MTF measurement system and MTF measurement method

Abstract

To provide an MTF measurement system for supporting MTF measurement of a three-dimensional display in an aerial image reproducing system.SOLUTION: An MTF measurement system 1 includes: a measurement camera 10; a light-transmitting diffusion plate 20 having a first surface 21 as a diffusion surface and a second surface 22 opposing the first surface 21; a knife edge target 40 having a knife edge as a sharp side in one edge section of a plate-like body and detachably adhered to the diffusion plate 20; and an MTF measuring device 30 for photographing the knife edge via the diffusion plate 20 by the measurement camera 10 in a state in which one surface of the knife edge target 40 is adhered to the first surface 21 of the diffusion plate 20 and performing image processing of image data of an obtained side image. The MTF measuring device 30 obtains a profile of an ESF as a side image distribution function from the image data, differentiates a result obtained by performing ensemble averaging the profile of the ESF, and calculates the MTF by obtaining its Fourier transform.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an MTF measurement system and an MTF measurement method, and more particularly to an MTF measurement system and an MTF measurement method that support MTF measurement of a spatial image reproduction type three-dimensional display. [Background technology]

[0002] So far, various methods have been proposed to measure the display spatial frequency characteristics of two-dimensional displays, i.e., the Modulation Transfer Function (MTF). For example, a common method is to display sine wave patterns or square wave patterns of various spatial frequencies on the display and capture the patterns with a measuring camera (hereafter referred to as the measuring camera) to obtain the MTF.

[0003] Recently, a method has been proposed in which a straight line pattern is displayed on a display, the straight line is photographed with a measurement camera to obtain a luminance profile, the profile is ensemble-averaged, the function is differentiated, and the function obtained by the differentiation is Fourier-transformed to obtain the MTF (see Patent Document 1 and Non-Patent Document 1). In the method disclosed in Patent Document 1 and Non-Patent Document 1, a video signal for displaying the narrowest straight line (line spread function (LSF)) that can be displayed on the display is input to the display, and the narrowest straight line (LSF) is photographed with a measurement camera. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-107804 [Non-patent literature]

[0005] [Non-Patent Document 1] K. Masaoka, "Line-Based Modulation Transfer Function Measurement of Pixelated Displays," in IEEE Access, vol. 8, pp. 196351-196362, 2020, doi: 10.1109 / ACCESS.2020.3033756. Summary of the Invention [Problem to be solved by the invention]

[0006] The methods disclosed in Patent Document 1 and Non-Patent Document 1 are methods for measuring the MTF of a two-dimensional display, and measuring the MTF of a three-dimensional display was not anticipated, leaving room for improvement. However, for example, in a two-viewpoint system (two-eye system) or a multi-viewpoint system (multiple-eye system), in a limited case where only the light rays from one eye (one viewpoint) enter the measurement camera, that is, when the light density is sufficiently low, it is also possible to measure the image for each viewpoint using a method similar to the method for measuring the MTF of a two-dimensional display.

[0007] However, depending on the 3D display method, there are cases where the measurement method for MTF of 2D displays cannot be used as is. Specifically, in spatial image reproduction 3D display methods such as a multi-viewpoint method with high light density, known as a light field method, and holography, the number of viewpoints is huge, which results in a long measurement time, and it is difficult to measure the image for each viewpoint, making this a non-realistic measurement method.

[0008] The present invention has been made in consideration of the above problems, and an object of the present invention is to provide an MTF measurement system and an MTF measurement method that support the MTF measurement of a spatial image reproduction type three-dimensional display. [Means for solving the problem]

[0009] In order to solve the above-mentioned problems, the MTF measurement system of the present invention includes a measurement camera, a light-transmitting diffuser plate having a first surface which is a diffusing surface and a second surface opposite to the first surface, a knife-edge target having a sharp edge on one edge of a plate-shaped body and the plate-shaped body being in close contact with the diffuser plate in a removably manner, and an MTF measurement device which processes image data of an edge image obtained by photographing the knife edge via the diffuser plate with the measurement camera while one surface of the knife-edge target is in close contact with the first surface of the diffuser plate, and the MTF measurement device obtains a profile of an ESF which is an edge spread function from the image data, differentiates the result obtained by ensemble averaging the ESF profile, and calculates the Fourier transform of the result to calculate the MTF.

[0010] Moreover, an MTF measurement method according to the present invention includes a preparation step of closely contacting a knife-edge target, which has a knife-edge that is a sharp edge at one edge portion of a plate-shaped body, to a first surface of a light-transmitting diffuser plate having a first surface that is a diffusing surface and a second surface opposite the first surface; an imaging step of photographing the knife-edge through the diffuser plate with a measurement camera to obtain image data of an edge image; and a processing step of processing the image data with an MTF measurement device, in which in the processing step, the MTF measurement device obtains a profile of an ESF, which is an edge spread function, from the image data, differentiates the result obtained by ensemble averaging the ESF profiles, and calculates the Fourier transform of the result to calculate the MTF. Effect of the Invention

[0011] The present invention provides the following excellent effects. The MTF measurement system can process the image data of the edge image obtained by photographing the knife edge of the knife edge target placed in close contact with the diffusing surface of the diffuser. This allows the MTF measurement system to calculate the MTF of the entire system including the measurement camera and the diffuser, thereby supporting the MTF measurement of aerial image reproduction type 3D displays. Furthermore, according to the MTF measurement method, it is not necessary to measure the image for each viewpoint as in the method for measuring the MTF of a two-dimensional display, and it is possible to assist in the measurement of the MTF of a spatial image reproduction type three-dimensional display in a short time. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram showing the configuration of an MTF measurement system according to a first embodiment. [Diagram 2] 2A and 2B are top views of the diffuser plate shown in FIG. 1, in which (a) shows the state in which the knife-edge target is attached and (b) shows the state in which the knife-edge target is removed. [Diagram 3] FIG. 2 is a schematic diagram showing a configuration example of the jig shown in FIG. [Figure 4] FIG. 13 is a schematic diagram showing a region of interest set in a captured image of a knife edge. [Diagram 5] FIG. 13 is a schematic diagram illustrating the process of binning pixels of a region of interest from a captured image that has been rotated so that its edges are upright. [Figure 6] 2A to 2C are graphs showing functions calculated by the MTF measuring device shown in FIG. 1, where (a) shows ESF, (b) shows LSF, and (c) shows MTF. [Figure 7] 2 is a schematic diagram showing how the MTF of a three-dimensional display is measured by the MTF measurement system of FIG. 1. [Figure 8] FIG. 8 is an enlarged view of the reconstructed stereoscopic image and the diffuser plate in FIG. [Figure 9] FIG. 8 is a schematic diagram showing a calculation process for determining an LSF from photographed images of the reconstructed stereoscopic image in FIG. 7. [Figure 10] 1A and 1B are schematic diagrams showing how a reconstructed stereoscopic image is captured, in which (a) shows a case where a diffuser plate is not used, and (b) shows a case where a diffuser plate is used. [Figure 11] FIG. 13 is a schematic diagram showing a modified example of the jig. [Figure 12] FIG. 13 is a schematic diagram showing another modified example of the jig. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] (First embodiment) The configuration of an MTF measurement system according to a first embodiment of the present invention will be described with reference to the drawings. In Fig. 1, the horizontal direction is the x-axis, the vertical direction is the y-axis, and the optical axis direction of the measurement camera 10 is the z-axis. The horizontal direction of the pixel array of the image sensor of the measurement camera 10 is the x-axis. C axis (≠ x-axis), and the vertical direction of the pixel array of the image sensor is the Y C axis (≠ y-axis).

[0014] [Outline of MTF measurement system] The MTF measurement system 1 shown in FIG. 1 includes a measurement camera 10, a diffusion plate 20, an MTF measurement device 30, and a knife-edge target 40. The diffusion plate 20 has a light-transmitting property and has a first surface 21 which is a diffusion surface, and a second surface 22 opposed to the first surface 21 (see FIG. 2(a)). The knife-edge target 40 has a plate-shaped body with a sharp edge, that is, a knife edge, at one edge thereof, and the plate-shaped body is detachably attached to the diffusion plate 20 . The MTF measuring device 30 processes image data of an edge image obtained by photographing the knife edge through the diffuser 20 with the measurement camera 10 while one side of the knife-edge target 40 is in close contact with the first surface 21 of the diffuser 20 as shown in Figure 2(a). MTF measurement device 30 obtains a profile of an ESF (Edge Spread Function), which is an edge spread function, from image data, differentiates the result obtained by ensemble averaging the ESF profiles, and calculates the MTF by obtaining a Fourier transform of the result.

[0015] [Two measurement phases] The MTF measurement system 1 can perform, at different times, a measurement phase P1 that is mainly related to the MTF of the measurement camera as shown in Fig. 2(a) and a measurement phase P2 that is mainly related to the MTF of the three-dimensional display as shown in Fig. 2(b). The measurement phase P1 corresponds to a preparation stage for performing the measurement phase P2. The details of the measurement phases P1 and P2 will be described later. The configuration of each part of the MTF measurement system 1 will be described below.

[0016] (Measuring camera) The measurement camera 10 is a camera for measurement, and is a camera that can turn off gamma correction and set γ = 1, such as a digital camera. The output signal indicating the knife-edge image captured by the measurement camera 10 corresponds one-to-one with each pixel.

[0017] The measurement camera 10 includes an imaging element with a square lattice pixel arrangement. The square lattice pixel arrangement is a pixel structure arranged in a square on a two-dimensional plane. In this pixel structure, for example, the vertical positions of pixels arranged in odd-numbered columns and the vertical positions of pixels arranged in even-numbered columns are aligned on the same straight line. In the case of a pixel structure arranged in a delta array on a two-dimensional plane, for example, the vertical positions of pixels arranged in odd-numbered columns and the vertical positions of pixels arranged in even-numbered columns are shifted by 1 / 2 the pixel interval. The optical axis 15 of the measurement camera 10 is set so as to be parallel to a direction perpendicular to the second surface 22 of the diffusion plate 20. The measurement camera 10 is placed in a state rotated at a predetermined angle around the optical axis 15 so that the knife edge is oblique to the pixel array of the imaging element. The rotation angle θ at this time is any angle (however, excluding angles near 0, 45, and 90 degrees). The measurement camera 10 is installed so as to be able to capture an image with an angle of view sufficient to cover the entirety of the diffusion plate 20 used for the measurement.

[0018] (Diffusion plate) The diffusion plate 20 has a translucency that diffuses and emits incident light. The diffusion plate 20 is arranged such that the first surface 21, which is a diffusion surface, faces the light source device 60, and the second surface 22, which is a flat surface, faces the measurement camera 10. The diffusion plate 20 diffuses and receives the light emitted from the light source device 60, and emits the diffused light from the second surface 22. The size of the diffusion plate 20 is, for example, approximately the same as the size of the three-dimensional display to be measured. The diffusion plate 20 may be a diffusion plate having a circular linear diffusion characteristic. For example, the diffusion plate 20 may be a holographic diffuser. When the diffusion plate 20 has a diffusion surface on only one side as shown in FIG. 2(a), the diffusion plate 20 is arranged so that the surface (first surface 21) opposite to the surface (second surface 22) facing the measurement camera 10 becomes a diffusion surface, and the knife-edge target 40 (real object) is placed in close contact with the diffusion surface. It should be noted that both the first surface 21 and the second surface 22 of the diffusion plate 20 may be diffusion surfaces.

[0019] (Light source device) The light source device 60 illuminates the knife edge. As shown in FIG. 2(a), the light source device 60 is disposed on the opposite side of the measurement camera 10 with respect to the diffusion plate 20 so as to face the first surface 21 of the diffusion plate 20. The light source device 60 may be any device that illuminates the knife edge with uniform gray or white light, and may be, for example, a two-dimensional display. The light source device 60 is mainly used only during the measurement phase P1 related to the MTF of the measurement camera. As a modification of the light source device 60, a three-dimensional display that is the measurement target may be used. In this case, the three-dimensional display may be used to display a uniform gray or white image to illuminate the knife edge only during the measurement phase P1. Here, saturation is not necessary, so the RGB values ​​are kept the same, and only the brightness is changed, taking into consideration the fact that it can be adjusted and displayed as is convenient for taking pictures with a camera, and is expressed as a uniform gray or white color. Note that in this specification, white refers to, for example, 8-bit gradation with RGB values ​​of 255, 255, 255, and gray is 127, 127, 127, which is the brightness lowered from white, and gray is a dark white.

[0020] (Knife-edge target) The knife-edge target 40 is an object having a sharp edge (knife edge) at one edge of a plate-shaped body. The knife-edge target 40 is sharply polished, and the knife-edge image is finer and more uniform than the pixel structure (grainy structure) of the display. The fineness of the knife-edge image is sufficiently finer than the fineness (maximum spatial frequency) of the three-dimensional image presented at a specified depth position by the three-dimensional display in the measurement phase P2 described later. The knife edge needs to be smooth and uniform. Specifically, the tip of a saw is not suitable, and the straight part of the tip of a kitchen knife or other knife can be used.

[0021] The knife-edge target 40 is necessary only during the measurement phase P1, which is mainly related to the MTF of the measurement camera. The knife-edge target 40 is attached to the diffuser plate 20 in the measurement phase P1, and after measurements are made mainly related to the MTF of the measurement camera, the knife-edge target 40 is removed from the diffuser plate 20, and then the measurement phase P2 is performed. In this manner, in the MTF measurement system 1, the knife-edge target 40 is used by being detached from the diffuser plate 20. The knife-edge target 40 may be attached to the diffuser plate 20 using a fastener such as a screw or a bolt, or may be temporarily attached and detached using an adhesive tape, double-sided tape, or the like.

[0022] (jig) The MTF measurement system 1 may include a jig 50 that connects the measurement camera 10 and the diffuser plate 20. The jig 50 holds the position and posture of the measurement camera 10 and the diffuser plate 20, and holds the distance L between the measurement camera 10 and the diffuser plate 20 at a constant value. The distance L may be any distance that allows the required shooting angle and normal focus adjustment. With the jig 50, the distance L during the measurement phase P2 in the series of measurements becomes the same as the distance L during the measurement phase P1, which is the preparation stage, so that the effort of re-measuring the distance between the measurement camera 10 and the diffuser plate 20 during the measurement phase P2 can be eliminated. In addition, by using the jig 50, it is possible to prevent the posture of the diffuser plate 20 from changing due to vibration or the like when the measurement phase P1 in the series of measurements is completed and the knife-edge target 40 is removed from the diffuser plate 20.

[0023] FIG. 3 is a schematic diagram showing a configuration example of the jig shown in FIG. 1. As shown in FIG. 3, the jig 50 includes a base 51, a first base 52, and a second base 53. The first base 52 is fixed on the base 51 and mounts the measurement camera 10, and the second base 53 is fixed on the base 51 at a distance from the first base 52 and mounts the diffusion plate 20. Here, the first base 52 includes a support 52a arranged on the base 51, an angle adjustment unit 52b arranged on the upper part of the support 52a, and a camera mounting unit 52c having a screw portion 52d provided on the angle adjustment unit 52b. Therefore, the measurement camera 10 is configured to be mounted on the first base 52 so that the female screw portion 10d of the measurement camera 10 is screwed into the screw portion 52d of the camera mounting unit 52c to be detachably attached, and the measurement camera 10 can be adjusted to a predetermined angle by the angle adjustment unit 52b and maintained in a certain position. The second base 53 includes a support 53a disposed at a preset distance from the first base 52, and a diffusion plate mounting portion 52c having a mounting groove 53b provided on the upper portion of the support 53a. Therefore, the diffusion plate 20 is configured so that it can be mounted on the second base 53 while maintaining a posture in which it is sandwiched in the mounting groove 53b of the diffusion plate mounting portion 52c. The diffusion plate mounting portion 52c is configured of an elastic member having a groove width narrower than the thickness of the diffusion plate 20, for example, and the diffusion plate 20 is fixed by sandwiching the diffusion plate 20 by pushing the groove width wide against the elastic force. In this example, the base 51 is a plate-like member made of metal or the like. The first base 52 and the second base 53 may be columnar members (poles) made of metal or the like as shown in FIG. 1. However, the configuration of the jig 50 is not limited to these. The measurement camera 10 may be fixed to the first base 52, for example, via a clamp member that clamps the measurement camera 10 so that the angle can be freely adjusted. The method of fixing the diffusion plate 20 to the second base 53 may be the same as or different from the method of fixing the measurement camera 10 to the first base 52.

[0024] In the MTF measurement system 1, the measurement camera 10 photographs the knife edge by focusing on the diffuser plate 20. At this time, light rays coming from multiple directions are incident on an arbitrary point on the diffuser plate 20, so the MTF measured by analyzing the profile of the knife edge in close contact with the diffuser plate 20 photographed by the measurement camera 10 can reflect the state in which the light rays coming from multiple directions are multiplexed by the diffuser plate 20.

[0025] (MTF measuring device) The MTF measurement device 30 includes a central processing unit (CPU), a random access memory (RAM), a read only memory (ROM), a hard disk drive (HDD), and an input / output interface, and is configured, for example, as a general personal computer. The display device (monitor) of the MTF measurement device 30 is configured, for example, as a liquid crystal display, and windows, icons, buttons, etc. are displayed on the monitor screen by a graphical user interface (GUI), and an operator (user) can select them using an input means such as a mouse. The MTF measurement device 30 is connected to the measurement camera 10 so as to be able to communicate with it, for example, by a wired (or wireless) connection such as a universal serial bus (USB).

[0026] [MTF measurement method] (Measurement Phase P1) The MTF measurement method will be described with reference to FIG. 1 and FIG. 2(a). The MTF measurement method includes a preparation step, an imaging step, and a processing step in a measurement phase P1. The preparation process is a process of closely adhering a knife-edge target 40 having a sharp edge, that is, a knife edge, at one edge end of a plate-shaped body to the first surface 21, which is a diffusion surface, of a light-transmitting diffusion plate 20 having a first surface 21 which is a diffusion surface and a second surface 22 opposite the first surface 21. The photographing step is a step of photographing the knife edge through the diffusion plate 20 by the measurement camera 10 to obtain image data of the edge image. The processing step is a step in which the image data is processed by the MTF measurement device 30 . In the processing step, MTF measurement device 30 obtains a profile of the ESF, which is an edge spread function, from the image data, differentiates the result obtained by ensemble averaging the ESF profiles, and calculates the MTF by obtaining a Fourier transform of the result.

[0027] In the preparation step, the knife edge of the knife edge target 40 is brought into close contact with the diffusion plate 20 as shown in FIG. 2(a). The measurement camera 10 preferably includes an imaging element with a square lattice pixel arrangement. In this case, in the preparation step, the optical axis 15 of the measurement camera 10 is set so as to be parallel to a direction perpendicular to the second surface 22 of the diffusion plate 20 (so that the z-axis is perpendicular to the diffusion plate 20). Then, the measurement camera 10 is placed in a state rotated at a predetermined angle around the optical axis 15 so that the knife edge is inclined with respect to the pixel arrangement of the imaging element. In this preparation step, it is preferable to use a jig 50 that keeps the distance L between the measurement camera 10 and the diffusion plate 20 at a constant value as shown in FIG. 1. The measurement camera 10 and the diffusion plate 20 are connected by the jig 50, which makes it easier to carry out a series of measurements.

[0028] In the photographing process, the light source device 60 illuminates the knife edge. In the example shown in Fig. 1, the light source device 60 displays a uniform white image in a rectangular area at the center of the display surface, and a black image around the rectangular area. The uniform white image displayed by the light source device 60 becomes the illumination light that illuminates the knife edge. Note that Fig. 2(a) shows a schematic diagram of the white image W displayed by the light source device 60 in the rectangular area at the center of the display surface, and the black image B displayed around the rectangular area.

[0029] In the processing step, the calculation process in which the MTF measurement device 30 calculates the MTF from the image data obtained by the measurement camera 10 is similar to the calculation process disclosed in Patent Document 1 and Non-Patent Document 1. Here, the methods disclosed in Patent Document 1 and Non-Patent Document 1 will be outlined in comparison with this embodiment. In the methods disclosed in Patent Document 1 and Non-Patent Document 1, a mechanism for generating a video signal and a display (two-dimensional display) for displaying the video signal perform filtering (sampling, quantization, etc.) required for display. Then, an algorithm (hereinafter referred to as a "signal display conversion algorithm") built into the display or its driver converts the video signal input to display a straight line into a specific pixel arrangement and pixel brightness of the display, and displays the straight line. The displayed result is a function that represents a straight line with some kind of brightness distribution in the width direction, i.e., an LSF with some kind of brightness distribution in the width direction. For example, if a straight line with a width of 0 is obtained as the LSF, the function resulting from ensemble averaging of that profile can be considered to be similar to a delta function. Therefore, if the function resulting from ensemble averaging of that profile is normalized with the DC intensity set to 1 and then Fourier transformed, an MTF is obtained that has characteristics such that the modulation depth at all frequencies from DC to the Nyquist frequency is equal to 1. The MTF obtained in this case is the MTF of a display that can display a virtual straight line with no width.

[0030] It is actually impossible to display a straight line without width. Therefore, in practice, for an LSF with a luminance distribution in the width direction, an appropriate region of interest (ROI) is arbitrarily determined, the ensemble average of the profile is calculated, and the MTF is calculated by Fourier transforming it. This measurement method is performed in accordance with the camera measurement method specified in "ISO 12233:2017 Photography-Electronic Still Picture Imaging-Resolution and Spatial Frequency Responses". This specification is intended to measure the MTF of a digital camera. As mentioned above, it is impossible to display a straight line that does not have a physical width on a display. Similarly, when measuring the MTF of a camera, it is impossible to use a straight line that does not have a physical width as a subject. An ideal thin "line" does not exist as a subject, but a knife edge (the right side of the knife edge target 40 in FIG. 1) does exist as an ideal "edge". In this method of measuring the MTF of a camera, a knife edge is photographed by a camera, an edge spread function (ESF) is obtained as an output signal indicating the knife edge image, this ESF is differentiated to obtain an LSF, and the MTF of the camera is obtained by Fourier transform of the LSF. The MTF measurement of displays disclosed in Patent Document 1 and Non-Patent Document 1 is an arrangement based on this method of measuring the MTF of a camera.

[0031] However, conventional methods for measuring the MTF of a two-dimensional display, including the methods disclosed in Patent Document 1 and Non-Patent Document 1, do not use a diffuser 20. On the other hand, in this embodiment, as shown in Fig. 1, the measurement camera 10 photographs the diffuser 20 with a knife edge in close contact with the first surface 21, which is the diffusing surface of the diffuser 20. Therefore, even if the calculation process performed by the MTF measuring device 30 to calculate the MTF from the image data obtained by the measurement camera 10 is the same as the calculation process disclosed in Patent Document 1 and Non-Patent Document 1, the obtained MTF will be completely different. The calculation process performed by the MTF measurement device 30 during the measurement phase P1 will now be described.

[0032] Measurement phase P1 is a preparation stage for measuring the MTF of a display. In measurement phase P1, as shown in Fig. 1 and Fig. 2(a), a knife edge is photographed by measurement camera 10, and the knife edge image is measured. At this time, measurement camera 10 photographs the knife edge placed in close contact with first surface 21, which is the diffusion surface of diffuser plate 20, from the diffuser plate 20 side.

[0033] FIG. 4 is a schematic diagram of an image generated by the output signal of the measuring camera 10 as a measurement result by the measuring camera 10. Here, an outer frame image area of ​​16 pixels horizontal by 21 pixels vertical, which is an extremely small part of the image of the output signal, is illustrated, and the types of luminance are simplified to three types, white, black, and neutral color pixels. The area where neutral color pixels are mainly arranged shows the knife-edge image. The operator (user) of the MTF measuring device 30 determines an appropriate region of interest (ROI) for the measurement result by the measuring camera 10 (FIG. 4). At this time, the user operates the MTF measuring device 30 to display the captured image of the knife edge on the monitor of the MTF measuring device 30 and observe it. Then, the user performs an operation to select a window, icon, button, etc. displayed by the GUI on the monitor using an input means such as a mouse. In this way, the user performs an operation to select a desired rectangular image area that is thought to include a sharp knife-edge image of the end of the knife-edge target 40 (the right end of the knife-edge target 40 in FIG. 1). In this example, the ROI is an area enclosed by a thick rectangular frame of 12 pixels horizontal by 19 pixels vertical inside the outer frame image area shown in FIG.

[0034] After determining the ROI, the user performs an operation of selecting a corresponding button or the like using input means such as a mouse in order to execute arithmetic processing for MTF measurement in MTF measurement device 30. The arithmetic processing by MTF measurement device 30 at this time is similar to the arithmetic processing disclosed in Patent Document 1 and Non-Patent Document 1, and an outline of this processing will be described with reference to FIG.

[0035] FIG. 5 is a diagram showing a state where the MTF measuring device 30 arranges the measurement results (FIG. 4) by the measuring camera 10 at an arbitrary angle in the virtual memory space. The angle at this time corresponds to the angle when the knife edge is rotated so as to stand upright, that is, the rotation angle θ of the measuring camera 10. In FIG. 5, the pixels shown in solid black in FIG. 4 are replaced with pixels with dot hatching to make them easier to see. Also, in FIG. 5, the white, black, and neutral color pixels shown in FIG. 4 are replaced for the purpose of explanation, and circles (pixel sampling points) filled with white, black, or neutral colors are shown as the pixel values ​​(luminance values) of the pixels. The MTF measuring device 30 performs a process in the virtual memory space where the sampling points (pixel values) of those pixels are dropped vertically downward in FIG. 5 as if following gravity, as shown by the long arrows hatched in FIG. 5. At this time, as shown by the tip of the short white arrow in Fig. 5, the sampling point of the pixel is input into one of many bins (containers) arranged in a horizontal row at small intervals. The MTF measurement device 30 resamples (called binning) the image data into many bins by this vertical drop process. Here, the resampling interval (bin interval) is at least narrower than the pixel interval of the measurement camera 10, and the narrower the better. This process makes it possible to perform ensemble averaging to remove the sampling influence caused by the pixel structure of the measurement camera 10.

[0036] In addition, since there are angles in binning where pixels are specifically placed in a particular bin in large or small numbers, the rotation angle θ of the measurement camera 10 is set to avoid such angles. If the rotation angle θ is, for example, 0, 45, or 90 degrees, multiple sampling points of a pixel will fall into the same bin when dropped for binning. Even at other angles, if the ratio between the representative spatial frequency components of the dropped sampling points of the pixel and the spatial frequency components of the binning sampling becomes close to an integer multiple, beats (moire) will occur, so it is preferable to set the rotation angle θ of the measurement camera 10 to avoid such cases as well.

[0037] The luminance distribution function obtained by photographing the knife edge is not an LSF but an edge spread function (ESF). The MTF measurement device 30 differentiates the result obtained by ensemble averaging the ESF profile (ESF as shown in FIG. 6(a)) to obtain an LSF as shown in FIG. 6(b), and performs a Fourier transform on the LSF to obtain an MTF as shown in FIG. 6(c). Note that the horizontal axis of FIG. 6(c) is spatial frequency (cycles / pixel), and the vertical axis is arbitrary unit, and the MTF value at the origin is set to 1. In this way, the MTF found in measurement phase P1 can be said to be the overall MTF of the entire system involved in capturing the knife-edge image, and is denoted here as MTF_tot1. In addition to this, the MTF of the knife edge itself is denoted as MTF_edge, the MTF of the measurement camera system is denoted as MTF_cam, and the MTF of the diffuser plate is denoted as MTF_diff. The overall MTF (MTF_tot1) obtained in the measurement phase P1 is a product of the MTF (MTF_edge) of the knife edge itself, the MTF (MTF_cam) of the measurement camera system, and the MTF (MTF_diff) of the diffuser. In other words, the overall MTF (MTF_tot1) satisfies the relationship shown in the following formula (1).

[0038] MTF_tot1 = MTF_edge × MTF_cam × MTF_diff … Equation (1)

[0039] Here, since a sharply polished knife edge can be considered to be equivalent to an ideal ESF compared to the pixel structure of a display, the result of differentiation and ensemble averaging can be considered approximately as a delta function. Therefore, the knife edge can be considered to have MTF characteristics close to the ideal state where the modulation depth is 1 regardless of the spatial frequency. In other words, the MTF of the knife edge itself (MTF_edge) has ideal MTF characteristics where the modulation depth is 1 regardless of the spatial frequency, and the above formula (1) can be approximately rewritten as the following formula (1B). As a result, in this system, an MTF is obtained by multiplying the MTF of the measurement camera system and the MTF of the diffuser, as shown in the following formula (1C).

[0040] MTF_tot1 = 1×MTF_cam ×MTF_diff … Formula (1B) MTF_tot1 = MTF_cam× MTF_diff … Formula (1C)

[0041] (Measurement Phase P2) Next, the MTF measurement method will be outlined with reference to Fig. 7. The MTF measurement method includes a preparation step, an imaging step, and a processing step in measurement phase P2. In the preparation step, a three-dimensional display 70 is placed in place of the light source device 60, and the knife-edge target 40 is removed from the diffusion plate 20 (see FIG. 7 and FIG. 2(b)).

[0042] In the photographing step, a reconstructed stereoscopic image 80 presented by the three-dimensional display 70 and focused on the diffusion plate 20 is photographed by the measurement camera 10 . The processing step is a step in which the captured image data is processed by the MTF measurement device 30, and the MTF is calculated in the same manner as in the measurement phase P1. The measurement phase P2 will be described in detail below.

[0043] 7 is an integral 3D display based on the fundamental principle of integral photography (IP), as an example. The 3D display 70 includes a display element 71 and a lens array 72. The display element 71 displays a group of element images input from the outside. Examples of the display element 71 include a liquid crystal display (LCD), a light emitting diode (LED) display, and an organic light emitting diode (OLED) display.

[0044] Lens array 72 is a typical one used in the IP system, and is placed in front of display element 71, with element lenses arranged in correspondence with element images. Here, lens array 72 has a large number of element lenses arranged in a delta array. The minimum horizontal sampling interval δ of lens array 72 is the reciprocal of the horizontal sampling frequency of lens array 72. Note that behind lens array 72 shown in FIG. 7 (in the negative direction of the z-axis), there are element images corresponding to the individual element lenses.

[0045] In the photographing process, as shown in Fig. 7, a video signal is input to the 3D display 70 to be measured, which displays a straight line as a spatial image at a desired depth position. This straight line is the thinnest straight line that the 3D display 70 can display. The input signal is converted into a luminance distribution at each pixel by a signal display conversion algorithm, and a specific method of displaying the thinnest straight line that can be displayed is determined. The position of the pattern (straight line) may be anywhere as long as it does not interfere with photographing by the measurement camera 10.

[0046] For example, when a line extending horizontally or vertically is displayed on a display having a square lattice pixel array, in most cases it will be displayed as a line having a width of one pixel. Even if the display has a square lattice pixel arrangement, the line width will change when measuring the MTF in any direction other than horizontal or vertical, or when the display has a pixel arrangement other than a square lattice (such as a delta arrangement). In these cases, the line width will change depending on the rules for how to represent the line with each pixel, that is, the rendering algorithm.

[0047] As a result of displaying the image signal input, the 3D display 70 under measurement presents an LSF (LSF image) as a spatial image at a desired depth position. The depth position of the stereoscopic image is, for example, a position spaced a distance D from the display surface of the display element 71 in the positive direction of the z axis. The distance D can be set arbitrarily, for example, to 5 cm, 10 cm, or the like.

[0048] Note that the spatial image will not necessarily be exactly the same as a single-pixel line display. For example, if the depth position of the stereoscopic image (the position in the z-axis direction of the spatial image) is very close to the display surface (near D=0), it is likely to be a single-pixel line display. On the other hand, if the spatial image is displayed at a position some distance away from the display surface in the z-axis direction, the spatial image is likely to be composed of light rays emitted from multiple locations on the display surface.

[0049] FIG. 8 is a partially enlarged view of FIG. 7. As shown in FIG. 7 and FIG. 8, the reproduced stereoscopic image 80 is formed at the position of the diffusion plate 20, and the reproduced stereoscopic image 80 and the diffusion plate 20 are at the same position. In the reproduced stereoscopic image 80, the spatial image (LSF image) of a straight line located near the center is bright (white), and the spatial image around it is dark (black). Note that in FIG. 2(b) corresponding to FIG. 7 and FIG. 8, the reproduced stereoscopic image formed at the position of the diffusion plate 20 from which the knife-edge target 40 has been removed is omitted, and a white image W of a straight line extending in the vertical direction (y-axis direction) at the center of the display surface displayed by the three-dimensional display 70 and a black image B displayed around it are shown in schematic form.

[0050] In the photographing process, the diffuser 20 is placed at the same depth as the LSF image, and the LSF is imaged on the diffuser 20. That is, the diffuser 20 is placed at the same position as the LSF as a spatial image. This is equivalent to placing the diffuser 20 at a position where the spatial image of the LSF appears to be most focused. At this time, the relative position and attitude of the measurement camera 10 and the diffuser 20 are made the same as the relative position and attitude of the measurement camera 10 and the diffuser 20 in the photographing process of the measurement phase P1. Therefore, in the photographing process of the measurement phase P2, it is preferable to use a jig 50 that relatively firmly fixes the measurement camera 10 and the diffuser 20 as shown in FIG. 7, and it is preferable that the jig 50 used in the measurement phase P2 is the same as the jig 50 used in the measurement phase P1.

[0051] In this way, in the photographing process, the LSF (a straight line image having some brightness distribution in the width direction) formed on the diffusion plate 20 is photographed by the measurement camera 10. 9 is an explanatory diagram of the processing steps of the measurement phase P2. The processing steps of the measurement phase P2 are similar to the processing steps of the measurement phase P1, but since the photographed image data acquired by the measurement camera 10 is different, some of the processing is also different. The operator (user) of the MTF measurement device 30 selects a desired rectangular image area that is likely to include the LSF image (image of a straight line) obtained by the measurement camera 10 as an ROI. In the upper diagram in FIG. 9, the area in which a group of white pixels (a group of pixels sandwiched between groups of pixels with dot hatching) are arranged indicates the LSF image. The MTF measurement device 30 arranges the ROI diagonally in the virtual memory space (upper diagram in FIG. 9). Then, the MTF measurement device performs a process (binning) of vertically dropping the pixel values ​​of the ROI (middle diagram in FIG. 9). Then, the MTF measurement device 30 obtains an LSF as a luminance distribution function as a result of ensemble averaging the profile of the LSF image (image of a straight line) (lower diagram in FIG. 9). Furthermore, the MTF measurement device 30 obtains an MTF by performing a Fourier transform of the obtained LSF. The MTF obtained here is naturally different from the MTF obtained in the measurement phase P1.

[0052] The MTF found in this measurement phase P2 can be said to be the overall MTF of the entire system involved in capturing the LSF image, and is denoted here as MTF_tot2. In addition to this, the MTF of the LSF image is denoted as MTF_image, the MTF of the measurement camera system is denoted as MTF_cam, and the MTF of the diffuser is denoted as MTF_diff. The overall MTF (MTF_tot2) calculated here is the product of the MTF of the LSF image (MTF_image), the MTF of the measurement camera system (MTF_cam), and the MTF of the diffuser (MTF_diff). In other words, the overall MTF (MTF_tot2) satisfies the relationship shown in the following formula (2).

[0053] MTF_tot2 = MTF_image × MTF_cam × MTF_diff … Equation (2)

[0054] The MTF of the LSF image alone (MTF_image) can be obtained by dividing the MTF (MTF_tot1 shown in formula (1)) resulting from the combination of the measurement camera system and the diffuser plate obtained in the preparation stage (measurement phase P1) shown in Fig. 1 from MTF_tot2 shown in formula (2). In other words, the MTF of the LSF image alone (MTF_image) satisfies the relationship shown in the following formula (3).

[0055] MTF_image = MTF_tot2 / MTF_tot1 = MTF_tot2 / (MTF_cam×MTF_diff) …Equation (3)

[0056] This represents the MTF of the 3D image at that depth. The MTF shown in formula (3) is the MTF at the depth position to be measured for a spatial image reproduction type 3D display, and is an index for objectively evaluating the image quality of 3D images.

[0057] The MTF for the LSF image presented as a spatial image is an MTF that includes all the effects of not only the rendering by the display and its driving device, but also the coding and transmission method used to transmit the input signal, the rendering to realize the spatial image, etc. Note that the rendering to realize the spatial image, for example in the case of an integral 3D display, is an algorithm for generating elemental images that correspond to the spatial image.

[0058] It is assumed that the line image (ideal reconstructed stereoscopic image) used as the measurement pattern to be displayed as a stereoscopic image is sampled using a square lattice (see Fig. 8). For example, in the case of an integral 3D display (three-dimensional display 70: see Fig. 7), the elemental images are resampled using the pixel structure of the lens array 72 and display element 71. The sampling interval Δ of the reconstructed stereoscopic image (one side of the square lattice in Fig. 8) is defined as being equal to the minimum sampling interval δ of the lens array 72 (Fig. 7), and the spatial frequency scale of the measurement results is normalized to the Nyquist frequency obtained by sampling the lens array 72 as the highest frequency (cutoff frequency) that can be displayed.

[0059] It is possible to measure the MTF distribution in the z direction by changing the z (depth) position of the line image (ideally reproduced stereoscopic image) as the measurement pattern displayed as a stereoscopic image and the diffuser 20. However, for positions behind the display surface, the spatial image becomes a virtual image and the diffuser 20 cannot be physically positioned there, so the MTF distribution in the z direction cannot be measured.

[0060] FIG. 7 shows an example in which a straight line parallel to the vertical side of the display surface of the display element 71 is displayed to obtain the MTF in the horizontal direction of the three-dimensional display 70. For example, when obtaining the MTF in the vertical direction of the three-dimensional display 70, the MTF can be obtained by displaying a straight line parallel to the horizontal side of the display surface of the display element 71. This is the same regardless of the direction of the MTF of the display surface of the three-dimensional display 70, and the MTF can be measured by displaying a straight line extending in a direction perpendicular to the desired direction to be measured on the display surface. The position of the pattern (straight line) may be anywhere as long as it does not interfere with the photographing by the measurement camera 10, and may be displayed anywhere on the display element 71, and the MTF at that position is measured. In fact, the pixel interval of the image sensor of the measurement camera 10 should be sufficiently small compared to the pixel interval of the display element 71 that displays the pattern (line). Note that if the sampling interval of the display element and the sampling interval of the image sensor are approximately the same, the measurement accuracy decreases.

[0061] When performing MTF measurement corresponding to the visual sensitivity characteristic of the human eye with respect to luminance, it is preferable to display a white straight line as a pattern and place a filter equivalent to the visual sensitivity distribution in the optical front stage of the measurement camera 10 before taking the image. In this case, the measurement camera 10 is a monochrome camera with a square pixel structure and no color filter, and is capable of linearly obtaining pixel values ​​for each pixel in response to light input to each pixel. The visual sensitivity distribution is defined as the CIE standard photometric observer specified by the International Commission on Illumination (CIE).

[0062] As described above, in the measurement phase P1 of the MTF measurement method, the camera 10 photographs the knife edge with the focus adjusted on the diffuser plate 20. At this time, light rays coming from multiple directions are incident on an arbitrary point on the diffuser plate 20, so the MTF measured by analyzing the profile of the knife edge in close contact with the diffuser plate 20 photographed by the measurement camera 10 can reflect the state in which the light rays coming from multiple directions are multiplexed by the diffuser plate 20. The MTF measurement method also uses the diffuser 20 in measurement phase P2, in which the MTF of the three-dimensional display 70 is measured. In measurement phase P2, the camera 10 captures a reproduced stereoscopic image 80 formed on the diffuser 20. At this time, light rays coming from multiple directions are incident on an arbitrary point on the diffuser 20, so the MTF measured by analyzing the profile of the reproduced stereoscopic image 80 captured by the measurement camera 10 can reflect the state in which the light rays coming from multiple directions are multiplexed by the diffuser 20.

[0063] [Comparison with Comparative Example] As described above, the MTF measurement system 1 measures the MTF using the diffuser 20. Here, as a comparative example, a system that measures the MTF by capturing a reproduced stereoscopic image without using a diffuser will be assumed and described in correspondence with the embodiment with reference to Figures 10(a) and 10(b). Note that the light rays shown in Figures 10(a) and 10(b) are schematic and not accurate.

[0064] In the comparative example, as shown in Fig. 10(a), a predetermined reproduced stereoscopic image 91 is displayed at a predetermined depth position (a position at a distance D from the display surface in the positive direction of the z-axis) by the 3D display 70 to be measured, and the measurement camera 10 focuses on the position at the distance D and captures the reproduced stereoscopic image 91. This makes it possible for the system of the comparative example to measure the MTF. However, the MTF measured by the system of the comparative example is the MTF at a single point, looking at the 3D display 70 from the camera position and camera orientation when the measurement camera 10 is set. If the measurement camera 10 is slightly shifted to the side, for example, and measures the 3D display 70, the measured MTF will be different.

[0065] In the case of a 2D display, the image seen by the viewer does not change even if the angle from which the viewer looks at the 2D display changes. In other words, the image displayed on a 2D display does not depend on the angle from which the viewer looks at the 2D display. Therefore, there is no need to change the viewing angle when measuring the MTF of a 2D display. On the other hand, in the case of a 3D display, if the viewer changes the angle from which the viewer looks at the 3D display, for example by moving left or right, the viewer sees different aspects of the 3D image. Therefore, in the case of a 3D display, it cannot be said that a good viewing method is for the viewer to stand still and view the 3D image directly in front of the 3D display. In other words, unlike a 2D display, it is difficult to objectively evaluate the image quality of the 3D image with the MTF measured to reflect the state as if the 3D image was viewed only from a point directly in front of the 3D display.

[0066] In contrast, as shown in FIG. 10(b), the MTF measurement system 1 according to the embodiment installs a diffuser 20 at a predetermined depth position (a position at a distance D from the display surface in the positive direction of the z-axis) by the three-dimensional display 70 to be measured, and forms a reproduced three-dimensional image 92 on the diffuser 20 at this position. Then, the measurement camera 10 focuses on the position at the distance D and captures the reproduced three-dimensional image 92. At this time, in the embodiment, the light rays coming from multiple directions are multiplexed by the diffuser 20 to form the three-dimensional image, which has the effect of performing a kind of averaging process as if the light rays coming from multiple directions were ensemble averaged. For example, the measured MTF is an excellent evaluation index for a three-dimensional image in the sense that it makes it easier to grasp the overall image of the three-dimensional image by reflecting a state as if the viewer observed not only a point directly in front of the three-dimensional display, but also the side of the three-dimensional image that appears when the viewer changes the angle at which the three-dimensional display is viewed.

[0067] Moreover, according to the MTF measurement system 1 of the embodiment, it is possible to quantitatively obtain the MTF (MTF of the LSF image alone (MTF_image) shown in formula (3)) at a depth position to be measured of a spatial image reproduction type three-dimensional display. The MTF at that depth position can itself be an objective evaluation value of the quality of a three-dimensional image. Moreover, the MTF at that depth position may be an objective evaluation value of the quality of a three-dimensional image by comparing different measured MTFs. When comparing different MTFs, for example, in FIG. 6(c), when two different MTFs are graphed with the values ​​at the origin aligned (normalized), a relatively high desired spatial frequency value is set as a reference spatial frequency, and one MTF is positioned above the other MTF at that reference spatial frequency. In this case, a 3D image presented at a position of distance D on a 3D display used for measurement corresponding to one MTF can be evaluated as being of better quality in that finer details (details corresponding to the reference spatial frequency) can be seen better than a 3D image presented at a position of distance D on a 3D display used for measurement corresponding to the other MTF. Alternatively, the quality of different MTFs may be judged, for example, by the high or low cutoff frequency of each MTF. Furthermore, for example, by changing the distance D, i.e., the depth position of the reproduced stereoscopic image, for one 3D display and comparing the MTFs measured at each position as described above, it is possible to objectively evaluate the quality of the 3D image at each depth position for one 3D display.

[0068] Although the MTF measurement system and the MTF measurement method according to the embodiment of the present invention have been described above, the gist of the present invention is not limited to these descriptions and should be interpreted broadly based on the claims. Furthermore, it goes without saying that various changes and modifications based on these descriptions are also included in the gist of the present invention.

[0069] For example, in the MTF measurement system 1, after a series of measurements in measurement phases P1 and P2 are completed, the knife-edge target 40 may be attached to the diffuser plate 20 in a new measurement phase P1 in the next series of measurements, and after a measurement related to the MTF of the camera is performed, the knife-edge target 40 may be removed from the diffuser plate 20. In this case, the knife-edge target 40 and the diffuser plate 20 can be reused in the new measurement phase P1. Alternatively, at least one of the knife-edge target 40 and the diffuser plate 20 may be replaced each time a new measurement phase P1 is performed.

[0070] In the jig 50 shown in FIG. 3, the positions of the first base 52 and the second base 53 are fixed on the base 51, but at least one of the positions may be variable. For example, the jig 50B shown in FIG. 11 is configured so that the positions of both the first base 52B and the second base 53B can be changed on the base 51B. The base 51B has a groove 54 formed on the upper surface so as to connect the first base 52B and the second base 53B. The groove 54 has a width equal to the width of the supports 52a and 53a. The jig 50B also has an auxiliary member 55 that stands the first base 52B at a desired position on the groove 54, and an auxiliary member 56 that stands the second base 53B at a desired position on the groove 54. The first base 52B is detachably fixed to the base 51B via an auxiliary member 55, and the second base 53B is detachably fixed to the base 51B via an auxiliary member 56. The auxiliary members 55, 56 are fixed to the base 51B by fixing members 55a, 55b such as screws. With this configuration, for example, when the measurement camera 10 or the diffusion plate 20 is replaced with a new part to change the performance or size of the measurement camera 10 or the diffusion plate 20, the distance between the measurement camera 10 and the diffusion plate 20 can be easily adjusted to correspond to the new part. Therefore, when the measurement camera 10 or the diffusion plate 20 is replaced with a new part, it is not necessary to replace the jig itself.

[0071] The jig 50 shown in FIG. 3 is configured so that the measurement camera 10 is attached to the screw portion 52d of the camera attachment portion 52c, but the camera attachment method is not limited to this. For example, the jig 50B shown in FIG. 11 has a first base 52B including a support 52a, an angle adjustment portion 52b arranged on the upper portion of the support 52a, and a camera attachment portion 52e provided on the angle adjustment portion 52b. The camera attachment portion 52e is configured in a box shape, and houses the measurement camera 10 with the camera lens protruding from the opening. At least the back plate of the box is made of a transparent material such as an acrylic plate. The camera attachment portion 52e is configured so that the rotation axis of the angle adjustment portion 52b and the optical axis of the measurement camera 10 coincide with each other. Even if the optical axis of the camera and the rotation axis of the jig do not coincide with each other, the MTF can be measured without any problem within the range where the knife edge portion is reflected within the angle of view of the camera. However, it is preferable for the jig to be configured so that the axis of rotation is located as close as possible to the optical axis, as this improves ease of use in practice.

[0072] In FIG. 7, the jig 50 may be configured to be movable. In the example shown in FIG. 12, the jig 50 is mounted on a support plate 57. The support plate 57 is larger than the base of the jig 50, and has a groove 58 in the z-axis direction on its upper surface. The jig 50 has a protrusion 51a on its bottom back surface that fits into the groove 58. The jig 50 can be slid in the z-axis direction on the support plate 57. The jig 50 is fixed to the support plate 57 by a fixing member 59a such as a screw. With this configuration, for example, in the measurement phase P2, the distance D, i.e., the depth position of the reproduced stereoscopic image, can be changed for one three-dimensional display, and the MTF can be easily measured at each position.

[0073] In the measurement phase P1, as shown in Fig. 1, the measurement camera 10 is placed rotated at a predetermined angle around the optical axis 15 so that the knife edge is oblique to the pixel array of the image sensor. However, since the placement of the measurement camera 10 and the placement of the knife-edge target 40 are relative, the knife-edge target 40 may be rotated without rotating the measurement camera 10. Even in this case, the MTF can be measured correctly. [Explanation of symbols]

[0074] 1. MTF measurement system 10. Measurement Camera 15 Optical axis 20 Diffuser 21 Page 1 22 Side 2 30 MTF measurement device 40 Knife Edge Target 50,50B Jig 51,51B base 52,52B 1st base 53,53B Second base 60 Light source device 70 3D Display 71 Display element 72 Lens Array 80,91,92 Reconstructed 3D image

Claims

1. A measuring camera; a light-transmitting diffusion plate having a first surface that is a diffusion surface and a second surface that faces the first surface; a knife-edge target having a sharp knife edge on one edge of a plate-like body, the plate-like body being detachably attached to the diffusion plate; an MTF measuring device that processes image data of an edge image obtained by photographing the knife edge through the diffusion plate by the measurement camera in a state where one surface of the knife-edge target is in close contact with the first surface of the diffusion plate, The MTF measurement system is characterized in that the MTF measurement device obtains a profile of an ESF, which is an edge spread function, from the image data, differentiates the result obtained by ensemble averaging the ESF profile, and calculates the MTF by obtaining a Fourier transform of the result.

2. the measurement camera includes an image sensor having pixels arranged in a square lattice, The optical axis of the measurement camera is arranged to be parallel to a direction perpendicular to the second surface of the diffusion plate, 2. The MTF measurement system according to claim 1, wherein the measurement camera is disposed in a state rotated at a predetermined angle around the optical axis so that the knife edge is inclined with respect to a pixel array of the image sensor.

3. a jig for connecting the measurement camera and the diffusion plate; 3. The MTF measurement system according to claim 2, wherein the jig holds the positions and orientations of the measurement camera and the diffusion plate, and holds the distance between the measurement camera and the diffusion plate at a constant value.

4. The jig is A base and a first base fixed on the base and carrying the measurement camera; 4. The MTF measurement system according to claim 3, further comprising: a second base fixed on the base portion at a distance from the first base, the second base carrying the diffusion plate.

5. a preparation step of closely contacting a knife-edge target having a sharp edge, which is a knife edge, at one edge of a plate-like body of the plate-like body to the first surface of a light-transmitting diffusion plate having a first surface that is a diffusion surface and a second surface that faces the first surface; an imaging step of imaging the knife edge through the diffusion plate by a measurement camera to obtain image data of an edge image; and a processing step of processing the image data by an MTF measuring device, the MTF measurement device determines a profile of an ESF, which is an edge spread function, from the image data in the processing step, differentiates a result obtained by ensemble averaging the ESF profiles, and calculates the MTF by determining a Fourier transform of the differentiated ESF profile.

6. the measurement camera includes an image sensor having pixels arranged in a square lattice, 6. The MTF measurement method according to claim 5, characterized in that, in the preparation step, the optical axis of the measurement camera is set so as to be parallel to a direction perpendicular to the second surface of the diffusion plate, and the measurement camera is disposed in a rotated state around the optical axis at a predetermined angle so that the knife edge is oblique with respect to the pixel array of the image sensor.

Images