JPH046961A - Noise reduction circuit for video signal - Google Patents

Abstract

PURPOSE:To reduce large input noise by using a spatial filter in real time when movement exists in an input video signal to increase the resolution timewise and applying timewise processing when no movement exists to improve the resolution in the spatial direction. CONSTITUTION:A mean value and a median or the like for 9 picture elements around a prescribed picture element are obtained from an input video signal V1 from a terminal 10 by a 2-dimension spatial filter 13. A difference between a filter output Vn and a signal Vn-1 stored precedingly in a memory 14 is detected by a movement detection circuit 15. When the absolute value of the difference is large (movement is in existence), a changeover circuit 11 is thrown to the position of a filter 13, and an adding mean circuit 24 applies arithmetic mean to a data with less number of times to the utmost and outputs the result. When the absolute value of the difference is small (movement is less), the changeover circuit 11 is thrown to the position of an input terminal 10, and the signal V1 is fed to the adding mean circuit 24, in which adding mean as many as possible number of times if implemented and the result is outputted. Thus, the changeover of the changeover circuit 11 and the sum of the adding mean circuit 24 are controlled in response to a variable of a movement detection circuit 17 thereby reducing noise in the video input signal effectively.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a video signal noise reduction circuit.

"Prior Art" Conventionally, there have been circuits of this type as shown in FIGS. 3 and 4 (Japanese Patent Publication No. 17580/1983).

Of these, the circuit in Figure 3 sends the video signal x1 applied to the input terminal (1) and the video signal y1-□ output from the frame memory (2) to the motion detection circuit (3), where both The difference between the two is calculated and outputs 1- and K regarding the movement are output. When the movement is large, K=O, and the motion detection circuit (3) outputs 1- = 1 and = O, and sends them to the addition circuit (6) via the multiplication circuits (4) and (5), respectively. . As a result, the input video signal remains unchanged at the output terminal (
7) is output.

When the movement is small, let = 0.5 and 1- K = 0.
5°K=0.5 is output, and the average value is outputted as +(Xl''Yt-x) through the multiplication circuits (4) and (5) and the addition circuit (6).

Next, the circuit in FIG. 4 is a circuit (8) that subtracts the difference signal between the input video signal
and adds it to the fixed memory (9) as an address signal. In this fixed memory (9), when the data corresponding to the address, that is, the absolute value of the difference signal is greater than a certain value, =O is recorded, indicating that there is movement at that address.
Further, when the absolute value of the difference signal is close to 0, a predetermined value K is recorded at that address, assuming that there is no movement. When there is movement, since the output xi-K (-yi-,) of the adder circuit (6) is 1, the input video signal xt is output as is from the output terminal (7). Also,
When there is little movement, for example, if = 1, then xl-K(
The average value is output as shown in Xt-yt-1).

"Problem to be solved by the invention" The circuit in Figure 3 uses a motion detection circuit (3) and a multiplication circuit (4).
(5) must be provided separately, and these multiplier circuits (
4) The configuration of the control circuit for (5) and variables is complicated.

Moreover, there were also problems with circuit technology.

The circuit of FIG. 4 solves the drawbacks of the circuit of FIG. 3 and has a simple circuit configuration. However, there is a problem in FIG. 4 that when input noise is large, this may be used as motion detection and the noise component may be output without being reduced.

An object of the present invention is to obtain a circuit that uses a real-time spatial filter to increase the resolution in the temporal direction when there is movement, and performs processing in the temporal direction when there is no movement to increase the resolution in the spatial direction. It is something.

"Means for Solving the Problems" The present invention includes a two-dimensional spatial filter that reduces noise in a two-dimensional space of an input video signal, a switching circuit that switches between the input video signal and an output signal of the two-dimensional spatial filter, comprising a motion detection circuit that detects motion based on the output signal of the two-dimensional spatial filter and the input video signal and outputs a variable, and an averaging circuit provided on the output side of the switching circuit, The switching of the switching circuit and the addition amount of the averaging circuit are controlled according to variables of the motion detection circuit.

"Effect" The input video signal is filtered by a two-dimensional spatial filter, e.g.
The average value, intermediate value (the fifth value when the values of nine pixels are arranged in order of magnitude), etc. of the pixels are obtained.

A difference signal between this two-dimensional spatial filter data and previously stored data is detected. When the absolute value of the difference is large (when there is movement), the switching circuit is switched to the two-dimensional spatial filter side, and the data is averaged as few times as possible in the averaging circuit and output. When the absolute value of the difference is small (when there is no movement), the switching circuit is switched to the input terminal side, and the input video signal is sent as is to the averaging circuit, where it is averaged as many times as possible and output.

Therefore, when there is movement, a real-time two-dimensional spatial filter is used to increase the resolution in the temporal direction, and when there is no movement, processing is performed in the temporal direction using an averaging circuit to increase the resolution in the spatial direction.

"Example" Hereinafter, an example of the present invention will be described based on FIGS. 1 and 2.

In FIG. 1 showing the first embodiment of the present invention, (10) is a video signal input terminal, and this input terminal (10) is coupled to one terminal (12) of a switching circuit (11) and two It is coupled to a dimensional space filter (13).

This two-dimensional spatial filter (13) is connected to the frame memory (1
4) is coupled to a motion detection circuit (17) consisting of a subtraction circuit (15) and a fixed ROM (16), and is also coupled to the other terminal (18) of the switching circuit (11). A common terminal (19) of this switching circuit (11) is connected to an output terminal ( 25).

In the above configuration, the video signal v1 input to the input terminal (10) is sent to the two-dimensional spatial filter (13). Here, if the number of pixels is, for example, 480 vertically by 640 horizontally, the two-dimensional spatial filter (13) outputs the average value, intermediate value, etc. of nine pixels surrounding a predetermined pixel.

This averaged data■. is stored in the frame memory (14) of the motion detection circuit (17). Then, at the time of the next frame signal, the subtraction circuit (15) inputs the frame memory (
14) and the previous data Vn-1 and the two-dimensional spatial filter (1
3) ■. Which difference signal is detected. When the absolute value of this difference is large, there is movement, and at this time the variable from the fixed ROM (16) is set to =1.

When this = 1, the switching circuit (11) is switched to the other terminal (18) by the output signal of the fixed ROM (16), the two-dimensional spatial filter (13) is used to increase the resolution in the time direction, and the averaging circuit ( 24). The subtracting circuit (20) of the averaging circuit (24) performs subtraction with the data in the frame memory (23), and sends the result to the multiple circuit (21). Multiplier circuit (
21), since =1 is sent from the fixed ROM (16), the output of the subtraction circuit (20) is directly sent to the addition circuit (
22), is added to the data in the frame memory (23), and is output from the output terminal (25).

When the absolute value of the difference signal in the subtraction circuit (15) of the motion detection circuit (17) is small, there is no motion, and the variable from the fixed ROM (16) is K+. The switching circuit (11) is switched to one terminal (12) by the 8-input signal of the fixed ROM (16) when K is +, and the video signal of the input terminal (10) is directly input to the averaging circuit (24
). This averaging circuit (24) takes the averaging with the data in the frame memory (23) as described above, but at this time, since the multiplier circuit (21) is a is increasing.

Next, a second embodiment of the present invention will be explained based on FIG. 2.

In this example, the motion detection circuit (17) uses the output of the two-dimensional spatial filter (13) and the corrected data from the frame memory (23) of the averaging circuit (24).
The motion detection circuit (17) does not have a frame memory. The other structure and operation are substantially the same as those in FIG. 1.

"Effects of the Invention" Since the present invention is configured as described above, the resolution in the time direction is increased by the real time two-dimensional spatial filter where there is movement in the input video signal, and the arithmetic averaging circuit is used where there is no movement. Processing is performed in the temporal direction to increase the resolution in the spatial direction. Therefore, even if large noise is input to the input video signal, it will be reduced and will not be output.

Further, the circuit configuration is simple and can be provided at low cost.

[Brief explanation of the drawing]

FIG. 1 shows a first circuit of a video signal noise reduction circuit according to the present invention.
FIG. 2 is a block diagram showing a second embodiment of the present invention, and FIGS. 3 and 4 are block diagrams showing different examples of conventional circuits. (10)...Input terminal, (11)...Switching circuit, (
12)...terminal, (13)...two-dimensional spatial filter, (14)...frame memory, (15)...subtraction circuit, (16)...fixed ROM, (17)...・Motion detection circuit, (18)...terminal, (19)...terminal, (20)...subtraction circuit, (21)...multiplier circuit,
(22)...Addition circuit, (23)...Frame memory, (24)...Additional averaging circuit, (25)...Output terminal.

Claims (3)

[Claims] (1) a two-dimensional spatial filter that reduces noise in the two-dimensional space of an input video signal; a switching circuit that switches between the input video signal and the output signal of the two-dimensional spatial filter;
A motion detection circuit that detects motion based on the output signal of the dimensional space filter and the input video signal and outputs a variable, and an averaging circuit provided on the output side of the switching circuit, A video signal noise reduction circuit, characterized in that the switching of the switching circuit and the addition amount of the averaging circuit are controlled in accordance with variables of the detection circuit. (2) The motion detection circuit detects the difference between the output data of the two-dimensional spatial filter and the output data of the frame memory one frame before, and stores the variable corresponding to this detected value in the fixed ROM.
2. The video signal noise reduction circuit according to claim 1, wherein the video signal is output from a video signal. (3) The motion detection circuit detects the difference between the output data of the two-dimensional spatial filter and the corrected output data of the frame memory of the averaging circuit, and fixes the variable corresponding to this detected value.
2. The video signal noise reduction circuit according to claim 1, wherein the video signal is output from an OM.