JPH0553221A - Photographic image information processing method - Google Patents

Abstract

PURPOSE:To optimize and automate color correction by evaluating the correlation between pieces of image information on two mutually different colors and deciding the attribute of an original image according to the evaluation result. CONSTITUTION:Pieces of secondary image density information regarding respective colors B, G, and R are taken out of a memory and set in variables (ST1). Then coefficients of correlation between two colors among the pieces of image density information on the colors B, G, and R are derived from a specific expression (ST2). The correlation coefficients which are thus found are numerals where differences in the deviation of distributions among the colors B, G, and R are reflected. Further, the correlation between the pieces of image information on the two mutually different colors is evaluated (ST3), the attribute of the original image is decided according to the evaluation result, and a parameter of correction level regarding the correction of an exposure quantity at an exposure control part is determined (ST4). Thus, the exposure quantity is corrected selectively according to the correction level found in consideration of the dominance of a specific color of a subject.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a photographic image information processing method when an original image of a photographic film is printed on a photographic paper, and more specifically, two of color image information obtained by color separation scanning the original image. The present invention relates to a method of discriminating the attribute of the original image by using color image information.

[0002]

2. Description of the Related Art In general photography, a subject's blue (B), green (G), red (R) (hereinafter simply referred to as B,
It is known as an empirical rule that the average reflectances of the three primary colors (described as G and R) are substantially constant. Therefore, in the conventional photographic printing apparatus, when determining the printing exposure amount, first, the average transmission density (LATD) of the entire area of the photographic original image is measured, and based on the measured average transmission density, B and G of the printing paper are measured. , R, the amount of exposure given to the photosensitive layer of each color is controlled to be constant, and a photographic print having a good color balance is prepared.

However, the LATD method has a problem that it is difficult to obtain a proper photographic print when the luminance and color distributions are uneven on the subject side. Such a photographic original image is called a subject feria, in particular, a thing caused by a bias of the luminance distribution of the subject is called a density feria, and a thing caused by a bias of the color distribution is called a color feria.

As a known technique for adjusting the exposure amount for color ferria, for example, "Exposure Determination Methods for CJ Bartleson and RW Huboi"
Color Printing: The concept of Optimum Correction L
evel ", J.SMPTE, 65, 205-215 (1956). In this paper, LAT is used for photographic originals of standard subjects.
Full correction is effective in which the exposure is adjusted in accordance with D, and the exposure amount given to each color photosensitive layer of the printing paper is set to a constant value. When the photographic exposure is appropriate, no correction is effective for subject ferria by exposing with a constant light flux or exposure time regardless of the LATD. As a realistic compromise, he says that the intermediate control method between the two, Roward Collection, is appropriate and that the optimum collection level should be selected from the overall quality of the photographic print.

By the way, recent photographic printing apparatuses are equipped with several levels of collection levels that can be selected. An operator observes a photographic original image and selects a collection level for each original image to adjust the color balance. Is commonly done. When selecting the collection level, full collection is suitable for original images that are affected by the improper shooting light source or latent image regression,
The lowward collection is considered appropriate for a color feria where the subject has an uneven color distribution.

As described above, the selection of the collection level in photographic printing must be made in accordance with the presence / absence of color distribution bias in the subject recorded in the color photographic original, that is, the predominance of the specific color. On the other hand, several methods have been conventionally proposed in relation to the method of discriminating the photographing light source. For example, by E. Goll, D. Hill and W. Severin
"Modern Exposure Determination for Custumizing
Photofinishing Printer Response ", JAPE, 5,93-104 (19
In 79), focusing on the fact that the influence of the color temperature of the photographing light source is remarkable for a specific hue in the chromaticity calculated based on the average density of the original photo, the collection level is adjusted depending on the hue. An exposure amount determination method is shown.

However, for example, there is no great difference in the chromaticity obtained from the average density between an abnormal photographic original image taken under a fluorescent lamp and a normal photographic original image taken on a green lawn, and therefore the hues of both Is difficult to identify. As seen in this case, the method of selecting the collection level according to the average concentration is not always effective.

Further, JP-A-55-26568 and JP-A-55-26570
In Japanese Patent Laid-Open No. 55-26571, based on the densities of the three primary colors of a large number of points in a color photographic original, the hue of the highest density point and the skin color under each light source of tungsten light, fluorescent lamp, daylight A method of classifying photographing light sources using the average hue of the determined points as a characteristic value is shown. However, the characteristics of photographic film generally differ greatly depending on the type of product and the storage condition after shooting, and even if the same skin-colored subject is shot under the same shooting conditions, it is recorded on each photographic film. There is no guarantee that the concentration will be constant. Therefore, these methods have the drawback that the determination results differ depending on the type of photographic film and the storage condition. Also,
Since the density of the highest density point is used, it is easily affected by noise contained in the image data, and there is a drawback that reproducibility is lacking.

Further, it is self-evident that the assumption that the subject includes a person is not always realistic. In addition, in JP-A-60-230129, the densities of a large number of areas of an exposed image are measured and the blue / red density difference of the area having the maximum blue density is derived as a function of the neutral density. A method of discriminating an exposure image created by artificial rays is shown by comparing the function blue / red density difference with a reference line.

This method is based on the premise that the entire area of the exposure image created by artificial rays has a strong red-yellow density, but many areas, such as the original image taken by a fluorescent lamp, No consideration is given to the original picture that has a high green density. In addition, there is a drawback that an original image of a blue object photographed with tungsten light is erroneously determined. Furthermore, since the characteristic value for discrimination is obtained from the region having the maximum blue density, there is a drawback that the characteristic value varies greatly and the reproducibility of the discrimination result is lacking.

[0011]

As described above, in the prior art relating to the discrimination of the photographing light source for color correction in photo printing, the following problems are likely to occur: an erroneous discrimination between an original image photographed under an artificial light source and a color ferria. -The discrimination result is affected by the characteristics of the photographic film depending on the type and storage condition of the photographic film-The discrimination cannot be performed when the subject does not include a person, especially the skin-colored portion-The characteristic values for discrimination vary widely However, there were many problems such as lack of accuracy and reproducibility. Therefore, it is extremely difficult to make such a determination automatically and stably.

The present invention has been made in view of the above problems, and has an attribute of a color photographic original image, that is, an uneven distribution of color in a subject, regardless of the type of photographic film, the storage state, and the type of photographic light source. It is an object of the present invention to provide a photographic image information processing method capable of accurately and stably discriminating the presence or absence of a color and realizing optimization and automation of color correction in photo printing represented by collection level selection.

[0013]

In order to achieve the above-mentioned object, the photographic image information processing method of the present invention uses two different color images among the image information of each color obtained by color-separating and scanning a color photographic original image. The correlation between image information was evaluated, the attribute of the original image was determined based on the evaluation result, and the bias of the color distribution can be determined regardless of the type of photographic film, the storage state, and the type of photographing light source. It is a thing.

Further, the two different colors are in a complementary color relationship with each other, and based on the statistical amount of the image density information,
This makes it possible to make highly accurate and stable determinations without being affected by noise. Also, the different 2
Assume that one of the colors is a neutral color, and based on the statistics of the image density information, without being affected by noise,
This makes it possible to make a stable determination with high accuracy.

The correlation is evaluated based on the correlation coefficient, and the attribute is not discriminated based on the density information of a specific pixel. Therefore, the correlation is not affected by noise included in the image information and is extremely small. This makes it possible to make a stable determination with high accuracy. Further, the attribute of the original image is related to the selection of the collection level in the photo printing, and the precise control of the exposure amount of the photo printing can be realized.

[0016]

The original image on the photographic film set in the exposure section is color-separated and scanned by the color filter and the CCD image sensor to extract the image density information of each color. First, based on this two-dimensional image density information, the density correction value for the exposure amount by the LATD method is calculated. Next, of the image information of each color, the correlation coefficient is calculated for two colors different from each other, and the bias of the color distribution in the original image is analyzed and evaluated. Then, according to this evaluation result, the collection level for the exposure amount by the LATD method is determined, and the final exposure amount is determined.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a photographic printing exposure amount control method to which the photographic image information processing method of the present invention is applied will be described below with reference to the drawings. 1 is an overall configuration diagram of a photographic printing apparatus to which the photographic image information processing method of the present application is applied, FIG. 2 is a block diagram showing an internal configuration of an imaging unit, FIG. 3 is a configuration diagram of a color separation filter, and FIG. 4 is an image processing unit. 5 is a block diagram showing the internal configuration of FIG. 5, FIG. 5 is a flowchart showing a process of image information processing, FIG. 6 is an explanatory diagram showing a correlation between image density information of two different colors, and FIG. This is an example.

In FIG. 1, F is a photographic film, and the photographic film F is set on a first spool 1 and wound on a second spool 2 via a predetermined transport path. The original image I formed on the photographic film F is positioned in the exposure section 3 and is irradiated from the light source 4, and the mixing section 5
Illuminated by the light homogenized by. Further, the lens 6 optically forms an image on the photographic printing paper P so that it can be exposed. Here, the average transmitted light of each color of B, G, and R of the original image I is passed through the photometric filters 7a, 7b, and 7c of each color of B, G, and R to the photodiode 8a,
The light is received by 8b and 8c. The B, G, and R photometric signals obtained by photoelectrically converting the received light amount are supplied to the exposure control unit 9 and then A / D converted, and as described later, the image processing unit 1
1, the exposure amount is determined based on the average photometric value of each of B, G, and R colors.

In the exposure section 9, the original image I formed on the photographic film F is color-separated into B, G, and R colors by the imaging section 10 and scanned. The image signals of B, G, and R colors obtained by the image pickup unit 10 are sent to the image processing unit 11, A / D converted, and shaped into image density information of a predetermined format. In the image processing unit 11, the exposure amount correction parameter is calculated from the image density information thus obtained by the method described below, and the obtained exposure amount correction parameter is transmitted to the exposure control unit 9 through the communication line 12. To be done.

The exposure control unit 9 corrects the exposure amount based on the above-mentioned average photometric value based on the exposure amount correction parameter transmitted from the image processing unit 11. Further, the exposure control unit 9
Then, the exposure amount thus obtained is set to yellow (Y), magenta (M), and cyan (C) arranged on the upper portion of the exposure unit 3.
The reduction time cut filters 13a, 13b, 13c and the operating time of the shutter 14 arranged under the exposure unit 3 are converted. Depending on the operating time, the cut filters 13a, 13
b, 13c and the shutter 14 are inserted into the exposure light path to adjust the exposure given to the photosensitive layer of each color of the photographic printing paper P.

After the completion of this exposure, the photographic printing paper P is transported by a predetermined distance in preparation for the next exposure, and the photographic film F is transported to position the original image I to be printed next in the exposure section 3. In this way, the photographic film F
The original image I formed in 1 is sequentially subjected to a baking process.

The original image I positioned on the exposure unit 3 and illuminated by the mixing unit 5 is optically projected by the lens 6 onto the two-dimensional imager 20. On the two-dimensional image sensor 20, stripe-shaped color separation filters 21 of B, G, and R as shown in FIG. 3 are placed corresponding to the individual photoelectric conversion elements, and are arranged at right angles to the stripes. The electric charges accumulated in the photoelectric conversion element are sequentially read out in the direction. As a result, the two-dimensional image sensor 20 is configured to output the image signals of B, G, and R colors in a mixed manner.

The image signal 20a thus obtained
In the signal processing circuit 22, the B, G, and R colors are color-separated, and the color-separated image signal is sample-held and amplified. The amplified image signals 22a, 22b, 22c of B, G, R colors are output to the image processing unit 11. Further, from the drive circuit 23, a clock signal 23a for driving the two-dimensional image sensor 20, a timing signal 23b for performing color separation in the signal processing circuit 22 as described above, and A / D conversion in the image processing unit 11. Further, a horizontal synchronizing signal 23c and a vertical synchronizing signal 23d for controlling writing to the image memory are supplied.

FIG. 4 shows a detailed structure of the image processing section 11. The image processing unit 11 performs the following signal processing. Image signal 22 supplied from the imaging unit 10
The a, 22b, and 22c are selected by the analog switch 30, sampled by the sample hold circuit 31, and converted into digital signals by the A / D converter 32. Image signals 22a, 2 in the analog switch 30
The timing of switching between 2b and 22c and the timing of sampling in the A / D converter 32 are controlled by the timing control circuit 33 based on the horizontal synchronizing signal 23c and the vertical synchronizing signal 23d output from the imaging unit 10. Here, the number of samplings is B, G, R per horizontal scanning.
128 for each color, such sampling is 1
It is designed to be performed in 128 horizontal scans per vertical scan.

Therefore, for each vertical scanning, 128 × 1
Digital image information of 28 pixels can be obtained for each of B, G, and R colors. The A / D conversion is processed with 10 bits. The digitized image signal is stored in ROM
A density value is converted through a LUT (look-up table) 34 configured by, for example, and stored in the image memory 35. The LUT 34 stores a conversion table represented by the following expression (1). Y = a × log (X + b) (1) where X is an input to the LUT 34 and Y is its output. Further, a is a constant relating to conversion from the photometric density determined by the spectral characteristic of the image pickup unit 10 to the printing density determined by the spectral sensitivity of the photographic printing paper, and b is a constant relating to removal of the influence of dark current in image pickup. The LUT 34 is provided with a plurality of conversion tables obtained by substituting several kinds of numerical values for a and b in Expression (1), and is previously selected by the CPU 36. The conversion table selected here does not necessarily have to be the same for each color of B, G, and R and may be different.

In this manner, 10-bit image density information of each color of B, G, and R consisting of 128 × 128 pixels is stored in the image memory 35. The image density information stored in this way is called primary image density information. The address when writing to the image memory 35 is the image pickup unit 1.
It is controlled by the timing control circuit 33 based on the horizontal synchronizing signal 23c and the vertical synchronizing signal 23d output from 0.

Here, there is a problem that the effective image area in the image density information differs depending on the format of the photographic film F. Therefore, the CPU 36 controls the photographic film F
Depending on the format of, the image area that is pre-stored for rounding and the number of pixels to be rounded are set,
A process of rounding each pixel in the set image area by the set number of pixels is performed.

The number of pixels obtained as a result of this rounding processing is 16 × 1 regardless of the format of the photographic film F.
It has 6 pixels. Regarding the method of setting the effective image area and the number of pixels, the applicant of the present invention has disclosed in Japanese Patent Laid-Open No.
The method proposed in Japanese Patent No. 60474 may be used. Here, the rounding processing is to take an arithmetic average of the image density information, which has an effect of reducing the influence of noise included in the image signal. However, the rounding does not necessarily have to target all the pixels in the selected image area, but the method of taking an upper arithmetic average by appropriately thinning out or only thinning out by a predetermined number of pixels The method that does not take the arithmetic mean in is also effective in speeding up the calculation.

By the above image processing, the secondary image density information having a predetermined number of pixels, here 16 × 16 pixels, is obtained regardless of the format of the photographic film F.
This secondary image density information is stored in the memory 37. Next, the internal processing in the image processing unit 11 will be described.

FIG. 5 shows a flowchart of this internal processing. First, in step 1, from memory 37 to B,
The secondary image density information (hereinafter referred to as image density information) regarding each color of G and R is taken out and set in the variable X _k (i, j). Here, i and j indicate pixel positions, and k indicates B, G, and R colors.
In step 2, the correlation coefficient between two different colors of the image density information of each of B, G, and R is derived by the following equation (2). CC _BG = Cov (X _B , X _G ) / {Var (X _B ) × Var (X _G )} ^1/2 CC _GR = Cov (X _G , X _R ) / {Var (X _G ) × Var (X _R )} ^1/2 CC _RB = Cov (X _R , X _B ) / {Var (X _R ) × Var (X _B )} ^1/2 (2) where CC _BG , CC _GR and CC _RB are the correlation coefficients between the colors BG, GR, and RB for the image density information, Cov is the covariance, and Var is the variance. The correlation coefficient thus obtained is B of the image density information X _k (i, j),
A value that reflects the difference in distribution bias between the G and R colors is taken.

FIG. 6 shows a graph plotting the image density information of two different colors on the vertical axis and the horizontal axis and plotting the relationship between the two for all pixels. It is possible to know the correlation between the image density information of. FIG. 6 (a) is an original image of a standard subject having no significant color deviation in daylight, and FIG. 6 (b) is a green image of a person against a background of dark green trees in daylight. 6 (c) shows an original image obtained by photographing a standard subject as described above under a fluorescent lamp.

As shown in FIG. 6A, in the case of an original image obtained by photographing a standard subject having no large color distribution bias in daylight, the densities of the pixels are almost the same for B, G, and R. Since it changes, the correlation of all the image density information of two different colors becomes extremely high. On the other hand, as shown in FIG. 6B, in the case of a green color feria where a person against a background of dark green trees is photographed under daylight, the image density information obtained from the area of the trees Has a relatively high G concentration,
Since the B and R densities are relatively low, the correlation between G and B and G and R image density information is low. As described above, in the case of the color ferria, the correlation between the image density information of the two colors relating to the dominant color in the subject and the other colors is low.

Further, in the case of an original image obtained by photographing a standard subject as described above under a fluorescent lamp, as shown in FIG. 6 (c), the G concentration tends to be almost uniformly high over the entire region (all pixels). As shown, since the densities of the respective colors change in substantially the same manner, the correlation of the image density information of all the two different colors becomes extremely high. In this way, the original image of photographic film that has deteriorated in characteristics due to the subject and storage conditions taken under artificial light is the original image of a photographic film that is taken in daylight on a standard subject and placed in good storage condition. Compared with the above, since the density relating to the specific color only changes over the entire area of the original image, the correlation between the image density information of all two different colors becomes high. This is the same when the types of photographic film are different and the tone reproduction characteristics are different.

Therefore, two different values obtained from equation (2)
It is rational to evaluate the dominance of the specific color in the subject based on the correlation coefficient between the image density information of the colors. For example, in the case of an original image of a standard subject, the three correlation coefficients show almost the same high value regardless of the characteristic difference due to the type of photographic film and the storage state and the type of the photographing light source.
In the case of a color ferria in which a specific color (for example, G) is dominant, a correlation coefficient CC _BG between the specific color (for example, G) and another color,
CC _GR shows a smaller value than other correlation coefficients CC _RB .

Therefore, as the evaluation amount of the dominance of the specific color in the subject as described above, for example, the minimum value of these three correlation coefficients can be mentioned. Alternatively, there is a method in which the difference between the maximum value and the minimum value of the three correlation coefficients is taken as the evaluation amount in case all the three correlation coefficients take small values. Here we follow the latter example.

In step 3, the evaluation amount CD relating to the dominance of the specific color in the subject as described above is calculated by the following equation (3).
Ask by. CD = Max (CC _BG , CC _GR , CC _RB ) −Min (CC _BG , CC _GR , CC _RB ) ... (3) Note that there is a complementary color relationship of 2 as expressed by the following equation (4). Correlation coefficients between color image density information CC _BY , CC _GM ,
Similar results are obtained with CC _RC . CC _BY = Cov (X _B , X _Y ) / {Var (X _B ) × Var (X _Y )} ^1/2 CC _GM = Cov (X _G , X _M ) / {Var (X _G ) × Var (X _M )} ^1/2 CC _RC = Cov (X _R , X _C ) / {Var (X _R ) × Var (X _C )} ^1/2 where _XY (I, J) = (X _G (I, J ) + X _R (I, J)) / 2 X _M (I, J) = (X _R (I, J) + X _B (I, J)) / 2 X _C (I, J) = (X _B (I, J) + X _G (I, J)) / 2 (4) In this case, the evaluation amount CD relating to the dominance of the specific color in the subject is expressed by the following equation (5). CD = Max (CC _BY , CC _GM , CC _RC ) -Min (CC _BY , CC _GM , CC _RC ) ... (5) Furthermore, one of the two colors as expressed by the following equation (6)
Similar results are obtained when the correlation coefficients CC _BN , CC _GN , and CC _RN between the image density information of two colors when the color is a neutral color are used. CC _BN = Cov (X _B , X _N ) / {Var (X _B ) × Var (X _N )} ^1/2 CC _GN = Cov (X _G , X _N ) / {Var (X _G ) × Var (X _N )} ^1/2 CC _RN = Cov (X _R , X _N ) / {Var (X _R ) × Var (X _N )} ^1/2 where X _N (I, J) = (X _B (I, J _{) + X G (I, J} ) + X R (I, J)) / 3 ··· (6) in this case, the evaluation amount CD on Control of the specific color in the subject represented by the formula (7). CD = Max (CC _BN , CC _GN , CC _RN ) -Min (CC _BN , CC _GN , CC _RN ) ... (7) As described above, in the present invention, the attribute of the color photographic original image, that is, color ferria or In order to evaluate the dominance of a specific color based on the statistical amount of image density information of the entire image, regardless of the density information of the specific pixel, in relation to the presence or absence of color distribution bias in the subject, such as the effect of the shooting light source. It is possible to perform stable determination with extremely high accuracy without being affected by noise included in the image information. Needless to say, this evaluation does not depend on whether or not the subject includes a person, especially a skin-colored portion.

In step 4, B, G, R as described above
Based on the analysis of the image density information of each color, the attribute of the original color photographic image is discriminated, and the collection level parameter CL relating to the correction of the exposure amount in the exposure control unit 9 described later is determined. Here, the determination of the attribute of the color photograph original image, that is, the presence / absence of the color distribution bias in the subject may be replaced with the determination as to whether lowward collection or full collection is necessary. If the CD value exceeds a predetermined value, it is a color ferria and thus requires low correction, and if the CD value is less than the predetermined value, it is an original image of a standard subject and full correction is necessary. Then, the collection level parameter CL is binaryly determined to a value corresponding to each.

With this method, however, when the CD is in the vicinity of the above-mentioned predetermined value, it can be discriminated by either of the two collection levels, so that variations easily occur. Therefore, the attribute of the color photograph original image to be discriminated may be a multivalued collection level. This corresponds to determining the collection level parameter CL according to a function of several orders including the first order of the CD or a look-up table (LUT) referred to by the CD to enable a non-linear transformation. Here we follow the latter example. CL = LUT (CD) (8) FIG. 7 shows an example of this LUT. In the figure, the horizontal axis represents the evaluation amount CD of the dominance of the specific color, and the vertical axis represents the collection level parameter CL. Here, the value of the parameter takes a value in the range of 0 to 1, but when it is 0, the exposure amount is not corrected, and when it is 1, the maximum correction is added.

Generally, about 90% of the images do not require the exposure amount correction for adjusting the color balance, and therefore the CL value is set to 0 in many cases. It The CL value is preferably a continuous numerical value, but for example, 0.0, 0.2, 0.4, ...
It may be a gradual or discrete number such as 1.0.

The lookup table is stored in the memory 37 of the image processing unit 11 prior to the processing.
The adjustment may be made by inputting from an operation unit or an auxiliary storage unit (not shown). Note that, here, CL is determined from one type of evaluation amount regarding the dominance of the specific color, but CL may be determined from more types of evaluation amount. In this case, as the evaluation amount, B of a standard image or a large number of images,
Standard chromaticity (X ₀ , Y ₀ ) obtained from the average value of the two-dimensional image density information of each of G and R, and the color obtained from the average value of the two-dimensional image density information of each of B, G, and R colors of the image. Degree (X, Y)
An example is one obtained by comparison with. The chromaticity (X, Y) is calculated by the following expression (9), for example. X = b * ^31/2 / 2-r * ^31/2 / 2Y = b / 2-g + r / 2 ... (9) Here, b, g, and r are B, G, and R. The average value of the two-dimensional image density information of each color is shown. Furthermore, examples of the evaluation amount relating to the dominance of the specific color based on the chromaticity include the saturation SAT and the hue HUE shown in the following expression (10). _{SAT = {(X-X 0} ) 2 + (Y-Y 0) 2} 1/2 HUE = Tan -1 {(Y-Y 0) / (X-X 0)} ··· (10) such The collection level CL relating to the correction of the exposure amount is determined from the evaluation amounts relating to the dominance of the plurality of specific colors obtained as described above. This determination complies with the following equation (11). CL = f (P ₁ , P ₂ , P ₃ , ...) (11) where P _i (i = 1,2,3, ..., N) is the control of a plurality of specific colors. The evaluation amount regarding sex is shown, and f shows a function. Further, this function may be a linear or non-linear mapping from the N-dimensional space to the one-dimensional space, or may be a logical expression.

In this way, the method of using the evaluation amount relating to the dominance of a plurality of specific colors when determining the correction level CL relating to the correction of the exposure amount is aimed at improving the reliability and accuracy of the correction level CL. Is effective in. The CL obtained as described above and the density correction value D calculated based on the two-dimensional image density information Xk (I, J) in the image processing unit 11 in parallel with the processing of steps 2 to 4 are It is transmitted to the exposure control unit 9 through the communication line 12 and is used for determining the exposure amount for photo printing.

Next, the determination of the photographic printing exposure amount in the exposure controller 9 will be described. Photographic exposure is calculated by the following formula (1
Determined according to 2). _{E k = LATD k -LATD 0k -C} k + D + E 0k ··· (12) where, E _k is B, G, the exposure amount of each color of R (logarithm of the exposure time), LATD _k in an exposed portion 3 The average transmitted light photometric value (density) from the original image to be printed which is provided by the photodiodes 7a, 7b and 7c, LATD _0k is the average transmitted light photometric value (density) from the standard original image, and C _k is the image processing unit 11. Correction amount based on the correction level transmitted from the device, D is the density correction value transmitted from the image processing unit, E _0k is the exposure amount (logarithmic value of exposure time) set for the standard original image, and k is B or G. , R are shown.

Here, the correction amount C _k is obtained by the following equation (13) using the collection level CL. _{C k = CL × (LATD k} -LATD 0k -ND + ND 0) ··· (13) However, ND, ND ₀ respectively LATD _k, LATD _0k
Is an average value for B, G, and R. In the right side of Expression (9), the values in parentheses are high values for relatively high density components and low values for relatively low density components. Therefore, for example, for an original image photographed against a background of a green grove, the amount of exposure of the G photosensitive layer of the photographic printing paper is reduced and magenta coloring is suppressed, so that a good color balance is obtained. Be done.

As described above, with the exception of such a case, the value of CL is 0 in most cases, so that the exposure amount is not corrected and most of the original images are exposed by full correction. Therefore, it is possible to obtain a photographic print that is hardly affected by the color temperature of the photographing light source and the characteristic difference depending on the type of photographic film. In the above, LATD0k
Is the average transmitted light photometric value (density) from the standard original image,
This value is the standard image or B, G, R of many images.
It may be obtained from the average transmitted light photometric value (density) of each color.

In this way, the exposure amount is obtained based on the average transmitted light photometric value of the original image, and is selectively corrected based on the collection level obtained in consideration of the dominance of the specific color in the subject.

[0046]

As described above, the photographic image information processing method of the present invention evaluates the correlation between the image information of two different colors among the image information of each color obtained by color separation scanning a color photographic original image. However, since it is characterized in that the attribute of the original image is determined based on the evaluation result, the attribute of the color original image, that is, the color feria is used regardless of the type of the photographic film, the storage state, and the type of the photographing light source. It is possible to stably and accurately determine whether or not there is a color distribution bias in the subject such as whether or not there is, and it is possible to optimize and automate color correction in photo printing represented by collection level selection.

Further, it is assumed that the two different colors have a complementary color relationship with each other, and based on the statistical amount of the image density information,
It is possible to make a stable determination with high accuracy without being affected by noise. Further, it is assumed that one of the two different colors is a neutral color, and stable determination can be performed with high accuracy without being affected by noise based on the statistical amount of the image density information. ..

Further, since the correlation is evaluated by the correlation coefficient, and the attribute is not discriminated by the density information of a specific pixel, it is not affected by the noise included in the image information and is extremely small. It is possible to make a stable determination with high accuracy. Further, the attribute of the original image is related to the selection of the collection level in the photo printing, and the precise control of the photo printing exposure amount can be realized.

Further, this discrimination is a multi-valued logical process, and since there is no variation in the discrimination around the threshold value as seen in the binary logical process, high reproducibility can be secured. Therefore, in particular, it brings a very beneficial effect on the photographic printing process.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram of a photographic printing apparatus to which a photographic image information processing method of the present application is applied.

FIG. 2 is a block diagram showing an internal configuration of an imaging unit.

FIG. 3 is a configuration diagram of a color separation filter.

FIG. 4 is a block diagram showing an internal configuration of an image processing unit.

FIG. 5 is a flowchart showing a process of image information processing.

FIG. 6 is a characteristic diagram showing a correlation between image density information of two different colors.

FIG. 7 is an example of a determination function that gives a collection level.

[Explanation of symbols]

 1 Spool 2 Spool 3 Exposure part 4 Light source 5 Mixing part 6 Lens 7a, 7b, 7c Filter 8a, 8b, 8c Photodiode 9 Exposure control part 10 Imaging part 11 Image processing part 12 Communication line 13a, 13b, 13c Cut filter 14 Shutter 20 CCD 20a Image signal 21 Color separation filter 22 Signal processing circuit 22a, 22b, 22c Image signal 23 Drive circuit 23a Clock signal 23b Timing signal 23c Horizontal sync signal 23d Vertical sync signal 30 Analog switch 31 Sample hold circuit 32 A / D converter 33 Timing Control Circuit 34 LUT (Look Up Table) 35 Primary Image Memory 36 CPU 37 Secondary Image Memory F Photographic Film P Photographic Paper

Claims (5)

[Claims] 1. A color image original image is color-separated and scanned, and among the image information of each color, the correlation between image information of two different colors is evaluated, and the attribute of the original image is determined based on the evaluation result. A photographic image information processing method comprising: 2. The photographic image information processing method according to claim 1, wherein the two different colors have a complementary color relationship with each other. 3. The photographic image information processing method according to claim 1, wherein one of the two different colors is a neutral color. 4. The photographic image information processing method according to claim 1, wherein the evaluation of the correlation is based on a correlation coefficient. 5. The photographic image information processing method according to claim 1, wherein the attribute of the original image relates to selection of a collection level in photo printing.