JP2000224586A - Method and device for coding video image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate a video compressed signal by which a plurality of MPEG signals with a plurality of spatial resolutions can be produced with less processinga amount by generating an orthogonal transform coefficient where part of the orthogonal transform coefficients with respect to an input image is replaced with an orthogonal transform coefficient with respect to a reduced input image. SOLUTION: A video reduction means 203 halves longitudinal/lateral resolutions of a received non-compression video image to reduce the input video image. An orthogonal transform means 202 applies orthogonal transform to the input non-compression video image acquired from an input means 101 and an orthogonal transform means 204 apples orthogonal transform to a reduced video image acquired from a video reduction means 203 respectively to generate a DCT coefficient. A coefficient synthesis means 205 synthesizes the DCT coefficients generated by the orthogonal transform means 202, 204. A quantization means 206 quantizes the synthesized DCT coefficient to generate a quantization coefficient. A structural coding means 208 adds information required for variable length coding to a header to generate a final MPEG stream.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video compression apparatus and a video compression method for compressing digital video signals, and more particularly to a video compression encoding system capable of extracting video signals of a plurality of resolutions from one compressed signal.

[0002]

2. Description of the Related Art There are many formats for compressing digital video signals, and MPEG (Motion P
The Picture Experts Group is one of the most popular formats. In recent years, it has become easy to encode and decode MPEG format video signals on a computer.
Many computer systems for transmitting and reproducing PEG format video via a network have been put to practical use.

In the video transmission system using such a network, it is necessary to change the data amount of the compressed video to be used depending on the transfer speed of the network for transmitting the video and the performance of the terminal for receiving and reproducing the video.
For example, when a 100 Mbit / s local area network is used for video transmission, 10 Mbit / s compressed video can be transmitted in real time.
Only a video signal of about 0 kilobits per second can be sent, and there is a 100-fold difference in the data amount between the two.

As described above, a method called scalability has been standardized in MPEG in order to generate video signals having the same contents but different data amounts. The scalability in MPEG is based on the required accuracy in the time direction, the spatial direction, and the image quality (signal-to-noise ratio). This is a method for transmitting and reproducing the information up to.

[0005] However, there are very few MPEG decoders that support the scalability of MPEG, and most of the decoders currently in use cannot reproduce, and thus have a problem in versatility.

[0006] MPE without scalability
Japanese Patent Laid-Open No. 4-3 discloses a technique for easily generating a compressed video compliant with G (hereinafter, general MPEG) with two types of spatial resolutions.
No. 58485. This technology encodes a low-resolution video, in which the image size of the input video is reduced by half both vertically and horizontally, in a general MPEG format, decodes the coded video, and decodes the image size of the input video. The signal of the difference between the video enlarged to the above and the input video is newly encoded in a general MPEG format or the like.
From a video signal encoded by this technique, a general MP
When generating an EG stream, a low-resolution video stream that has already been encoded can be used as it is. However, when generating a high-resolution video, the low-resolution stream and the high-resolution stream are each decoded once, the decoded low-resolution image is enlarged, the decoded high-resolution image is added, and the decoded image is added again. There is a problem that a process of encoding to general MPEG is required, and a large amount of processing is required.

[0007]

SUMMARY OF THE INVENTION An object of the present invention is to solve the above problems. That is, MPEG signals of a plurality of spatial resolutions without using scalability are
It is an object to generate a video compression signal that can be generated with a small processing amount.

[0008]

In order to solve the above-mentioned problems, the present invention firstly reduces the size of the input image and the size of the input image when compressing and encoding the input image using orthogonal transform. The reduced input image is subjected to orthogonal transformation in block units to generate an orthogonal transformation coefficient, and the orthogonal transformation coefficient obtained by replacing a part of the orthogonal transformation coefficient for the input image with the orthogonal transformation coefficient for the reduced input image is compressed. In the above-mentioned method, among the above methods, an orthogonal transform coefficient for a reduced input image is generated from an orthogonal transform coefficient for an input image by a coefficient transformation process (a reduced input process is performed by an operation in an orthogonally transformed base axis region). To generate an orthogonal transform coefficient for the image).

Further, control means, input means for inputting a compressed video signal including orthogonal transform coefficients for input images of a plurality of resolutions, and code conversion means for generating a low-resolution compressed video signal from the input compressed video signal And an output unit for individually outputting compressed video signals of a plurality of resolutions, wherein the code conversion unit deletes and changes a part of the variable length code of the compressed video signal input by the input unit. The low-resolution compressed video signal is generated by adding a code.

[0010] This makes it possible to extract video signals of a plurality of resolutions from one compressed signal without having to perform an enormous amount of calculation such as decoding, difference (hierarchical resolution information) adjustment, and encoding as in the prior art. It is possible to provide a video compression coding system.

Second, the input image and the reduced input image are orthogonally transformed in block units, and among the orthogonal transformation coefficients generated by the orthogonal transformation, a part of the orthogonal transformation coefficients of the input image is converted into a part of the reduced input image. When generating a compressed video signal for reproducing the reduced input image from a multi-resolution compressed video signal obtained by performing variable length encoding on the orthogonal transform coefficient replaced with a part of the orthogonal transform coefficient, the multi-resolution compressed video signal A compressed video signal for reproducing the reduced input image is generated by deleting a component other than the orthogonal transform coefficient component of the reduced input image included, and thereby, in addition to the first effect, more compressed video signals are generated. Even if the orthogonal transform coefficients corresponding to the low resolution are combined, the influence on the decoded video of the high resolution can be reduced.

Third, when the request, generation and analysis of a compressed video signal having a plurality of resolutions are performed in a distributed manner on a network, the apparatus according to the first and second aspects of the present invention transmits the compressed video signal to the network. Can be transmitted from the transmitting side, and a plurality of resolutions can be simultaneously provided from the transmitting side only by sending the data amount for one stream (without increasing the network addition).

[0013]

Embodiments of the present invention will be described below with reference to the drawings. The present invention is not limited to these embodiments at all, and can be implemented in various modes without departing from the gist of the invention.

(First Embodiment) In the first embodiment, a video encoding device for generating a video compression signal capable of extracting an MPEG stream of a plurality of resolutions at high speed, and a video encoding device A description will be given of a resolution converter for extracting MPEG streams of a plurality of resolutions from a generated compressed signal at a high speed. Note that the present apparatus does not perform inter-frame compression, but performs only intra-frame compression (it is mainly performed on intra frames as conversion targets).

First, a description will be given of a video encoding apparatus for generating a video compression signal capable of extracting MPEG streams of a plurality of resolutions at a high speed. The configuration of this device
This is explained using Fig. 1. FIG. 1 is a functional block diagram of the present apparatus. As shown in FIG. 1, the present apparatus includes an overall control unit 10, an input unit 100, an encoding unit 200, and an output unit 300.

The overall control unit 10 controls the entire apparatus, that is, the input unit 100, the encoding unit 200, and the output unit 300. The input unit 100 includes input means 101 for inputting an uncompressed digital video signal, and an input video buffer 102 for temporarily storing the digital video signal input from the input means 101.
The output unit 300 includes an output unit 301 that outputs the compressed signal of the MPEG standard generated by the encoding unit 200.

The encoding unit 200 includes an encoding control unit 201 for controlling the operation of each unit in the encoding unit 200, an orthogonal transformation unit 202 for applying DCT to the video signal from the input video buffer 102 in block units, Video reduction means 203 for reducing the resolution of the video signal from the input video buffer to half both vertically and horizontally; orthogonal transformation means 204 for applying DCT to the video signal reduced by the video reduction means 203 in block units; 202
And a coefficient synthesizing unit 205 that synthesizes the DCT coefficients generated by the orthogonal transform unit 204 into one DCT coefficient,
A quantizing unit 206 for quantizing the DCT coefficients generated by the coefficient synthesizing unit 205; a variable-length encoding unit 207 for scanning the coefficients quantized by the quantizing unit 206 to perform variable-length encoding; Encoding means 208 which adds information such as various headers necessary for the MPEG standard to the code generated by the encoding means 207 to generate a compressed signal of the MPEG standard.
And Here, the orthogonal transform means 202 and the orthogonal transform means 204 have the same function except that the resolution of the input video signal is different.

Next, the operation of the present apparatus will be described. Figure
2 is a flowchart showing the flow of the overall operation of the present apparatus. As shown in FIG. 2, the present apparatus performs processing according to the following flow.

Step 1100: If there is an unprocessed video to be input, Step 1200: the input means 101 inputs one frame of uncompressed video signal, and Step 1300: the video reduction means 203 Step 1400: Decreasing the input uncompressed video obtained by the orthogonal transformation means 202 from the input means 101 and reducing the input uncompressed video obtained by the orthogonal transformation means 202 from the video reduction means 203 The image is orthogonally transformed,
Generate DCT coefficients. Step 1500: Next, the coefficient synthesizing means 205 synthesizes the DCT coefficients generated by the orthogonal transform means 202 and the orthogonal transform means 204. Step 1600: The quantizing means 206 modifies the DCT coefficients synthesized by the coefficient synthesizing means 205. Is quantized to generate a quantized coefficient.

Step 1700: and the variable length coding means 20
7 performs variable-length coding on the quantized coefficients generated by the quantizing unit 206. Step 1800: The structure coding unit 208 adds necessary information to the variable-length code as a header to generate a final MPEG stream. Step 1900: The output unit 301 outputs the MPEG stream generated by the structure encoding unit 208, and continues processing the next frame again from the beginning. (Step 1100
Back to).

Here, the coefficient synthesizing process (step 1500) will be described in detail with reference to the drawings. FIG. 3 shows the luminance signal of the original image and the luminance signal of the reduced image, and their DCTs.
FIG. 9 is a diagram showing a relationship between coefficients and DCT coefficients synthesized by a coefficient synthesis process. An area corresponding to one macroblock in an uncompressed video signal (original image) input from the input unit 101, that is, an area of 16 pixels in height and 16 pixels in width, corresponds to an image (reduced image) reduced by the image reduction unit 203. 1
It contains an image of the same content as the DCT block area, that is, the area of 8 pixels vertically and 8 pixels horizontally.

The coefficient synthesizing process includes images having the same contents.
The DCT coefficient of the low-frequency portion of any one of the DCT blocks (the upper left block in FIG. 3) of one macroblock of the original image is converted to the DCT of the corresponding reduced image.
This is performed by replacing the DCT coefficient with the low frequency part of the block. In FIG. 3C, the darker portions are replaced.

Here, the low frequency part of the DCT block is
After quantizing the DCT coefficients, it indicates an arbitrary number of coefficients from the next AC component excluding the DC component in the scan order when performing variable length coding. Also, as for the color difference signal, only one DCT block exists in one macro block, so that the only difference is that the color difference signals of four macro blocks of the original image correspond to the color difference signals of one macro block of the reduced image. The coefficient synthesis processing is performed in the same manner as the luminance signal.

Further specific examples are shown in FIGS. 4 and 5 and will be described. In the examples shown in FIGS. 4 and 5, nine DCT coefficients are used as the low frequency part. FIG. 4 shows (A) an original image before synthesis, (B) a reduced image before synthesis, and (C) a 1DCT after synthesis.
The block is shown. One square in each block represents one DCT coefficient. FIG. 5 shows these DCT coefficients arranged in the order of scanning at the time of variable-length encoding.

The DCT coefficients thus synthesized can be restored to image information by general inverse DCT processing.
The restored image is different from the original image or the reduced image. If the number of low-frequency coefficients used for synthesis is large,
The restored image is an image closer to the reduced image, and when the number of coefficients in the low-frequency portion is small, the restored image is an image closer to the original image. However, since the combined DCT coefficient uses the DC component of the DCT block of the original image as the DC component, the image has much of the properties of the original image.

In the present embodiment, one DCT block is subjected to coefficient synthesis processing for every four DCT blocks.
It is assumed that the order is fixed, such as the last block in the coding order among the four DCT blocks. In addition, the number of DCT coefficients in the low-frequency portion whose coefficients are operated in the coefficient synthesis processing is fixed in advance.

The MPEG stream including the above-described synthesizing process and output from the video encoding apparatus conforms to the MPEG standard, and can be decoded by a general MPEG decoder. Most of the decoded video is a video of the original image itself.
One DCT block in T blocks contains not only the original image,
Some of them include information on reduced images.

In the present embodiment, D for low resolution is used.
To generate the CT coefficients, reduce the input image and then
Although the CT is performed, if the apparatus having the configuration shown in FIG. 6 is used, the coefficient transforming unit 209 performs the coefficient transforming process on the DCT coefficient for high resolution, so that the DCT coefficient of low resolution can be directly obtained. . Here, the coefficient deformation processing is DC
This is a process of thinning out the image resolution in the T region.

That is, by this processing, one DCT block for low resolution is directly generated from four DCT blocks adjacent to each other by two blocks vertically and horizontally. If this stream is used, an MPEG stream having a plurality of resolutions can be generated at high speed by the resolution conversion device described below.

The processing amount required by this apparatus is a general M
Compared with the processing amount of the PEG encoder, only the reduction of the input image and the calculation of DCT for the reduced input image are added, and the calculation amount does not greatly increase.

Next, a description will be given of a resolution converter for extracting an MPEG stream having a plurality of resolutions at a high speed from the MPEG stream generated by the video encoding apparatus. FIG. 7 shows the configuration of the present apparatus. FIG. 7 is a functional block diagram of the present apparatus. As shown in FIG. 7, the present apparatus comprises a control means 410 for controlling the entire apparatus, an input means 420 for inputting an MPEG stream generated by the video encoding apparatus described above, and a low-speed Code conversion means 43 for generating a resolution MPEG stream
0 and an output unit 440 that outputs an MPEG stream of a plurality of resolutions.

Next, the operation of the present apparatus will be described. FIG. 8 is a flowchart showing the flow of the entire operation of the present apparatus. As shown in FIG. 8, the present apparatus performs processing in the following flow. Step 2100: The input means 420 inputs the MPEG stream generated by the video encoding device. Step 2200: The code conversion means 430
The G stream is subjected to code conversion processing to generate an MPEG stream whose vertical and horizontal resolutions are halved. Step 2300: Finally, the output unit 440 converts the MPEG stream input from the input unit 420 to a high-level image. The MPEG stream generated by the transcoding means 430 is output as a low-resolution MPEG stream.

Here, the code conversion processing (step
2200) will be described in detail. As described with reference to FIGS. 3 to 5 in the present embodiment, the MPEG stream generated by the video encoding device according to the present embodiment includes four DCT blocks (in the case of a luminance signal, one DC block). Of the DCT blocks) in one DCT block,
Information on DCT coefficients of low-resolution video is included. The main processing performed in the code conversion processing of the present apparatus is to extract only the information of the low-resolution DCT coefficient and reconstruct the MPEG stream.

FIG. 9 shows the flow of the code conversion process. As shown in FIG. 9, the present apparatus performs a code conversion process in the following flow. Step 2210: If there is an unprocessed part in the input stream, Step 2220: Input a part of the input stream. Step 2230: If the input portion is MPEG header information, Step 2231: Generate low-resolution header information corresponding to the input header information. For example, if the input information is information shared by a high-resolution stream and a low-resolution stream, such as a macroblock layer and a slice layer, the information input as low-resolution header information is used. If the value of the low-resolution stream is different from the value of the high-resolution stream, such as the information of the picture layer, the value is replaced for the low-resolution stream to generate header information.

Step 2232: The generated header information is output as an output stream. Step 2240: If the input portion is a coefficient of a DCT block, and if the DCT block does not include information for low-resolution video, Step 2241: Obtain the value of the coefficient of the DC component and Enter the new part. (Continue processing from step 2210.) Step 2250: If the DCT block includes information for a low-resolution video, four DCT blocks for high resolution corresponding to the low-resolution video included in the DCT block Calculate the average of the DC component of, and use it as the value of the DC component of the low-resolution block. Step 2260: And at the end of the information for low resolution video
EOB (EndOf Block) code is added. Step 2260: The subsequent information for high-resolution video is deleted and output as an output stream.

FIG. 10 is a diagram showing the relationship between (A) a part of an input stream and (B) a low-resolution stream generated from the part. As shown in FIG. 10, the low-resolution stream has DCT coefficients of only the low-frequency part, and the high-frequency part outside the low-frequency part has an EOB in front of it.
Are inserted, the values are all 0. Therefore, the video obtained when this low-frequency stream is composited is a video without deleted high-frequency components, that is, a video without a sharp change, but is composed of only low-resolution information without high-resolution information. ing.

By the processing described above, a high-resolution MP
This means that a low-resolution MPEG stream can be generated from the EG stream. Conventionally, DCT coefficients differ between high-resolution information and low-resolution information. To generate a low-resolution MPEG stream from a high-resolution MPEG stream, complete decoding is performed once, and then the image size is reduced. Had to be re-encoded to MPEG. On the other hand, according to the present embodiment, a low-resolution MPEG stream is generated by only replacing header information, adding an EOB in units of DCT blocks, and deleting unnecessary codes to a high-resolution MPEG stream. it can. Therefore, it is not necessary to perform a process with a large amount of operation such as DCT, inverse DCT, variable length coding, and variable length decoding, and the resolution can be converted at high speed, and the practical effect is large.

(Second Embodiment) In a second embodiment, a video encoding device for generating a video compression signal capable of extracting MPEG streams of a plurality of resolutions at high speed, and a video encoding device A description will be given of a resolution converter for extracting MPEG streams of a plurality of resolutions from a generated compressed signal at a high speed.

The difference between the apparatus of the present embodiment and the apparatus of the first embodiment is that the video encoding apparatus converts the low-resolution DCT coefficients into all the high-resolution DCT coefficients by coefficient synthesis processing. The point is that they are dispersed and combined. Further, the resolution conversion apparatus is also different from the first embodiment only in that the resolution conversion apparatus performs a code conversion process corresponding to the MPEG stream generated by the video coding apparatus.
This is different from the first embodiment.

First, the video encoding device will be described.
The configuration of this apparatus is the same as the configuration of the video encoding apparatus shown in FIG. 1 in the first embodiment. Further, a large processing flow of the present apparatus is the same as the processing flow shown in FIG. 2 in the first embodiment. This apparatus differs from the first embodiment in the coefficient synthesis processing (step 1500). The coefficient synthesis processing of the present apparatus will be described with reference to the drawings.

FIG. 11 is a diagram showing the relationship between the luminance signal of the original image and the luminance signal of the reduced image, their DCT coefficients, and the DCT coefficients synthesized by the coefficient synthesis processing. As shown in FIG. 11, unlike the first embodiment, the DCT block after synthesis includes low-resolution DCT coefficients in the DCT coefficients of all four blocks. Also, the number of DCT coefficients for low resolution included in each DCT block is the number of DCT coefficients after synthesis according to the first embodiment.
The number can be reduced as compared with the T block. Therefore, the MP for high resolution using the DCT block after synthesis is used.
When the EG stream is coded, an image closer to the original image is restored as compared with the first embodiment. It should be noted that this video encoding apparatus can also directly generate low-resolution DCT coefficients in the DCT domain by adopting the configuration shown in FIG.

Next, the resolution converter will be described.
The configuration of this device is the same as that shown in FIG. 7 in the first embodiment. Further, the flow of the entire processing is the same as the flow shown in FIG. This apparatus differs from the first embodiment in the code conversion processing 2200. FIG. 12 shows a flow of the code conversion process of the present apparatus. As shown in FIG. 12, the present apparatus performs a code conversion process in the following flow.

Step 2210: If there is an unprocessed portion in the input stream, Step 2220: Input a part of the input stream. Step 2230: If the input portion is MPEG header information, Step 2231: Generate low-resolution header information corresponding to the input header information.

For example, if the input information is information shared by a high-resolution stream and a low-resolution stream, such as a macroblock layer and a slice layer, the information input as low-resolution header information is used. If the value of the low-resolution stream is different from the value of the high-resolution stream, such as the information of the picture layer, the header information is generated by replacing the value for the low-resolution stream. Step 2232: The generated header information is output as an output stream. Step 2240: If the input part is a coefficient of a DCT block, obtain the value of the coefficient of the DC component of the DCT block. Step 2250: Change the sign of the part representing the low-resolution DCT coefficient included in the DCT block. get. Step 2260: If the information of four blocks necessary for forming the DCT coefficient for low resolution has not been obtained, a new part is input from the input stream and the processing is continued. (Continue processing from step 2210.) Step 2270: If the information of the four blocks has already been obtained, calculate the average of the coefficient values of the four DC components already obtained, and use them as the coefficient values of the DC component for low resolution. Step 2280: Connect the codes representing the low-resolution DCT coefficients that have already been obtained so that the DCT coefficients in the code are arranged in the scanning order.
The data is output as a PEG stream, and a new portion is input from the input stream (the processing is continued from step 2210).

As is clear from the above description, according to the present embodiment, as in the first embodiment, low-resolution processing can be performed without performing a computationally intensive process such as DCT or variable-length coding. MPEG stream can be generated at high speed.

Further, information of the DCT coefficient for low resolution is
Since the DCT coefficients are dispersedly included in the four DCT blocks, the effect on the decoded image of the high-resolution MPEG stream can be reduced even if more DCT coefficients for low resolution are combined as compared with the first embodiment. And its practical effect is great.

(Third Embodiment) In a third embodiment, a variable resolution video is provided, which allows a user to select a resolution and receive an MPEG stream video transmitted via a network. The system will be described.

First, the configuration of the present system will be described. FIG. 13 is a block diagram showing the configuration of the present system. As shown in FIG. 13, the system includes a video generation unit 10
00, a resolution conversion unit 3000, and a video reception unit 4000 are connected and configured via a network 2000.

A video generation unit 1000 has a function of generating and transmitting a base MPEG stream of a video used in the present system, and obtains a video from a video device 1100 that provides a video source and a video source 1100. And multi-resolution MPEG
A video encoding device 1200 that generates an MPEG stream that can easily generate a stream, and an M code generated by the video encoding device.
A video transmission device 1300 for transmitting the PEG stream to the resolution conversion unit 3000.

The resolution conversion unit 3000 receives and processes a user request sent from the video reception unit 4000, processes the request, and controls all the devices of the resolution conversion unit. The video receiving device 3200 that receives the stream and the M received by the video receiving device 3200
A resolution conversion device that converts a PEG stream into an MPEG stream having the resolution specified by the request processing device 3100
3300 and MPEG converted by resolution converter 3300
Video transmitter 34 that transmits a stream to video receiver 4000
00.

The video receiving unit 4000 receives a request from the user, requests the selected resolution to the resolution converting unit 3000,
A video receiving / reproducing device 4100 for receiving, decoding, and displaying the MPEG stream transmitted from the resolution conversion unit 3000 is provided.

In the present system, the video encoding device 1200
The resolution conversion device 3300 uses the video encoding device and the resolution conversion device described in the first and second embodiments. However, the resolution converter described in the first and second embodiments simultaneously outputs a high-resolution MPEG stream and a low-resolution MPEG stream.
The resolution conversion device 3300 used in the present system can output either one of the high-resolution or low-resolution MPEG streams according to the instruction of the request processing device 3100.

Next, the operation of the present system will be described. FIG.
FIG. 4 is a flowchart showing a flow of processing from generation to reproduction of a video in the present system. As shown in FIG. 14, the present system provides an image in the following flow.

Step 5100: First, the video signal generated by the video equipment 1100 in the video generation unit 1000 is converted to a video encoding device.
Step 1200: Generate MPEG stream. Step 5300: The video transmitting device 1300 is connected to the video encoding device 12
00 to the resolution conversion unit 3000
Send to. Step 5400: On the other hand, when the user requests video reproduction in the video reception / reproduction unit 4000 and selects a desired resolution, the video reception / reproduction device 4100 transmits the user's request to the resolution conversion unit 3000.
Step 5500: The resolution conversion unit 3000 requests the request processing device 3100
Receives the request transmitted from the video reception and playback unit 4000,
The operation of each device of the resolution conversion unit 3000 is instructed. Step 5600: First, the video receiving device 3200 receives the instructed video from the video generating unit 1000, and the resolution converting device 3300
Pass to. Step 5700: The resolution conversion device 3300 is the video receiving device 32
The MPEG stream of the video passed from 00 is converted into an MPEG stream of the resolution specified by the request processing device 3100 and passed to the video transmission device 3400. Step 5800: The video transmission device 3400 transmits the MPEG stream passed from the resolution conversion device 3300 to the video reception / reproduction unit 4000 specified by the request processing device. Step 5900: The video receiving / playing device 4100 is a resolution conversion unit
The MPEG stream transmitted from 3000 is received, decoded and displayed.

With the above processing, a plurality of resolution images are provided.

According to the present embodiment, even when there are a plurality of receiving / reproducing devices and a request for the same video at a plurality of resolutions is generated, the MPE generated by the video generation unit is generated.
The G stream requires only one stream, and does not increase the amount of processing required for video encoding and transmission by the video generation unit, and further increases the network load required for video transmission from the video generation unit to the resolution conversion unit. A plurality of resolution images can be provided without any problem, and the practical effect is great.

[0057]

As described above, according to the present invention, first, when an input image is subjected to compression coding using orthogonal transform, the input image and the reduced input image obtained by reducing the input image are compared. An orthogonal transform coefficient is generated by performing an orthogonal transform in block units, and an orthogonal transform coefficient obtained by replacing a part of the orthogonal transform coefficient for the input image with the orthogonal transform coefficient for the reduced input image is used for compression encoding. Also, among the above methods, a coefficient transformation process for generating an orthogonal transform coefficient for a reduced input image from an orthogonal transform coefficient for an input image (generating an orthogonal transform coefficient for a reduced input image by an operation in an orthogonally transformed base axis region) ).

Control means, input means for inputting a compressed video signal containing orthogonal transform coefficients for input images of a plurality of resolutions, and code conversion means for generating a low-resolution compressed video signal from the input compressed video signal And an output unit for individually outputting compressed video signals of a plurality of resolutions, wherein the code conversion unit deletes and changes a part of the variable length code of the compressed video signal input by the input unit. The low-resolution compressed video signal is generated by adding a code.

This makes it possible to extract video signals of a plurality of resolutions from one compressed signal without having to perform an enormous amount of calculation such as decoding, adjustment of difference (hierarchical resolution information), and encoding. It is possible to provide a video compression coding system.

Second, the input image and the reduced input image are orthogonally transformed in units of blocks, and among the orthogonal transformation coefficients generated by the orthogonal transformation, a part of the orthogonal transformation coefficients of the input image is replaced with a part of the reduced input image. When generating a compressed video signal for reproducing the reduced input image from a multi-resolution compressed video signal obtained by performing variable length encoding on the orthogonal transform coefficient replaced with a part of the orthogonal transform coefficient, the multi-resolution compressed video signal A compressed video signal for reproducing the reduced input image is generated by deleting a component other than the orthogonal transform coefficient component of the reduced input image included, and thereby, in addition to the first effect, more compressed video signals are generated. Even if the orthogonal transform coefficients corresponding to the low resolution are combined, the influence on the decoded video of the high resolution can be reduced.

Third, when the request, generation and analysis of a compressed video signal having a plurality of resolutions are performed in a distributed manner on a network, the apparatus according to the first and second aspects of the present invention transmits the compressed video signal to the network. Can be transmitted from the transmitting side, and a plurality of resolutions can be simultaneously provided from the transmitting side only by sending the data amount for one stream (without increasing the network addition).

[Brief description of the drawings]

FIG. 1 is a block diagram of a video encoding device.

FIG. 2 is a flowchart showing overall processing of the video encoding device.

FIG. 3 is a conceptual diagram of DCT coefficients generated by a coefficient synthesis process.

FIG. 4 is a detailed diagram of a DCT block generated by a coefficient synthesis process.

FIG. 5 is a conceptual diagram of a variable length code for a coefficient generated by a coefficient synthesis process.

FIG. 6 is a block diagram of a video encoding device that generates DCT coefficients of a reduced image using coefficient conversion means.

FIG. 7 is a block diagram of a resolution conversion device.

FIG. 8 is a flowchart showing the overall processing of the resolution conversion device.

FIG. 9 is a flowchart showing a code conversion process.

FIG. 10 is a conceptual diagram of a stream generated by a code conversion process.

FIG. 11 is a conceptual diagram of a coefficient synthesis process for dispersing and storing low-resolution DCT coefficients.

FIG. 12 is a flowchart showing processing of the resolution conversion device for a stream in which low-resolution DCT coefficients are stored in a distributed manner;

FIG. 13 is a block diagram of a video information providing system.

FIG. 14 is a flowchart showing processing of the video information providing system.

[Explanation of symbols]

 Reference Signs List 10 Overall control unit 100 Input unit 101 Input unit 102 Input video buffer 200 Encoding unit 201 Encoding control unit 202 Orthogonal transformation unit 203 Image reduction unit 204 Orthogonal transformation unit 205 Coefficient synthesis unit 206 Quantization unit 207 Variable length encoding unit 208 Structure encoding means 209 Coefficient deformation means 300 Output unit 301 Output means 410 Control means 420 Input means 430 Code conversion means 440 Output means 1000 Video generation unit 1100 Video equipment 1200 Video encoding device 1300 Video transmission device 2000 Network 3000 Resolution editing unit 3100 Ideal place 3200 Video receiver 3300 Resolution converter 3400 Video transmitter 4000 Video receiver 4100 Video receiver / player

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Hiroaki Yoshio 1006 Kazuma Kadoma, Osaka Prefecture Inside Matsushita Electric Industrial Co., Ltd. Terms (reference) 5C053 FA20 GA11 GB23 GB26 5C059 KK15 KK41 LC09 MA00 MA21 MA32 MC01 MC11 MC32 MC34 ME01 PP04 SS06 UA02 UA39 5C064 BA07 BB05 BC10 BC27 BD13 5J064 AA04 BA09 BA16 BB13 BD02 BD03

Claims (10)

[Claims] 1. A first orthogonal transformation step for orthogonally transforming an input video signal to obtain a first orthogonal transformation coefficient, a video signal reduction step for reducing the input video signal, and a reduced video signal obtained in the video signal reduction step Is orthogonally transformed to obtain a second orthogonal transform coefficient, a part of the first orthogonal transform coefficient obtained by the first orthogonal transform step is An orthogonal transform coefficient synthesizing step of generating a third orthogonal transform coefficient by substituting all or part thereof;
And a video encoding method. 2. A video signal input means for inputting a first video signal, a first orthogonal transform means for performing an orthogonal transform on the input video signal, and a second orthogonal resolution means for reducing a resolution of the input video signal. Reduced video generating means for generating a video signal, second orthogonal transform means for performing orthogonal transform on the second video signal, and third orthogonal transform coefficient by combining the first and second orthogonal transform coefficients , A quantizing means for quantizing the third orthogonal transform coefficient generated by the coefficient synthesizing means, and a variable-length code for variable-length encoding the coefficient quantized by the quantizing means. A video coding apparatus comprising: coding means; and structure coding means for adding a code other than a video to the code string generated by the variable length coding means. 3. A first orthogonal transformation step of orthogonally transforming an input video signal to obtain a first orthogonal transformation coefficient, and performing a coefficient transformation process on the first orthogonal transformation coefficient to obtain a second orthogonal transformation coefficient. The second orthogonal transformation step of obtaining the third orthogonal transformation by replacing a part of the first orthogonal transformation coefficient obtained by the first orthogonal transformation step with a part or all of the second orthogonal transformation coefficient An orthogonal transform coefficient synthesizing step of generating coefficients. 4. An image signal input means for inputting a first image signal, a first orthogonal transformation means for performing an orthogonal transformation on the inputted image signal to obtain a first orthogonal transformation coefficient, Applying a coefficient coefficient deformation process to the orthogonal transform coefficients to obtain a second orthogonal transform coefficient, and generating a third orthogonal transform coefficient by combining the first and second orthogonal transform coefficients. Coefficient synthesizing means, quantizing means for quantizing the third orthogonal transform coefficient generated by the coefficient synthesizing means, and variable length coding means for performing variable length coding on the coefficients quantized by the quantizing means. And a structure encoding unit for adding a code other than a video to the code sequence generated by the variable length encoding unit. 5. When replacing an orthogonal transform coefficient for an input image with an orthogonal transform coefficient of a reduced input image, a plurality of orthogonal transform coefficients of each orthogonal transform block of the reduced input image are arranged in a scanning order at the time of variable length coding. 4. The method according to claim 1, wherein the set is divided into sets, and each set of orthogonal transform coefficients of the reduced input image is replaced with a part of orthogonal transform coefficients of a plurality of blocks corresponding to the input image. 2. The video encoding method according to 1. 6. A coefficient synthesizing unit arranging orthogonal transform coefficients of a reduced input image in a scanning order at the time of variable-length coding, dividing the set into a plurality of sets, and replacing a part of the orthogonal transform coefficients of a corresponding plurality of blocks of the input image. The video encoding device according to claim 2, wherein a coefficient synthesis process is performed. 7. A first orthogonal transformation step for orthogonally transforming an input video signal to obtain a first orthogonal transformation coefficient, a video signal reduction step for reducing the input video signal, and a reduced video signal obtained in the video signal reduction step Is orthogonally transformed to obtain a second orthogonal transform coefficient, a part of the first orthogonal transform coefficient obtained by the first orthogonal transform step is An orthogonal transform coefficient synthesizing step of generating a third orthogonal transform coefficient by substituting all or part thereof;
A compressed video signal generating step of generating a multi-resolution compressed video signal including a third orthogonal transform coefficient; and a step of deleting components other than the orthogonal transform coefficient component of the reduced video signal included in the multi-resolution compressed signal. . 8. Input means for inputting a compressed video signal composed of a variable-length code including orthogonal transform coefficients corresponding to a plurality of resolution images among orthogonal transform coefficients corresponding to at least one orthogonal transform block, and said input means. Code conversion means for generating at least a low-resolution compressed video signal of the multi-resolution video signal by deleting a part of the variable-length code or adding a variable-length code from the compressed video that has been generated, A compressed image signal having a plurality of resolutions including a low-resolution compressed signal. 9. A compressed video signal having the same content and a plurality of spatial resolutions is provided by the compressed video signal generated by the video compression device according to any one of claims 2, 4, 6, and 8, A video information providing system characterized by reproducing. 10. A video generating unit, a resolution converting unit, a video receiving unit, and a network, wherein the network is connected to the video generating unit, the resolution converting unit, and the video receiving unit for compression. Transmitting a video signal or a control signal, the video generation unit, a video providing device for providing a video source,
A video encoding device that acquires a video from the video providing device and generates a compressed video signal, and a video transmission device that transmits the compressed video signal generated by the video encoding device to the resolution conversion unit, A conversion unit receives and processes a user request sent from the video reception unit, processes the request, and controls all devices of the resolution conversion unit, and a video that receives a compressed video signal transmitted from the video generation unit. A receiving device, a resolution conversion device that converts a compressed video signal received by the video receiving device into a compressed video signal having a resolution requested by the request processing device, and a compressed video signal converted by the resolution conversion device, A video transmitting device for transmitting to a receiving unit, wherein the video receiving unit receives a request from a user and sends a resolution selected by the user to the resolution converting unit. And a video receiving and reproducing device that receives, decodes, and displays the compressed video signal transmitted from the resolution conversion unit, wherein the video encoding device is a video according to any one of claims 2, 4, and 6. An image information providing system, comprising: an encoding device; wherein the resolution conversion device is the image encoding device according to claim 8.