JPH04145774A - Noise reducer - Google Patents

Abstract

PURPOSE:To suppress the generation of an after image and a block distortion by shifting a block in the horizontal direction and applying a sequential conversion and a reverse conversion. CONSTITUTION:An Hadamard converter 4 constitutes a block by a field difference signal obtained by subtracting an output of a field memory 2 from an input video signal by a subtracter 1, and a signal obtained by delaying it by 1H by a 1H delay element 3, and performs an Hadamard conversion. A non- linear processing part 5 executes a non-linear processing of a converted signal, and a subtracter 7 subtracts an output of an Hadamard reverse converter 6 from an input video signal. In such a way, by applying a sequential conversion and a reverse conversion by moving the block by one sample each, an after image and a block distortion can be suppressed.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention relates to a noise reduction device (noise reducer) installed in a home VTR or the like.

BACKGROUND OF THE INVENTION In recent years, video tape recorders have been equipped with noise reducers to reduce noise during playback. A signal processing method of a conventional noise reducer will be explained below.

A noise reducer takes advantage of the fact that while video signals have a strong correlation between fields or frames, noise is uncorrelated, and by taking a weighted average with the signal from one field or frame before. , to reduce noise.

The general configuration of a field cyclic noise reducer is expressed by the following equation.

yn= (1-k) X, + k 3'1.11"X
n −k (X n −Y n −+ ), where 0
≦to≦l Xo: Input signal Vn-+: Field memory output signal y1: Output signal The configuration of the above equation is shown in a block diagram as shown in FIG. The first subtracter 21 subtracts the output signal of the field memory 22 from the input video signal to obtain a field difference signal.

The nonlinear processing unit 23 calculates k in the above equation and the field difference signal (Xn
3'n-+), and an example of its input/output characteristics is shown in FIG. For the input field difference signal, when the field difference signal is small (close to a still image), k (=b/a) shown in the figure is set to 1.
Set the value close to , increase the weight of the signal one field before,
As the field difference signal becomes larger (the movement becomes larger), k is brought closer to O, and nonlinear processing is performed to increase the weight of the current signal. The output signal of this nonlinear processing section 23 is
The subtracter 24 subtracts the signal from the input video signal to reduce noise. From a different perspective, the output of the nonlinear processing unit 23 can also be regarded as extracted noise.

Problems to be Solved by the Invention However, in the conventional noise reducer as described above, since the above processing is performed individually for each pixel, it is difficult to distinguish between signal and noise for small movements, resulting in afterimages. The problem was that it was easy to use.

In response, noise reducers have been proposed that apply Hadamard transform, which is an orthogonal transform, to field or frame difference signals to extract image features and make noise separation relatively easy (for example, in television Academic journal V-ol, 37°No, 12. 1983
).

The principle of this type of noise reducer will be briefly described using FIG. 4(A) and FIG. 6. FIG. 6 shows a block diagram of a noise reducer using a two-dimensional Hadamard transform. FIG. 4(A) shows the state of the data in the portions indicated by ■ to ■ in FIG. 6. The clock is the basic clock signal of this system. The first subtractor 31 is
Field or frame memory 32 from input video signal
subtracts the output of and outputs a field or frame difference signal (■). IH (H: horizontal scanning period) delay element 39
delays the output of the first subtractor 31 by IH (■). The Hadamard transformer 33 forms a block consisting of a plurality of sample points as shown in FIG. 2 from the output ■ of the first subtractor 31 and the output ■ of the IH delay element 39, and forms a block consisting of a plurality of sample points as shown in FIG.
Perform a 7-Damar transformation as follows (■). Nonlinear processing f
! E34 performs nonlinear processing on each transform component as shown in Figure 3 to extract noise components, and then generates S, T, ... (■)
shall be. The Hadamard inverse transformer 35 inversely transforms the output (■) of the nonlinear processing unit 34, and converts it into extracted noise (■, ■) corresponding to each sample point of the original block. Of the outputs of the Hadamard inverse transformer 35, ■ is direct, ■ is the IH delay element 3
A selection is output from the selector switch 42 via 7. The changeover switch 42 connects the Hadamard inverse converter 35 and IH every IH.
The output of the H delay element 37 is selected and output to the second subtracter 36. A second subtracter 36 subtracts the extracted noise from the input signal to perform noise reduction.

By applying Hadamard transform to each sample point within a block, it is thought that the signal is concentrated in some transformed components and the noise is evenly distributed to each component, so the difference between the signal and noise values becomes large. It becomes easier to differentiate. Therefore, it is possible to configure a noise reducer with less occurrence of afterimages.

However, in the noise reducer described above, if blocks are taken so that sample points do not overlap, data discontinuity, so-called block distortion, tends to occur at the boundaries of the blocks. Explaining this using FIG. 4A, the block boundaries indicated by broken lines in the figure are easily noticeable because there is no overlapping data between blocks.

Note that if the data is duplicated during conversion, the inversely converted data will also be duplicated, and processing such as averaging the data will be required, increasing the circuit size.

Also, as shown in FIG. 6, the Hadamard transformer 33
IH delay elements 37 and 39 are required before and after Hadamard inverse transformer 35, respectively.

Moreover, in order to match the timing of the extracted noise obtained by the Hadamard inverse transformer 35 and the input video signal,
Since an IH delay element 38 is also required before the second subtracter 36, a total of three elements are required for the entire system, which also increases the circuit scale.

The present invention solves the above-mentioned conventional problems, and aims to provide a noise reducer that causes less afterimage generation and suppresses block distortion. Another purpose is to minimize the increase in circuit scale.

Means for Solving the Problems To achieve this object, the noise reduction device of the present invention comprises:
A delay means for delaying a video signal, a first subtracter for subtracting an output signal of the delay means from an input video signal, and a signal block composed of a plurality of elements from the output signal of the first subtractor. a signal block forming means for forming a signal block, a conversion means for converting the signal block into a feature component, a nonlinear processing section for nonlinearly processing the output signal of the conversion means, and one of the signal blocks from the output signal of the nonlinear processing section.
The apparatus includes an inverse transformer that obtains only the elements, and a second subtracter that subtracts the output of the inverse transformer from the input video signal to provide an input signal to the delay means.

Effect of the present invention With the above-described configuration, by shifting the block one pixel at a time in the horizontal direction and sequentially performing transformation and inverse transformation,
Make block distortion less noticeable.

Furthermore, it is possible to reduce the number of delay elements, which has the effect of reducing hardware.

EXAMPLE Hereinafter, a noise reducer which is an example of the present invention will be described with reference to the drawings.

FIG. 1 shows a block diagram of a noise reducer that is an embodiment of the present invention. In addition, Fig. 4 (B) shows the first
This figure shows the state of the data in the parts ■ to ■ in the figure.

The first subtractor 1 subtracts the input video signal from the field memory 2.
The field difference signal obtained by subtracting the output of is input to the IH delay element 3 and the Hadamard transformer 4 (■).

In the Hadamard transformer 4, a block is constructed as shown in FIG. 2 from the field difference signal (■) output from the first subtractor 1 and the signal (■) which is IH delayed by the turtle IH delay element 3, and a 2X
A four-dimensional two-dimensional Hadamard transformation is applied. The transformation formula is expressed as F '' Ha・X@H,...(1) where F is the Hadamard transform component, and X is the block N H21H4 shown in FIG.
It is a Hadamard matrix of the following order and the fourth order, and is expressed by the following equation.

The Hadamard-transformed signals Fsi~F, 3 are each Fag
: Two-dimensional low frequency component Fa+~F13: Horizontal direction high frequency component F, @Two-dimensional vertical direction high frequency component~F+3: Represents oblique direction component. The field difference signal concentrates on at least one of them. For example, if a portion with almost no movement between fields is converted, the F2 acid component will be concentrated, and if there is movement in the lateral direction, the Fs+ to F'si components will be concentrated. In contrast, noise has no correlation between fields and is divided equally into all components, making it easy to distinguish between signals and noise.

The converted signal F (■) is subjected to nonlinear processing in the nonlinear processing section 5 according to the input/output characteristics shown in FIG. 3, and the output becomes extracted noise (■).

Since the output signal of the nonlinear processing unit 5 is Hadamard-transformed noise, it is necessary to perform Hadamard inverse transformation to return it to noise on the time axis. The Hadamard inverse transform of order 2×4 is (1
) is expressed by formula (2) using the same F, X, H2, and H4 as in formula.

X” (1/8)”H2”F IIH-・=(2) Of these, X+@, that is, X+a=1/8・(Faa+Fs++Fa2+Fa*
-F+a-F++ F+2-F++) is obtained by the Hadamard inverse transformer 6 (■), and is subtracted from the input video signal by the second subtractor 7. The block is moved horizontally one sample at a time, and sequential transformation and inverse transformation are performed. Looking at Figure 4 (B), each block of 8 samples is S+L.
...and perform nonlinear processing to obtain S, T,
...and... S, T.

. . . Only samples corresponding to Xl. are found.

If this is done, the boundaries between blocks will be one pixel at a time, making them less noticeable. Furthermore, Hadamard inverse transformer 6
Both the output and the input video signal can be input to the second subtracter 7 without passing through the IH delay element.

In this embodiment, only X+a was obtained during the Hadamard inverse transformation, but the same effect can be expected even if X + + to X + a are obtained instead. Further, during the Hadamard transform, a block is formed of four consecutive samples in the horizontal direction and two consecutive samples in the vertical direction, but the samples do not need to be consecutive.

Effects of the Invention As described above, the present invention includes a delay means for delaying a video signal, a first subtracter for subtracting an output signal of the delay means from an input video signal, and a subtracter for subtracting an output signal of the first subtractor from the output signal of the first subtractor. ,
A signal block forming means for forming a signal block composed of a plurality of elements, a converting means for converting the signal block into a feature component, a nonlinear processing section for nonlinearly processing an output signal of the converting means, and the nonlinear processing section. inverse transformation means for obtaining only one element of the signal block from the output signal of;
a second subtracter that subtracts the output of the inverse conversion means from the input video signal to provide an input signal to the delay means, the output signal of the second subtractor or the output signal of the delay means; By using the output as an output, it is possible to realize a noise reducer with less generation of afterimages and block distortion and with minimal increase in circuit scale, and its practical effects are significant.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the noise reducer in the embodiment of the present invention, Fig. 2 is a schematic diagram showing the input signal block of Hadamard transform, Fig. 3 is a characteristic diagram showing the input/output characteristics of the nonlinear processing section, and Fig. 4 is a timing diagram showing the state of data in each part of Figure 1 and Figure 1,
FIG. 5 is a block diagram of a conventional noise reducer, and FIG. 6 is a block diagram of a conventional noise reducer using Hadamard transform. DESCRIPTION OF SYMBOLS 1... First subtractor, 2... Field memory, 3... IH delay element (signal block forming means), 4
... Hadamard transformer (conversion means), 5 ... nonlinear processing section, 6 ... Hadamard inverse transformer (inverse transformer stage), 7 ... second subtractor.

Claims (4)

[Claims] (1) Consisting of a plurality of elements: a delay means for delaying a video signal; a first subtracter for subtracting the output signal of the delay means from the input video signal; and an output signal of the first subtractor. Signal block forming means for forming a signal block; Conversion means for converting the signal block into a feature component; A nonlinear processing section for performing nonlinear processing on the output signal of the conversion means; From the output signal of the nonlinear processing section, the signal comprising an inverse transformer for obtaining only one element of a block, and a second subtracter that subtracts the output of the inverse transformer from the input video signal to provide an input signal to the delay means, and the second subtractor A noise reduction device whose output is an output signal of the delay means or an output signal of the delay means. (2) The noise reduction device according to claim 1, wherein the delay amount of the delay means is one frame, one field, or n lines (n: natural number). (3) The signal block forming means is an n-stage shift register (n
: natural number). The noise reduction device according to claim 1. (4) The noise reduction device according to claim 1, wherein the signal block forming means includes an m-stage shift register and n one-line delay elements (m, n: natural numbers).