KR100460947B1 - Device for processing image signal and method thereof - Google Patents

Abstract

An image signal processing apparatus and method are disclosed. The pixel value sorter aligns each pixel value extracted for the input video signal according to a predetermined alignment order, and generates position information of the aligned pixel values. The encoder encodes the aligned pixel values and the position information. The decoding unit receives and decodes the input coded data and restores the pixel value and the position information on the pixel value. The pixel value reordering unit rearranges the pixel values based on the restored position information. As a result, each pixel value extracted with respect to the video signal is aligned according to a predetermined alignment order to increase the correlation between the pixels, so that only a partial coefficient of the low frequency region can be encoded, thereby reconstructing a high quality image even with a small amount of information. can do.

Description

Device for processing image signal and method

The present invention relates to a video signal processing apparatus and method, and more particularly, to a video signal processing method for encoding and decoding an input video signal.

Recently, in order to improve image quality, a method of encoding and processing a video signal into digital data has become popular. When encoding a video signal, it is necessary to remove the redundant data included in the video signal to reduce the total amount of data. To this end, transform coding, differential pulse code modulation (DPCM), vector quantization, and variable length coding are performed.

1 is a block diagram of an embodiment of a conventional video encoding apparatus.

Referring to FIG. 1, the image encoding apparatus includes a DCT unit 100, a zigzag scan unit 110, and a variable length encoder 120. When the video signal is input, the pixel value is selected using the motion vector. The DCT unit 100 performs a discrete cosine transform (DCT) on each of the selected pixel values. By performing the DCT transformation, the pixel values of the input time domain are transformed into the frequency domain which is an entirely different dimension and offers the possibility for energy compaction.

The zigzag scan unit 110 scans the pixel values subjected to DCT conversion in a predetermined order from the data of the low frequency region to the data of the high frequency region for energy compression. The scanned pixel values are variable length encoded and output.

Meanwhile, the encoded image signals are decoded through an inverse process of encoding, inversely transformed by inverse zig-zag scanning in a predetermined order, and decoded into a signal corresponding to the original image signal.

As described above, since the conventional video signal processing apparatus directly performs DCT conversion on the input pixels without any processing, there is a high possibility that many conversion coefficients are generated in the high frequency region according to the type of the input image. Therefore, in order to induce proper image quality, all coefficients must be coded, which causes a problem of increasing the amount of information.

In order to solve the problem that the amount of information increases, the conventional video signal processing apparatus controls the amount of information by adjusting the quantization value. However, when the quantization value is lowered, many coefficients in the high frequency region as well as the low frequency region occur. In this case, when all coefficients are encoded, the image may be restored to an image close to the original image quality, but there is a problem in that the amount of information increases.

In contrast, when the quantization value is large, the coefficients are concentrated in the low frequency region and hardly exist in the high frequency region. Therefore, the amount of information to be coded together is less than when using a low quantization value, but there is a problem that the reconstructed image quality is lower than that of the original image.

An object of the present invention is to provide a video signal processing apparatus and method for restoring a high quality original video signal with a small amount of information by increasing the correlation between pixels through alignment of pixel values of an input video signal. To provide.

1 is a block diagram of an embodiment of a conventional video encoding apparatus.

2 is a block diagram of an embodiment of an image encoding apparatus according to the present invention;

3 is a view showing an example of an alignment table showing a predetermined alignment order according to the present invention;

4 is a view showing an example of a location information table according to the present invention;

5 is a block diagram of an embodiment of an image decoding apparatus according to the present invention;

6 is a flowchart of an embodiment of a video signal processing method according to the present invention;

7 is a flowchart of an embodiment of an image decoding method according to the present invention.

Explanation of symbols on the main parts of the drawings

100: DCT unit 110: zigzag scan unit

120: variable length encoder 200: storage unit

210: pixel value alignment unit 212: alignment unit

214: information generator 220: encoder

300: decoding unit 310: pixel value reordering unit

According to an aspect of the present invention, there is provided a video encoding apparatus comprising: a pixel value sorting unit for sorting according to a predetermined sorting order extracted for an input video signal and generating the sorted position information; And an encoder which encodes the aligned pixel values and the position information.

On the other hand, the pixel value sorting unit, the alignment unit for aligning the pixel values in the sorting order; And an information generator for generating a position information table including the position information corresponding to each of the aligned pixel values.

The apparatus may further include a storage unit configured to store at least one sorting table for sorting each of the extracted pixel values according to a predetermined sorting order, wherein the pixel value sorting unit is arranged according to the sorting order of the selected sorting table. Align pixel values. Here, the sorting order is based on the size of the extracted pixel value.

In accordance with another aspect of the present invention, an image decoding apparatus includes: a decoder configured to receive and decode input coded data to restore a pixel value and position information on the pixel value; And a pixel value reordering unit for rearranging the pixel values based on the restored position information.

According to the present invention for achieving the above object, there is provided a video signal processing method comprising: aligning each pixel value extracted for an input video signal according to a predetermined alignment order; Generating position information of the aligned pixel values; And encoding the aligned pixel value and the position information.

The method may further include receiving and decoding the encoded pixel value and the position information; And rearranging the pixel values based on the decoded position information. Here, the sorting order is based on the size of the extracted pixel value.

In accordance with another aspect of the present invention, there is provided a method of decoding an image, the method comprising: receiving and decoding input coded data and restoring position information corresponding to pixel values and aligned pixel values; And rearranging the pixel values based on the restored position information.

As a result, each pixel value extracted with respect to the input image signal may be aligned according to a predetermined alignment order to increase the correlation between the pixels, and thus the coefficients converted through the DCT transform may be distributed in the low frequency region. Therefore, it is possible to encode only a part of coefficients in the low frequency region, and to decode a high quality image even with a small amount of information.

Hereinafter, with reference to the accompanying drawings will be described in detail the present invention.

2 is a block diagram of an embodiment of an image encoding apparatus according to the present invention.

Referring to FIG. 2, the image encoding apparatus includes a storage unit 200, a pixel value sorter 210, and an encoder 220. The pixel value sorter 210 sorts each pixel value extracted for the input video signal according to a predetermined sorting order. Here, the sorting order is determined based on the size of the extracted pixel value.

Meanwhile, the pixel value sorter 210 includes an sorter 212 and an information generator 214. The alignment unit 212 aligns each pixel value extracted for the input video signal according to a predetermined alignment order. The information generator 214 generates a location information table that includes location information corresponding to each of the aligned pixel values. The encoder 220 scans and encodes the position information and the aligned pixel values according to a predetermined alignment order.

In general, when a video signal is received from an external device, the received video signal is divided into block units having a predetermined size (for example, 4x4, 8x8, 16x16, ...). For each block, the motion vector between the blocks is detected by predicting the motion from the previous frame and the current frame. When the motion vector is detected, the pixel value corresponding to each block is extracted by the detected motion vector.

In this case, when the discrete cosine transform is performed with the pixel values corresponding to each extracted block out of alignment, many coefficients may appear in the high frequency region according to the type of the input image. This leads to an increase in the amount of information during encoding.

Discrete Cosine Transform (DCT) means DCT transforming an input image signal, thereby converting pixel values in a spatial domain into a signal in a frequency domain. This means that when DCT conversion of the pixel value in the space domain and this signal, the conversion coefficients in the frequency domain are much higher in energy concentration efficiency than the signal in the space domain. The energy of this transformed conversion coefficient is mainly concentrated in the low frequency region.

Therefore, it is preferable to perform DCT conversion by sorting in descending or ascending order based on the size of the extracted pixel value. That is, when DCT conversion of pixel values aligned based on the size of pixel values, the coefficients in the low frequency region have a significantly larger value, and the coefficients of the high frequency components have smaller values, so that only low frequency components of the DCT coefficients can be encoded.

The pixel value sorter 210 sorts each extracted pixel value according to a predetermined sorting order. Meanwhile, in order to sort pixel values according to a predetermined sorting order, one of a plurality of previously created sorting tables may be selected and sorted. The storage unit 200 stores at least one sorting table for sorting each extracted pixel value according to a predetermined sorting order. In this case, the pixel value sorter 210 sorts the pixel values according to the sorting order of the selected sorting table.

3 is a view showing an example of an alignment table showing a predetermined alignment order according to the present invention.

Referring to FIG. 3, the sorting order may be arranged based on the size of the pixel value, and the sorting order may be sorted in the sorting order such as sorting tables (a), (b), and (c), and may be reversed. The pixel value sorter 210 selects one of a plurality of sorting tables stored in the storage 200. The pixel value sorter 210 sorts pixel values according to the sorting order of the selected sorting table. The alignment may remove high frequency components existing between the pixels. In addition, the coefficients of the high frequency region may be reduced in the DCT conversion of the aligned pixel values.

Although the embodiment has described the sorting table sorted in ascending order based on the size of the pixel value as shown in FIG. 3, the present invention is not limited thereto and may be sorted in any manner based on the size of the pixel value. Do.

When the pixel values are aligned according to a predetermined sorting order, the information generator 214 generates a position information table including position information corresponding to each aligned pixel value. This position information is necessary for decoding the encoded video signal.

4 is a diagram illustrating an example of a location information table according to the present invention.

Referring to Fig. 4, information on the original position corresponding to each pixel value is recorded in the position information table before being aligned in a predetermined alignment order. For example, the position of pixel value 1 after alignment is P11, but the original position corresponding to pixel value 1 before alignment is P24. In this case, the positional information of the pixel value stored in the position P11 of the positional information table is P24. That is, information about the original position of the pixel value is recorded in the position information table before being aligned.

The encoder 220 scans and encodes the aligned pixel values according to a predetermined alignment order, and also encodes the position information.

5 is a block diagram of an embodiment of an image decoding apparatus according to the present invention.

Referring to FIG. 5, the image decoding apparatus includes a decoder 300 and a pixel value reordering unit 310. The decoder 300 receives and decodes the input coded data to restore the pixel value and the position information corresponding to the aligned pixel value from the image data. The pixel value reordering unit 310 rearranges the pixel values based on the restored position information.

When receiving encoded data from an external device, the encoded data includes pixel values of the video signal and position information corresponding to each of the aligned pixel values. Therefore, after receiving the quantized digital signal, it is dequantized by performing a process opposite to that of the quantization unit (not shown). In addition, the inverse quantized signal is inversely DCT transformed to inversely transform the signal in the spatial domain.

The decoder 300 receives and decodes the encoded data and restores the position information corresponding to the pixel value and the aligned pixel value. When each pixel value and position information for the image signal are restored, the pixel value reordering unit 310 rearranges the pixel values based on the restored position information. Here, the position information means information about the original position corresponding to each pixel value before being aligned with respect to the aligned pixel values.

6 is a flowchart of an embodiment of a video signal processing method according to the present invention.

Referring to FIG. 6, when a video signal is received from an external device, the received video signal is divided into block units having a predetermined size (for example, 4x4, 8x8, 16x16, ...), and motion is predicted and detected for each block. The pixel value is extracted from one motion vector. The alignment unit 212 aligns the extracted pixel values according to a predetermined alignment order (S600). In this case, the sorting order is based on the size of the extracted pixel value. For example, the image may be arranged in ascending or descending order based on the size of the pixel value.

Alternatively, one of a plurality of previously created sorting tables is selected to sort the extracted pixel values according to a predetermined sorting order. The sorting unit 212 sorts pixel values according to the sorting order of the selected sorting table. When the pixel values are aligned according to a predetermined sorting order, the information generator 214 generates a position information table including position information corresponding to each of the aligned pixel values (S610). In the position information table, information about the original position corresponding to each pixel value is recorded before being aligned.

When the pixel values are aligned according to a predetermined alignment order and position information on the aligned pixel values is generated, the encoder 220 adds the generated position information to the aligned pixel values and encodes the signal (S620). The decoder 300 decodes the encoded pixel value and the position information to restore the position information corresponding to the pixel value and the aligned pixel value (S630). The pixel value reordering unit 310 rearranges pixel values based on the restored position information (S640). Here, the position information means information on the original position corresponding to each pixel value before the aligned pixel values are aligned.

7 is a flowchart of an embodiment of an image decoding method according to the present invention.

Referring to FIG. 7, when receiving encoded data from an external device (S700), the encoded data includes pixel values of the image signal and position information corresponding to each of the aligned pixel values. Therefore, after receiving the quantized digital signal, it is dequantized by performing a process opposite to that of the quantization unit (not shown). In addition, the inverse quantized signal is inversely DCT transformed to inversely transform the signal in the spatial domain.

The decoder 300 decodes the encoded data to restore pixel values and position information corresponding to each pixel value from the data (S710). When the pixel value and the position information corresponding to the pixel value are restored through decoding, the pixel value reordering unit 310 rearranges the pixel values based on the restored position information (S720). The position information means information about the original position corresponding to each pixel value before being aligned with respect to the aligned pixel values.

According to the image encoding / decoding apparatus and method according to the present invention, coefficients obtained through DCT conversion are added to a low frequency region by increasing the correlation between pixels by aligning each pixel value extracted for a video signal according to a predetermined alignment order. It can be distributed a lot. Therefore, it is possible to encode only a part of coefficients in the low frequency region, and to restore a high quality image even with a small amount of information.

Although the present invention has been described in detail through the representative embodiments, those skilled in the art to which the present invention pertains can make various modifications without departing from the scope of the present invention with respect to the embodiments described above. Will understand. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims below and equivalents thereof.

Claims (9)

A storage unit for storing at least one sorting table; The input image signal is divided into predetermined block units, and pixel values existing in the divided blocks are sorted based on the sorting order of the selected sorting table among the at least one sorting table, and for each sorted pixel value. A pixel value sorter for generating position information; And And an encoder which encodes the aligned pixel values and the position information. The method of claim 1, The pixel value sorter, An alignment unit to align the pixel values based on an alignment order of the selected alignment table; And And an information generator generating position information on each of the aligned pixel values and generating a position information table based on the generated position information. delete The method of claim 2, And the sorting order is based on the size of the extracted pixel value. The pixel values existing in the blocks divided into block units corresponding to the video signal are arranged based on the alignment order of the alignment table, and the encoded data is obtained by adding the position information for each pixel value to the aligned pixel values. In the video decoding apparatus to decode, A decoder which receives and decodes the encoded data to restore each pixel value and position information on each pixel value; And And a pixel value rearranging unit for rearranging the pixel values based on the reconstructed position information. Dividing an input video signal into predetermined block units and extracting pixel values existing in the divided blocks; Sorting the extracted pixel values based on a sorting order of a selected sorting table; Generating location information for each of the aligned pixel values; And And encoding the sorted pixel values and the position information. The method of claim 6, Restoring pixel values and position information corresponding to each pixel value by decoding the encoded data; And rearranging the pixel values based on the reconstructed position information. The method of claim 7, wherein And the sorting order is based on the size of the extracted pixel value. delete