JPH0553103B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to an image processing device that records each pixel in black and white binary and reproduces halftones depending on the density of black pixels. It is suitable for application to facsimile and the like.

[Prior art]

Conventionally, this type of halftone processing method compares the pixel signal level of each pixel with a predetermined threshold level for each pixel position, and determines whether the pixel signal level is higher than the threshold level. The main method used was a systematic design method that approximated halftones by using black for large areas and white for small areas.

However, this method has the disadvantage that the resolution is significantly degraded compared to simple binary recording, resulting in poor image quality in areas that require high resolution, such as character areas.

In order to eliminate such drawbacks, Japanese Patent Application Laid-Open No. 58-3374
In this issue, the screen is divided into multiple blocks,
The difference between the maximum and minimum density levels of pixels within each block is detected, and this determines whether the block is a halftone image area such as a photograph or painting, or a binary image area such as a character or line drawing. A method has been proposed to determine whether there is one.

However, this method has problems such as a complicated circuit for determining the maximum and minimum density levels, and the need to increase the capacity of the image buffer memory to store halftone images. .

On the other hand, Japanese Patent Application Laid-open No. 100549/1983 discloses a method for determining halftone image regions based on input binary data. It was not something to judge, and there was still room for improvement.

[Objective] In view of the above-mentioned points, it is an object of the present invention to accurately determine a halftone image area based on an input multivalued image signal with a simple configuration.

For this purpose, an image processing apparatus according to the present invention is an image processing apparatus that discriminates a halftone image area based on an input multi-value image signal, and includes an input means for inputting the multi-value image signal, and an input means for inputting the multi-value image signal. Binarizing means for binarizing the input multivalued image signal using a threshold matrix of a predetermined size, and binarizing the binarized signal binarized by the binarizing means for each block of a predetermined size. the size of the threshold matrix used by the binarization means and the size of the threshold matrix used by the binarization means; This method is characterized in that the reference pattern matrices used by the comparison means have the same size.

Hereinafter, the present invention will be explained in detail with reference to the drawings.

〔Example〕

First, an image processing method that is a premise of the present invention will be explained. This was separately proposed by the applicant in order to eliminate the drawbacks of the prior art. death,
The number of white pixels (or the number of black pixels) in each block is counted, and if the counted value is 0 or equal to or close to the total number of pixels in the block, the block is determined to be a halftone block, and the systematic dither method is applied. An image processing method (hereinafter referred to as binary image block judgment) that outputs a corresponding block of an image that has been binarized in advance, and if not, outputs a block that has been binarized using the above fixed threshold value. (referred to as method)
It is.

However, in such a binary image block determination method, blocks that should originally be determined to be halftone blocks are often mistakenly determined to be character blocks, especially regarding blocks with data values near a fixed threshold value. , which significantly degraded the quality of the reproduced image.

FIGS. 1A to 1C are diagrams illustrating a one-dimensional binary image block determination method. Here, A in the same figure is 8
This shows image data quantized in bits and a fixed threshold (level 128).
A signal obtained by binarizing image data using the fixed threshold value is shown.

Furthermore, FIG. 1C shows the results of binary image block determination using four pixels as one block. Here, B indicates a character block and H indicates a halftone block. As is clear from C in the same figure, bk1
Blocks are correctly recognized as characters, but
Both blocks bk5 and bk6 are erroneously determined to be character blocks as a result of binary image block determination, even though they are clearly halftone image areas in the original signal.

The cause of this misjudgment is that the halftone image data fluctuates in the vicinity of a fixed threshold value, and it is necessary to take some countermeasure.
Therefore, in the present invention, such erroneous determination is avoided by the configuration shown in FIG.

FIG. 2 is an overall configuration diagram of an image processing apparatus to which the present invention is applied. In this figure, 1 is a fixed threshold binarization circuit that binarizes image information using a fixed threshold value, 2 is a memory that stores the fixed threshold value, and 3 is a circuit that binarizes image information using a systematic dither method. 4 is a threshold memory for systematic dither processing; 5 to 10 are changeover switches; 11 to 14 are image memories for storing binarized images; 15 is an image quality determination circuit; 16 17 and 18 are output binary image memories; 20 is a misjudgment prevention circuit; and 21 is a fixed threshold pattern memory.

In this configuration, when the switches 5 to 10 are in the illustrated states, image information read from a document is binarized by the fixed threshold binarization circuit 1 and stored in the image memory 11. At the same time, the image information is binarized by the dither processing circuit 3 based on the dither method and stored in the image memory 13.

Now, when one block is composed of m×n mn pixels, when the image information corresponding to n scanning lines has been binarized by the above method, the switches 5 to 10 shown in the figure are all switched from the X side to the Y side. This enables continuous processing of image information.

Also, when the switch is set as shown,
The image quality judgment circuit 15 determines the contents of the image memory 12 by m×
The n blocks are sequentially read out and the number of white pixels or black pixels in the block is counted.

If this count is equal to 0 or mn, or close to 0 or mn, the block is identified as a halftone image area.

Similarly, in the switch setting state shown in FIG.
It is compared with the contents of the fixed threshold pattern memory 21, and if the degree of matching is high, it is identified as a halftone image area.

If the identification signal sent from the misjudgment prevention circuit 20 indicates a halftone image area, the selection circuit 16 determines that the block of interest is a halftone block,
If the identification signal does not indicate a halftone image area, the identification signal from the image quality determining circuit 15 is used to determine whether the block of interest is a halftone block or a character block.

If it is determined that the block is a halftone block, the image data is transferred from the image memory 14, and if it is determined to be a character block, the image data is transferred from the image memory 12 to the output image memory 18 in units of m×n blocks.

Here, the image quality determination circuit 15 includes a CPU,
Microcomputer with ROM, RAM, etc.
Alternatively, it can be made into various forms such as a logic circuit. For example, when an image quality determination circuit is configured using a microcomputer, the image quality of one block can be determined using the following procedure.

FIG. 3A shows such an image quality determination procedure,
An example of a procedure for identifying a halftone image area when the number of white pixels is equal to mn will be shown. First, in step S1, 1-bit binary image information I ^* for one pixel is read out from the image memory on the side selected by switch 7, and then the information I ^* corresponds to a white pixel. Determine whether or not. Here, if the determination is affirmative, the count value C is incremented by +1 and the process proceeds to step S4, and if the determination is negative, the process directly proceeds to step S4. In step S4, it is determined whether the reading process for one block has been completed, and if it is determined that the reading process has been completed, the process advances to step S5.

In step S5, the count value C is 0 or
Determine whether it is equal to mn. If the determination is affirmative, the process proceeds to step S6, and if the determination is negative, the process proceeds to step S7, in which it is determined that the block related to the determination is a halftone image area or a binary image area, and an identification signal to that effect is sent. is sent to the selection circuit 16. Next, in step S8, the count value C is cleared, and the image quality determination process for one block is completed.

FIG. 3B shows an example of the erroneous determination prevention circuit 20 shown in FIG. 2 in more detail. In this figure, 141 is an address circuit for processing blocks from the image memory 14, 211 is an address circuit for accessing the fixed threshold pattern memory, and 201 is for incrementing the addresses of the address circuits 141 and 211. 202 is a circuit that performs an exclusive OR on the 1-bit output from the image memory 14 and the fixed threshold pattern memory 21 (hereinafter referred to as an EX-OR circuit); 203 is an EX-OR circuit 202 204 is a tolerance memory that stores a predetermined constant; 205 is a comparator that compares the output of the counter 203 and the output of the tolerance memory 204.

FIGS. 4 to 7 are diagrams for explaining the operation of this embodiment in more detail.

FIG. 4 shows an example of the dither threshold value stored in the systematic dither threshold memory 4, which is 8×8
It expresses 33 levels of intermediate tones with a matrix size of . Note that here, the original image consists of 8 pixels each.
It is assumed that it also has 256 levels of intermediate tones.

FIG. 5 shows an example of original picture information.

FIG. 6A is an image obtained by binarizing the image shown in FIG.
is stored in.

FIG. 6B shows the image shown in FIG. 5 binarized using the dither threshold shown in FIG.
It is stored in the image memory 14.

FIG. 7 is an example of an image stored in the fixed threshold pattern memory 21, which is a pattern created by pre-binarizing the fixed threshold value 128 with the dither threshold shown in FIG. be.

Now, assuming that the image quality determination circuit 15 performs image quality determination using 8×8 as one block, reference is made to FIG. 6A. In this case, since white pixels and black pixels coexist within the block, the original image shown in Figure 5 is erroneously determined to be a character block.

The misjudgment prevention circuit 20 compares the binarized images of FIG. 6B and FIG.
In this example, the count value is 1. It is assumed that, for example, the value 4 is stored in the allowable value memory 204. Then, since this count value is within the permissible value, the image shown in FIG. 6B is identified as a halftone image near the threshold value. Therefore, selection circuit 1
6 gives priority to the judgment output of the misjudgment prevention circuit 20;
The image shown in FIG. 6B is transferred to the output image memory 18.

Note that in the above-mentioned false judgment prevention circuit, the size of the judgment block is made to match the size of the dither matrix, but depending on the method of arranging the dither threshold values, the size of the judgment block may be made to match the size of the dither matrix. It is also possible to make it less than For example, the dithered threshold matrix shown in Figure 4 is an 8x8 matrix based on a 4x4 matrix, so the dithered image (see Figure 6B) also has a 4x4 unit. The images are highly similar.

Therefore, it is possible to design the above-mentioned erroneous judgment prevention circuit with a judgment block size of 4 x 4 or a horizontally long 4 x 8. According to such a method, the image memory can be reduced and the device size can be reduced. It can also be made cheaper.

Furthermore, by inputting image information directly into a memory such as a ROM as an address, it is also possible to configure the device to output dithered data or data processed using a fixed threshold value. .

〔effect〕

As described above, according to the present invention, it is possible to accurately determine a halftone image area based on an input multivalued image signal with a simple configuration.

[Brief explanation of the drawing]

1A to 1C are diagrams one-dimensionally showing erroneous judgments that occur in the binary image block judgment method that is the premise of the present invention; FIG. 2 is a block diagram showing one embodiment of the present invention; Figure A is a flowchart showing the image quality judgment procedure in Figure 2, Figure 3B is a diagram showing the detailed configuration of the misjudgment prevention circuit, Figure 4 is a diagram showing the systematic dither threshold array, and Figure 5. 6A is a diagram showing an example of the original image, FIG. 6A is a diagram showing an image obtained by binarizing the image in FIG. 5 using a fixed threshold, and FIG. 6B is a diagram showing the image in FIG. FIG. 7 is a diagram showing a dither pattern image for a fixed threshold value. DESCRIPTION OF SYMBOLS 1...Fixed threshold value binarization circuit, 2...Fixed threshold value memory, 3...Dither processing circuit, 4...Organized dither threshold memory, 5, 6, 7, 8, 9, 10...
Switch, 11, 12, 13, 14... Image memory, 15... Image quality judgment circuit, 16... Selection circuit, 1
7, 18... Image memory, 20... Misjudgment prevention circuit,
21...Fixed threshold pattern memory, 141...Address circuit, 201...Clock signal generation circuit, 2
02...Exclusive OR circuit, 203...Counter, 2
04...Tolerance value memory, 205...Comparator, 211...
Address circuit.

Claims (1)

[Scope of Claims] 1. An image processing device for determining a halftone image area based on an input multi-value image signal, comprising: an input means for inputting a multi-value image signal; and a multi-value image signal input by the input means. a binarization means for binarizing the data using a threshold matrix of a predetermined size; and comparing the binarized signal binarized by the binarization means with a reference pattern matrix for each block of a predetermined size. a comparison means, and a discrimination means for discriminating a halftone image area according to a comparison result by the comparison means, and a size of a threshold matrix used by the binarization means and a reference value used by the comparison means. An image processing device characterized in that pattern matrices have the same size.