JPH06284290A - Picture information processor - Google Patents

Abstract

PURPOSE:To improve the reading speed of an original by implementing binary error spread processing attended with binarization processing of picture data read from the original at a high speed. CONSTITUTION:Multi-value picture data inputted to a binarization processing circuit pass through a delay circuit, an adder circuit and a line memory 21 till the data reach a binarization comparator 23. Then the multi-value picture data on a line 19 to which a binarization error of surrounding picture elements is added are compared with a threshold level fed from a threshold level supply device 22 and converted into binary picture data. Then a subtractor 26 calculates a binarization error of the multi-value picture data for a black reference level from a black reference level supply device 24 in parallel with the binarization processing of the multi-value picture data. Then an output from the comparator 23 becomes binary picture data. On the other hand, a binarization error as to a noted picture element selected by a selector 28 is shared by an error distribution device 29. Thus, high speed error spread processing is executed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image information processing apparatus capable of speeding up error diffusion processing and improving reading speed of a document.

[0002]

2. Description of the Related Art Generally, in a document image processing apparatus which handles code information and image information, image information having a strong contrast such as characters and diagrams with respect to an original read by a reading means such as a scanner is set by a fixed threshold value. Simple binarization is performed, and image information having gradations such as photographs is binarized by pseudo gradation means. There is an error diffusion method as a binarization method that satisfies the gradation of the area of the photographic image and has a higher resolution than the systematic dither method for the area of the character and line drawing images. The error diffusion method adds the density of the pixel of interest, which is obtained by multiplying the binarization error of the already binarized peripheral pixel by a certain weighting coefficient,
This is a method of binarizing with a fixed threshold, and a binarizing error generated by binarizing a pixel of interest is diffused to a peripheral error and error compensation is performed to minimize the binarizing error.

FIG. 7 is a circuit diagram showing the main configuration of a conventional binarization processing circuit by the error diffusion method. For example, 256-level multivalued image data A is supplied to one input of the binary comparator 70 and the subtractor 72. Binary comparator 7
0 is the input multi-valued image data A and the threshold supply means 71.
The multi-valued image data A is binarized by 1 or 0 by comparing with a predetermined threshold value supplied from each pixel. The subtractor 72 is
The binarized signal B from the binary comparator 70 is subtracted from the multi-valued image data A with positive and negative signs, and the binary value between the input multi-valued image data A and the binarized signal B is subtracted for each pixel. The digitization error e is calculated. The error distributor 73 distributes the calculated binarization error e to the peripheral pixels of the target pixel to perform error diffusion.

FIG. 8 is a block diagram for explaining the binarization processing time by the error diffusion method of FIG. In the conventional binarization process by the error diffusion method, the binarization process, the binarization error calculation process, the binarization error distribution process, and the binarization error addition process are sequentially performed in series. Therefore, the processing cycle T1 of the binarization processing by the conventional error diffusion method is the binarization time Tc, the binarization error calculation time Tec, the binarization error distribution time Ted, and the binarization error addition time. It is the total time with Tfb.

[0005]

As described above, the prior art is as follows.
After performing the binarization process, it takes time to calculate the binarization error based on the result of the binarization process.
There has been a problem that the binarization process and the binarization error diffusion process are serialized, which hinders the speedup of the error diffusion process.

The present invention has been made to solve the above-mentioned problem, and it is possible to perform a binarization error diffusion process associated with a binarization process of image data read from an original at a high speed on a pixel-by-pixel basis. It is an object of the present invention to provide such an image information processing device.

[0007]

In order to solve the above problems, the present invention provides a supply means for supplying multi-valued image data represented by a plurality of bits in pixel units to a target pixel and peripheral pixels, and by this supply means. Binarizing means for converting the supplied multivalued image data into a binarized signal by comparing it with a predetermined threshold value, and multivalued image data supplied by the supplying means and the binarized signal. 2 for each pixel
An error calculating means for calculating a binary error of a type in parallel with the conversion of the binary signal in advance, and an error selecting one of the binary errors of the two types according to the output of the binary signal. And selecting means.

Further, according to the present invention, the supply means for supplying the multivalued image data represented by a plurality of bits in pixel units to the target pixel and the peripheral pixels, the multivalued image data supplied by the supply means, and a predetermined threshold value. A binarizing means for converting into a binarized signal by comparing the two values, and the multivalued image data supplied by the supplying means and the binarized signal, and a binary value of a black pixel and a white pixel in each pixel unit. Error calculation means for calculating a binarization error in parallel with the conversion of the binarized signal in advance, and 2 of the supplied multi-valued image data and the black pixel and the white pixel according to the output of the binarized signal. An error selecting unit that selects one of the binarizing errors, an error distributing unit that distributes the pixel-unit binarizing error selected by the error selecting unit to the peripheral pixels of the pixel of interest, and an error distributing unit that distributes the binarizing error. Wrong binarization The are provided with an error adding means for adding to each pixel.

[0009]

According to the present invention, two kinds of binarization error calculation processing corresponding to the binarization result for each pixel of input multi-valued image data are performed in parallel with the binarization processing in advance. Depending on the binarization result of the input multi-valued image data, one of the two types of binarization errors that have already been calculated is selected, and error diffusion is performed based on this. The error selection processing time by the selector is faster than the error calculation processing time by the subtractor, so the overall error diffusion processing time becomes shorter,
The processing speed can be increased. As a result, the error diffusion process can be sped up, the reading speed of the original can be improved, and the pseudo halftone reading of the original can be performed at high speed.

[0010]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 2 is a block diagram showing a schematic configuration of the image information processing apparatus of the present invention. The image information processing apparatus of the present invention is, for example, an original document G placed on the original table glass 11.
Is provided with a document reading device 10 for outputting multi-valued image data read by a reading optical system 12 such as a scanner, a mechanical control unit, an image processing unit for performing image processing such as error diffusion, and a CPU for controlling these. , Reading optical system 1
Processing device 1 for processing multi-valued image data read in 2
3, a host computer 14 that functions as a controller of an electronic file, an interface 15 that connects the processing device 13 and the host computer 14, for example, a general-purpose interface SCSI-2, and each unit connected to the processing device 13. It is composed of an operating motor 16.

FIG. 1 shows the error diffusion method according to the present invention.
It is a circuit diagram which shows the structure of a binarization processing circuit. This binarization processing circuit is arranged in the processing device 13 of FIG. Figure 1
In the figure, D is a delay circuit, + is an addition circuit, 21 is a line memory having a storage capacity for one line of pixels, and 20 is a binarization circuit. The delay circuit and the adder circuit are provided for three peripheral pixels in front of the line memory 21 and one peripheral pixel for one pixel after the line memory 21 with the target pixel as a reference. The binarization circuit 20 compares the multi-valued image data input of 256 levels with a threshold supply means 22 for supplying a predetermined fixed threshold value to a predetermined fixed threshold value, and binarizes it to output binary image data. Quantization comparator 23
And a black reference supply means 2 for supplying a reference corresponding to FFh.
4, a subtractor 26 for subtracting a reference corresponding to FFh from the multi-valued image data input on the line 25, and a subtractor 2 according to the binary image data output from the binarization comparator 23.
The selector 28 for selecting either the output of 6 or the multi-valued image data input on the line 27, and the output of this selector 28, the output of which is pixel-by-pixel through the lines 30-33. The error distributor 29 is fed back to the input.

Next, referring to FIGS. 1 and 3 to 5,
The operation of the binarization processing circuit according to the error diffusion method in the present invention will be described. The multi-valued image data input to the binarization processing circuit passes through the delay circuit, the addition circuit, and the line memory 21 before reaching the binary comparator 23.
The error distributor 29 adds the binarization errors of the three peripheral pixels b, c, and d shown in FIG. 3 to the pixel of interest (the pixel of interest is the pixel of interest) shown in FIG. 3 by lines 30, 31, and 32. The line 33 takes into account the binarization error of the peripheral one pixel a in FIG. The synchronization required for adding the binarization error between the target pixel and the peripheral pixels is performed by the address management of the line memory 21 based on the processing clock. As shown in FIG. 4, the multi-valued image data on the line 19 in which the binarization error of the peripheral pixels is added is
The predetermined threshold value supplied from the threshold value supply means 22 is compared and converted into binary image data. The subtractor 26 performs the black reference supply means 2 in parallel with the binarization processing of the multivalued image data.
The calculation of the binarization error (AB ′ in FIG. 4) of the multivalued image data with respect to the black reference corresponding to FFh supplied from No. 4 is performed in parallel. The binarization error (AB in FIG. 4) of the multivalued image data with respect to the white reference (00h) is directly obtained as the multivalued image data on the line 27. Where h is he
It indicates that it is a hexadecimal number with x. That is, the binarization error of multi-valued image data can be obtained by the following equation. (Multi-valued image data considering the binarization error of peripheral pixels)-(Binary image data) = (Binarization error) Binary image data is 0 or 1, and the hexadecimal system is used when applying the above equation. Therefore, it is 00h or FFh. That is, if (input image data) ≧ (binarization threshold), “1” (black pixel) is output from the binary comparator 23 as the binarized image data. On the other hand, (input image data) <(binarization threshold)

If so, "0" (white pixel) is output from the binary comparator 23 as the binary image data. Two
When the value image data is 0, it is given by (input image data) = (binarization error). According to the above, the binarization error is (a) (multi-valued image data in which the binarization error of peripheral pixels is added) -00h or (b) (multi-valued image data in which the binarization error of peripheral pixels is added)ー FFh

It is clear that there are only two ways. In the present invention, the subtractor 26 uses the binary comparator 23
The process (b) is performed in parallel with the value conversion process.
The value corresponding to the process of (a) is obtained as it is on the line 27 as input multi-valued image data. The selector 28 selects either the result (a) or the result (b) according to the result of the binarization processing by the binary comparator 23. That is, in the selector 28, the output of the binarization comparator 23 is 1
When the output of the binarizing comparator 23 is 0, the input multi-valued image data on the line 27 is output. Conventionally, either the operation (a) or the operation (b) is performed after waiting for the binarization processing result of the binarization comparator 70. The result of binarization by the binary comparator 23 is output as binary image data. On the other hand, the binarization error for the pixel of interest selected by the selector 28 is detected by the error distributor 29 as shown in FIG.
It is distributed to B, C and D.

FIG. 6 is a diagram for explaining the binarization processing time by the error diffusion method of the present invention. In the present invention, the binarization process and the error calculation process, both of which require time, are simultaneously performed in parallel. Therefore, in the present invention, the processing time T2 is either the binarization processing time Tc or the error calculation processing time Tec, and the error selection processing time Ts.
Error distribution processing time Ted and error addition processing time Tf
It becomes the sum with b. Here, the error calculation processing time Tec>
The error selection processing time Ts. Therefore, it is understood that the conventional processing time T1> the processing time T2 of the present invention.

[0017]

As described above, according to the present invention,
It is possible to realize high-speed error diffusion processing corresponding to high-speed reading of a document.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing a configuration of a binarization processing circuit by an error diffusion method according to the present invention.

FIG. 2 is a block diagram showing a schematic configuration of an image information processing apparatus of the present invention.

FIG. 3 is a diagram illustrating addition of a binarization error from a peripheral pixel to a target pixel.

FIG. 4 is a diagram for explaining a binarization error in FIG.

FIG. 5 is a diagram illustrating distribution of binarization error from a target pixel to peripheral pixels.

FIG. 6 is a diagram illustrating a binarization processing time by the error diffusion method of the present invention.

FIG. 7 is a block diagram showing a schematic configuration of a binarization processing circuit according to a conventional error diffusion method.

FIG. 8 is a diagram illustrating a binarization processing time by a conventional error diffusion method.

[Explanation of symbols]

G ... manuscript, 10 ... manuscript reading device, 11 ... manuscript table glass,
12 ... Reading optical system, 13 ... Processing device, 14 ... Host computer, 15 ... Interface, 16 ... Motor, 20
... Binarization circuit, 21 ... Line memory, 22, 71 ... Threshold supply means, 23, 70 ... Binarization comparator, 24 ... Black reference supply means, 26, 72 ... Subtractor, 28 ... Selector, 2
9, 73 ... Error distributor.

Claims (2)

[Claims] 1. A supply means for supplying multivalued image data represented by a plurality of bits in pixel units to a target pixel and a peripheral pixel, and comparing the multivalued image data supplied by this supply means with a predetermined threshold value. Binarizing means for converting into a binarized signal by the multivalued image data supplied by the supplying means,
An error calculation unit that calculates two types of binarization errors in advance for each pixel in parallel with the conversion of the binarized signal based on the binarized signal; An image information processing apparatus, comprising: an error selecting unit that selects one of binarization errors. 2. A supply means for supplying multivalued image data represented by a plurality of bits in pixel units to a target pixel and a peripheral pixel, and comparing the multivalued image data supplied by this supply means with a predetermined threshold value. Binarizing means for converting into a binarized signal by the multivalued image data supplied by the supplying means,
2 with a black pixel and a white pixel in each pixel unit by the binarization signal
An error calculating means for calculating a binarization error in parallel with the conversion of the binarized signal in advance; and a multi-valued image data supplied to the black pixel and the white pixel in accordance with the output of the binarized signal. An error selection unit that selects one of the binarization errors, an error distribution unit that distributes the pixel-unit binarization error selected by the error selection unit to peripheral pixels of the target pixel, and an error distribution unit An image information processing apparatus, comprising: an error adding means for adding the binarized error thus generated to each pixel unit.