JPH0658504B2 - Image exposure method capable of obtaining proper exposure amount - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an image exposure method, and more specifically, it divides two-dimensionally into a plurality of pixels for detection, and automatically detects them using a calculation means. The image exposure method is adapted to determine the exposure amount and / or the exposure correction supplement amount for image exposure.

[Background of the Invention]

When automatically determining the exposure amount in this type of technology, for example, in the exposure method for photographs, the RG of the entire screen area is usually used.
The amount of transmitted light of the B3 color component is controlled to be constant, so that an image having a well-balanced color balance and print quality is created. This is based on an empirical rule that the average reflectances of the three colors obtained by integrating the entire normal shooting scene are almost constant. That is, when a general negative neutral subject is photographed with a color negative film under a white light source, the average transmission density is excessive or insufficient in the exposure amount, the light quality of the light source at the time of photographing, and the RGB of the color negative film.
Although it changes depending on the sensitivity of the photosensitive layer, the presence or absence of a mask, and the like, these changes can be absorbed by making the exposure exposure amount of RGB constant during printing, and a good print can be obtained.

On the other hand, when the color distribution of the subject deviates from the statistical average, the energy distribution of the light reflected by the subject is also not RGB constant, so that the three colors of the color negative film produced in that case. The concentration change of can not be correctly controlled by the above method. In addition, when the brightness configuration is extremely biased compared to the normal brightness distribution, such as when there are extremely high brightness areas or extremely high low brightness areas compared to normal scenes, the density of the negative film Since the change is due to the area change of the density of the subject, it cannot be corrected by the method of controlling the printing exposure amount only by the average transmission density. Similarly, even when the main subject constitutes a shadow portion or an extreme highlight portion as compared with surrounding subjects, the setting conditions of the printer are largely different and correction cannot be performed. As a method of determining a printing exposure amount for solving such a problem, a negative screen is divided and image information characteristic values are individually obtained as in JP-A-56-23936 and JP-B-56-2691. There is a method of obtaining an appropriate exposure amount for the scene from the respective divided screen information characteristic values. For example, as an example thereof, the average transmission density of the entire screen is D _a , the maximum density of the entire screen is D _max , and the average density of the central portion of the screen is D _max .
The exposure amount X ₁ of the 135F size film when the _center is set is X ₁ = a ₁ · D _a + b ₁ · D _max + c ₁ · D
_center + D ₁ (1), the exposure amount X ₂ of the 110 size film is X ₂ = a ₂ · D _a + b ₂ · D _max + c ₂ · D
_center + D ₂ It is expressed by (2). If the correction formula X _s such as X _s = K _i + K _j · X (3) is prepared for each film size with respect to the exposure amount X thus obtained, it is possible to correspond to the size. It is possible to print a negative film with the exposure amount X _s corrected correctly. Where the coefficients a ₁ , b ₁ ,
_{c 1, D 1, a 2} , b 2, c 2, D 2, K i, K j is what is determined in advance by experiments or the like for each film size.

In addition, instead of measuring the image information characteristic value for the divided area of the desired shape from the beginning, the screen is made two-dimensional by an image sensor such as CCD or photodiode array.
Desirably, the image information characteristic values such as the density or the transmittance are measured for each pixel by substantially evenly dividing, and the characteristic information for each divided area is obtained by appropriately combining the characteristic values for each pixel based on the shape of the screen division. It can be used to determine the exposure dose.

Therefore, when the amount of exposure is determined or the amount of exposure is corrected based on the information from the divided areas by dividing the screen as described above, how to divide the screen becomes a problem. If the division is different due to the difference in the original image film format such as the difference in film size printing direction, there is a problem that the arithmetic processing becomes complicated. This problem arises because the number of divided areas and the positions of divided screens can be made to correspond one-to-one even between different original picture formats. The problem can be solved by classifying and re-arranging in the format of. However, in this case, it has been difficult in the prior art to easily and efficiently classify various original picture formats into intermediate formats. Moreover, the original format sometimes changes. Disc film, for example, is the newest emerging format, but it has a different original size than any conventional film. It is highly possible that newer original formats will appear in the future. Also, if there is a deviation in the distance or angle between the image sensor and the original image due to work accuracy (variation in mounting) or aging, the size and position of the original image in the light receiving area of the image sensor will change even if the original image format is the same. It will change. Even in such a case, we will respond flexibly,
Conventionally, there has not been an image exposure method capable of automatically calculating an exposure amount and performing exposure.

OBJECT OF THE INVENTION The present invention has been made as a result of taking the above points into consideration, and has a simple structure and various formats of an original image such as a film which may exist now and in the future, and an image sensor and an original image. It can flexibly cope with deviations in distance and angle, efficiently and at high speed, accurately and precisely detects the image information of the original image such as negative film, and automatically adjusts the exposure amount according to the image information of the original image. The automatic exposure amount can be calculated flexibly in the event of a deviation in the distance or angle between the image sensor, which is the image reading means, and the original image. It is an object of the present invention to provide an image exposure method capable of exposing.

[Structure of Invention]

The present invention relates to an image exposure method for automatically determining a printing exposure amount and / or an exposure correction amount, which is obtained in pixel units by dividing image information including an original image on a color negative film into a plurality of pixels. Detect the characteristic value,
Based on the classification table, the characteristic values are classified into a plurality of virtual pixel sets each composed of 0 or more pixels, and the characteristic values of the pixels obtained by dividing the original image into screens In the image exposure method of obtaining information and determining the exposure amount and / or the exposure correction of the original image based on at least a part of the information, the classification table is an image characteristic value obtained by reading the reference original image. It is characterized by being created based on.

In the present specification, a split screen in an intermediate format that does not depend on the format of the original image is called a virtual screen, pixels constituting the virtual screen are called virtual pixels, and the original image is screen-divided by an image sensor or the like and captured. The original pixel is called a physical pixel, and the screen composed of the physical pixel is called a physical screen. In general, one virtual pixel is composed of zero or more (not exceeding the maximum number of all physical pixels) physical pixels.

The reference original image in the present invention refers to an original image having a certain density and pattern, and for example, an image such as a negative blank portion can be used.

The classification table in the present invention is used as a classification means for converting a physical pixel set into a virtual pixel set, and it should be in a table format that can reproduce a state in which the original image film is divided into screens for each pixel. Is a means of facilitating arithmetic processing.

The image information is read by photoelectric conversion using reflected light or transmitted light of the original image.

For the light-sensitive material of the present invention, for example, range coated paper, baryta paper, white terephthalate, etc., color photographic paper having a support as a support can be used.

In the present invention, the exposure amount and / or the exposure correction amount is determined based on the two-dimensional image information. There are a case where the exposure amount is determined only by the two-dimensional image information and a case where the conventionally used LATD control method is used together.

Two-dimensional image information can be read using an image sensor such as a CCD.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings. The examples described below are examples of the present invention, and the present invention is not limited thereto.

In this embodiment, the following method is adopted. That is, for a color negative film which is an original image, image information including a negative is divided into a plurality of pixels by a two-dimensional image sensor such as a CCD, a characteristic value obtained in each pixel is detected, and the characteristic value is classified. On the basis of the characteristic values of the pixels obtained by dividing the original image into screens by dividing the original image into a plurality of virtual pixel sets each of which has a number of pixels of 0 or more. And an exposure amount correction of the original image is determined based on at least a part of the above, and the classification table is created based on the image characteristic value obtained by reading the reference original image. did.

Various methods are conceivable as a method of dividing the image information including the original image into a plurality of pixels (physical pixels) and detecting the characteristic value in pixel units. For example, a CC using a light receiving element is used.
It can be easily detected by using an image sensor such as D (one-dimensional or two-dimensional) (Japanese Patent Publication No. 56-2691 and Japanese Patent Publication No. 60-177337).

The characteristic value is image information in which an original image is detected for each pixel by an image sensor or the like, and means light intensity in this embodiment, but may be transmittance or density.

The above classification table is created based on the image characteristic value obtained by reading the reference original image means that the image characteristic value is detected for each pixel from the reference original image by using the image sensor or the like, and based on this, The non-fixed classification table is created according to the image. Here, the light intensity was detected by setting the negative part of the negative as the reference original image, but it is also possible to detect the light intensity by setting the image without setting anything or the standard original image of the printer control set. it can.

More specifically, first, the physical pixel group is classified into two valid / invalid pixel groups by setting a threshold value by a function of a characteristic value in the physical pixel group forming the physical screen. The function of the characteristic value of the entire screen is obtained based on the obtained characteristic value of each pixel, and the threshold value for classifying the effective pixel and the invalid pixel is obtained from this. Specifically, the third
As shown in the figure, the two-dimensional image information including the original image (eg, 135F size negative film) obtained by an image sensor such as CCD is roughly divided into effective pixels that can be used as information for determining the exposure amount. There is a group and an invalid pixel group that cannot be used as information. Actually, the boundary portion is not clear, and the position, size, shape, etc. of the effective pixel group constantly change due to the image sensor and the displacement of the image position. Therefore, it is necessary that at least the area for obtaining appropriate image information is within the effective pixel group, and therefore, it is necessary to determine the effective pixel group as a preliminary problem. Then, using a histogram (frequency distribution function), an image as shown in FIG. 4 (a) is represented in a graph as shown in FIG. 4 (b).
(In this embodiment, each pixel is represented by 8 bits, so the light intensity in the histogram takes a value from 0 to 255.) Here, by dividing by a constant threshold value (a value indicated by a dashed line, for example, x) , Valid / invalid can be classified into two pixel groups. In the present embodiment, the cumulative frequency distribution function (hereinafter also referred to as CDF) is preferably adopted as the function of the characteristic value for setting the threshold value, but the frequency distribution function can also be used.

When actually using this cumulative frequency distribution function to set a threshold value, the dynamic range (Dr) is first determined. As shown in FIG. 5, the maximum value (here, 240) and the minimum value (here, 10) of the light intensity are taken and calculated by the following formula.

Dr = maximum value-minimum value = 240-10 = 230 Next, the values near the maximum and minimum values of the dynamic range are cut at a constant rate. This is often done in image information processing, since effective information is often not included near the maximum and minimum values of the dynamic range. In the present embodiment, the following processing is performed by using a value obtained by cutting the fixed ratio by 5% each from the maximum value and the minimum value, but the ratio is not limited to this. The respective points after cutting are designated as P ₉₅ and P ₅ as shown in FIG.

Further, the CDF value from that point is taken, and CDF (P
₉₅ ) → C ₁ and CDF (P ₅ ) → C ₂ .

The point X that internally divides between C ₁ and C ₂ into m: n is obtained by the following formula. Here, although the optimum value of the ratio of m: n varies depending on the image pickup system, m: n = 7: 3 in this embodiment.

Then, the inverse function of X, that is, the value of the light intensity at the X value of CDF, CDF ^-1 (X), is used as the threshold value to classify the valid / invalid pixel group.

Based on the threshold value obtained as described above, a group of pixels having the light intensity or more is set as an effective pixel group. Further, an image information area which is an area for actually obtaining image information is set in the effective pixel group. The image information area can be set using the following procedure, for example. The following procedure is an example, and the present invention is not limited to this.

In order to obtain the approximate center of gravity within the effective pixel group, first the effective pixels are counted to obtain the total area of the effective pixel group, and this is designated as A. Next, as shown in the figure of the image classified into valid / invalid pixels in FIG. 6, each valid pixel is sequentially counted from top to bottom in the horizontal direction (X-axis direction) of the image. Then, the Y coordinate of the pixel whose accumulated area exceeds A / 2 is defined as Cy. X-coordinates of pixels whose cumulative area exceeds A / 2 by sequentially counting each effective pixel from left to right (or right to left) toward the image (subordinate direction) Is Cx. X,
A pixel whose Y coordinate is Cx, Cy is substantially the center of gravity. The top and bottom are examined from the approximate center of gravity to detect the edge part. Set the top and bottom widths slightly inside, looking at the margin. This is to obtain a proper image information by taking a margin because the information near the edge has variations. After going up and down from the approximate center of gravity, the edge part is detected in each pixel of the upper and lower limits of the upper and lower width after the margin is taken (upper part 2 points, lower part 2 points, total 4 points) Compare the lower part with the lower part and choose the narrower width. Here, the width of the upper portion is the width on the left and right. As with the setting of the top and bottom width, set the final left and right width inside with a margin for the narrow left and right width.

As described above, in order to create the classification table prior to the actual printing method, the threshold value is used to set the area of the effective pixel group, but as another means, the differential processing is performed as an image to detect the edge. The area may be set by doing so. A predetermined standard original image film can be used as the reference original image film, but a blank negative film may be used.

Here, the size of the area formed by the pixel group determined to be effective is obtained. For example, as shown in FIG. 7, the pixel having the maximum row coordinate and the pixel having the minimum row coordinate among the pixels determined to be valid are the row coordinates R _max and R _min , which are similarly determined to be valid. The column coordinates of the pixel having the maximum column coordinate and those having the minimum column coordinate are C _max ,
_Let C _min . Moreover, the format of the virtual screen is predetermined and is assumed to be U × V. Here, if (R _max −R _min +1) <U or (C _max −C _min +1) <V, the size is insufficient, so that, for example, a warning signal such as an alarm is output, or It is also possible to stop the image processing together with the output and wait for the next treatment.

On the other hand, the physical screen is M × N. Therefore, an array R (i) consisting of M elements is prepared as a row classification table, an array C (j) consisting of N elements is prepared as a column classification table, and an auxiliary array K (i) is used as follows. Can be used to create row and column sort tables. The eighth
The flowchart of the do loop in the figure shows the equivalent of (M ≦ N) do i = MtoN processing end do.

Creation of row classification table X = 2 · ((R _max −R _min +1) div2) Y = XdivU Z = (XmodU) div2 W = ((M−1) div2) +1 do i = 1 toM R (i) = 0 end do do i = 1 to U K (i) = Y end do do i = W−Z to W + Z−1 K (i) = Y + 1 end do a = R _min do i = 1 to M do j = 1 to K (i) R (a) = I + 1 a = a + 1 end do end do Creating a column classification table X = 2 · (((C _max −C _min +1) div2) Y = XdivV Z = (XmodV) div2 W = ((N−1) div2) +1 do i = 1toN C (i) = 0 end do do i = 1toV K (i) = Y end do do i = W−ZtoW + Z−1 K (i) = Y + 1 end do a = C _min do i = 1toV do j = 1toK (i) C (a) = i + 1 a = a + 1 end do end do However, the above mod means a remainder, and div means an integer division (when a remainder is cut off).

A method of storing the classification table created as described above by using a non-volatile storage means, that is, one that does not lose the stored information even when the power is turned off (for example, a proppy disk or a memory with a battery backup) It can also be adopted. As a result, a classification table corresponding to the usage status is created once, and the stored classification table can be repeatedly used until the usage status changes, which facilitates handling.

Classifying the screen-divided physical pixel set into a virtual pixel set using the created classification table means that the characteristic value detected in physical pixel units is used at high speed using the created classification table. This is performed in order to absorb the difference due to the film size and / or the printing method by once replacing with an intermediate format (here, a virtual pixel set) that does not depend on the format of the film and / or the printing direction.

The above description will be specifically described below.

The light intensity at the i-th row and the j-th column of the negative film captured by the image sensor for each physical pixel is represented by DN _{ij in} FIG.
By preparing the classification table as shown in (a) in advance, the classification value of DN _ij (that is, the coordinate value of the virtual pixel)
Can be obtained by simply referring to the i-th row and the j-th column of the classification table. Values are written in the classification table in the same number as all the physical pixels captured by the image sensor in a divided manner. That is, for example, as shown in FIG. 1 (a), the physical screen captured by the image sensor is
As long as the classification table is composed of 12 × 8 physical pixels, the classification table of the same 12 × 8 size is prepared. In this example, the classification table is created on the assumption that the virtual screen is composed of 4 × 4 16 pixels. The classification table contains 0 and integers from 1 to 16. Here, the value of the classification table is n (1
The physical pixel corresponding to the position ≦ n ≦ 16) is classified as the nth virtual pixel. However, the physical pixel corresponding to the position where the value written in the classification table is 0 means that the corresponding virtual pixel does not exist, and is invalid data and is ignored in the processing. For example, assuming that FIG. 1 (a) is a classification table for a 110-size negative film, when classifying pixels of a 110-size negative film, the classification value of pixels in the third row and fourth column is 1, The classification value of the pixel in the fourth row and fifth column is 6. Therefore, with respect to all the characteristic values taken in for each pixel, it is possible to easily associate the virtual pixels with each other by referring to the classification table in which the physical pixels are in one-to-one correspondence. High-speed classification is possible. If the negative film format is different,
In the case of the 135F size, it is possible to perform the same classification by preparing a classification table as shown in FIG. 1 (b).

In this way, when printing a plurality of negative film formats, it is possible to make the classification method the same by preparing a plurality of classification tables, and the processing after conversion to virtual pixels is The processing is easy because they can be exactly the same. Note that it is not always necessary to prepare a classification table for every negative film format.

Further, it is possible to achieve the same technical idea without necessarily using a classification table having the same size as the physical screen. Two examples will be given below.

As a first example, when the classification table of FIG. 1 (a) is desirable, the minimum row and column values in which a classification table other than 0 is written as the classification table value and the maximum row and column values Of the values and the original classification table, only the part in which a non-zero value is entered may be prepared.

This row is shown in FIG. 2 (a). In this case, the smallest row value is R
_min , the maximum row value is R _{max, and the} minimum column value is C
When _min maximum column values to _{C max,} the classification values _{V ij} for _{DN ij} is Required by. However, 0 in the classification table means that there is no virtual pixel corresponding to the physical pixel. In this case as well, conversion to a virtual screen is performed in the same classification procedure by preparing a plurality of classification tables for different formats of negative film.

As another example, if the classification table of FIG. 1 (a) is desirable, as shown in FIG. 2 (b), the classification table is retained in a decomposed form by orthogonal decomposition in the row direction and the column direction. You can do it. The classification table of FIG. 2 (b) can be easily reproduced by performing the following calculation. After the reproduction, the same image processing can be performed, and the classification table can be stored with a small storage capacity.

Here, the classification table in the row direction and the classification table in the column direction can be obtained by the following means. For example, in this example,
Since the virtual screen is composed of 4 × 4, the classification value V _ij
Takes values from 1 to 16 and 0. Here, the following calculation work is performed.

The row classification table R _i and the column classification table C _j are obtained using the U _ij thus obtained.

To use the row classification table and column classification table obtained in the above procedure, By performing the above calculation, the classification value is uniquely obtained from the row and column coordinates _ij .

By adopting this method, the classification table can be stored with a small storage capacity, and when used, the original classification table can be easily derived by calculation.

Further, the configuration of the classification table is not limited to the above example, and generally, any configuration can be considered as long as it involves a mapping unit that uniquely finds the coordinates of the virtual pixel based on the coordinate information of the physical pixel. To be For example, the classification table is decomposed into a row classification table and a column classification table, and the non-zero portion and the non-zero
It is also possible to retain the values of the first row and column of and the values of the last row and column of as a classification table. Alternatively, the same effect can be obtained by holding the values of the first row and column, the number of non-zero elements in the row, and the number of non-zero elements in the column.

Each of the above-mentioned methods has advantages and disadvantages, and which method is adopted is arbitrary, but can be appropriately selected according to the storage capacity of the computer, the calculation speed, the target of information processing, and the like.

Next, as a method for obtaining an information group for each virtual pixel unit from the characteristic value detected in pixel units by dividing the original image such as the negative film, screen division is performed for the intermediate format as in the conventional technique. It is performed again and is extracted by extracting a very small amount from the re-divided screen. Specifically, the exposure amount determination formula (1) for each film size and
Formula (2) can be expressed as, for example, X = a · Da + b · D _MAX + c · D _MIN + Dd.
The image information of the split screen common to each film size can be obtained for each virtual pixel by using a method such as obtaining the characteristic value of the physical pixel included in each virtual pixel. With such a calculation method, the characteristic amount of each virtual pixel (for example, the average value, the maximum value, the minimum value of the characteristic values, etc.) can be easily calculated from the data of the physical pixel.

Using at least a portion of the information set obtained by the above method, the exposure and / or the printing exposure for a negative film is used.
Alternatively, the exposure correction amount is determined by substituting the feature amount for each virtual pixel obtained by the method into a common exposure amount determination formula to determine the exposure amount for the original image. Specifically, the exposure amount X is obtained from the above equation (4) for negative films of various sizes conveyed to the printing section. Further, the exposure amount X can be corrected according to the size according to the above formula (3).

Further, according to this embodiment, it is used as an image reading means.
Even if there is a gap in the distance or angle between the image sensor such as CCD and the original image due to work precision (variation in mounting) or changes over time, it is possible to flexibly deal with this, even if the original image format is the same. It is possible to prevent the size and position of the original image from changing in the light receiving area of the sensor.

〔The invention's effect〕

As described above, according to the present invention, it is possible to flexibly cope with a change in various formats of an original image and a shift of the position of the image sensor or a shift of the optical axis with a simple configuration, that is, for example, an image sensor as an image reading unit Even if there is a deviation in the distance or angle between the original image and the original image, this can be flexibly dealt with, and effectively, and therefore the image information of the original image can be detected accurately and in detail at high speed. Image printing can be performed with an appropriate exposure amount according to original images of various sizes.

[Brief description of drawings]

FIGS. 1 (a) and 1 (b) are explanatory views of a classification table having different film sizes in the embodiment of the present invention, and FIGS. 2 (a) and 2 (b) are explanatory views showing a storage means of a simpler classification table. Figure 3rd
The figure shows 2 of original film obtained by image sensor.
It is a figure of three-dimensional image information, and FIG. 4 (a) (b) is a figure which respectively shows the corresponding relationship of the two-dimensional image information of an image sensor and its histogram, and FIG. 5 uses a cumulative frequency distribution function. FIG. 6 is an explanatory diagram of a method of classifying valid / invalid pixel groups, and FIG. 6 is an explanatory diagram of a method of setting an image information area from within the valid pixel group. FIG. 7 is an explanatory diagram of a method of determining the size of the constituent area of the effective pixel group, and FIG.
It is a flowchart figure of a loop.

Claims (3)

[Claims] 1. Image information including an original image on a color negative film is screen-divided into a plurality of pixels, a characteristic value obtained for each pixel is detected, and the characteristic value is 0 or more pixels each based on a classification table. The virtual image is classified into a plurality of virtual pixel sets, the information of the previous virtual pixel is obtained from the characteristic value of the pixel obtained by dividing the original image into screens, and the original image is obtained based on at least part of the information. In the image exposure method for determining the exposure amount and / or the exposure amount correction, the classification table is created based on the image characteristic value obtained by reading the reference original image. 2. The image exposure method according to claim 1, wherein the divided areas of the classification table are formed symmetrically with respect to the center. 3. The image exposure method according to claim 1, wherein the divided areas of the classification table are formed such that the central area is wider than the peripheral area.