JP3254749B2 - Image reading 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 processing technique for an image reading apparatus for reading a still image by photoelectric conversion.

[0002]

2. Description of the Related Art An image reading apparatus reads each pixel of a still image as an analog image signal by a photoelectric conversion means, and digitizes the analog pixel signal by an A / D conversion means, so that numerical data is stored in a storage means such as a memory. As an image.

Conventionally, an A / D converter for converting an analog image signal into numerical data outputs a numerical value proportional to an input value. The original image can be restored by outputting the image data stored as the numerical data to an output device such as a CRT or a printer.

CRTs and printers, which are output devices for restoring original images, each have characteristics of density reproducibility, and numerically converted image data obtained by A / D conversion means is directly input to the output device. In many cases, the density and the like are large and the original image is often restored in a different form. Therefore, the image data is corrected by gamma correction means for converting the image data so as to match the characteristics of each output device, and input to each output device. As a result, each output device can faithfully restore the original image.

In order to correct output distortion caused by non-uniformity of an optical system that forms an original image on a photoelectric conversion unit,
It is common practice to provide shading correction means.

[0006] By these techniques, density gradation unevenness and output distortion of the output image are removed, and as a result, the output image read by the image reading device becomes closer to the original image.

[0007]

However, in the A / D conversion by the conventional method, the CCD output voltage as a photoelectric conversion element is equally spaced (for example, 0 to 1 V, 1-2 V, 2-3 V,...).
.) Are assigned to numerical values converted into integers at equal intervals, and when gamma correction (FIG. 6A) that changes sharply in a portion where the CCD output voltage is small is performed, the corrected CC
As for the density gradation output of the image with respect to the D output voltage, the density gradation output value that can be obtained near the small CCD output voltage greatly fluctuates (FIG. 6B). There is considerable unevenness in the tone expression.

In other words, although the original density according to the conventional method and the image data after A / D conversion are in a proportional relationship, the density density is determined by gamma correction performed when the image data is output to the output device. There is a problem that density gradation unevenness occurs in a part of the tone.

As one method for solving such a problem, it is possible to use a high-resolution A / D converter.
It was inevitable that the equipment would be expensive.

The present invention has been conceived in order to solve such a problem, and an object of the present invention is to provide a novel method for solving the drawbacks of the conventional gamma correction in consideration of shading correction. A / D conversion means or equivalent pre-processing means for gamma correction is created to make the density tone output of the image correspond to the CCD output voltage more finely, eliminate the density tone unevenness, and more accurately reproduce the density tone of the original image. An object of the present invention is to provide a reading image device that outputs a high density gradation.

[0011]

In order to solve such a problem, an image reading apparatus according to the present invention comprises: a photoelectric conversion unit for reading a still image; and a digital signal converting the analog image signal read by the photoelectric conversion unit. A / D conversion means for converting the digital signal into a digital signal, shading correction means for correcting output distortion of the digital signal obtained by the A / D conversion means, and correction of the output obtained by the shading correction means. Gamma correction means, and in particular, the A / D conversion means satisfies the following relational expression.

Y = f (ax) = b · f (x) However, this excludes the case where the output value y is proportional to the input value x.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram for explaining the flow of image processing in the present invention.
3 is A / D conversion means, 4 is shading correction means, 5
Denotes gamma correction means. The photoelectric conversion unit 2 forms an image on a CCD 6 through a lens 8 and a light source 7 for irradiating the original 1 and reflected light from the original 1. In the drawing, x is a value indicating a voltage output from the CCD 6 in proportion to the reflectance of the read original 1. From this point on, each pixel data obtained by the photoelectric conversion means 2 is subjected to A / D conversion and gamma conversion. Is performed.

First, the CCD output x is input to the A / D conversion means 3, where it is converted through a function y = f (x) and y is output. Next, the value of y is input to the shading correction means 4, where the output distortion of one line of pixels is corrected, converted to Y, and input to the gamma correction means 5. z is an output from the gamma correction means 5 having a function of z = h (Y). Based on this value, the amount of YMC toner is determined based on this value, or the RGB amount is determined based on CRT. The color amount is determined.

FIG. 2 is a diagram specifically showing the A / D conversion means 3 of the present invention. Reference numeral 9 denotes a resistor network for generating a CCD output voltage obtained by dividing the reference voltage between -Vr and + Vr into 32 steps by 32 resistors having different resistance values from R1 to R32, and 10 denotes an R33. This is a resistor network for producing a CCD output voltage equally divided in 32 steps by 32 identical resistors from R to R64. 11 is the resistance network 9 or 1
This is a switch circuit for switching the CCD output voltage x from 0 by the SEL signal and inputting it to the comparator 12. SE
The resistance network is switched between 9 and 10 by the L signal,
An A / D conversion that selects different y = f (x) functions can be employed. Reference numeral 13 denotes an encoder for converting the output of the comparator 12 into a binary number, and the result is output in five bits from D0 to D4.

Now, assuming that an ideal voltage obtained from the actual density of the original image is X, the shading correction means performs correction according to the following relational expression.

G (X) = (f (X) −f (Xb)) / (f (Xw) −f (Xb)) (1) Xb: ideal output voltage based on black Xw: ideal output voltage based on white Output Voltage With this, the output voltage X is linearly normalized using the white reference data and the black reference data.

By the way, the actual CCD output voltage x has a relation of X = a · x due to the influence of the distortion of the optical system for forming the original image on the CCD as described above. Therefore, the output f (x) that satisfies the following equation is provided to the A / D conversion means.

[0019] _{f (a x · x) =} b x · f (x) (2) a x, b x: where Output distortion coefficient caused by the CCD optical system, in a manner that takes into account the output distortion (1) Is transformed into: g (ax _× x) = (f (ax _× x) −f (ax _× xb)) / (f (ax _× xw) −f (ax _× xb)) = b _x · (f (x) -f ( xb)) / b x · (f (xw) -f (xb)) = (f (x) -f (xb)) / (f (xw) -f (xb) ) = G (x) (3) xb and xw are the black reference and white reference densities of the original image itself. This means that if the CCD output x is subjected to A / D conversion through a function that satisfies the equation (2), it is not necessary for the gamma correction means 5 to consider the influence of the output distortion of each pixel. are doing. In other words, by giving the characteristic of equation (2) to the A / D conversion means, an effect is produced that the data after shading correction does not include a term of output distortion.

In the following, the resolution is set to 32 steps from 0 to 31 by setting the data bit length to 5 bits, and the A / D
The case where two function expressions (4) and (5) satisfying the expression (2) are used as the function f (x) of the conversion means 3 will be described while comparing.

Y = x (4) y = x ^{1 / 2.2} (5) Here, the output z with respect to the input x that is finally obtained is both equations (6).

Z = 1 + (1 / 2.42) logx (6) FIG. 3 shows values obtained by converting the CCD output voltage x into an integer using an A / D converter having functions of equations (4) and (5). I have. Here, since the input is set so that the value of f (x) takes a continuous integer of 0, 1, 2, 3}, for example, in equation (4), when x is 0 to 1, f (X) is 0,
f (x) is 1 when x is 1-2, and f (x) is x when 2-3 is
(X) is 2, for example.

On the other hand, in the equation (5), x is 0 to 0.01.
6, f (x) is 0, x is 0.016 to 0.074, f (x) is 1, x is 0.074 to 0.182, f (x) is 2, x Is 0.182 to 0.342, f
(X) is 3, for example.

A / D having characteristics of formula (4) or (5)
For the output Y obtained by performing shading correction on the image data y obtained by the conversion, the function h (Y) of gamma correction performed to obtain the output result z of Expression (6) is expressed by Expressions (7) and (8).

Z = 1 + (1 / 2.42) logY (7) z = 1 + (2.2 / 2.42) logY (8) FIG. 4 shows possible values of f (x) from 0 to 31 shown in FIG.
FIG. 11 is a diagram in which possible integer values of Expressions (7) and (8) are plotted with respect to Y obtained by shading-correcting the integer value y of FIG. From this figure, in the gamma correction according to the conventional equation (7), possible values are only 16 levels such as 0, 11, 15, 18,.... On the other hand, in the gamma correction by the equation (8), 0, 2, 6, 8, 1
It is possible to take 23 levels such as 0,.

The final output z with respect to the CCD output voltage x finally obtained is obtained by combining FIGS. 3 and 4, and can be expressed by the following formula. ) And the method according to the present invention (combination of equations (5) and (8)), the final output z for input x can be expressed by equation (6). However, in the present invention, the number of possible levels of the final output z can be greater than in the conventional method.

FIG. 5 shows this conventional method (FIG. 5 (a)).
FIG. 6 is a diagram visually illustrating a difference between a possible level of an output value by the method according to the present invention (FIG. 5B). In (a), the point of the final output z that is rounded to an integer value with respect to the value of the CCD output x that can be actually taken is x =
16 points such as 0, 1, 2, 3, 4, 5,..., X = 0, 0.18, 0.34, 0.5 in FIG.
23, such as 6, 0.83. As is clear from this graph, in the portion where the value of the CCD output voltage x is small, the value suddenly becomes 11 next to 0 in the conventional example, whereas it is 0, 2, 6, 8, 10, 12 in the present invention. The resolution can be increased in such a manner. Moreover, there are 23 punctuation points of x in comparison with the 16 points in the conventional example, and a portion having a large value of x is almost equal to the resolution of the conventional example.

These can be realized by forming the resistor network in the A / D converter with detection resistors having different resistance values (FIG. 2). That is, assuming a function to obtain a final output z which is finally desired with respect to the CCD output voltage x, and determining a detection resistor network of the A / D converter to be used in accordance with the function, the density gradation unevenness is reduced. A suppressed high-quality image reading device can be realized.

Further, if a plurality of types of resistance networks in FIG. 2 are prepared and a plurality of types of resistance networks of the A / D converter to be used according to a function to be finally obtained are used, the variation of correction can be increased.

In the embodiment of the present invention, a method of providing the A / D conversion means with a correction function has been described.
It is clear that the same effect can be obtained by performing the same function using an amplifier having a function represented by equation (2) before the D conversion means.

[0031]

As described above, according to the present invention, considering the shading correction for correcting the output distortion of the optical system of the image reading apparatus, y = f (ax) = b.f
By providing the function (x) in the A / D converter, the output distortion term is removed from the output after shading correction. Further, since the detection resistor network of the A / D converter is constituted by resistors having different resistance values, an expensive high-resolution A
It is possible to increase the resolution of the digital output after gamma correction for the CCD output voltage without using a / D converter. As a result, the density gradation unevenness of the obtained image can be significantly reduced as compared with the conventional method.

[Brief description of the drawings]

FIG. 1 is a block diagram of image processing according to the present invention.

FIG. 2 is a diagram showing a circuit configuration of an A / D converter according to an embodiment of the present invention.

FIG. 3 is a diagram showing the input / output relationship of an A / D converter.

FIG. 4 is a diagram showing the relationship between input and output of a gamma converter.

FIG. 5 is a diagram illustrating a relationship between a final output z and a CCD voltage x.

FIG. 6 is a diagram illustrating the principle of conventional gamma correction.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Document 2 Photoelectric conversion part 3 A / D conversion means 4 Shading correction means 5 Gamma correction means 6 CCD 7 Light source 8 Lens 9 Resistance network 10 Resistance network 11 Switch 12 Comparator 13 Encoder

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 1/40-1/409 H03M 1/00-1/88

Claims (2)

(57) [Claims] 1. A photoelectric conversion unit for reading a still image, an A / D conversion unit for converting an analog image signal read by the photoelectric conversion unit into a digital signal, and a signal obtained by the A / D conversion unit. An image reading apparatus having shading correction means for correcting output distortion of a digital signal and gamma correction means for correcting an output obtained by the shading correction means, wherein the A / D conversion means is as follows: An image reading apparatus satisfying the following relational expression: y = f (ax) = b · f (x) However, this excludes the case where the output value y is proportional to the input value x. 2. The A / D conversion means according to claim 1, wherein
2. The apparatus according to claim 1, comprising a plurality of functions satisfying the relational expression described above, and selecting one of the plurality of relational expressions of the A / D conversion according to the gamma correction unit used. Image reading device.