JPH02271423A - Pseudo half tone picture storage device - Google Patents

Abstract

PURPOSE:To make resolution and the number of gradation compatible and to obtain a high picture quality by preparing plural different threshold matrixes, extracting the characteristic quantity of a picture signal, using the respective threshold matrixes according to the characteristic quantity, and binarizing the picture signal. CONSTITUTION:Threshold storage devices 3 and 4 which store the sizes of threshold matrix 8 X 8 picture elements and 4 X 4 picture elements, and their output are inputted to a data selector 8. A picture signal from a picture scanning input device 1 is once stored into a line memory 5, and the absolute value of a Laplacian is calculated by a spatial filter 6. A comparator 7 compares a Laplacian absolute value L of the filter 6 with a constant value Ts, and outputs it to a data selector 8. The selector 8 selects the threshold of the device 3 at the time of L < Ts, and selects the threshold of the device 4 at the time of L > Ts. Further a comparator 9 compares the threshold from the selector 8 with a picture signal from the device 1, binarizes the picture signal, and records it by a printer 10.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a pseudo-halftone image recording apparatus, such as a printer, for obtaining high-quality reproduced images.

(Prior Art) Conventionally, a systematic dither method is generally known as a method for expressing the gradation of a grayscale image as a binary image. This method sets a different threshold for each dot, which is the "minimum recording unit" on the recording surface, and compares the image signal with this threshold, thereby creating a pseudo-binary image while maintaining the gradation. It is something that becomes.

Pixels and dots, which are the minimum spatial units of image signals, are 1
It corresponds to 1:1. The threshold value for each dot is set by a matrix called a threshold value matrix. In this way, thresholds of different gray levels are set for each dot. This situation is shown in FIG. Note that although the size of the threshold value matrix is 4×4 pixels in FIG. 6, any size is possible.

By the way, conventionally, the same threshold matrix has been used over the entire screen. For this reason, for example, there are the following problems.

When recording a gray scale image using the systematic dither method, various methods have been proposed for arranging the threshold values in the threshold matrix. Gaining weight, for example f as shown in Figure 7
The attening type arrangement method is known as a method that can easily obtain relatively high image quality.

As can be seen from FIG. 7, according to this method, one point is created in one threshold matrix area in a medium density area. On the screen, the pitch of these points determines the effective resolution. Therefore, in order to reproduce fine parts of an image, it is sufficient to reduce the pitch of the points by reducing the size of the threshold matrix and increase the effective resolution.

On the other hand, the number of gradations that can be recorded using this method is the number of elements in the threshold matrix plus "1".

That is, for example, in a 4×4 pixel threshold matrix, the number of gradations is 17 (4×4+1). Therefore, if the threshold matrix is made smaller in order to increase the resolution, the number of recordable gradations will be reduced.

On the other hand, if the threshold matrix is increased in order to increase the number of gradations, the resolution will decrease.

Here, generally high resolution is not necessary for the entire screen. This is especially necessary in areas where there are dramatic changes in shading on the screen, such as images of small objects. Parts where the shading changes slowly, such as images of large objects, do not require much resolution. On the other hand, a large number of gradations is required in areas where the shade changes gradually, and a smaller number is required in areas where the shade changes rapidly. For example, in the case of an image of a person's face used in an ID photo, a small threshold matrix is preferable because the boundary between the hair and the background requires more resolution than the number of gradations. On the other hand, for the skin of the face, a larger number of gradations is required, so a large threshold matrix is suitable.

In this way, the optimal threshold matrix (size) may differ depending on the part of the screen.

However, the conventional organized dither method uses a constant threshold matrix over the entire screen, and therefore has the problem of not being able to achieve both resolution and number of gradations, resulting in lower overall image quality.

(Problems to be Solved by the Invention) As described above, conventional systems have been configured to perform binarization processing using a fixed threshold value matrix, which has led to problems such as a decrease in overall image quality.

The present invention has been made in view of the above points, and it is an object of the present invention to provide a pseudo-halftone image recording device that can achieve both resolution and the number of gradation levels and obtain high overall image quality.

[Configuration of the Invention (Means for Solving the Problems) In other words, the pseudo halftone image recording device according to the present invention inputs an image signal representing shading, and calculates the difference by comparing this image signal with the elements of a threshold matrix. In a pseudo-halftone image recording device that converts and stores values, a plurality of different threshold matrices are prepared, the feature amount of the image signal is extracted, and each of the threshold matrices is used properly according to the feature amount to record the image signal. The configuration was configured to binarize.

(Function) According to the above configuration, an optimal threshold matrix can be set according to the feature amount in each part of the image, and thereby a pseudo-halftone image recording device that achieves both resolution and number of gradations can be It is possible to obtain high image quality.

(Embodiment) Hereinafter, a pseudo halftone image recording apparatus according to an embodiment of the present invention will be described with reference to the drawings.

Figure 1 is a block diagram showing the circuit configuration.
is an image scanning input device that scans an original image to generate an image signal. This image scanning input device 1 simultaneously generates a dot clock indicating the period of one pixel and one main scanning of the image signal. A dot counter 2 counts dot clocks from the image scanning input device 1 and generates addresses for the first and second threshold storage devices 3.4. The first and second threshold value storage devices 3.4 each store values of respective elements of a mutually different threshold value matrix.

A line memory 5 stores several lines of image signals generated by the image scanning input device 1. 6 is line memory 5
This is a spatial filter that performs spatial calculations on the image signal stored in the image signal and extracts the feature amount of the image signal. 7 is a first comparison device that compares the feature quantity obtained by the spatial filter 6 with a certain constant value Ts. A data selector 8 selects the output of either the first or second threshold storage device 3.4 depending on the comparison result of the comparison device 7.

9 is an image signal obtained by the image scanning input device 1,
This is a second comparison device that performs binarization by comparing with the output of the data selector 8. Reference numeral 10 denotes a printer, such as a laser beam printer, which records the binary image signal output from the second comparator 9 to generate a duplicate image.

The present invention is characterized by comprising three to eight of these devices. The other parts 1.2.9.10 are the same parts as in the case of realizing the well-known organized dither method.

Next, the operation of this embodiment will be explained.

Here, explanation will be given assuming that the size of the threshold matrix is changed depending on the Laplacian of the image signal.

That is, first, the image scanning input device ii scans the original image to generate an image signal. Image signals are sequentially output pixel by pixel as the document is scanned.

This image scanning input device 1 simultaneously generates a dot clock that indicates one pixel period of an image signal and one main scanning period. The dot counter 2 counts these dot clocks and generates addresses for the first and second threshold storage devices 3.4. In this example, the dot counter 2 consists of two 3-bit counters, one of which counts the address X in the main scanning direction of the threshold storage device, and the other counts the address Y in the sub-scanning direction. Further, the size of the threshold value matrix stored in the first threshold value storage device 3 is 8×8 pixels, and the size of the threshold value matrix stored in the second threshold value storage device 3 is 4×4 pixels.

The threshold matrix in each threshold storage 3.4 is shown in FIG. In FIG. 2, (a) shows an 8×8 pixel threshold matrix of the first threshold storage device 3, and (b) shows a 4×4 pixel threshold matrix of the second threshold storage device 4. In FIG. In this case, in the second threshold storage device 4, the dot counter 2
By accessing the lower two bits of each XSY address, the threshold matrix is made continuous on the recording surface as shown in FIG. Each threshold storage 3.4
The output is input to the data selector 8.

On the other hand, the image signal output from the image scanning input device 1 is
- is stored in the line memory. Then, the spatial filter 6 calculates the absolute value of the Laplacian of this stored image signal. In this example, the absolute value of the Laplacian is used as a measure of image fineness. The absolute value L (x, y) of the Laplacian at the point (x, y) is calculated by the following equation.

L(x, y)・1p(x, y)−1/4(p(x,
r-I)+p(x-1,7)+P(1+1.ri+P
(1, y+I)) where p(x, y) is the point (x, y)
is the value of the image signal. As shown in FIG. 4, the above formula means subtracting the image signals of four pixels on the upper, lower, left and right sides from the image signal of a certain pixel with weights, and then finding the absolute value thereof. In the case of detailed images, the size of the image signal differs greatly between a certain pixel and surrounding pixels, so
The absolute value of Laplacian increases. Thus, it is generally known that the Laplacian can be used as a measure of image fineness.

In this case, as input to the spatial filter 6, image signals of not only the pixel of interest whose Laplacian is currently being determined, but also the pixels above, below, and to the left and right of the pixel of interest are required. A line memory 5 is required to store these image signals.

That is, when using the spatial filter 6 that receives four pixels around the pixel of interest as shown in FIG. 4, the line memory 5
As shown in FIG. 5, this can be realized by cascading memory cells capable of storing one pixel's image signal in a number equivalent to two lines. The image signals sequentially flow through the line memory 5, and the image signals of the pixel of interest and the pixels above, below, left and right of the pixel are given to the spatial filter 6.

The first comparator 7 compares the absolute value of the Laplacian output from the spatial filter 6 with a certain value T6.

This comparison result is output to the data selector 8. As a result, the data selector 8 selects the threshold stored in the first threshold storage device 3 when the absolute value of the Laplacian is smaller than Ts, and selects the second threshold when the absolute value is larger than Ts.
The threshold value stored in the threshold value storage device 4 is selected and output. The second comparison device 9 binarizes the image signal by comparing the threshold value output from the data selector 8 with the image signal. The value of Ts is determined by the size of the two threshold matrices, and there is no need to change this value depending on the image content.

The binary image signal obtained in this way is sent to the printer 1.
Recorded as 0. This printer 10 is, for example, a laser beam printer, and generates a duplicate image based on this binary image signal.

In this way, when the Laplacian indicating the feature amount of the image signal is smaller than a certain value Ts, the threshold matrix is increased,
Moreover, when it is larger than Ts, the threshold value matrix can be made smaller and recorded.

Therefore, by performing image processing that achieves both resolution and the number of gradations, it is possible to obtain an output image with a higher overall quality than before.

The above is one example for realizing the present invention.

Regarding equipment, only the minimum necessary equipment was described.

Note that in order to slightly improve the image quality, it is possible to add the following devices.

(1) Correction device for the top, bottom, left and right edges of the screen. When the line memory 5 as described above is used, an accurate Laplacian cannot be obtained for pixels at the top, bottom, left, and right ends of the image because some surrounding pixels do not exist. Therefore, for example, by adding a correction device between the line memory 5 and the spatial filter 6, which changes the value of the image signal of the pixel in the non-existent part to "0",
This mistake can be avoided.

(2) Accumulator for the output of the spatial filter 6. In this embodiment, a threshold matrix is determined for each dot. For this reason, it is conceivable that there will be parts on the screen where the threshold matrix changes frequently.

Image quality may deteriorate in such areas.

This problem can be solved by using the same threshold matrix within the region of the larger threshold matrix. Therefore, an accumulating device may be inserted between the spatial filter 6 and the first comparator 7 to accumulate the output of the spatial filter 6 for that area. This accumulator can be realized by storing the output of the spatial filter 6 in the same way as the line memory 5 and adding them in order. A threshold matrix is determined based on the cumulative result.

In addition to the above-described embodiments, the present invention can also be applied to the following cases, for example.

(1) When using a multi-gradation output device. In the above embodiment, an example of application in the case of binarizing an image signal was shown, but
For example, even when using a printer that records each pixel in four gradations, the present invention can be similarly applied if the number of gradations in the original image signal is greater than that. It is also conceivable to use a plurality of recording dots, such as four dots, for one image signal, and the present invention can be similarly applied to such a case.

(2) When using a recording device other than a printer.

It can be applied not only to printers but also to other recording/display devices such as CRTs.

(3) When using a spatial filter other than the above embodiments. By changing the weights of surrounding pixels, spatial filters with various spatial frequency characteristics can be realized. For example, even if an image signal from a pixel further away, such as a pixel two above the pixel of interest, is required, this can be applied by increasing the capacity of the line memory 5.

(4) When using more threshold storage devices.

In the above embodiment, two threshold storage devices are used, but
A similar method can be used in the case of three or more.

In this case, the first comparison device 7 compares the output of the spatial filter 6 with two or more Ts, and depending on the comparison result, the data selector 8 selects one of the outputs of the three or more threshold storage devices.
You should choose one.

[Effects of the Invention] As described above, according to the present invention, an optimal threshold matrix can be set according to the feature amount of each part in an image. Therefore, it is possible to achieve both resolution and the number of gradations, and to obtain overall high image quality.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a circuit configuration according to an embodiment of the present invention, FIG. 2 is a diagram showing a configuration of a two-well threshold matrix in the embodiment, and FIG. 3 is a 2FJt in the embodiment.
4 is a diagram for explaining the method of calculating the Laplacian in the same embodiment, and FIG. 5 is the configuration of the line memory in the same embodiment. FIG. 6 is a diagram for explaining image processing using a conventional systematic dither method, and FIG. 7 is a diagram for explaining a threshold array method using a conventional systematic dither method. 1... Image scanning input device, 2... Dot counter,
3... First threshold storage device, 4... Second threshold storage device, 5... Line memory, 6... Spatial filter,
7... First comparison device, 8... Data selector, 9
...Second comparison device, 10...Printer. Applicant's representative Patent attorney Takehiko Suzue! Δbe〉− ward～t qtomΔbe〉− (i in I411 box)

Claims (1)

[Scope of Claims] In a pseudo-halftone image recording device that inputs an image signal representing light and shade, binarizes this image signal by comparing it with an element of a threshold value matrix, and stores it, a line that stores several lines of image signals. A memory, a spatial filter that extracts the feature amount of the image signal in this line memory, and a first filter that compares the output of this spatial filter with a certain value.
a plurality of threshold value storage means storing elements of mutually different threshold value matrices, and a selection means for selecting one of the outputs of the respective threshold value storage means according to the comparison result of the first comparison means. and second comparison means for comparing the image signal with the output of the selection means and binarizing it, and using a plurality of different threshold matrices depending on the feature amount of the image signal. Halftone image recording device.