JPH0161272B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a grayscale image signal correction system used in image recording apparatuses capable of expressing halftones, such as inkjet printers and facsimile machines.

Conventionally, in recording apparatuses such as inkjet printers that are poor in the ability to express density levels, dither methods, pseudo halftone dots, density pattern methods, and the like have been used to improve the apparent image quality of density characteristics. However, most of the conventional methods have been one-dimensional, and even two-dimensional methods have not been adapted to the characteristics of the image to be recorded, resulting in the disadvantage that the recorded images remain unnatural.

The present invention has been made in view of the above-mentioned problems, and the density reproduction range in which the reproduced output density can follow the input is limited, and the image input device is used in an image recording thermal device, such as an inkjet printer, which has poor shading reproducibility. It is an object of the present invention to provide a grayscale image signal correction method that allows natural recording and reproduction characteristics to be obtained over the entire recording surface by superimposing a special random number as a correction signal on the signal.

The principle of the present invention is to superimpose a two-dimensional signal with alternating polarity on a density input signal for recording a grayscale image, and use this as an image recording output signal. Now, in two-dimensional recording of images, etc., the recording direction of one line is X, with the recording start position as the origin.
The feeding direction of a line perpendicular to X is Y, the pixel unit in the X direction is x, and the pixel unit in the y direction is y.
Further, if the density input signal at image coordinates (x, y) is expressed as D _xyio , the two-dimensional alternating polarity signal superimposed on this is expressed as D _xycpo , and the image recording output signal is expressed as D _xyput , then D _xyput = D _xyio + D _xycpo ...(1). On the other hand, the superimposed signal D _xycpo is selected to be expressed by the following equation (2).

D _xycpo = (-1) ^x+y △D _xy ... (2) Here, △D _xy is a positive random number.

This superimposed signal D _xycpo is characterized by a two-dimensional polarized random number, and is a signal as shown in FIG. 1, for example. In the figure, a indicates a signal in the X direction, and b indicates a signal in the Y direction.

In the case of Fig. 1, the size of the random numbers is completely random, but as shown in Fig. 2 a and b, the _superimposed signal D It is also possible to make it symmetrical in both positive and negative directions.

1 and 2 show the case where the superimposed signal is a polarized random number signal completely unrelated to the concentration input signal D _xyio , but the concentration input signal D _xyio
It may be proportional to. That is, the superimposed signal D _xycpo is selected as follows: D _xycpo = (-1) ^x+y D _xyio △D _xy (3). In this case, the polar alternating random number becomes a two-dimensional concentration input signal proportional alternating random number that changes while being modulated by the concentration input signal D _xyio . In this case as well, the alternating random values of the polarized random numbers proportional to the two-dimensional concentration input signal can be made symmetrical in positive and negative directions.

Next, an example of a circuit configuration for concretely implementing each of the above signal correction methods will be explained. FIG. 3 shows an example in which the two-dimensional polar random numbers shown in FIG. 1 are used, in which 1 is a pseudo random number generator, 2 is an adder/subtractor,
3 and 4 are flip-flops, and 5 is an exclusive OR circuit. A pseudorandom number generator 1 generates a positive random number 11 in synchronization with an X-direction pixel clock 14. This positive random number 11 is added to the adder/subtractor 2 together with the concentration input signal 10. On the other hand, the X direction pixel clock 14
and Y-direction line feed signal 15 are applied to exclusive OR circuit 5 via flip-flops 3 and 4, respectively, to generate polarity signals 16 for each of x and y. This polarity signal 16 acts to cause the adder/subtractor 2 to perform addition and subtraction alternately.
Then, the random number 11 is alternately added and subtracted from the concentration input signal 10 according to the polarity signal 16. As a result, the output 13 of the adder/subtractor 2 is an image output signal obtained by adding a polarized random number signal as shown in FIG. 1 to the density input signal 10. By recording an image using this output signal, a natural recorded image with good gradation reproducibility can be obtained.

FIG. 4 shows an embodiment in which the symmetrical two-dimensional polar random numbers shown in FIG. 2 are used, and the same parts as in FIG. 3 are given the same reference numerals and their explanation will be omitted. 6 is a presettable X-direction pseudorandom number generator, and 7 is a Y-direction pseudorandom number generator. The Y-direction pseudo-random number generator 7 synchronizes the Y-direction line sending signal 15 with the random number generation synchronization clock obtained through the flip-flop 4 to generate a positive random number signal, and this random number signal is
directional pseudorandom number generator 1. The X-direction pseudo-random number generator 6 generates a positive random number signal by synchronizing the X-direction pixel clock 14 with the clock obtained through the flip-flop 3, and generates a positive random number signal from the Y-direction pseudo-random number generator 7 by the X-direction recording start signal 17. Preset positive random numbers. The output 11 of the X-direction pseudo-random number generator 6 is added to the adder/subtractor 2, and the polarity signal 1 obtained by passing the outputs of the flip-flops 3 and 4 through the exclusive OR circuit 5 is
6 is alternately added to and subtracted from the concentration input signal 10.

FIG. 5 shows an embodiment in which two-dimensional concentration input signal proportional type polarized random numbers are superimposed. This embodiment differs from the embodiment shown in FIG. 3 in that a multiplier 8 multiplies the output of the pseudorandom number generator 1 and the concentration input signal 10, and this signal is input to the adder/subtractor. Other details are the same as in Figure 3. Therefore, the output 13 of the adder/subtractor 2 in this implementation is the result of adding or subtracting a random number signal proportional to the concentration input signal to the concentration input signal.

FIG. 6 shows an embodiment in which symmetrical two-dimensional concentration input signal proportional polarized random numbers are superimposed. This embodiment is the same as that of FIG. 4 except that a multiplier 8 is added to the embodiment of FIG. The operation of multiplier 8 is the fifth
It is exactly the same as the multiplier 8 shown in the figure.

Next, the effect of using a polar random number as a correction signal, which is a feature of the present invention, will be explained based on FIG.

If polar random numbers are not used as in the present invention,
When a simple random number signal is used, a problem arises in that grainy noise due to the random numbers themselves becomes noticeable in the image.
The reason why this granular noise is noticeable is due to the spatial random elasticity of pixel gradation, and speaking of spatial frequency,
As shown at A in FIG. 7, the low frequency noise spectrum is especially an obstacle.

On the other hand, when polarized random numbers are used as in the present invention, it is possible to shift the power distribution of output noise to a higher frequency range as shown in FIG. The feeling can be reduced. This can be considered to be because the alternating period adds the effect of a carrier wave corresponding to the screen frequency in halftone dot printing, so that the spatial regularity produces the effect of suppressing unpleasant random elasticity.

In this way, using polar random numbers, the main scanning direction,
Furthermore, in order to correct for the sub-scanning direction, the seventh
It is clear that the roughness removal characteristics shown in the figure occur throughout the two-dimensional image.

As described above, in the present invention, with the recording start position as the origin, the recording direction X of one line is the main scanning direction X, and the feeding direction Y of the line perpendicular to the main scanning direction When a correction signal is sequentially added to each input signal of the read two-dimensional image by controlling the timing signal generated by the position control means based on the two-dimensional position (x, y) of each input signal according to the gray scale. , With respect to the main scanning direction Also in the sub-scanning direction Y, by using a correction signal whose signal polarity is inverted from that added to the input signal at the same position in the main-scanning direction X one line before, inkjet recording can be performed. In image recording apparatuses with poor gradation reproducibility such as the above, the apparent image quality of gradation characteristics can be significantly improved.

[Brief explanation of drawings]

FIGS. 1a and 1b are waveform diagrams showing an example of a correction signal used in the grayscale image signal correction method according to the present invention;
FIGS. 2a and 2b are waveform diagrams showing other examples of correction signals used in the grayscale image signal correction method according to the present invention;
3 to 6 are block diagrams each showing an example of a circuit configuration for carrying out the grayscale image correction method according to the present invention, and FIG. FIG. 1... Pseudo-random number generator, 2... Addition/subtraction device, 3,
4...Flip-flop, 5...Exclusive OR circuit, 6...X-direction pseudorandom number generator, 7...Y-direction pseudorandom number generator, 8...multiplier.

Claims (1)

[Claims] 1. With the recording start position as the origin, the recording direction X of one line is the main scanning direction X, and the feeding direction Y of the line perpendicular to the main scanning direction When sequentially adding a correction signal to each input signal of the read two-dimensional image by controlling the timing signal generated by the position control means based on the two-dimensional position (x, y) of each input signal according to the density, With respect to the main scanning direction Also in the sub-scanning direction Y, a correction signal added to an input signal at the same position in the main-scanning direction X one line before is used, and the signal polarity is inverted. Image signal correction method. 2. The position control means includes a first flip-flop that is inverted when correction processing for one input signal in the main scanning direction X is completed, a second flip-flop that is inverted when one line in the sub-scanning direction Y is completed, and 2. The grayscale image signal correction method according to claim 1, further comprising an exclusive OR circuit which receives the outputs of the first and second flip-flops as inputs and generates a timing signal.