smoothing filter for each pixel of the image based on the specific statistical values; and a processing step of processing the image using the edge-preserving smoothing filter for which parameters are set for each pixel of the image.

[0012] Alternatively, the image processing apparatus according to the second aspect of the embodiment includes: a relation calculation unit that calculates the relationship between luminance values ​​and statistical values ​​of luminance values ​​in an image; and an image processing unit that processes an image of a processing object, wherein the image processing unit, based on the luminance value and the relationship of each pixel in the image, calculates specific statistical values ​​for each pixel of the image, sets parameters for an edge preservation smoothing filter for each pixel of the image based on the specific statistical values, and processes the image using the edge preservation smoothing filter with parameters set for each pixel of the image.

[0013] Alternatively, the image processing program according to the third aspect of the implementation involves a processor that functions as a relationship calculation unit for calculating the relationship between luminance values ​​and statistical values ​​of luminance values ​​in an image, and an image processing unit for processing the image of the object being processed. The image processing unit, based on the luminance value and relationship of each pixel in the image, sets parameters for an edge-preserving smoothing filter for each pixel in the image based on the specific statistical values, and processes the image using the edge-preserving smoothing filter with parameters set for each pixel in the image.

[0014] According to any one of the first to third aspects described above, a parameter suitable for the brightness value is set for each pixel of the image, and an edge-preserving smoothing filter reflecting the set parameters of each pixel is used to process the image. This removes noise corresponding to the statistical values ​​of the brightness values ​​of each pixel in the image. As a result, noise in the overall image can be effectively removed.

[0015] The effects of the invention

[0016] According to any aspect of this embodiment, noise removal in the overall image can be effectively achieved. Attached Figure Description

[0017] Figure 1 This is a structural diagram of the image acquisition device 1, which is an image processing apparatus according to the first embodiment.

[0018] Figure 2 It means Figure 1 A diagram of the hardware structure of the control device.

[0019] Figure 3 It means Figure 1 A block diagram of the functional structure of the control device.

[0020] Figure 4 It means Figure 3 An example of an X-ray image acquired by the image acquisition unit.

[0021] Figure 5 This is a flowchart showing the sequence of observation processing performed by the image acquisition device 1.

[0022] Figure 6 This is a diagram showing an example of an X-ray image acquired by the image acquisition device 1 according to the first embodiment.

[0023] Figure 7 This is a diagram showing an example of an X-ray image acquired by the image acquisition device 1 according to the first embodiment.

[0024] Figure 8 This is a block diagram showing the functional structure of the optical image processing system 1A, which is the image processing apparatus of the second embodiment.

[0025] Figure 9 yes Figure 8 A diagram of an example of an optical image G1 acquired by the image acquisition unit.

[0026] Figure 10 This is a flowchart showing the sequence of observation processing performed by the optical image processing system 1A. Detailed Implementation

[0027] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. Furthermore, in the drawings, the same or equivalent parts are labeled with the same symbols, and repeated descriptions are omitted.

[0028] [First Implementation Method]

[0029] Figure 1 This is a structural diagram of the image acquisition device 1, which is an image processing apparatus according to the first embodiment. (See diagram below.) Figure 1As shown, the image acquisition device 1 is an apparatus that irradiates an object F transported along the transport direction TD with X-rays (energy rays, radiation), and acquires an X-ray image (radiation image) of the object F based on the X-rays that have passed through it. The image acquisition device 1 uses the X-ray image to perform foreign object inspection, weight inspection, and commodity inspection on the object F. Examples of its applications include food inspection, baggage inspection, substrate inspection, battery inspection, and material inspection. The image acquisition device 1 is configured to include a belt conveyor (transport device) 60, an X-ray irradiator (generating source) 50, an X-ray inspection camera (imaging device, inspection unit) 10, a control device 20, a display device 30, and an input device (input unit) 40 for various inputs. Furthermore, in the embodiments of this disclosure, the image only needs to be an image captured by energy rays that have passed through the object F. Energy rays can be, for example, X-rays and gamma rays, or any of visible light, infrared light, ultraviolet light, and electron beams. In this embodiment, the image is a radiation image such as an X-ray image, but other images are also possible.

[0030] The belt conveyor 60 has a conveyor belt section for carrying the object F. By moving this conveyor belt section along the conveying direction TD, the object F is conveyed at a predetermined conveying speed along the conveying direction TD. The conveying speed of the object F is, for example, 48 m / min. The belt conveyor 60 can change the conveying speed as needed, for example, to 24 m / min or 96 m / min. In addition, the belt conveyor 60 can appropriately change the height position of the conveyor belt section to change the distance between the X-ray irradiator 50 and the object F. Furthermore, the object F conveyed by the belt conveyor 60 can include, for example, various items such as edible meat, seafood, crops, snacks and other food products, rubber products such as tires, resin products, metal products, mineral resources and other materials, waste, and electronic components or electronic boards. The X-ray irradiator 50 is a device that irradiates (outputs) X-rays to the object F as an X-ray source. The X-ray irradiator 50 is a point source that diffuses X-rays within a predetermined angle range in a certain irradiation direction to irradiate. The X-ray irradiator 50 is positioned above the belt conveyor 60 at a predetermined distance, such that the X-ray irradiation direction is towards the belt conveyor 60, and the diffused X-rays cover the entire width direction (the direction intersecting the conveying direction TD) of the object F. Furthermore, the irradiation range of the X-ray irradiator 50 is defined as a predetermined segment along the length direction of the object F (parallel to the conveying direction TD). By conveying the object F along the conveying direction TD from the belt conveyor 60, the entire length direction of the object F is irradiated with X-rays. The tube voltage and tube current of the X-ray irradiator 50 are set by the control device 20. The X-ray irradiator 50 irradiates the belt conveyor 60 with X-rays of a predetermined energy and radiation dose corresponding to the set tube voltage and tube current. Additionally, a filter 51 is provided near the belt conveyor 60 side of the X-ray irradiator 50 to allow X-rays to pass through a predetermined wavelength range. The filter 51 is not necessarily required and can be omitted appropriately.

[0031] X-ray inspection camera 10 detects X-rays that have passed through the object F from the X-rays irradiated by X-ray irradiator 50, and outputs a signal detecting an image based on these X-rays. X-ray inspection camera 10 is a dual-line X-ray camera equipped with two sets of X-ray detection structures. In the image acquisition apparatus 1 of this embodiment, X-ray images are generated (detected) based on the X-rays detected by each line (first line and second line) of the dual-line X-ray camera. Then, by averaging or adding the two generated X-ray images, a clear (bright) image can be obtained with less X-ray radiation compared to the case where an X-ray image is generated based on X-rays detected by a single line. Furthermore, X-ray inspection camera 10 can also be configured by arranging two or more single-line X-ray cameras, including single-line X-ray cameras equipped with one set of X-ray detection structures, multi-line X-ray cameras equipped with two or more sets of X-ray detection structures, and single-line X-ray cameras in general.

[0032] The X-ray inspection camera 10 includes a filter 19, scintillators 11a and 11b, line scan cameras 12a and 12b, a sensor control unit 13, amplifiers 14a and 14b, an AD converter 15a and 15b, a correction circuit 16a and 16b, output interfaces 17a and 17b, and an amplifier control unit 18. The scintillator 11a, line scan camera 12a, amplifier 14a, AD converter 15a, correction circuit 16a, and output interface 17a are electrically connected, which is the structure described in the first line. Furthermore, the scintillator 11b, line scan camera 12b, amplifier 14b, AD converter 15b, correction circuit 16b, and output interface 17b are electrically connected, which is the structure described in the second line. The line scan cameras 12a of the first line and 12b of the second line are arranged along the transport direction TD. Furthermore, the following description of the common structures in the first and second lines will be based on the structure of the first line.

[0033] The scintillator 11a is fixed to the line scan camera 12a by adhesive or other means, and converts the X-rays that have passed through the object F into scintillation light. The scintillator 11a outputs the scintillation light to the line scan camera 12a. The filter 19 directs the X-rays of a specified wavelength range toward the scintillator 11a. The filter 19 is not necessarily required and can be omitted appropriately.

[0034] The line scan camera 12a detects the flashing light from the scintillator 11a and converts it into electrical charge, which is output as a detection signal (electrical signal) to the amplifier 14a. The line scan camera 12a has multiple line sensors arranged side-by-side in a direction intersecting the transport direction TD. These line sensors are, for example, CCD (Charge Coupled Device) image sensors, CMOS (Complementary Metal-Oxide Semiconductor) image sensors, etc., and contain multiple photodiodes.

[0035] The sensor control unit 13 controls the line scan cameras 12a and 12b to repeatedly capture images according to a predetermined detection cycle, so that the line scan cameras 12a and 12b can capture X-rays that have passed through the same area of ​​the object F. The predetermined detection cycle can be set based on, for example, the distance between the line scan cameras 12a and 12b, the speed of the belt conveyor 60, the distance between the X-ray irradiator 50 and the object F on the belt conveyor 60 (FOD (Focus Object Distance)), and the distance between the X-ray irradiator 50 and the line scan cameras 12a and 12b (FDD (Focus Detector Distance)). Alternatively, the predetermined cycle can be set individually based on the pixel width of the photodiodes in a direction orthogonal to the pixel arrangement direction of the respective line sensors of the line scan cameras 12a and 12b. In this case, individual cycles can be set based on the distance between line scan cameras 12a and 12b, the speed of the belt conveyor 60, the distance (FOD) between the X-ray irradiator 50 and the object F on the belt conveyor 60, and the distance (FDD) between the X-ray irradiator 50 and the line scan cameras 12a and 12b, and the deviation (delay time) of the detection cycle between specific line scan cameras 12a and 12b. Amplifier 14a amplifies the detection signal at a predetermined set amplification rate to generate an amplified signal, and outputs this amplified signal to the AD converter 15a. The set amplification rate is the amplification rate set by the amplifier control unit 18. The amplifier control unit 18 sets the set amplification rate of amplifiers 14a and 14b based on predetermined shooting conditions.

[0036] The AD converter 15a converts the amplified signal (voltage signal) output from amplifier 14a into a digital signal and outputs it to the correction circuit 16a. The correction circuit 16a performs prescribed corrections, such as signal amplification, on the digital signal and outputs the corrected digital signal to the output interface 17a. The output interface 17a outputs the digital signal to the outside of the X-ray inspection camera 10. Figure 1In this system, the AD converter, correction circuit, and output interface exist separately, but they can also be combined into one.

[0037] The control device 20 is, for example, a computer such as a PC (Personal Computer). The control device 20 generates an X-ray image based on the digital signals (amplified signals) output from the X-ray inspection camera 10 (more specifically, output interfaces 17a and 17b). The control device 20 generates an X-ray image by averaging or adding the two digital signals output from output interfaces 17a and 17b. The generated X-ray image, as the image to be processed, is output to the display device 30 after noise removal processing as described later, and is displayed by the display device 30. Furthermore, the control device 20 controls the X-ray irradiator 50, the amplifier control unit 18, and the sensor control unit 13. In this embodiment, the control device 20 is a device independently installed outside the X-ray inspection camera 10, but it can also be integrated into the interior of the X-ray inspection camera 10.

[0038] Figure 2 This indicates the hardware structure of the control device 20. For example... Figure 2 As shown, the control device 20 is a computer that physically includes a CPU (Central Processing Unit) 101 and a GPU 105 as processors, RAM (Random Access Memory) 102 and ROM (Read Only Memory) 103 as recording media, a communication module 104, and an input / output module 106, all electrically connected to each other. Furthermore, as an input device 40 and a display device 30, the control device 20 may include a monitor, keyboard, mouse, touch panel display, etc., or it may include a hard disk drive, semiconductor memory, or other data recording devices. Additionally, the control device 20 may be composed of multiple computers.

[0039] Figure 3 This is a block diagram showing the functional structure of the control device 20. The control device 20 includes an image acquisition unit 201, a relation calculation unit 202, a noise map generation unit 203, a parameter setting unit (image processing unit) 204, a filter processing unit (image processing unit) 205, and an image storage unit 207. Figure 3The functional units of the control device 20 shown are implemented as follows: by loading a program (the image processing program of this embodiment) onto hardware such as the CPU 101, GPU 105, and RAM 102, the communication module 104 and input / output module 106 are operated under the control of the CPU 101 and GPU 105, and data is read from and written to the RAM 102. The CPU 101 and GPU 105 of the control device 20 execute this computer program, enabling the control device 20 to function as... Figure 3 Each functional unit operates, sequentially executing the processing corresponding to the image processing method described later. Furthermore, the CPU 101 and GPU 105 can be standalone hardware units, or either one can be the only component. Alternatively, the CPU 101 and GPU 105 can be installed within programmable logic such as an FPGA, similar to a software processor. RAM and ROM can also be standalone hardware units, or they can be integrated into programmable logic such as an FPGA. All data required for executing the computer program, and all data generated by executing the computer program, are stored in built-in memory such as ROM 103 and RAM 102, or in storage media such as hard disk drives or solid-state drives.

[0040] The functions of each functional unit of the control device 20 will be explained in detail below.

[0041] Image acquisition unit 201 irradiates an object F with energy rays and acquires an image of the energy rays that have passed through the object F. Specifically, image acquisition unit 201 generates an X-ray image based on digital signals (amplified signals) output from X-ray detection camera 10 (more specifically, output interfaces 17a and 17b). Image acquisition unit 201 generates an X-ray image by averaging or adding the two digital signals output from output interfaces 17a and 17b, and stores the generated X-ray image in image storage unit 207. Figure 4 This is a diagram showing an example of an X-ray image G1 acquired by the image acquisition unit 201.

[0042] The relationship calculation unit 202 calculates the relationship between the brightness value (hereinafter, also referred to as pixel value) and the statistical value of the brightness value in the X-ray image acquired by the image acquisition unit 201. In this embodiment, the noise evaluation value, which evaluates the extended noise value, and the noise sigma value, which is the standard deviation of the pixel value, can be used as the statistical value of the pixel value. Alternatively, the variance of the pixel value can also be used as the statistical value of the pixel value.

[0043] That is, the relation calculation unit 202 calculates first relation data representing the relationship between pixel value and standard deviation of pixel value through simulation. The relation between pixel value and standard deviation of pixel value used by the relation calculation unit 202 in the simulation is expressed by the following equations (1) to (4).

[0044] [Number 1]

[0045]

[0046] M = Mcoeff M (3)

[0047] F p =F p coeff Fp (4)

[0048] In equations (1) to (3) above, the variable Noise represents the standard deviation of the pixel value, the variable Signal represents the signal value (pixel value) of the pixel, the constant F represents the noise factor, and the variable E m To represent the average energy of the X-rays, the constant M represents the multiplication rate based on the scintillator, and the constant coeff... M To represent information about the multiplication rate adjustment coefficient used to adjust the multiplication rate based on the scintillator, the constant C represents information about the coupling efficiency between the line scan camera 12a and the scintillator 11a, or between the line scan camera 12b and the scintillator 11b in the X-ray inspection camera 10, and the constant Q represents information about the quantum efficiency of the line scan camera 12a or the line scan camera 12b.

[0049] In equation (1) above, the constant cf represents the conversion coefficient in line scan camera 12a or line scan camera 12b that converts the signal value of a pixel into charge, and the constant R represents the readout noise in line scan camera 12a or line scan camera 12b. The conversion coefficient cf and the readout noise R are determined by the gain setting in line scan camera 12a or line scan camera 12b.

[0050] In equation (1) above, the constant M Si The constant rate represents the multiplication rate of a line scan camera (silicon) when X-rays incident on scintillator 11a or scintillator 11b are not converted into visible light and are instead incident on line scan camera 12a or line scan camera 12b. si To represent information about the direct silicon generation rate, which represents the probability that X-rays incident on scintillator 11a or scintillator 11b will not be converted into visible light and will instead be incident on line scan camera 12a or line scan camera 12b, the constant C is used.S To represent the information of the shadow correction value, the constant offset represents the camera offset information, which represents the offset value of line scan cameras 12a and 12b. In the above equation (1), the constant N represents the information of the number of sensors. The constant N can be, for example, information representing the number of line scan cameras (line count), or information representing the merging setting in line scan camera 12a or line scan camera 12b. In the above equation (1), coeff noise This is the noise adjustment coefficient used to adjust the noise. In equations (2) and (4) above, the constant F... p To represent the information of the fuzzy coefficient, which represents fuzziness, the constant coeff... Fp This refers to information about the fuzzy coefficient adjustment coefficients used to adjust the coefficients representing fuzziness.

[0051] Here, when using equations (1) to (4) above, the relational calculation unit 202 substitutes the pixel values ​​of each pixel of the X-ray image acquired by the image acquisition unit 201 into the variable Signal, and sets other parameters in equations (1) to (4) above based on the input information received from the user by the input device 40. Then, the relational calculation unit 202 obtains the value of the standard deviation of the pixel values ​​of each pixel of the X-ray image as the variable Noise calculated using equations (1) to (4) above. Furthermore, the parameters in equations (1) to (4) above can be set based on the information received from the user, or they can be set by reading data such as RAM 102 or ROM 103 stored in the control device 20 based on the received information.

[0052] For example, the relational calculation unit 202 receives the following information input from the user.

[0053] • Determine camera model

[0054] Gain settings

[0055] · Number of merges, number of lines

[0056] • Average energy of X-rays E m

[0057] Camera offset

[0058] Silicon direct production rate si

[0059] •Shadow correction value C S

[0060] Noise adjustment factor coeff noise

[0061] • Multiplication rate adjustment factor coeff M

[0062] • Fuzzy coefficient adjustment coefficient coeff Fp

[0063] Then, based on the received information "measuring camera model", the relational calculation unit 202 reads from the control device 20 the scintillator multiplication rate M, coupling efficiency C, and ambiguity coefficient F stored corresponding to "measuring camera model". p Parameters such as quantum efficiency Q are also considered. For example, when the "camera model being measured" is "camera 1", the relational calculation unit 202 reads the parameters corresponding to "camera 1", such as scintillator multiplication rate M = 10, coupling efficiency C = 0.7, and ambiguity coefficient F. p With parameters of 0.9 and quantum efficiency Q = 0.9, when the "measuring camera model" is "camera 2", the relational calculation unit 202 reads the scintillator multiplication rate M = 100, coupling efficiency C = 0.8, and ambiguity coefficient F corresponding to "camera 2". p The parameters are 0.8 and quantum efficiency Q = 0.7. Similarly, based on the received information "gain setting", the relational calculation unit 202 reads parameters such as the conversion count cf and readout noise R stored in the control device 20 corresponding to the "gain setting". Based on the received information "number of merges, number of lines", it reads parameters such as the number of sensors N stored in the control device 20 corresponding to the "number of merges, number of lines". Furthermore, the relational calculation unit 202 sets other X-ray average energy E based on information received from the user. m Parameters such as these.

[0064] Furthermore, the relational calculation unit 202 can also set some of the other parameters in equations (1) to (4) above by performing calculations based on information received from the user. For example, the relational calculation unit 202 can obtain the average energy E by calculation. m As an example, the relational calculation unit 202 calculates the average energy related to the X-rays (radiation) that have passed through the object F based on the conditional information input by the user. The conditional information refers to either the conditions of the energy ray source or the imaging conditions when the object F is irradiated with energy rays. Furthermore, the relational calculation unit 202 can receive the conditional information input as direct input of numerical information, or as a selection input of numerical information pre-set in its internal memory. The relational calculation unit 202 can also receive the aforementioned conditional information input from the user and acquire a portion of the conditional information (such as tube voltage) based on the detection results of the control state based on the control device 20.

[0065] The condition information refers to, for example, the operating conditions of the X-ray irradiator (source) 50 when taking an X-ray image of an object F, or the imaging conditions based on the X-ray detection camera 10. Examples of operating conditions may include all or some of the following: the type of X-ray source, tube voltage, tube current, target angle, target material, etc. As conditional information indicating the shooting conditions, examples may include all or a portion of the following: the material, thickness, and density of filters 51 and 19 disposed between X-ray irradiator 50 and X-ray inspection camera 10; the distance (FDD) between X-ray irradiator 50 and X-ray inspection camera 10; the type and thickness of the window material of X-ray inspection camera 10; information related to scintillators 11a and 11b of X-ray inspection camera 10 (e.g., thickness, material, density, multiplication rate, surface reflectivity, diffusion coefficient, or absorption coefficient); X-ray inspection camera information (e.g., gain setting, circuit noise value, saturation charge, conversion coefficient value (e- / count), camera linear rate (Hz) or linear velocity (m / min)); and information about the object F (measuring material, thickness, density).

[0066] For example, the relational calculation unit 202 calculates the spectrum of X-rays detected by the X-ray detection camera 10 using approximations such as the well-known Tucker formula, based on information contained in the conditional information, including tube voltage, target angle, target material, material and thickness of filters 51 and 19 and their presence or absence, type of window material of the X-ray detection camera 10 and its presence or absence, and material and thickness of scintillators 11a and 11b of the X-ray detection camera 10. The relational calculation unit 202 further calculates the spectral intensity integral value and photon number integral value based on the X-ray spectrum, and calculates the average energy value of the X-rays by dividing the spectral intensity integral value by the photon number integral value. Furthermore, the calculation of the X-ray spectrum can also use approximations based on well-known Kramer or Birch formulas.

[0067] Furthermore, the relation calculation unit 202 can also derive first relational data representing the relationship between pixel values ​​and the standard deviation of pixel values ​​based on images obtained from actual imaging. As an example, the relation calculation unit 202 can also acquire an X-ray image (test image) taken by irradiating a fixture with X-rays. The fixture can be a flat plate-shaped component with known thickness and material. The relation calculation unit 202 can also derive relational data representing the relationship between pixel values ​​and the standard deviation of pixel values ​​from the acquired test image.

[0068] The derivation of first relational data, representing the relationship between pixel values ​​and standard deviations of pixel values, from an X-ray image of an imaging fixture, performed by the relational calculation unit 202, will be explained. The fixture may, for example, be a component whose thickness varies in a step-like manner along one direction. First, in a test image of the imaging fixture, the relational calculation unit 202 derives the pixel values ​​(hereinafter referred to as true pixel values) for each step of the fixture when there is no noise, and derives the standard deviation of the pixel values ​​based on the true pixel values. Specifically, the relational calculation unit 202 derives the average value of the pixel values ​​in a certain step of the fixture. Then, the relational calculation unit 202 sets the average value of the derived pixel values ​​as the true pixel value in that step. In that step, the relational calculation unit 202 derives the difference between each pixel value and the true pixel value as a noise value. The relational calculation unit 202 derives the standard deviation of the pixel values ​​from the noise value of each derived pixel value.

[0069] Then, the relation calculation unit 202 derives first relation data based on the relationship between the true pixel values ​​and the standard deviation of the pixel values. Specifically, the relation calculation unit 202 derives the true pixel values ​​and the standard deviation of the pixel values ​​for each step of the fixture. The relation calculation unit 202 plots the relationship between the derived true pixel values ​​and the standard deviation of the pixel values ​​into a graph, and derives a relation graph representing the relationship between the pixel values ​​and the standard deviation of the pixel values ​​by drawing an approximate curve. Furthermore, the approximate curve uses exponential approximation, linear approximation, logarithmic approximation, polynomial approximation, power approximation, etc.

[0070] The relationship calculation unit 202 can also use multiple X-ray images taken with varying brightness when the object is not present, instead of X-ray images taken with the fixture as the object, as the test images. In this case, multiple test images obtained by changing the output intensity of the X-ray irradiator 50 or by changing the exposure time in the X-ray detection camera 10 are used. Then, in the same order as described above, the relationship calculation unit 202 can derive a relationship chart showing the relationship between pixel values ​​and the standard deviation of pixel values.

[0071] The noise map generation unit 203 generates a noise map based on first relational data calculated by the relational calculation unit 202, which represents the relationship between the pixel value and the standard deviation of each pixel in the X-ray image. The noise map generation unit 203 generates data that corresponds the derived standard deviation to each pixel in the X-ray image.

[0072] The parameter setting unit 204 sets parameters for the edge preservation smoothing filter for each pixel based on the noise map generated by the noise map generation unit 203. In this embodiment, a bilateral filter can be used as the edge preservation smoothing filter. That is, the parameter setting unit 204 calculates 2×σ for each pixel based on the standard deviation σ corresponding to each pixel on the noise map. 2 Generate a value of 2×σ for each of the calculated pixels. 2 The data corresponding to each pixel is the parameter map.

[0073] The filter processing unit 205 takes the X-ray image stored in the image storage unit 207 as the processing target and performs noise removal processing using a bilateral filter with parameters calculated for each pixel by the parameter setting unit 204. Specifically, when (i, j) represents the position of each pixel in the X-ray image as a two-dimensional image (i, j: natural numbers), f(i, j) represents the pixel value of each pixel in the X-ray image being processed, g(i, j) represents the pixel value of each pixel in the output X-ray image, σ1 represents the control parameter for the spatial direction, and σ2 represents the control parameter for the brightness direction, the filter processing unit 205 performs noise removal processing using a bilateral filter by calculating the following formula (5) for all pixels of the X-ray image.

[0074] [Number 2]

[0075]

[0076] At this time, when the filter processing unit 205 calculates the pixel value g(i, j) of each pixel, it refers to the parameter map generated by the parameter setting unit 204 and sets the parameter 2×σ corresponding to each pixel on the parameter map. 2 Substitute the parameters into equation (5) above;

[0077] [Number 3]

[0078]

[0079] In this process, the pre-defined parameters are set to the spatial direction control parameter σ1 in the above formula (5), and the pixel value is calculated accordingly.

[0080] The filter processing unit 205 stores the X-ray image that has undergone noise removal processing as described above in the image storage unit 207. At this time, the filter processing unit 205 can either display the noise-removed X-ray image on the display device 30 or transmit the noise-removed X-ray image to an external device.

[0081] Next, the sequence of observation and processing of the energy ray transmitted image of the object F using the image acquisition device 1 of this embodiment, i.e., the flow of the image processing method of this embodiment, will be explained. Figure 5 This is a flowchart showing the sequence of observation processing performed by the image acquisition device 1.

[0082] First, in the control device 20, information for parameter setting is received from the user via the input device 40 (step S101, input step). Next, the object F is placed in the image acquisition device 1 and an X-ray image of the object F is captured using the control device 20 (step S102, acquisition step). Then, using the control device 20, based on the information received in step S101, first relational data representing the relationship between pixel values ​​and standard deviation is calculated (step S103: calculation step).

[0083] Next, using the control device 20, a noise map and a parameter map are generated based on the X-ray image and the first relational data acquired in step S102 (step S104: setting step). Then, using the control device 20, the parameters of a bilateral filter are set for each pixel based on the parameter map, and noise removal processing is performed on the X-ray image using the set bilateral filter (step S105: processing step). Finally, using the control device 20, the X-ray image after noise removal processing is stored (step S106).

[0084] In the image acquisition apparatus 1 of the first embodiment described above, control parameters suitable for the brightness value are set for each pixel of the X-ray image, and an edge-preserving smoothing filter reflecting the set control parameters for each pixel is used to process the X-ray image to be processed. This performs noise removal processing corresponding to the standard deviation of the brightness values ​​of each pixel in the X-ray image. As a result, noise in the overall image can be effectively removed while preserving the edges of the image.

[0085] Currently, bilateral filters, known as edge-preserving smoothing filters, are effective at removing Gaussian noise. A bilateral filter is a Gaussian filter used to apply weights to each local region. In this embodiment, by setting the control parameter σ2 of the brightness direction of the bilateral filter for each pixel based on the standard deviation of the pixel value, the weight value for each local region can be optimized.

[0086] In this embodiment, the first relational data is calculated through simulation. This allows for setting appropriate parameters for each pixel of the X-ray image, and enables the proper removal of noise from the overall image.

[0087] Furthermore, in this embodiment, first relational data is calculated based on information input by the user. Therefore, appropriate parameters can be set for each pixel of the X-ray image using the input information, and noise in the overall image can be appropriately removed.

[0088] Furthermore, in this embodiment, first relational data is also calculated based on the acquired test image. According to this process, by analyzing the test image, appropriate parameters can be set for each pixel of the X-ray image, and noise in the overall image can be appropriately removed.

[0089] Furthermore, in this embodiment, a bilateral filter can be used as the edge-preserving smoothing filter. In this case, by setting the control parameter for the brightness direction based on the standard deviation of the pixel value of each pixel, noise in the overall image can be appropriately removed.

[0090] exist Figure 6 and Figure 7 The image shown is an X-ray image acquired by image acquisition device 1 with an object F whose thickness varies in a stepped manner. Figure 6 In the image, section (a) shows the X-ray image before noise removal processing, and section (b) shows the X-ray image after noise removal processing. Additionally, in... Figure 7 In the diagram, sections (a) and (b) show X-ray images after noise removal processing when the control parameter σ2 in the brightness direction is set to a fixed value throughout the X-ray image. These results demonstrate that, according to this embodiment, noise removal is optimized across all thicknesses. Figure 6 (b) part). On the other hand, when the control parameter σ2 is set to a fixed value, a partially blurred image can be obtained. Figure 7 (a) or images where noise has not been adequately removed ( Figure 7 The result of (b) part.

[0091] [Second Implementation]

[0092] Figure 8 This is a block diagram illustrating the functional structure of the optical image processing system 1A, which is the image processing apparatus of the second embodiment. For example... Figure 8 As shown, the optical image processing system 1A is a system that acquires an optical image of an object F based on light L received from the object F. Light L can be, for example, light emitted by the object F, transmitted light from the object F, reflected light from the object F, or scattered light from the object F. Examples of the aforementioned light L include ultraviolet light, visible light, and infrared light. Hereinafter, this light will sometimes be referred to as observation light. The optical image processing system 1A includes a camera (imaging device, detection unit) 82, an optical image processing module 83, a display device 84, and an input device (input unit) 85.

[0093] Camera 82 acquires an optical image by capturing light L from an object F. Camera 82 includes a light detector 121 and an image control unit 122. The light detector 121 is an imaging element with multiple pixels. Examples of photodetectors 121 include CCD (Charge Coupled Device) image sensors, CMOS (Complementary Metal-Oxide Semiconductor) image sensors, photodiodes, InGaAs sensors, TDI (Time Delay Integration) – CCD image sensors, TDI – CMOS image sensors, imaging tubes, EM (Electron Multiplying) – CCD image sensors, EB (Electron Bombarded) – CMOS image sensors, SPAD (Single Photon Avalanche Diode), SPPC (Single-Pixel Photon Counter), MPPC (Multi-Pixel Photon Counter), SiPM (Silicon Photomultiplier), HPD (Hybrid PhotoDetector), APD (Avalanche Photodiode), and PMT (Photomultiplier Tube). Tube). Alternatively, the photodetector 121 can also be a photodetector that combines an image intensifier (II) or a micro-channel plate (MCP) with a CCD image sensor, a CMOS image sensor, etc. Examples of the shape of the photodetector 121 include a region sensor, a line sensor that acquires images by line scanning, a TDI sensor, and a point sensor that acquires images by two-dimensional scanning. The camera 82 captures the light L from the object F imaged by the imaging optical system 124 through the objective lens 123, and outputs a digital signal based on the image capture result to the image control unit 122.

[0094] The image control unit 122 performs image processing based on the digital signal received from the photodetector 121. The image control unit 122 may be configured as, for example, a CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field-Programmable Gate Array), ASIC (Application Specific Integrated Circuit), or SoC (System on a Chip). The image control unit 122 generates image data based on the digital signal received from the photodetector 121, applies prescribed image processing to the generated image data, and then outputs it to the optical image processing module 83.

[0095] The optical image processing module 83 is, for example, a computer such as a PC (Personal Computer). The optical image processing module 83 processes the image data output from the camera 82 to generate a noise-removed optical image. The optical image processing module 83 can be connected to the camera 82, the display device 84, and the input device (input unit) 85 via wired or wireless means, respectively. After the noise removal processing described later is performed, the generated optical image is output to the display device 84 and displayed by the display device 84. Based on user operations, various input information, such as the shooting conditions of the object F, is input from the input device 85 to the optical image processing module 83. Furthermore, the optical image processing module 83 controls the camera 82. In addition, the optical image processing module 83 of the second embodiment is a device independently installed outside the camera 82, but it can also be integrated into the internal structure of the camera 82. For example, the optical image processing module 83 can also be a module equivalent to a CPU or GPU installed in the camera's processing circuitry. Similar to the first embodiment, the optical image processing module 83 has... Figure 2 The hardware structure shown.

[0096] like Figure 8 As shown, the optical image processing module 83 includes an image acquisition unit 131, a relation calculation unit 132, a noise map generation unit 133, a parameter setting unit (image processing unit) 134, a filter processing unit (image processing unit) 135, and an image storage unit 136. Figure 8 The functional units of the optical image processing module 83 shown are implemented through the following: CPU 101, GPU 105, and RAM 102, etc. Figure 2The program (image processing program of the second embodiment) is read from the hardware of the optical image processing module 83, thereby causing the communication module 104 and input / output module 106 to operate under the control of the CPU 101 and GPU 105, and performing data reading and writing in the RAM 102. The CPU 101 and GPU 105 of the optical image processing module 83 execute this computer program, causing the optical image processing module 83 to function as... Figure 8 Each of the functional units shown in the diagram performs its function and sequentially executes the processing corresponding to the image processing method described later.

[0097] The functions of each functional unit of the optical image processing module 83 will be explained in detail below.

[0098] The image acquisition unit 131 acquires an optical image of the light emitted from the object F. Specifically, the image acquisition unit 131 acquires an optical image output from the camera 82. Then, the image acquisition unit 131 stores the acquired optical image in the image storage unit 136 each time. Figure 9 This is a diagram showing an example of an optical image G1 acquired by the image acquisition unit 131.

[0099] The relationship calculation unit 132 calculates the relationship between pixel values ​​and statistical values ​​of pixel values ​​in the optical image acquired by the image acquisition unit 131 as follows. In this embodiment, as in the first embodiment, the noise sigma value, which is the standard deviation of the pixel values, can be used as the statistical value of the pixel values. First, the relationship calculation unit 132 receives input from the user using the input device 85. The received information includes, for example, information about the type of camera 82 and information indicating the gain of the camera 82. Then, if the information about the type of camera 82 indicates a non-multiplier camera such as CMOS, CCD, PD, InGaAs, TDI CCD, TDI CMOS, or imaging tube, the relationship calculation unit 132 uses the following formula (6) to calculate first relationship data representing the relationship between pixel values ​​and the standard deviation of pixel values ​​by simulation.

[0100] [Number 4]

[0101]

[0102] In the above formula, Noise is the standard deviation of the pixel value, Signal is the signal value of the pixel (pixel value), Cf is the conversion coefficient in camera 82 that converts the signal value of the pixel into charge, D is the dark current noise in camera 82, and R is the readout noise in camera 82.

[0103] Here, when using the above formula (6), the relation calculation unit 132 substitutes the pixel values ​​of each pixel of the optical image acquired by the image acquisition unit 131 into the variable Signal, and sets other parameters in the above formula (6) based on the information received from the user by the input device 85. Then, the relation calculation unit 132 obtains the value of the standard deviation of the pixel values ​​of each pixel of the optical image as the variable Noise calculated using the above formula (6). Furthermore, the parameters in the above formula (6) can be set based on the information received from the user, or can be set by reading data such as RAM 102 or ROM 103 stored in the control device 20 based on the received information, or can be preset as fixed values ​​in the control device 20.

[0104] On the other hand, in the case of a camera that doubles the information of the type of camera 82, such as EMCCD, PMT, MPPC (SiPM), SSPD, or APD, the relationship calculation unit 132 uses the following formula (7) to calculate the first relationship data that represents the relationship between the pixel value and the standard deviation of the pixel value by simulation.

[0105] [Number 5]

[0106]

[0107] In the above formula, the constant F is the noise figure of the camera 82, and G is the gain of the camera 82. At this time, the relational calculation unit 132 sets the variables and constants in the above formula (7) in the same way as above. However, in the case of a camera that uses a photon counting method such as PMT to indicate the type of camera 82, the constant R is set to 0 in the above formula (7).

[0108] On the other hand, in the case of a camera that performs avalanche multiplication, such as an MPPC or APD, the information of the type of camera 82 is represented by the following formula (8), the relationship calculation unit 132 uses the following formula (8) to calculate the first relationship data representing the relationship between the pixel value and the standard deviation of the pixel value by simulation calculation.

[0109] [Number 6]

[0110] Noise = 2q(I L +I dg )BM 2 F+2qI ds B,

[0111] F=Mk+(2-1 / M)(1-k) (8)

[0112] In the above formula, the constant q is the charge of each electron, and the constant I... L The photocurrent is given by constant M = 1, and constant I is given by... dgThe current generated inside the substrate of camera 82 is given by constant B, which is the bandwidth; constant M is the multiplication rate; constant F is the excess noise coefficient; and constant I is... ds The surface leakage current is given by k, and the constant is the ionization ratio. In this case, the relationship calculation unit 132 sets the variables and constants in the above equation (8) in the same way as above. However, in the case of hole injection avalanche layer in camera 82, the constant k in the above equation (7) is set to 1 / k.

[0113] Furthermore, the relation calculation unit 132 can also derive first relational data representing the relationship between pixel values ​​and the standard deviation of pixel values ​​based on images obtained from actual shooting. As an example, the relation calculation unit 132 can also acquire an optical image (test image) captured by the camera 82 with a fixture as the object. The fixture can be a flat plate-shaped component with known density, etc. The relation calculation unit 132 can also derive relational data representing the relationship between pixel values ​​and the standard deviation of pixel values ​​from the acquired test image.

[0114] The derivation of first relational data, representing the relationship between pixel values ​​and the standard deviation of pixel values, from an optical image obtained by the relational calculation unit 132, will be described. The fixture may, for example, be a component whose concentration changes in a step-like manner along one direction. First, in a test image of the fixture, the relational calculation unit 132 derives the pixel values ​​(hereinafter referred to as true pixel values) for each step of the fixture when there is no noise, and derives the standard deviation of the pixel values ​​based on the true pixel values. Specifically, the relational calculation unit 132 derives the average value of the pixel values ​​in a certain step of the fixture. Then, the relational calculation unit 132 sets the average value of the derived pixel values ​​as the true pixel value in that step. The relational calculation unit 132 derives the difference between each pixel value and the true pixel value in that step as a noise value. The relational calculation unit 132 derives the standard deviation of the pixel values ​​from the noise value of each derived pixel value.

[0115] Then, the relation calculation unit 132 derives first relation data based on the relationship between the true pixel value and the standard deviation of the pixel value. Specifically, the relation calculation unit 132 derives the true pixel value and the standard deviation of the pixel value for each step of the fixture. The relation calculation unit 132 plots the relationship between the derived true pixel value and the standard deviation of the pixel value into a graph, and derives a relation graph representing the relationship between the pixel value and the standard deviation of the pixel value by drawing an approximate curve. In addition, the approximate curve uses exponential approximation, linear approximation, logarithmic approximation, polynomial approximation, power approximation, etc.

[0116] The relationship calculation unit 132 can also use multiple optical images taken with the fixture as the object, obtained by changing the brightness when the object is not present, instead of optical images taken with the fixture as the object. In this case, multiple test images obtained by changing the illumination intensity of the detection range of the camera 82, or multiple test images obtained by changing the exposure time in the camera 82, are used. Then, in the same order as described above, the relationship calculation unit 132 can derive a relationship chart showing the relationship between pixel values ​​and the standard deviation of pixel values.

[0117] Similar to the noise map generation unit 203 in the first embodiment, the noise map generation unit 133 generates a noise map based on the first relational data calculated by the relational calculation unit 132 and the optical image. Similar to the parameter setting unit 204 in the first embodiment, the parameter setting unit 134 generates a parameter map based on the noise map generated by the noise map generation unit 133. Similar to the filter processing unit 205 in the first embodiment, the filter processing unit 135 performs noise removal processing on the optical image using a bilateral filter that sets the parameters of each pixel on the parameter map generated by the parameter setting unit 134.

[0118] Next, the sequence of observation and processing of the optical image of the object F using the optical image processing system 1A of this embodiment, i.e., the flow of the image processing method of this embodiment, will be explained. Figure 10 This is a flowchart showing the sequence of observation processing performed by the optical image processing system 1A.

[0119] First, in the optical image processing module 83, information for parameter setting is received from the user via the input device 85 (step S201, input step). Next, the object F is placed in the optical image processing system 1A and photographed, and the optical image of the object F is acquired using the optical image processing module 83 (step S202, acquisition step). Next, using the optical image processing module 83, based on the information received in step S201, first relational data representing the relationship between pixel values ​​and standard deviation is calculated (step S203: calculation step).

[0120] Next, using the optical image processing module 83, a noise map and a parameter map are generated based on the optical image and the first relational data obtained in step S202 (step S204: setting step). Furthermore, using the optical image processing module 83, while setting the parameters of the bilateral filter for each pixel based on the parameter map, noise removal processing is performed on the optical image using the set bilateral filter (step S205: processing step). Finally, using the optical image processing module 83, the optical image that has undergone noise removal processing is stored (step S106).

[0121] The optical image processing system 1A of the second embodiment described above also performs noise removal processing corresponding to the standard deviation of the brightness values ​​of each pixel in the optical image. As a result, noise in the overall image can be effectively removed.

[0122] [Variation Example]

[0123] The various embodiments of the present invention have been described above, but the embodiments of the present invention are not limited to the above embodiments.

[0124] In the image acquisition apparatus 1 of the first embodiment and the optical image processing system 1A of the second embodiment described above, a bilateral filter is used as an edge-preserving smoothing filter when performing noise removal processing on an image. As a variation, in these embodiments, other edge-preserving smoothing filters, such as smoothing filters utilizing NLM filters (Non-Local Means Filter), wavelets, and TV filters (Total Variation Filter), may also be used.

[0125] A variation of the NLM filter used in the first embodiment will be described. The NLM filter is a non-local region averaging filter. If the surrounding pixels are similar to the pixel of interest, the weights are increased; if the surrounding pixels are not similar to the pixel of interest, the weights are decreased. When obtaining the pixel value of the pixel of interest, the weights are used to perform a weighted average of the surrounding pixels.

[0126] The filter processing unit 205 of the control device 20 calculates the weight parameters w(p,q) of the peripheral pixel q relative to the pixel of interest p used in the NLM filter using the following formula (9).

[0127] [Number 7]

[0128]

[0129] In equation (9) above, v(p) is a vector for the similarity of the pixel of interest, v(q) is a vector for the similarity of the surrounding pixels, and σ is the noise standard deviation of the image. This weight parameter w(p, q) is calculated such that it increases as the similarity between the surrounding pixels and the pixel of interest increases. When calculating the weight parameter w(p, q) for each pixel of interest in equation (9) above, the filter processing unit 205 refers to the parameters generated by the parameter setting unit 204. Figure 1 The edges will be mapped to the parameters 2×σ corresponding to each pixel of interest on the parametric graph. 2 Substituting "2σ" into equation (9) above 2Therefore, calculations are performed. Furthermore, the filter processing unit 205 uses an NLM filter that applies weight parameters w(p, q) calculated for each pixel of interest and each surrounding pixel to perform noise removal processing on the X-ray image.

[0130] In this variation, the parameters of the NLM filter are also optimized for each pixel using the standard deviation of the pixel values. As a result, noise can be effectively removed while preserving the edges of the image as a whole.

[0131] A variation of the first embodiment using a wavelet-based smoothing filter will be described. The noise removal process based on the wavelet-based smoothing filter is performed sequentially by wavelet transformation, shrinkage, and inverse wavelet transformation. In the shrinkage, the wavelet coefficients are thresholded; in the inverse wavelet transformation, the thresholded wavelet coefficients are used for the inverse wavelet transformation.

[0132] The filter processing unit 205 of the control device 20 performs noise removal processing on the X-ray image using a smoothing filter that utilizes wavelet parameters set for each pixel. During the shrinkage process, the filter processing unit 205 determines a threshold for each pixel based on the standard deviation value corresponding to each pixel on the noise map.

[0133] This variation utilizes the threshold used in wavelet smoothing filters, optimizing each pixel using the standard deviation of its pixel value. As a result, noise can be effectively removed while preserving the edges of the image as a whole.

[0134] In the above embodiments, the relationship between brightness values ​​and statistical values ​​can also be calculated through simulation in the calculation step. In the above embodiments, the relationship calculation unit can also calculate the relationship between brightness values ​​and statistical values ​​through simulation. Therefore, it is possible to use simulation to set appropriate parameters for each pixel of the image, and to appropriately remove noise from the overall image.

[0135] Furthermore, in the above embodiment, an input step may be included to receive information from the user, and in the calculation step, the relationship between the brightness value and the statistical value is calculated based on the input information. Additionally, in the above embodiment, an input unit may be included to receive information from the user, and a relationship calculation unit calculates the relationship between the brightness value and the statistical value based on the input information. According to the above structure, appropriate parameters can be set for each pixel of the image using the input information, and noise in the overall image can be appropriately removed.

[0136] In the above embodiment, a detection unit for detecting an image of the object to be processed may also be included. According to the above structure, noise in the image acquired within the image detection device can be effectively removed.

[0137] Furthermore, in the above embodiment, an additional acquisition step for acquiring a test image may be included, and in the calculation step, the relationship between the brightness value and the statistical value is calculated based on the acquired test image. Additionally, in the above embodiment, the relationship calculation unit may also calculate the relationship between the brightness value and the statistical value based on the measured test image. According to this structure, by analyzing the test image, appropriate parameters can be set for each pixel of the image, and noise in the overall image can be appropriately removed.

[0138] In the above embodiments, the statistical values ​​can be standard deviation or variance. Therefore, the parameters of the edge-preserving smoothing filter can be set to values ​​suitable for each pixel, effectively removing noise from the overall image.

[0139] In the above embodiments, the edge-preserving smoothing filter can be a bilateral filter or a non-local averaging filter. Therefore, by setting parameters based on statistical values ​​of the brightness of each pixel, noise in the overall image can be appropriately removed.

[0140] The image processing method of the embodiment is [1] "An image processing method comprising: a calculation step of calculating the relationship between a brightness value in an image and a statistical value of the brightness value; an acquisition step of acquiring an image of the object to be processed; a setting step of specifying the statistical value for each pixel of the image based on the brightness value of each pixel of the image and the relationship thereon, and setting parameters of an edge preservation smoothing filter for each pixel of the image based on the specified statistical value; and a processing step of processing the image using the edge preservation smoothing filter for which the parameters are set for each pixel of the image."

[0141] The image processing method of the implementation method may also be [2] "According to the image processing method described in [1] above, wherein, in the above calculation step, the relationship between the above brightness value and the above statistical value is calculated by simulation."

[0142] The image processing method of the implementation method may also be [3] "According to the image processing method described in [2] above, there is also an input step of receiving information input from the user, and in the above calculation step, the relationship between the above brightness value and the above statistical value is calculated based on the input information."

[0143] The image processing method of the implementation method may also be [4] "According to the image processing method described in [1] above, there is an additional acquisition step of acquiring a test image, and in the above calculation step, the relationship between the brightness value and the statistical value is calculated based on the acquired test image."

[0144] The image processing method of the implementation method may also be [5] "the image processing method described in any one of [1] to [4] above, wherein the above statistical value is the standard deviation or variance".

[0145] The image processing method of the embodiment may also be [6] "the image processing method described in any one of [1] to [5] above, wherein the edge preservation smoothing filter is a bilateral filter or a non-local averaging filter."

[0146] The image processing apparatus of the embodiment is [7] "An image processing apparatus comprising: a relation calculation unit that calculates the relationship between a brightness value in an image and a statistical value of the brightness value; and an image processing unit that processes an image of a processing object, wherein the image processing unit specifies the statistical value for each pixel of the image based on the brightness value of each pixel of the image and the relationship, sets parameters of an edge preservation smoothing filter for each pixel of the image based on the specified statistical value, and processes the image using the edge preservation smoothing filter for each pixel of the image with the parameters set thereon."

[0147] The image processing apparatus of the embodiment may also be [8] "the image processing apparatus described in [7] above, wherein the relationship calculation unit calculates the relationship between the brightness value and the statistical value by simulation."

[0148] The image processing apparatus of the embodiment may also be [9] "The image processing apparatus described in [8] above, wherein it further includes an input unit that receives information input from a user, and the relationship calculation unit calculates the relationship between the brightness value and the statistical value based on the input information."

[0149] The image processing apparatus of the embodiment may also be

[10] "the image processing apparatus according to any one of [7] to [9] above, wherein it further includes a detection unit for detecting the image of the object to be processed".

[0150] The image processing apparatus of the embodiment may also be

[11] "the image processing apparatus described in [7] above, wherein the relationship calculation unit calculates the relationship between the brightness value and the statistical value based on the measured test image."

[0151] The image processing apparatus of the embodiment may also be

[12] "the image processing apparatus described in any one of [7] to

[11] above, wherein the above statistical values ​​are standard deviation or variance".

[0152] The image processing apparatus of the embodiment may also be

[13] "the image processing apparatus according to any one of [7] to

[12] above, wherein the edge preservation smoothing filter is a bilateral filter or a non-local averaging filter."

[0153] Explanation of symbols

[0154] 1…Image acquisition device, 1A…Optical image processing system, 20…Control device, 40, 85…Input device (input unit), 10…X-ray inspection camera (inspection unit), 82…Camera (inspection unit), 83…Optical image processing module, 132, 202…Relationship calculation unit, 134, 204…Parameter setting unit (image processing unit), 135, 205…Filter processing unit (image processing unit), F…Object.

Claims

1. An image processing method, wherein, have: The calculation steps for determining the relationship between brightness values ​​in an image and statistical values ​​of those brightness values; The steps for acquiring the image of the object to be processed; Based on the brightness value of each pixel in the image and the relationship, a specific statistical value is assigned to each pixel of the image, and based on the assigned statistical value, a parameter setting step is performed for each pixel of the image to set the parameters of the edge preservation smoothing filter. as well as The processing steps involve processing the image using the edge-preserving smoothing filter with parameters set for each pixel of the image.

2. The image processing method according to claim 1, wherein, In the calculation step, the relationship between the brightness value and the statistical value is calculated by simulation.

3. The image processing method according to claim 2, wherein, It also has an input step for receiving information from the user. In the calculation step, the relationship between the brightness value and the statistical value is calculated based on the input information.

4. The image processing method according to claim 1, wherein, It also has an additional step for acquiring test images. In the calculation step, the relationship between the brightness value and the statistical value is calculated based on the acquired test image.

5. The image processing method according to any one of claims 1 to 4, wherein, The statistical values ​​are standard deviation or variance.

6. The image processing method according to any one of claims 1 to 5, wherein, The edge-preserving smoothing filter is a bilateral filter or a non-local averaging filter.

7. An image processing apparatus, wherein, have: The relationship calculation unit calculates the relationship between the brightness values ​​in the image and the statistical values ​​of those brightness values; and The image processing unit processes the image of the target object. The image processing unit, Based on the brightness value of each pixel in the image and the relationship, specific statistical values ​​are assigned to each pixel of the image. Based on these specific statistical values, parameters for an edge-preserving smoothing filter are set for each pixel of the image. The image is processed using the edge-preserving smoothing filter, which sets the parameters for each pixel of the image.

8. The image processing apparatus according to claim 7, wherein, The relationship calculation unit calculates the relationship between the brightness value and the statistical value through simulation.

9. The image processing apparatus according to claim 8, wherein, It also has an input section for receiving information from the user. The relationship calculation unit calculates the relationship between the brightness value and the statistical value based on the input information.

10. The image processing apparatus according to any one of claims 7 to 9, wherein, It also includes a detection unit for detecting the image of the object to be processed.

11. The image processing apparatus according to claim 7, wherein, The relationship calculation unit calculates the relationship between the brightness value and the statistical value based on the measured test image.

12. The image processing apparatus according to any one of claims 7 to 11, wherein, The statistical values ​​are standard deviation or variance.

13. The image processing apparatus according to any one of claims 7 to 12, wherein, The edge-preserving smoothing filter is a bilateral filter or a non-local averaging filter.

14. An image processing program, wherein, This enables the processor to function as both a relational computing unit and an image processing unit. The relationship calculation unit calculates the relationship between the brightness values ​​in the image and the statistical values ​​of those brightness values. The image processing unit processes the image of the object to be processed. The image processing unit, Based on the brightness value of each pixel in the image and the relationship, specific statistical values ​​are assigned to each pixel of the image. Based on these specific statistical values, parameters for an edge-preserving smoothing filter are set for each pixel of the image. The image is processed using the edge-preserving smoothing filter, which sets the parameters for each pixel of the image.

Images