JPS61147670A - Video signal binarization system - Google Patents

Abstract

PURPOSE:To make a video signal binary by a simple constitution by a high accuracy and in an actual time by setting suitably a threshold in accordance with a luminance level showing a minimum value out of luminance levels of one picture element or above located at a luminance level of a picture element and the periphery of a picture element, converting a video signal to two level signals based upon the change of positive and negative of an edge differential value and following automatically the change of the luminance distribution for a picture element each. CONSTITUTION:From a television camera 10, a video signal is obtained in accordance with a luminance level of an object and a background, and an area border detecting device 11 detects an edge in the scanning line direction, matching with a scanning timing of a video signal outputted from the television camera 10. At the area border detecting circuit 11, a video signal outputted from the television camera is inputted, at an A/D converting device, an analog digital conversion is executed, at a binary device 12, three-value signal (+, -and 0) and a picture element signal e'u ('1' or '0') outputted from the area border detecting device 11 are inputted, at an AND gate, an AND gate is obtained and the binary operation is executed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a video signal binarization method at a brightness level in a fax machine or the like.

[Prior art and its problems]

Traditionally, the video signal binarization method used for faxes, etc. used a fixed threshold value, so it was affected by lighting conditions, video tube sensitivity, shading, etc., and it was difficult to faithfully reproduce the raw image of the subject. There were no drawbacks. In order to eliminate this drawback, the video signal of the subject is simultaneously input, and each scanning line is compared with a predetermined reference signal, and a constant adjustment voltage is generated for each scanning line. A system for performing binarization has been proposed (Patent Application Publication No. 117861/1988 entitled "R Video Signal Binarization Apparatus"). However, this proposed method can only track each scanning line, and cannot track each pixel. Also,
There is a drawback that an adjustment voltage generation circuit is required for each scanning line.

[Purpose of the invention]

The purpose of the present invention is to eliminate these drawbacks of the conventional ones, and to provide a video signal that automatically follows the drift of the video signal, illumination changes on the subject, etc., and can binarize the video signal with high precision and in real time at low cost. The purpose is to provide a binarization method.

[Structure of the invention]

The present invention is a video signal binarization method that performs differential processing on a video signal, treats pixels for which the absolute value of the difference value is greater than or equal to a threshold value as an edge, and converts the video signal into a two-level signal based on fluctuations in the edge difference value. Among the brightness level of the pixel and the brightness level of one or more pixels located around the pixel,
The present invention is characterized in that the threshold value is adaptively set according to the luminance level showing the minimum value, and the video signal is converted into a two-level signal based on the positive/negative change in the edge difference value.

[Function/principle of the invention]

Due to the element characteristics of photoelectric conversion elements used in television cameras, etc., the output voltage is non-uniform even when the luminance of the measurement target is constant, and the degree of non-uniformity is ±Δ% or less of the output voltage (Δ
: fixed by the element).

Here, assuming that the output voltage distribution is a normal distribution f (x), where σ: standard deviation V: expected value (standard output voltage), 99.7% of the output voltage in a region where the brightness is constant is V It is included in -3σ to V+3σ.

Therefore, the coefficient of variation p (=standard deviation/expected value) is a constant value Δ/300, and pixels in which a voltage difference of ±3 pV or more from the expected value V occurs belong to different regions with different brightness.

Furthermore, the degree of non-uniformity of the output voltages of characters, figures, etc. of the subject generally satisfies the above-mentioned element characteristics. Therefore, the edge difference values corresponding to the region boundaries alternately appear as positive and negative groups in the scanning line direction.

In the present invention, using the above-mentioned principle, the following method is adopted which can immediately follow temporal and spatial variations in brightness distribution and can binarize video signals with high precision.

That is, the following equation (1) is used for edge detection in the scanning line direction.

F l vl, J Vt, jvx l□ ・Min
(VLJ+ vi, j++<)... (1) T h e n e'IJ = 1 E 1 s e e'1' = Q However, V(1): The concerned (i-th scanning line, j-th column ) Output voltage value (luminance level) of pixel (i, j) k: Number of delayed pixels Ko: Fixed value "IJ: (in j) Pixel characteristic signal, that is,
Difference value 1 v 1. jv t, JIkI are fixed values K.

and the minimum value Min (Vl, JI Vt, jvk), the pixel feature signal e/,
The pixel (1+j) where is "1" and the pixel feature signal e' II is "11" corresponds to an edge (area boundary).The threshold value KO・Min (vt, jv vt, "4+c) is Adaptively configured.

Note that the minimum value detection of Min (Vl, J, Vt,''-m) is performed for the brightness level of the pixel and the brightness level of the pixels of 1 (ml) located around the pixel. The brightness level of a pixel and the brightness levels of two or more pixels located around the pixel may be calculated.In the above formula (1), (a) e'1j=
In case of 1, 1F (Vl, J Vt, JIx) >O...(
2) Then elJ=f)) Elss etj=θ (b) If e'lJ = 0, 1j=O... (3) However, f3I) : (t, j) Pixel edge difference value ■ : Positive e : Negative The above equation (2) classifies the edge difference value elJ into positive and negative, and for pixels other than edges, e+3=o is set according to the above equation (3). The sign of the edge difference value elj is
An area that changes twice (■-e-■ or e→■-θ) in the scanning line direction is detected. In detecting this area, correction is made only to the number of delayed pixels in order to prevent area shift.

In the scanning line direction in the area, the first pixel (t, j) with a positive edge difference value and the last pixel (t, j) with a negative edge difference value
i, 1), and pixels (1, j+k) ~ (i, l
) as “1” and other pixel values as “0゛”.
Perform value conversion.

The above processing will be explained using FIG. 4, FIG. 5, and FIG. 6. Figure 4 shows the white letters rA written on the blue mount.
It is an explanatory diagram when J is photoelectrically converted, and 1.2 in the figure indicates a horizontal scanning line. FIG. 5 shows a video signal representing the white character rAJ in FIG. 4, and 3.4 in the diagram indicates the video signal portions corresponding to the horizontal scanning line 1°2 in FIG. 4, respectively.

FIG. 6 is a process explanatory diagram of the present invention, and FIG. 6(a) shows the video signal portion of FIG. 2 in which the brightness level of each pixel is clarified. In Figure 0, which shows the individual edge difference values when the number k of delayed pixels used is 1, 5.7 represents the edge difference value ■, 6.
8 indicates the edge difference value e. FIG. 6(b) shows a video binarized signal obtained by binarizing the video signal portion of FIG. 6(a) by the binarization process as described above.

In the area where the edge difference value changes from ■5-96-4)7 in the scanning line direction, from the pixel next to the pixel whose edge difference value is Φ5 to the pixel whose edge difference value is C6. The pixel value is "1", and the other pixel values are 10". In the area where the edge difference value changes from C6-■7-68 in the scanning line direction, from the pixel next to the pixel whose edge difference value is Φ7 , the pixel value is set to "1" for the edge difference value up to C8, and the other pixel values are set to "0".In this way, the video signal is converted into a two-level signal based on the positive/negative change in the edge difference value. Ru.

〔Example〕

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

FIG. 2 is a diagram for explaining one embodiment of the present invention, in which 10 is a television camera, 11 is an area boundary detection device,
12 is a binarization device, and 13 is a timing generator.

A video signal corresponding to the brightness level of the subject and background is obtained from the television camera 10, and the area boundary detection device 11 detects edges in the scanning line direction in accordance with the scanning timing of the video signal output from the television camera 10. and
A pixel characteristic signal e/ representing a region boundary and a ternary signal (+, -, O) representing an increase/decrease in pixel output voltage are obtained. The binarization device 12 uses the pixel feature signal “+313 value signal” output from the area boundary detection device 11 and the pixel scanning clock (C1ock) output from the timing generator 13.
The video signal is binarized using the horizontal synchronization signal (Hsyn, ), and the number of continuous pixels of "1" representing a high level and the number of continuous pixels of "O" representing a low level are obtained alternately.

FIG. 2 is a block diagram of the area boundary detection device 11. Note that the brightness level will be explained below as an output voltage.

The area boundary detection device 11 inputs the video signal output from the television camera 10, performs analog-to-digital conversion in the A/D conversion device 14, and converts the output voltage V1. The delay circuit 15 delays J by the pixel, and the comparator 16 outputs the voltage v+, J,! : The output voltage of Ii element (i, j+k) is determined to be larger or smaller than vido1<, and the minimum value Min (
Vl, JI Vt, J-et) to the multiplier 17, and 3
The value signal is transferred to the binarization device 12. In the differential circuit 18, I V 1. The multiplier 17 calculates the fixed value K stored in the buffer 19.
The product of O and the minimum value Min (Vl, JI Vi, j+k) is used as a threshold value. In the comparator 20, the difference value l v
LJ-vt, j+kl and the threshold value □ -Min (vt,
b), and if the former is greater than or equal to the latter, 1 is set as the pixel characteristic signal e/, J.
In the opposite case, O is transferred to the binarization device 12.

FIG. 3 is a block diagram of the binarization device 12. 2
The value conversion device 12 converts the ternary signal (+, -, O) output from the region boundary detection device 1 and the pixel feature signal e'lj
The same 1"or"0") is input, the AND gate 21 performs a logical product, and the resulting edge difference value e+j(■, 0.θ
) is output to the comparator 22. The comparator 22 determines only the edge difference values ■ and θ, and the shift register 23 and A
The signal is sent to the ND gate 24. In the shift register 23,
Every time the edge difference value (2) or e is transferred from the comparator 22, the previous edge difference value is sent to the AND gate 24.

Further, the contents of the shift register 23 are cleared by a horizontal synchronization signal (Hsyn, ) sent from the timing generator 13. The AND gate 24 sends -1 to the comparator 25 if the edge difference value is inverted, and +1 if it is not inverted. In comparison 25, AND gate 2
4 is output, and a trigger pulse is sent to the counter 26. The comparator 27 determines whether the value of the counter 26 is equal to the value 2 stored in the buffer 28, and if they are equal (the edge difference value is inverted twice), a trigger pulse is sent to the counter 26. It is sent to buffer 29 and buffer 30. The value of the counter 26 is cleared by the horizontal synchronization signal of the timing generator 13 and the trigger pulse from the comparator 27. The value of the counter 31 is cleared by the horizontal synchronization signal of the timing generator 13, and is incremented when the pixel characteristic signal e'lJ is "1". The comparator 32 compares the value of the counter 31 with the value 1 stored in the buffer 33, and if they are equal (corresponding to the first edge), the trigger pulse is transferred to the buffer 2.
Send on 9th. The value of the counter 34 is cleared by the horizontal synchronization signal of the timing generator 13 and incremented by the pixel scanning clock. Further, the counter value (the number of pixels between edges) is sent to the buffer 29 and its contents are cleared by the value 1 of the pixel feature signal e':'' output from the area boundary detection device 11. In buffer 29,
The contents (the number of continuous pixels at low level) are added by the trigger pulse from the comparator 32 (corresponding to the first edge in the same scanning line) and the trigger pulse from the comparator 27 (corresponding to inverting the edge difference value twice). The data is sent to the circuit 35, and the contents are then cleared. The adder circuit 36 adds the contents of the buffers 29 and 30 in response to a trigger pulse from the delay circuit 37, and stores the result in the buffer 30. buffer 3
0, the trigger pulse from comparator 27 causes the content (
The number of consecutive high-level pixels) is sent to the subtraction circuit 38, and the contents are then cleared. In the subtraction circuit 38, the buffer 3
The value stored in the buffer 39 (number of delayed pixels k) is subtracted from the value sent from 0 (correction of area deviation), and the result (number of consecutive pixels of "1") is output. The adder circuit 35 adds the value k stored in the buffer 39 to the value sent from the buffer 29, and outputs the result (the number of consecutive "O" pixels).

Although one embodiment of the present invention has been described above, the present invention can of course be implemented not only by the area boundary detection device and the binarization device as shown in FIGS. 2 and 3, but also by a microprocessor sensor. Therefore, the description of Examples is not intended to limit the scope of the present invention.

〔Effect of the invention〕

As explained above, according to the present invention, the threshold value is adaptively set according to the luminance level indicating the minimum value among the luminance level of the pixel and the luminance levels of one or more pixels located around the pixel. Since the video signal is converted to a two-level signal based on the settings and changes in the edge difference value between positive and negative, it automatically follows changes in the brightness distribution for each pixel.
With a simple configuration, video signals can be binarized with high accuracy and in real time, and the effects are significant.

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining one embodiment of the present invention, FIG. 2 is a block diagram of an area boundary detection device used in the embodiment of FIG. 1, and FIG. 3 is a diagram for explaining an embodiment of the present invention. A block diagram of the binarization device used in the example. Figure 4 is an explanatory diagram for photoelectrically converting the white letter "A" written on a blue mount. Figure 5 represents the white letter rAJ. The video signal, Figure 6, is
FIG. 3 is a process explanatory diagram of the present invention. 10... Television camera 11... Area boundary detection device 12... Binarization device 13... Timing generator 14...
A/D converter 15.37... Delay circuit 16.20.22... Comparator 17... Multiplier 18... Difference circuit 19.28.29... Buffer 21.24... AND gate 23...Shift lens star 26.31.34...Counter 35.36...Addition circuit 38...Subtraction circuit Fig. 4 Fig. 5 Time Fig. 6

Claims (1)

[Claims] (1) A video signal binarization method that performs differential processing on the video signal, treats pixels for which the absolute value of the difference value is greater than or equal to a threshold value as an edge, and converts the video signal into a two-level signal based on fluctuations in the edge difference value. Then, the threshold value is adaptively set according to the luminance level showing the minimum value among the luminance level of the pixel and the luminance level of one or more pixels located around the pixel, and the positive and negative values of the edge difference value are set. Video signal 2 characterized in that the video signal is converted into a two-level signal based on the change.
Value method.