JPH01135267A - Color film reader - Google Patents

Abstract

PURPOSE:To easily find out necessary correcting conditions by storing plural kinds of coefficients of color correcting operations and selecting one from them, when the density signals of respective colors are independently correcting- operated mutually. CONSTITUTION:A color film 103 is image-formed to a CCD 105 in which the filters of RGB three colors are alternately arranged. The output of line CCD 105 is A/D-converted 106, shading-corrected 107 and given to a logarithm converter 108. The converter 108 converts an RGB signal to a CMY density signal. A UCR circuit 109 executes a color correcting processing to the CMY density signal outputted from the converter 108. In such a case, the converter 108 stores plural kinds of the coefficients of the color correcting operations as to the C, M and Y, respectively, and one among them is selected by a switching signal given from an external part. Thus, the correcting conditions necessary for obtaining an output image having a proper color balance can be easily found out.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to a color film reading device.

[Prior Art] Conventional color film reading devices digitally read color films, output RGB density signals for each pixel, and send the output data to a color printer such as a laser beam printer. ,Digital scanner writers, etc. exist on a large scale.

[Problems to be Solved by the Invention] In the conventional device described above, depending on the lighting conditions at the time of photographing,
The color balance of images on color film varies greatly, and it is impossible to know the lighting conditions at the time of shooting when the images are shot on film. Therefore, there is a problem in that the operator's skill and a large amount of work are required to find the correction conditions necessary to obtain an output image with appropriate color balance.

[Means for Solving the Problems] The present invention focuses a transmissive color film on a three-primary color separation sensor, logarithmically transforms the transmittance signal of the color film, and obtains yellow and magenta signals for each pixel. , when performing color correction calculations on the density signals of cyan independently from each other, a plurality of types of coefficients for the color correction calculation are stored for each of the yellow, magenta, and cyan, and one of these is selected. .

[Function] The present invention focuses a transmissive color film on a three-primary color separation sensor, logarithmically transforms the transmittance signal of the color film, and mutually compares yellow, magenta, and cyan density signals obtained for each pixel. When performing color correction calculations independently, multiple types of coefficients for the color correction calculation are stored for each of the yellow, magenta, and cyan colors, and one is selected from among them, resulting in output with appropriate color balance. Correction conditions necessary to obtain an image can be easily found.

[Example] FIG. 1 is a block diagram showing one embodiment of the present invention.

This embodiment includes an illumination light source 101 that illuminates the film;
A condenser lens 102, a color film 103 to be read, an imaging lens 104, an in-line CCD 105 in which RGB three color filters are arranged alternately, and a CCD
Sample and hold the output of CD105, A/D convert it and R
An A/D converter 106 that outputs GB digital data, and a shading correction circuit 1 that corrects the sensitivity unevenness of one line of the COD I O 5, the light amount unevenness of the illumination light source, and the color temperature.
07, a logarithmic converter 108 that converts RGB signals into CMY density signals, and a UCR circuit 1 that performs well-known color correction processing.
09, a masking circuit 110, a γ correction circuit 111,
It has a color printer 112 such as a laser beam printer.

The film 103 is fed at a predetermined speed in the direction of the arrow by a scanning mechanism (not shown), and an image is output by the printer 112 for each scanning line. Alternatively, the film 103 may be fixed, and the lens 104 and CCD 105 may be scanned. The shading correction circuit 107 has the role of normalizing the input signal by, for example, COD output with the film 103 removed.

Next, the operation of the above embodiment will be explained.

Here, a case will be considered in which the film 103 is a color reversal film. The gradation recording characteristics of color reversal film (the relationship between the amount of light E of the subject and the transmittance T of the film that records this) are generally expressed on a logarithmic scale.
It is represented as shown in the figure.

The curves of 2-R52-G and 2-B are R, G, and
This corresponds to the transmittance of H. In other words, GR, GG
, GB are predetermined constants, and the absolute values of the slopes of the straight line portions of 2-Rl2-G and 2-B are γR1γG and γB, respectively.

-1ogTR=GR-yRII 1ogER-logT
G=GG-1G・1agEG-1ogT B=G
B -y B @1ogE B... (1) formula % formula % Transmittance of each film, in Figure 1, R, G
, B. Also, ER, EG, E
B indicates the amount of R, G, and B subject light. Assuming that the logarithmic converter 108 outputs the object density signals C, M, and Y, it is sufficient to output C=-1ogE RM=-1ogEG Y=-1ogE B, so the logarithmic converter 108 is The following equation (2) is converted.

C=-(1/γR)・1ogT R M=-(1/γG)・1ogTG Y=-(17γB)・1ogTB...Equation (2) Here, in the shading correction circuit 107, TR
, TG, and TB are normalized by a certain transmittance, so (
GR, GG, and GB in equation (1) have disappeared from equation (2).

Next, consider the RGB light intensity balance (ERlEG, E
Consider a case where the ratio of B) deviates from a predetermined condition. In other words, suppose that the ratio of ER, EG, and EB of sunlight outdoors during a clear day is standardized to be 1=1:1, and under certain shooting conditions it becomes 1:kG:kB (i.e. ,ER+
ER1EG+kG-EG, EB+kB @EB). In this case, the above equation (2) must be changed to the following equation (3).

C=-(1/γR)・1ogT R M=-(1/γG)・IogT G + Iagk
GY--(1/γB)* 1ogTB+ Iog
kB...(3) In other words, each of the logarithmically transformed M and Y signals is
Bias signals for ogk G and Iogk B must be provided. UCR circuit 109 provides the bias signal. By converting equation (4) shown below, UC
The R circuit 109 inputs C', M'', Y from the input C, M, Y signals.
′, generates Ko.

K=min (Y, M, C) Y'=Y-GY
(K-βM) M'=M-6M(K-βM) C'=C-αC(K-βC) K'=αK(K-βK) ...Formula (4), Ko is the black signal. , and α and β are constants determined in advance for each color.

Here, by changing the equation regarding Y' and M'' in the above equation (4) to the following equation (5), equation (3) can be realized.

Y'=Y-GY (K-βM) + IogkBM'
=M-6M(K-βM)+1ogkG...(5
) Formula The above formula (5) is actually expressed as the following formula (6) due to the circuit configuration.

Y'=Y-GY [K-(βY+(1ogkB)/αY
)]M' = M-cx M [K-(βM+(1ogk
G)/a M)]... (6) Formula (6) The value in the curly brackets is stored in multiple types according to the values of kB and kG, and one of them is selected by a switching signal given from the outside. All you have to do is select.

For reference, the values of kG and kB depending on the color temperature of the light source are:
It is shown in Table 1 below.

Table 1 The above switching signal may be generated automatically by prescanning the film image and reading the image condition, or may be generated automatically by prescanning the film image and reading the image condition.
By using Table 1 above, which may be generated manually by an operator, it is possible to switch to four patterns.

Note that the correction of equation (3) above can be approximately realized by the masking circuit 110. The masking circuit 110 normally performs the calculation of the following equation (7) on the input Y, M, and C signals to obtain the Y', M', and C' signals.

Y'=a++Y+a+2M+a+3cM'=
a2+Y + &22M+ a23cC' = aH
Y+a32M+a33c...(7) Formula % Formula % is a constant. Here, all is all(1+(log
kB) /Y O), &22 as a22(1+(1
ogkG)/MO), equation (3) holds true only when Y=Yo and M=Mo, and becomes an approximation of equation (3) in other cases. Therefore, if Yo and MO are set to a halftone density (for example, 128 of 255 levels), good approximations can be obtained over almost the entire Y and M.

Similar to the above example, a++ (1+ (logkB)
/Yo), a22(1+ (IogkB)/Mo
), a plurality of sets of values may be stored according to kG and kB, and may be switched as appropriate.

Furthermore, as a modification of the above, the same result as above can be obtained in the γ correction circuit 111 as well. In the γ correction circuit 111, for example, the input Y, M, and C signals are
Y', M', and C' times signal conversion is performed using equation (8) shown below.

Y'=b+Y M'=b2M C'=b3C ... (8) formula% formula%) If you store multiple sets of these values, you can approximate (3) in the same way as above. The formula can be established. However, in this case, since the processing is performed on Y, M, and C after masking, the accuracy of the approximation is slightly worse than in the "-criminal example."

By doing the above, the color balance of the color film can be corrected without much skill required by the operator and with a simple circuit configuration, which standardizes the color correction work and improves operability. can be achieved.

Furthermore, in the above embodiments, the present invention can be applied not only to color correction of the illumination light source but also to fading of the film, discoloration due to development conditions, etc., as long as the values of kG and kB are determined in advance.

[Effects of the Invention] According to the present invention, when a color film is digitally read, a KGB density signal is output for each pixel, and the output data is sent to a color printer such as a laser beam printer, the color film can be read with appropriate color balance. This has the effect that correction conditions necessary to obtain an output image can be easily found.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention. FIG. 2 is a diagram showing the gradation recording characteristics of a color reversal film. 103... Input target film, 107... Shading correction circuit, 108... Logarithmic converter, 109... UCR circuit, 110... Masking circuit, 111... γ correction circuit. Patent applicant Canon Co., Ltd. Agent Arata Yokubo −

Claims (2)

[Claims] (1) A transmissive color film is imaged on a three-primary color separation sensor, the transmittance of the color film is digitally read, this transmittance signal is logarithmically converted, and yellow, magenta, and cyan density signals are converted to pixels. Yellow,
In a color film reading device that performs color correction calculations on magenta and cyan density signals independently of each other, for each of the yellow, magenta, and cyan,
A color film reading device characterized in that a plurality of types of coefficients for the color correction calculation are stored and one is selected from the coefficients. (2) The color film reading device according to claim 1, wherein the coefficients for the color correction calculation are selected according to the spectral distribution of an illumination light source at the time of photographing.