JPS5977771A - Picture signal processor - Google Patents

Abstract

PURPOSE:To obtain a sharp picture signal by a simple optical system by quantizing and storing a read picture in a line memory, and reading and delaying the picture specifically and then performing contour emphasizing operation. CONSTITUTION:An original 15 set around a drum 16 is irradiated with light from a light source 22 and its reflected light is converged to form an image on the surface of an aperture 25. Light passed through the aperture 25 is converted by a photoelectric converter 26 into an electric signal, which is amplified 27 and then applied to an A/D converter 28. A sampling pulse generator 29, on the other hand, is initialized by a phase signal (g) on every turn of the drum 16 to generate sampling pulses (h) of frequency corresponding to scanning density. The converter 28 converts the input signal from analog to digital by the pulses (h) to supply an output signal to a memory part 30. The contents of the memory part 30 are read out in specific order and applied to an arithmetic part 31. The arithmetic part 31 delays the picture signal specifically and then processes the signal by using a contour emphasis constant (n) set on a constant setting panel 32 as a parameter to generate output data O and an output clock P.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention is applied to facsimile machines, scanners, etc. when scanning originals with halftones, enhancing the outline of blurred images, and increasing the sharpness of images. The present invention relates to a signal processing device.

Conventional Structures and Problems Many facsimile machines and scanners that can process originals with halftones have conventionally been analog systems, one example of which is shown in FIG. This is a method using a blur mask, and is often used as a means of edge enhancement. A document 2 pasted on a rotating drum 1 is irradiated with light 9 by a light source 3 and a lens 4, and the reflected light is collected by a lens 5, divided by a half mirror 6, and sent to an aperture 7 and Focus the image on the aperture 8 image. The aperture has an aperture size corresponding to the scanning density, and the aperture 8 has an aperture size 2 to 4 times that of the normal aperture 7. The reflected light that has passed through the aperture 7 is turned into an electrical signal by a photoelectric converter 9, and is amplified by an amplifier 10 to become a sharp signal a. Further, the reflected light that has passed through the aperture 8f becomes an electric signal at a photoelectric converter 11, and is amplified and level-adjusted by an amplifier 12 to become an unsharp signal. Sharp signal a and unsharp signal S are constant setting panel 13
The signal is added to the arithmetic unit 14 together with the contour enhancement constant G set by , and a predetermined calculation is performed to obtain the output signal d. The calculation performed by the calculator 14 is as follows: 0UT=A+Kx(A -B) is the calculation. Through this calculation, as shown in Fig. 2, the image signal value e when no contour enhancement is applied, that is, at K=00, becomes the image signal value f when contour enhancement is applied, and the contours of the image are emphasized. Ru.

However, such a conventional method requires a No. 7 mirror, and requires two sets each of an aperture, a photoelectric exchanger, and an amplifier. Therefore, the optical system becomes complicated,
This is an expensive device. Also, to make the scanning density variable,
The aperture size must be made variable, and if the two apertures are made variable, the level will also fluctuate, so it is necessary to correct the level in order to match the two amplifier outputs, which poses a problem in maintaining stability. Ta.

OBJECTS OF THE INVENTION It is an object of the present invention to provide an image signal processing device that digitally processes an image signal obtained by a simple optical system instead of the conventional complicated optical system and obtains a sharp image signal. do.

Structure of the Invention The present invention reads an image with a single reading optical system, and
With a configuration in which the image signal is quantized by a D converter, stored in a selected line memory in synchronization with scanning, read out in a predetermined order, and subjected to a predetermined delay, the image signal is subjected to contour enhancement calculations.
This allows a complicated optical system to be simplified.

Description of examples Embodiments of the present invention will be described below with reference to the drawings.

FIG. 3 is a schematic configuration diagram showing an embodiment of the image signal processing device of the present invention, and FIG. 4 is a timing chart of signals of the main parts of the embodiment. The drum 16 to which the original 16 is attached is rotated by a drive motor 17 . drum 16
Coaxially with the drum 16, a notched disc 18 for generating a phase signal is provided, which rotates between the phase signal generators 190 and generates one phase signal g (see FIG. 4) per rotation of the drum 16. It has occurred. On the other hand, the sub-scanning movable table 2o carrying the scanning optical system is moved by the sub-scanning drive section 21 in the axial direction along the drum 16 at a speed corresponding to the scanning density. On the sub-scanning movable table 20, a light source 22, a lens 23, a lens 24 are arranged.
, an aperture 25, a photoelectric converter 26, an amplifier 27, and other scanning optical systems are mounted thereon. The document 16 attached to the drum 16 is irradiated with light by a light source 22 and a lens 23, and the reflected light is collected by a lens 24 and sent to an aperture 25.
The image is formed on the surface of The aperture 25 has an aperture size corresponding to the scanning density, and the light passing through the aperture 25 becomes an electrical signal in a photoelectric converter 26, is amplified by an amplifier 27, and is applied to the human/D converter 2. The sampling pulse generator 29 is initialized every rotation of the drum 16 by the phase signal g, and generates a sampling pulse h of a frequency corresponding to the scanning density. The A/D converter 28 A/D converts the output signal i from the amplifier 27 using the sampling pulse h, and outputs sampling data l and data clock j. The memory section 3o has a plurality of main scanning line memory configurations, and outputs -N data no N+8 data m to Na data, and is added to the calculation section 31.

The calculation unit 31 uses the contour enhancement constant % f harammeter set by the constant setting panel 32, and converts the N-s data to -N data! by the data clock controller. , N-s data m to generate output data O and output clock P.

Note that details of the memory section 30 and the calculation section 31 will be explained below with reference to the drawings.

FIG. 6 is a detailed configuration diagram of the memory section 3o.

The line memory 33 has a capacity for 17 main scanning lines, and has input data, output data, read line control, and address control terminals. The address counter 34 is initialized for each main scanning line by the phase signal g, and is counted up by the data clock. The output of the address counter 34 is applied to an address control terminal of the line memory 33 to designate a pixel position in the main scanning direction. The sampling data i is applied to the input data terminal of the line memory 33, and the data clock j
The data is latched by the latch 35 and output as N+8 data m. The write selector 36ij and the data clock j are applied to the read/write control terminal of any one of the 17 main scanning lines of the line memory 33, and the memo I) of that main scanning line is placed in a writing state. The other main scanning line memories are in a read state. The N-s selector 37 and the N-selector 38 select any one output data terminal of the 17 main scanning lines of the line memory 33 and output it as N-8 data k and N data, respectively. The write selector 36, N-s selector 37, and N selector 38 are controlled by a selector controller 39 controlled by a phase signal g. The main scanning line in the line memory 33 indicated by the symbol is selected.

Table 1 FIG. 6 is a detailed configuration diagram of the calculation section 31.

The latch 40 and pixel shift registers 41 and 42 are connected in columns to form a 17-pixel shift register, and a data clock j is applied to each to shift each pixel. Further, each main scanning line is initialized by the phase signal g. latch 43 and 8 pixel shift register 44,
45, latch 46 and 8 pixel shift register 47+'
8f'j, latch 40 and 8 pixel shift register 41.4
2, each has a complete configuration of 17 pixel shift registers, and is connected to a data clock j and a phase signal gK. A latch 40 is attached to the output N-f3 data of the memory section 3o.
The data no. is connected to the latch 43, and the N+8 data m is connected to the latch 46, respectively, and serve as inputs to each of the 17-pixel shift registers. The respective outputs of the latch 40°43.46 and the 8-pixel shift registers 41, 42°44.45, 47.48 are input to an adder 49 and added. The addition result is multiplied by 91/8 by a 3-bit shifter 60 and becomes unsharp data q. The output of the 8-pixel shift register 44 is input to a subtracter 61 as sharp data γ, and the value of unsharp data q is subtracted therefrom. The output of the subtracter 51 is input to the multiplier 52 together with the contour enhancement constant n, and is calculated.
The output is added to sharp data γ by an adder 63. The delay circuit 54 transfers the data clock j to the arithmetic circuit 6.
After a delay of 1 processing time, the output of the adder 5o is latched pixel by pixel in addition to the latch 65, and the output data is set to Ok. Note that the output of the delay circuit 54 is the output clock P.
It is also output as Like other circuits, the latch 55 is initialized for each main scanning line using the phase signal g.

FIG. 7 illustrates the processing of the arithmetic unit 31 explained in FIG. 6 using another expression.

Arrow 66 indicates the main scanning direction, and arrow 57 indicates the sub-scanning direction. Each square shows one full pixel.

Therefore, -67 indicates a scanning window of 17 pixels in the main scanning direction and 17 pixels in the sub-scanning direction. 6 and 7, the output of one latch 40, 43, 46 is the pixel 68, 61, .
64, 8 pixel shift registers 41, 42, 44, 45
, 47.48 outputs pixels 59, 60, 62, 63, 6
5.66 respectively. That is, in this embodiment, the document 15 is scanned while moving pixel by pixel in the main scanning direction in one scanning window 67, and when the main scanning line ends, it is moved by one pixel in the sub-scanning direction, and then the document 15 is scanned in the same manner in the main scanning direction. This is a process in which scanning is repeated over and over again.

Also pixels 58, 59.60.61, 62.63.6
The respective values of 4.65°66 are DI, D2. Da,
D4゜Ds, Da, D'y, DB, D9, the value of sharp data γ is A, the value of unsharp data q is tB, the value of edge enhancement constant n is k, and the value of appearance data 0 is all OUT. , B = 178X (DI+D2+D3+D4+D6+
D7+DB+D9)mu=ns OUT=person ten Kx(mu-B) The following calculation will be performed.

In this embodiment, a scanning window of 17 pixels x 17 pixels has been described. However, in another embodiment, in order to make the size of the scanning window arbitrary, the line memory 33 shown in FIG. The number of lines equal to the number of pixels in the sub-scanning direction is prepared, and the write selector 36° N-8 selector 37 and N selector 38 are also configured to allow selection of the number of lines, and the selector controller 39 uses the selection table shown in Table 1. It has a configuration that can be arbitrarily rewritten, and in FIG. 6, the 8-pixel shift registers 41, 42, 44, 45, 47, and 48 can be shifted by the number of stages of pixels in the main scanning direction of the scanning window, and output from any stage. A scanning window of any size can be constructed by using a shift register that can take out . According to this method, although the configuration becomes complicated, the scanning window size is not limited as in the above embodiment, and a scanning window of any size can be selected according to the image processing method.

As another embodiment, an example in which the detailed configuration of the memory section 30 in FIG. 3 is different from that in FIG. 5 will be described below with reference to FIG. 8.

The line memory 68 has a capacity for 18 main scanning lines, and has input data, output data, read line control, and address control terminals. The address counter 69 is initialized for each main scanning line by the phase signal g and counted up by the data clock j. The output of the address counter 69 is applied to the address control terminal of the line memory 68 to set the image position in the main scanning direction.

Sampling data i is applied to the input data terminal of line memory 68. The write selector 70 applies the data clock j to the read/write control terminal of any one of the 18 main scanning lines of the line memory 68 and puts the memory of that main scanning line into a write state. The other main scanning line memories are in a read state. N-8 selector 7
1 and N selector 72 and N+8 selector 73 (d) selects any one output data terminal of the 18 main scanning lines of line memory 68 and outputs it as N-8 data k, N data no, and N+8 data m .

The write selector 70, the N-s selector 71, the N selector 72, and the N+8 selector 73 are controlled by a selector controller 74 controlled by a phase signal g. On the other hand, all main scanning lines in the line memory 68 indicated by the symbol of each selector are selected.

According to this embodiment, all data is stored in the line memory 68.
Since the output of the three data can be controlled by the same type of selector, the configuration of the memory section 3o becomes orderly.

DETAILED DESCRIPTION OF THE INVENTION As described above, the present invention reads image sound using a single reading optical system, quantizes it using an A/D converter, stores it in a selected line memory in synchronization with scanning, and stores it in a predetermined order. By using a configuration that performs contour enhancement calculations on the image signal read out with a predetermined delay, it is possible to simplify the complicated optical system, obtain a stable and sharp image signal, and realize an inexpensive image signal processing device. .

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of a conventional image signal processing device, FIG. 2 is a waveform diagram of an output signal of the conventional image signal processing device, and FIG. 3 is a schematic diagram of an image signal processing device showing an embodiment of the present invention. 4 is a timing chart of the main signals of the same embodiment, FIG. 5 is a detailed configuration diagram of the memory section, FIG. 6 is a detailed configuration diagram of the calculation section, FIG. 7 is a configuration diagram of the scanning window, FIG. 8 is a diagram showing details of a memory in another embodiment of the present invention. 1.16...Drum, 2,16...Manuscript, 3.32°゛°゛°Light source -4, 5, 23, 24...
... Runes, 6 ... Half mirror - 17, 8
, 25°゛°°° Apatya, 9.11.26...
・Photoelectric converter, 10, 12.27...Amplifier,
14... Arithmetic unit-13, 32... Constant setting panel, 17... Drive motor, 18...
...Knurled disc, 19...Phase signal generator, 2o...Sub-scanning moving table, 21...
... Sub-scanning drive unit, 28... A/D converter,
29... Sampling pulse generator, 30...
...Memory section, 31...Arithmetic section, 33.6
8...Line memory, 34,69°°°゛°. Address counter 35.40.43.46.55...
・°・Latch, 36.70...Write selector, 37.71...N+8 selector, 38.72
...H selector, 39.74°...Selector controller, 41, 42°44, 45.47.
48...8 pixel shift register 49.53.
...Adder, 60...3 bit shifter. 61... Subtractor, 52... Multiplier, 5
4... Delay circuit, 56... Main scanning direction, 67... Sub-scanning direction, 5B, 59, 60, 6
1.62,63,64.6666...pixels, 6
7...Scanning window, 73...N+8 selector. Figure 2 Min −゛～, −X ′″′ − Tobun Mino (=

Claims (3)

[Claims] (1) A first means for optically scanning an image including halftones and generating an electrical signal for each scan, and a second means for generating a sampling pulse having a reading frequency in synchronization with the scanning. , a third means for quantizing the electrical signal in synchronization with the sampling pulse and converting it into sampling data; and a fourth means for storing the sampling data in a predetermined order in a line memory having a plurality of line numbers; a fifth means for reading the sampling data from the line memory in the order of predetermined line numbers; a sixth means for outputting the sampling data in synchronization with the fifth means; the fifth means and the sixth means; a seventh means for delaying the plurality of image signals obtained by the means by a predetermined clock cycle; an eighth means for setting a predetermined contour enhancement constant; and a plurality of images delayed by the seventh means. It is characterized by comprising a ninth means for performing an edge enhancement calculation using the signal and the edge enhancement constant set by the eighth means, and a tenth means for outputting the calculation result of the ninth means. image signal processing device. (2) The line memory has line numbers from 1 to 17, and the fourth means has line numbers 1.2.3.4.5.6°7, 8.9.10.11.12.13.14 .15.1
6.Repeat the order of 17 to store the sampling data,
The fifth means has first and second reading means, and the first reading means reads line numbers 2.3.4.5.6.7.8.9.10°11. 12,
13, 14, 15, 16, 17.1, and the second reading means reads line numbers 10, 11, 12, 13, 14, 15, 16 from the line memory.
, 17, 1, 2, 3, 4, 5, 6, 7, 8, 9 degrees, and the seventh means reads out the three types obtained by the sixth means and the sixth means. One cycle and nine cycles of a data clock synchronizing each image signal with the sampling data.
and 17 periods, and the first
The values of the image signals delayed by the reading means and the seventh means are respectively Dl, D2 and D3, and the values of the image signals delayed by the second reading means and the seventh means are respectively D4, D5, and D6, the values of the image signals delayed by the sixth means and the seventh means are D7, D8, and D9, respectively, and the value of the edge enhancement constant is K, then the ninth The means is D
5 +K x (D 5 1/8X(DI+D2+'D
The image signal processing device according to claim 1, characterized in that the image signal processing device performs the following calculation. (3) a first means for fully optically scanning an image including halftones and generating an electrical signal for each scan; and a second means for generating a sampling pulse having a reading frequency in synchronization with the scanning. , a third means for quantizing the electrical signal in synchronization with the sampling pulse and converting it into sampling data; and a fourth means for storing the sampling data in a predetermined order in a line memory having a plurality of line numbers.
a fifth means for reading the sampling data from the line memory in the order of a predetermined line number; and a sixth means for delaying each of the plurality of image signals obtained by the sixth means by a predetermined clock period. and seventh means for setting a predetermined contour enhancement constant, and performing contour enhancement calculation using the plurality of image signals delayed by the sixth means and the contour enhancement constant set by the seventh means. An image signal processing device comprising: an eighth means; and a ninth means for outputting a calculation result of the eighth means.