Embodiment
Below, with reference to the accompanying drawings of embodiments of the present invention.
Fig. 1 is the integrally-built figure that the color adjusting device of an embodiment of the invention is shown.Color adjusting device 20 shown in Fig. 1 has housing, and this housing possesses light source chamber 20a and these two chambers of photograph unit room 20b.These light source chambers 20a and photograph unit room 20b are separated by circular hole slab 23.On circular hole slab 23, be formed with circular hole 23a.
In light source chamber 20a, dispose camera bellows 21.On the face relative with photograph unit room 20b of camera bellows 21, be formed with circular hole 21a.In addition, in camera bellows 21, be provided with light source 22.Light source 22 is for example the such standard sources of A light source.The spectral sensitivity characteristic of this A light source shown in Figure 2.Herein, as light source 22, so long as the light source that radiation grade increases continuously monotonously from short wavelength side to long wavelength side also can be used A light source light source in addition.In the structure shown in Fig. 1, from the light of the light source 22 in camera bellows 21, via circular hole 21a, 23a, inject to photograph unit room 20b.
In photograph unit room 20b, be provided with digital camera 26.In addition, between digital camera 26 and circular hole slab 23, dispose BPF disk 24.In this structure, can in 24 pairs of camera bellows 21 of BPF disk, be taken by digital camera 26.
Fig. 3 is the structure chart of the camera system parts that use of digital camera 26.As shown in the drawing, through the light from light source 22 of imaging lens system group 11 incidents, by infrared (IR) light cutoff filter mirror 12 as optics low pass filtered light microscopic, remove harmful infrared composition, then incide imaging apparatus portion 13.
Imaging apparatus portion 13 has microlens array 13a, filter 13b and imaging apparatus 13c.The filter 13b as color component cutting unit that the light that microlens array 13a by imaging apparatus portion 13 is converged according to each pixel is for example arranged via former colour system Baeyer images on imaging apparatus 13c.Herein, the filter that so-called former colour system Baeyer is arranged, refers to the filter using 4 pixels of R, Gr, Gb, B as Yi Ge Pixel arrangement unit.
As shown in Figure 4, using a plurality of, for example 7 peak wavelengths respectively the different logical optical filtering (BPF) 241~247 of band be configured in and on same circumference, form the BPF disk 24 as spectrophotometric unit.And, motor 25 is installed on the central shaft of BPF disk 24.Can pass through the rotary actuation of this motor 25, will be with arbitrarily logical optical filtering 241~247 to be inserted in the light path between light source 22 and digital camera 26.Herein, peak wavelength that will the logical optical filtering 241~247 of band is made as 440[nm], 480[nm], 520[nm], 560[nm], 600[nm], 620[nm], 640[nm], half breadth is made as respectively to 20[nm].
And, in Fig. 1, on digital camera 26, being connected with stube cable 27, stube cable 27 is connected with personal computer (PC) 28.Stube cable 27 is the stube cables such as USB etc.PC 28 has as the function of spectral sensitivity detecting unit, as the function of white balance coefficient calculation unit, as the function of color of object correction coefficient calculation, calculate the corrected value of each digital camera 26, and set the corrected value calculating in digital camera 26.In addition, PC 28 also controls by the rotary actuation of motor 25 and is inserted into the selection of the logical optical filtering 241~247 of band of the BPF disk 24 in light path, the actions such as photography in digital camera 26.
The action of the color adjusting device of present embodiment then, is described.
Fig. 5 illustrates the flow chart of the contents processing of the color adjustment method of the digital camera 26 of setting as shown in Figure 1.According to the processing that is stored in the program execution 5 in for example hard disk of PC 28 of Fig. 1.
First, as initial setting, the variable n that PC 28 leads to the wavelength of optical filtering by the band that represents BPF disk 24 is set as initial value " 1 " (step S1).
Afterwards, PC 28 by the band being represented by variable n logical optical filtering 24n (n is 1~7) be inserted into the light path of digital camera 26 in after, the photography (step S2) of the light source 22 that combine digital camera 26 carries out.Color signal data corresponding to the color alignment with filter 13b that obtain as the photography result of digital camera 26 are input to PC 28 via stube cable 27.After carrying out photography, whether the value of PC 28 decision variable n has reached the quantity N (N=7 in the example of Fig. 4) (step S3) of the logical optical filtering of all bands forming on BPF disk 24.In the judgement of step S3, in the situation that the value of variable n does not reach N, the value of 28 couples of variable n of PC adds 1 (step S4).Afterwards, PC 28 returns to processing to step S2, uses the logical optical filtering of new band to carry out photography.
In having used the photography of the logical optical filtering of each band, as shown in Figure 6, for example vertical R, the Gr of 32 pixels * horizontal 32 pixels of the central portion of the color signal data that calculating obtains by photography are, the mean value of each color component of Gb, B.Particularly Gr, Gb are further calculated as follows.
G=(Gr+Gb)/2
Thus, can calculate the mean value of each color component of R, G, B.The in the situation that of Fig. 4, R, B are respectively color signal data average of 256 pixels, and G is color signal data average of 512 pixels.By the processing of repeated execution of steps S2~S4, calculate the average color signal data of the sheet number 7 * 3 (RGB)=21 of the logical optical filtering of band, as the result of photography of using the logical optical filtering 241~247 of all bands of BPF disk 24.
In the judgement of step S3, be judged to be in the situation that the value of variable n and all quantity N of the logical optical filtering of band equate, PC 28 is used 21 color signal data that obtain by 7 photographies to calculate the relative spectral sensitivity data (step S5) of the camera system parts of digital cameras 26.
In step S5, PC 28 carries out computing to eliminate the impact of the spectral sensitivity of light source 22 and the spectral transmission of the logical optical filtering of band from color signal data.Herein, in the situation that for example peak wavelength is 440[nm] the logical optical filtering 241 of band, in the spectral sensitivity characteristic of light source 22, as shown in Figure 2 time, clip the wavelength 440[nm in this characteristic], 20[nm] be half breadth.Herein, if be located at wavelength 420[nm]~460[nm] the spectral sensitivity characteristic in scope interior focusing source 22 value of carrying out after integration be Arsc_440, the wavelength 440[nm of the logical optical filtering 241 of band] spectral transmission located is Tr_440 (0 < Tr_440 < 1), the color signal data that calculate mean value when photography are respectively R_440in, G_440in, B_440in, when the spectral sensitivity data of each color component of obtaining are respectively R_cam_440, G_cam_440, B_cam_440, can obtain as follows these values.
R_cam_440=R_440_in/(Asrc_440×Tr_440)
G_cam_440=G_440_in/(Asrc_440×Tr_440)
B_cam_440=B_440_in/ (Asrc_440 * Tr_440) (formula 1)
The logical optical filtering 242~247 of other bands is also carried out to the computing same with (formula 1).After obtaining 21 data of the result of the computing of logical optical filtering for all bands, the computing starting for Shi Cong subordinate is easy, 21 data is normalized, so that the maximum in the result obtaining as operation result becomes 1024.
Then, PC 28 judges whether the IR light cutoff filter mirror 12 of the camera system parts that are used as digital camera 26 is infrared external reflection type (step S6).If the pre-stored information that has the kind that represents IR light cutoff filter mirror 12 in digital camera 26, can be by obtaining from digital camera 26 judgement that information is carried out step S6.
Generally speaking, the relative spectral sensitivity characteristic of known imaging apparatus has characteristic as shown in Figure 7.As shown in Figure 7, R composition also has sensitivity in region of ultra-red.Therefore,, for the correct light that detects R composition, need to remove the infrared composition in the light that is included in R composition.Therefore, at the former configuration IR of imaging apparatus portion 13 light cutoff filter mirror 12.Herein, there is INFRARED ABSORPTION type and infrared external reflection type in known IR light cutoff filter mirror.
The IR light cutoff filter mirror of INFRARED ABSORPTION type is the IR light cutoff filter mirror that is configured to the infrared composition of the light that absorbs institute's incident.
In addition, the IR light cutoff filter mirror of infrared external reflection type is the IR light cutoff filter mirror that is configured to the infrared composition of the light that reflects institute's incident.The example of the spectral transmission characteristic of the IR light cutoff filter mirror of infrared external reflection type shown in Figure 8.The emulation of being undertaken by applicant obtains the characteristic of Fig. 8.
Generally speaking, the manufacture deviation of the cut-off wavelength of the IR light cutoff filter mirror of known infrared reflection type is larger.Fig. 8 considers that this deviation shows three specific characters.Characteristic TYP shown in Fig. 8 show IR-cut start wavelength and IR-cut finish wavelength be respectively predetermined value (in Fig. 8 cut-off to start wavelength be 620[nm], to finish wavelength be 640[nm in cut-off]) the spectral transmission characteristic of representative IR light cutoff filter mirror.In addition, to show the result of the spectral transmission characteristic that detects a plurality of IR light cutoff filter mirrors be that IR-cut starts the characteristic that wavelength has moved to lowest wavelength side to characteristic MIN.In addition, to show the result of the spectral transmission characteristic that detects a plurality of IR light cutoff filter mirrors be the characteristic that IR cut off wavelength has moved to long wavelength side to characteristic MAX.
Known have wavelength beyond obtaining the wavelength that also comprises the logical optical filtering of band that BPF disk 24 has, during as the relative spectral sensitivity characteristic of camera system parts integral body the relative spectral sensitivity characteristic of the combination of IR light cutoff filter mirror 12 and imaging apparatus portion 13 (as), if be wavelength 620[nm about infrared band]~640[nm] scope carry out merely spline interpolation, become large with the error of actual sensitivity.If the IR light cutoff filter mirror of INFRARED ABSORPTION type, can be by wavelength 620[nm] and wavelength 640[nm] between carry out linear interpolation, reduce the error with actual sensitivity.On the other hand, in the situation that the IR light cutoff filter mirror of infrared external reflection type, as shown in the A of Fig. 9, to wavelength 620[nm] and wavelength 640[nm] between while carrying out linear interpolation, become large with the error of actual sensitivity.Therefore in the present embodiment, according to the IR light cutoff filter mirror 12 that is used as camera system parts, whether be, that infrared external reflection type is carried out different interpolation processing.
In the judgement of step S6, in the situation that IR light cutoff filter mirror 12 is infrared external reflection type, PC 28 obtains the relative spectral sensitivity data (step S7) of the interpolation wavelength beyond 7 kinds of specific wavelengths obtaining in step S5.Obtain relative spectral sensitivity performance data R_cam_x, G_cam_x, the B_cam_x of the wavelength shown in following table 1 herein.X represents the value of the wavelength of following table 1.The value of the wavelength shown in table 1 is to have set the value after the wavelength that is suitable for spline interpolation described later according to each color component.These wavelength are corresponding with the part that the relative spectral sensitivity data of each color component are carried out respectively characteristic variation.
[table 1]
Table 1
When obtaining relative spectral sensitivity data R_cam_x, G_cam_x, B_cam_x, suppose that the relative spectral sensitivity performance data of each wavelength of camera system parts of the representativeness combination of the IR light cutoff filter mirror that is provided with characteristic TYP is known herein.Below, by R composition, the G composition of camera system parts of IR light cutoff filter mirror that is provided with this characteristic TYP, the relative spectral sensitivity performance data of each wavelength of B composition is made as reference data, and be expressed as R_ref_x respectively, G_ref_x, B_ref_x.X represents the value of wavelength, at wavelength 380[nm]~wavelength 700[nm] between according to every 1[nm] grade change.And, specifically by aftermentioned, but in order to obtain wavelength 630[nm] the relative spectral sensitivity performance data of R composition, suppose the wavelength 620[nm in the R composition of camera system parts of representativeness combination of the IR light cutoff filter mirror that is provided with characteristic MIN and characteristic MAX]~wavelength 640[nm] relative spectral sensitivity data be also known.The relative spectral sensitivity performance data of camera system parts that is provided with the IR light cutoff filter mirror of characteristic TYP, characteristic MIN and characteristic MAX all can be obtained by actual measurement or emulation etc.
When supposition be provided with characteristic TYP IR light cutoff filter mirror camera system parts each wavelength relative spectral sensitivity performance data and be provided with characteristic MIN and the R composition of the camera system parts of the representativeness of the IR light cutoff filter mirror of characteristic MAX combination in wavelength 620[nm]~wavelength 640[nm] and relative spectral sensitivity data when known, can as following (formula 2), obtain in each the relative spectral sensitivity data shown in table 1.
R_cam_380=R_ref_380
G_cam_380=G_ref_380
B_cam_380=B_ref_380
R_cam_420=R_cam_440×(R_ref_420/R_ref_440)
G_cam_420=G_cam_440×(G_ref_420/G_ref_440)
B_cam_420=B_cam_440×(B_ref_420/B_ref_440)
G_cam_460=G_cam_440×(G_ref_460/G_fef_440)
B_cam_460=B_cam_440×(B_ref_460/B_ref_440)
(formula 2)
R_cam_500=R_cam_520×(R_ref_500/R_ref_520)
G_cam_500=G_cam_520×(G_ref_500/G_ref_520)
R_cam_540=R_cam_560×(R_ref_540/R_ref_560)
G_cam_540=G_cam_560×(G_ref_540/G_ref_560)
R_cam_570=R_cam_600×(R_ref_570/R_ref_600)
R_cam_590=R_cam_600×(R_ref_590/R_ref_600)
G_cam_590=G_cam_600×(G_ref_590/G_ref_600)
R_cam_610=R_cam_600×(R_ref_610/R_ref_600)
R_cam_630=b
Wherein, when following a < R_cam_620, b=a
During following a >=R_cam_620, b=R_cam_620
a=α×R_cam_620+β×R_cam_640
R_cam_670=R_cam_660×(R_ref_670/R_ref_660)
R_cam_680=R_ref_680
R_cam_700=R_ref_700
G_cam_700=G_ref_700
B_cam_700=B_ref_700
From (formula 2), the little relative spectral sensitivity data corresponding to interpolation wavelength of deviation of the relative spectral sensitivity characteristic causing about the deviation of the spectral transmission with IR light cutoff filter mirror, are directly used reference data.In addition, the larger relative spectral sensitivity data corresponding to interpolation wavelength of deviation ratio of the relative spectral sensitivity characteristic causing about the deviation of the spectral transmission with IR light cutoff filter mirror, obtain according to relative spectral sensitivity data and the ratio of immediate reference data.
On the other hand, as shown in Figure 9, the king-sized wavelength 630[nm of deviation of the relative spectral sensitivity characteristic causing about the deviation of the spectral transmission of IR light cutoff filter mirror] the relative spectral sensitivity data of R composition, use predetermined interpolation formula to obtain.For this wavelength 630[nm] the method for obtaining of relative spectral sensitivity data of R composition describe.
In the following description, the relative spectral sensitivity data of each wavelength of R composition are expressed as to H (x).Herein, x represents the value of wavelength.Now, for the IR light cutoff filter mirror of the infrared external reflection type of various spectral transmission characteristics, obtain the value of H (620)/H (640) and the value of H (630)/H (640).And, by H (620)/H (640) be set as transverse axis, H (630)/H (640) is set as the longitudinal axis and makes curve chart.Now, according to applicant's simulation result is known, can access the curve chart shown in Figure 10.As shown in figure 10, the pass of the value of the value of H (620)/H (640) and H (630)/H (640) is linear relationship.Orthoscopic shown in Figure 10 can represent as follows.
H (630)/H (640)=α * (H (620)/H (640))+β (formula 3)
If solve (formula 3) about H (630), can access the relation of following (formula 4).
H (630)=α * H (620)+β * H (640) (formula 4)
At the interpolation formula shown in (formula 4) with corresponding at " a " shown in (formula 2).Herein, if the wavelength 620[nm of characteristic TYP, characteristic MIN and characteristic MAX]~640[nm] relative spectral sensitivity data known respectively, can obtain factor alpha, the β of (formula 3).That is,, if the R11 of Fig. 9, R12, R13, R21, R22, R23, R31, R32, these 9 points of R33 are known, can obtain factor alpha, β.In the situation that the example of Fig. 8, α=0.83, β=0.19.
According to (formula 4) discussed above, can obtain the 630[nm for the IR light cutoff filter mirror of the infrared external reflection type of any characteristic] relative spectral sensitivity data.In addition, as also illustrated, in the situation that the value of " a " is more than R_cam_620, the value of R_cam_630 is restricted to the value of R_cam_620 in (formula 2).As shown in Figure 8, conventionally in the relative spectral sensitivity characteristic of R composition, wavelength 620[nm] relative spectral sensitivity data than wavelength 630[nm] relative spectral sensitivity data little.Therefore,, in the situation that the value of " a " is more than R_cam_620, does not use the value of " a " now and use the value of R_cam_620.
Obtaining as above and wavelength 380[nm]~wavelength 700[nm] relative spectral sensitivity performance data corresponding to each interpolation wavelength after, PC 28, by use the relative spectral sensitivity data of wavelength and the spline interpolation of the relative spectral sensitivity data of the interpolation wavelength of obtaining obtain in step S7 in step S5, obtains every wavelength 1[nm] the relative spectral sensitivity performance data (step S8) of camera system parts.Herein, in the situation that the value that the result of spline interpolation is relative spectral sensitivity data is less than 0, by the relative spectral sensitivity data limit of this wavelength, be 0.The result of spline interpolation shown in Figure 11.Herein, Figure 11 (a) shows the spectral sensitivity performance data of the R composition of camera system parts, Figure 11 (b) shows the spectral sensitivity performance data of the G composition of camera system parts, and Figure 11 (c) shows the spectral sensitivity performance data of the B composition of camera system parts.In addition the point that, the black circle of the use of each figure is drawn represents the relative spectral sensitivity data of the specific wavelength obtained by actual measurement.On the other hand, the point that the white circle of the use of each figure is drawn represents the relative spectral sensitivity data of the interpolation wavelength obtained by interpolation processing.
In addition, in the judgement of step S6, in the situation that IR light cutoff filter mirror 12 is INFRARED ABSORPTION type, PC28, by use the spline interpolation of the relative spectral sensitivity data of 7 kinds of wavelength obtaining in step S5, obtains every 1[nm] the relative spectral sensitivity performance data (step S9) of the camera system parts of wavelength.Herein, in the situation that the value that the result of spline interpolation is relative spectral sensitivity data is less than 0, by the relative spectral sensitivity data limit of this wavelength, be 0.In addition, about wavelength 620[nm]~640[nm] scope, by linear interpolation, obtain relative spectral sensitivity data.
After the relative spectral sensitivity performance data calculating as described above as camera system parts integral body, PC 28 calculates the corrected value (step S10) of white balance according to each light source.And, the special color correction value (step S11) of calculating about color of object of PC 28.By the calculating of these corrected values, finish a series of processing of Fig. 5.For calculating the corrected value of white balance and the method for color correction value according to relative spectral sensitivity performance data, for example, can use No. 4460717 disclosed method of communique of Japanese Patent herein.Therefore, description is omitted herein.
As described above, according to present embodiment, when calculating the relative spectral sensitivity performance data of using with color correction for the white balance correction of camera system parts that is provided with the IR light cutoff filter mirror of infrared external reflection type, in the relative spectral sensitivity data that use calculates according to color signal data, with respect to wavelength 630[nm] for the specific wavelength of short wavelength side be wavelength 620[nm] relative spectral sensitivity data with respect to wavelength 630[nm] for the specific wavelength of long wavelength side be wavelength 640[nm] relative spectral sensitivity data, calculate and become the wavelength 630[nm that large wavelength is interpolation object carrying out linear interpolation time error] relative spectral sensitivity data.There is linear relationship in the value of the value of known H (620)/H (640) and H (630)/H (640), therefore can also be according to 620[nm] relative spectral sensitivity data and wavelength 640[nm] relative spectral sensitivity data, the wavelength correctly calculating between it is wavelength 630[nm] relative spectral sensitivity data.Thus, can also carry out correct color adjustment for the digital camera 26 with the camera system parts of the IR light cutoff filter mirror that is provided with infrared external reflection type.In addition,, for INFRARED ABSORPTION type, also can carry out correct color adjustment samely.
In addition, by setting interpolation wavelength to be suitable for spline interpolation as table 1, if the IR light cutoff filter mirror of infrared external reflection type can only calculate the correct spectral sensitivity performance data of each color component by spline interpolation.Now, do not need to combine spline interpolation and linear interpolation, its result can realize reduction in processing time.
Herein, in the present embodiment, IR light cutoff filter mirror about INFRARED ABSORPTION type, combination spline interpolation and linear interpolation are calculated spectral sensitivity performance data, if but similarly obtain the relative spectral sensitivity data of interpolation wavelength with the IR light cutoff filter mirror of infrared external reflection type, can also only by spline interpolation, calculate spectral sensitivity performance data.Now, need to the relative spectral sensitivity data of representative camera system parts of IR light cutoff filter mirror that be provided with INFRARED ABSORPTION type are pre-stored in PC 28 as reference data.
In addition, in the present embodiment, in order to obtain α, the β of (formula 4), use the relative spectral sensitivity data of 3 kinds of representative camera system parts of the IR light cutoff filter mirror be respectively arranged with characteristic TYP, characteristic MIN and characteristic MAX to make the curve chart of Figure 10.As long as there is linear relationship in the value of the value of H (620)/H (640) and H (630)/H (640), therefore 2 curve charts that just can make Figure 10 of minimum drafting.That is,, in order to obtain α, β, not necessarily need to obtain to be respectively arranged with 3 kinds of relative spectral sensitivity data whole of the IR light cutoff filter mirror of characteristic TYP, characteristic MIN and characteristic MAX.That is, as long as obtain at least any two in characteristic TYP, characteristic MIN and characteristic MAX.Otherwise, thereby also can to come up more correctly to make the curve chart of Figure 10, obtain α, β by drawing at 4.
According to above execution mode, the present invention has been described, but has the invention is not restricted to above-mentioned execution mode, certainly can in the scope of aim of the present invention, carry out various distortion and application.
And the invention that above-mentioned execution mode has comprised the various stages, can be by the various inventions of appropriately combined extraction of disclosed a plurality of inscapes.For example, even delete several inscapes from all inscapes shown in execution mode, can solve above-mentioned problem, and obtain in the situation of above-mentioned effect, the structure of having deleted after this inscape also can be used as invention extraction.