JPH10276433A - Block noise eliminating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decode data subjected to image compression and to reduce noise caused at the time of reproducing an image. SOLUTION: Filters 12 and 13 and an adder 14 take out a high-pass component, a block noise extracting part 15, block noise memory 21 and a square noise detector 22 detect what is square noise and does not have correlativity in a field direction and obtain position information of block noise. A vertical correction circuit 32 and a horizontal correction circuit 33 perform filtering to smooth a noise position based on the position information of the noise block.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a block noise eliminator for reducing the occurrence of so-called block noise when a signal subjected to discrete cosine transform processing (hereinafter referred to as DCT processing) for each pixel block is restored. About.

[0002]

2. Description of the Related Art In the field of image compression technology, the following processing is generally performed. That is, an original signal is divided into pixel blocks, and each block is subjected to DCT processing in a DCT processor to convert signal components from a time axis to a frequency axis. The converted value is called a DCT coefficient, and the DCT coefficient is input to a quantizer,
It is quantized and compressed by an appropriately selected quantization table (corresponding to the activity of the image). Next, the quantized coefficients quantized by the quantizer are encoded by a Huffman encoder and recorded on a transmission line or a recording medium.

On the decoding side, Huffman-decoded data is input to an inverse quantizer and inversely quantized. The quantized coefficients are then subjected to inverse DCT processing in an inverse DCT processor to obtain the original pixel block. , And the block image is arranged and reconstructed into a frame image.

[0004]

By the way, when a frame image is constructed as described above, there is a problem that boundaries between adjacent pixel blocks are visually noticeable. This phenomenon is called block noise. The generation of this noise is D
The point is that the CT processing is divided into blocks, the relation between blocks is cut off, and signal processing is performed. Image compression is also applied to recorded information on digital video disks (DVDs).

It is known that, in DVD reproduction, the above-mentioned block noise often occurs at points where the frequency component of an image is high or in a gradation pattern. SUMMARY OF THE INVENTION It is an object of the present invention to provide a block noise eliminator capable of reducing block noise generated when decoding image-compressed data and reproducing the image.

[0006]

According to the present invention, there is provided a high frequency component extracting means for extracting a high frequency component of image data, and a square image is detected from the high frequency component obtained from the high frequency component extracting means. A square image detection unit, a noise determination unit that identifies an original image and block noise from the square image detected by the square image detection unit, and determines block noise, and the noise determination unit Correction means for reducing the block noise determined by the above by filtering processing using the preceding and following pixel data. By the above means, block noise can be reduced without deteriorating image quality.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. In FIG. 1A, digital image data subjected to inverse DCT processing is input to an input terminal 11. This image data is input to the vertical filter 12, and a vertical high-frequency component is extracted. Furthermore, the image data is input to the horizontal filter 13 to extract a horizontal high-frequency component. The vertical high frequency component and the horizontal high frequency component are added to an adder 14.
Are added to form block edge information, which is input to the block noise extraction unit 15.

The block noise extracting section 15 is composed of a field switch (or a frame switch) 16, edge component memories 17, 18, and a subtractor 19 for subtracting the outputs of the edge component memories 17, 18 to obtain an absolute value. ing.

The field switch 16 is switched for each field, and supplies block edge information to the edge component memory 17 in the first field and supplies block edge information to the edge component memory 18 in the second field. Here, when the information of the corresponding addresses in the edge component memories 17 and 18 is subtracted, components having a correlation in the field direction are deleted.

This is because the edge information having a correlation in the field (frame) direction is an original video, and is not subject to noise. That is, the present invention utilizes the fact that block noise has no correlation in the temporal direction, and the original image (having a square edge) has correlation in the field direction.

The output of the block noise extraction unit 15 is temporarily stored in a block noise memory 21 to detect whether or not square noise exists therein. The square noise is determined by the square noise detector 22 accessing the block noise memory 21 at high speed and inspecting the arrangement of the noise information. If there is a square noise, this noise information is stored in the square noise memory 23 as it is. The square noise is stored as position information, as shown in FIG.

The contents of the square noise memory 23 are horizontal /
The signal is input to the vertical edge processing unit 24, and the edge in the horizontal direction and the edge in the vertical direction are determined. On the other hand, the image data at the input terminal 11 is input to a time adjustment circuit 31, where the time adjustment is performed, and the image data is input to a vertical correction circuit 32 to perform vertical filtering of a noise portion. The signal is input to the filter 33, and filtering of the noise portion in the horizontal direction is performed. The information for performing the filtering in the noise portion is given from the horizontal / vertical noise processing unit 24 described above.

The overall timing and control are executed by the system control unit 35. System control unit 3
5 is input with reference to a horizontal synchronizing pulse and a vertical synchronizing pulse.

FIG. 2A shows a specific configuration of the vertical fill 12. By obtaining the difference between the delayed signal delayed by the line memory 12a and the direct signal, a vertical high frequency component can be obtained. The absolute value processing unit for this output is omitted.

FIG. 2B shows a specific configuration of the horizontal fill 13. The horizontal high frequency component can be obtained by obtaining the difference between the delayed signal delayed by the unit delay memory 13a and the direct signal. The absolute value processing unit for this output is omitted.

FIG. 3A shows the configuration of the vertical correction circuit 32. The image data of the input terminal 321 is input to the line memory 322, is delayed, and is input to the line memory 323. The image data of the input terminal 321 is input to the coefficient unit 322. Further, outputs of the line memories 322 and 323 are output to coefficient units 324 and 32, respectively.
5 is input. The outputs of the coefficient units 323, 324, and 325 are added by the adder 326 and supplied to one input terminal of the switch 327. The output of the line memory 322 is directly supplied to the other input terminal of the switch 327.

Thus, if the switch 327 is directly switched to the signal side, a signal without filtering can be obtained, and if it is switched to the signal on the adder 326 side, a vertically filtered signal can be obtained. That is, the horizontal line of the block noise can be reduced.

FIG. 3B shows a horizontal correction circuit 33. The image data at the input terminal 331 is input to the unit delay memory 332, delayed by one pixel, and further input continuously to the unit delay memories 333 and 334. The image data of the input terminal 331 is input to the coefficient unit 335. Further, outputs of the unit delay memories 332, 333, 334 are input to coefficient units 336, 337, 338, respectively. The outputs of the coefficient units 335 to 338 are added to an adder 339.
And is supplied to one input terminal of the switch 340. The output of the memory 332 is directly supplied to the other input terminal of the switch 340.

By switching the switch 340 directly to the signal side, a signal without filtering can be obtained, and by switching to the signal on the adder 339 side, a horizontally filtered signal can be obtained. That is, the vertical lines of the block noise can be reduced.

The present invention is not limited to the above embodiment. FIG. 4 shows another embodiment of the present invention. Image data is input to the input terminal 401, and the image data is input to the high-frequency component extraction unit 402. Here, the high-frequency components (image contours) in the vertical and horizontal directions are extracted, and the positional information thereof is sent to the rectangular image detection unit 403 together with the image data. The square detection unit 403 detects the outline of the square using the field memory unit 404. This detection includes not only the block noise but also the outline of the original square image. Next, the image data and the outline information of the square are sent to the original square image detection unit 405. The square image detection unit 504 uses the field memory 406 to detect an image having a correlation between fields. The correlated image is determined to be the original square image. Block noise utilizes a characteristic that it does not appear at the same coordinates on the screen over a plurality of fields.

Next, the information of the original square image and the information of all the square images are input to the block noise determination unit 407 together with the image data. The block noise determination unit 407 determines an actual block noise position using the field memory unit 408, and holds the position information. Next, the block noise position information and the image data are input to a block noise vertical contour correcting unit 409, and noise removal processing from the image data is performed.
In this process, the image data is corrected so that the portion where the value of the data arranged in the horizontal direction changes sharply using left and right (horizontal) image data so as to have a gentle change.

The corrected image data is input to a block noise horizontal contour correction unit 410, and noise removal processing from the image data is performed. In this process, the part where the value of the data arranged in the vertical direction changes sharply,
The image data is corrected so as to make a gradual change by using the upper and lower (vertical) image data.

As a result, image data from which block noise has been removed can be obtained at the output terminal 411. FIG.
Is an example of an image encoding / decoding device useful by applying the present invention.

That is, the original image data is divided into blocks of 8 * 8 pixels to form a block scanning row, which is input to the DCT processor 501 and subjected to DCT processing. The DCT coefficient obtained by the DCT processing is quantized by the quantizer 502 using the coefficient given from the quantization table 503. The quantized coefficients output from the quantizer 502 are input to the Huffman encoder 504. The quantized coefficients are Huffman-encoded with the coefficients given from the encoding table 505.
The coded data is transmitted or recorded on a transmission path or a recording medium 507 with a parameter added thereto by a parameter adding unit 506.

The parameters include information on the used quantization table and information on the coding table. Information from the transmission path or the recording medium 507 is transmitted to a parameter extraction unit 511.
Is input to The parameters extracted by the parameter extraction unit 511 are stored in the quantization table 503 and the encoding table 5.
05 and used as information for decoding.
In the drawing, the quantization table 503, the encoding table 50
5 shows both the transmitting side and the receiving side in common, but is actually independent.

The coded data from which the parameters have been removed by the parameter extracting section 511 are input to a Huffman decoder 512, which performs Huffman decoding, and then performs an inverse quantizer 513.
And is inversely quantized. The DCT count obtained by the inverse quantization is input to the inverse DCT processor 514 and restored as an original block image.

This block image corresponds to the circuit 515 of the present invention.
And the block noise is removed.
Although the above-described block noise was likely to occur at a position of an image having a fast movement, the noise was effectively removed by the circuit of the present invention.

The present invention is effective for removing block noise from image data subjected to inverse DCT processing as described above, and is particularly applicable to processing of image data reproduced from an MPEG decoder of a video disk reproducing apparatus. Good. This is because on the production side of a video disc, there is a difference in quality between the discs, and there is a possibility that some disc noises are generated.

The present invention is not limited to the above embodiment, but can be realized by software.
Further, it is needless to say that realization by a DSP (digital signal processor) using a plurality of microprocessors is also possible.

[0030]

As described above, according to the present invention,
It is possible to reduce the block noise generated when decoding the image-compressed data and reproducing the image.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of the present invention.

FIG. 2 is a diagram illustrating a vertical filter and a horizontal filter of FIG. 1;

FIG. 3 is a diagram showing a vertical correction circuit and a horizontal correction circuit of FIG. 1;

FIG. 4 is a diagram showing another embodiment of the present invention.

FIG. 5 is a diagram showing an image compression encoder and a decoder.

[Explanation of symbols]

 12 vertical filter, 13 horizontal filter, 14 adder, 15 block noise extractor, 16 switch, 17, 18 edge component memory, 19 subtractor, 21 block noise memory, 22 square noise Detector, 23: square noise memory, 24: horizontal / vertical noise processing unit, 32: vertical correction circuit, 33: horizontal correction circuit

Claims (4)

[Claims] A high-frequency component extracting means for extracting a high-frequency component of the image data; a square image detecting means for detecting a square image from the high-frequency component obtained from the high-frequency component extracting means; A noise determination unit that identifies an original image and block noise from the square image detected by the square image detection unit and determines block noise; And a correction means for reducing by a filtering process using the pixel data. 2. The noise determination unit determines a square image having a correlation in the field or frame direction as an original image and discriminates between the original image and block noise in order to distinguish between the original image and block noise. The block noise eliminator according to claim 1, wherein the block noise is determined in such a case. 3. The block noise eliminator according to claim 2, wherein the high-frequency component extracting means extracts horizontal and vertical contour components of the image. 4. A rectangular component extracting means for receiving image data subjected to inverse DCT processing and extracting a rectangular component from high-frequency components included in the image data; Among them, it is determined whether there is a correlation between frames (or fields),
A noise discriminating means for judging the contour of the image if there is a correlation, and a block noise if there is no correlativity, and a filter means for filtering the image data of the location where the block noise occurs based on the judgment result of the noise discriminating means. It is provided with. A block noise removing apparatus comprising the above means.