JPS5938789B2 - Image signal predictive restoration device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image signal predictive restoration device that performs image processing to convert a low pixel density image signal into a high pixel density image signal.

Generally, in digital copying machines, facsimile machines, etc., an electrical signal obtained by scanning a document image with a scanner is converted into a digital signal by an A-D converter, and then sampled according to the density level. By doing this, it is possible to read the image signal for each pixel.

In this case, in order to simplify the document image reading device and reduce the transmission line capacity required for digitally transmitting image signals, coarse sampling is performed to reduce the number of bits per pixel, and the number of bits per pixel is reduced. Image processing is performed on the high-density image signal to convert it into a high-pixel density image signal in order to improve the quality of the reproduced image.

Conventionally, such image processing means uses an electronic computer to predict, through simulation, how the input image will change when it is played back, and to reduce the appearance of the image due to low pixel density. Computation processing is performed to improve calculations and other issues.

In this case, digitized image signals for each pixel are temporarily stored in a recording medium such as a magnetic tape, and the magnetic tape is run on a computer to perform predetermined image processing according to an instruction program written in advance. I am trying to have it perform calculations. However, such conventional image processing means using a computer requires a computer with a large data processing capacity and an input/output device for the computer, making the entire device complex and large. In addition, it is impossible to perform simulation calculations on an input image in a single time, and it takes a long calculation time to use simulation to investigate the effects of cases where the density level changes rapidly from frame to frame. It has the disadvantage of being

The present invention has been made in consideration of the above points, and is capable of converting low pixel density image signals to high pixel density using a relatively simple circuit configuration without having to perform arithmetic processing by an electronic computer as in the past. An object of the present invention is to provide an image signal predictive restoration device that can perform processing quickly. The image signal predictive restoration device according to the present invention divides each pixel of a sampled five-level density level configuration into four subpixels, and calculates the density level distribution state of surrounding pixels around the pixel of interest. The ratio of the black or white level states of the four small pixels is determined according to the density level of the pixel of interest, and the black or white level of each small pixel is positioned according to the density level distribution of surrounding pixels. It was designed to make the students do the following.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an image signal predictive restoration device that converts an input image signal read at a low pixel density into a four-times high pixel density signal and outputs the image signal. (not shown) A/D converted by 5-value signal line with 3 bits per pixel (white level is displayed as 0, black level as 4, and halftone as 1, 2, and 3). It is sent to the pixel memory S9 of the register SR and stored therein.

The contents of the pixel memory S9 of the shift register SR are sequentially transferred to the pixel memories S8 and S7 each time one pixel of the input image signal VIN is transferred. Furthermore, the input image signal N sent to the pixel memory S9 of the shift register SR is 1
It is also transferred to the line memory and BMl, and this memory BMl
When one pixel of the next input image signal IN is sent to this, the contents of are transferred to the pixel memory S6 of the shift register SR, and thereafter the contents are sequentially sent to the pixel memories S5 and S4.

That is, the contents of the pixel memory S9 of the shift register SR3 and the contents of the pixel memory S6 are delayed by one line in time. Similarly, the pixel memory S6 of the shift register SR is sent to the 1-line memory BM2, and is 2 times larger than the input image signal VIN sent to the pixel memory S9.
The contents of the previous line are in the pixel memory S of the shift register SR.
3, and thereafter its contents are sequentially sent to pixel memories S2 and Sl. Such shift registers SR and 1
Due to the configuration of the line memories BM1 and BM2, the density level state in a pixel area of a 3×3 configuration around the pixel of interest S5 (stored contents of the pixel memory) that is the object of image processing is stored. Next, the contents of each pixel memory S, ~S1 of the shift register SR are sent to the calculator ADD, where calculations are performed according to the following formula to select the pixel of interest S5.
Outputs T1 to T4 are produced by dividing each pixel (eight pixels) around the pixel into four pixel sections. ” Each output T1 to T4 of the calculator ADD is sent to the next-stage comparator CMP, where the following predetermined comparison calculations are performed in order to detect the state of distribution of pixel density levels around the pixel of interest S5. This produces 6-bit binary outputs C1-C6, where C1-C6 are the respective outputs of the comparators CMP1-CMP6.

Next, the outputs C1 to C6 of this comparator CMP are sent to each address of the read-only memory ROM, and this R
The 3-bit signals S5, O, S5-1, S5- of the target pixel S5 of the shift register SR are sent to other addresses of OM.
2 is inputted respectively, and according to the combination of each address input, the content stored in this in advance is read out and the pixel of interest S5 is
The density level outputs 01 to 04 of each small pixel divided into four are determined.

Here, a memory code is incorporated in the ROM so as to determine outputs 01 to 04 according to Bj using an algorithm described below. At this time, among the outputs T1 to T4 of the Calo calculator ADD, 0 to 04 are set in order from the maximum level according to the outputs C1 to C6 of the density distribution state detected by the comparator CMP as the density of the target pixel S5. Set to 1 (black level) according to the level.

In this case, for example, when S5=2, any two outputs from 01 to 04 are set to 1, and further T,〉T2〉T3〉
If it is T4, as shown in Figure 2, 2 of 01 and 02
Set the output to 1 (black level J, and the remaining 2 outputs 03 and 04 to O)
(white level). Furthermore, the parallel signals 01 to 04 simultaneously output from the ROM are converted into parallel/serial conversion circuits P
S, and is converted into a serial signal 0ut (digital signal) suitable for transmission. This parallel/series circuit PS is composed of buffer memories for 4 lines.
As shown in the figure, the memory for the first two lines is 01 to 0.
4, the memory for the second two lines is 0.
One line is output by repeating 1, 02, 01, 02, and one line is further output in the order of 03, 04, 03, 04 to send out the serial signal VOut. In the image signal predictive restoration device according to the present invention configured as described above, the input image signal VIN of the five-value pixel of interest S5 read at a low pixel density is converted into an output image of small pixels obtained by dividing the pixel of interest S5 into four. The density level of each small pixel in the pixel of interest S5 that is converted into the signal VOut and processed into a 4 times higher pixel density image and is in the high pixel density state is only the density level of the read pixel of interest S5. pixel S1 in a specific area around it
Since the determination is made taking into consideration the density level distribution state of S4, S6 to S9, it is possible to obtain a reproduced image with good clarity and high quality that can give a visually natural feeling. Furthermore, in the embodiment described above, the density levels of each of the three pixels around the pixel of interest S5 are added by the l calculator ADD according to the above equation 1). This processing by the calculator ADD, which is a process in determining the density level distribution of surrounding pixels, may be performed using the following equation.The addition processing of image signals is performed for pixels S2, directly adjacent to the pixel of interest S5, S4, S6
, S8 are divided into four pixel sections S. Therefore, by this adder ADD, the above (2)
When performing arithmetic processing on formulas, it is possible to perform almost the same image processing as described above, and in addition, the adder ADD
This has the advantage that the configuration can be further simplified. As described above, in the image signal predictive restoration device according to the present invention, a specific pixel region consisting of the target pixel of the input image signal for each pixel and its surrounding pixels obtained by scanning and sampling the original image is provided. Each of the five values of image information is sequentially temporarily stored and accumulated in a storage circuit consisting of a shift register and two 1-line memories, and the pixel of interest is divided into four When divided into small pixels, the black or white level of each small pixel is determined according to the density distribution state of surrounding pixels (the density level of one surrounding pixel or the sum of the density levels of at least two surrounding pixels). An arithmetic circuit consisting of a 77n calculator and a comparator performs the calculations necessary for the calculation, and reads the density level corresponding to each position of the small pixel from the ROM according to the density level of the pixel of interest and the result of the calculation. If necessary, the four parallel outputs of the ROM are converted into serial signals that can be easily transmitted by a parallel/serial conversion circuit to perform image processing.

Therefore, with a relatively simple configuration, it is possible to predictably restore an image signal obtained by reading a document image at a low pixel density into an image with good clarity and a pixel density four times higher than that of conventional electronic It has an excellent advantage of being able to perform image processing more quickly than when using a computer.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
The figure shows an example of the density level of each subpixel when the pixel of interest is divided into four subpixels, and Figure 3 shows the division into subpixels to explain the operation of the parallel/serial conversion circuit in the same example. FIG. 3 is a diagram showing a pixel configuration. SR...Shift register, BMl, BM2...
...1 line memory, ADD...--force n calculator, CMP...comparator, ROM...read-only memory, PS...parallel Series conversion circuit.

Claims (1)

[Claims] 1. A first means for sequentially accumulating density level information of each pixel in a specific pixel area consisting of a pixel of interest and adjacent pixels around it based on an image signal obtained by reading a document image in pixel units at multi-value density levels; A second means for determining the density level distribution in pixels surrounding the accumulated pixel of interest, and a ratio of black and white distribution corresponding to each density level of the pixel of interest when the pixel of interest is divided in advance into a plurality of small pixels, and Data on the black and white positions of each small pixel according to each density level distribution state in surrounding pixels of the pixel is stored in a memory, and the density level of the pixel of interest and the density level distribution of surrounding pixels determined by the second means are stored in the memory. and a third means for determining black and white of each small pixel by reading predetermined data from the memory according to the following.