JPH082101B2 - Motion adaptive scan line interpolation circuit - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion adaptive scanning line interpolation circuit used when converting an interlaced scanning video signal into a progressive scanning video signal.

Conventional technology One of the problems that occurs in current television receivers such as NTSC and PAL systems is line flicker interference. This is the interference caused by the current TV receiver's 2: 1 interlace scanning, and in order to remove this interference, the interpolation line is used to interpolate the scanning lines, and the method of progressive scanning is used. . In this method, generally, the motion of the subject of the video signal is detected, field interpolation is performed using the signal of the previous field when the subject is stationary, and the signal within the same field is used when the subject is moving. To interpolate in the field. This is called motion adaptive scanning line interpolation processing.

FIG. 7 shows an example of a conventionally known motion adaptive scanning line interpolation circuit. The operation of the conventional example will be described below with reference to the drawings.

In FIG. 7, 1 is an input terminal for inputting a composite video signal, 2 is a Y / C separation circuit for separating and extracting a luminance signal from the composite video signal, and 3 is a one-line delay device for delaying the input signal for one line period. 4 is a 262 line delay device, 5 is an adder, 6a and 6b are multipliers for multiplying coefficients, 7 is an adder, 8 is a double speed converter for obtaining a sequential scanning signal by the current scanning line and the interpolation scanning line, A motion detection circuit 9 generates a coefficient (motion control signal) according to the motion of the subject in the video signal.

First, the composite video signal input from input terminal 1 is Y / C
The color signal is removed in the separation circuit 2, and the luminance signal is obtained. Next, the luminance signal is delayed by the 1-line delay unit 3 and the 26-line delay unit 4. The signals at points a, b, and c in FIG. 7 correspond to the scanning lines shown in FIG. In FIG. 7, a signal (field interpolation signal) corresponding to c is obtained at the output of the 262 line delay unit 4, and a signal (in-field interpolation signal) corresponding to (a + b) / 2 is obtained at the output of the adder 5. can get. The weights are added to the respective interpolation signals in the multipliers 6a and 6b and the adder 7, and the result is used as the interpolation scanning line signal I. Coefficient k is 0 ≦ k
When it changes within ≦ 1, k approaches 1 when the subject of the video signal is moving, and approaches 0 when the subject is stationary. Therefore, when moving, the weight of the intra-field interpolated signal (moving image processing) becomes large and the occurrence of double image interference is suppressed. On the other hand, when stationary, the weight of the field interpolation signal (still image processing) is increased to suppress the occurrence of line flicker interference. Generally, the motion detection circuit 9 is composed of the circuit shown in FIG. In FIG. 9, 91 is a 1-frame delay device, 92 is a subtractor, 93 is a low-pass filter (hereinafter referred to as LPF), 9
4 is an absolute value circuit, 95 is a coefficient generator, and 96 is a 1-line delay device for timing adjustment. In FIG. 9, the input video signal is timing-adjusted by the 1-line delay unit 96, then delayed by 1 frame period by the 1-frame delay unit 91, and the subtracter 92 obtains the difference signal between 1-frames of the video signal. . After the LPF 93 removes noise from the one-frame difference signal and the absolute value circuit 94 takes the absolute value, the coefficient generator 95 determines the coefficient based on the absolute value signal. For example, when the subject of the video signal is stationary, there is no difference between frames, so the absolute value applied to the coefficient generator 95 is zero and the coefficient k is zero. 1 if the subject of the video signal is moving
Since there is a difference between the frames and it has a certain value as an absolute value, the coefficient k is determined to be a value between 0 ≦ k ≦ 1 according to its magnitude.

The interpolation scanning line signal I and the current scanning line signal b obtained as described above are input to the double speed converter 8. The double speed converter 8 is composed of, for example, a DRAM and realizes double speed conversion by reading data written at a normal speed at twice the speed. In this way, sequential scanning signals are obtained at the output terminal 10. For example, as shown in Fig. 10, when a video signal such that only one scanning line is at the black level is input to the scanning line at the white level (circle in the figure), the display image by the interlaced scanning is the scanning at the black level. Lines interfere with line flicker. When progressive scanning is performed using the motion adaptive scanning line interpolation circuit shown in the conventional example of FIG. 7, the coefficient becomes k = 0 because the difference between frames is 0, and as shown in FIG. A field-interpolated progressive scan image without line flicker interference is reproduced (see, for example, "Journal of the Television Society of Japan").
40, No. 5, pp. 357-365).

Problems to be Solved by the Invention However, in the above-described configuration, when displaying still superimposed characters etc. (hereinafter referred to as telop characters) that are often seen in news programs, etc., the telop characters are If it is incomplete, it has a problem of causing line flicker interference.

For example, a normal ideal telop character is displayed by forming one character in two fields as shown in FIG. 12 and repeating it thereafter, and line flicker interference occurs when it is displayed by interlaced scanning. When this is processed by the motion adaptive scanning line interpolation circuit of the conventional example, a 1-frame difference signal is not generated, so field interpolation is performed, and line flicker interference does not occur in a progressively scanned image. However, in reality, as shown in FIG. 13, the signal may differ from frame to frame due to factors on the transmission side (for example, the part indicated by a in FIG. 13) (this state is called an incomplete telop character). . In this case, the one-frame difference signal is detected in the portion a in FIG. 13 in the motion detection circuit, even though it is a stationary telop character, and the coefficient k becomes 1, for example. When K = 1, as the interpolation method, in-field interpolation is selected, and line flicker interference occurs near the portion corresponding to a of the reproduced image sequentially scanned.

In view of the above problems, the present invention provides a motion adaptive scanning line interpolation circuit that does not cause line flicker interference in a reproduced image even when an incomplete telop character is input.

Means for Solving the Problems In order to solve the above problems, a motion adaptive scanning line interpolation circuit of the present invention detects a motion that generates a motion control signal for switching a scanning line interpolation method according to the motion of a subject of a video signal. Means, a motion determination means for determining "moving" or "static" for each pixel based on the inter-frame difference signal of the subject of the video signal, the determination result from the determination means, and the pixel of interest, Extraction means for extracting and outputting the determination results of a plurality of pixels above and below, detection means for detecting that the determination result in the vertical direction from the extraction means has a specific pattern, and the detection means for specifying And a motion control signal converting means for converting the motion control signal from the motion detecting means into a value for selecting the scanning line interpolation method for a still image only when such a pattern is detected. .

The present invention detects line artifacts caused by incomplete telop characters and the like by detecting and eliminating false motion detection signals caused by incomplete telop characters and the like from the motion detection signals by the above-described configuration. be able to.

Embodiment A motion adaptive scanning line interpolation circuit according to an embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, 91 is a one-frame delay device, 92 is a subtractor, 93 is an LPF, 94 is an absolute value circuit, 95 is a coefficient generator, 97 is a timing adjustment delay device, 101 is a comparator, and 102a and 102a.
b is a delay unit between lines, 103a and 103b are inverters, 1
04 is a 3-input AND gate, and 105 is a coefficient converter. The circuit shown in FIG. 1 applies to the motion detection circuit 9 in the conventional example shown in FIG.

In FIG. 1, the input video signal is delayed by a one-frame delay device 91 for one frame period, and a one-frame difference signal is obtained at the output of the subtractor 92. The one-frame difference signal is noise-removed by the LPF 93 and converted into an absolute value by the absolute value circuit 94. Next, after the timing is adjusted by the timing adjusting delay device 97, the coefficient k is determined by the coefficient generator 95. The operation up to this point is similar to that described in the conventional example. In the present embodiment, the output of the absolute value circuit 94 is added to the timing adjusting delay device 97 and at the same time to the comparator 104. An appropriate threshold value is given to the comparator 104, and it is compared with the absolute value of the one-frame difference signal. For example, when the absolute value is smaller than the threshold value, it is a theoretical value. Outputs the value of "L". The theoretical value output from the comparator 104 is delayed by the 1-line delay devices 102a and 102b. The theoretical values at points A, B, and C in the figure correspond to logical values at each pixel in the scanning diagram shown in FIG. 2, for example. The logical value at point A is inverted by the inverter 103a and has a 3-input AND gate 1
Added to 04. The logical value at point C is also added to the inverted 3-input AND gate 104 by the inverter 103b. The logical value at point B is added to the 3-input AND gate 104 as it is. The logical value state at points A, B, and C and the output logical value of the 3-input AND gate 104 are the third
It has the relationship shown in the figure. According to FIG. 3, a 3-input AND gate 1
As for the output logical value of 04, the input logical value is A = "L", B = "H", C =
Only becomes "H" when the pattern is "L". That is, n-1
With the logical values of 3 pixels in the vertical direction of the line, n line, and n + 1 line, only the logical value of the pixel of interest (n line) is “H”, if expressed differently, only the difference between the 1 frame of the pixel of interest is sudden in the vertical direction. Then, a state in which it becomes larger than the threshold of the comparator 101 is detected. Most of the false one-frame difference signals caused by the defective telop character correspond to this state. Therefore, it is understood that the output logic value of the 3-input AND gate 104 becomes "H" only when the false motion occurs, and the false motion signal can be detected.
This detection signal (hereinafter referred to as a false motion detection signal) is connected to the coefficient converter 105 shown in FIG. The coefficient converter 105 converts the coefficients (1-k) and k from the coefficient generator 95 into other values according to the false motion detection signal. The coefficient converter 105 is composed of, for example, the circuit shown in FIG. 1051 in FIG.
Reference numerals a and 1051b are 2-input switches for selecting an input signal by the logical value of the false motion detection signal. The coefficient (1-k) from the coefficient generator 95 is added to one input terminal of the 2-input switch 1051a, and 1 is added to the other input terminal. Further, the coefficient k from the coefficient generator 95 is added to one input terminal of the two-input switch 1051b, and 0 is added to the other input terminal. For example, when a false motion is detected, the false motion detection signal has a logical value of "H", and the 2-input switches 1051a and 1051b select 1 and 0 respectively and output them. Further, when the false motion is not detected, the logical value of the false motion detection signal becomes “L”, and the coefficient generator 95
The coefficients (1-k) and k from are selected and output as they are. The output of the coefficient converter 105 is connected to the multipliers 6a and 6b shown in the conventional example of FIG. 7, and when false motion is not detected, the coefficient is controlled according to the movement of the subject as described in the conventional example. Then, a motion adaptive scanning line interpolation process is performed in which the intra-field interpolation signal and the inter-field interpolation signal are weighted and added. When the false motion is detected, the coefficients are set to (1-k) = 1, k = 0, so that the 262 line delay signal, that is, the field interpolation signal is selected as the interpolation signal. In the conventional example, when a false motion occurs, intra-field interpolation is selected as the interpolation method, resulting in line flicker interference. However, according to the present invention, even if a false motion occurs, the coefficient is k = 0, so that field interpolation is selected as the interpolation method and the line flicker interference can be removed.

FIG. 5 shows the configuration of the coefficient converter in the second embodiment of the present invention. 1051a and 1051b in FIG.
Is a two-input switch for selecting an input signal by the logical value of the false motion detection signal, 1052 is a subtracter, and 1053 is a coefficient setting device. Coefficient setter 10
At 53, an arbitrary coefficient k of 0 ≦ k ≦ 1 is set. The coefficient k is applied to one input terminal of the two-input switch 1051b and simultaneously to one input of the subtractor 1052. Subtractor 10
One is added to the other input of 52, and (1-k) is obtained as a subtraction result, which is added to one input terminal of the two-input switch 1051a. Coefficients (1-k) and k from the coefficient generator 95 are added to the other inputs of the two-input switches 1051a and 1051b, respectively. The control method by the false motion detection signal is the same as that described in the first embodiment, and by using the coefficient converter of the second embodiment, the coefficient at the time of detecting the false motion signal can be set to an arbitrary value. For example, false motion detection methods are not without risk of false detections. Therefore, if erroneous detection occurs, using the coefficient converter shown in FIG. 4, the coefficient is set to k = 0 even though the error is detected in the erroneous detection portion, and field interpolation is selected as the scanning line interpolation method. As a result, double image disturbance occurs. Therefore, the coefficient is set to a value near k = 0.5, for example, and intra-field interpolation and field interpolation are used by intermediately adding weights to suppress double image interference at the time of erroneous detection and a false motion signal is detected. It also suppresses the occurrence of line flicker interference in some cases.

FIG. 6 shows the configuration of the coefficient converter in the third embodiment of the present invention. 1051a and 1051b in FIG.
Is a two-input switch for selecting an input signal by the logical value of the false motion detection signal, 1052 is a subtractor, 1053 is a coefficient setting device, and 1054 is a minimum value selector for selecting and outputting the smaller one of the two input signals. Is. The arbitrary coefficient set by the coefficient setter 1053 is added to one input of the minimum value selector 1054. A coefficient generator is provided at the other input of the minimum value selector 1054.
The coefficient k from 95 is added, and the coefficient k is added to one input terminal of the 2-input switch 1051b. The output signal of the minimum value selector 1054 is added to the other input terminal. The output signal of the minimum value selector 1054 is simultaneously subtracted by the subtractor 10
Added to one of the 52 inputs. One is added to the other input of the subtractor 1052, and the subtraction result (1-
k) is obtained. The subtraction result is added to one input terminal of the 2-input switch 1051a. The coefficient (1-k) from the coefficient generator 95 is added to the other input terminal. For example, in the coefficient converter according to the second embodiment shown in FIG. 5, when the coefficient is set to k = 0.5 by the coefficient setting unit 1053, the coefficient from the coefficient generator 95 is detected when a false motion signal is detected. There are cases where k = 0.
In this case, since the coefficient from the coefficient generator 95 is k = 0, it is desirable to perform field interpolation, but since a false motion signal has been detected, the coefficient k = 0.5 from the coefficient setter 1053 is selected. Then, the interpolated signal becomes a signal in which the intra-field interpolation and the field interpolation are weighted and added, and some line flicker interference occurs. Therefore, in the coefficient converter in the third embodiment shown in FIG. 6, the minimum value selector 1054 compares the coefficient k from the coefficient setter 1053 with the value of the coefficient k from the coefficient generator 95, and the smaller one is always found. Is output (this is equivalent to giving priority to a process closer to a still image process). As a result, even in the state where the false motion signal is detected, the coefficient from the coefficient generator 95 is preferentially selected, the weight of the field interpolation signal is increased, and the occurrence of line flicker interference is suppressed.

In the embodiment described above, the case where the field interpolation is used as the interpolation method for the stationary of the subject of the video signal is shown, but the invention is not limited to this interpolation method, and any line flicker interference can be removed at the stationary time. Any structure may be used.

Further, as the interpolation method, the method of adding the weights of the intra-field interpolation and the field interpolation is shown, but the method is not limited to this, and a method of preparing a plurality of interpolation methods and switching these interpolation methods in multiple stages may be used.

EFFECTS OF THE INVENTION As described above, the present invention also includes a motion detection unit that generates a motion control signal for switching the scanning line interpolation method according to the motion of the subject of the video signal, and the inter-frame difference signal of the subject of the video signal. Further, the motion determination means for determining "moving" or "static" for each pixel and the determination result from the determination means are given, and the determination results of the target pixel and a plurality of pixels above and below it are extracted and output. Extraction means, detection means for detecting that the vertical determination result from the extraction means has a specific pattern, and the detection means for detecting movement from the motion detection means only when the detection means detects a specific pattern. By providing a motion control signal converting means for converting the motion control signal into a value for selecting the scanning line interpolation method for a still image, an incomplete telop character or the like is included in the motion detection signal. Since the generated false motion detection signal is detected and removed, it is possible to remove the line flicker interference generated by incomplete telop characters and the like.

[Brief description of drawings]

FIG. 1 is a block diagram of a motion detection circuit in the first embodiment of the present invention, FIG. 2 is a scanning line diagram for explaining the operation of the present invention, and FIG. 3 is an explanation showing a logical value of false motion detection. FIG. 4 is a block diagram of the coefficient converter in the first embodiment, FIG. 5 is a block diagram of the coefficient converter in the second embodiment, and FIG. 6 is a block diagram of the coefficient converter in the third embodiment. Configuration diagram, FIG. 7 is a configuration diagram of a conventional motion adaptive scanning line interpolation circuit, FIG. 8 is a scanning line diagram for explaining the operation of the conventional example, and FIG. 9 is a configuration diagram of a motion detection circuit in the conventional example. , FIG. 10 is a scanning line diagram for explaining line flicker interference, FIG. 11 is a scanning line diagram after progressive scan conversion, FIG. 12 is a scanning line diagram for explaining a display method of telop characters, and FIG. The figure is a scanning diagram for explaining incomplete telop characters. 91 …… 1-frame delay device, 92 …… Subtractor, 93 …… Low-pass filter, 94 …… Absolute value unit, 95 …… Coefficient generator, 97…
... delay device for timing adjustment, 101 ... comparator, 102 ... 1
Line delay device, 103 ... Inverter, 104 ... 3-input AN
D-gate, 105 ... Coefficient converter.

Claims (3)

[Claims] 1. A motion adaptive scanning line interpolation circuit used when converting an interlaced scanning video signal into a progressive scanning video signal, for switching the scanning line interpolation method according to the movement of the subject of the video signal. Motion detection means for generating a motion control signal, motion determination means for determining "moving" or "static" for each pixel based on the inter-frame difference signal of the subject of the video signal, and the determination result from the determination means Extraction means for extracting and outputting the determination result of the target pixel given and a plurality of pixels above and below it, and detection means for detecting that the determination result in the vertical direction from the extraction means has a specific pattern, Motion control signal converting means for converting the motion control signal from the motion detecting means into a value for selecting the scanning line interpolation method for a still image only when the detecting means detects a specific pattern. Motion adaptive scanning line interpolation circuit, characterized by comprising. 2. The motion adaptive scanning line interpolation circuit according to claim 1, wherein the motion control signal converting means comprises a setting means capable of arbitrarily setting the motion control signal. 3. The motion control signal converting means is a motion control signal for controlling a scanning line interpolation method closer to the processing for a still image, of the motion control signal from the motion detecting means and the motion control signal arbitrarily set. 2. The motion adaptive scanning line interpolation circuit according to claim 1, further comprising a selection means for preferentially outputting.