JPH0652931B2 - Gradation conversion curve generator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention automatically generates a gradation conversion curve used in the field of image processing (for example, a plate-making scanner) according to the density range of an original image. Regarding the device.

[Conventional technology]

As is well known, a plate-making scanner or the like is provided with a gradation conversion device for performing gradation conversion of image data obtained by reading an original image having gradation. The gradation conversion device is provided with a gradation conversion curve determined according to the density distribution state of the original image, plate making application, and the like.

However, in general, it is not easy to accurately set the characteristics of the density distribution of the original image to set an appropriate gradation conversion curve, and conventionally, the setting of the gradation conversion curve relies largely on the experience of the operator.

[Problems to be Solved by the Invention]

In order to deal with such a situation, it is desired to automate the setting of the gradation conversion curve. However, it is difficult to effectively achieve such automation unless a technology that expresses the characteristics of the original image with simple parameters and automatically sets the gradation conversion curve that matches the empirical rule according to the parameters is developed. Is.

[Object of the Invention]

The present invention has been made in view of the above problems, and an object of the present invention is to provide an apparatus for automatically generating a gradation conversion curve that also conforms to the empirical rule depending on the density range of an original image.

[Means for Solving the Problems]

With reference to FIG. 1, a tone conversion curve generating device according to a first configuration of the present invention inputs data indicating a density range of an original image, and bends over the entire density range according to the density range of the original image. A model curve generating means for generating a model curve whose state is determined, and a predetermined highlight point and a shadow point are passed on the gradation conversion coordinate plane showing the relationship between the original image density and the copy halftone dot% or the copy density value. As described above, there is provided a correction means for correcting the model curve to generate a gradation conversion curve for the original image.

Then, the model curve generating means (a) storage means for storing first and second reference gradation conversion curves which are prepared in advance and have different curved states, and (b) in the density range of the original image. Accordingly, a combination ratio characteristic means for specifying a value of a combination ratio of the first and second reference gradation conversion curves, and (c) combining the first and second reference gradation conversion curves with the combination ratio. And a synthesizing means for generating the model curve thereby, and by the synthesizing, the curved state of the model curve includes a state group including at least an upward convex state and a downward convex state. It is selected from the inside according to the density range of the original image.

Further, the tone conversion curve generating device having the second configuration further includes a density range detecting means for obtaining the density range based on the density histogram of the original image and giving data indicating the density range to the model curve generating means. I have it.

It should be noted that the term "density" in the present invention is a general term for not only the optical density but also a quantity representing the optical density such as a signal level obtained by photoelectrically reading the optical density or a Munsell value.

Further, in the present invention, the term “curve” is used in a broad concept including straight lines and the like. Therefore, "curved state"
Can also include a “non-curved state”. Further, the "density area" in the present invention is a generic term including not only the difference between the maximum occurrence density and the minimum occurrence density of the original image but also the difference between a predetermined shadow point density and a highlight point density.

[Action]

In the first configuration of the present invention, the curved state of the model curve is determined according to the density range of the original image, and the gradation conversion curve can be obtained by correcting it. Therefore, a gradation conversion curve that matches the empirical rule described later can be obtained. Can be generated automatically.

Further, the model curve (hence, the gradation conversion curve) is determined by combining the two types of reference gradation conversion curves with the combination ratio determined by the density range.

Further, in the second configuration, by further including the density range detecting means for obtaining the density range based on the density histogram of the original image, the density range is also automatically and accurately detected.

〔Example〕

A. Overall Structure FIG. 2 is a schematic block diagram of a plate-making scanner having a gradation converting apparatus according to an embodiment of the present invention. This is based on FIG. 1 of Japanese Patent Laid-Open No. 63-42575, but its internal processing is peculiar to the present invention as will be described later. In this scanner, the image of the original image 1 having gradation is read by the scanning reading device 2 pixel by pixel, and the image data obtained thereby becomes digital image data in the A / D converter 3. The digital image data is subjected to shading correction in the shading correction circuit 4 and then given to the selector 5.

In the setup process described later, the selector 5 selects the histogram counting circuit 6 side, but selects the gradation conversion circuit 7 side in actual operation. The gradation conversion circuit 7 has a RAM 8 and performs gradation conversion on the input image data based on the gradation conversion table in the RAM 8. The image data after the gradation conversion is subjected to processing such as unsharp masking and magnification conversion in the image processing circuit 9, and then given to the halftone dot signal generation circuit 10.

Then, the halftone dot signal generation circuit 10 converts the input image data into a halftone dot signal and outputs it to the scanning exposure device 11. The scanning exposure device 11 is based on this halftone dot signal,
A halftone image is exposed and recorded on the photosensitive film 12.

Each of these circuits operates under the control of the computer 13. The computer 13 has a CPU 14 and a memory 15, and also performs various data processing described later. Transmission and reception of signals (data) between each circuit and the computer 13 are performed through the signal line group 16. Also, the keyboard 17
Is for performing a command operation to the computer 13.

B. Reference gradation conversion curve reference step FIG. 3 is a flow chart showing the gradation conversion curve generating operation according to this embodiment, and FIG. 4 is a conceptual diagram showing its contents schematically. The symbols (a) to (f) displayed on the right side of each block in FIG. 3 correspond to those in FIG. 4 (a) to (f), respectively.

First, in step S1, two types of reference gradation conversion curves F ₁ (D) and F ₁ having different curved states in a single direction are provided.
₂ (D) are prepared, and data expressing them are stored in the memory 1
Store in 5 (D is concentration variable). In this embodiment, an upwardly convex curve (FIG. 4 (a)) obtained empirically is used as F ₁ (D), and a straight line (FIG. 4 (b)) is used as F ₂ (D). ) Is used. However, Q indicates the dot area ratio (dot%), and D _max indicates the maximum density of appearance density of the original image that is assumed in advance.

C. Synthesis Ratio Function Preparation Step In the next step S2, the shape of the synthesis ratio f (ΔD) shown in FIG. 4 (d) is determined and shown as preparation for synthesizing these two reference gradation conversion functions. The data is also stored in the memory 15. However, [Delta] D is the difference between the shadow point density value D _S and a highlight point density value D _H: ΔD≡D S _{-D H} ... is defined as _(1), a parameter indicating the size of the density range of the original ing.

As can be seen from FIG. 4 (d), this composite ratio f (ΔD)
The value of is the constant value ΔD ₁ , ΔD ₂ (ΔD
₁ <ΔD ₂ ) and ΔD, the constant (maximum value) f _max in the first section: 0 ≦ ΔD <ΔD ₁ (2) and the second section: ΔD ₁ ≦ ΔD <ΔD ₂ ((2)) In 3), a value on a straight line having a negative slope is taken, and in the third section: ΔD ₂ ≦ ΔD (4), it is set to have a constant value of “0”. However, in this embodiment, f _max = 2.0 (5)

The reason why such a function form is adopted as the composite ratio function f (ΔD) will be described later. It should be noted that ΔD _{C in} FIG. 4 (d) indicates a critical agriculture range value such that the value of the function f (ΔD) becomes 1.0.

D. Density value D _H , D _S determination step In the next step S3, the original image 1 is pre-scanned to obtain a cumulative density histogram h in the reading target area of the original image 1.
Find (D) (Fig. 4 (c)). This is done using the histogram counting circuit 6 of FIG.

Then, the values Y ₁ , Y of the cumulative appearance rate Y, which are empirically considered to have a high probability of becoming a highlight point and a shadow point,
₂ is specified, and in the cumulative density histogram h (D),
A pair of density values corresponding to these values Y ₁ and Y ₂ are respectively defined as a highlight point density value D _H and a shadow point density value D _S. In addition, in FIG. 4 (c),
D _min and D _max are the minimum generated density and the maximum generated density of the original image 1, respectively.

E. Synthesis ratio specifying step and synthesis step the thus obtained density values _D H, based on the _{D S,} the concentration range: ΔD _{= D} S -D calculates H _... (6). The composition ratio f (ΔD) of FIG. 4 (d) is stored by the look-up table method, and the density range Δ
By inputting the value of D, the materialized composite ratio f
(ΔD) is obtained (step S4).

Then, using the synthesis ratio f (ΔD), the reference gradation conversion curves F ₁ (D) and F ₂ (D) are synthesized according to the following equation (7), and thereby the model curve K (D) is obtained. (Step S5).

K (D) = {1-f (ΔD)} F ₁ (D) + f (ΔD) F ₂ (D) (7) This model curve K (D) has the following properties. First, when the difference ΔD, that is, the density range of the original image 1 belongs to the narrow density range: 0 ≦ ΔD <ΔD _c (8), f (ΔD)> according to FIG. 4 (d).
It is 1.0, and the coefficient of F ₁ (D) becomes negative according to the equation (7). Therefore, from FIG. 4 (a), (b), the model curve K
The curved state of (D) is convex downward (K _{3 in} Fig. 4 (e)
(D)).

On the other hand, in a wide concentration range: ΔD _c <ΔD (9), f (ΔD) <1.0 is obtained from FIG. 4 (d), and F ₁ (D) and F ₂ (D) on the right side of the equation (7) are obtained. The coefficients of and are both positive. Therefore, the model curve K (D) becomes convex (K ₁ (D) in FIG. 4 (e)).

Further, in the critical concentration range between them: ΔD = ΔD _c (10), f (ΔD) = 1.0, and from the equation (7), K
(D) becomes a straight line (K ₂ (D) in FIG. 4 (e)).

As described above, by synthesizing in the above-described manner, the curved state of the model curve K (D) is changed according to the value of the concentration range ΔD.
One of them is selected from the three types of upward convex, downward convex, and straight line. As can be seen from FIG. 4 (d), the absolute value of the curvature of the curve K (D) also changes within the range of ΔD ₁ <ΔD <ΔD ₂ (11).

F. Generation of the tone conversion curve then the thus obtained model curve K (D) is, on the gradation conversion coordinate plane (D / Q plane), the designated highlight point: P _{H = (D} H , Q _H ) ... (12) and the shadow point: P _S = (D _S , Q _S ) ... (13) so that the tone conversion curve G is corrected.
(D) is obtained (step S6, FIG. 4 (f)). However,
Q _S and Q _H are highlight dot net% and shadow dot net% found empirically, and are stored in the memory 15 beforehand.
It is stored in. Specifically, this fix is in (14)
~ It is executed by using the equation (17).

G (D) = {ΔQ · K (D) + Δ (QK)} / ΔK ... (14) ΔQ≡Q S -Q H ... (15) Δ (QK) ≡Q H K (D S) -Q S K _{(D H) ... (16)} ΔK≡K (D S) -K (D H) ... (17) Figure 4 in (f) is obtained from the model curve K (D) of FIG. 4 (e) The gradation conversion curve G (D) in the case of positive plate making is illustrated. The gradation conversion curve G obtained in this way
(D) is stored in the RAM 8 in a look-up table format, thereby completing the setup of the gradation conversion circuit 7.

Since the actual plate making process can be understood from the above description, the detailed description thereof will be omitted.

G. Features and Modifications of the Embodiment Since the gradation conversion curve G (D) is obtained by scaling and shifting the model curve K (D), as can be seen from equations (14) to (17), The curved state of the model curve K (D) is substantially reflected in the curved state of the gradation conversion curve G (D). Therefore, in the case of positive output, the gradation conversion curve G (D) is convex downward for the original image in the narrow density range and is convex upward for the wide density range.

This property is consistent with the rule of thumb. That is, the gradation conversion curve for positive output, which is created empirically by a skilled operator, is a downwardly convex curve G _3E (D) for an original image in a narrow density range, as shown in FIG. The curve G _1E (D) is upwardly convex in a wide concentration range, and is a linear curve G in the middle.
_2E (D) has been found by the inventor's analysis. Therefore, the above-described embodiment is a technique that realizes automation of gradation conversion curve generation while satisfying the empirical rule.

By the way, in the apparatus of this embodiment, it should be noted that the curved state of the model curve K (D) over the entire density range is determined according to the density range of the original image. That is, since the composition ratio f (ΔD) is a constant that does not depend on the concentration value D, the composition ratio is common to the entire range of the concentration value D, and the curved state of the model curve K (D) over the entire concentration range is It depends on the density range of the original image.

If only the local (partial) curved state of the model curve K (D) is changed according to the density range of the original image, it is possible to obtain a result satisfying the above empirical rule. Although not possible, the above empirical rule can be satisfied only by applying a curved state over the entire concentration range as in the present invention.

As described above, the gradation conversion curve G (D) is the model curve K
Obtained through scaling and shift of (D), and if necessary, processing is performed so as to saturate Q = 0% and 100% at finite concentration values D _H and D _S as shown in FIG. 4 (f). You can also Therefore, regarding the gradation conversion curve G (D), the curved state reflects the density range of the original image only in the range of D _H ≦ D ≦ D _S , and the curved state of “total density range” reflects the density range of the original image. However, this is not against the spirit of the present invention. That is, in the present invention, the gradation conversion curve G
In the model curve K (D) as a basis for generating (D), the curved state over the entire density range may be determined by the density range of the original image, and the final gradation conversion curve G (D) is partially deformed. It is not the essence even if it is given.

Needless to say, the present invention can be applied to the case of negative plate making. The gradation conversion curve G (D) in the case of this negative output is convex upward for an original image in a narrow density range, and conversely is convex downward for an original image in a wide density range. Further, in this invention, two types of reference gradation conversion curves F
₁ (D), F ₂ (D) are synthesized, and thereby a curved state over the entire concentration range of the model curve is selected from a group of states including at least an upward convex state and a downward convex state. The reference gradation conversion curves F ₁ (D) and F ₂ (D) may have different curved states, for example, F ₂ (D).
May be a downwardly convex curve. Munsell curves can also be used. The composition does not have to be the linear composition as shown in equation (7).

Further, in the above embodiment, the difference (D _S −D) between the highlight point density value _DH and the shadow point density value D _S is set as the density range ΔD.
_H ) was used, but instead of this, FIG.
Maximum density D _max and minimum density D _min in (c)
The difference from (D _max −D _min ) may be used.

〔The invention's effect〕

As described above, according to the first configuration of the present invention,
Since the curved state of the gradation conversion curve can be automatically determined according to the density range of the original image, it is possible to automatically generate the gradation conversion curve that matches the empirical rule while reflecting the characteristics of the original image.

Further, since the first configuration is realized by changing the composition ratio of the two types of reference gradation conversion curves according to the density range, the above effect can be obtained with a relatively simple configuration.

Further, in the second configuration, by further including a density range detection unit that obtains a density range based on the density histogram of the original image, the density range is also automatically and accurately detected.
Not only does it require a skilled operator, but it is also not affected by individual differences among operators.

[Brief description of drawings]

FIG. 1 is a block diagram showing the concept of the configuration of the present invention, FIG. 2 is a schematic block diagram of a plate-making scanner incorporating this embodiment, and FIG. 3 is a flow chart showing the operation of the embodiment. FIG. 5 is a conceptual diagram schematically showing the procedure of the embodiment, and FIG. 5 is an explanatory diagram of an empirical rule for creating a gradation conversion curve. 1 ...... original, 7 ...... gradation conversion circuit, _{D H} ...... highlight point density value, _{D S} ...... shadow point density value, _F 1 (D) ...... first reference gradation conversion curve, _{F 2} (D) -second reference gradation conversion curve, K (D) -model curve, G (D) -gradation conversion curve, f (ΔD) -synthesis ratio, D-density value, Q ...... Mesh%, ΔD ...... Concentration range

Front page continuation (72) Inventor Yasuo Kurusu 1 Horikawa-dori Teranouchi, Kamikyo-ku, Kyoto-shi, Kyoto Prefecture 1-chome, Tenjin Kitamachi 1-Dai Nippon Screen Mfg. Co., Ltd. (72) Inventor Kunio Tomohisa Kyoto, Kyoto 4-chome, Tenjin Kitamachi, 1-chome, Horikawa-dōji, Kamigyo-ku, 1st in Japan Screen Manufacturing Co., Ltd. (72) Inventor, Sakamoto Takashi 4-chome, Tenjin-kita-cho, 4-chome, Horikawa-duji, Kyoto, Kyoto No. 1 in Dainippon Screen Mfg. Co., Ltd. (72) Inventor Shinhiro Takita 1-5-9 Tenmabashi, Kita-ku, Osaka City, Osaka Prefecture Dainippon Screen Mfg. Co., Ltd. in Osaka Technical Center (56) Reference JP-A-63 -2463 (JP, A)

Claims (2)

[Claims] 1. A device for generating a gradation conversion curve for gradation conversion of image data, wherein data for instructing a density range of an original image is input, and the entire density range is covered according to the density range of the original image. A model curve generating means for generating a model curve whose curved state is determined, and a predetermined highlight point and a shadow point on the gradation conversion coordinate plane showing the relationship between the original image density and the copy halftone dot% or the copy density value. Correction means for correcting the model curve so as to pass therethrough and generating a gradation conversion curve for the original image, wherein the model curve generation means is prepared in advance and has different curved states. Storage unit for storing the reference gradation conversion curve of No. 1, and a combination ratio specifying unit for specifying the value of the combination ratio of the first and second reference gradation conversion curves according to the density range of the original image. 1 and 2 And a synthesizing means for synthesizing the quasi-tone conversion curve at the synthesizing ratio, thereby generating the model curve, wherein the synthesizing causes the curved state of the model curve to be at least an upward convex state and a downward convex The gradation conversion curve generating device is selected from a group of states including the following states according to the density range of the original image. 2. The gradation conversion according to claim 1, further comprising density range detecting means for obtaining the density range based on the density histogram of the original image and for giving data indicating the density range to the model curve generating means. Curve generator.