JPH0414988A - printing device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention mainly relates to a recording device that uses a video signal as an input source.

[Prior Art] Conventionally, devices for printing video signals have used analog low-pass filters and high-frequency emphasis filters that use differential effects in order to improve image quality.

Digitally, there were noise removal using smoothing filters and high-frequency emphasis filters using Laplacian filters.

[Problems and Objectives to be Solved by the Invention] However, in conventional examples, whether using an analog filter or a digital filter, filtering is performed uniformly on the entire screen, so although noise is reduced, high frequencies are attenuated. There was a problem with this.

Another possible method is to digitally extract the contour, emphasize only the contour, and smooth the area other than the contour, but this requires two types of filters: a smoothing filter and a high-frequency emphasis filter. Furthermore, there is a problem in that both methods must be used selectively.

The present invention is intended to solve the above-mentioned problems, and its purpose is to provide a recording device that can output high-band image quality with low noise through simple processing.

[Means for Solving the Problems] The recording device of the present invention includes an NTSC signal input means and an A/D
demodulation processing means having a converter and outputting a luminance signal component and a color difference signal component quantized from an NTSC signal; a contour extraction filter unit that outputs the contour data and contour data); a digital data processing unit that regenerates a luminance signal component from the output of the contour extraction filter unit; and print data using the output of the digital data processing unit. The apparatus is characterized in that it includes print data generation means, and prints the image output by the print data generation means.

Further, the present invention is characterized in that non-linear transformation is performed on the contour data when regenerating the luminance signal.

[Example] A full color gradation recording device using the present invention was created. The input image signal is an NTSC signal and a line head is used.
The recording density is 6.0 dat/mm, and the number of pixels is 480 in the main scanning direction and 640 in the sub-scanning direction. The recording screen size is approximately 80 mm x 107 m.

FIG. 1 shows a schematic diagram of a system according to the present invention.

101 is an NTSC signal, 102 is a synchronous processing circuit, 103
104 is a Y/C separation circuit, 104 is a chroma demodulation circuit, 105 is an A/D converter, 106 is a video memory, 107 is a digital image processing section, 108 is a printing data conversion circuit, 109 is a sampling clock generation circuit, and 110 is a memory. 111 is a central processing unit CPU; 112 is a printing processing unit having a look-up table; 113 is a line head having a drive circuit; and 114 is a printing mechanism.

FIG. 2 shows a block diagram inside the digital image processing section 107. 1071 is a neighborhood averaging circuit, 1072 is a peripheral averaging circuit, 1073 is a subtracter, 1074 is a contour data conversion circuit (ROM), and 1075 is an adder.

The manually generated NTSC signal 101 is transmitted to the Y/C separation circuit 10
3 and is separated into a luminance signal (Y signal) and a chroma signal (C signal). The chroma signal is further processed by the chroma demodulation circuit 1.
04 and is converted into a color difference signal (RY, BY signal). On the other hand, the NTSC signal 101 is also input to a synchronization processing circuit 102, which outputs a horizontal synchronization signal and a vertical synchronization signal.

The horizontal synchronization signal and the vertical synchronization signal are input to the CPU 11 and the print processing section 112.

The Y signal and color difference signals (R-Y, B-Y) are input to the A/D converter 105.

Luminance and color difference signals are for one frame (64 in the horizontal scanning line direction)
The CPU 11 sends a control signal to the sampling clock generation circuit 109 and the memory address control unit so that the data is sampled (0 dat, 480 dat in the vertical direction). The sampling clock generation circuit 109 is CPUI 1
According to the control signal No. 1, a sampling clock is generated and sent to the A/D converter 105. The frequency (fs) of the sampling clock is set to 12.23776 MHz so that the aspect ratio of the output image is 1:1. A/D
The luminance and color difference signals converted into digital data by the converter 105 are once written into the video memory 106.

The digital image processing unit 107 first operates a band pass filter (BPF) having a peak at 1/8 of the sampling frequency on the sampled luminance signal, and outputs the size of the contour (Eg).

When BPF is expressed as a weighted matrix, it becomes as shown in FIGS. 3, 4, and 5. The frequency characteristics of these BPFs are shown in Figures 7, 8, and 9, respectively, and when fs is 1
It has a peak of /8. The operation of the BPF will be explained with reference to FIG. 6 when the filter shown in FIG. 5 is used. A
, B, C, D, and E represent 5 pixels within 3×3 pixels near the pixel of interest, Fil, F12, F13.011, GI2,
G13, Hll, HI3, HI3, Ill, I12.1
13 each represents a pixel that is a distance 4 away from the pixel of interest in the horizontal or vertical direction. The neighborhood averaging circuit 1071 outputs the average (a) of 5 pixels within the 3×3 pixels in the neighborhood of the pixel of interest, a=(A+B+C+D+E)15 The neighborhood averaging circuit 1072 outputs the average (a) of 5 pixels within the 3×3 pixels in the neighborhood of the pixel of interest, a=(A+B+C+D+E)15 The neighborhood averaging circuit 1072 outputs the average (a) of 5 pixels in the 3×3 neighborhood of the pixel of interest.
Output the average (b) of two pixels.

b = (F11+F12+F13+G11+G12
+013+H11+H12+H13+111+112+
113) /12 subtractor 1073 is neighborhood averaging circuit 10
The output of the peripheral averaging circuit 1072 is subtracted from the output of the peripheral averaging circuit 71.

The output of this subtracter 1073 is the output Eg of the BPF.

Eg=a-b Next, the output of the subtracter 1073 is the contour data conversion circuit 107
4, the data is converted (Eg').

The Eg'-Eg contour data conversion circuit 1074 is a ROM and internally contains the 10th
As shown in the figure, nonlinear data consisting of three areas is written. These three areas are a zero area that returns the O value, an emphasis area that performs monotonically increasing linear transformation on the input Eg value, and an upper limit area that returns the upper limit value of the degree of edge enhancement. The contour data converted in this manner is added to the neighborhood averaging circuit 1071 by an adder 1075 and output as new luminance data (Y'').

Y'=a+Eg In other words, in the zero region, Eg'=O, so Y'=a, and the average value of the 3×3 pixels in the vicinity of the pixel of interest becomes the new luminance. In the emphasis region, Eg' is a linear transformation of Eg, so Y'=a+Eg' constitutes an emphasis filter. In the upper limit region, Y' = a + Eg (max), and a constant value of contour emphasis is performed.

In the print data conversion circuit 108, the digital image processing section 1
07 output Y' and BY and RY stored in the video memory 106 are input and converted into RGB data.

First, color difference G-Y data is created from R-Y and B-Y.

G-Y = (-110,59) (0,3OR-Y+0.
11 B-Y) Next, add Y'' to obtain RGB data.

R" = Y" + (RY) G' = y' + (G-Y) B' = Y' + (B-Y) The created print data is sent to the print processing section 112. In the print processing unit 112, after gamma conversion, the line rad 11 is
The data is converted so as to be compatible with the No. 3 driver IC, sent to the line RAD 113, and printed in the main scanning direction.

Data conversion and printing are repeated in this way to print the entire screen.

Note that, as described above, the output (Eg) of the BPF is converted into a table in the digital image processing unit 107, but a threshold value may be set for the magnitude of Eg and subsequent processing may be branched.

In the embodiment, the digital image processing section is made into a custom IC and can be easily configured as hardware, and the data bus is connected to the CPU.
If the configuration is such that it can be directly read, filter processing can be performed using software calculations. With such a configuration, it is possible to further reduce the price.

In the embodiment, the demodulation processing is performed in an analog manner with Y/C separation and color difference demodulation, and the luminance signal and color difference signal are A/D converted, but the composite video signal is directly A/D converted and digitally converted. Demodulation processing is also possible.

[Effects of the Invention] As described above, according to the present invention, by using a filter that performs both smoothing and contour extraction, contour enhancement processing and smoothing processing can be realized without using multiple filters, and furthermore, non-linear By converting the data, it is possible to use both separately without performing conditional branch processing.
This has the advantage that it is possible to provide a printing device that outputs images with high bandwidth and low noise.

[Brief explanation of the drawing]

FIG. 1 is a block diagram schematically showing a system according to the present invention. FIG. 2 is a block diagram inside the digital image processing section. FIG. 3 is a diagram for explaining the digital filter 1. FIG. 4 is a diagram for explaining the digital filter 2. FIG. 5 is a diagram for explaining the digital filter 3. FIG. 6 is a national policy for explaining contour extraction. FIG. 7 is a diagram of the frequency characteristics of the digital filter 1. FIG. 8 is a diagram of the frequency characteristics of the digital filter 2. FIG. 9 is a diagram of the frequency characteristics of the digital filter 3. FIG. 10 is a diagram for explaining the contour data conversion circuit. 101...NTSC signal 102...Synchronization processing circuit 103...Y/C separation circuit 104...Chroma demodulation circuit 105...106...107...1071...1072...1073 ... 1074 ... 1075 ... 108 ... 109 ... 110 ... 111 ... 112 ... 113 ... 114 ... A/D converter video memory digital image processing section Neighborhood averaging circuit Peripheral averaging circuit Subtracter Contour data conversion circuit Adder Print data conversion circuit Sampling clock generation circuit Memory address control unit PU Print processing unit Line head Print mechanism Part 3, 1 other personFigure 2Figure 3Figure 4Figure 7Figure 5Figure 6Figure 8

Claims (2)

[Claims] (1) Demodulation that has a composite video signal (hereinafter referred to as NTSC signal) input means and an analog-to-digital converter (hereinafter referred to as A/D converter) and outputs quantized luminance signal components and color difference signal components from the NTSC signal. a processing means, a contour extraction filter unit that inputs the luminance signal component, performs both smoothing and contour extraction, and outputs both data (smoothed data and contour data); The apparatus comprises digital data processing means for regenerating signal components, and print data generation means for generating print data using the output of the digital data processing means, and for printing the image output by the print data generation means. Characteristic printing device. (2) The printing apparatus according to claim 1, wherein the contour data is subjected to nonlinear transformation when regenerating the luminance signal.