JP2002247398A - Device and method for processing image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prepare a gamma correction table from which an excellent gradation property can be obtained in order to conquer such a problem that the table becomes a gamma table which makes a whole image brighter by making the whole image darker when the image is too bright or brighter when the image is too dark and is not able to excellently reproduce gradations due to too bright shadow sections. SOLUTION: A histogram is prepared from image data (S401) and white and black points and the mean value of the image data are acquired from the histogram (S402). Then a gamma value is calculated by selectively using the calculating method corresponding to the positional relation between the acquired white and black points and the mean value of the image data and the gamma correction table is prepared based on the calculated gamma value and white and black points (S403).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method, and more particularly to an image processing apparatus and method for creating a correction table for correcting an image read from a document.

[0002]

2. Description of the Related Art A film scanner is a typical example of a device for inputting a transparent original image. Film scanners can scan both positive and negative films. Since a positive film has a narrow exposure area, a photographed image is often biased toward a highlight portion or a shadow portion. The film scanner has a function of automatically correcting gradation in preparation for scanning such an image.

FIG. 1 is a flowchart showing automatic gradation correction processing in a film scanner.

In step S101, a target transparent original image is prescanned. The pre-scanned image is used for analyzing image characteristics and displaying the scanned image on the screen.
In pre-scan, to reduce the processing load,
Scanned at a lower resolution than the actual scan.

In step S102, a histogram of each RGB color signal and luminance signal is created based on the pixel data of the prescan image. For example, the luminance signal Y is created from the RGB signal by Expression (1). Y = 0.30R + 0.59G + 0.11B… (1)

When each of the RGB signals is 8 bits (256 gradations),
The luminance signal Y is also an 8-bit signal. Therefore, the histogram counts the frequency of each of the 256 signal levels from 0 to 255.

In step S103, a white point and a black point are determined. Highlight point for RGB signal and luminance signal Y (255 level side in this case)
And a signal level at which the cumulative value of the frequency from the shadow point (in this case, the 0 level side) becomes a predetermined value (shown as a ratio to the total number of pixels), and determines the signal level as a white point and a black point. . The optimal value of the ratio for setting the white point and the black point cannot be unconditionally determined because it depends on the image input device, but is generally set to about 0.5% in many cases.

In step S104, a gamma value is determined from the histogram of the luminance signal Y, and a gamma table is created in consideration of a white point and a black point. The gamma value is calculated so that the average value of the luminance signal Y becomes 50%. If the output signal is 8 bits for each color, 50% is "128"
It is. Specifically, the gamma value γ is calculated by equation (2). _{γ = log {(Y AV -} Y LO) / (Y HI - Y LO)} / log (128/255) ... (2) where, Y _HI: white point of the luminance signal Y _LO: Black Point luminance signal Y _AV : Average value of luminance signal

To create a gamma table, a one-dimensional lookup table (LUT) for each of RGB used for pre-scanning is used.
Convert using white point, black point and gamma value. First, the output value of the prescan LUT is
Input values RGB (XR _HI prescan LUT comprising a white point and black _{point, XG HI, XB HI, XR} LO, XG LO
And XB _LO ).

[0010] Next, the output value from the highlight point of the input prescan LUT to XR _HI, XG _HI and XB _{H I} 25
At 5, the output values from the shadow point to XR _LO , XG _LO and XB _LO are fixed to 0. By this processing, the input values XR _{LO to} XR _HI , XG _{LO to} XG _HI and XB _{LO to} XB of the prescan LUT
_The output value corresponding to _HI will be in the range from the black point to the white point. Then, the output value of the prescan LUT existing in the above range is expanded to the range of 0 to 255, and the output value of the prescan LUT is multiplied by the gamma value γ to create a gamma table.

In step S105, a positive film or the like is scanned (called "main scan") using the gamma table created in step S104.

[0012]

There is room for improvement in the creation of the gamma table for automatic correction described above.

Since the gamma value γ is calculated by a formula for setting the average value of the luminance signal Y to 50%, the gamma value γ increases or decreases as the average luminance deviates from the center. That is,
In the case of an image that is too bright, the entire image is darkened, and in the case of an image that is too dark, the function of brightening the entire image is enhanced. As a result, when an image having a low average luminance such as a night view is scanned, a gamma table for brightening the entire image is obtained, so that the shadow portion becomes too bright and gradation cannot be reproduced well.

In preparing a gamma table, a normal gamma curve is calculated by equation (3). Y = 255 × (y / 255) ^{1 / γ} … (3) where output data before and after 8-bit conversion are y and Y, respectively.

FIG. 2 is a diagram showing an example of a gamma table. As shown in FIG. 2A, when the gamma value γ> 1.0, in a region where the input luminance is low where the gamma curve is steep, the gradation becomes rougher as the gamma value γ increases. On the other hand, as shown in FIG. 2A, when the gamma value is γ <1.0, in a region where the input luminance is low where the gamma curve is gentle, the gradation becomes smaller as the gamma value becomes smaller.

As another method, when gamma value γ> 1.0, a gamma curve as shown in FIG. 2C may be used in order to prevent gradation from being roughened in a region where input luminance is low. The gamma curve shown in FIG. 2 (c) is calculated by equation (4). Y = 255 × {1-(1-y / 255) ^γ }… (4)

However, even if a gamma curve as shown in FIG. 2 (c) is used, there remains a problem that the gradation of the highlight portion is destroyed when the gamma value γ increases.

An object of the present invention is to solve the above-mentioned problem, and an object of the present invention is to create a gamma correction table capable of obtaining good gradation.

[0019]

The present invention has the following configuration as one means for achieving the above object.

An image processing apparatus according to the present invention comprises a creating means for creating a histogram from image data; a white and black point from the histogram;
Obtaining means for obtaining an average value of image data; calculating means for calculating a gamma value by selectively using a calculation method according to a positional relationship between the obtained white and black points and the average value of the image data A gamma correction table based on the calculated gamma value and the white and black points.

According to the image processing method of the present invention, a histogram is created from image data, and white and black points and an average value of the image data are acquired from the histogram. Calculating a gamma value by selectively using a calculation method according to the positional relationship with the average value of the data, and creating a gamma correction table based on the calculated gamma value and the white and black points. Features.

Preferably, the calculation of the gamma value includes increasing or decreasing the gamma value from the center of the range of the white and black points as the average value of the image data deviates, and decreasing or increasing the gamma value further from the average value. Features.

Further, preferably, the gamma correction table is created by selectively using a creation method according to the gamma value.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image processing apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Configuration] FIG. 3 is a block diagram showing a configuration example of an image processing apparatus according to the embodiment.

The host computer 300 is connected to a film scanner 301 and a monitor 302. In the host computer 300, an application 303 capable of performing scanner input and a scanner driver 304 are installed as software.

A film scanner 301 capable of scanning a positive film has an RGB
Load each one-dimensional LUT and perform low-resolution pre-scan and high-resolution main scan.

[Scanning Procedure] Next, a series of procedures for scanning a transparent original image will be described.

The user sets a transparent original on the film scanner 301 and requests the scanner driver 304 via the application 303 to scan the transparent original image. Upon receiving the request, the scanner driver 304
Requests 01 to load the default prescan LUT and perform prescan. Film scanner 30
Step 1 uses the loaded prescan LUT to prescan the transparent original image.

Low resolution RG obtained by prescan
The B image is sent to the scanner driver 304 and processed by the image processing function of the scanner driver 304 to create a main scan LUT for a transparent original image. Note that the main scan LUT is created for each transparent document image.
Further, the scanner driver 304 simulates a case where a transparent original image is scanned by the created main scan LUT using the pre-scan image, and causes the monitor 302 to display an image of the simulation result via the application 303.

The user confirms the image displayed on the monitor 302 and requests the scanner driver 304 via the application 303 to scan the transparent original image again (main scan). Scanner driver 30 requested
4 issues a request to the film scanner 301 to load the created main scan LUT and perform the main scan. The film scanner 301 uses the loaded main scan LUT to perform a main scan on the transparent original image.

The high-resolution image obtained by the main scan is transferred to the application 303 via the scanner driver 304.

[Image processing function] FIG.
14 is a flowchart for describing an image processing function of 04.

For the pixel data of the prescan image,
A histogram is created for the RGB signal and the luminance signal Y (S401). Note that the luminance signal Y is obtained by the above equation (1). At this time, noises due to external factors may be included at the four ends of the upper, lower, left, and right sides of the scan image. If a histogram is created from the entire prescan image, accurate image characteristics may not be obtained. Therefore, an accurate image characteristic is obtained by creating a histogram from a central portion of the pre-scan image, for example, a 7/8 × 7/8 area.

Next, white and black points are determined from the histogram (S402). As mentioned above, RG
A signal level at which the cumulative value of the frequency from the highlight and shadow points of the B signal and the luminance signal Y becomes a predetermined value is determined, and this signal level is determined as a white and black point.

From the histogram of the luminance signal Y, the gamma value γ
Is determined and a gamma table (main scan LUT) is created using the white and black points (S40
3). Here, details of the method of determining the gamma value γ and the method of creating the gamma table will be described later.

The prescan image is subjected to conversion processing for monitor display to simulate an image obtained by the main scan (S404). Main scan LUT is pre-scan LUT
In this step, the input and output of the same bit information as the pre-scan image is based on a linear LUT,
A gamma table is created, and the pre-scan image is subjected to gamma conversion using the created gamma table to obtain a simulation image.

[Preparation of gamma table]
The creation of the gamma table in S403 will be described in detail.

First, a method for determining the gamma value γ will be described. The gamma value γ is calculated using the average value of the luminance signal Y and the white and black points.

Hereinafter, the position of the average value of the luminance signal Y with respect to the range of the luminance signal Y (the range from the white point to the black point) will be described as Yp. The average value of the luminance signal and the white and black points respectively
Y _AV, Y _HI, When Y _LO Yp is expressed by Equation (5). Yp = (Y _AV -Y _LO ) / (Y _HI -Y _LO )… (5)

The gamma value γ is basically calculated by the following equation. In Expression (2), the target of Expression (6) is 128/255 ≒
0.5. γ = log (Yp) / log (Target)… (6)

A specific method of calculating the gamma value will be described below.

(1) When 0.35 ≦ Yp ≦ 0.65: Yp is Target =
Set the gamma value γ to be 0.5. That is, the gamma value γ is such that the average value of the luminance is 128.

(2) When Yp> 0.65: Yp is Target = 0.5 to 0.
Set the gamma value γ to be 65. That is, the gamma value γ is such that the average value of the luminance is 128 to 169. Targ
The value of et changes stepwise according to the value of Yp.

(3) If Yp <0.35: Yp is Target = 0.35-
Set the gamma value γ to be 0.65. That is, the gamma value γ is such that the average value of the luminance is 89 to 128. Tar
The value of get changes stepwise according to the value of Yp.

(4) In the case of Yp> 0.75: In the above item 2, the gamma value γ decreases as the value of Yp departs from 0.5. Therefore, the boundary (vertex) is set to Yp = 0.75, and the gamma value γ is set to approach 1.0 in the case of Yp = 0.75.

(5) When Yp <0.25: In the above item 3, the gamma value γ increases as the value of Yp departs from 0.5. Therefore, the boundary (vertex) is set at Yp = 0.25, and if it is less than that, the gamma value γ is set so as to approach 1.0.

FIG. 5 is a diagram in which Yp is plotted on the horizontal axis and gamma value γ is plotted on the vertical axis. In the above description and FIG. 5, specific numerical values are used for ease of description, but these numerical values depend on the device and change.

Next, a method for creating a gamma table using white and black points and a gamma value γ will be described.

(6) RGB which is the input value of the pre-scan LUT whose output value becomes the white and black points
Value _{(XR HI, XG HI, XB} HI, XR LO, XG LO and XB _LO) Request.

(7) From the highlight point, XR _HI and XG _HI
And the output value of the pre-scanning LUT to XB _HI fixed to 255, to fix the shadow point XR _LO, the output value of the LUT for pre-scan until XG _LO and XB _LO 0.

(8) By the processing of the above item 7, the input values of the pre-scan LUT are XR _LO to XR _HI , XG _{LO to} XG _HI and XB
The output values of _{LO to} XB _HI are in the range from the black point to the white point. Then, the output value of the prescan LUT existing in this range is expanded to the range of 0 to 255.

(9) All output values of the pre-scanning LUT after the processing of the above item 8 are gamma-converted by both equations (3) and (4).

(10) One of the output values of the pre-scanning LUT calculated for each two sets in the above item 9 is selected based on the gamma value γ to complete the gamma table. If γ ≧ 1.0, use the calculated small value If γ <1.0, use the calculated large value

FIG. 6A shows an example of a gamma table created when γ ≧ 1.0, and FIG. 6A shows an example of a gamma table created when γ <1.0.

In the above-described embodiment, an example has been described in which, when white and black points are obtained, a predetermined cumulative frequency is determined, and signal levels that match (or exceed) the frequency are set as white and black points. However, the cumulative frequency for determining the white and black points may be varied according to the characteristics of the histogram.

For example, the range of the histogram (the range from the white point to the black point) is narrow and
If the average brightness is higher than a certain value (bright image)
If the value is lower than a certain value (dark image), data in a narrow range is expanded and gamma converted, so that noise in highlights and shadows is expanded. Therefore, in the case of an image having the above characteristics, if the noise component is cut by increasing the cumulative frequency for determining the white and black points, the tone of the image can be reproduced well.

In the above embodiment, the white and black points are determined for each of the RGB signals.
The example in which the gamma value γ is determined by the luminance signal Y has been described. For example, the white and black points may be determined by the luminance signal Y, and the gamma value γ may be determined for each of the RGB signals.

In the above-described embodiment, an example has been described in which the present invention is applied to a film scanner that reads a transparent original such as a positive film. However, the application of the present invention is not limited to positive films and film scanners.
In other words, the present invention can be applied to a device for reading a negative film or a non-transparent document as long as the device performs pre-scan of a document image and performs automatic gradation correction.

As described above, according to the present embodiment, it is possible to determine the gamma value γ by analyzing the image characteristics of each document image and create a gamma table from which an image with rich gradation can be obtained. it can. Therefore, the gradation of the original image can be reproduced well.

[0061]

[Other Embodiments] Even if the present invention is applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), an apparatus (for example, a copying machine) Machine, facsimile machine, etc.).

Further, an object of the present invention is to supply a storage medium (or a recording medium) in which a program code of software for realizing the functions of the above-described embodiments is recorded to a system or an apparatus, and a computer (a computer) of the system or the apparatus. Alternatively, it is needless to say that the present invention can also be achieved by a CPU or an MPU) reading and executing the program code stored in the storage medium. In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention. Also,
When the computer executes the readout program code, not only the functions of the above-described embodiments are realized, but also the operating system (OS) running on the computer based on the instructions of the program code.
It goes without saying that a case where the functions of the above-described embodiments are implemented by performing some or all of the actual processing, and the processing performs the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into the memory provided in the function expansion card inserted into the computer or the function expansion unit connected to the computer, the program code is read based on the instruction of the program code. Needless to say, the CPU included in the function expansion card or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

When the present invention is applied to the storage medium, the storage medium stores program codes corresponding to the flowcharts described above.

[0065]

As described above, according to the present invention,
It is possible to create a gamma correction table from which good gradation can be obtained.

[Brief description of the drawings]

FIG. 1 is a flowchart showing an automatic key correction process in a film scanner.

FIG. 2 is a diagram showing an example of a gamma table;

FIG. 3 is a block diagram illustrating a configuration example of an image processing apparatus according to the embodiment;

FIG. 4 is a flowchart for explaining an image processing function of the scanner driver;

FIG. 5 is a diagram for explaining a method of determining a gamma value;

FIG. 6 is a diagram illustrating an example of a created gamma table.

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B057 BA11 CA01 CA08 CA12 CB01 CB08 CB12 CE11 DB02 DB06 DB09 DC23 DC36 5B069 AA01 BB07 HA11 5C077 LL09 LL19 MP08 NN02 NN03 PP09 PP44 PP45 PP46 PQ19 PQ23 RR06 MA12 LA PA08

Claims (10)

[Claims] A generating means for generating a histogram from image data; an obtaining means for obtaining a white and a black point from the histogram and an average value of the image data; a white and a black point to be obtained; Calculating means for calculating a gamma value by selectively using a calculation method according to the positional relationship with the average value of the image data; and generating a gamma correction table based on the calculated gamma value and the white and black points. An image processing apparatus comprising a creating unit. 2. The calculation of a gamma value by the calculation means,
2. The image processing apparatus according to claim 1, wherein the gamma value increases or decreases as the average value of the image data deviates from the center of the range of the white and black points, and decreases or increases as the average value further deviates. . 3. The image according to claim 1, wherein the creation unit creates the gamma correction table by selectively using a creation method according to the gamma value. Processing equipment. 4. The creation means clips a predetermined gamma correction table at the white and black points, and converts all output values of the clipped table according to the following two formulas, so that the gamma value γ is The method according to claim 1, wherein a small value obtained by the conversion is selected when γ ≧ 1, and a large value obtained by the conversion is selected when γ <1, and the gamma correction table is created based on the selected value. 4. The image processing device according to 3. Y1 = (2k-1) {y / (2k-1)} ^{1 / γ} Y2 = (2k-1) [1-{1-y / (2k-1)} ^γ ] where y is a table input Value, Y is the output value of the table, k is the number of quantization bits of the image data 5. The image processing apparatus according to claim 1, wherein the obtaining unit changes a condition for obtaining the white and black points according to an image characteristic indicated by the histogram. The described image processing device. 6. A histogram is created from the image data, a white and black point and an average value of the image data are obtained from the histogram, and the positions of the obtained white and black points and the average value of the image data are obtained. An image processing method, wherein a gamma value is calculated by selectively using a calculation method according to a relationship, and a gamma correction table is created based on the calculated gamma value and the white and black points. 7. The calculation of the gamma value, wherein the gamma value is increased or decreased as the average value of the image data deviates from the center of the range of the white and black points, and decreased or increased as the average deviates. 7. The image processing method according to claim 6, wherein: 8. The image processing method according to claim 6, wherein the gamma correction table is created by selectively using a creation method according to the gamma value. 9. A program for controlling an image processing device to execute the image processing according to claim 6. Description: 10. A recording medium on which the program according to claim 7 is recorded.