JP3928054B2 - Discrimination performance evaluation method, apparatus, program, and recording medium for noise of visible camera - Google Patents

Description

  The present invention relates to a discrimination performance evaluation method for noise of a visible camera, a program for causing a computer to execute each step of the method, a computer-readable recording medium recording the program, and a discrimination performance evaluation for noise of a visible camera The present invention relates to an apparatus, a program for causing a computer to function as each means of the apparatus, and a computer-readable recording medium (hereinafter simply referred to as a visible camera identification performance evaluation system) on which the program is recorded. In particular, the present invention relates to a visible camera identification performance evaluation system that can objectively and specifically evaluate the identification performance of a visible camera against noise.

  2. Description of the Related Art In recent years, image processing technology has been the main focus in security systems for safety guards, traffic monitoring systems, ship navigation systems, aircraft control systems, national park animal monitoring systems, and the like.

  As means for capturing the image, for example, a visible camera such as a high-definition camera (so-called high-vision camera) is used. In the visible camera, means for evaluating performance include the total number of pixels and the S / N ratio.

  However, the performance evaluation means in the conventional visible camera has no means for objectively and specifically evaluating the discrimination performance against noise numerically.

  Here, the identification performance of the visible camera with respect to noise refers to the ability of a captured image to be identified without being buried in the noise, that is, the ability to see. If the discrimination performance against this noise is good, it is possible to identify a long-distance object. Therefore, when a visible camera is installed in the system that performs the image processing, it is important to grasp the identification performance of the visible camera objectively and specifically by numerical values.

  An object of the present invention is to provide a visible camera identification performance evaluation system that can objectively and specifically evaluate the identification performance of a visible camera against noise.

  Other objects and novel features of the present invention will be clarified in embodiments described later.

In order to achieve the above object, according to the first aspect of the present invention, the first calculation means includes the numerical value of the illumination spectral characteristic of the reference illumination input by the input means and the numerical value of the spectral reflectance of the reference chart. A first calculation step for calculating a numerical value of the reflected light characteristic, and a first conversion in which the first conversion means converts the numerical value of the reflected light characteristic calculated in the first calculation step into a numerical value of the tristimulus value of the reflected light. A step, a second calculation means, the numerical value of the illumination spectral characteristic inputted by the input means, the numerical value of the spectral reflectance, the numerical value of the spectral characteristic of the visible camera, the numerical value of the focal length of the lens, and the lens A second calculation step of calculating a numerical value of the susceptibility characteristic of the visible camera from the value of the spectral transmittance of the image, and the second conversion means converts the numerical value of the susceptibility characteristic calculated in the second calculation step into the tristimulus of the signal To the numeric value A second conversion step for converting, and a third calculation means, the numerical value of the tristimulus value of the reflected light converted by the first conversion step, and the numerical value of the tristimulus value of the signal converted by the second conversion step, A third calculation step for calculating a numerical value of the stimulus value per unit light, and a third conversion means for converting the numerical value of the system noise of the visible camera input by the input means into the numerical value of the tristimulus value of noise. A third conversion step, a fourth conversion unit, wherein the fourth conversion unit converts the numerical value of the noise tristimulus value converted by the third conversion step into a numerical value of the noise stimulus value; A fourth calculation for calculating a noise equivalent value from the value of the stimulus value per unit light calculated in the third calculation step and the value of the noise stimulus value converted in the fourth conversion step. A step, fifth conversion means, and a fifth conversion step of converting said fourth numerical noise equivalent calculated by the calculating step L ^* a ^* b ^* to a number of parameters in the color system, the fifth calculating means A fifth calculation step of calculating a noise equivalent bright color difference value in Lab space from a parameter value in the L ^* a ^* b ^* color system converted by the fifth conversion step. To do.

As a result, according to the first aspect of the present invention, the noise equivalent is first calculated from the tristimulus value on the signal side of the visible camera and the tristimulus value on the noise side, and then the noise equivalent is expressed as L ^* a ^* b ^* color specification. Is converted to a parameter in the system, and then a numerical value of the noise equivalent bright color difference in the Lab space is calculated from the parameter in the L ^* a ^* b ^* color system.

  That is, the invention according to claim 1 can express a noise equivalent bright color difference in a three-dimensional color space. Thus, the invention according to claim 1 can objectively and specifically evaluate the discrimination performance against noise of the visible camera numerically.

  According to a second aspect of the present invention, there is provided a program for causing a computer to execute each step of the identification performance evaluation method for noise of the visible camera according to the first aspect.

  As a result, the invention according to claim 2 allows the computer to execute each step of the identification performance evaluation method for noise of the visible camera according to claim 1 by causing the computer to read the program. . Therefore, as in the invention according to claim 1, the invention according to claim 2 can objectively and specifically evaluate the discrimination performance against noise of the visible camera numerically.

  According to a third aspect of the present invention, there is provided a computer-readable recording medium storing a program for causing a computer to execute the steps of the discrimination performance evaluation method for noise of the visible camera according to the first aspect. Features.

  As a result, the invention according to claim 3 causes the computer to read the program recorded on the recording medium, thereby causing the computer to perform each step of the identification performance evaluation method for noise of the visible camera according to claim 1. Can be executed. Therefore, similarly to the invention according to claim 1, the invention according to claim 3 can objectively and specifically evaluate the identification performance of the visible camera against noise.

Further, the invention according to claim 4 is that the numerical value of the illumination spectral characteristic of the reference illumination, the numerical value of the spectral reflectance of the reference chart, the numerical value of the spectral characteristic of the visible camera, the numerical value of the focal length of the lens, and the spectral of the lens. The input means for inputting the numerical value of the transmittance and the numerical value of the system noise of the visible camera, the numerical value of the reflected light characteristic from the numerical value of the illumination spectral characteristic input by the input means and the numerical value of the spectral reflectance First calculating means for calculating the reflected light characteristic numerical value calculated by the first calculating means into a reflected light tristimulus numerical value, and the input means input the input A second value for calculating the sensitivity characteristic of the visible camera from the numerical value of the illumination spectral characteristic, the numerical value of the spectral reflectance, the numerical value of the spectral characteristic, the numerical value of the focal length, and the numerical value of the spectral transmittance. Calculation means and front A second conversion means for converting the numerical value of the sensitive characteristic calculated by the second calculation means into a numerical value of a tristimulus value of a signal; a numerical value of the tristimulus value of the reflected light converted by the first conversion means; Third arithmetic means for calculating a numerical value of the stimulus value per unit light from the numerical value of the tristimulus value of the signal converted by the second conversion means, and the numerical value of the system noise input by the input means Third conversion means for converting to a tristimulus value, fourth conversion means for converting the noise tristimulus value converted by the third conversion means to a noise stimulus value, and the third Fourth arithmetic means for calculating a noise equivalent numerical value from the numerical value of the stimulus value per unit light calculated by the arithmetic means and the numerical value of the stimulus value of the noise converted by the fourth conversion means; By 4th computing means A fifth converting means for converting the numerical value of the computed the noise equivalent L ^* a ^* b ^* to a number of parameters in the color system, in the converted L ^* a ^* b ^* color system by the fifth conversion unit And fifth calculating means for calculating a numerical value of the noise equivalent bright color difference in the Lab space from the numerical value of the parameter.

  As a result, as in the invention according to claim 1, the invention according to claim 4 can objectively and specifically evaluate the discrimination performance against noise of the visible camera numerically.

  According to a fifth aspect of the present invention, there is provided a program for causing a computer to function as each means of the discrimination performance evaluation apparatus for visible camera noise described in the fourth aspect.

  As a result, the invention according to claim 5 can cause the computer to function as each means of the discrimination performance evaluation apparatus for noise of the visible camera according to claim 4 by causing the computer to read the program. . Therefore, the invention according to claim 5 can objectively and specifically evaluate the discrimination performance against noise of the visible camera in the same manner as the invention according to claim 4.

  According to a sixth aspect of the present invention, there is provided a computer-readable recording medium storing a program for causing a computer to function as each means of the identification performance evaluation apparatus for visible camera noise described in the fourth aspect. Features.

  As a result, the invention according to claim 6 makes the computer read the program recorded on the recording medium, thereby causing the computer to execute each means of the discrimination performance evaluation apparatus for noise of the visible camera according to claim 4. Can be made to function. Therefore, the invention according to claim 6 can objectively and specifically evaluate the discrimination performance against noise of the visible camera in the same manner as the invention according to claim 4.

  According to the discrimination performance evaluation method for noise of a visible camera according to the present invention (claim 1), the noise equivalent bright color difference can be expressed in a three-dimensional color space. Can be evaluated numerically and specifically.

  Moreover, according to the program concerning this invention (Claim 2), a computer can be made to perform each step of the identification performance evaluation method with respect to the noise of a visible camera by making a computer read a program. As a result, the invention according to claim 2 can express the noise equivalent bright color difference in the three-dimensional color space in the same manner as the invention according to claim 1, so that the discrimination performance for the noise of the visible camera is objective. In addition, it can be specifically evaluated by numerical values.

  Moreover, according to the recording medium (Claim 3) of the present invention, by causing the computer to read the program recorded on the recording medium, each step of the discrimination performance evaluation method for noise of the visible camera is performed by the computer. Can be executed. As a result, the invention according to claim 3 can express the noise equivalent bright color difference in the three-dimensional color space in the same manner as the invention according to claim 1, so that the discrimination performance against the noise of the visible camera is objective. In addition, it can be specifically evaluated by numerical values.

  In addition, according to the discrimination performance evaluation apparatus for noise of a visible camera according to the present invention (claim 4), the noise equivalent bright color difference can be expressed in a three-dimensional color space, as in the invention according to claim 1. Therefore, it is possible to objectively and specifically evaluate the identification performance of the visible camera against noise.

  Moreover, according to the program concerning this invention (Claim 5), a computer can be functioned as each means of the identification performance evaluation apparatus with respect to the noise of a visible camera by making a computer read a program. As a result, since the invention according to claim 5 can express the noise equivalent bright color difference in the three-dimensional color space, similarly to the invention according to claim 4, the discrimination performance against the noise of the visible camera is objective. In addition, it can be specifically evaluated by numerical values.

  Moreover, according to the recording medium (Claim 6) of the present invention, by causing the computer to read the program recorded on the recording medium, the computer can be used as each unit of the identification performance evaluation apparatus for noise of the visible camera. Can function. As a result, since the invention according to claim 6 can express the noise equivalent bright color difference in the three-dimensional color space, similarly to the invention according to claim 4, the discrimination performance against the noise of the visible camera is objective. In addition, it can be specifically evaluated by numerical values.

  Hereinafter, an example of an embodiment of an identification performance evaluation system for noise of a visible camera according to the present invention will be described with reference to the accompanying drawings. Note that this embodiment does not limit the identification performance evaluation system for noise of the visible camera.

(Explanation of hardware configuration)
FIG. 2 is a block diagram showing a hardware configuration. In FIG. 2, 1 is an input device, 2 is an arithmetic device, and 3 is an output device.

  The input device 1 includes a keyboard, OCR, mouse, tablet, light pen, image scanner, and the like. The arithmetic device 2 includes a CPU, a parallel processor, a vector processor, and the like. Further, the output device 3 includes a display device, a printer, and the like.

(Explanation of hardware functions)
FIG. 1 is a block diagram showing functions of the hardware shown in FIG. In FIG. 1, 1 is the input device, 2 is the arithmetic device, and 3 is the output device.

The input device 1 includes a numerical value of the illumination spectral characteristic S (λ) 11 of the reference illumination 4, a numerical value of the spectral reflectances R _R , _G and _B (λ) 12 of the reference chart 5, and a spectral characteristic bar of the visible camera 6. The numerical values of r (λ), bar g (λ), bar b (λ) 13, the focal length (optical specification) F 14 of the lens 7, and the numerical value of the spectral transmittance μ (λ) 15 of the lens 7 The system noise [V] R _N , G _N , and B _N 16 values of the visible camera 6 are input.

  The arithmetic device 2 includes first to fifth arithmetic means 201, 203, 205, 208, 210 and first to fifth conversion means 202, 204, 206, 207, 209.

The first calculation unit 201 calculates the numerical value of the illumination spectral characteristic S (λ) 11 of the reference illumination input by the input device 1 and the spectral reflectances R _R , _{G of the} reference chart based on the following equation (1). , _B (λ) 12, and the reflected light characteristics B _R , _G , _B (λ) are calculated.

The first converter 202 converts the numerical values of the reflected light characteristics B _R , _G and _B (λ) calculated by the first calculator 201 into the tristimulus value B of reflected light based on the following equation (2). _{Converts to X} , _Y , _Z numbers. In the following formula (2), the transformation matrix of A is an example, and is not limited to this numerical value.

Based on the following equation (3), the second arithmetic unit 203 calculates the numerical value of the illumination spectral characteristic S (λ) 11 input by the input device 1 and the spectral reflectances R _R , _G , _B (λ ) 12, the spectral characteristics bar r (λ), bar g (λ), bar b (λ) 13 of the visible camera 6, and the focal length (optical specification) F 14 of the lens 7 From the numerical value of the spectral transmittance μ (λ) 15 of the lens 7, the numerical values of the sensitive characteristics [V] R, G, B of the visible camera 6 are calculated.

  The second conversion means 204 converts the numerical values of the sensitive characteristics [V] R, G, and B calculated by the second calculation means 203 based on the following equations (4) and (5) into tristimulus values of signals. This is converted into numerical values of X, Y, and Z. In the following formula (4), the conversion matrix A is an example, and is not limited to this numerical value.

The third calculation means 205 calculates the values of the tristimulus values B _X , _Y and _Z of the reflected light converted by the first conversion means 202 and the second conversion means 204 based on the following equation (6). The stimulation values [V] V _SX , _SY , _SZ per unit light are calculated from the tristimulus values X, Y, Z of the converted signal.

Based on the following equation (7), the third conversion unit 206 converts the system noise [V] R _N , G _N , and B _N values of the visible camera 6 input by the input device 1 into tristimulus values of noise. This is converted into numerical values of X _N , Y _N and Z _N. In the following equation (7), A uses the transformation matrix A of the above equations (2) and (4).

The fourth conversion means 207 also converts the noise tristimulus values X _N , Y _N , and Z _N converted by the third conversion means 206 into the noise stimulus values [ V] V _NX , _NY and _NZ are converted into numerical values.

The fourth calculation means 208 calculates the stimulation values [V] V _SX , _SY , _SZ per unit light calculated by the third calculation means 205 based on the following equation (8), and the fourth conversion The noise equivalent values NEXD, NEYD, NEZD are calculated from the noise stimulus values [V] V _NX , _NY , _NZ converted by the means 207.

The fifth conversion unit 209 converts the numerical values of the noise equivalents NEXD, NEYD, and NEZD calculated by the fourth calculation unit 208 based on the following equation (9) to the parameter ΔL ^* in the L ^* a ^* b ^* color system ^. , Δa ^* and Δb ^* . In the following formula (9), the specific numerical values “116”, “16”, “500”, and “200” surrounded by a square are examples, and are not limited to these numerical values.

Based on the following equation (10), the fifth calculation means 210 calculates the values of the parameters ΔL ^* , Δa ^* , Δb ^{* in} the L ^* a ^* b ^* color system converted by the fifth conversion means 209, The numerical value of the noise equivalent bright color difference NEED in the Lab space is calculated.

  The output device 3 displays and outputs the calculation result of the calculation device 2.

(Description of overview)
FIG. 3 is a block diagram showing an outline of this embodiment. In FIG. 3, 4 is the reference illumination, 5 is the reference chart, 6 is the visible camera for performance evaluation, and 7 is the lens of the visible camera. The reference illumination 4 is turned on, and the reference chart 5 is irradiated with light from the reference illumination 4. On the other hand, the reference chart 5 is imaged by the visible camera 6.

  The discrimination performance evaluation system for noise of the visible camera according to this embodiment objectively evaluates how much the image captured by the visible camera 6 can be identified without being buried in the noise, that is, the discrimination performance of the visible camera 6. It evaluates numerically and specifically.

That is, the numerical value of the illumination spectral characteristic S (λ) 11 of the reference illumination 4, the numerical value of the spectral reflectance R _R , _G , _B (λ) 12 of the reference chart 5, and the spectral characteristic bar r of the visible camera 6. (Λ), bar g (λ), bar b (λ) 13, focal length (optical specification) F 14 of the lens 7, and spectral transmittance μ (λ) 15 of the lens 7 When the visible light camera 6 of system noise _{[V] R N, G N} , and numerical values of B _N 16 are those each of which is grasped. These numerical values are respectively input to the input device 1.

Then, the first calculation means 201 calculates the numerical values of the reflected light characteristics B _R , _G and _B (λ) based on the above equation (1). Next, the first conversion means 202 converts the reflected light into tristimulus values B _X , _Y and _Z based on the above equation (2). Next, the second calculation means 203 calculates numerical values of the sensitive characteristics [V] R, G, B of the visible camera 6 based on the above equation (3). Next, the second conversion means 204 converts the signal into tristimulus values X, Y, and Z based on the above equations (4) and (5). The third calculating means 205 calculates the stimulation values [V] V _SX , _SY , _SZ per unit light based on the above equation (6). On the other hand, the third conversion means 206 converts the camera system noise [V] R _N , G _N , B _N into numerical values of noise tristimulus values X _N , Y _N , Z _N based on the above equation (7). To do. Next, the fourth conversion means 207 also converts the noise stimulus values [V] V _NX , _NY and _NZ into numerical values based on the above equation (7). Then, the fourth computing means 208 computes numerical values of noise equivalents NEXD, NEYD, NEZD based on the above equation (8). Next, the fifth conversion means 209 converts the numerical values of the parameters ΔL ^* , Δa ^* , Δb ^{* in} the L ^* a ^* b ^* color system based on the above equation (9). Then, the fifth computing means 210 computes the numerical value of the noise equivalent bright color difference NEED in the Lab space based on the above equation (10). The obtained noise equivalent bright color difference NEED is output to the output device 3.

(Description of images taken by different visible cameras)
FIG. 4 is an explanatory diagram schematically showing images taken by different visible cameras. In FIG. 4, 4 is the reference illumination, and 5 is the reference chart. The A and B images obtained by imaging the reference chart 5 of the same color (red in this example) illuminated by the same reference illumination 4 with different visible cameras A and B will be compared.

Signal level of the A image (small noise) captured by the visible camera A and the B image (large noise) captured by the visible camera B (sensitive characteristics [V] R, G, B of the visible camera) Are equal. However, p-p of the noise of the A image, i.e., the noise level (visible camera system noise _{[V] R N, G N} , B N) is about 0.011MV. In contrast, p-p of the noise of the B image, that is, the noise level (visible camera system noise _{[V] R N, G N} , B N) is about 0.045MV.

  For this reason, as shown in FIG. 4, the B image is more noisy than the A image. When this is expressed by the noise equivalent bright color difference NEED, the A image is 125, whereas the B image is 201, and the B image is numerically about twice as loud as the A image. Thereby, the discrimination performance with respect to the noise of the visible camera can be objectively and specifically evaluated by numerical values.

(Explanation using specific numerical values)
Hereinafter, NEED by the visible camera B will be obtained from specific numerical values.

First, numerical values of the reflected light characteristics B _R , _G and _B (λ) calculated by the first calculating means 201 are shown in the spectral column of FIG. Next, the numerical values of the perceptual characteristics [V] R, G, and B of the visible camera calculated by the second calculating means 203 are shown in the signal level column of FIG. Next, the numerical values of the tristimulus values X, Y, and Z of the signal converted by the second conversion means 204 are shown in the signal column of FIG. On the other hand, the numerical value of the visible camera 6 of system noise _{[V] R N, G N} , B N input by the input device 1 is shown in the column and the column of the noise level of the noise p-p shown in FIG. 5 (C) . Next, the numerical values of the tristimulus values X _N , Y _N and Z _N of the noise converted by the third conversion means 206 are shown in the noise column of FIG.

Then, the numerical values of noise equivalents NEXD, NEYD, and NEZD calculated by the fourth calculation means 208 are shown in the NEYZD column of FIG. Next, the numerical values of the parameters ΔL ^* , Δa ^* , Δb ^* in the L ^* a ^* b ^* color system converted by the fifth converting means 209 are shown in the Lab column of FIG. 6 (B). The numerical value of the noise equivalent bright color difference NEED in the Lab space calculated by the fifth calculation means 210 is shown in the NEED column in FIG. 6C and the NEED B image column in FIG. 6D.

When obtaining a specific value of NEED by the visible camera B, the value of the illumination spectral characteristic S (λ) 11 of the reference illumination 4 and the spectral reflectance R _R , _G , _B (λ) 12 of the reference chart 5 are calculated. Of the spectral characteristics bar r (λ), bar g (λ), bar b (λ) 13 of the visible camera 6, numerical value of the focal length (optical specification) F 14 of the lens 7, The numerical value of the spectral transmittance μ (λ) 15, the numerical values of the tristimulus values B _X , _Y and _Z of the reflected light converted by the first conversion unit 202, and the unit light calculated by the third calculation unit 205 The numerical values of the stimulation values [V] V _SX , _SY and _{SZ and} the numerical values of the noise stimulation values [V] V _NX , _NY and _NZ converted by the fourth conversion means 207 are omitted.

  On the other hand, the specific value of NEED by the visible camera A is shown in the NE image A column of FIG.

FIG. 7 is an explanatory diagram showing an XYZ (Yxy) color system. This XYZ (Yxy) color system includes the tristimulus values B _X , _Y and _{Z of the} reflected light converted by the first conversion unit 202 and the tristimulus values X and Y of the signal converted by the second conversion unit 204. , Z and the noise tristimulus values X _N , Y _N , Z _N converted by the third conversion means 206 are used.

FIG. 8 is an explanatory diagram showing the L ^* a ^* b ^* color system. This L ^* a ^* b ^* color system is used to indicate the parameters ΔL ^* , Δa ^* , Δb ^* in the L ^* a ^* b ^* color system converted by the fifth conversion means 209.

FIG. 9 is an explanatory diagram showing a color difference diagram of the L ^* a ^* b ^* color system. The color difference diagram of the L ^* a ^* b ^* color system is used when the noise equivalent bright color difference NEED in the Lab space calculated by the fifth calculation means 210 is shown.

In FIG. 9, A is the reference color, B is the color of the measurement sample, and A ′ is the reference color at the same brightness as the color B of the measurement sample. Here, ΔL is the difference in brightness between the reference color A ′ and the reference color A at the same brightness as the color B of the measurement sample. Δa is a difference in the red direction between the reference color A ′ at the same brightness as the color B of the measurement sample and the color B of the measurement sample. Δb is a difference in the yellow direction between the reference color A ′ at the same brightness as the color B of the measurement sample and the color B of the measurement sample. ΔE ^* ab is a color difference between the reference color A and the color B of the measurement sample, that is, a noise equivalent bright color difference NEED in the Lab space.

(Explanation of effect of embodiment)
As described above, in the discrimination performance evaluation system for noise of the visible camera according to this embodiment, first, the tristimulus values B _X , _Y , _Z , X, Y, Z on the signal side of the visible camera 6 and three of the noise Noise equivalents NEXD, NEYD, NEZD are calculated from the stimulus values X _N , Y _N , Z _N. Next, the noise equivalents NEXD, NEYD, NEZD are converted into parameters ΔL ^* , Δa ^* , Δb ^* in the L ^* a ^* b ^* color system. Then, the numerical value of the noise equivalent bright color difference NEED in the Lab space is calculated from the parameters ΔL ^* , Δa ^* , Δb ^* in the L ^* a ^* b ^* color system.

  That is, the visual recognition noise evaluation performance evaluation system according to this embodiment can express a noise equivalent bright color difference in a three-dimensional color space. Thereby, the identification performance evaluation system for noise of the visible camera according to this embodiment can objectively and specifically evaluate the identification performance for noise of the visible camera numerically.

  Although the embodiments of the present invention have been described above, it will be obvious to those skilled in the art that the present invention is not limited to these embodiments, and various modifications and changes can be made within the scope of the claims.

It is a block diagram which shows the function of the hardware of one example of embodiment concerning the identification performance evaluation system with respect to the noise of the visible camera of this invention. It is a block diagram which shows the hardware structure of an example of embodiment concerning the identification performance evaluation system with respect to the noise of the visible camera of this invention. Similarly, it is a block diagram which shows an outline. Similarly, it is explanatory drawing which shows the image imaged with the different visible camera similarly. Similarly, (A) is an explanatory diagram showing specific numerical values of the spectral and signal levels of the visible camera B, (B) is an explanatory diagram showing specific numerical values of the signal of the visible camera B, and (C) is noise of the visible camera B. FIG. 3D is an explanatory diagram showing specific numerical values of pp and noise level, and FIG. 4D is an explanatory diagram showing specific numerical values of noise of the visible camera B. FIG. Similarly, (A) is an explanatory diagram showing specific numerical values of NEXYZD of the visible camera B, (B) is an explanatory diagram showing specific numerical values of the Lab of the visible camera B, and (C) is a specific NEED of the visible camera B. Explanatory diagrams showing numerical values, (D) are explanatory diagrams showing specific numerical values of NEED of the A image by the visible camera A and NEED of the B image by the visible camera B. FIG. It is explanatory drawing which shows XYZ (Yxy) color system. It is explanatory drawing which shows a L ^* a ^* b ^* color system. It is explanatory drawing which shows the color difference figure of L ^* a ^* b ^* color system.

Explanation of symbols

1 Input device (input means)
11 Illumination spectral characteristics of reference illumination 12 Spectral reflectance of reference chart 13 Spectral characteristics of visible camera 14 Focal length of lens 15 Spectral transmittance of lens 16 System noise of visible camera 2 Calculation means 201 First calculation means 202 First conversion means 203 2nd operation means 204 2nd conversion means 205 3rd operation means 206 3rd conversion means 207 4th conversion means 208 4th operation means 209 5th conversion means 210 5th operation means 3 Output device (output means)
4 Reference illumination 5 Reference chart 6 Visible camera 7 Lens

Claims (6)

A method for evaluating the discrimination performance of a visible camera with respect to noise by numerical values,
A first calculation step in which a first calculation means calculates a numerical value of the reflected light characteristic from the numerical value of the illumination spectral characteristic of the reference illumination input by the input means and the numerical value of the spectral reflectance of the reference chart;
A first conversion step in which a first conversion means converts the value of the reflected light characteristic calculated in the first calculation step into a value of a tristimulus value of reflected light;
The second calculation means includes the numerical value of the illumination spectral characteristic, the numerical value of the spectral reflectance, the numerical value of the spectral characteristic of the visible camera, the numerical value of the focal length of the lens, and the spectral transmission of the lens. A second calculation step for calculating the sensitivity characteristic value of the visible camera from the rate value;
A second conversion step in which a second conversion means converts the numerical value of the susceptibility characteristic calculated in the second calculation step into a numerical value of a tristimulus value of a signal;
The third arithmetic means calculates the stimulation value per unit light from the numerical value of the tristimulus value of the reflected light converted by the first conversion step and the numerical value of the tristimulus value of the signal converted by the second conversion step. A third calculation step for calculating the numerical value of
A third conversion step, wherein the third conversion means converts the numerical value of the system noise of the visible camera input by the input means into the numerical value of the tristimulus value of noise;
A fourth conversion step, wherein a fourth conversion means converts the numerical value of the tristimulus value of the noise converted by the third conversion step into a numerical value of the stimulus value of noise;
The fourth calculation means calculates a noise equivalent value from the value of the stimulus value per unit light calculated in the third calculation step and the value of the stimulus value of the noise converted in the fourth conversion step. A fourth calculation step;
A fifth conversion step, wherein the fifth conversion means converts the noise-equivalent numerical value calculated in the fourth calculating step into a numerical value of a parameter in the L ^* a ^* b ^* color system;
A fifth calculation step in which a fifth calculation means calculates a numerical value of a noise equivalent bright color difference in the Lab space from a numerical value of a parameter in the L ^* a ^* b ^* color system converted by the fifth conversion step;
A method for evaluating discrimination performance against noise of a visible camera, comprising:   A non-transitory computer-readable storage medium storing a program for causing a computer to execute each step of the identification performance evaluation method for noise of a visible camera according to claim 1.   A computer-readable recording medium recording a program for causing a computer to execute each step of the identification performance evaluation method for noise of a visible camera according to claim 1. A device that evaluates the discrimination performance against noise of a visible camera numerically,
The numerical value of the illumination spectral characteristic of the reference illumination, the numerical value of the spectral reflectance of the reference chart, the numerical value of the spectral characteristic of the visible camera, the numerical value of the focal length of the lens, the numerical value of the spectral transmittance of the lens, and the system of the visible camera An input means for inputting a numerical value of noise;
First calculation means for calculating a numerical value of reflected light characteristics from the numerical value of the illumination spectral characteristic input by the input means and the numerical value of the spectral reflectance;
First conversion means for converting the numerical value of the reflected light characteristic calculated by the first calculation means into a numerical value of tristimulus values of reflected light;
From the numerical value of the illumination spectral characteristic, the numerical value of the spectral reflectance, the numerical value of the spectral characteristic, the numerical value of the focal length, and the numerical value of the spectral transmittance input by the input unit, the sensitivity of the visible camera is obtained. A second calculating means for calculating a numerical value of the characteristic;
Second conversion means for converting the numerical value of the sensibility characteristic calculated by the second calculation means into a numerical value of a tristimulus value of a signal;
A first value for calculating a stimulus value per unit light is calculated from the value of the tristimulus value of the reflected light converted by the first conversion unit and the value of the tristimulus value of the signal converted by the second conversion unit. 3 computing means;
Third conversion means for converting the numerical value of the system noise input by the input means into a numerical value of a tristimulus value of noise;
Fourth conversion means for converting the value of the tristimulus value of the noise converted by the third conversion means into the value of the stimulus value of noise;
Fourth arithmetic means for calculating a noise equivalent numerical value from the numerical value of the stimulus value per unit light calculated by the third arithmetic means and the numerical value of the stimulus value of the noise converted by the fourth conversion means. When,
Fifth conversion means for converting the noise-equivalent numerical value calculated by the fourth calculating means into a numerical value of a parameter in the L ^* a ^* b ^* color system;
Fifth calculation means for calculating a numerical value of noise equivalent bright color difference in Lab space from numerical values of parameters in the L ^* a ^* b ^* color system converted by the fifth conversion means;
An apparatus for evaluating discrimination performance against noise of a visible camera, comprising:   The program for functioning a computer as each means of the identification performance evaluation apparatus with respect to the noise of the visible camera described in the said Claim 4.   A computer-readable recording medium storing a program for causing a computer to function as each means of the identification performance evaluation apparatus for noise of a visible camera according to claim 4.

Images