JPH0344172A - Contour correction circuit - Google Patents

Abstract

PURPOSE:To improve the sharpness suitable for the visual sense characteristic by adding an edge signal whose low amplitude level of a high frequency component is suppressed at a flat region to a picture signal and adding the edge signal whose negative level of the high frequency component is subject to amplitude expansion at a contour region to the picture signal. CONSTITUTION:An input signal S is inputted to an HPF circuit 1, in which the high frequency component is extracted as a signal SH and a contour extraction circuit 2 detects the contour part of a picture and its vicinity as contour information EDT from the SH signal. A nonlinear processing circuit 3 generates a signal SH' resulting from a prescribed nonlinear processing to the signal SH according to the contour information EDT, and the signal SH' is added to a signal SD subject to delay adjustment at a delay circuit 4 by an adder circuit 5 resulting in obtaining signal SENH subjected to contour correction. Then edge emphasis is applied only in the vicinity of the contour region and the picture is displayed by so-called black level supplement. Thus, the characteristic in matching with the visual sense characteristic is obtained without losing a natural sense.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an outline correction circuit that performs correction to emphasize the outline of a video signal in order to increase the sharpness of an image.

[Conventional technology]

Conventionally, in order to clearly display video signals such as television signals, television receivers and the like have performed contour correction to emphasize the contours of the video signals. As a circuit for realizing contour correction, for example, Japanese Patent Application Laid-Open No. 64-516
No. 8, JP-A-63-311875.

As described in Japanese Patent Application Laid-Open No. 63-310279, there are various forms.

[Problem to be solved by the invention]

The above-mentioned conventional technology does not take into consideration noise and the like, and has problems such as a decrease in S/N as a result of contour correction. Further, visual characteristics are not taken into consideration with respect to characteristics of contour correction, and there are problems such as loss of natural feeling due to this correction.

An object of the present invention is to provide a contour correction circuit that achieves sharpness improvement that is compatible with visual characteristics without causing a reduction in S/N due to the influence of noise or the like and without impairing the natural feeling.

[Means to solve the problem]

In order to achieve the above object, the present invention provides a function to extract the contour region of an image, and performs contour correction on the contour portion of the video signal and the region in the vicinity thereof.

In addition, in one contour correction, a function is provided to add an edge signal that has been subjected to nonlinear processing, so that a so-called darkened image is displayed.

[Effect]

According to the present invention, the function of extracting the contour area of an image allows the contour correction operation to be performed only in the vicinity of this contour area. Contour correction operates to emphasize the high frequency components of the video signal, so that edge enhancement improves sharpness in the contour portion.

On the other hand, since this contour correction operation is not performed in a flat part of one image, the noise component contained in the video signal in this area is not emphasized and remains at its original level. In other words, since edge enhancement is performed only in the vicinity of one contour area, noise in flat areas of the image that are easily noticeable does not increase;
Problems such as reduction in S/N can be resolved.

Further, as the edge signal added in the contour correction, for example, a signal obtained by performing nonlinear processing to expand the amplitude value several times with respect to a negative level signal of the high frequency component of the video signal is used. As a result, the contour-corrected video signal becomes a signal in which the undershoot portion is more emphasized, and the image is displayed with so-called black margins, so that characteristics matching the visual characteristics can be realized without impairing the natural feeling.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG. This is an overall configuration diagram of the present invention.

The input signal S is input to the HPF circuit 1, and a contour extraction path 2 extracts this high frequency component as a signal SH.
The contour part of the s-times signal image and its neighboring area are detected as contour information EDT.

The nonlinear processing circuit 3 generates a signal SH' by subjecting the signal SH to predetermined nonlinear processing in accordance with the contour information EDT. This SH signal is sent to the delay circuit 4 in the adder circuit 5.
is added to the delay-adjusted signal go to obtain the contour-corrected signal 5asH.

Hereinafter, the configuration and operation of each part will be explained in detail based on an example.

FIGS. 2 and 3 show an embodiment of the HPF circuit 1. FIG. FIG. 2 shows an example of a horizontal one-dimensional bypass filter, in which a signal sequence delayed by one pixel in a one-pixel delay circuit 6 is processed by a coefficient weighting circuit 7 to 11/4°1/2. The adder circuit 8 adds the weighted coefficients of -1/4 to extract the high frequency component SH.

On the other hand, FIG. 3 shows an example of a horizontal and vertical two-dimensional bypass filter, including a one-pixel delay circuit 6. Coefficient weighting circuit 7. A horizontal LPF circuit consisting of an adder circuit 8.

and IH delay circuit 9. Coefficient weighting circuit 7. Vertical LPF circuits composed of adder circuits 8 are connected in cascade, and the subtracter circuit 11 converts the HI signal from the signal whose delay has been adjusted by the delay circuit 10.
By subtracting , the high frequency component SH is extracted.

Next, an embodiment of the contour extraction circuit 2 is shown in FIG. 4. The signal SH previously extracted by the HPF circuit is passed through the absolute value quantization concave path 12 to generate the absolute value quantized signal DH. An example of the characteristics of the quantization circuit is shown in Section 5I14. Figure 6 shows an example of binary quantization, where Do=O when the amplitude value of the input signal Sn is between -ΔH and Δ■, and D in other areas.
Binarize as s=1. Also, Figure 6 is an example of multi-level quantization.

When the SH amplitude value is between -ΔH and ΔH, Du=O1. In other areas, DHI, L) ox, etc. depending on the amplitude value.
・L) Quantize to Hm.

Since the operation of the contour extraction circuit is the same for both binary quantization and multi-value quantization, the following operation will be explained for the case of 24Th quantization.

The smoothing circuit 13 removes noise components from the L)H signal;
Also, smoothing of the contour area is performed, specifically, smoothing processing is performed by removing isolated points, replenishing discontinuous points, etc. An example of this operation is shown in FIGS. 7(a) and 7(b).

The signal sequence DI is as shown in FIG.

In addition to the original contour components, noise components are also included. this house,
Since the noise component is generated by random noise, l
) appears as an isolated point in the H signal series. Therefore, the noise component is removed by removing this type of isolated point, as shown in FIG. 2(b). On the other hand, since areas where l is continuous correspond to contour areas, if discontinuous points occur in these areas (corresponding to O), add 1 as shown in Figure (b). Remove discontinuities and perform smoothing.

As a result, the output signal DAV of the smoothing circuit 13 is
b) is obtained as a continuous signal sequence.

Further, the contour identification circuit 14 1) generates contour information E DT in which the area before and after the contour segment is expanded, as shown in FIG. 7(c), based on the AI/signal;

Note that the absolute value quantization circuit 12. The smoothing 13° contour identification circuit 14 can be easily realized using a ROM or the like.

Next, one embodiment of the nonlinear processing circuit 3 is shown in FIG. 8, and examples of its human output characteristics are shown in FIGS. 9 and 10. These nonlinear processes are performed using the human power signal SH, @Kuo Information 1! , D'1' as RO
This can be realized by setting the address signal sequence of the M circuit 15 and setting the characteristic corresponding to this as the output signal S 11'.

Now, for the contour information E D 'r = 1, that is, for the area corresponding to the contour of the image, when the high frequency component SH is at a negative level, it is higher than when it is at a positive level, as shown in Figure 10. The gain is also set several times higher, and a negative level signal generates a signal SH' whose amplitude has been expanded several times as compared to a positive level signal.

This SH signal is added to the original signal as an edge signal.

On the other hand, for the area corresponding to the flat part of the image, Eυ'l' = Ol, as shown in FIG.
, to ΔN, the output signal SH' is suppressed to zero, for example. Note that it is desirable to set the value of ΔN to a value larger than the normally generated noise level. Therefore, in the flat area where ED'r=0, normally S, t
Since the signal is zero, the problem of increased noise in this area due to the contour correction operation does not occur.

As described above, according to the present invention, it is possible to realize contour correction that matches visual characteristics without deteriorating the S/N ratio or impairing the natural feeling.

Next, another embodiment of the present invention is shown in the first drawing. In this embodiment, the amplitude level determination circuit 16 determines the amplitude level SA of the human input signal S, and the nonlinear processing circuit 17 Then, for example, nonlinear processing with characteristics as shown in FIG. 12 is performed to generate an edge signal SH' to be added.

According to this embodiment, even finer contour correction can be realized.

Now, in the embodiments so far, signal processing is performed to detect the outline of the image based on the high frequency SH signal extracted by the H)'F circuit 1, and the circuit is simplified. The characteristics of zero recursion and contour extraction that we are aiming for are HP Fl! ! The degree of kneading, which is a characteristic of route 1, is defined.

For this reason, a configuration in which the contour area is directly detected from the human input signal S may also be considered. This embodiment is shown in FIGS. 13 and 14. It is also possible to use a technique used in pattern recognition or the like to extract the six contours. Note that since the operations and the like are almost the same as in the previous case, their explanation will be omitted.

Furthermore, by combining the contour correction circuit according to the present invention with a non-linear circuit, it is possible to improve the characteristics of the intermediate brightness level of the video signal and reproduce images with a soft feel.
It is Noh. An example of this is shown in Fig. 15.
One characteristic of 0 is shown in FIG.

In the embodiments described above, the human signal S is as follows:
Various types of UJ functions include luminance signals, three primary color signals of red, green, and blue, and color difference signals.

〔Effect of the invention〕

According to the present invention, it is possible to improve the sharpness without problems such as low S/N and without impairing the natural feeling, so it is possible to improve the image quality of television images, etc. It has a great effect on quality.

Note that the present invention is applicable to any of the current interlaced scanning video signals, video signals converted to progressive scanning video signals through signal processing such as scanning line interpolation, or progressive scanning video signals. It is clear that IJJ is capable.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the overall configuration of a circuit according to an embodiment of the present invention, FIG. @Figure 3 is a block diagram showing an example of the configuration of the HPF outlet, and Figure 4 is contour extraction I! l! The block diagrams shown in FIGS. 5 to 7 are block diagrams showing an example of the configuration of 1wI. !
Figure 8 is a block diagram showing an example of the configuration of a nonlinear processing circuit. Figures 9 and 10 are characteristic diagrams of the circuit in Figure 8, Figures 11, 13, and 14. 12 is a block diagram showing the overall configuration of a circuit according to another embodiment of the present invention, FIG. 12 is a characteristic diagram of the nonlinear processing circuit shown in FIG. 11, and FIG. A block diagram showing an example, FIG. 16, is a characteristic diagram of a nonlinear circuit. 1...HPF circuit, 2...contour extraction circuit, 3...
Nonlinear processing circuit, 4... Delay circuit, 5... Adding circuit, 6... 1 pixel delay circuit, 7... Coefficient weighting circuit, 8
... Addition circuit, 9 ... LH delay circuit, 10 ... Delay circuit, 11 ... Subtraction tV path, 12 ... Absolute value quantization circuit, 13 ... Smoothing I! ! 1 path, 14... contour identification circuit, 15... ROM circuit. 16... Amplitude level determination concave path, 1'7... Nonlinear processing circuit, 18... Contour extraction circuit, 19... Four contour correction paths according to the present invention, 20... Nonlinear path. 1st mouth 1st fast eye concave broomstick 0 concave broom 1 ■ 12th concave 3rd 8 ring fraud kichihiki 5th rule / 6 kot three-way Ikichigo SHshi Ll 79 Honkin Il@l = Yoru outline Correction enclosure 20 non-linear white feathering

Claims (1)

[Claims] 1. It has a means for extracting the high frequency component S_H of the image signal, a means for detecting the contour area and flat slope of the image signal. A contour correction circuit characterized in that an edge signal S_H' obtained by expanding the amplitude of a negative level of a high frequency component S_H is added to an image signal.