JP2934971B2 - Image binarization processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to a binary image useful when an image having a halftone level is reproduced using a binary recording device such as a dot printer. Device.

2. Description of the Related Art Recently, an image having a halftone level such as a photographic image is displayed and output using a simple binary recording device such as a dot printer. Since this type of recording apparatus can take only a binary display mode of white and black, it is necessary to effectively binarize the halftone image. In this binarization process, for example, in a photograph described in a newspaper or the like, a screen is used to generate a shading effect using the size of a mesh, but this method is applied to a dot printer or the like as it is. It is extremely difficult.

Therefore, conventionally, binarization processing such as a tissue dither method or an average error minimization method is typically introduced to binarize a halftone image. The tissue dither method is achieved by a simple process of determining an appropriate threshold value in advance according to the pixel position and thereby discriminating and binarizing a pixel signal. Has the disadvantage that the image quality is poor. On the other hand, the average error minimization method determines binarization so that the average of the difference between the binarized image and the original image in a small area composed of a plurality of pixels is reduced, and the image quality is sufficiently improved. On the other hand, it has a drawback that the binarization process is complicated and requires a large-scale system configuration.

In consideration of such circumstances, Japanese Patent Application Laid-Open No. 57-104369 described above can obtain a binarized image having an image quality equivalent to the image quality obtained by the average error minimization method, and has a simple hardware configuration and is inexpensive. There has been proposed a halftone image binarization apparatus called an average density approximation method which can be realized in the following manner. In this method, binarization is performed based on the area ratio occupied by black points in a small area, that is, the average density. This average density is calculated for the dot to be binarized and the surrounding dots that have already been binarized. The dot to be binarized now is determined by the average density when it is black and the average density when it is white. This is performed by calculating the average density of the input signal, examining which is closer to the input signal, and employing the closer one.

However, in the above-described method, when the density and the γ characteristic of an image are converted and reproduced, conversion processing is performed in advance, and image processing is performed in the future, and the hardware is increased.

(Problems to be Solved by the Invention) As described above, in the above-described apparatus, when converting and reproducing the density, γ characteristic, and the like of an image, the conversion processing is performed in advance, and then the image processing is performed. Therefore, an object of the present invention is to provide an image binarization processing device that can reduce hardware by performing these image processes at once.

[Means for Solving the Problems] In order to achieve the above object, the present invention provides an input unit for inputting an image signal obtained by raster-scanning an image, and a conversion for setting a density conversion function. A function setting means for performing a predetermined weighting process based on the binarized past binarized data with respect to the current input image signal and the previously input image signal; Means for creating binary threshold data for the current input image signal based on the density conversion function set by the means, and a binary threshold created by the binary threshold data creating means Comparing means for forming binary data for the current input image signal by comparing the data with the current input image signal; outputting the binary data formed by the comparing means and temporarily storing the binary data; Characterized by comprising a feedback means for feeding back a past binarized data creation means.

(Operation) In the present invention, binarization threshold data to be compared with the current input image signal to form binarized data is binarized with respect to the current input image signal and the previously input image signal. It is created based on the result of a predetermined weighting process performed based on the past binarized data and the density conversion function set by the conversion function setting means.

According to such a configuration, the density conversion of the image and the conversion of the γ characteristic can be performed at one time by the above-described processing of creating the binarization threshold data, so that the density conversion and the γ There is no need to convert the characteristics, and the hardware configuration of the device can be significantly reduced.

(Embodiment) Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 3 shows the relationship between the halftone image 1 and the pixels 2 obtained by raster-scanning the halftone image 1 and separating pixels by sampling. FIG. 3 (a) shows the overall configuration, and FIG. ) Shows an enlarged part. Now, if the input image is defined as f (i, j) and the binarized data at this image position (i, j) is defined as g (i, j), 0 ≦ f (i, j) ≦ 1 g (i , j) = 0or1. That is, the input image f (i, j) at the image position (i, j) has a normalized level of 0 to 1, and the binary data g (i, j) is 0 or 1 for this level. To take. In FIG. 1 (b), it is assumed that the pixel 2a indicated by oblique lines is a data position to be processed, the pixel 2b has already completed the binarization processing, and the binarized data has been determined. 2c indicates an unprocessed pixel. In this manner, the pixels 2a, 2b, and 2c shown in FIG. 1 (b) are obtained by extracting the periphery of the pixel 2a to be processed, and a set of pixels 2b for which processing has been completed is denoted by Q. Assuming that a set composed of the pixels 2a is Q ′, the algorithm of the binarization process according to this method is shown as follows.

g (M, N) = F {f (M, N), g (i, j) [(i, j) {Q]} where the coordinates (M, N) indicate the pixel position to be processed. Where F is a function. In other words, the binarized data g (M, N) of the pixel position (M, N) to be processed is calculated based on the neighboring pixel position (i) included in the input image signal data f (M, N) at the same pixel position. , j) with the already binarized data g (i, j), and is determined by some weighting function F.
This function F is, for example, It is shown as. Here, α (i, j) is a weight function having a weight as shown in FIG. 2, for example.
As shown in this algorithm, by simply determining a weighting function in advance, the binarized data for the image signal data is easily created from the past binarized data and the input image signal data at the current coordinate position, This can be output. In addition, since all past data required for the binarization processing is binarized data, it is easy to store and store the binarized data.

Further, according to this configuration, the value X is obtained,
By comparing the value X with the coefficient α (0,0), the binarized data g (M, N) can be obtained. In other words, Was determined. Therefore, focusing on the above equation, Can be shown as Is determined, the binarized data g (M, N) can be directly obtained. Here, the right side of the above expression is the past data g (i,
j) is shown as a function.

By the way, in the conventional average density approximation method, when the density and the γ characteristic of an image are converted and reproduced, these image processings are performed after performing a density conversion process in a preceding stage in advance. That is, density level conversion is performed on the input image f (M, N).

That is, if the conversion function at this time is h (M, N), Execute the algorithm of

However, if we focus on the above equation, It can be seen that the density and γ characteristics of the reproduced image can be converted and reproduced when the binarized threshold data is created.

FIG. 3 is a block diagram showing an embodiment of the image binarization processing apparatus according to the present invention which realizes the above algorithm. An image signal input by line-scanning the halftone image is converted into n-bit digital data via the A / D converter 11, for example, and is taken in. The n-bit image signal data is compared with n-bit binarization threshold data from a ROM (read only memory) 13 to be described later in a comparator 12, and as a result, a binarized digital signal corresponding to the image signal is obtained. Output as data. Needless to say A /
A circuit for image correction, such as a shading correction circuit or an automatic gain correction circuit, may be provided between the D converter 11 and the comparator 12, if necessary. The ROM 13 contains 12
Bit binary digital data and m-bit density threshold selection data from the density threshold selector 14 are input as address data, and n-bit binary threshold data created based on the algorithm described above is output. You.

The binarized 1-bit data output from the comparator 12 is supplied to a predetermined output device (not shown), and after being delayed by two samplings via the shift register 15, the shift registers 16 and 17 are sequentially output. , 18, and 19. The shift registers 16 and 17 are serially configured to delay the one-bit data by one line scanning period. Each tap of the shift register 17 at the subsequent stage is used to shift one bit of, for example, 5 samples (5 pixels) adjacent to each other. Value data is extracted in parallel. Similarly, the shift registers 18 and 19 are similarly configured in series so as to delay 1-bit data by one line scanning period, and take out 1-bit binary data at 5 sample positions in parallel from the subsequent stage.
A total of 12 bits of parallel data extracted in parallel from each of these shift registers 15, 17, and 19 is stored in the ROM 1
3 is used as address data.

That is, when the output from the A / D converter when the current becomes the timing t ₀ is indicated as the image signal data of n bits _{(C 0, C 2 ~C n} -1), at the same timing t _0, the shift register
Binary data D _-1 one sample before 15 and binary data D _-2 two samples before are output. The parallel output of the shift register 17 is binarized data D- _(L-2) , D- _(L-1) , D- _L , D- _{(L + 1)} ,
D- _(2L-2), and the parallel output of the shift register 19 is binary data D- _(2L-2) , D- _(2L-1) , D _-2L , D- _{(2L + 1 )} , D
_{-(2L + 2)} . Therefore, the ROM 13 stores the data D _-1 , D _-2 , D
_{-(L-2) to} D- _{(L + 2)} , D- _{(2L-2) to} D- _{(2L + 2)} 12-bit data and m-bit density threshold selection data from density threshold selector 14 (12 + m) bits of data.

Here, the data C _{0 to} C _n−1 correspond to f (M, N) in the above-described algorithm, and D ₋₁ , D ₋₂ , D
_{-(L-2) to} D- _{(L + 2)} and D- _{(2L-2) to} D- _{(2L + 2)} are 12-bit data, respectively, g (i, j) [i, j∈Q ].

The n-bit binarization threshold data registered in the ROM 13 in advance is created by calculation based on the above algorithm, and the m-bit density threshold selection data from the density threshold selector 14 is h ( M, N) is data for arbitrarily selecting binarization threshold data having different density levels, γ characteristics, and the like created in the ROM in advance in consideration of a function corresponding to (M, N). The m-bit density threshold value selection data is arbitrarily selected data for each screen, and the selection method and the size of m are arbitrary. After all, there are 2 ^m types of concentrations that can be selected.

In the above-described embodiment, the shift register is used for the line delay of the binary data. However, the shift register may be stored in a RAM (random access memory) and controlled by an address controller. Further, the comparator 12 has been comparing with n-bit digital data, but this is input directly using an analog comparator without using an A / D converter, and the n-bit digital data from the ROM 13 is used. Digital data D / A
The data may be converted into an analog value by a converter and compared.

FIG. 4 shows another embodiment of the present invention.
This embodiment is constructed by replacing the ROM 13 and the comparator 12 shown in FIG. In the case of this embodiment, a total of 12 bits of parallel data respectively taken in parallel from the shift registers 15, 17, and 19, m-bit density threshold selection data from the density threshold selector 14, and n from the A / D converter 12. The RO20 is addressed by a total of (12 + m + n) bits of bit data, and a desired binarized image signal is output. Other configurations are the same as those shown in FIG.

The present invention is not limited only to the above embodiment. For example, the pixel area Q and the number of pixels of the past binarized data to be fed back to the binarization processing, the weight function α
The density conversion function h (M, N) may be determined according to the specifications. Of course, it is also possible to configure so that the calculation data is not registered in the ROM in advance, but is calculated each time.

[Effects of the Invention] As described above, according to the present invention, a relatively simple configuration can be realized at low cost, and conversion processing of image density, γ characteristics, and the like is performed at the time of generation of binary threshold data. Therefore, it is possible to provide an image binarization processing device whose hardware is significantly reduced.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the concept of pixel position and image processing, and FIG.
FIG. 3 is a diagram illustrating an example of a weight function, FIG. 3 is a block diagram showing one embodiment of the present invention, and FIG. 4 is a block diagram showing another embodiment of the present invention. 11 ... A / D converter, 12 ... Comparator, 13, 20 ... ROM, 14 ...
… Density conversion selector, 15,16,17,18,19 shift register.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (1)

(57) [Claims] 1. An input means for inputting an image signal obtained by raster-scanning an image, a conversion function setting means for setting a density conversion function, and a current input image signal and a previously input image signal. A predetermined weighting process is performed based on the binarized past binarized data, and binarization of the current input image signal is performed based on the result of the weighting process and the density conversion function set by the conversion function setting means. Binarization threshold data generation means for generating threshold data, and binarization for the current input image signal by comparing the binarization threshold data generated by the binarization threshold data generation means with the current input image signal A comparing means for forming data; a feedback means for outputting and temporarily storing the binary data formed by the comparing means and feeding it back to the creating means as past binary data; An image binarization processing device comprising: a stage;