Optical performance evaluation system, optical performance evaluation device, optical performance evaluation method, size determination device and size determination method

Abstract

To enable appropriate optical performance evaluation regardless of the infrared sensor resolution.SOLUTION: An evaluation target 11 comprises a background portion having a given emissivity and a rectangular linear pattern portion arranged at the center of the background portion, the linear pattern portion having a sufficiently lower emissivity than the background portion. An image capturing unit 12 captures an image of the evaluation target 11 from the sky. An image evaluation unit 13 evaluates the image of the evaluation target 11 captured by the image capturing unit 12.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present disclosure relates to evaluation of optical performance. The present disclosure particularly relates to MTF evaluation of thermal infrared images (hereinafter simply referred to as images) captured by infrared sensors for remote sensing purposes mounted on artificial satellites (hereinafter simply referred to as satellites), aircraft, etc. [Background technology]

[0002] When capturing an image, MTF (Modulation Transfer Function) measurements have traditionally been used to evaluate the resolution, which is one of the optical performance characteristics of an infrared sensor, from the image. Various methods have been proposed for measuring MTF. Among these, the "edge method" extracts edge information from objects present in the image and measures the MTF from that edge. The edge method is useful for evaluating the performance of infrared sensors installed on satellites and aircraft for remote sensing applications. Infrared sensors measure the temperature of materials. Edge information with steep temperature gradients is rare in thermal infrared images, making it difficult to extract the edges required for MTF measurement.

[0003] Non-Patent Document 1 gives an example of evaluating the MTF of an infrared sensor mounted on a commercial satellite using the edge method.

[0004] FIG. 1 is a block diagram showing the processing of Non-Patent Document 1. The processing in Non-Patent Document 1 is configured by an evaluation target section that is an object to be imaged, an imaging device section that is an infrared sensor on orbit, and an image evaluation device section that evaluates the MTF.

[0005] FIG. 2 is a flow diagram showing the image evaluation device section. The image evaluation device section is composed of an edge angle extraction section, an ESF (Edge Spread Function) calculation section, an LSF (Line Spread Function) calculation section, and an MTF calculation section. In Non-Patent Document 1, the boundary between land and sea in a desert area as shown in FIG. 3 is used as the evaluation target portion. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Brian. N. Wenny, et. al., “Pre- and Post-Launch Spatial Quality of the Landsat 8 Thermal Infrared Sensor”, Remote Sens., volume. 7, issue. 2, pp. 1962-1980, Jan. 2015. Summary of the Invention [Problem to be solved by the invention]

[0007] As mentioned above, MTF is evaluated using the boundaries of natural features in Non-Patent Document 1. Natural features do not have steep temperature changes at their boundaries, and a temperature gradient occurs on the land side of the object surface. The satellite-mounted infrared sensor used in the evaluation in Non-Patent Document 1 has low resolution with respect to the range of temperature change and the temperature gradient on the land side, so MTF evaluation was feasible. However, if the resolution of the onboard infrared sensor is high enough to resolve the range of temperature changes and temperature gradients at the boundaries of features, the boundary between land and sea will be blurred, making it impossible to capture sharp edges and making MTF evaluation difficult.

[0008] The main object of the present disclosure is to solve these problems, and more specifically, to enable appropriate evaluation of optical performance regardless of the resolution of the infrared sensor. [Means for solving the problem]

[0009] The optical performance evaluation system according to the present disclosure comprises: Has a predetermined emissivity The ground is the backKagebe and ,before Background The emissivity of the long part is 1 / 20 or less than the emissivity of the an evaluation target portion having a rectangular linear pattern portion; 、 Onboard a satellite or aircraft, Evaluation target section Top features containing an imaging device unit that captures images from the sky; and an image evaluation device section for evaluating the image of the evaluation target section captured by the imaging device section. [Effects of the Invention]

[0010] According to the present disclosure, optical performance evaluation can be performed appropriately regardless of the resolution of the infrared sensor. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram showing an MTF evaluation device of Non-Patent Document 1. [Figure 2] FIG. 10 is a flow diagram showing an example of the configuration of the image evaluation device unit in Non-Patent Document 1. [Figure 3] FIG. 1 is a schematic diagram illustrating an evaluation target portion of Non-Patent Document 1. [Figure 4] 1 is a block diagram showing an MTF evaluation system according to a first embodiment. [Figure 5] FIG. 2 is a schematic diagram illustrating an evaluation target portion according to the first embodiment. [Figure 6] 4 is a flow diagram showing an example of the configuration of an image evaluation device unit according to the first embodiment. FIG. [Figure 7] 3 is a schematic diagram illustrating an edge angle extraction unit according to the first embodiment. FIG. [Figure 8] FIG. 2 is a schematic diagram illustrating an LSF calculation unit according to the first embodiment. [Figure 9] FIG. 2 is a schematic diagram illustrating an MTF calculation unit according to the first embodiment. [Figure 10] 1 is a block diagram showing a linear pattern portion size optimization device according to a first embodiment. [Figure 11] FIG. 2 is a flow diagram illustrating the linear pattern portion size optimization device according to the first embodiment. [Figure 12]2 is a block diagram showing an example of a hardware configuration of an image evaluation device unit according to the first embodiment. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0012] Embodiment 1 FIG. 4 is a block diagram showing the MTF evaluation system 1 according to the first embodiment. The MTF evaluation system 1 evaluates the MTF of an infrared sensor mounted on a satellite, etc. The MTF evaluation system 1 corresponds to an optical performance evaluation system. The MTF evaluation system 1 comprises an MTF evaluation device 10 and an evaluation target unit 11 . The MTF evaluation device 10 corresponds to an optical performance evaluation device, and the operations performed by the MTF evaluation device 10 correspond to an optical performance evaluation method. The MTF evaluation device 10 comprises an image pickup device section 12 and an image evaluation device section 13 .

[0013] 5 is a schematic diagram of the evaluation target section 11. Details of the evaluation target section 11 will be described later.

[0014] The image capturing device unit 12 is an infrared sensor for remote sensing, and is mounted on a satellite or an aircraft. The imaging device unit 12 captures an image of a feature D11 including the evaluation target portion 11 from the sky, and transmits the captured image D12 to an image evaluation device unit 13 on the ground.

[0015] The image evaluation device unit 13 is a computer located on the ground. The image evaluation device section 13 performs MTF evaluation of the captured image D12 captured by the imaging device section 12. Fig. 6 shows an example of the configuration of the image evaluation device section 13. Details of Fig. 6 will be described later. The image evaluation device unit 13 has, for example, a hardware configuration shown in Fig. 12. The details of Fig. 12 will be described later.

[0016] Here, the evaluation target unit 11 will be described in detail with reference to FIG.

[0017] As shown in FIG. 5, the evaluation target portion 11 has a background portion 111 and a linear pattern portion 112 . The linear pattern portion 112 is disposed in the center of the background portion 111 . In the case of a visible light sensor, the MTF can be measured using an edge with sufficient luminance difference and uniform low and high luminance areas, as shown in Figure 3. In this embodiment, an infrared sensor is assumed which has a higher resolution than the one described in Non-Patent Document 1 but a lower resolution than the visible light sensor. Therefore, if an attempt is made to artificially install a shape such as that shown in Figure 3, the evaluation target will become large. Therefore, by preparing a linear pattern portion 112 as shown in FIG. 5, it becomes possible to reduce the scale of the evaluation target. However, the evaluation target portion 11 shown in FIG. 5 has a smaller width on the low-luminance side than that shown in FIG. 3, and is therefore susceptible to the influence of S / N (Signal to Noise).

[0018] The background 111 is a flat ground with a predetermined emissivity. The background 111 has a high emissivity. The linear pattern portion 112 has a sufficiently low emissivity compared to the background portion 111 . Since the linear pattern portion 112 has a sufficiently low emissivity compared to the background portion 111, a steep edge can be realized even when a high-resolution infrared sensor is used. As a result, even when a high-resolution infrared sensor is used, optical performance evaluation can be performed appropriately. In addition, the influence of S / N can be reduced. For example, it is preferable that the emissivity of the linear pattern portion 112 is 1 / 10 or less of the emissivity of the background portion 111. It is even more preferable that the emissivity of the linear pattern portion 112 is 1 / 20 or less of the emissivity of the background portion 111.

[0019] In this embodiment, it is assumed that the emissivity of the linear pattern portion 112 is 1 / 20 or less of the emissivity of the background portion 111, for example. Specifically, in this embodiment, it is assumed that concrete is used for the background portion 111 and a metal plate is used for the linear pattern portion 112. It is desirable to use an aluminum plate as the metal plate. Concrete has a high emissivity of approximately 0.9 in the thermal infrared wavelength range (8 to 14 μm), and is widely installed in various places such as parking lots, making it easy to use. Aluminum has a low emissivity of approximately 0.025 in the thermal infrared wavelength range (8 to 14 μm). Aluminum can also be purchased relatively inexpensively, and is easy to process if it is thin. For this reason, aluminum plates are suitable for the linear pattern portion 112.

[0020] The linear pattern portion 112 is rectangular. In this embodiment, it is assumed that the long sides (vertical sides in FIG. 5) are five times as long as the short sides (horizontal sides in FIG. 5), for example. The size of the linear pattern portion 112 can be optimized by designing it according to the GSD (Ground Sample Distance), resolution, and S / N of the infrared sensor.

[0021] Next, the image evaluation device section 13 will be described in detail with reference to FIG.

[0022] The image evaluation device unit 13 acquires the captured image D12 and outputs the MTF calculation result DOUT. The image evaluation device unit 13 is made up of an edge angle extraction unit 131 , an LSF calculation unit 132 , and an MTF calculation unit 133 . The image evaluation device 13 according to this embodiment does not include an ESF calculation unit, as compared with the configuration shown in FIG. 2. In other words, the image evaluation device 13 according to this embodiment evaluates the captured image D12 without calculating the ESF. This is due to the difference in shape between the evaluation target 11 in this embodiment and the evaluation target 11 in Non-Patent Document 1 shown in FIG. 3. The evaluation target 11 in FIG. 3 is configured with a binary pattern of low and high emissivity, in which the change in luminance value in the captured image follows a step function. However, to accurately calculate the ESF, the evaluation target 11 must have uniform and flat luminance characteristics on both the low and high emissivity sides. When implementing the evaluation target 11 using an artificial object, as in this embodiment, a large evaluation target is required. For example, if the GSD is Nm and the evaluation target requires at least M pixels, an evaluation target of size N × Mm is required.

[0023] In this embodiment, instead of using a binary pattern on a step function, a linear pattern is used as the evaluation target unit 11, and the LSF calculation unit 132 calculates an LSF instead of an ESF.

[0024] However, if the sensor (imaging device section 12) has high resolution, the size of the evaluation target section 11 can be reduced, so a binary pattern on a step function can be used as the evaluation target to calculate the ESF.

[0025] The edge angle extraction unit 131 calculates the edge angle D131 in the captured image D12.

[0026] FIG. 7 explains the edge angle D131 calculated by the edge angle extraction unit 131. The captured image D12 includes a linear pattern portion captured image D112 that is a captured image of the linear pattern portion 112. In FIG. 7, the angle between the y axis of the captured image D12 and the center line D113 parallel to the long side of the captured image D112 of the linear pattern portion is defined as θ. The edge angle extraction unit 131 calculates the angle θ and outputs the angle θ to the LSF calculation unit 132 as the edge angle D131.

[0027] The LSF calculation unit 132 calculates the LSF in the captured image D12 and outputs the calculation result as an LSF calculation result D132.

[0028] The LSF calculation unit 132 plots the luminance values ​​in the captured image D12 pixel by pixel along the x-axis of the captured image D12 on a graph.

[0029] In this case, to reduce the influence of noise, the LSF calculation unit 132 plots the luminance values ​​of multiple lines in the captured image D12 on a graph along the x-axis of the captured image D12, as shown in Fig. 8. The x-axis values ​​of the graph at this time are values ​​shifted by i*tan θ from the positions of pixel values ​​along the x-axis in the captured image D12, where i is the number of lines from the top edge of the captured image D12, and θ is the edge angle D131. That is, the LSF calculation unit 132 picks up a line profile at the same x-axis start position in the captured image D12, and shifts and plots the line profile on the graph by tan θ for i=1 and by n*tan θ for i=n.

[0030] The LSF calculation unit 132 outputs the plotted result to the MTF calculation unit 133 as an LSF calculation result D132.

[0031] The MTF calculation unit 133 performs a Fourier transform on the LSF calculation result D132 to calculate the MTF calculation result DOUT.

[0032] In this case, if the linear pattern portion captured image D112 is larger than the GSD of the infrared sensor, the size of the linear pattern portion captured image D112 will affect the MTF calculated as shown in Fig. 9. For this reason, it is necessary to divide by the MTF of the linear pattern portion captured image D112 itself. In FIG. 9, the upper part represents the spatial domain and the lower part represents the spatial frequency domain, and each domain can be transformed by Fourier transform / inverse Fourier transform. 9A shows a linear target (i.e., a linear pattern portion captured image D112). In FIG. 9B, this linear target is captured by an infrared sensor (i.e., the image capture device unit 12). As a result, an LSF such as that shown in FIG. 10C is observed, which is broadened due to the sensor performance. In this case, the straight line target in Figure 9(a) is ideally a target without width, but in reality, the straight line target has width, which results in a sinc function in the spatial frequency domain. Since the LSF in FIG. 9C is affected by the sinc function in FIG. 9A, it is necessary to divide it by the MTF (sinc function) of the linear pattern portion captured image D112 itself, as described above.

[0033] 8, when calculating the LSF calculation result D132, it is necessary to plot multiple lines in the y-axis direction to reduce the influence of noise, and therefore the vertical size of the linear pattern portion 112 in FIG.

[0034] Furthermore, as described above, when the linear pattern portion captured image D112 is larger than the GSD of the infrared sensor, it is necessary to exclude the influence of the linear pattern portion 112 from the MTF calculation result DOUT. For this reason, it is desirable that the horizontal size of the linear pattern portion 112 be equal to or smaller than the GSD. However, if the horizontal size of the linear pattern portion 112 is too small, the luminance will be buried in noise, so a minimum width is necessary.

[0035] The specific size of the linear pattern portion 112 is determined by optimizing the number of pixels in the vertical and horizontal directions through simulations taking into account the S / N ratio of the infrared sensor, and multiplying by the GSD to determine the design value.

[0036] The specific size optimization of the evaluation target portion 11 can be performed, for example, by a linear pattern portion size optimization device 2 shown in FIG. The linear pattern part size optimization device 2 determines the size (horizontal width / vertical width) of the linear pattern part 112. The linear pattern portion size optimization device 2 corresponds to a size determination device. The operation performed by the linear pattern portion size optimization device 2 corresponds to a size determination method.

[0037] As shown in FIG. 10, the linear pattern portion size optimization device 2 is composed of an evaluation target portion simulation unit 21, a sensor GSD effect simulation unit 22, a sensor MTF effect simulation unit 23, a sensor S / N effect simulation unit 24, an image evaluation device unit 13, a linear pattern portion size determination unit 26, and a linear pattern portion size correction unit 25.

[0038] The evaluation target portion simulation portion 21 generates, for each of one or more candidate sizes (candidate width / candidate height), a candidate for the evaluation target portion 11 including the linear pattern portion 112 formed in the candidate size.

[0039] The sensor GSD effect simulator 22 calculates, for each candidate size, an effect value of the sensor GSD when the candidate for the evaluation target 11 is measured by a sensor in the sky.

[0040] The sensor MTF effect simulator 23 calculates, for each candidate size, an effect value of the sensor MTF when it is assumed that the candidate for the evaluation target portion 11 is measured by a sensor.

[0041] The sensor S / N influence simulator 24 calculates, for each candidate size, an influence value of the sensor S / N when it is assumed that the candidate for the evaluation target portion 11 is measured by a sensor.

[0042] The image evaluation device unit 13 is similar to that shown in Fig. 4. Here, the image evaluation device unit 13 estimates the imaging MTF obtained when the candidate evaluation target unit 11 is measured by the sensor, using the influence value of the MTF, the influence value of the sensor GSD, and the influence value of the sensor S / N for each candidate size.

[0043] The linear pattern portion size determination unit 26 acquires the sensor MTF and imaging MTF for each candidate size and compares the acquired sensor MTF with the imaging MTF. The sensor MTF is the MTF obtained when the candidate evaluation target portion 11 is measured by a ground sensor. The imaging MTF contains noise components originating from the sensor, and the MTF deteriorates due to the effects of noise. The narrower the line pattern size, the smaller the number of usable pixels, so the S / N ratio decreases, and the wider the line pattern size, the more the LSF characteristics change.

[0044] In order to search for the minimum linear pattern size for which the imaging MTF satisfies the S / N ratio required for measurement, the linear pattern portion size determination unit 26 performs a comparative evaluation of the imaging MTF and the sensor MTF. Examples of evaluation methods include a method of comparing the amplitude of the Nyquist frequency of each MTF, or a method of comparing frequencies that are N (N≧2) times the Nyquist frequency of each MTF and comparing the ratio of the target frequency component to the noise (high frequency) component.

[0045] Based on the comparison result, the linear pattern portion size determination unit 26 determines one of the one or more candidate sizes as the size of the linear pattern portion 112. The linear pattern portion size determination unit 26 selects a candidate size where the sensor MTF and the imaging MTF match. The linear pattern portion size determination unit 26 corresponds to an MTF acquisition unit and a size determination unit.

[0046] The linear pattern portion size correction unit 25 corrects the candidate size for which the sensor MTF and the imaging MTF do not match as a result of the comparison by the linear pattern portion size determination unit 26, and outputs the corrected candidate size to the evaluation target portion simulation unit 21 as a new candidate size.

[0047] Next, the processing flow of the linear pattern portion size optimization device 2 will be described with reference to FIG.

[0048] The evaluation target portion simulation unit 21 receives the width / length information of the linear pattern portion 112 and creates an array (candidate for evaluation target) that simulates the evaluation target portion 11 as shown in Fig. 5. In this case, the initial width / length value DIN3 is used for the first simulation, and the width / length correction value D27 is used when the linear pattern portion size correction unit 25 has corrected it.

[0049] The sensor GSD effect simulation unit 22 simulates the effect of the sensor GSD value DIN1 during remote sensing on the array D22 of the simulated evaluation target unit 11.

[0050] The sensor MTF effect simulator 23 simulates the effect of the sensor MTF value DIN3 measured on the ground on the array D23 of the simulated evaluation target portion 11.

[0051] The sensor S / N influence simulation unit 24 simulates the influence of the sensor S / N value DIN2 on the array D24 of the simulated evaluation target portion 11.

[0052] The image evaluation device section 13 calculates the MTF (D26) using the array D25 of the evaluation target section 11 that simulates the influence of the sensor.

[0053] If the calculated MTF (D26) matches the sensor MTF value DIN3 measured on the ground, the linear pattern portion size determination unit 26 outputs the width / length size at that time.

[0054] If the calculated MTF (D26) does not match the sensor MTF value DIN3, the linear pattern portion size corrector 25 adjusts the horizontal / vertical widths and inputs them to the evaluation target portion simulator 21 again.

[0055] The vertical and horizontal widths of the linear pattern are corrected until the calculated MTF (D26) matches the sensor MTF value DIN3 measured on the ground.

[0056] In the above, in this embodiment, it has been explained that the evaluation target portion has a background portion with high emissivity and a linear pattern portion with low emissivity. As a result, according to this embodiment, even with a high-resolution infrared sensor, a steep edge can be obtained and the MTF can be measured appropriately.

[0057] The procedure described in this embodiment is an example. Therefore, it is possible to carry out only a part of the procedure described in this embodiment. Furthermore, at least part of the procedure described in this embodiment may be combined with a procedure not described in this embodiment. Furthermore, the configuration and procedures described in this embodiment may be modified as necessary.

[0058] Finally, the hardware configuration of the image evaluation device section 13 will be described with reference to FIG. The image evaluation device unit 13 includes a processor 901, a main memory device 902, an auxiliary memory device 903, and a communication device 904 as hardware. The processor 901 is a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or the like. The main storage device 902 is a RAM (Random Access Memory). The auxiliary storage device 903 is a ROM (Read Only Memory), a flash memory, an HDD (Hard Disk Drive), or the like. The communication device 904 is an electronic circuit that performs the communication processing of data. The communication device 904 is, for example, a communication chip or a NIC (Network Interface Card).

[0059] The functions of the edge angle extraction unit 131, the LSF calculation unit 132, and the MTF calculation unit 133 are realized by, for example, a program. The auxiliary storage device 903 stores programs that realize the functions of the edge angle extraction unit 131, the LSF calculation unit 132, and the MTF calculation unit 133. These programs are loaded from the auxiliary storage device 903 to the main storage device 902. Then, the processor 901 executes these programs to perform the operations of the edge angle extraction unit 131, the LSF calculation unit 132, and the MTF calculation unit 133. The auxiliary storage device 903 also stores an OS (Operating System). At least a part of the OS is executed by the processor 901 . The processor 901 executes the OS, which performs task management, memory management, file management, communication control, and the like. The image evaluation device unit 13 may also be realized by a processing circuit, such as a logic IC (Integrated Circuit), a GA (Gate Array), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array).

[0060] Although not shown, the linear pattern portion size optimization device 2 is also, for example, a computer having the same hardware configuration as that shown in FIG. In the hardware configuration of the linear pattern portion size optimization device 2, the components shown in Fig. 10 are arranged in the processor 901 of Fig. 12 instead of the edge angle extraction unit 131, the LSF calculation unit 132, and the MTF calculation unit 133. Except for this, the above description of Fig. 12 also applies to the hardware configuration of the linear pattern portion size optimization device 2. [Explanation of symbols]

[0061] 1 MTF evaluation system, 10 MTF evaluation device, 11 evaluation target section, 111 background section, 112 linear pattern section, 12 imaging device section, 13 image evaluation device section, 131 edge angle extraction section, 132 LSF calculation section, 133 MTF calculation section, 2 linear pattern section size optimization device, 21 evaluation target section simulation section, 22 sensor GSD effect simulation section, 23 sensor MTF effect simulation section, 24 sensor S / N effect simulation section, 25 linear pattern section size correction section, 26 linear pattern section size determination section.

Claims

1. an evaluation target portion having a background portion which is the ground having a predetermined emissivity and a rectangular linear pattern portion having an emissivity which is 1 / 20 or less of the emissivity of the background portion; an imaging device unit mounted on a satellite or an aircraft, which captures images of features including the evaluation target portion from above; an optical performance evaluation system comprising an image evaluation device section that evaluates an image of the evaluation target section captured by the imaging device section;

2. 2. The optical performance evaluation system according to claim 1, wherein the background portion is concrete and the linear pattern portion is a metal plate.

3. 3. The optical performance evaluation system according to claim 2, wherein the linear pattern portion is an aluminum plate.

4. The image evaluation device unit includes:

2. The optical performance evaluation system according to claim 1, wherein the image of the evaluation target portion is evaluated without calculating an ESF (Edge Spread Function).

5. an imaging device unit mounted on a satellite or aircraft that captures from above an object including an evaluation target portion having a background portion, which is the ground having a predetermined emissivity, and a rectangular linear pattern portion having an emissivity that is 1 / 20 or less of the emissivity of the background portion; an optical performance evaluation device comprising an image evaluation device section that evaluates an image of the evaluation target section captured by the imaging device section;

6. An infrared sensor mounted on a satellite or an aircraft captures an image of a feature from above, the feature including an evaluation target portion having a background portion which is the ground having a predetermined emissivity and a rectangular linear pattern portion having an emissivity of 1 / 20 or less of the emissivity of the background portion; An optical performance evaluation method in which a computer evaluates the image of the evaluation target portion captured by the infrared sensor.

7. A size determination device for determining a size of a linear pattern portion of an evaluation target portion having a background portion having a predetermined emissivity and a rectangular linear pattern portion disposed at the center of the background portion and having a sufficiently lower emissivity than the background portion, an MTF acquisition unit that acquires, for each candidate size, a sensor MTF, which is an MTF (Modulation Transfer Function) obtained when a candidate evaluation target portion in which the linear pattern portion formed in the candidate size is placed at the center of the background portion is measured by a sensor on the ground, and an imaging MTF, which is an MTF obtained when the candidate evaluation target portion is measured by a sensor in the sky, for one or more candidate sizes that are candidates for the size of the linear pattern portion; and a size determination unit that compares the sensor MTF acquired by the MTF acquisition unit with the imaging MTF for each candidate size, and determines one of the one or more candidate sizes as the size of the linear pattern portion based on the comparison result.

8. The size determination unit 8. The size determination device according to claim 7, wherein a candidate size for which the sensor MTF and the imaging MTF match is determined as the size of the linear pattern portion.

9. The size determination unit 8. The size determination device according to claim 7, wherein the sensor MTF and the imager MTF are compared by one of a method of comparing the amplitudes of the Nyquist frequencies of the sensor MTF and the imager MTF, and a method of comparing the frequency of the sensor MTF and the frequency of the imager MTF that is N (N≧2) times the Nyquist frequency, and comparing the ratios of the target frequency component and the noise component.

10. 1. A size determination method for determining a size of a linear pattern portion of an evaluation target portion having a background portion with a predetermined emissivity and a rectangular linear pattern portion disposed at the center of the background portion and having a sufficiently lower emissivity than the background portion, the method comprising: The computer acquires, for each candidate size, a sensor MTF, which is an MTF (Modulation Transfer Function) obtained when a candidate evaluation target portion in which the linear pattern portion formed in the candidate size is placed at the center of the background portion is measured by a sensor on the ground, and an imaging MTF, which is an MTF obtained when the candidate evaluation target portion is measured by a sensor in the sky, for one or more candidate sizes that are candidates for the size of the linear pattern portion. the computer compares the acquired sensor MTF with the imaging MTF for each candidate size, and determines one of the one or more candidate sizes as the size of the linear pattern portion based on the comparison result.

Images