JPS6222304B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a coded signal transmission device, and in particular,
This paper attempts to restore low-density sampled images to high density using a predictive restoration method of digital image processing, thereby achieving high image quality.

Conventionally, when an encoded image signal is damaged due to transmission line noise, etc., the image signal of the damaged scanning period is erased and replaced with the image signal of one line before. , in order to restore a low-density sampling image to a high density, the sampling pixel Sij and the output pixel Oij(i,
j; positive integer), O ₁₁ = S ₁₁ O ₂₁ = S ₁₁ O ₃₁ = S ₂₁ O ₁₂ = S ₁₂ O ₂₂ = S ₁₂ O ₃₂ = S ₂₂ : That is, Oi, j = S [ i+1/2], j where [ ] is a Gauss symbol.

However, in the conventional type, the predictive reproduction of pixels was not sufficient, and the quality of reproduction of complex characters was difficult.
Also, if images are read at high density, reproduced images can be obtained with high fidelity, but high-density reading not only makes the image sensor expensive;
It takes a lot of time to process and transmit image signals.

In view of the above, the present invention aims to reproduce high-quality images while keeping the reading density constant, and will be described below with reference to embodiments.

FIG. 2 shows the relationship between sampling pixels and output pixels in the present invention. In FIG. 2A, the output pixels are compared to the sampling pixels in the vertical direction.
In FIG. 2B, it is increased by 1.5 times in the horizontal direction. In this case, there are two types of aperture sizes for the image sensor (small and large), and this can be easily achieved by providing two types of image scanning time and, as a result, two types of apparent apertures. can be achieved. Next, regarding the algorithm for predictive restoration of pixels, the case shown in FIG. 2A will be explained with reference to FIG. First, the information on the small aperture is
Output as is to the corresponding pixel. That is,
S ₁₁ →O ₁₁ , S ₁₂ →O ₁₂ ,…, S ₃₁ →O ₃₁ , S ₃₂ →O ₃₂ ,
..., becomes. Therefore, i of the sampling pixel Sij
is odd (corresponds to a small aperture)
is output as is to the corresponding pixel (Fig. 3). In the case of a large aperture (when i of Sij is an even number), if the information on Si,j is black or white, Si,j is directly divided into two pixels and output (FIG. 3). If the information on Sij is halftone (gray), the output pixel is determined by comparing information on adjacent pixels. For example, black pixels rarely exist alone and tend to gather together, so you can collect black pixels together, or
When distributing the halftone signal to two pixels, the signal level of each output pixel is determined at an arbitrary appropriate ratio determined in advance according to the result of the above comparison.
Therefore, in this embodiment, Si-1,j and Sij are compared, and if Sij>Si-1,j, when dividing into two, the first pixel is set to 0 signal, and the latter one is outputted as Sij. (Figure 3). Further, in the case of SijSi-1,j, when dividing into two, the first pixel is set to Sij, and the latter one is set to 0 signal (FIG. 3). Below i→i+
1, and output pixels are determined in the same way. A circuit block diagram for realizing the above is shown in FIG.
First, an analog sampling signal is extracted by the sampling gate 10, and after AD conversion, it is applied to the digital memory 12, comparator 13, decoder 14, and AND circuit G1 through the latch circuit ₁₁ . If the output from the line counter 15 is an odd number signal (in the case of a small aperture), the output from the latch circuit 11 is sent through the output line 4 to the OR
Output from circuit _G2 . (Equivalent to the one in Figure 3).
Next, if the output from the line counter 16 is an even signal, and if the output from the decoder 14 is a black or white signal, the output from the latch circuit 11 is output from the OR circuit G ₂ through the output line 3. (Equivalent to the one in Figure 3). Further, if the output from the decoder 14 is a halftone (gray) signal, the comparator 13
Then, the contents of the memory 12 and the output of the latch circuit 11 are compared, and depending on the result, V _CC (0 signal) is output from the OR circuit through the output line 1 or 2 through Sij or the resistor R, respectively. Output (3rd
). The line counter 16 and the decoder 17 count the number of lines, decode it, and output whether it is an odd number or an even number. This is necessary to output Si-1, if j≧Sij, then Sij, 0, then Si-1, and if j<Sij, then 0, Sij, in this order, depending on the signal comparison results. Note that FIG. 5 shows line counters 15a, 15
b and 16 show input and output waveforms.

The above describes the restoration method when the output is binary in the case of Figure 2A, but when the output is 3-valued or in the case of Figure 2B, the output pixel can be converted in the same way. can be determined.

As described above, according to the present invention, when an image signal is encoded and transmitted, the reading density is kept constant by making the aperture for the image sensor small and large, while at the same time reproducing it on the receiving side. When outputting small aperture information as is and dividing large aperture information into two parts, output pixels are determined by comparing with adjacent image signals, allowing high-density, high-quality image reproduction. can be done.

[Brief explanation of the drawing]

1 is a diagram for explaining a conventional image restoration method, FIG. 2 is a diagram for explaining the relationship between sampling pixels and output pixels in the present invention, and FIG. 3 is a diagram for explaining the relationship between sampling pixels and output pixels in the present invention.
4 is a flowchart for explaining the operation of the present invention, FIG. 4 is an electrical block diagram for explaining one embodiment of the present invention, and FIG. FIG. 3 is a diagram showing an output waveform. DESCRIPTION OF SYMBOLS 10... Sampling gate, 11... Latch circuit, 12... Digital memory, 13... Comparator, 14... Decoder, 15a, 15b, 16... Line counter, 17... Decoder.

Claims (1)

[Claims] 1. In a coded signal transmission device that encodes and transmits image signals such as facsimile or television, two types of reading sensor apertures are installed to read, code and transmit, and when played back on the receiving side,
When outputting the information of one aperture as is and dividing the information of the other aperture into two,
1. An encoded signal transmission device comprising means for determining and filling in two divided output pixels by comparing with adjacent image signals.