JPH0282874A - Noise cancel circuit - Google Patents

Abstract

PURPOSE:To realize noise cancellation taking the visual characteristic into account by using a cancellation quantity of noise component through the use of a lightness function. CONSTITUTION:A cancellation quantity control means 5 controls the amplitude of noise extracted according to the coefficient obtained by a coefficient calculation means 4. A noise component extraction means 2 is so constituted to utilize 1H correlation of a video signal similar to conventional 1H noise scan, the detection of a background level is implemented by obtaining a video signal level around a picture element subject to cancellation. Then a noise component whose amplitude is controlled in response to the level of YN, YD it given to a subtraction circuit 6 and noise cancellation is executed in response to the visual characteristic. Thus, when the noise is visually remarkably, the cancellation quantity is increased and when the noise is not visually remarkably, the cancellation quantity is decreased, then the loss of resolution is avoided.

Description

[Detailed description of the invention] (b) Industrial application fields The present invention relates to a noise canceling circuit.

(b) Conventional technology Japanese Patent Application Laid-open No. 130075 (HO4N5/21)
2 shows a noise canceling circuit used in a camera-integrated VTR. This circuit controls the amount of noise cancellation based on the output of the detection circuit that detects the output signal of the CCD, so that when the amount of imaging light is small, noise cancellation is performed with a large amount of noise cancellation, and the appropriate amount of imaging light is maintained. Problems to be Solved by the Invention In this case, the amount of noise cancellation can be suppressed, and the problem to be solved by the invention.By the way, in the prior art described in L, the amount of noise cancellation is controlled by detecting the CCD output, but the actual control is It is only a two-step operation, and it does not take human visual characteristics into consideration. Therefore, the present invention provides a noise canceling circuit that takes human visual characteristics into consideration.

(d) Means for Solving the Problems In the present invention, in a noise canceling circuit that extracts a noise component from an input video signal and subtracts this noise component from the input video signal to cancel the noise, the amount of cancellation of the noise component is It is controlled using a brightness function.

(e) Effect: In other words, since the brightness function based on the human visual characteristics is used, noise cancellation can be realized taking visual characteristics into consideration.

One of the brightness functions that determines human visual characteristics is the following (Judd's brightness function).

Here, R is the reflectance of the optotype, Ro is the reflectance of the background, and ■ is the brightness. This characteristic is illustrated in FIG. 4.

What is clear from Figure 4 is that the brightness (human perception) is not proportional to the reflectance of the optotype; in general, when the reflectance of the optotype is small, the change in brightness is large, and when the reflectance is large, the brightness changes. change is smaller (slope of brightness function). Furthermore, even if the reflectance of the optotype (the part to which the eye is focused, ie, the point of fixation) is the same, the brightness will change if the reflectance of the background is different. The darker the background, the brighter the visual target appears to be.

The above explanation is about the reflectance of objects, but V
The above relationship is considered to remain the same even when the reproduced image of the TR is displayed on a monitor TV.

Next, let's consider the visual conspicuousness of noise based on this brightness function. The noise is considered to be something superimposed on the original video signal. In other words, it is considered to be a minute amplitude fluctuation in the original video signal that has nothing to do with the information.

Considering this and the characteristics shown in Figure 4, even if the amplitude of the superimposed noise is the same, the corresponding brightness will change depending on the signal level at the point of interest and the signal level of the background (around the point of interest). It can be seen that the changes are different. For example, when the background is dark (for curve (a), Ro = 0.1) and when the point of gaze is bright, there is almost no change in brightness due to noise, but
If the gaze point is dark, the brightness change will be quite large. Therefore, the conspicuousness of noise differs.

Since the change in brightness is proportional to the slope of the graph of the brightness function in FIG. 4, the conspicuousness of noise can be expressed by the following equation obtained by differentiating equation (1). In other words, when formula (2) is graphed, it looks like Figure 5.

Note that the above relationship coincides with the empirical fact that "the brighter the screen, the less noticeable noise is."

From the above, by controlling the amount of noise cancellation using equation (2), it is possible to achieve visually optimal noise cancellation and to suppress deterioration of resolution.

(f) Example Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the basic configuration. In FIG. 1, (1) is the input terminal for the reproduced luminance signal, (2) is the noise component extraction means, (3) is the background level detection means, (
4) is a coefficient calculation means for calculating the equation (2); (5) is a cancellation amount control means for controlling the amplitude of the extracted noise according to the obtained coefficient; and (6) is a subtraction means.

The noise component extracting means may have a configuration that utilizes the IH (H is the horizontal synchronization period) correlation of the video signal, similar to normal IH noise canceling. This treats uncorrelated signals as noise.

The background level can be detected by using a field memory to find the video signal level around the pixel where the cancellation operation is to be performed, or by approximating the average level before the -field to the background level.

For further simplification, a configuration using a recursive filter using a 1H delay line can also be considered. Figure 2 shows
This is what happens when you use this method.

In Figure 2, (7), (8), and (9) are adder circuits, (6
) (10) is a subtraction circuit, (5) is a multiplication circuit, (11) is a division circuit, (12) is an l/2 multiplication circuit, (13) is an IH delay line, (14) and (15) are squaring circuits. , (16) is a 1/2 doubler circuit, and (17) is a J-mitter.

In Figure 2, the non-delayed luminance signal is transferred to YN, recursive filter (18) (adder circuit (7),
2), having an IH delay line (13)) is denoted as YD. Then, the subtraction circuit (lO) calculates Y, -Y, and extracts a signal without IH correlation (which can be considered to be noise).

Furthermore, the output YD of the recursive filter (18) is also used as a signal representing the background level. Coefficient calculation means (4
), perform the calculation of equation (2) to obtain I(yo”+YI))
/ (Yl)+Y)1)" outputs the result of I. This output is divided by a predetermined value (multiplied by 17) and is supplied to the multiplier circuit (5) via the limiter (17). It operates to control the amplitude of the noise component.

The reason why the limiter (17) is provided is to limit it to a predetermined value since the value shown in FIG. 5 becomes quite large.

Then, the noise component whose width is controlled according to the levels of Y9 and YD is given to the subtraction circuit (6), and noise cancellation is executed according to the visual characteristics. Generally, when the level is low, the coefficient becomes large, and when the level is high, the coefficient becomes small. Therefore, when the noise is visually noticeable, the amount of cancellation is made large, and when it is not noticeable, the amount of cancellation is made small, so that the resolution is not impaired.

In the configuration shown in Figure 2, the output of the recursive filter (18) is used as a delayed luminance signal for noise component extraction, but as shown in Figure 3, another IH delay line (19) is used for noise extraction. It may be provided.

(G) Effects of the Invention -E As described above, according to the present invention, the amount of noise cancellation is controlled according to the characteristics of vision, and noise cancellation that is optimal for vision can be performed with less risk of deterioration of resolution. , the effect is great.

[Brief explanation of the drawing]

1, 2, and 3 are block diagrams showing embodiments of the present invention, FIG. 4 is a characteristic diagram of a brightness function, and FIG. 5 is a characteristic diagram showing brightness change characteristics. (2)... Noise component extraction, (3)... Background level detection, (4)... Coefficient calculation, (5)... Cancellation amount control.

Claims (1)

[Claims] (1) In a noise canceling circuit that extracts a noise component from an input video signal and subtracts it from the input video signal to cancel the noise, the amount of cancellation of the noise component is controlled using a brightness function. cancellation circuit.