JP3352112B2 - Image information processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image information processing apparatus which divides image information of an original image into pixels by using an image sensor to read a digital image, and performs information processing on the original image using the digital image. The present invention relates to an image information processing apparatus suitable for use in a photographic printing apparatus that reads an original image on a color photographic film and determines a printing exposure amount of the original image based on the image information.

[0002]

2. Description of the Related Art In general photographing, a subject B (blue), 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 a conventional photographic printing apparatus, the average transmission density (LA) of the entire area of the photographic original is
TD) is measured, and the exposure in photographic printing is determined based on the measured average transmission density, whereby the exposure to be given to the B, G, and R photosensitive layers of the photographic paper is controlled to a constant value, and the color balance is adjusted. (Referred to as LATD control method).

However, this LATD control method has a disadvantage that it is difficult to obtain a proper photographic print when the luminance distribution and the color distribution are uneven in the subject. These original photos are called subject feria, especially if they are caused by uneven brightness distribution of the subject.
In addition, a case where a color distribution is biased is called a color feria.

For this reason, in the current high-performance photographic printing apparatus, a negative image is divided into fine pixels, and image characteristic values (for example, optical transmission density and optical transmittance) of each pixel are measured. Or a method of determining an exposure correction amount for the exposure amount determined by the LATD method. Such an apparatus is generally called a scanner. In a scanner, image information of a substantially entire negative image is two-dimensionally input by image input means, and preferably sampled in substantially equally divided pixel units. The image characteristic values are measured in the following manner, and the characteristic values of the respective pixels are appropriately combined based on the shape of the screen division to obtain the image characteristic values of each divided region.

Conventionally, in such a scanner, as described in, for example, Japanese Patent Application Laid-Open No. 60-177337, an image sensor including a plurality of light receiving elements is driven by a driving circuit, The signal is sampled and held by a sample and hold circuit, the value is digitized by an A / D converter, and the digital value is written to the memory in accordance with the driving speed of the image sensor. Image information is read in pixel units corresponding to the numbers.

[0006]

However, the above configuration has the following problems. One is that such a configuration strongly depends on the number of light receiving elements of the image sensor. An image sensor (two-dimensional CCD) that is widely used at present is configured by arranging a plurality of light receiving elements (photodiodes). The output signal obtained by driving such an image sensor in a time series manner is not continuous, but is a rectangular wave signal in which the voltage drops sharply at a portion corresponding to the boundary of each light receiving element. For this reason, an internal sample-and-hold circuit is added to the drive circuit of the image sensor, and a signal resulting from sample-and-hold of the output is output in synchronization with the reading of the output from each light receiving element constituting the image sensor. ing.

In order to obtain a digital image signal from an analog image signal of such an image sensor, it is necessary to perform sampling as well as digitization by a sampling means. In this case, in order to obtain a correct image signal, the sampling timing by the sampling means must be synchronized with the drive circuit of the image sensor, and the process of digitizing the analog sample signal (ie, A / A
D conversion) must also be performed in synchronization with it.

On the other hand, since a recent image sensor is provided with a dedicated drive circuit in advance, it is desirable to drive the sensor using the dedicated drive circuit from the viewpoint of cost and performance. Therefore, the driving frequency of the image sensor is
It is almost fixedly defined, and the operation speed and timing of the sample and hold circuit and the A / D conversion circuit constituting the sampling means are strongly restricted. That is, if the sampling pitch is kept constant, it is limited to whether sampling and A / D conversion are performed in units of pixels constituting the sensor or at a sampling pitch that is a fraction of the integer. In addition, as described above, since the driving frequency of the image sensor is limited, the driving frequency of the sampling circuit and the A / D conversion circuit is not the same as the read frequency of the image sensor or is not an integral number thereof. It will not be. Actually, these circuits have conventionally been operated at the same frequency as the readout pitch of the image sensor.

However, as long as such a configuration is employed, there is a disadvantage that the number of samples of the original image information depends on the number of pixels of the image sensor, and arbitrary setting cannot be performed. Further, the sampling pitch is also limited by the number of pixels of the image sensor. As a result of these restrictions, ・ The circuit configuration of the image processing system becomes complicated and the number of parts increases. ・ Use of expensive elements capable of high-speed operation as circuit components such as A / D converters and memories. This has the disadvantage of increasing costs. ・ Adjustment of circuit assembly is troublesome.

Further, as described above, in order to write digitized pixel values in the memory in synchronization with the drive circuit of the image sensor, a buffer memory having substantially the same capacity as the number of pixels of the image sensor is required. I will. This memory capacity is enormous, it must be able to operate at high speed,
There is a problem that a very expensive memory element is required. Moreover, saving the memory capacity causes another problem that the write control circuit for the memory becomes complicated.

Another disadvantage is that if image information in pixel units constituting an image sensor is taken into a memory, the pixels are too fine and the information becomes excessive depending on the purpose of use. For example, in image processing for determining an exposure amount in a photographic printing apparatus, it is necessary to perform various kinds of statistical processing on image information, such as average density, maximum density, minimum density, and variance of each divided area on a screen. In order to shorten the processing time and to use different formats (shapes and dimensions of images) for each film, a rounded image obtained by appropriately simplifying the original image is used. However, if the original source image information is enormous, the rounding operation itself takes a lot of time, and even though there is fine sample data, rough thinning processing in units of "rows" or "columns" can be performed. And have to use only a small part of the valid data. However, in this case, data collected using an expensive circuit is wasted and discarded, which is a major concern in the circuit configuration of the scanner.

SUMMARY OF THE INVENTION In view of the above, the present invention provides an image information processing apparatus comprising an inexpensive circuit which achieves a reduction in the number of components and a lower operating speed while maintaining the quality of image information. It is an object. It is also an object of the present invention to provide an image information processing apparatus which has a degree of design freedom to determine the number of sampling pixels irrespective of the number of photometric elements of an image sensor, and which can shorten the processing time while making full use of effective image information. It is an object.

[0013]

In order to achieve the above object, the present invention provides an image sensor unit comprising a plurality of light receiving elements and an original image signal at a predetermined read frequency from a light receiving element group of the sensor unit. And a signal filter unit that obtains a smooth image signal by passing the original image signal through a low-pass filter, and a sampling unit that obtains a sample image by sampling the smooth image signal at a predetermined sample frequency. A rounding unit that performs a rounding process on the sample image to obtain a rounded image, and an image processing unit that performs a predetermined image processing based on the rounded image, wherein the number of pixels Norigin of the image sensor is , The number of pixels of the sample image Nsamp
le and the number of pixels Nprocess of the rounded image satisfy the relationship of Norigin> Nsample ≧ Nprocess, and achieve a reduction in the number of components and a reduction in the operating speed while maintaining the quality of image information. It is made possible.

Further, the cutoff of the low-pass filter
Frequency F_cutAnd the sampling frequency F _sampleAnd the rounding
And the average rounding rate R of the_sample ≧ 2 ・ F_cut≧ R ・ F_sample Configuration that satisfies the relationship of
The number of sampled pixels can be determined regardless of the number of photometric elements.
To reduce processing time while fully utilizing effective image information
(However, the rounding coefficient R is
The number of pixels of the rounded image is calculated by
This is the value divided by the number of pixels that contributed to the configuration of the image.)

[0015]

According to the present invention, since the high-frequency signal component is removed from the original image signal output from the image sensor and then sampled, the signal output from the image sensor is a smoothed continuous signal. Therefore, the number of pixels to be sampled can be determined regardless of the number of light receiving elements of the image sensor. Therefore, an appropriate sample point can be set with a smaller number than the number of light receiving elements, and a sample image can be formed. However, this sample image depends on the film format.

Thereafter, a rounding process is performed on the sample image in order to provide consistency with each film format, and a rounded image independent of the film format is obtained.
Image information processing is performed using the rounded image. At this time, since the number of pixels of the sample image is reduced to an appropriate number, the rounding process can be performed at a high speed, and the rounded image can be formed by utilizing all the effective information.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to an embodiment shown in the accompanying drawings. First, the outline function of the image information processing apparatus of the present application will be described with reference to the principle diagram of FIG. In the figure, F is a photographic film, which is set on a first spool 1 and wound around a second spool 2 via a predetermined transport path. The photographic film F is formed by joining a plurality of films into one at a splice portion (not shown), and a notch portion (not shown) for detecting the position of the original image is formed at the center edge of each original image I. Have been. The start and end of the film are determined by detecting the splice, and the original image position can be confirmed by detecting the notch.

The original image I formed on the photographic film F is positioned at the exposure unit 3 and irradiated from the light source 4 to the mixing unit 5.
Illuminated by the light that has been made uniform. Further, the transmitted light is optically imaged on the photographic printing paper P by the lens 6 so that the light can be exposed. Where B,
Through the photometric filters 7a, 7b, 7c of G, R colors,
The average transmitted light of each color of B, G and R of the original image I is received by the photodiodes 8a, 8b and 8c. The B, G, and R photometric signals obtained by photoelectrically converting the average photometric value are supplied to the exposure control unit 9 and A / D-converted. As described later, the B, G, and The basic exposure amount for each of the R colors is determined.

In the exposure section 3, the original image I formed on the photographic film F is transferred to an imaging section (image sensor section) 10
Are separated into B, G, and R colors and scanned. The B, G, and R color image signals obtained by the imaging unit 10 are sent to the image processing unit 11, A / D converted, and shaped into image density information in a predetermined format. The image processing unit 11 calculates an exposure correction parameter from the image density information obtained in this manner by an exposure control method described later.
The obtained exposure correction parameter is transmitted to the exposure controller 9 via the communication line 12.

The exposure controller 9 corrects the basic exposure based on the above-mentioned average photometric value based on the exposure correction parameter transmitted from the image processor 11. Further, in the exposure control unit 9, the exposure amount obtained in this manner is reduced by the color-cut type cut filters 13 a, 13 b, 13 a, 13 b, 13 a, and 13 b arranged on the upper part of the exposure unit 3.
3c and the operation time of the shutter 14 disposed below the exposure unit 3. That is, the cut filters 13a, 13b, 13c and the shutter 1
4 is inserted into the exposure light path, and the exposure amount given to each color photosensitive layer of the photographic printing paper P is adjusted. After 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 I to be printed next on the exposure unit 3. I have.

The original image I formed on the photographic film F in this manner is configured to be sequentially printed. Next, the internal configuration of the imaging unit 10 will be described with reference to the block diagram of FIG. The original image I positioned in the exposure unit 3 and illuminated by the mixing unit 5 is optically projected on a two-dimensional image sensor (CCD) 20 by a lens 24 in the imaging unit 10 and imaged. On the two-dimensional image sensor 20, color-separated B, G, and R color filters 21 as shown in FIG. 3 are mounted so as to correspond to the individual photoelectric conversion elements. Charges accumulated in the photoelectric conversion elements are sequentially read in a direction perpendicular to the direction. As a result, the two-dimensional image sensor 2
From 0, the image signals 20a for the B, G, and R colors are mixed and output.

The image signal 20a thus obtained
In the signal processing circuit 22, color separation is performed for each of B, G, and R colors, and the color-separated image signal is sampled and held and amplified. Further, the amplified image signals 22a, 22b, and 22c of B, G, and R are output to the image processing unit 11 at the subsequent stage. In addition, a driving circuit (sensor driving unit) 2
3, 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 an A / D conversion and image memory in the image processing unit 11. A horizontal synchronizing signal 23c and a vertical synchronizing signal 23d for controlling writing of data are supplied respectively.

As the CCD two-dimensional image sensor of the present embodiment, one having a light receiving element of 510 × 493 pixels = approximately 250,000 pixels is preferably used, but it is a matter of course that the present invention is not limited to this. . In recent years, the number of pixels of a CCD image sensor for imaging has been increasing steadily in conjunction with the spread of home video devices, and application of an image sensor including a larger number of light receiving elements will be sufficiently considered in the future. On the other hand, in order to prevent the sampling circuit from becoming complicated, it is necessary to read image information with a sampling number different from the number of pixels of the image sensor and perform efficient processing.

In order to solve this problem, as shown in the block diagram of FIG. 4, the image processing section 11 performs the following signal processing. The image signals 22 a, 22 b, and 22 c supplied from the imaging unit 10 are selected by the analog switch 30. Since this output signal is sampled and held by the internal sample and hold circuit of the image sensor drive circuit according to the read frequency of the image sensor, it is an analog signal whose value changes discretely. If the signal is directly sampled at the frequency F _sample , accurate image information may not be obtained due to interference between the sampling frequency and the reading frequency of the image sensor.

Then, this signal is passed through a low-pass filter (signal filter unit) 100 to cut off the cutoff frequency F.
High-frequency components above the _cut are removed to obtain a smoothed signal. The smoothed signal is sampled at a predetermined frequency F _sample by an A / D converter 32 via a sample and hold circuit 31, and is converted into a digital signal.

Here, any known circuit may be used as the signal filter unit 100 as long as it can substantially filter a signal component having a frequency equal to or higher than a predetermined frequency. Is used. Image signal 22 in analog switch 30
Switching between a, 22b, and 22c and sampling timing in the A / D converter 32 are performed by the timing control circuit 33 based on the horizontal synchronization signal 23c and the vertical synchronization signal 23d output from the driving circuit 23 of the imaging unit 10. It is controlled.

Here, the number of samplings is 128 for each of B, G, and R colors in one horizontal scan, and such sampling is performed in each of 128 horizontal scans in one vertical scan. Has become. Therefore,
Each time vertical scanning is performed, digital image information of 128 × 128 pixels is obtained for each of B, G, and R colors. As described above, by smoothing the signal in the signal filter unit 100, it is possible to determine the number of samples / pitch independently of the number of pixels of the image sensor.
The A / D conversion is processed by 10 bits.

Next, the digitized image signal is given by RO
Are converted into image characteristic values (density values) via an LUT (look-up table, sampling unit) 34 composed of M and the like,
It is stored in the primary image memory 35. The LUT 34 includes
A conversion table represented by the following equation is stored. Y = a × log (X + b) where X is an input to the LUT 34 and Y is its output. A is a constant for converting the photometric density determined by the spectral characteristics of the imaging unit 10 in the scanning unit 19 into a printing density determined by the spectral sensitivity of the photographic paper, and b is a constant for removing a dark current component in the imaging. It is.

This conversion table can be appropriately changed according to the purpose. For example, a linear conversion represented by Y = c × X + d may be used.

The LUT 34 includes a constant a of the equation (1),
A plurality of conversion tables obtained by substituting several types of numerical values into b are prepared, and can be selected in advance by the CPU 36. However, the selected conversion table does not necessarily have to be the same for each of the B, G, and R colors. In this way, 12
The 10-bit image density information of each color of B, G, and R consisting of 8 × 128 pixels is stored in the primary image memory 35. The image density information stored in the primary image memory 35 is called a sample image. Note that an address at the time of writing to the primary image memory 35 is controlled by the timing control circuit 33 by a horizontal synchronization signal 23c and a vertical synchronization signal 23d output from the imaging unit 10.

Here, the effective image area in the image density information is the format (screen size) of the photographic film F.
In addition, since the number of output pixels of the two-dimensional CCD reaches tens of thousands of pixels, there is a problem that it is too large to perform various image processing operations necessary for controlling the exposure amount. So, CP
U36 (rounding section and proofreading means) performs processing of rounding (simplifying) an image area previously set as an effective screen according to the format of the photographic film F to an appropriate number of pixels. That is, regarding the density information of the sample image, the image is simplified by calculating an arithmetic average of a predetermined number of pixels as one group or thinning out the image at predetermined intervals. For example, Japanese Patent Application Laid-Open No. Sho 62-2
The technique disclosed in Japanese Patent No. 60136 can be used.

In this embodiment, the number of pixels obtained by this rounding process is independent of the format of the photographic film F.
Secondary image density information composed of 16 × 16 pixels is stored as a rounded image in a rounded image memory 37 (rounded image storage means). However, in this embodiment, since the image information is read after being separated into three colors of B, G, and R, a 16 × 16 rounded image is obtained for each color.
In practice, × 16 × 3 pixels will be obtained. In this case, the area determined as an effective screen is generally 128 × 12
8 is a part of the 8 sampling pixel areas. Further, the pixels subjected to the rounding are usually further a subset thereof.
That is, the average rounding coefficient R is calculated based on the number of pixels (16
× 16) by the number of pixels that contributed to the configuration of the rounded image among the pixels of the sample image.

The cutoff frequency F _cut of the low-pass filter, the sampling frequency F _sample, and the average rounding ratio R of the rounding section are as follows: F _sample ≧ 2 · F _cut ≧ R · F _sample Equation (2) It is configured so as to satisfy the following magnitude relationship, so that the number of sampling pixels can be determined regardless of the number of photometric elements of the image sensor. This makes it possible to shorten the processing time while making full use of the effective image information.

This equation is based on the sampling theorem. That is, in order to avoid high-frequency components from being mixed as noise, the sample frequency F _sample is set to the cut-off frequency F
It is desirable that the cutoff frequency is twice or more of the _cut , while, in order to obtain valid image information correctly, the rounding frequency R · F
_This is because it is desired that the _sample be less than twice the cutoff frequency _Fcut .

After the rounded image calibration processing is performed, the rounded image memory 37 is rewritten and stored as a calibrated image. If the arithmetic mean of the image density information is employed in the rounding process, an effect of reducing a noise component in the image signal can be expected. The processing time can also be shortened by performing arithmetic averaging after appropriate thinning processing and rounding by thinning out by a predetermined number of pixels without performing arithmetic averaging.

The number N _{origin of the} light receiving elements constituting the image sensor, the number N _{sample of} the pixels sampled by the sampling means, and the number N _process of the pixels obtained by performing the rounding process are as follows. The relationship of Expression (1) is established. N _origin > N _sample ≧ N _process Equation (1) However, the equal sign means that the rounding process is not substantially performed. This is the case when processing very small screen size film such as DISC film.
This is because a sufficient number of sample pixels (N _sample ) for performing the rounding process may not be obtained.

Here, N _origin is a value determined by the standard of the image sensor to be used. Therefore, as long as an image sensor supplied as a standard product from a device maker is used, the standard must be followed. on the other hand,
N _sample is determined by the operating frequency of the sampling unit.
This value can be appropriately determined according to the convenience of the device designer. Further, N _process is determined by the convenience of the subsequent image information processing, and is determined independently of the film format.

As described above, in the present invention, N _origin
Sampling the pixel information sampled in N _{samp le} pieces from pixel information of pieces of original image, N by means rounding it
_By rounding down to _{proc ess} and sequentially reducing the number of pixels, the sampling circuit is configured independently of the image sensor, and it is necessary and sufficient without complicating the hardware configuration, and furthermore, noise and the like are reduced. It is possible to obtain stable image information that is not easily affected.

Reference numeral 38 denotes a calibration image memory in which a calibration image composed of pixels that can correspond one-to-one with the pixels of the rounded image is stored in advance, and the calibration image is stored in shading correction (calibration work) described later. It can be supplied. It is preferable that the calibration images are configured to store three images for each color of B, G, and R, and to calibrate the images of the colors corresponding to the image information to be printed, respectively. It is also possible to substitute only one color, for example, G.

Further, for example, according to the following equation (3), B for each pixel, G, image characteristic values D _B of R, D _G, calculates the D _R weighted average D _N, composed from the average value D _N The obtained image can be stored as a calibration image. By doing so, the memory capacity required for storing the calibration image can be reduced to 1/3.
It is possible to save money. _{D N = (A B · D} B + A G · D G + A R · D R) ··· (3) where A _B, A _G, A _R is a constant, A _B + A _G + A
_R = 1 The image reading means of the imaging unit 10 is replaced with a line sensor (1) instead of the two-dimensional image sensor (CCD) 20.
While reading a one-dimensional image of the original image I in the main scanning direction using a two-dimensional CCD, the original image I is conveyed in a direction perpendicular to the line sensor by a stepping motor or the like and sub-scanned to obtain a two-dimensional digital image. May be configured.

That is, a one-dimensional CCD is used to perform a main scan to input a one-dimensional image, and then a sub-scan is performed line by line by transporting the original image to create a two-dimensional image (for example, 0.
(Convey at 25 mm / line to create a one-dimensional image of 128 lines.) This image may be converted into a density value by the LUT 34 to form a sample image.

In this case, the two-dimensional image obtained by the one-dimensional CCD may have a different number of pixels than the image obtained by the two-dimensional image sensor. The same pixel configuration can be obtained.
For example, when a 1024-pixel one-dimensional CCD is used,
The output signal from the sensor is 1 /
The sample-and-hold circuit is controlled by the signal obtained by dividing the signal into eight, and is input as a 128-pixel one-dimensional image, and then rounded at appropriate intervals in the same manner as in the rounding process described above, thereby forming 16 × 16 pixels. A rounded image can be obtained. This rounded image is stored in the rounded image memory 37.
(Rounded image storage means), the subsequent processing is 2
This is exactly the same as the case of reading with a dimensional CCD.

Next, the shading correction for the rounded image thus obtained will be described. First, a process of setting a reference calibration image used for performing a calibration operation will be described. FIG. 5 is a flowchart showing a setting process of the reference calibration image.

First, temporary calibration image information in which a uniform negative value is set for each pixel is stored in the calibration image storage means 38.
(Step S12, Step S1)
3). The reason why a uniform negative value is set as the temporary calibration image information is to prevent a negative value from being mixed into the pixel value of the reference calibration image information serving as a calibration reference for a large number of printed original images. . In this embodiment, the negative value is −
Although 100 is preferably used, it is not limited to this.

Next, as a reference image that directly reflects the shading state of the exposure system, for example, a blank negative is set in the exposure unit 3, and the image is calibrated in the same process as a normal reading operation for an original printed image. . That is, the imaging negative is read by the imaging unit 10 as a reference image (step S14) and stored as a reference sample image in the primary image memory 35 (sampling unit) (step S15).

Further, after the reference sample image is rounded by the CPU 36 (rounding means) to obtain a reference rounded image, the reference rounded image is stored in the memory 37 (storage means) (step S16). Then, a uniform negative value set in the calibration image storage means 38 is subtracted from the reference rounded image to create a reference calibration image (step S1).
7). At this time, it is clear that there is no possibility that a pixel having a negative value will occur in the reference calibration image.

Further, in the conventional apparatus, the read image is stored as it is as the calibration information. On the other hand, in the apparatus of the present invention, the rounding process is performed once, and then stored as the calibration image. The memory capacity of the storage means may be significantly smaller. In addition, the process of subtracting a uniform negative value, which is a temporary calibration image, from the reference rounded image is simple and is performed at high speed.

The reference calibration image thus obtained is
It is configured to be stored in the proof image memory 37 and to be used as a proof image in the subsequent baked original image (in the proof for the baked original image, the image information in the rounded image memory 37 is rewritten). Next, the procedure of shading correction for a normal photographic image is shown in FIG.
This will be described with reference to the flowchart shown in FIG.

In the present apparatus, as described above, the two-dimensional CC
A digital image is read by D (or one-dimensional CCD) (step S22), and the digital image is converted and sampled into a sample image composed of density information (step S2).
3) Further, a rounding process is performed by the CPU 36 (rounding means) (step S24). Then, calibrated image information is obtained by subtracting the calibration image in the calibration image memory 38 from the rounded image in the rounded image memory 37 (storage means) (S25, S26). The obtained calibrated image is stored in the rounded image memory 37, replacing the rounded image before calibration.

In the calibration for this burn-in, since the calibration is performed on the image after the rounding processing, the subtraction processing is simple and the high-speed processing is possible. That is, while the conventional apparatus performs calibration on a sample image of 128 × 128 pixels (FIG. 8), the apparatus of the present application performs calibration on a rounded image compressed to 1/64 pixels. It has become. Thus, by reducing the number of pixels of the image handled in the calibration stage, it is possible to save storage capacity and to speed up the calibration work.

As is clear from the above description, the main part of the process of creating the reference calibration image shown in FIG. 5 Image reading (step S14) → image sampling (step S15) → rounding processing (step S16) → calibration Subtraction of image information (step S17) and calibration process for normal burned-in image shown in FIG. 6 Image reading (step S22) → image sampling (step S23) → rounding process (step S24) → subtraction of calibration image information (step S24) Step S25) is substantially the same process.

In the above embodiment, the burn-in image and the calibration image are respectively converted into density information. However, each image corresponds to the amount of received light in accordance with the content of image information processing in exposure amount control. Processing may be performed with the output value as it is, and calibration may be performed. However, in the calibration operation in this case, since the burn-in image information is divided by the calibration image information, the processing time is longer than that of the density information image.

The calibrated image thus obtained is subjected to predetermined image processing, and the exposure controller 9 calculates desired parameters to determine the amount of exposure. The photographic printing exposure amount in the exposure control section 9 is determined according to the following equation. Here _{E KL = LATD KL -LATD0 K +} γ × CC KL + E0 K, E K is B, G, the exposure amount of each R color (logarithmic value of the exposure time), LATD _K is the exposed portion 3 the photodiode 8
a, 8b, the average transmitted light photometric value from the original image to be subjected to baking to Motasa by 8c (logarithmic), LATD0 _K is the average transmitted light photometric value from the standard original (logarithmic), CC _K color correction value,
γ is a coefficient for adjusting the strength of color correction, E0 _K is an exposure amount (log value of exposure time) set for a standard original image,
K is each color of B, G, and R, and L is the processing order of the original image.

As described above, the exposure amount is determined based on the average transmitted light photometric value of the original image, and is corrected by the color correction value determined in consideration of the tone reproduction characteristics of the photographic film. In the above embodiment, the original image I of the film F conveyed to the exposure unit 3 is illuminated by the light source 4 and the mixing unit 5, and the transmitted light is imaged by the imaging unit 10. That is, the original image formed on the two-dimensional CCD by the lens 24 is color-separated by the signal processing circuit 22, digitized by the A / D converter 32 in the image processing unit 11, and then subjected to LUT
At 34, it is converted into image density information and stored in the primary image memory 35 as a sample image.

Next, the sample image is rounded by the CPU 36 into a rounded image having a smaller number of pixels, and is stored in the rounded image memory 37. If the rounded image is a burned-in original image, shading correction (calibration) is performed using a calibration image. If the rounded image is a reference original image that is the basis of the calibration image, calibration is performed using a preset uniform negative value. Will be

First, when the rounded image is a reference original image for calibration, a uniform negative value is preset in the calibration image memory 38 as temporary calibration image information. Then, the negative value is subtracted from the rounded image of the reference original image (plain negative) obtained by the normal reading process, to obtain a calibration image serving as a calibration reference for the subsequent printing original image. The obtained calibration image is stored in the calibration image memory 38, and thereafter,
Used as a proofreading image for the original print.

On the other hand, if the rounded image is composed of a burned-in original image, a shading corrected image is obtained by subtracting the calibration image in the calibration image memory 38 from the rounded image. Then, the exposure control unit 9 performs predetermined image processing to calculate various parameters necessary for determining the exposure amount. In accordance with the obtained exposure amount, the operation time of the color-reduction type cut filters 13a, 13b, and 13c is adjusted above the exposure unit 3, and the operation time of the shutter 14 disposed below the exposure unit 3 is adjusted. Is done. The cut filters 13a, 13b, 13c and the shutter 14 are inserted into the exposure optical path in accordance with the operation time, and the exposure amount given to each color photosensitive layer of the photographic printing paper P is adjusted.

In the above embodiment, 2 is used as the image sensor.
Although a one-dimensional or one-dimensional CCD is preferably used, the present invention is not limited to this. When a light receiving element configured as a set of a plurality of light receiving elements such as a MOS image sensor or a general photodiode array is used, The present invention is applicable. Also, here, an example in which the image information processing apparatus of the present application is applied to a photographic printing apparatus has been described. However, the apparatus of the present application performs optical reading of image information and stably converts image information including a relatively small number of pixels. Obviously, it is effective for reading images of various devices for which it is important to keep the device cost low.

[0059]

As described above, the present invention provides an image sensor unit including a plurality of light receiving elements, a sensor driving unit that obtains an original image signal at a predetermined readout frequency from a group of light receiving elements of the sensor unit, A signal filter unit that obtains a smooth image signal by passing the image signal through a low-pass filter; a sampling unit that samples the smoothed image signal at a predetermined sampling frequency to obtain a sample image; and performs a rounding process on the sample image. A rounding unit that obtains a rounded image, and an image processing unit that performs predetermined image processing based on the rounded image.
The number of pixels Norigin of the image sensor, the number Nsample of pixels of the sample image, and the number N of pixels of the rounded image
The process is configured to satisfy the magnitude relationship of Norigin> Nsample.gtoreq.Nprocess. Therefore, it is possible to reduce the number of components and reduce the operating speed while maintaining the quality of image information.

Also, the cutoff of the low-pass filter
Frequency F_cutAnd the sampling frequency F _sampleAnd the rounding
And the average rounding rate R of the_sample ≧ 2F_cut≧ R ・ F_sample Because it is configured to satisfy the relationship of
Determines the number of sampling pixels regardless of the number of photometric elements in the disensor
Yes, and when processing while fully utilizing the effective image information
The time can be shortened.

As a result, when the image information is optically read by the high-resolution imaging means, the image information having a relatively small number of pixels can be stably obtained, and the apparatus cost can be reduced. It has an excellent effect of being able to do so.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a photographic printing apparatus to which the present apparatus is applied.

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

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

FIG. 4 is an explanatory diagram illustrating a detailed configuration of an image processing unit.

FIG. 5 is a flowchart showing calibration for a reference original image.

FIG. 6 is a flowchart showing calibration for a burned original image.

FIG. 7 is a flowchart showing calibration of a reference original image in a conventional apparatus.

FIG. 8 is a flowchart showing calibration of a printing original image in a conventional apparatus.

[Explanation of symbols]

 Reference Signs List 3 exposure unit 4 light source 5 mixing unit 6 lens 7a, 7b, 7c filter 8a, 8b, 8c photodiode 9 exposure control unit 10 imaging unit (image sensor unit) 11 image processing unit 20 two-dimensional CCD 21 color separation filter 22 signal processing Circuit 23 Drive circuit (sensor drive unit) 23c Horizontal synchronization signal 23d Vertical synchronization signal 24 Lens 34 LUT (Look-up table, sampling unit) 35 Primary image memory (Sample image storage unit) 36 CPU (Rounding unit, Calibration unit) 37 rounded image memory (rounded image storage means) 38 calibration image memory (calibrated image storage means) 100 signal filter section (low-pass filter) F photographic film P photographic printing paper

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G03B 27/80 G06T 3/40 H04N 5/253

Claims (2)

(57) [Claims] 1. An image sensor unit including a plurality of light receiving elements, a sensor driving unit that obtains an original image signal at a predetermined read frequency from a light receiving element group of the sensor unit, and a low pass filter that converts the original image signal to a low-pass filter. A signal filter unit that obtains a smoothed image signal through a sampling unit that obtains a sample image by sampling the smoothed image signal at a predetermined sampling frequency; and a rounding unit that performs a rounding process on the sample image to obtain a rounded image. And an image processing unit for performing predetermined image processing based on the rounded image, the number of pixels N of the image sensor
origin and the number of pixels Nsample of the sample image
And the number of pixels Nprocess of the rounded image satisfies the magnitude relationship of Equation 1. (Equation 1) Norigin> Nsample ≧ Nproces s ” 2. The method according to claim 1, wherein the cutoff frequency Fcut of the low-pass filter, the sampling frequency Fsample, and the average rounding ratio R of the rounding section satisfy a magnitude relationship of Equation 2. Item 2. The image information processing apparatus according to Item 1. (Equation 2) Fsample ≧ 2 · Fcut ≧ R · Fsample