JPS6221471B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a grayscale image recording method,
This invention aims to improve the density reproduction characteristics of image recording devices, such as inkjet printers, which have a limited density reproduction range and poor density reproduction, such as reproduced output density that can only follow the input at a certain threshold or higher. purpose.

FIG. 1 shows an example of recording density versus applied voltage in a conventionally used inkjet printer.

V _T in the figure is a so-called threshold value, and it is impossible to record an image at a level smaller than this. Therefore, simply increasing or decreasing the applied voltage according to the image signal will not reproduce the density below D _nio . It is not possible.

In order to compensate for this drawback, there are methods to obtain reproducibility in low-density regions by configuring pixels with multiple dots, and to reduce the average density by thinning out dots in low-density regions. There is a method, but it results in degraded resolution and visual unnaturalness.

The present invention provides a grayscale image recording method that does not cause these problems and can obtain natural density reproducibility with an extremely simple configuration.

A detailed explanation will be given below with reference to the drawings.

FIG. 2 schematically shows an embodiment of an apparatus for realizing the grayscale image recording method according to the present invention. The image signal generator 1 is a camera or a shading pattern generator that scans an image with shading and generates an image signal according to the density, and the image signal is generated based on a function described later. It is added to the function signal from the signal generator 3 in the adder 2, passes through the corrector 4, and becomes an input signal to the recorder 5. The recorder 5 forms recording dots on the recording medium 6 with characteristics as illustrated in FIG. As shown in _FIG . 3, the input/output characteristics of the corrector 4 are expressed by a straight line connecting the characteristic curve of FIG. 1 with D _nio and D _nax . At the same time, for inputs in the range of V _T /2 to V _T , it is clamped to V _T , and for inputs below V _T /2, it is corrected to the recorder 5. prevents the application of Therefore,
As shown in FIG. 4, the characteristics of the recording density of the recorder 5 with respect to the input of the corrector 4 are linear for input V _T or more, 0 for 0 to V _T /2, and D _nio for V _T /2 to V _{T .} Become.

On the other hand, the function signal generated from the signal generator 3 is
It has an instantaneous value with a predetermined probability distribution using the instantaneous value as a variable. For example, as shown in Figure 5a, the instantaneous value in the range of positive and negative from n _- to n ₊ is set to equal probability P _(o). arise. Examples include a signal as shown in FIG. 2(b), and a signal whose polarity is inverted between positive and negative in synchronization with the scanning clock. At this time, the input concentration signal
The expected value ₁ of recording density for v ₁ is ₁ = −log ₁₀ ∫ ^v1+n+ _v1+o− 10−δ(v)
dv ₁ ......(1) is given. However, δ is a function representing the input versus recording density function in FIG. If the function signal is from Figure 5b and takes values in the range n _- ~ n ₊ with uniform probability of occurrence, and the value of |n ₊ +n _- | is sufficiently small compared to the input range, then , input is V
_{When T} ⁺ n _{- , equation (} ¹ ) ^{approximately becomes} +δ′(n _- ))...
...It is expressed as (2). However, δ' has the same slope as δ in the high concentration region, so it is a function as shown in FIG. Therefore, if n ₊ =-n _- = n, then for inputs greater than or equal to (V _T +n), _{1≈D 1} and the reproduced pixels are proportional to the input signal on average.

Also, for inputs below V _T +n, the characteristics will be as shown in Figure 7 depending on the value of n, and V _T /4≦
Good input/output characteristics are obtained in the range of n≦2V _T , and particularly optimal input/output characteristics are obtained when n is _VT /2.
Also, as shown in FIG. 8, the recording density for the corrector input as a corrector is V _T /2 of the input, and V _T +
If a discrimination point is provided for each V _s after V _s /2 and the recording density is divided into discrete values D _n +mD _s (m=0, 1, 2, etc.), then n=V _T /2. hour, 7th
The undulations seen in the figure disappear, and the stability of the output density can be considered at the above-mentioned discrete points, making driving easier.

As described above, according to this embodiment, the signal generator 3
The image signal generator 1 generates a function signal in which the probability of occurrence of an instantaneous value at each time as shown in FIG.
Compared to regular function signals such as regular sine waves, by adding this to the image signal sent from the image signal, not only can moiré be prevented from occurring, but gradation fidelity will not deteriorate. That is, image quality (degree of influence on human vision) affects gradation fidelity and noise, and the lower the sum of the two, the better. Moreover, when comparing tone fidelity and noise, the decrease in tone fidelity is more problematic than the noise. According to the reliable density function signal in this embodiment, data with clearly better gradation fidelity has been experimentally obtained compared to a regular function signal such as a normal sine wave.

Further, according to this embodiment, the addition signal in the adder 2 becomes 0 at a level lower than V _{T /} 2 as shown in _FIG . Since _T is corrected by the corrector 4, the characteristics of the recorder 5, which has poor gradation reproducibility, can be significantly improved.

In addition, in the corrector 4, the level of the above addition signal is V.
For those larger than _T , the corrector 4 performs correction to obtain an inverse characteristic value by reversing the input vs. recording density characteristic shown in FIG. Although it is possible to improve the overall quality, this correction may be performed when it is necessary to take into account a specific linearity due to the input versus recording density characteristics, and is not necessarily necessary.

In addition, in this example, the case where the instantaneous value is uniformly distributed in the positive and negative sections as a function signal has been described, but
For signals or pseudo-noise that follow other probability distributions, the expected value of recording density can similarly be made to correspond to the input continuously from the low density region.

In addition, as another method, the maximum amplitude of the function signal (n in the previous example) is input to the image signal generator 1,
The same effect can be obtained by controlling the correction range by increasing/decreasing or turning on/off accordingly.

As described above, the present invention provides a grayscale image recording method using a recording means with a limited density expression range that does not respond when the recording density is below a certain threshold V _T of an input signal, in which the input signal is given by a probability density function. Add the function signals, and for the added signal v,
The addition signal v is converted to 0 or V _T in two regions: v≦V _T /2 and V _T /2≦v≦V _T /2, and the function signal to be added in this way is By using the probability density function, it is possible to significantly improve the characteristics of the recording means, which has poor shading reproducibility, without causing moiré or deteriorating gradation fidelity, and the effect is significant. There is.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an example of input/output characteristics of a grayscale image recording device having a threshold value, FIG. 2 is a block diagram of an embodiment of a device for realizing the grayscale image recording method according to the present invention, and FIG. 3 is a diagram of the same device. FIG. 4 is a diagram showing an example of the recording characteristics after the corrector, and FIG. 5a is a diagram showing an example of the instantaneous value of the output signal of the signal generator versus the probability of occurrence. , FIG. 6 shows the shape of the function δ' used in the present invention, and FIG. 7 shows the output characteristics in the low concentration region obtained by the present invention. FIG. 8 is a diagram showing the characteristics of the corrector and subsequent parts according to another embodiment of the corrector. 1... Image signal generator, 2... Function signal adder, 3... Signal generator, 4... Input/output characteristic corrector, 5... Grayscale image recorder, 6... Recording medium.

Claims (1)

[Claims] 1. In a grayscale image recording method using a recording means with a limited grayscale expression range that does not respond when the recording density is below a certain threshold V _T of an input signal, the probability of occurrence of an instantaneous value at each time is A function signal given by a probability density function with that value as a variable is added to the input signal, and for the added signal v obtained by the addition, the added signal v at a level where v≦V _T /2 is converted to 0. The grayscale image recording is characterized in that the addition signal v at a level such that V _T /2≦v≦V _T is converted into V _T and these converted signals are supplied to the recording means having a limited gray scale expression range. Method. 2. The grayscale image recording method according to claim 1, wherein the function signal occurs with equal probability within an interval whose instantaneous value is positive or negative. 3. Claim 1, wherein the polarity of the function signal is inverted in synchronization with the scanning clock.
Grayscale image recording method described in Section 1. 4. Any one of claims 2 and 3, characterized in that the possible range of the instantaneous value of the function signal is an interval from 0.25 to twice the threshold value V _T . The gray scale image recording method described in .