JP4713983B2 - Color adjustment method and color adjustment apparatus - Google Patents

Description

  The present invention relates to a color adjustment method and a color adjustment apparatus that perform color adjustment of an imaging system member included in a digital camera or the like.

  An imaging system member of a general digital camera often has an imaging lens group and an imaging element (CCD element). In such an imaging system member, the light incident through the imaging lens group is imaged on the CCD element and becomes an image signal.

  Here, a color filter is arranged in front of the CCD element. This color filter has a large variation in spectral sensitivity characteristics, and has a great influence on color reproduction of an image by a digital camera. In order to minimize this effect in the digital camera, it is necessary to adjust the correction value in the color adjustment processing in the image processing unit in the subsequent stage for each digital camera.

As an example of such a color adjustment apparatus, in Patent Document 1, a color chart or a gray scale is photographed using a light source that emits light of a reference luminance level, and an image signal obtained by this photographing is within the adjustment standard. Thus, the gain value at the time of generating the color signal is adjusted.
Japanese Patent Laid-Open No. 11-167661

  Here, the imaging system member of the digital camera includes an infrared cut filter (IR cut) as an optical low-pass filter that removes an infrared component of light incident through the imaging lens group between the imaging lens group and the imaging element. In many cases, a filter is provided. The spectral sensitivity characteristics of the IR cut filter also vary widely, which greatly affects the color reproduction of the image by the digital camera. However, in Patent Document 1 described above, the spectral sensitivity characteristic of the IR cut filter inside the digital camera is considered. There is no particular mention of color adjustment.

  The present invention has been made in view of the above circumstances, and a color adjustment method capable of obtaining the spectral sensitivity characteristic of a digital camera with higher accuracy in consideration of the influence of the spectral sensitivity characteristic of an IR cut filter in the digital camera. And a color adjusting device.

In order to achieve the above object, a color adjustment method according to a first aspect of the present invention includes a predetermined light source that is one of the constituent members of the color adjustment apparatus and an imaging system member that is one of the constituent members of the digital camera. The infrared cut filter is interposed between the predetermined light source and the imaging system member and the photographing performed without interposing the infrared cut filter which is one of the constituent members of the color adjusting device between At least imaging to be performed, a plurality of image data is acquired, and the acquired plurality of image data is processed by paying attention to a cutoff start wavelength of the infrared cut filter, so that the imaging system member Spectral sensitivity characteristics are provisionally predicted. Spectral sensitivity characteristic data of the imaging system member is obtained from the provisionally predicted spectral sensitivity characteristics, the spectral sensitivity characteristics of the predetermined light source, and the spectral sensitivity characteristics of the infrared cut filter. Final plan Characterized in that it.

In order to achieve the above object, a color adjustment apparatus according to a second aspect of the present invention includes a predetermined light source, a digital camera that captures the predetermined light source and acquires image data, and the predetermined light source. An infrared cut filter that removes an infrared light component of light from the predetermined light source, spectral sensitivity characteristics of the predetermined light source, and a digital camera that serves as a reference A reference data storage unit that stores at least the spectral sensitivity characteristics of the reference infrared spectral filter and the spectral sensitivity characteristics of a reference infrared cut filter as reference data, and the infrared cut filter interposed between the predetermined light source and the digital camera. let not line intends imaging and upper SL at least performing an imaging intends row by interposing the infrared cut filter between the predetermined light source and the digital camera, a plurality of Acquires image data, a plurality of image data the acquired, that processing is performed focusing on the cutoff start wavelength of the infrared cut filter with the reference data stored in the reference data storage unit, a provisional spectral sensitivity prediction unit which tentatively predicted spectral sensitivity characteristics of the digital camera, the spectral sensitivity characteristic of the tentatively predicted the digital camera in the interim spectral sensitivity prediction unit, the spectral sensitivity characteristic of the predetermined light source And a final spectral sensitivity prediction unit that finally predicts spectral sensitivity characteristic data of the digital camera from the spectral sensitivity characteristics of the infrared cut filter.

  According to the first and second aspects, it is possible to obtain the spectral sensitivity characteristic with higher accuracy by finally predicting the spectral sensitivity characteristic of the digital camera from the temporarily estimated spectral sensitivity characteristic.

  According to the present invention, it is possible to provide a color adjustment method and a color adjustment apparatus capable of obtaining the spectral sensitivity characteristic of a digital camera with higher accuracy in consideration of the influence of the spectral sensitivity characteristic of the IR cut filter in the digital camera. it can.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is an overall configuration diagram of a color adjustment apparatus 10 according to an embodiment of the present invention.
Referring to the configuration of FIG. 1 schematically, a viewer 13 having a light source 12 and a digital camera (hereinafter simply referred to as a digital camera) 14 having a light source 12 in a light shielding member 11 that shields external light from entering. And the viewer 13 can be photographed by the digital camera 14. As the light source 12, for example, an A light source is used. The spectral sensitivity characteristic of the A light source is shown as characteristic L21 in FIG. As the light source 12, a light source other than the A light source can be used as long as the radiation level continuously and monotonously increases from the short wavelength side to the long wavelength side.

  In FIG. 1, an equipment IR cut filter 15 is disposed between the viewer 13 and the digital camera 14 so as to be freely inserted or retracted. Here, the equipment IR cut filter 15 is attached to the attachment member 16, and the attachment member 16 is further attached to the filter moving device 17.

  FIG. 3 is a front view of the equipment IR cut filter 15, the attachment member 16, and the filter moving device 17. As shown in FIG. 3, the attachment member 16 is movably attached to the filter moving device 17, and the equipment IR cut filter 15 is moved by sliding the attachment member 16 in the direction of the arrow shown in the figure. It is inserted between the viewer 13 and the digital camera 14 to remove a specific wavelength component (infrared component) of light incident through the viewer 13, or the equipment IR cut filter 15 is inserted between the viewer 13 and the digital camera 14. It can be evacuated.

  In FIG. 1, the digital camera 14 and the filter moving device 17 are connected to a personal computer 20 via connection cables 18 and 19, respectively. A memory 21 and a CPU 22 are provided in the personal computer 20 as a reference data storage unit, provisional spectral sensitivity prediction unit, and final spectral sensitivity prediction unit. The memory 21 is a memory including an adjustment program storage unit 21a for storing an adjustment program, a reference data storage unit 21b for storing reference data to be described later, and the like. The CPU 22 controls the digital camera 14 and the filter moving device 17 to operate at a desired timing based on data stored in the memory 21 and controls adjustment by the color adjusting device 10.

  FIG. 4 is a configuration diagram of imaging system members used in the digital camera 14. As shown in the figure, the light incident through the imaging lens group 30 passes through an IR cut filter 31 as an optical low-pass filter and is removed from the harmful infrared component, and then incident on the CCD unit 32. The Here, as the equipment IR cut filter 15 attached to the filter moving device 17 via the attachment member 16, an infrared cut wavelength having a shorter wavelength than the cut wavelength of the IR cut filter 31 of the digital camera 14 is used. To.

  The CCD unit 32 includes a microlens array 32a, a color filter 32b, and a CCD element 32c. The light collected for each pixel by the micro lens array 32a of the CCD unit 32 is imaged on the CCD element 32c via, for example, a color filter 32b having a primary color Bayer array. Here, the primary color Bayer array color filter is a color filter having four pixels of R, Gr, Gb, and B as one pixel array unit.

  FIG. 5 is a flowchart showing the entire processing contents of the color adjustment method of the digital camera 14 set as shown in FIG. The processing in FIG. 5 is executed by the CPU 22 in accordance with a program stored in the adjustment program storage unit 21a in the personal computer 20 in FIG.

In FIG. 5, first, the CPU 22 causes the digital camera 14 to photograph the viewer 13 with the equipment IR cut filter 15 retracted from between the viewer 13 and the digital camera 14, and digitally emits the light from the light source 12. The camera 14 receives light (step S1). Then, the CPU 22 acquires image data obtained by photographing at this time (step S2). As shown in FIG. 6, the image data acquired in step S <b> 2 calculates an average value of each color component of R, Gr, Gb, and B of 32 pixels in the center × 32 pixels in the center of the image obtained by shooting. , Especially for Gr and Gb,
G = (Gr + Gb) / 2 (Formula 1)
And As a result, the average value (white level of the image data) of the R, G, and B color components is calculated. In this case, the R and B components are the average of 256 pixels, and the G component is the average of 512 pixels. Hereinafter, the white level R component, G component, and B component obtained in step S2 in this way are denoted as White_r, White_g, and White_b, respectively.

  Next, the CPU 22 arranges the equipment IR cut filter 15 between the viewer 13 and the digital camera 14, causes the digital camera 14 to photograph the viewer 13, and converts the light from the light source 12 into the equipment IR cut filter 15. The light is received by the digital camera 14 via (step S3). Then, the CPU 22 acquires the white level of the image data, which is the average value of the R, G, and B color components, from the image data obtained by the photographing at this time (step S4). Hereinafter, the white level R component, G component, and B component obtained in step S4 are denoted as White_ir_r, White_ir_g, and White_ir_b, respectively.

  Next, the CPU 22 determines the spectral sensitivity characteristics of the digital camera 14 from the white level acquired in steps S2 and S4 and reference data (details will be described later) stored in the reference data storage unit 21b inside the personal computer 20. Prediction is performed (step S5), and the series of processes in FIG. 5 is terminated.

Next, the spectral sensitivity characteristic prediction calculation in step S5 in FIG. 5 will be described.
First, reference data stored in the reference data storage unit 21b will be described. This reference data is reference data having a fixed value used in later calculations.

  One of the reference data is the spectral sensitivity characteristic of the light source 12 shown in FIG. Hereinafter, the spectral sensitivity characteristic of the light source 12 is expressed as HGVWR_spct_ref_x. Here, x represents the horizontal axis in FIG. 2, that is, the wavelength value, and is a value that changes every 1 [nm] within a range of 380 ≦ x ≦ 720. Also in the reference data thereafter, x represents a wavelength value that changes every 1 [nm] within a range of 380 ≦ x ≦ 720.

  One of the reference data is a spectral sensitivity characteristic of an IR cut filter (hereinafter referred to as a reference IR cut filter) serving as a predetermined reference. This characteristic is indicated by L41 in FIG. As shown in FIG. 7, in the reference IR cut filter, the spectral transmittance starts to decrease near 620 [nm], and the spectral transmittance becomes almost zero near 640 [nm]. Hereinafter, the wavelength of 620 [nm] is referred to as a cutoff start wavelength, and the wavelength of 640 [nm] is referred to as a cutoff end wavelength. The spectral sensitivity characteristic of the reference IR cut filter is denoted as IR_spct_ref_x. As the reference IR cut filter, for example, the equipment IR cut filter 15 may be used, or another IR cut filter may be used.

  One of the reference data is a long wavelength band of the CCD element, that is, a reference spectral sensitivity characteristic of the R component. This characteristic is indicated by L81 in FIG. Hereinafter, this R reference spectral sensitivity characteristic is expressed as R_ccd_ref_x.

  One of the reference data is an intermediate wavelength band of the CCD element, that is, a reference spectral sensitivity characteristic of the G component. This characteristic is indicated by L82 in FIG. Hereinafter, this G reference spectral sensitivity characteristic is expressed as G_ccd_ref_x.

  One of the reference data is a short wavelength band of the CCD element, that is, a reference spectral sensitivity characteristic of the B component. This characteristic is indicated by L83 in FIG. Hereinafter, this B reference spectral sensitivity characteristic is expressed as B_ccd_ref_x.

  Here, the R reference spectral sensitivity characteristic L81, the G reference spectral sensitivity characteristic L82, and the B reference spectral sensitivity characteristic L83 shown in FIG. 8 are normalized relative spectra so that the maximum spectral sensitivity of the G reference spectral sensitivity characteristic L82 is 100. Shown as sensitivity.

  One of the reference data is a product of the R reference spectral sensitivity characteristic L81 and the spectral sensitivity characteristic L41 of the reference IR cut filter, which is 1 [nm] in the visible wavelength band (380 [nm] to 720 [nm]). This is a waveform obtained every time. This characteristic is indicated by L84 in FIG. Hereinafter, this reference data is denoted as R_IR_ccd_ref_x.

  One of the reference data is the product of the R reference spectral sensitivity characteristic L81 and the spectral sensitivity characteristic L21 of the light source 12 for each 1 [nm] in the visible wavelength range (380 [nm] to 720 [nm]). It is an integrated value obtained by calculating and summing them. Hereinafter, this reference data is expressed as White_r_ref.

  One of the reference data is a product of the G reference spectral sensitivity characteristic L82 and the spectral sensitivity characteristic L21 of the light source 12 for each [nm] in the visible wavelength range (380 [nm] to 720 [nm]). , An integrated value obtained by calculating the sum thereof. Hereinafter, this reference data is expressed as White_g_ref.

  One of the reference data is a product of the B reference spectral sensitivity characteristic L83 and the spectral sensitivity characteristic L21 of the light source 12 for each [nm] in the visible wavelength band (380 [nm] to 720 [nm]). , An integrated value obtained by calculating the sum thereof. Hereinafter, this reference data is expressed as White_b_ref.

  Also, one of the reference data is the product of the R reference spectral sensitivity characteristic L81, the spectral sensitivity characteristic L21 of the light source 12, and the spectral sensitivity characteristic L41 of the reference IR cut filter in the visible wavelength band (380 [nm] to 720 [nm]. ) Is an integral value obtained every 1 [nm] in the range of) and the sum thereof. Hereinafter, this reference data is expressed as White_ir_r_ref.

  One of the reference data is the product of the G reference spectral sensitivity characteristic L82, the spectral sensitivity characteristic L21 of the light source 12, and the spectral sensitivity characteristic L41 of the reference IR cut filter in the visible wavelength band (380 [nm] to 720 [nm]). ) Is an integral value obtained every 1 [nm] in the range of) and the sum thereof. Hereinafter, this reference data is expressed as White_ir_g_ref.

  Hereinafter, a method for predicting the spectral sensitivity characteristic of the digital camera 14 using the above-described reference data will be described.

First, based on the white level acquired in step S2 of FIG. 5, a relative sensitivity coefficient kR for G of R and a relative sensitivity coefficient kB for G of B are respectively determined.
kR = (White_r / White_g) / (White_r_ref / White_g_ref)
kB = (White_b / White_g) / (White_b_ref / White_g_ref) (Formula 2)
Ask from.

Next, R spectral first detection data R_ccd_est1_x, G spectral first detection data G_ccd_est1_x, and B spectral first detection data B_ccd_est1_x defined as follows are obtained.
R_ccd_est1_x = R_ccd_ref_x × kR (380 ≦ x ≦ 720)
G_ccd_est1_x = G_ccd_ref_x × 1 (380 ≦ x ≦ 720) (Formula 3)
B_ccd_est1_x = B_ccd_ref_x × kB (380 ≦ x ≦ 720)
Next, a value SR_ccd_est1 obtained by integrating R_ccd_est1_x in (Equation 3) in the range of 380 ≦ x ≦ 720,
SR_ccd_est1 = R_ccd_est1_380 + R_ccd_est1_381 +… + R_ccd_est1_720 (Formula 4)
Ask from.

Here, the influence of the spectral sensitivity characteristic of the IR cut filter 31 inside the digital camera 14 on the spectral sensitivity characteristic of the R component is corrected. For this purpose, the relative sensitivity coefficient kIRR for G of R when passing through the equipment IR cut filter 15 from the white level acquired in step S4 of FIG.
kIRR = (White_ir_r / White_ir_g) / (White_ir_r_ref / White_ir_g_ref) (Formula 5)
The R relative spectral sensitivity characteristic R_IR_ccd_est_x through the equipment IR cut filter 15 is predicted by multiplying the reference data R_IR_ccd_ref_x every 1 nm by the result of (Equation 5). That is,
R_IR_ccd_est_x = R_IR_ccd_ref_x × kIRR (380 ≦ x ≦ 720) (Formula 6)
Perform the operation.

Next, from the result of (Equation 6), paying attention to the spectral sensitivity at the cutoff start wavelength 620 [nm], the R relative spectral sensitivity R_ccd_est2_620 at the cutoff start wavelength 620 [nm]
R_ccd_est2_620 = R_IR_ccd_est_620 (Formula 7)
Ask from. From this result, the height coefficient kR_h is
kR_h = R_ccd_est2_620 / R_ccd_ref_620 (Formula 8)
Ask from.

Next, R spectral second detection data R_ccd_est2_x
R_ccd_est2_x = R_ccd_ref_x (380 ≦ x ≦ 540)
R_ccd_est2_x = R_ccd_ref_x × kR_h (541 ≦ x ≦ 720) (Formula 9)
It is defined as The R-spectral second detection data defined as described above is shown as a characteristic L85 in FIG. As shown in FIG. 9, the R spectral second detection data is the R reference spectral sensitivity characteristic itself indicated by the characteristic L81 in the range of 380 ≦ x ≦ 540, and the R reference spectral sensitivity characteristic in the range of 541 ≦ x ≦ 720. The height coefficient is multiplied by kR_h.

Next, the R spectral sensitivity characteristic in the wavelength band of 621 [nm] to 720 [nm] is determined in more detail from the R spectral second detection data defined as described above. For this purpose, the integrated value SR_ccd_est2 of the R spectroscopy second detection data is
SR_ccd_est2 = R_ccd_est2_380 + R_ccd_est2_381 +… + R_ccd_est2_720 (Formula 10)
Ask from.

  Thereafter, the spectral sensitivity characteristic of the R component is provisionally predicted using the value obtained as described above. The provisional calculation of R spectral sensitivity is shown in the flowcharts of FIGS.

  First, the CPU 22 calculates a difference dR = SR_ccd_est2-SR_ccd_est1 between the integrated value SR_ccd_est2 of the R spectral second detection data and the integrated value SR_ccd_est1 of the R spectral first detection data (step S21). After obtaining the difference dR in step S21, the CPU 22 determines whether the difference dR is positive or negative (step S22). If it is determined in step S22 that dR is positive, the process proceeds to step 23. If dR is negative, the process proceeds to step S35.

  If the difference dR is positive in the determination in step S22, initialization is performed by substituting the R_ccd_est2_x 621 [nm] spectral sensitivity R_ccd_est2_621 for the R spectral movement amount calculation integrated value diff (step S23). Next, initialization is performed by substituting 0 for the count value n of a counter (not shown) (step S24).

  Next, the R spectral movement amount calculation integrated value diff initialized as described above is compared with the difference dR to determine whether or not diff> dR is satisfied (step S25). If it is determined in step S25 that diff ≦ dR, the value of R_ccd_est2_ (622 + n) is added to diff (step S26). Since n = 0 at the first time, the value of R_ccd_est2_622 is added. Subsequently, n is incremented by 1 (step S27), and the process returns to step S25. That is, in the determination of step S25, the process from step S25 to step S27 is repeated while adding the value after 622 [nm] of R_ccd_est2_x in increments of 1 [nm] until diff> dR.

If it is determined in step S25 that diff> dR, the CPU 22 determines whether the count value n of the counter is 20 or less (step S28). If it is determined in step S28 that n is 20 or less, the R spectral third detection data R_ccd_est3_x is obtained as follows (step S29).
R_ccd_est3_x = R_ccd_est2_ (x + n) (621 ≦ x ≦ 700) (Formula 11)
On the other hand, if n exceeds 20 in the determination of step S28, the R spectral third detection data R_ccd_est3_x is obtained as follows (step S30).
R_ccd_est3_x = R_ccd_est2_ (x + n) (621 ≦ x ≦ 700-n)
R_ccd_est3_x = 0 (720-n + 1 ≦ x ≦ 700) (Formula 12)
In step S29 or step S30, after obtaining the R spectral third detection data within the range of 621 ≦ x ≦ 700, the CPU 22 determines the value of the R spectral third detection data R_ccd_est3_x within the range of 701 ≦ x ≦ 720. Clip to 0 (step S31).

  Next, the validity of the count value n is confirmed by determining whether or not the count value n is less than or equal to the predetermined value Max_shift (step S32). This Max_shift is assumed to be a value larger than 20, for example, 30. If n exceeds Max_shift in the determination in step S32, it is determined that there is an error in the predicted spectral characteristics, and this is output, and the provisional spectral prediction calculation is terminated.

On the other hand, if it is determined in step S32 that n is equal to or smaller than Max_shift, n is substituted into the R spectral shift amount shift (step S33). Subsequently, the x-coordinate a and the y-coordinate b of the point P82 shown in FIG. 9 and the x-coordinate c and the y-coordinate d of the point P83 shown in FIG.
a = 640-shift
b = R_ccd_est2_640
c = 620
d = R_ccd_est2_620 (Formula 13)
(Step S34). Here, a point P82 in FIG. 9 is a point obtained by laterally shifting the point P81 indicating the R spectral second detection data at the cutoff end wavelength 640 [nm] to the short wavelength side by the R spectral movement amount shift. A point P83 in FIG. 9 is a point indicating the spectral sensitivity of the R spectral second detection data at the cutoff start wavelength 620 [nm].

  In the determination of step S22, when dR is negative, the value obtained by multiplying the R spectral movement amount calculation integrated value diff by -1 to the spectral sensitivity R_ccd_est2_621 at the wavelength 621 [nm] of the R spectral second detection data R_ccd_est2_x. Is initialized by substituting (step S35). Further, the count value n is initialized to 0 as in step S24 (step S36).

  Next, the CPU 22 compares the integrated value diff for calculating the R spectral movement amount diff and the difference dR initialized as in step S35, and determines whether or not diff <dR (step S37). In the determination of step S37, if diff ≧ dR, the value of R_ccd_est2_ (622 + n) is subtracted from diff (step S38). Since n = 0 at the first time, the value of R_ccd_est2_622 is subtracted. Subsequently, n is incremented by 1 (step S39), and the process returns to step S37. That is, in the determination of step S37, the operations of step S37 to step S39 are repeated until diff <dR.

If it is determined in step S37 that diff <dR, the CPU 22 determines whether n is 0 (step S40). If n is 0 in the determination in step S40, the R spectral third detection data R_ccd_est3_x in 621 ≦ x ≦ 700 is obtained as follows (step S41).
R_ccd_est3_x = R_ccd_est2_x (621 ≦ x ≦ 700) (Formula 14)
On the other hand, if n is not 0 in the determination in step S40, first, 1 is added to n (step S42), and then the R spectral third detection data in 621 ≦ x ≦ 700 is obtained as follows (step) S43).
R_ccd_est3_x = R_ccd_est2_ (xn) (621 ≦ x ≦ 700) (Equation 15)
In step S41 or step S43, after obtaining the R spectral third detection data within the range of 621 ≦ x ≦ 700, the CPU 22 determines the value of the R spectral third detection data R_ccd_est3_x within the range of 701 ≦ x ≦ 720. Clip to 0 (step S44).

  Next, the validity of the count value n is confirmed by determining whether or not the count value n is less than or equal to the predetermined value Max_shift (step S45). If n exceeds Max_shift in the determination in step S45, it is determined that there is an error in spectral prediction, and that is output, and the provisional spectral prediction calculation is terminated.

On the other hand, if it is determined in step S45 that n is equal to or smaller than Max_shift, n is substituted into the R spectral shift amount shift (step S46). Subsequently, the above a, b, c, d are respectively
a = 640 + shift
b = R_ccd_est2_640
c = 620
d = R_ccd_est2_620 (Formula 16)
(Step S47). The point P82 in this case is a point obtained by laterally shifting the point P81 indicating the spectral sensitivity at the wavelength 640 [nm] of the R spectral second detection data to the long wavelength side by the R spectral movement amount shift (illustrated in FIG. 9). Omitted).

After a, b, c, and d are obtained in step S34 or step S47, smooth interpolation is performed between the points P82 and P83 shown in FIG. 9 using the obtained a, b, c, and d. R spectral fourth detection data R_ccd_est4_x, which is final provisional spectral characteristics, is obtained. To this end, first from the above a, b, c, d
R_ccd_est4_x = {(bd) / (ac)} × (xc) + d (621 ≦ x ≦ a-1) (Expression 17)
Is calculated (step S48). This section becomes a portion of a straight line 801 shown in FIG. Here, in the present embodiment, the point P82 and the point P83 are interpolated by the straight line 801, but curve interpolation may be performed between these points. Next, R_ccd_est4_x from the wavelength a to the wavelength 720 [nm] at the point P82 is obtained as follows (step S49).
R_ccd_est4_x = R_ccd_est3_x (Formula 18)
Then, the R spectral sensitivity characteristic R_cam_cal_x is obtained as follows based on the R spectral fourth detection data obtained in steps S46 and S47 (step S50), and the provisional spectral prediction calculation is terminated.
R_cam_cal_x = R_ccd_est2_x (380 ≦ x ≦ 620) + R_ccd_est4_x (621 ≦ x ≦ 720)
(Formula 19)
As shown in (Equation 19), the R spectral sensitivity characteristic is the R spectral second detection data as it is for the spectral sensitivity characteristic from the wavelength 380 [nm] to 620 [nm], and from 621 [nm] to 720 [nm]. ], The R spectral fourth detection data obtained from the R spectral second detection data is used. That is, the R spectral sensitivity characteristic is the sum of the characteristic L85, the straight line 801, and the characteristic L86 in FIG. By obtaining the R spectral sensitivity characteristic in this way, the R spectral sensitivity characteristic considering the influence of the spectral sensitivity characteristic of the IR cut filter 31 inside the digital camera 14 can be obtained.

Further, since the G spectral sensitivity characteristic and the B spectral sensitivity characteristic are less influenced by the IR cut filter 31 in the digital camera 14, the first spectral detection data obtained from the white level obtained without using the equipment IR cut filter 15. Is used as is. That is,
G_cam_cal_x = G_ccd_est1_x (380 ≦ x ≦ 720)
B_cam_cal_x = B_ccd_est1_x (380 ≦ x ≦ 720) (Formula 20)
It becomes. That is, the G spectral sensitivity characteristic is the characteristic L82 in FIG. 9, and the B spectral sensitivity characteristic is the characteristic L87 in FIG.

  Next, a method for predicting the final R spectral sensitivity characteristic by correcting the R spectral sensitivity characteristic predicted as described above will be described. 13 and 14 are flowcharts of the final prediction calculation of the R spectral sensitivity characteristic.

First, the CPU 22 determines whether or not “a” is not less than 640 [nm], which is a cutoff end wavelength (step S61). If it is determined in step S61 that a is 640 [nm] or more, the R spectral sensitivity characteristic is corrected. In this case, R sensitivity coefficients k_R_get and k_R_ir_get indicating the ratio of R to G as a reference of the white level from the white level acquired in step S2 and step S4 in FIG.
k_R_get = White_r / White_g
k_R_ir_get = White_ir_r / White_ir_g (Formula 21)
(Step S62).

Next, white level verification is performed to determine whether or not the spectral sensitivity characteristics provisionally predicted as described above are correct (step S63). In this white level verification, first, the first R white level verification data R_spct_chk1_x and the first G white level verification data G_spct_chk1_x are
R_spct_chk1_x = R_cam_cal_x × HGVWR_spct_ref_x (380 ≦ x ≦ 720)
G_spct_chk1_x = G_cam_cal_x × HGVWR_spct_ref_x (380 ≦ x ≦ 720) (Formula 22)
As described above, the R spectral sensitivity characteristic and the G spectral sensitivity characteristic are respectively multiplied by the spectral sensitivity characteristic of the light source 12.

Next, the first R (IR) white level calculation data R_spct_ir_chk1_x and the first R white level calculation data and the first G white level calculation data are multiplied by the spectral sensitivity characteristic of the reference IR cut filter, respectively. First G (IR) white level verification data G_spct_ir_chk1_x is obtained. That is,
R_spct_ir_chk1_x = R_spct_chk1_x × IR_spct_ref_x (380 ≦ x ≦ 720)
G_spct_ir_chk1_x = G_spct_chk1_x × IR_spct_ref_x (380 ≦ x ≦ 720) (Equation 23)
Perform the operation.

Next, by integrating the result of (Expression 22) and the result of (Expression 23) in the range of 380 ≦ x ≦ 720, the first R white level calculation data White_r_chk1, the first G white level calculation data White_g_chk1, First R (IR) white level verification data White_ir_r_chk1 and first G (IR) white level verification data White_ir_g_chk1 are obtained. That is,
White_r_chk1 = R_spct_chk1_380 + R_spct_chk1_381 +… + R_spct_chk1_720
White_g_chk1 = G_spct_chk1_380 + G_spct_chk1_381 +… + G_spct_chk1_720
White_ir_r_chk1 = R_spct_ir_chk1_380 + R_spct_ir_chk1_381 +… + R_spct_ir_chk1_720
White_ir_g_chk1 = G_spct_ir_chk1_380 + G_spct_ir_chk1_381 +… + G_spct_ir_chk1_720
(Formula 24)
Perform the operation. Here, the first R white level verification data and the first G white level verification data are the R component of the white level of the image data acquired when the viewer 13 is photographed without passing through the equipment IR cut filter 15. And the G component are obtained from the provisionally predicted spectral sensitivity characteristics, and the first R (IR) white level calculation data and the first G (IR) white level calculation data are passed through the equipment IR cut filter 15. The white level of the image data acquired when the viewer 13 is photographed in a state is obtained from the spectral sensitivity characteristic that is provisionally predicted.

After the white level verification in step S63, the CPU 22 calculates R sensitivity coefficients k_R_chk1 and k_R_ir_chk1 based on the predicted values from the result of (Equation 24).
k_R_chk1 = White_r_chk1 / White_g_chk1
k_R_ir_chk1 = White_ir_r_chk1 / White_ir_g_chk1 (Formula 25)
(Step S64).

Next, the difference dk_R_chk1 and dk_R_ir_chk1 between the R sensitivity coefficient based on the actual measurement value obtained as in (Equation 21) and the R sensitivity coefficient based on the prediction value obtained as in (Equation 25),
dk_R_chk1 = | (k_R_get / k_R_chk1) -1 |
dk_R_ir_chk1 = | (k_R_ir_get / k_R_ir_chk1) -1 | (Formula 26)
(Step S65). Thereafter, a vertical correction coefficient k_adj_vt for correcting in the vertical direction of the R spectral sensitivity characteristic (that is, the direction of the relative spectral sensitivity axis which is the vertical axis in FIG. 9),
k_adj_vt = k_R_ir_get / k_R_ir_chk1 (Formula 27)
(Step S66).

From the vertical correction coefficient k_adj_vt thus determined, the CPU 22 corrects the R spectral sensitivity characteristic in the vertical direction to obtain vertical correction R spectral sensitivity characteristic data R_cam_adjvt_x (step S67). However, this vertical correction is performed only within the range of 541 [nm] to a [nm]. That is, the vertical correction R spectral sensitivity characteristic data obtained by this vertical correction is
R_cam_adjvt_x = R_cam_cal_x (380 ≦ x ≦ 540)
R_cam_adjvt_x = R_cam_cal_x × k_adj_vt (541 ≦ x ≦ a)
R_cam_adjvt_x = R_cam_cal_x (a + 1 ≦ x ≦ 720) (Formula 28)
It becomes. Next, these vertical correction R spectral sensitivity characteristic data are set as corrected R spectral sensitivity characteristic data R_cam_adj_x (step S68). The corrected R spectral sensitivity characteristic data indicates the R spectral sensitivity characteristic data after the correction is completed.

  Next, the CPU 22 corrects the R spectral sensitivity coefficient in the horizontal direction (the direction of the wavelength axis, which is the horizontal axis in FIG. 9). For this purpose, the CPU 22 initializes the count value m of a counter (not shown) to 0 (step S69). Next, the CPU 22 determines whether or not the count value m of the counter is equal to or lower than the upper limit value mLim (step S70). This mLim is, for example, 5 but a changeable value.

When the count value m is not more than mLim in the determination in step S70, the CPU 22 performs the second white level verification (step S71). The white level verification here is performed by replacing the R spectral sensitivity characteristic R_cam_cal_x in (Equation 22) described above with R_cam_adj_x.
R_spct_chk2_x = R_cam_adj_x × HGVWR_spct_ref_x (380 ≦ x ≦ 720) (Equation 29)
The second R white level check data R_spct_ir_chk2_x is obtained by multiplying the second R white level check data obtained thereby by the spectral sensitivity characteristic of the reference IR cut filter. That is,
R_spct_ir_chk2_x = R_spct_chk2_x × IR_spct_ref_x (380 ≦ x ≦ 720) (Equation 30)
Perform the operation. Next, by integrating the result of (Expression 29) and the result of (Expression 30) in the range of 380 ≦ x ≦ 720, the second R white level verification data White_r_chk2 and the second R (IR) white level verification are performed. Data White_ir_r_chk2 is obtained. That is,
White_r_chk2 = R_spct_chk2_380 + R_spct_chk2_381 +… + R_spct_chk2_720
White_ir_r_chk2 = R_spct_ir_chk2_380 + R_spct_ir_chk2_381 +… + R_spct_ir_chk2_720
(Formula 31)
Perform the operation. After this, R sensitivity coefficients k_R_chk2 and k_R_ir_chk2 are calculated from the result of (Equation 31).
k_R_chk2 = White_r_chk2 / White_g_chk1
k_R_ir_chk2 = White_ir_r_chk2 / White_ir_g_chk1 (Formula 32)
It asks like this.

After performing the white level verification in step S71, the CPU 22 calculates the difference dk_R_chk2 and dk_R_ir_chk2 of the R sensitivity coefficient,
dk_R_chk2 = | (k_R_get / k_R_chk2) -1 |
dk_R_ir_chk2 = | (k_R_ir_get / k_R_ir_chk2) -1 | (Formula 33)
(Step S72). Next, the CPU 22 determines whether or not the difference dk_R_chk2 of the R sensitivity coefficient is equal to or less than a predetermined value 0.01, thereby determining the R sensitivity coefficient obtained based on the actual measurement value and the R sensitivity coefficient obtained based on the prediction. It is determined whether or not the error (accuracy error) is within an allowable range (step S73). The predetermined value 0.01 is an example and can be changed.

If the difference dk_R_chk2 exceeds 0.01 in the determination in step S73, horizontal correction is performed. For this purpose, the CPU 22 calculates a horizontal correction coefficient k_adj_hz defined below (step S74).
k_adj_hz = k_R_get / k_R_chk2 (Formula 34)
Next, the CPU 22 increments the count value m of the counter by 1 (step S75). Thereafter, the CPU 22 determines whether or not the horizontal correction coefficient k_adj_hz obtained as in (Equation 34) is smaller than 1 (step S76). If it is determined in step S76 that the horizontal correction coefficient k_adj_hz is smaller than 1, the vertical correction R spectral sensitivity characteristic data R_cam_adjvt_x in step S67 is leftward (short wavelength direction) within the range of a−m ≦ x ≦ 700. Horizontal corrected R spectral sensitivity characteristic data R_cam_adjhz_x, which is the shifted characteristic data, is obtained (step S77). That is,
R_cam_adjhz_x = R_cam_adjvt_ (x + m) (am ≦ x ≦ 700) (Formula 35)
And Note that horizontal correction is not performed for other ranges.

  Next, the horizontal correction R spectral sensitivity characteristic data obtained in this way is smoothed (step S78). This smoothing is a process of comparing the relative spectral sensitivities between adjacent wavelengths and smoothly interpolating the spectral waveform based on this magnitude relationship.

In this smoothing, first, the horizontal correction R spectral sensitivity characteristic data R_cam_adjhz_ (a−m + 1) obtained in step S77 is compared with R_cam_adjvt_ (am) which is spectral sensitivity characteristic data at the previous wavelength. , R_cam_adjhz_ (a−m + 1)> R_cam_adjvt_ (am) is determined. When R_cam_adjhz_ (a−m + 1)> R_cam_adjvt_ (am), the spectral sensitivity characteristic data R_cam_adjsmth_ (a−m + 1) after smoothing at the wavelength a−m + 1 [nm] is converted to the previous wavelength. The spectral sensitivity characteristic data at a-m [nm]. That is,
R_cam_adjsmth_ (a-m + 1) = R_cam_adjvt_ (am) (Formula 36)
And On the other hand, when R_cam_adjhz_ (a−m + 1) ≦ R_cam_adjvt_ (am), the spectral sensitivity characteristic data R_cam_adjsmth_ (a−m + 1) after smoothing at the wavelength a−m + 1 [nm] is changed to the same wavelength a. The horizontal correction R spectral sensitivity characteristic data at −m + 1 [nm]. That is,
R_cam_adjsmth_ (a-m + 1) = R_cam_adjhz_ (a-m + 1) (Formula 37)
And

Thereafter, it is similarly determined whether or not R_cam_adjhz_ (x + 1)> R_cam_adjsmth_x while increasing x by 1 in the range of a−m + 1 ≦ x ≦ 700. If R_cam_adjhz_ (x + 1)> R_cam_adjsmth_x Uses the spectral sensitivity characteristic data R_cam_adjsmth_ (x + 1) after smoothing at the wavelength x + 1 [nm] as spectral sensitivity characteristic data at the previous wavelength x [nm]. That is,
R_cam_adjsmth_ (x + 1) = R_cam_adjsmth_x (Formula 38)
And On the other hand, when R_cam_adjhz_ (x + 1) ≦ R_cam_adjsmth_x, the spectral sensitivity characteristic data R_cam_adjsmth_ (x + 1) after smoothing at the wavelength x + 1 [nm] is converted into the horizontal correction R spectral sensitivity at the same wavelength x + 1 [nm]. Characteristic data. That is,
R_cam_adjsmth_ (x + 1) = R_cam_adjhz_ (x + 1) (Formula 39)
And

On the other hand, if it is determined in step S76 that the horizontal correction coefficient k_adj_hz is 1 or more, the vertical correction R spectral sensitivity characteristic data R_cam_adjvt_x in step S67 is set to the right (long wavelength direction) within the range of a ≦ x ≦ 700. Horizontal corrected R spectral sensitivity characteristic data R_cam_adjhz_x, which is the shifted characteristic data, is obtained (step S79). That is,
R_cam_adjhz_x = R_cam_adjvt_ (xm) (a ≦ x ≦ 700) (Formula 40)
And Note that horizontal correction is not performed for other ranges.

Next, the horizontal correction R spectral sensitivity characteristic data obtained in this way is smoothed (step S80).
In this smoothing, first, it is determined whether or not R_cam_adjhz_ (a + 1)> R_cam_adjvt_ (a) in the horizontal correction R spectral sensitivity characteristic data obtained in step S79, and R_cam_adjhz_ (a + 1)> In the case of R_cam_adjvt_ (a), the spectral sensitivity characteristic data R_cam_adjsmth_ (a + 1) after smoothing at the wavelength a + 1 [nm] is set as the spectral sensitivity characteristic data at the previous wavelength a [nm]. That is,
R_cam_adjsmth_ (a + 1) = R_cam_adjvt_ (a) (Formula 41)
And On the other hand, when R_cam_adjhz_ (a + 1) ≦ R_cam_adjvt_ (a), the spectral sensitivity characteristic data R_cam_adjsmth_ (a + 1) after smoothing at the wavelength a + 1 [nm] is horizontally corrected at the same wavelength a + 1 [nm]. R spectral sensitivity characteristic data. That is,
R_cam_adjsmth_ (a + 1) = R_cam_adjhz_ (a + 1) (Formula 42)
And

Thereafter, it is similarly determined whether or not R_cam_adjhz_ (x + 1)> R_cam_adjsmth_x while increasing x by 1 in the range of a + 1 ≦ x ≦ 700. If R_cam_adjhz_ (x + 1)> R_cam_adjsmth_x, the wavelength is determined. The spectral sensitivity characteristic data R_cam_adjsmth_ (x + 1) after smoothing at x + 1 [nm] is taken as spectral sensitivity characteristic data at the previous wavelength x [nm]. That is,
R_cam_adjsmth_ (x + 1) = R_cam_adjsmth_x (Formula 43)
And On the other hand, when R_cam_adjhz_ (x + 1) ≦ R_cam_adjsmth_x, the spectral sensitivity characteristic data R_cam_adjsmth_ (x + 1) after smoothing at the wavelength x + 1 [nm] is converted into the horizontal correction R spectral sensitivity at the same wavelength x + 1 [nm]. Characteristic data. That is,
R_cam_adjsmth_ (x + 1) = R_cam_adjhz_ (x + 1) (Formula 44)
And

Next, final corrected R spectral sensitivity characteristic data R_cam_adj_x is calculated from the result of step S78 or step S80 (step S81). If the horizontal correction coefficient k_adj_hz is smaller than 1 in the determination in step S76, the processing after step S77 is performed, so that the corrected R spectral sensitivity characteristic data R_cam_adj_x is
R_cam_adj_x = R_cam_adjvt_x (380 ≦ x ≦ am)
R_cam_adj_x = R_cam_adjsmth_x (a-m + 1 ≦ x ≦ 700)
R_cam_adj_x = R_cam_adjvt_x (701 ≦ x ≦ 720) (Formula 45)
It becomes. On the other hand, when the horizontal correction coefficient k_adj_hz is 1 or more in the determination in step S76, the processing after step S79 is performed, so that the corrected R spectral sensitivity characteristic data R_cam_adj_x is:
R_cam_adj_x = R_cam_adjvt_x (380 ≦ x ≦ a)
R_cam_adj_x = R_cam_adjsmth_x (a + 1 ≦ x ≦ 700)
R_cam_adj_x = R_cam_adjvt_x (701 ≦ x ≦ 720) (Formula 46)
It becomes. Note that the corrected R spectral sensitivity characteristic data in a state in which the correction in the horizontal direction is performed is indicated by a characteristic L88 in FIG.

  After calculating the corrected R spectral sensitivity characteristic data as described above, the process returns to step S70. Thereafter, the processing of step S70 to step S81 is repeated until the count value m of the counter exceeds mLim in step S70 or until the difference dk_R_chk2 becomes 0.01 or less in the determination of step S73.

When m exceeds mLim in the determination in step S70 or when the difference dk_R_chk2 is 0.01 or less in the determination in step S73, the CPU 22 ends the correction and the difference dk_R_ir_chk2 is 0.01 or less. By determining whether or not, it is determined whether or not the above correction is appropriate (step S82). If the difference dk_R_ir_chk2 is equal to or smaller than 0.01 in the determination in step S82, the CPU 22 determines final R spectral sensitivity characteristic data R_cam_calfinal_x as follows (step S83). ), Final prediction of R spectral sensitivity characteristic data is terminated.
R_cam_calfinal_x = R_cam_adj_x (380 ≦ x ≦ 720) (Formula 47)
That is, in this case, one of the results of (Expression 28), (Expression 45), and (Expression 46) is used as the final R spectral sensitivity characteristic data.

On the other hand, if the difference dk_R_ir_chk2 exceeds 0.01 in the determination of step S82, the CPU 22 determines the final R spectral sensitivity characteristic data R_cam_calfinal_x as follows, assuming that the correction so far is inappropriate. (Step S84), and final prediction of the R spectral sensitivity characteristic data is completed.
R_cam_calfinal_x = R_cam_cal_x (380 ≦ x ≦ 720) (Formula 48)
That is, in this case, the result of (Equation 19) is used as final R spectral sensitivity characteristic data. If it is determined in step S21 that a does not exceed 640 [nm], the R spectral sensitivity characteristic data is not corrected. Also in this case, the process branches to step S84, and the result of (Equation 19) is used as the final R spectral sensitivity characteristic data.

  The G spectral sensitivity characteristic data and the B spectral sensitivity characteristic data use the result of (Equation 20).

  After calculating the spectral sensitivity characteristics of the RGB color components as described above, color adjustment can be performed using these spectral sensitivity characteristics. As the color adjustment method, a known method disclosed in, for example, Japanese Patent Application Laid-Open No. 2001-320716 may be used, and the description thereof is omitted here.

  As described above, according to the present embodiment, the R spectral sensitivity characteristic is tentatively predicted by paying attention to the wavelength 620 [nm] which is the cutoff start wavelength of the IR cut filter. Focusing on the wavelength 640 [nm], which is the cutoff end wavelength of the cut filter, corrects the waveform of the R spectral sensitivity characteristic in the vertical and horizontal directions, so the spectral sensitivity of the IR cut filter 31 provided in the digital camera 14 The R spectral sensitivity characteristic can be obtained with higher accuracy in consideration of the influence of the characteristic.

  Further, since the calculation is performed from the difference between the R sensitivity coefficient based on the actual measurement value and the R sensitivity coefficient based on the predicted value, the prediction accuracy of the R spectral sensitivity characteristic can be further improved.

  Although the present invention has been described based on the above embodiments, the present invention is not limited to the above-described embodiments, and various modifications and applications are naturally possible within the scope of the gist of the present invention.

  Further, the above-described embodiments include various stages of the invention, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be extracted as an invention.

It is a figure which shows the structure of the whole color adjustment apparatus which concerns on one Embodiment of this invention. It is a figure which shows the spectral sensitivity characteristic of A light source. It is a front view of an equipment IR cut filter, a mounting member, and a filter moving device. It is a figure which shows the structure of the general imaging system member of a digital camera. It is a flowchart shown about the flow of a process of the color adjustment method which concerns on one Embodiment of this invention. It is a figure which shows the color component in an image. It is a figure which shows the spectral sensitivity characteristic of a reference IR cut filter. It is a figure shown about R reference spectral sensitivity characteristic, G reference spectral sensitivity characteristic, and B reference spectral sensitivity characteristic. It is a figure shown about each characteristic calculated | required by R spectroscopy provisional prediction calculation. It is FIG. 1 of the flowchart shown about the flow of a process of R spectroscopy provisional prediction calculation. It is FIG. 2 of the flowchart shown about the flow of a process of R spectrum provisional prediction calculation. It is FIG. 3 of the flowchart shown about the flow of a process of R spectrum provisional prediction calculation. It is FIG. 1 of the flowchart shown about the flow of a process of R spectroscopy final prediction calculation. It is FIG. 2 of the flowchart shown about the flow of a process of R spectroscopy final prediction calculation. It is a figure shown about an example of the R spectral sensitivity characteristic estimated by R spectral final prediction calculation.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Color adjusting device, 11 ... Shading member, 12 ... Light source, 13 ... Viewer, 14 ... Digital camera, 15 ... Equipment IR cut filter, 16 ... Mounting member, 17 ... Filter moving device, 18, 19 ... Connection cable, 20 ... Personal computer, 21 ... Memory, 21a ... Adjustment program storage unit, 21b ... Reference data storage unit, 22 ... CPU, 30 ... Imaging lens group, 31 ... IR cut filter, 32 ... CCD unit, 32a ... Micro lens array, 32b ... Color filters, 32c ... CCD elements

Claims (9)

An infrared cut filter that is one of the components of the color adjustment device is interposed between a predetermined light source that is one of the components of the color adjustment device and an imaging system member that is one of the components of the digital camera. Performing at least photographing performed without performing the photographing performed with the infrared cut filter interposed between the predetermined light source and the imaging system member, and acquiring a plurality of image data,
By processing the acquired plurality of image data by paying attention to the cutoff start wavelength of the infrared cut filter, the spectral sensitivity characteristic of the imaging system member is provisionally predicted,
Final prediction of spectral sensitivity characteristic data of the imaging system member from the provisionally predicted spectral sensitivity characteristic, the spectral sensitivity characteristic of the predetermined light source, and the spectral sensitivity characteristic of the infrared cut filter,
A color adjustment method for a digital camera. The plurality of image data and the provisionally predicted spectral sensitivity characteristic include data relating to a red component and data relating to a green component, respectively.
The final forecast is
Based on the data regarding the red component of the provisionally predicted spectral sensitivity characteristic and the spectral sensitivity characteristic of the predetermined light source, the infrared cut filter is not interposed between the predetermined light source and the imaging system member. Calculate the data for the red component of the corresponding white level for verification,
Based on the data on the green component of the provisionally predicted spectral sensitivity characteristic and the spectral sensitivity characteristic of the predetermined light source, the infrared cut filter is not interposed between the predetermined light source and the imaging system member. Calculate the data for the green component of the corresponding white level for verification,
From the data regarding the red component of the provisionally predicted spectral sensitivity characteristic, the spectral sensitivity characteristic of the predetermined light source, and the spectral sensitivity characteristic of the infrared cut filter, between the predetermined light source and the imaging system member. Calculate data on the red component of the white level for verification corresponding to the state where the infrared cut filter is interposed,
From the data on the green component of the provisionally predicted spectral sensitivity characteristic, the spectral sensitivity characteristic of the predetermined light source, and the spectral sensitivity characteristic of the infrared cut filter, between the predetermined light source and the imaging system member. Calculate data on the green component of the white level for verification corresponding to the state where the infrared cut filter is interposed,
Data on red and green components of image data acquired without the infrared cut filter, data on red and green components of image data acquired with the infrared cut filter interposed , Data regarding the red component and green component of the check white level corresponding to a state where the infrared cut filter is not interposed, and the red component of the check white level corresponding to a state where the infrared cut filter is interposed From the data relating to the green component, a correction coefficient for correcting the data relating to the red component of the tentatively predicted spectral sensitivity characteristic is calculated,
Correcting the data regarding the red component of the provisionally predicted spectral sensitivity characteristic based on the correction coefficient;
The color adjustment method for a digital camera according to claim 1, wherein the color adjustment method is performed. The data regarding the red component of the provisionally predicted spectral sensitivity characteristic is function data of the relative spectral sensitivity level regarding the red component of the imaging system member with respect to the wavelength,
The color adjustment method according to claim 2, wherein the correction coefficient includes a correction coefficient related to a wavelength direction of a spectral waveform and a correction coefficient related to a sensitivity direction based on the function data.   The correction of the data regarding the red component of the provisionally predicted spectral sensitivity characteristic is performed by correcting the sensitivity direction of the spectral waveform based on the correction coefficient related to the sensitivity direction, and then correcting the spectral spectrum based on the correction coefficient related to the wavelength direction. The color adjustment method according to claim 3, wherein the correction is performed in the wavelength direction of the waveform.   When correcting the data regarding the red component of the provisionally predicted spectral sensitivity characteristic in the wavelength direction, the spectral waveform is smoothed while comparing the relative spectral sensitivity levels at adjacent wavelengths within a certain wavelength range. 5. The color adjustment method according to claim 4, further comprising performing smoothing.   Correction of the tentatively predicted spectral sensitivity characteristic in the wavelength direction is performed by using the relative sensitivity coefficient obtained from the data regarding the red component and the green component of the image data acquired without the infrared cut filter, and the sensitivity. Data on the tentatively predicted spectral sensitivity characteristic with the direction corrected, the spectral sensitivity characteristic of the predetermined light source, and the green component of the white level for verification corresponding to the state where the infrared cut filter is not interposed The color adjustment method according to claim 4, wherein the color adjustment method is repeated until the difference from the relative sensitivity coefficient obtained from the above is within an allowable range.   Prior to the final prediction, it is determined whether the characteristic wavelength in the data regarding the red component of the provisionally predicted spectral sensitivity characteristic is larger than the cutoff end wavelength of the infrared cut filter. In the data regarding the red component of the provisionally predicted spectral sensitivity characteristic, the correction is performed when the characteristic wavelength in the data regarding the red component of the spectral sensitivity characteristic is larger than the cutoff end wavelength of the infrared cut filter. 5. The color adjustment method according to claim 4, wherein when the characteristic wavelength is smaller than a cutoff end wavelength of the infrared cut filter, the provisionally predicted spectral sensitivity characteristic is set as a final spectral sensitivity characteristic.   The color adjustment method according to claim 7, wherein the characteristic wavelength is a wavelength at which the relative spectral sensitivity level is lower than a relative spectral sensitivity level at a cutoff end wavelength of the infrared cut filter by a predetermined level. A predetermined light source;
A digital camera that captures the predetermined light source and acquires image data;
An infrared cut filter configured to freely move back and forth between the predetermined light source and the digital camera, and removing an infrared light component of light from the predetermined light source;
A reference data storage unit that stores at least the spectral sensitivity characteristics of the predetermined light source, the spectral sensitivity characteristics of the reference digital camera, and the spectral sensitivity characteristics of the reference infrared cut filter as reference data;
It intends row by interposing the infrared cut filter between the infrared cut filter intends line without interposing the imaging and the upper Symbol predetermined light source and the digital camera between the predetermined light source and the digital camera At least photographing is performed to acquire a plurality of image data, and the acquired plurality of image data are cut off using the reference data stored in the reference data storage unit, and the cutoff start wavelength of the infrared cut filter. The provisional spectral sensitivity prediction unit that tentatively predicts the spectral sensitivity characteristics of the digital camera by processing focusing on
From the spectral sensitivity characteristics of the digital camera , the spectral sensitivity characteristics of the predetermined light source, and the spectral sensitivity characteristics of the infrared cut filter that are provisionally predicted by the temporary spectral sensitivity prediction unit, the spectral sensitivity of the digital camera is calculated. A final spectral sensitivity prediction unit for final prediction of characteristic data;
A color adjusting apparatus comprising:

Images