JP2000152248A - Device and method for composing video signal and recording medium where video signal composing program is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To speed up an operation by outputting compressed data of a composite image by performing variable-length encoding of a quantized DCT coefficient and side information including a moving vector. SOLUTION: A quantization part 8 quantizes a composite DCT coefficient from a DCT coefficient replacement part 7 and outputs the result as a quantized DCT coefficient to a variable-length encoding (VLC) part 9 and a reverse quantization part 10a of a local decoder 10. Then the VLC part 9 performs the variable-length encoding of the quantized DCT coefficient from the quantization part 8 and the moving vector from a movement detection part 11 and outputs compressed data 103 in the original moving picture expert group(MPEG) format. The local decoder 10 decodes the quantized DCT coefficient from the quantization part 8 by the reverse quantization part 10a and an IDCT part 10b by a decoding method prescribed to the MPEG format as well as the decoder 1 and outputs a 2nd reference image compensated on the basis of the moving vector obtained by a movement compensation part 10e to a subtracter 5 when further obtaining the moving vector.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video signal synthesizing apparatus, a video signal synthesizing method, and a video signal synthesizing apparatus which synthesizes another image signal with a compressed video signal and recompresses the synthesized image signal in the original compression format. The present invention relates to a recording medium on which a video signal synthesizing program is recorded.

[0002]

2. Description of the Related Art A conventional video signal synthesizing apparatus will be described below with reference to FIG. FIG. 6 is a block diagram showing a configuration of a conventional video signal synthesizing apparatus. As shown in FIG. 6, the conventional video signal synthesizing device is a decoder 61 for decoding compressed data 50, and is connected to the decoder 61, and synthesizes the image data 51 with the decoded image from the decoder 61 based on the position information 52. Unit 62 and the combining unit 62
And an encoder 63 for encoding the composite image from the compressed data 50 in the same format as the compression format of the compressed data 50. The decoder 61 decodes the compressed data 50 for each frame (image) according to the decoding method of the compression format to generate a decoded image (data). The synthesizing unit 62 externally inputs image data 51 and position information 52 that specifies a synthesizing position of the image data 51 in the decoded image to be synthesized. The combining unit 62 combines the decoded image from the decoder 61 with the image data 51 based on the position information 52 to generate a combined image (data).
The encoder 63 generates the compressed data 53 by compression-encoding the synthesized image from the synthesizing unit 62 in the same format as the compression format of the input compressed data 50. As described above, the conventional video signal synthesizing device completely decodes the compressed data 50, synthesizes the image data 51 obtained from the outside, and compress-encodes the synthesized image for one frame to generate the compressed data 53. Was output as.

[0003]

In the conventional video signal synthesizing apparatus as described above, although the image data is synthesized only in a part of the image before the synthesis, all the compressed data of one frame is synthesized. After decoding the image data and synthesizing the image data, all the data for one frame (synthesized image) is recompressed in the original compression format and encoded. For this reason,
In the conventional video signal synthesizing apparatus, the amount of arithmetic processing is the sum of the respective processing amounts of decoding, synthesizing, and encoding, and the video signal synthesizing process including the series of processes cannot be performed at high speed. There was a point.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and simplifies a video signal synthesizing process including a compression encoding process for compressing and encoding in a compression format before synthesis to achieve high speed. It is an object of the present invention to provide a video signal synthesizing apparatus, a video signal synthesizing method, and a recording medium on which a video signal synthesizing program is recorded.

[0005]

A video signal synthesizing device according to the present invention synthesizes another video signal with a compressed video signal,
A video signal synthesizing device for re-compressing, wherein a decoder for decoding the compressed data and decoding the decoded image, and synthesizing the image data with the decoded image from the decoder based on the position information to generate a synthesized image A reference MB detecting unit that detects a reference position of a macro block for performing motion compensation in the decoder and sets the position information as a reference macro block position; Using a position information conversion unit that generates a converted macroblock position converted to the position of a macroblock that performs partial encoding on the composite image, connected to the synthesis unit, according to the type of the composite image A subtractor that outputs pixel data or difference data of the composite image as output data, and among the macroblocks of output data from the subtractor, A DCT unit for performing a DCT operation on a macroblock designated by a conversion macroblock position from the location information conversion unit to obtain a partial DCT coefficient; and a DCT coefficient among the DCT coefficients of the decoded image input from the decoder. The DCT coefficient of the macroblock designated by the position of the converted macroblock is replaced with the partial DCT coefficient from the DCT unit to obtain a composite DCT coefficient.
A CT coefficient replacement unit, a quantization unit that quantizes the combined DCT coefficient from the DCT coefficient replacement unit to obtain a quantized DCT coefficient, and decodes again the quantized DCT coefficient of the combined image from the quantization unit. Generating a first reference image not subjected to motion compensation for detecting a motion vector of the composite image, and performing motion compensation on the first reference image to obtain difference data of the composite image. A local decoder that generates a second reference image, and a motion that detects a motion vector of the synthesized image for a macroblock specified by a converted macroblock position from the position information conversion unit in accordance with the type of the synthesized image. A variable length code is applied to a detection unit and side information including a quantized DCT coefficient from the quantization unit and a motion vector from the motion detection unit, and And a VLC unit for outputting the compressed data. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing apparatus according to another aspect of the present invention is a video signal synthesizing apparatus for synthesizing another video signal with a compressed video signal and recompressing the compressed video signal. A combining unit that combines a decoded image from the decoder with image data based on the position information to generate a combined image, and a reference position of a macro block for performing motion compensation in the decoder. And a reference MB detection unit that detects the position of the reference macroblock and uses the position information as a reference macroblock position from the reference MB detection unit to determine a position of a macroblock for performing partial encoding on the composite image. A position information conversion unit that generates a converted converted macroblock position; and a pixel data or a difference of the synthesized image that is connected to the synthesizing unit according to the type of the synthesized image. A DCT operation is performed on a macroblock specified by the position of the conversion macroblock from the position information conversion unit, from among the macroblocks of the output data from the subtractor, which outputs the data as output data. A DCT unit for obtaining a DCT coefficient, and a quantization unit for performing a quantization operation on a partial DCT coefficient from the DCT unit using a quantization matrix of a decoded image input from the decoder to obtain a partially quantized DCT coefficient. ,
Among the quantized DCT coefficients of the decoded image input from the decoder, the quantized DCT coefficients of the macroblock designated by the transformed macroblock position from the position information converting unit are used as the partially quantized DCT coefficients from the quantizing unit. Replace
A quantized DCT coefficient replacement unit for converting the quantized DCT coefficient, and a motion compensation for detecting a motion vector of the synthesized image by decoding again the quantized DCT coefficient of the synthesized image from the quantized DCT coefficient replacement unit. No 1
And a local decoder that generates a second reference image for obtaining difference data of the synthesized image by performing motion compensation on the first reference image and a type of the synthesized image. A motion detection unit that detects a motion vector of the composite image for a macroblock specified by a converted macroblock position from the position information conversion unit;
A VLC unit that applies a variable length code to the quantized DCT coefficient from the quantized DCT coefficient replacement unit and side information including the motion vector from the motion detection unit, and outputs compressed data of the synthesized image. I have. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing apparatus according to another aspect of the present invention is a video signal synthesizing apparatus for synthesizing another video signal with a compressed video signal and recompressing the compressed video signal. A combining unit that combines the decoded image from the decoder with image data based on the position information to generate a combined image, and performs partial encoding of the position information on the combined image A position information conversion unit for generating a converted macroblock position converted to the position of the macroblock, and a macroblock specified by the converted macroblock position from the position information conversion unit among the macroblocks of the pixel data from the synthesis unit About DCT
Performing a quantization operation on a partial DCT coefficient from the DCT unit by using a DCT unit for performing an operation to obtain a partial DCT coefficient and a quantization matrix of a decoded image input from the decoder; , And a quantized DCT of the decoded image input from the decoder.
Of the coefficients, the quantized DCT coefficient of the macroblock specified by the transform macroblock position from the position information transforming unit is replaced with the partially quantized DCT coefficient from the quantizing unit, and is used as the quantized DCT coefficient. A coefficient replacement unit, and quantization D from the quantized DCT coefficient replacement unit.
A VLC unit that applies a variable length code to the side information including the CT coefficient and the quantization matrix and outputs compressed data of the synthesized image. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing apparatus according to another aspect of the present invention is a video signal synthesizing apparatus for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein the synthesized image generates input position information. A position information conversion unit for generating a converted macroblock position converted to a position of a macroblock to be subjected to the partial encoding described above, and a variable length decoding of the compressed data, and among the obtained quantized DCT coefficients, the position information conversion A decoder that performs an inverse quantization operation and an IDCT operation on the quantized DCT coefficient of the macroblock designated by the converted macroblock position from the unit to generate a partially decoded image, and a decoder based on the position information. A synthesizing unit that synthesizes image data with the partially decoded image from the decoder to generate a partially synthesized image; and a macro block of pixel data from the synthesizing unit. The DCT unit performs a DCT operation on a macroblock designated by the position of the converted macroblock from the position information conversion unit to obtain a partial DCT coefficient, and uses the quantization matrix of a decoded image input from the decoder to perform the DCT operation. A quantization unit that performs a quantization operation on the partial DCT coefficient from the unit to obtain a partially quantized DCT coefficient, and a quantization D of the decoded image input from the decoder.
Among the CT coefficients, the quantization DC of the macroblock specified by the position of the conversion macroblock from the position information conversion unit.
A quantized DCT coefficient replacing unit that replaces a T coefficient with a partially quantized DCT coefficient from the quantizing unit to obtain a quantized DCT coefficient; a quantized DCT coefficient from the quantized DCT coefficient replacing unit; And a VLC unit that applies a variable-length code to side information including the compressed image data and outputs compressed data of the synthesized image. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing method according to the present invention is a video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again. Decoding step, combining image data with the decoded image decoded in the decoding step based on the position information,
A synthesis step for generating a synthesized image, a reference MB detection step for detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the decoding step and setting the reference position as a reference macroblock position, and a reference MB detection step. A position information converting step of generating a converted macroblock position obtained by converting the position information into a position of a macroblock for performing partial encoding on the synthesized image using a reference macroblock position; and a synthesized image generated in the synthesizing step. Outputting pixel data or difference data of the synthesized image as output data in accordance with the type of DCT. DCT is performed on the macroblock specified by the conversion macroblock position among the macroblocks of the output data.
Performing a DCT operation to obtain a partial DCT coefficient; and obtaining, in the DCT operation step, a DCT coefficient of a macroblock designated by the transformed macroblock position among DCT coefficients of the decoded image decoded in the decoding step. DCT coefficient replacement step of replacing the partial DCT coefficient with the obtained partial DCT coefficient to obtain a combined DCT coefficient, quantizing the combined DCT coefficient obtained in the DCT coefficient replacement step,
A quantization step of obtaining a quantized DCT coefficient, a first step of decoding the quantized DCT coefficient of the composite image obtained in the quantization step again and not performing a motion compensation for detecting a motion vector of the composite image; Generating a second reference image for obtaining difference data of the synthesized image by performing motion compensation on the first reference image, and generating a second reference image for obtaining difference data of the synthesized image. A motion detecting step of detecting a motion vector of the synthesized image for a macroblock specified at the converted macroblock position, a quantized DCT coefficient obtained in the quantizing step, and a motion detecting step. A variable length code step of applying a variable length code to the obtained side information including the motion vector to generate compressed data of the synthesized image. There. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing method according to another aspect of the present invention is a video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again. A combining step of combining image data with the decoded image decoded in the decoding step based on the position information to generate a combined image, and a macro for performing motion compensation on the decoded image decoded in the decoding step. A reference MB detection step of detecting a reference position of a block and setting the reference macro block position, using the reference macro block position detected in the reference MB detection step,
A position information converting step of generating a converted macroblock position obtained by converting the position information into a macroblock position for performing partial encoding on the synthesized image, and synthesizing the macroblock according to a type of the synthesized image generated in the synthesizing step; Outputting pixel data or difference data of an image as output data; and performing a DCT operation on a macroblock of the output data designated by the converted macroblock position to obtain a partial DCT coefficient. Performing a quantization operation on the partial DCT coefficients obtained in the DCT operation step using a quantization matrix of the decoded image decoded in the decoding step,
A quantization step of obtaining a partially quantized DCT coefficient, and a quantization DCT coefficient of a macroblock designated by the transformed macroblock position among the quantized DCT coefficients of the decoded image decoded in the decoding step. Replacing the obtained partially quantized DCT coefficients with the quantized DCT coefficients, and decoding the quantized DCT coefficients of the synthesized image obtained in the quantized DCT coefficient replacement step again to obtain the synthesized image. A first reference image not subjected to motion compensation for detecting the motion vector of the first reference image, and performing a motion compensation on the first reference image to obtain difference data of the composite image.
Generating a reference image, a motion detecting step of detecting a motion vector of the synthesized image for a macroblock specified by the converted macroblock position according to a type of the synthesized image generated in the synthesizing step, and the quantum Variable-length code step of applying a variable-length code to the quantized DCT coefficient obtained in the transformed DCT coefficient replacement step and side information including the motion vector obtained in the motion detection step to generate compressed data of the composite image It has. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A video signal synthesizing method according to another aspect of the present invention is a video signal synthesizing method for synthesizing another video signal with a compressed video signal and recompressing the compressed video signal. A decoding step of combining the decoded image decoded in the decoding step with the image data based on the position information to generate a combined image; and a macro for performing partial encoding of the position information on the combined image. A position information conversion step of generating a converted macroblock position converted to a block position, and performing a DCT operation on a macroblock specified by the converted macroblock position among macroblocks of pixel data of the synthesized image generated in the synthesis step. And a DCT operation step of obtaining a partial DCT coefficient, and a quantization matrix of the decoded image decoded in the decoding step. Using a scan, by performing quantization operation on a portion DCT coefficients obtained by the DCT calculation step,
A quantization step of obtaining a partially quantized DCT coefficient, and a quantization DCT coefficient of a macroblock designated by the transform macroblock position among the quantized DCT coefficients of the decoded image decoded in the decoding step. A quantized DCT coefficient replacing step of replacing the obtained partially quantized DCT coefficient with a quantized DCT coefficient, and the quantization D obtained in the quantized DCT coefficient replacing step.
A variable length code step of applying a variable length code to the side information including the CT coefficient and the quantization matrix of the decoded image decoded in the decoding step to generate compressed data of the composite image. With this configuration,
The decoding process of the video signal, and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format can be simplified, and the amount of calculation can be appropriately reduced,
And it can be performed at high speed.

A video signal synthesizing method according to another aspect of the present invention is a video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein the synthesized image generates input position information. A position information conversion step of generating a converted macroblock position converted to a position of a macroblock to be subjected to the partial encoding described above, variable-length decoding of the compressed data, and among the obtained quantized DCT coefficients, the converted macroblock position DC of the macroblock specified by
A decoding step of performing an inverse quantization operation and an IDCT operation on the T coefficient to generate a partially decoded partially decoded image, and combining the partially decoded image decoded in the decoding step with image data based on the position information, A synthesizing step of generating a partially synthesized image, and performing a DCT operation on a macroblock specified at the converted macroblock position among macroblocks of pixel data of the partially synthesized image generated in the synthesizing step to obtain a partial DCT coefficient A DCT operation step, a quantization step of performing a quantization operation on the partial DCT coefficients obtained in the DCT operation step using a quantization matrix of the decoded image decoded in the decoding step, to obtain a partially quantized DCT coefficient; The transform macro block among the quantized DCT coefficients of the decoded image decoded in the decoding step. The quantized DCT coefficients for the specified macroblock at a position replacing the resulting partially quantized DCT coefficients in the quantization step, the quantized DCT coefficients replacement step of the quantized DCT coefficients, and the quantization D
A variable-length code for generating compressed data of the synthesized image by applying a variable-length code to the quantized DCT coefficient obtained in the CT coefficient replacement step and side information including the quantization matrix of the decoded image decoded in the decoding step It has steps. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

A recording medium on which the video signal synthesizing program of the present invention is recorded decodes the compressed data and decodes the decoded data into a decoded image, and synthesizes the decoded image decoded in the decoding step with the image data based on the position information. A synthesizing step of generating a synthesized image, a reference MB detecting step of detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the decoding step, and setting the reference position as a reference macroblock position. Using the detected reference macroblock position, the position information is generated in the position information conversion step of generating a converted macroblock position in which the position information is converted into the position of a macroblock for performing partial encoding on the synthesized image. Depending on the type of the composite image, pixel data or difference data of the composite image is used as output data. Performing a DCT operation on a macroblock specified at the converted macroblock position among macroblocks of the output data to obtain a partial DCT coefficient; and outputting a decoded image of the decoded image decoded in the decoding step. Of the DCT coefficients, the DCT coefficient of the macroblock specified at the position of the transform macroblock is replaced with the partial DCT coefficient obtained in the DCT operation step, and is used as a composite DCT coefficient.
A CT coefficient replacement step, a quantization step of quantizing the combined DCT coefficient obtained in the DCT coefficient replacement step to obtain a quantized DCT coefficient, and decoding again of the quantized DCT coefficient of the combined image obtained in the quantization step Then, a first reference image not subjected to motion compensation for detecting a motion vector of the composite image is generated, and motion compensation is performed on the first reference image to calculate difference data of the composite image. Generating a second reference image to be obtained, and motion detection for detecting a motion vector of the synthesized image for a macroblock specified by the converted macroblock position in accordance with a type of the synthesized image generated in the synthesizing step. Step, and the quantized DCT coefficient obtained in the quantization step, and the motion vector obtained in the motion detection step. Subjected to variable length code the free side information, and a variable-length code generating compressed data of the composite image. With this configuration,
The decoding process of the video signal, and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format can be simplified, and the amount of calculation can be appropriately reduced,
And it can be performed at high speed.

According to another aspect of the present invention, there is provided a recording medium storing a video signal synthesizing program according to another aspect of the present invention, wherein a decoding step of decoding compressed data and decoding the decoded data into a decoded image; A reference MB detection step of detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the decoding step and setting the reference position as a reference macroblock position; Using the reference macroblock position detected in the detection step, a position information conversion step of generating a converted macroblock position in which the position information is converted into a position of a macroblock for performing partial encoding on the synthesized image, the synthesis step Depending on the type of the composite image generated in step 2, pixel data or difference data of the composite image Step of outputting as the force data,
Of the macroblocks of the output data, the macroblock specified at the converted macroblock position is D
Performing a CT operation to obtain a partial DCT coefficient, performing a quantization operation on the partial DCT coefficient obtained in the DCT operation step using a quantization matrix of the decoded image decoded in the decoding step, A quantization step of obtaining a partially quantized DCT coefficient, and a quantization DCT coefficient of a macroblock designated by the transform macroblock position among the quantized DCT coefficients of the decoded image decoded in the decoding step. A quantized DCT coefficient replacing step of replacing the obtained partially quantized DCT coefficient as a quantized DCT coefficient, and a quantized DCT of the composite image obtained in the quantized DCT coefficient replacing step.
Decoding the coefficients again to generate a first reference image that has not been subjected to motion compensation for detecting a motion vector of the composite image, and performs motion compensation on the first reference image;
A step of generating a second reference image for obtaining difference data of the composite image; and, according to a type of the composite image generated in the composite step, a motion vector of the composite image is designated by the position of the conversion macroblock. A motion detection step for detecting a macroblock, and the quantization D
A variable length code step of applying a variable length code to the quantized DCT coefficient obtained in the CT coefficient replacement step and the side information including the motion vector obtained in the motion detection step to generate compressed data of the synthesized image; Have. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

According to another aspect of the present invention, there is provided a recording medium storing a video signal synthesizing program according to another aspect of the present invention, wherein a decoding step of decoding compressed data and decoding the decoded data into a decoded image is performed. A combining step of combining based on the position information to generate a converted macroblock position obtained by converting the position information into a position of a macroblock for performing partial encoding on the combined image; and the combining step. Out of the macroblocks of the pixel data of the composite image generated in
Performing a CT operation to obtain a partial DCT coefficient, using a quantization matrix of the decoded image decoded in the decoding step, performing a quantization operation on the partial DCT coefficient obtained in the DCT operation step, A quantization step of obtaining a partially quantized DCT coefficient, and a quantization DCT coefficient of a macroblock designated by the transformed macroblock position among the quantized DCT coefficients of the decoded image decoded in the decoding step. Replacing the quantized DCT coefficient with the obtained partially quantized DCT coefficient to obtain a quantized DCT coefficient;
The quantized DCT coefficients obtained in the T coefficient replacement step,
And a variable-length code step of applying a variable-length code to side information including a quantization matrix of the decoded image decoded in the decoding step to generate compressed data of the composite image. With this configuration, it is possible to simplify the video signal decoding process and the video signal synthesis process including the compression encoding process of compressing and encoding again in the original format, and appropriately reduce the calculation amount. And at high speed.

According to another aspect of the present invention, there is provided a recording medium on which a video signal synthesizing program according to the present invention is recorded, wherein a converted macroblock position converted into a macroblock position for performing partial encoding on a synthesized image for generating input position information is generated. And performing variable length decoding of the compressed data, and performing an inverse quantization operation and an IDCT operation on the quantized DCT coefficients of the macroblock designated by the converted macroblock position among the obtained quantized DCT coefficients. A decoding step of generating a partially decoded image partially decoded, synthesizing image data with the partially decoded image decoded in the decoding step based on the position information, and generating a partially synthesized image; Of the macroblocks of the pixel data of the partial composite image generated in By performing DCT calculation for black blocks, DCT computation step of determining a partial DCT coefficients,
Using the quantization matrix of the decoded image decoded in the decoding step, the part D obtained in the DCT operation step
A quantization step of performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient; and a quantization step of a macroblock designated by the transformed macroblock position among quantized DCT coefficients of the decoded image decoded in the decoding step. DCT
The coefficients are partially quantized DCT obtained in the quantization step.
Quantized DCT that replaces coefficients with the quantized DCT coefficients
Applying a variable length code to the coefficient replacement step, side information including the quantization matrix of the decoded image decoded in the decoding step, and the quantized DCT coefficient obtained in the quantization DCT coefficient replacement step, A variable length code step for generating compressed data is provided.
With such a configuration, decoding processing of a video signal,
In addition, the video signal synthesizing process including the compression encoding process of recompressing and encoding in the original format can be simplified, and the amount of calculation can be appropriately reduced, and the operation can be performed at high speed.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A preferred embodiment showing a video signal synthesizing apparatus and a video signal synthesizing method of the present invention will be described below with reference to the drawings.

<< First Embodiment >> FIG. 1 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a first embodiment of the present invention. In the following description, MPEG (Moving Pictu
A configuration in which another image data is combined with the compressed data compressed in the re Experts Group format (ISO / IEC 11172 standard and 13818 standard) and compressed again in the original MPEG format will be described. As shown in FIG. 1, the video signal synthesizing apparatus of the present embodiment
A decoder 1 that inputs the image data 101 from the outside and performs a decoding process in units of one frame (image), and combines the decoded image input from the decoder 1 with the image data 101 input from the outside based on the position information 102 from the outside. The image processing apparatus includes a synthesizing unit 2 that generates an image, and a reference MB detection unit 3 that detects a reference position of a macroblock (Macroblock) for performing motion compensation in the decoder 1 and sets the reference position as a reference macroblock position. In order to decode the compressed data 100, the decoder 1 receives a VLD (Variable Length Decorder) unit 1a for inputting the compressed data 100, an inverse quantization unit 1b sequentially connected to the VLD unit 1a, and an IDCT (inverse). Inverse Discrete Cosine Transfor
m) A section 1c and an adder 1d are provided. Further, the decoder 1 is connected to the VLD unit 1a and the motion compensator 1e connected to the adder 1d, and to the adder 1d and the motion compensator 1e in order to perform motion compensation on P and B pictures described later. A frame memory 1f is provided.

The video signal synthesizing apparatus of this embodiment is a position information converter for inputting position information 102 and a reference macroblock position in order to apply partial coding to the synthesized image from the synthesizing unit 2 in units of macroblocks. 4 and a subtractor 5 sequentially connected to the synthesizing unit 2, a DCT (Discrete Cosine Transform, Discrete
te Cosine Transform section 6, DCT coefficient replacement section 7, quantization section 8, and VLC (variable length code, Variable L
(length code) unit 9 is provided. Furthermore, the video signal synthesizing device of the present embodiment decodes the synthesized image quantized by the quantization unit 8 again, and performs the first reference without performing motion compensation for detecting a motion vector of the synthesized image. A local decoder 10 that performs motion compensation on the image and the first reference image to generate a second reference image for obtaining difference data of a composite image, and a motion detection unit that detects a motion vector in macroblock units 11 is provided.
The local decoder 10 includes an inverse quantization unit 10a and an IDCT unit 10b for decoding the composite image again,
The first reference image is connected to the CT unit 10b via an adder 10c, and a second reference image is stored in the second reference unit using the frame memory 10d for storing the first reference image and the motion vector from the motion detection unit 11. Motion compensator 10e for generating an image
It has. The motion detection unit 11 detects a motion vector according to the picture type of the decoded image (synthesized image) to be synthesized and outputs the motion vector to the VLC unit 9 and the local decoder 10 (details will be described later).

The position information conversion unit 4 uses the reference macroblock position from the reference MB detection unit 3 to convert the input position information 102 into the position of a macroblock for performing partial encoding on a composite image. Generate a block position. The subtractor 5, as described in detail later, according to the picture type (kind) of the composite image from the composite unit 2, or the pixel data of the composite image or the second reference from the local image and the composite image. The difference data obtained by calculating the difference from the image is output to the subsequent DCT unit 6 as output data. Here, the pixel data and the difference data are, for example, data of a pixel block unit of (8 × 8) pixels, and data of one frame is constituted by data of a predetermined number of pixel blocks. The DCT unit 6 performs the conversion based on the transformed macroblock position input from the position information
The output data from the subtracter 5 is subjected to a DCT operation,
The CT coefficients are obtained and output to the DCT coefficient replacement unit 7.

The DCT coefficient replacement unit 7 performs a conversion based on the position of the transformed macroblock input from the position information
The DCT coefficient input from the inverse quantization unit 1b of the decoder 1 is replaced with the partial DCT coefficient input from the DCT unit 7, and is output to the quantization unit 8 as a combined DCT coefficient in frame units. The quantization unit 8 quantizes the combined DCT coefficient input from the DCT coefficient replacement unit 7 and outputs the quantized DCT coefficient to the VLC unit 9 and the inverse quantization unit 10a of the local decoder 10 as a quantized DCT coefficient. The quantization unit 8 performs a quantization operation on the synthesized DCT coefficients in frame units, that is, all macroblocks of the processing target image. Therefore, the quantization unit 8 can perform control to set the bit rate of the image to a desired value, and the VLC unit 9 can generate a stream (compressed data 103) at a constant rate. The VLC unit 9 performs VLC operation on the quantized DCT coefficient from the quantization unit 8 and side information including the motion vector from the motion detection unit 11, and outputs the result as compressed data 103.

Hereinafter, detailed functions and operations of the video signal synthesizing apparatus according to the present embodiment will be described. The decoder 1
The decoder decodes the compressed data 100 in accordance with the decoding method specified in the MPEG format.
The decoded image is combined with the I picture (intra-frame coded image), P picture (inter-frame forward coded image), and B picture (bidirectional coded image) specified in the PEG format. Output to More specifically, the VLD unit 1a performs variable length decoding on the input compressed data 100 and decodes the compressed data 100 into quantized DCT coefficients and side information. Further, in the case of a P or B picture, the VLD unit 1a outputs a motion vector included in the decoded side information to the motion compensation unit 1e and the motion detection unit 11. Inverse quantization unit 1b
Calculates the DCT coefficient by inversely quantizing the quantized DCT coefficient from the VLD unit 1a. The IDCT unit 1c performs an IDCT operation on the DCT coefficient from the inverse quantization unit 1b, and outputs pixel data or difference data to the adder 1d.

In detail, in the case of an I picture, since it is not necessary to perform motion compensation, the IDCT unit 1c outputs pixel data to the adder 1d, and the adder 1d outputs the input pixel data to the output (decoding) of the decoder 1. (Image) to the synthesizing unit 2. Further, the adder 1d also outputs the pixel data to the frame memory 1f, and holds the decoded picture of the I picture as a reference picture. On the other hand, in the case of a P picture and a B picture, the motion compensation unit 1e performs motion compensation on the reference image stored in the frame memory 1f using the motion vector from the VLD unit 1a, and adds it as a motion-compensated reference image. Output to the container 1d. Further, the quantized DCT coefficient decoded by the VLD unit 1a is subjected to inverse quantization and IDCT operation similarly to the I picture, and is output as difference data from the IDCT unit 1c to the adder 1d. Adder 1d
Adds the difference data to the input compensated reference image and outputs the result to the synthesizing unit 2 as a decoded image. Also, P
In the case of a picture, the decoded image is output from the adder 1d to the frame memory 1f, and is held as a reference image in the same manner as the I picture. The decoder 1 outputs the decoded images to the synthesizing unit 2 in the decoding order without rearranging the images in the display order. Thereby, the memory for rearranging the image data from the adder 1d in the display order can be reduced, and the image data of the synthesized image after the synthesis need not be rearranged in the encoding order. Therefore, the image data is rearranged in the encoding order. Memory can be reduced, and processing and configuration in the device can be simplified.

Subsequently, the synthesizing unit 2 rearranges the image data 101 and the position information 102 input from the outside in the order of encoding so as to match the order of the decoded images output from the decoder 1 and synthesizes the decoded images. I do. For example, when the decoded image is output from the decoder 1 in the order of the I, P, B1, and B2 pictures, the combining unit 2 outputs the I, P,
Image data 101 and position information 102 corresponding to the image data 101 are input in the order of B1, B2, and P pictures. Therefore, the synthesizing unit 2 first inputs the input image data 1
01 and the position information 102 are rearranged in the order of 1, 4, 2, and 3 in accordance with the order of the above-described decoded images. Thereafter, the synthesizing unit 2 synthesizes the image data 101 with the position of the decoded image to be synthesized indicated by the rearranged position information 102, and outputs a synthesized image. Here, the position information 102 uses coordinates on the decoded image, and is constituted by coordinates connecting the outline of the shape formed by the corresponding image data 101. For example, if the shape of the image data 101 is a rectangle, the position information 102 is constituted by the coordinates of the four vertices of the rectangle.

The reference MB detection unit 3 obtains, for each macroblock, a reference position of the macroblock, which is information of motion compensation in the motion compensation unit 1e. Reference MB detection unit 3
Detects which macroblock in the reference image the macroblock refers to based on the acquired reference position of the macroblock, and sets the detected macroblock as the reference macroblock position. For example, specifically explaining with reference to FIG. 2, the macroblock 13 on which motion compensation has been performed refers to a part of the macroblocks 1, 2, 5, and 6 in the reference image 12. In this case, the reference MB detection unit 3 sets the reference macroblock position of the macroblock 13 as 1, 2, 5,
6 macroblocks are detected. Thus, the reference MB
The detection unit 3 detects the position of each reference macroblock for all the macroblocks included in the input reference image and notifies the position information conversion unit 4 of the position.

Next, the position information converter 4 calculates the position of the macroblock to be partially coded on the composite image based on the input position information 102 and the reference macroblock position, and outputs the calculated position as the converted macroblock position. I do. The position information converter 4 changes the processing operation according to the picture type of the composite image, and sets the converted macroblock position to an appropriate position. More specifically, as a processing operation commonly performed for I, P, and B pictures, the position information conversion unit 4 first detects a synthesized macroblock position using the input position information 102. The method for detecting the position of this synthesized macroblock is described in reference M.
This is performed for each macroblock in the same manner as the method of detecting the reference macroblock position in the B detection unit 3, and the image data 101 is synthesized with any macroblock based on the position information 102 of the image data 101 to be synthesized. Is obtained, and the obtained position of the macro block is set as a synthesized macro block position.

Next, in the case of an I picture, the position information conversion unit 4 outputs the above-described synthesized macroblock position to the DCT unit 6, the DCT coefficient replacement unit 7, and the motion detection unit 11 as a converted macroblock position. At the same time, the position information conversion unit 4 sets the converted macroblock position to the first or second macroblock position.
It is held in a bank format stored as a synthesized macroblock position. That is, when the previous converted macroblock position is held as the first (second) synthesized macroblock position, the position information conversion unit 4 uses the converted macroblock position obtained this time as the second (first) synthesized macroblock. Hold as block position. The position information conversion unit 4 performs the operation of holding the converted macroblock position using the above-described bank format for I and P pictures that can be reference images. For example, the order of synthesized images is I0, P0, B1, B2, P
Assuming that 1, B3, B4, I1, B5, B6, P2, B7, B8, the position information converter 4 sets the converted macroblock positions of I0, P0, P1, I1, P2 to the first, second, and first, respectively. , And the second and first synthesized macroblock positions are sequentially changed and held.

In the case of a P-picture, the position information conversion unit 4 selects, from the previous converted macroblock position held as the first or second synthesized macroblock position, the input reference macroblock position. A macro block that overlaps with the included macro block is detected, and the detected macro block position is set. Further, the position information converter 4 detects the detected macroblock position and the position information 102.
And outputs the converted macroblock position to the DCT unit 6, the DCT coefficient replacement unit 7, and the motion detection unit 11 as a converted macroblock position. at the same time,
As described above, the position information conversion unit 4 performs the same processing operation as that of the I picture, and holds the converted macroblock position as the first or second combined macroblock position. Further, in the case of a B picture, the position information conversion unit 4 determines, from among the previous converted macroblock positions held as the first and second combined macroblock positions, a macroblock included in the input reference macroblock position. A duplicated macroblock is detected, and the detected macroblock position is set. Further, the position information conversion unit 4 combines the detected macroblock position with the synthesized macroblock position obtained based on the position information 102 to obtain a converted macroblock position, which is used as the DCT unit 6, the DCT coefficient replacement unit 7, and the motion detection unit. 11 is output.

Further, the motion detection unit 11 performs motion detection on P and B pictures since I pictures do not require motion compensation. More specifically, when the decoded image to be synthesized (synthesized image) is a P or B picture, the motion detection unit 11 transmits the first reference image stored in the frame memory 10 d of the local decoder 10 and the synthesized image from the synthesis unit 2. And a motion vector is detected by comparing the motion vector with the synthesized image. In the motion vector detection process, the motion detection unit 11 performs motion detection only on the macroblock specified by the converted macroblock position from the position information conversion unit 4, and detects the motion vector of the composite image. The motion detection unit 11 extracts a motion vector from the VLD unit 1a of the decoder 1 for a motion vector of a macroblock located at a position other than the above-described converted macroblock position. The motion detection unit 11 outputs the detected motion vector and the extracted motion vector to the VLC unit 9 and the motion compensation unit 10e of the local decoder 10. As described above, since the motion detection unit 11 performs motion detection of a composite image only for a macroblock specified at a converted macroblock position, the video signal synthesis device of the present embodiment greatly reduces the processing load of motion detection. It is possible to perform processing in the device at high speed.

The subtracter 5 determines whether the decoded image to be synthesized is P, B
In the case of a picture, the difference between the synthesized image output from the synthesis unit 2 and the second reference image output from the motion compensation unit 10e of the local decoder 10 is calculated. And the subtractor 5
Outputs the difference data, which is the calculation result, to the DCT unit 6 as output data. On the other hand, in the case of an I picture that does not require motion compensation, the subtracter 5 outputs the pixel data of the synthesized image itself from the synthesis unit 2 as output data without taking a difference. Subsequently, the DCT unit 6 is designated by the converted macroblock position from the position information conversion unit 4 among the macroblocks of the difference data (P and B pictures) or the pixel data (I picture) input from the subtractor 5. The DCT operation is performed only on the macro block. DCT unit 6
Outputs the calculation result to the DCT coefficient replacement unit 7 as a partial DCT coefficient. As described above, since the DCT unit 6 performs the DCT operation only on the macroblock specified at the position of the transformed macroblock, the video signal synthesizing apparatus of the present embodiment can reduce the amount of DCT operation and perform high-speed processing. .

Next, the DCT coefficient replacement unit 7 converts the DCT coefficients of the decoded image input from the inverse quantization unit 1 b of the decoder 1 into the macroblock specified by the converted macroblock position from the position information conversion unit 4. DCT coefficient only, D
The DCT coefficient is replaced with the partial DCT coefficient from the CT unit 6 and is output to the quantization unit 8 as a combined DCT coefficient for each frame. As described above, the DCT coefficient replacement unit 7 obtains the partial DCT coefficient and the DCT coefficient of the decoded image from the DCT unit 6 and the inverse quantization unit 1b of the decoder 1, respectively, and replaces the DCT coefficient based on the position of the transformed macroblock. are doing. As a result, in the video signal synthesizing apparatus of the present embodiment, the DCT coefficients decoded by the decoder 1 can be used for the macroblock of the decoded image in which the image data 101 is not synthesized, and the processing in the apparatus can be performed at high speed. It is possible to do it. It should be noted that each processing operation in the DCT coefficient replacement unit 7, the subsequent quantization unit 8, and the VLC unit 9 is performed for all I, P, and B pictures regardless of the picture type of the decoded image.

Subsequently, the quantizing section 8 quantizes the combined DCT coefficient from the DCT coefficient replacing section 7 to quantize the DCT coefficient.
The coefficients are output to the VLC unit 9 and the inverse quantization unit 10a of the local decoder 10 as coefficients. As described above, since the quantization operation in the quantization unit 8 is performed on the synthesized DCT coefficients in frame units, the bit rate of the image to be processed can be set to a desired value. Next, the VLC unit 9 performs variable-length coding on the quantized DCT coefficient from the quantization unit 8 and side information including the motion vector from the motion detection unit 11, and converts the compressed data 103 compressed in the original MPEG format. Output.

Similarly to the decoder 1, the local decoder 10 has an inverse quantization unit 10a and an IDCT unit 10 according to a decoding method defined in the MPEG format.
b, the quantized DCT coefficient from the quantization unit 8 is decoded. However, unlike the decoder 1, the local decoder 10 inputs a quantized DCT coefficient, so that there is no need to provide a VLD unit for performing variable length decoding. Local decoder 10
Decodes the input quantized DCT coefficients for the I and P pictures of the decoded image, and stores the decoded DCT coefficients as a first reference image in the frame memory 10d. Local decoder 1
0 indicates that if the synthesized image to be encoded is a P or B picture, the frame memory 10d
Are output to the motion detecting unit 11. Further, when the local decoder 10 obtains a motion vector from the motion detecting unit 11, the local reference 10 performs a motion compensation based on the motion vector obtained by the motion compensating unit 10e in order to generate difference data from the synthesized image. Is output to the subtractor 5.

As described above, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the position information converting section 4
Uses the reference macroblock position from the reference MB detection unit 3 to generate a converted macroblock position in which the input position information 102 is converted into a macroblock position to be subjected to partial encoding on the composite image. Further, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the DCT
The unit 6 and the motion detecting unit 11 perform the DCT operation of the synthesized image and the detection of the motion vector only for the macroblock designated by the above-described converted macroblock position. Further, the DCT coefficient replacement unit 7 replaces only the DCT coefficients of the macroblock designated at the above-described transformed macroblock position with the partial DCT coefficients from the DCT unit 6 among the DCT coefficients of the decoded image input from the decoder 1. I have. For other macroblocks not specified by the above-mentioned conversion macroblock position, D
The CT coefficient replacement unit 7 and the motion detection unit 11 use the DCT coefficient and the motion vector of the decoded image obtained in the decoder 1, respectively. As a result, the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment simplify the video signal synthesizing process including the decoding process of the video signal and the compression encoding process of re-compression encoding in the original format. Thus, the amount of calculation can be appropriately reduced, and the calculation can be performed at high speed. Further, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the quantization unit 8
Performs a quantization operation on all macroblocks, so that the bit rate can be controlled, and a stream with a constant rate can be generated.

<< Second Embodiment >> FIG. 3 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a second embodiment of the present invention. In this embodiment, in the configuration of the video signal synthesizing device, only the macroblock designated by the transformed macroblock position is subjected to DCT operation and quantization operation to calculate a partially quantized DCT coefficient, and further obtained by a decoder. The quantized DCT coefficients of the decoded image are replaced with the partially quantized DCT coefficients calculated only for the specified macroblock. The other parts are the same as those of the first embodiment, and the duplicated description thereof will be omitted. As shown in FIG. 3, in the video signal synthesizing device of the present embodiment, the quantization unit 14 and the quantized DCT
A coefficient replacement unit 15 is provided between the DCT unit 6 and the VLC unit 9. The quantization unit 14 calculates the partial D from the DCT unit 6
Enter the CT coefficient. The quantization unit 14 determines the VL of the decoder 1
It is connected to the D section 1a and inputs a quantization matrix included in the side information decoded by the VLD section 1a. The quantization unit 14 uses the quantization matrix to
The CT coefficient is subjected to a quantization operation, and is output to the quantized DCT coefficient replacing unit 15 as a partially quantized DCT coefficient. The quantized DCT coefficient replacement unit 15 includes the VLD unit 1a,
And a quantized DCT coefficient of the decoded image decoded by the VLD unit 1a and a transformed macroblock position. The quantized DCT coefficient replacement unit 15 receives the quantized DCT from the VLD unit 1a.
Of the coefficients, only the macroblock specified by the transform macroblock position is replaced with the partially quantized DCT coefficient from the quantization unit 14, and the VLC unit 9 and the inverse quantization unit 1 of the local decoder 10 are used as the quantized DCT coefficients.
0a.

Hereinafter, the operation of the video signal synthesizing apparatus according to this embodiment will be described. In the following description, the decoder 1,
Synthesis unit 2, reference MB detection unit 3, position information conversion unit 4, subtractor 5, DCT unit 6, VLC unit 9, local decoder 1
0 and the motion detection unit 11 perform the same operations as those of the first embodiment, so that the quantization unit 14 and the quantization D
The operation of the CT coefficient replacement unit 15 will be mainly described.
The quantization unit 14 performs a quantization operation on the partial DCT coefficient from the DCT unit 6 using the quantization matrix output from the VLD unit 1a of the decoder 1. Then, the quantization unit 14
Is output to the quantized DCT coefficient replacement unit 15 as a partially quantized DCT coefficient. Thereby, the position information converter 4
The DCT operation and the quantization operation are performed only on the macroblock designated by the position of the converted macroblock. Next, the quantized DCT coefficient replacing unit 15 converts the quantized DCT of the decoded image input from the VLD unit 1 a of the decoder 1.
Of the T coefficients, only the quantized DCT coefficients of the macroblock specified at the transformed macroblock position are replaced with the partially quantized DCT coefficients from the quantization unit 14. After that, the quantized DCT coefficient replacement unit 15
The coefficients are output to the VLC unit 9 and the inverse quantization unit 10a of the local decoder 10 as coefficients. Subsequently, the VLC unit 9 performs a variable length code on the quantized DCT coefficients and the side information including the motion vector from the motion detection unit 11, and performs the original MPEG
The compressed data 103 compressed by the format is output.

As described above, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the DCT section 6 and the quantization section 14 are designated by the position of the transformed macroblock from the position information converting section 4. The DCT operation and the quantization operation of the composite image are performed only for the macroblock. Further, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the quantized DCT coefficient replacement unit 1
Reference numeral 5 denotes only the quantized DCT coefficients of the macroblock designated by the above-described transformed macroblock position, among the quantized DCT coefficients of the decoded image obtained in the decoder 1, and the partially quantized DCT coefficients from the quantizing unit 14. Has been replaced. Thus, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, compared with the video signal synthesizing apparatus of the first embodiment, when recompressing into a variable rate stream (compressed data 103), It is possible to simplify the quantization processing in the case where it is not necessary to control the bit rate of the image to a desired value, and it is possible to perform the video signal synthesis processing in the device at high speed.

In the above-described first and second embodiments, the configuration in which image data is combined with the compressed data compressed in the MPEG format has been described. However, the compressed data to be processed is limited to the MPEG format. Instead, any format may be used as long as the format is a hybrid coding scheme using a combination of intra-frame coding and inter-frame prediction coding. For example, H.261 or H.263
It is also possible to combine the image data with the compressed data compressed in the format and re-compress it in the original format.

<< Third Embodiment >> FIG. 4 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a third embodiment of the present invention. In this embodiment, in the configuration of the video signal synthesizing apparatus, the image data is synthesized with the compressed data using only the intra-frame encoding, and the image data is compressed again. The other parts are the same as those of the second embodiment, and the duplicated description thereof will be omitted. In the following description, J
PEG (Joint Photographic Coding Experts Group)
Combines another image data with the compressed data compressed by
A configuration for recompressing in the original JPEG format will be described. As shown in FIG. 4, the video signal synthesizing apparatus according to the present embodiment includes a decoder 1 that inputs compressed data 200 to be processed from the outside and performs a decoding process on a frame (image) basis.
6. A synthesizing unit 17 for synthesizing the decoded image input from the decoder 16 with the image data 201 input from the outside based on the position information 202 from the outside, and the synthesizing unit 17
In order to perform partial encoding on a macroblock unit for a synthesized image synthesized by the above, a position information conversion unit 18 that converts input position information 202 into a converted macroblock position is provided. Further, the video signal synthesizing device of the present embodiment is connected to the synthesizing unit 17 and the position information converting unit 18, and
DCT unit 19 that performs a DCT operation on the pixel data from 7
Connected to the decoder 16 and the DCT unit 19,
A quantization unit 20 for performing a quantization operation on the partial DCT coefficients from the unit 19 is provided. The video signal synthesizing apparatus according to the present embodiment is connected to the decoder 16, the position information converting unit 18, and the quantizing unit 20, and receives the partial quantized DCT from the quantizing unit 20.
A quantized DCT coefficient replacing unit 21 for replacing with a coefficient, and a VLC unit 22 connected to the quantized DCT coefficient replacing unit 21 for performing a VLC operation on the input quantized DCT coefficient and outputting as compressed data 203 are provided.

The decoder 16 compresses the compressed data 200 according to the decoding method specified in the JPEG format.
To a decoded image, and the sequentially connected V
LD unit 16a, inverse quantization unit 16b, and IDCT unit 16
c. The VLD unit 16a performs variable length decoding on the input compressed data 200, and decodes the compressed data 200 into quantized DCT coefficients and side information. The VLD unit 16a outputs the quantized DCT coefficients to the inverse quantization unit 16b. Further, the VLD unit 16
“a” outputs the quantized DCT coefficient and the quantization matrix included in the decoded side information to the quantized DCT coefficient replacement unit 21 and the quantization unit 20, respectively. The inverse quantization unit 16b inversely quantizes the quantized DCT coefficient from the VLD unit 16a to calculate a DCT coefficient. IDCT unit 16c
Performs an IDCT operation on the DCT coefficient from the inverse quantization unit 16b, and outputs the calculated pixel data to the synthesis unit 17 as an output (decoded image) of the decoder 16. The synthesizing unit 17 receives the decoded image from the decoder 16, and further externally inputs the image data 201 and the position information 202 to be synthesized. The combining unit 17 combines the image data 201 with the decoded image based on the position information 202, and outputs the combined image to the DCT unit 19. Here, the position information 202 uses coordinates on the decoded image, and is configured by coordinates connecting the outline of the shape formed by the corresponding image data 201.
For example, if the shape of the image data 201 is a rectangle, the position information 202 is constituted by the coordinates of four vertices of the rectangle.

The position information converter 18 receives the position information 202 and calculates the position of the macroblock to be partially encoded on the composite image from the input position information 202.
The converted macro block position is output to the DCT unit 19 and the quantized DCT coefficient replacement unit 21. DCT unit 19
Inputs the pixel data of the synthesized image and the converted macroblock position from the synthesizing unit 17 and the position information converting unit 18, respectively. The DCT unit 19 performs a DCT operation on the pixel data based on the transformed macroblock position, and outputs the result to the quantization unit 20 as a partial DCT coefficient.

The quantization unit 20 receives the quantization matrix and the partial DCT from the VLD unit 16a and the DCT unit 19.
Enter each coefficient. The quantization unit 20 performs a quantization operation on the partial DCT coefficient using the quantization matrix, and outputs the result to the quantized DCT coefficient replacement unit 21 as a partially quantized DCT coefficient. The quantized DCT coefficient replacement unit 21
VLD unit 16a, position information conversion unit 18, and quantization unit 2
From 0, the quantized DCT coefficient of the decoded image, the transformed macroblock position, and the partially quantized DCT coefficient are input. Further, the quantized DCT coefficient replacement unit 21 outputs the VL
Of the quantized DCT coefficients from the D unit 16a, only the quantized DCT coefficients of the macroblock designated at the above-described transformed macroblock position are replaced with the partially quantized DCT coefficients from the quantization unit 20 to quantize the DCT coefficients. As VLC
Output to the unit 22. The VLC unit 22 performs a variable length code on the quantized DCT coefficient from the quantized DCT coefficient replacement unit 21 and side information including the quantization matrix, and
The compressed data 203 compressed according to the PEG format is output.

The operation of the video signal synthesizing apparatus according to the present embodiment will be described below. The decoder 16 stores the compressed data 200
Is input, the VLD unit 16a performs variable length decoding on the input compressed data 200 and decodes the compressed data 200 into side information including a quantized DCT coefficient and a quantization matrix. And V
The LD unit 16a converts the quantized DCT coefficient into an inverse quantized unit 16b.
And a quantized DCT coefficient replacement unit 21, and outputs a quantization matrix to the quantization unit 20. Inverse quantization section 16
b dequantizes the quantized DCT coefficients and outputs the DCT coefficients to the IDCT unit 16c. Thereafter, the IDCT unit 16
c performs the IDCT operation on the DCT coefficient and outputs the result to the synthesizing unit 17 as a decoded image. Subsequently, the synthesizing unit 17 synthesizes the image data 201 with the above-described decoded image based on the input position information 202, and outputs the decoded image to the DCT unit 19 as a synthesized image. In addition, the position information conversion unit 18 outputs the position information 20.
2, the position of the macroblock on which the partial encoding is to be performed on the synthesized image is obtained from the input position information 202 and is output to the DCT unit 19 and the quantized DCT coefficient replacement unit 21 as the transformed macroblock position. .

Next, the DCT section 19 performs a DCT operation only on the macroblock specified by the transformed macroblock position among the macroblocks of the pixel data of the synthesized image, and quantizes the operation result as a partial DCT coefficient. Output to the unit 20. Subsequently, the quantization unit 20 converts the partial DCT coefficient from the DCT unit 19 into the VLD unit 1 of the decoder 16.
It quantizes using the quantization matrix obtained by 6a, and outputs it to the quantized DCT coefficient replacement unit 21 as a partially quantized DCT coefficient. After that, the quantized DCT coefficient replacement unit 21 outputs only the quantized DCT coefficients of the macroblock designated by the transformed macroblock position among the quantized DCT coefficients of the decoded image input from the VLD unit 16a. With the partially quantized DCT coefficients from. Then, the quantized DCT coefficient replacement unit 21 outputs the quantized DCT coefficient to the VLC unit 22 as a quantized DCT coefficient. next,
The VLC unit 22 applies a variable length code to the side information including the quantized DCT coefficient and the quantization matrix from the quantized DCT coefficient replacing unit 21 and outputs compressed data 203 compressed according to the original JPEG format.

As described above, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the DCT unit 19 and the quantization unit 20 are specified by the position of the converted macroblock from the position information conversion unit 18. The DCT operation and the quantization operation of the composite image are performed only for the macroblock. Further, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment, the quantized DCT coefficient replacing unit 21 performs the above-described conversion macroblock position conversion on the quantized DCT coefficients of the decoded image obtained in the decoder 16. Only the quantized DCT coefficients of the macroblock specified by
It is replaced with a partially quantized DCT coefficient from the quantization unit 20. As a result, the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment simplify the video signal synthesizing process including the decoding process of the video signal and the compression encoding process of re-compression encoding in the original format. Thus, the amount of calculation can be appropriately reduced, and the calculation can be performed at high speed.

<< Fourth Embodiment >> FIG. 5 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a fourth embodiment of the present invention. In this embodiment, in the configuration of the video signal synthesizing apparatus, only the macroblock at the position where the image data is synthesized is partially decoded, and the above-described image data is synthesized to perform the DCT operation and the quantization operation. The other parts are the same as those of the third embodiment, and the duplicated description thereof will be omitted. As shown in FIG. 5, the video signal synthesizing apparatus according to the present embodiment receives a compressed data 200 to be processed and a converted macroblock position indicating a synthesizing position of image data 201 to be synthesized with the compressed data 200 and partially inputs the converted macroblock position. A decoding unit 23 that performs an appropriate decoding process, a synthesizing unit 24 that synthesizes the partially decoded image input from the decoder 23 with image data 201 input from the outside based on the position information 202 from the outside, In order to apply partial encoding to the synthesized image in units of macroblocks, a position information conversion unit 25 that converts the input position information 202 into the converted macroblock positions is provided. Further, the video signal synthesizing device of the present embodiment is connected to the synthesizing unit 24 and performs a DCT operation on the pixel data from the synthesizing unit 24, and is connected to the decoder 23 and the DCT unit 26. Is provided with a quantization unit 27 for performing a quantization operation on the partial DCT coefficient from. The video signal synthesizing apparatus according to the present embodiment includes a decoder 23, a position information converting unit 2
5, a quantized DCT coefficient replacing unit 28 connected to the quantizing unit 27 and replacing with the partially quantized DCT coefficient from the quantizing unit 27, and the quantized DCT coefficient replacing unit 2
8 is provided with a VLC unit 29 that performs a VLC operation on the input quantized DCT coefficient and outputs the result as compressed data 203.

The decoder 23 compresses the compressed data 200 according to the decoding method specified in the JPEG format.
Into a decoded image, and a position information conversion unit 2
5, the inverse quantization operation and the IDCT operation are performed only on the macroblock designated by the converted macroblock position, and a partially decoded partially decoded image is output. The decoder 23 includes a VLD unit 23a, an inverse quantization unit 23b, and an IDCT unit 23c that are sequentially connected. VLD
The unit 23a performs variable length decoding on the input compressed data 200 and decodes it into quantized DCT coefficients and side information. VLD
The unit 23a outputs the quantized DCT coefficients to the inverse quantization unit 23b. Further, the VLD unit 23a converts the quantized DCT coefficients and the quantization matrix included in the decoded side information into the quantized DCT coefficient replacement unit 21 and the quantizing unit 2.
0 is output. The inverse quantization unit 23b inversely quantizes the quantized DCT coefficient from the VLD unit 23a and calculates a partial DCT coefficient only for the macroblock specified by the transformed macroblock position from the position information conversion unit 25. The IDCT unit 23c performs an IDCT operation on the partial DCT coefficient from the inverse quantization unit 23b, and outputs the calculated pixel data to the combining unit 24 as an output (partially decoded image) of the decoder 23.

The synthesizing section 24 inputs the partially decoded image from the decoder 23, and further inputs image data 201 and position information 202 to be synthesized from outside. The combining unit 24 adds the image data 201 to the partially decoded image based on the position information 202.
Are combined and output to the DCT unit 26 as a partially combined image.
The position information conversion unit 25 receives the position information 202, calculates the position of the macroblock to be partially encoded on the composite image from the input position information 202, and determines the position of the macroblock as the converted macroblock position by the inverse of the decoder 23. The quantization unit 23b,
And a quantized DCT coefficient replacement unit 28. DC
The T unit 26 receives the pixel data of the partially synthesized image from the synthesizing unit 24. The DCT unit 26 adds D to the input pixel data.
A CT operation is performed, and the result is output to the quantization unit 27 as a partial DCT coefficient.

The quantization section 27 receives the quantization matrix and partial DCT from the VLD section 23a and DCT section 26.
Enter each coefficient. The quantization unit 27 performs a quantization operation on the partial DCT coefficient using the quantization matrix, and outputs the result to the quantized DCT coefficient replacement unit 28 as a partially quantized DCT coefficient. The quantized DCT coefficient replacement unit 28
VLD unit 23a, position information conversion unit 25, and quantization unit 2
7, the quantized DCT coefficient of the decoded image, the transformed macroblock position, and the partially quantized DCT coefficient are input. Further, the quantized DCT coefficient replacement unit 28
Of the quantized DCT coefficients from the D unit 23a, only the quantized DCT coefficients of the macroblock designated by the above-described transformed macroblock position are replaced with the partially quantized DCT coefficients from the quantization unit 27, and As VLC
Output to the unit 29. The VLC unit 29 performs variable length coding on the side information including the quantization DCT coefficient and the quantization matrix from the quantization DCT coefficient replacement unit 28, and
The compressed data 203 compressed according to the PEG format is output.

The operation of the video signal synthesizing apparatus according to the present embodiment will be described below. First, the position information conversion unit 25 obtains the position of the macroblock to be subjected to partial encoding on the generated composite image from the input position information 202 and determines the position of the macroblock as a converted macroblock position.
And a quantized DCT coefficient replacement unit 28. Next, when the decoder 23 inputs the compressed data 200, the VLD unit 23a performs variable length decoding on the input compressed data 200 and decodes the compressed data 200 into side information including a quantized DCT coefficient and a quantization matrix. Then, the VLD unit 23a
The quantized DCT coefficient is inversely quantized by the inverse quantization unit 23b and the quantized DCT.
The quantization matrix is output to the coefficient replacement unit 28 and the quantization matrix is output to the quantization unit 27. The inverse quantization unit 23b partially inversely quantizes the quantized DCT coefficient from the VLD unit 23a based on the transformed macroblock position input from the position information conversion unit 25. That is, the inverse quantization unit 23b inversely quantizes the above-described quantized DCT coefficient only for the macroblock specified by the transformed macroblock position, and performs partial DCT.
The coefficient is output to the IDCT section 23c. After that, IDCT
The unit 23c performs an IDCT operation on the above-described partial DCT coefficient, and outputs the result to the synthesizing unit 24 as a partially decoded image.

Subsequently, the synthesizing unit 24 converts the image data 2 into the partially decoded image based on the input position information 202.
01 is synthesized and output to the DCT unit 26 as a partially synthesized image. Next, the DCT unit 26 performs a DCT operation only on the macroblock specified by the transformed macroblock position among the macroblocks of the pixel data of the partial composite image, and uses the operation result as a partial DCT coefficient as the quantization unit 2.
7 is output. Subsequently, the quantization unit 27 includes the DCT unit 26
Of the partial DCT coefficient from the VLD unit 23a of the decoder 23
Is quantized using the quantization matrix obtained by the above, and is output to the quantized DCT coefficient replacement unit 28 as a partially quantized DCT coefficient.

Next, the quantized DCT coefficient replacement unit 28
Is the quantized DC of the decoded image input from the VLD unit 23a.
Of the T coefficients, only the quantized DCT coefficients of the macroblock designated at the transformed macroblock position are replaced with the partially quantized DCT coefficients from the quantization unit 27. After that, the quantized DCT coefficient replacement unit 28
It outputs to the VLC unit 29 as a coefficient. After that, the VLC unit 29 applies a variable length code to the side information including the quantization DCT coefficient and the quantization matrix from the quantization DCT coefficient replacement unit 28, and outputs compressed data 203 compressed in the original JPEG format.

As described above, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present embodiment, the image data 201 is synthesized based on the position information 202 input by the position information conversion unit 25. A converted macroblock position that specifies the position of the macroblock is generated. further,
In the video signal synthesizing apparatus and the video signal synthesizing method according to the present embodiment, the compressed data 200 input to the decoder 23 is variable-length decoded, and only the macroblock specified by the above-described converted macroblock position is subjected to the inverse quantization operation. IDCT
The partially decoded image subjected to the operation is output to the synthesizing unit 24. Thereby, in the video signal synthesizing apparatus and the video signal synthesizing method of the present embodiment of the present embodiment, the decoding process of the compressed data 200 can be appropriately reduced as compared with the third embodiment. The video signal synthesizing process in the device can be performed at high speed.

Although the third and fourth embodiments have been described in connection with a configuration in which image data is combined with compressed data compressed in the JPEG format, the processing target compressed data is limited to the JPEG format. Instead, any format may be used as long as the format is an encoding scheme using only intra-frame encoding. For example, even in the case of the MPEG format, image data is synthesized with compressed data of a moving image composed of only I pictures, and the original MPEG
It is also possible to recompress in format. In addition to the description of the third and fourth embodiments, the decoder is provided with a deshuffling function of rearranging the pixel data after the IDCT operation in units of macroblocks.
Quantized DCT of a decoded image by shuffling CT coefficients
A configuration in which a function of replacing a coefficient is added to the quantized DCT coefficient replacement unit may be used. With this configuration, the SD-VCR (Standard Definition Video Casset) can be used.
eRecorder) format data.

Also, since the video signal synthesizing methods of the video signal synthesizing apparatuses in the first to fourth embodiments can be computer-programmed, the present invention is applied to a computer-readable recording medium. It is also possible to provide a video signal combining method. The recording medium referred to here is a floppy disk, CD-RO
M, DVD, magneto-optical disk, removable hard disk, and the like.

[0056]

As described above, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present invention, the position information conversion unit uses the reference macroblock position from the reference MB detection unit to convert the input position information. A converted macroblock position converted to a macroblock position to be subjected to partial encoding on the composite image is generated. Further, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present invention, the DCT unit and the motion detection unit perform DCT calculation of the synthesized image and motion vector only for the macroblock specified by the above-described converted macroblock position. Are respectively detected. Further, the DCT coefficient replacement unit performs, among the DCT coefficients of the decoded image input from the decoder, only the DCT coefficient of the macroblock designated by the above-described transformed macroblock position.
It is replaced with the partial DCT coefficient from the DCT section. For other macroblocks not specified by the above-described transformed macroblock position, the DCT coefficient replacement unit and the motion detection unit use the DCT coefficients and the motion vectors of the decoded image obtained in the decoder, respectively.
Accordingly, the video signal synthesizing apparatus and the video signal synthesizing method of the present invention simplify the video signal synthesizing process including the decoding process of the video signal and the compression encoding process of re-compression encoding in the original format. , The amount of calculation can be appropriately reduced and can be performed at high speed.

In the video signal synthesizing apparatus and the video signal synthesizing method according to another aspect of the present invention, the DCT unit and the quantization unit are used only for the macroblock designated by the converted macroblock position from the position information conversion unit. , And a DCT operation and a quantization operation of the composite image, respectively.
Further, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present invention, the quantized DCT coefficient replacement unit is specified by the above-described transformed macroblock position in the quantized DCT coefficients of the decoded image obtained in the decoder. Only the quantized DCT coefficients of the macro block are replaced with the partially quantized DCT coefficients from the quantization unit. Accordingly, in the video signal synthesizing apparatus and the video signal synthesizing method of the present invention, when the video signal is compressed again into a stream with a variable rate, that is, when it is unnecessary to control the bit rate of the image to be processed to a desired value. Can be simplified, and the video signal synthesizing process in the device can be performed at high speed.

In the video signal synthesizing apparatus and the video signal synthesizing method according to another aspect of the present invention, the DCT unit and the quantization unit are used only for the macroblock specified by the converted macroblock position from the position information conversion unit. , And a DCT operation and a quantization operation of the composite image, respectively.
Further, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present invention, the quantized DCT coefficient replacement unit is specified by the above-described transformed macroblock position in the quantized DCT coefficients of the decoded image obtained in the decoder. Only the quantized DCT coefficients of the macro block are replaced with the partially quantized DCT coefficients from the quantization unit. Accordingly, the video signal synthesizing apparatus and the video signal synthesizing method of the present invention simplify the video signal synthesizing process including the decoding process of the video signal and the compression encoding process of re-compression encoding in the original format. , The amount of calculation can be appropriately reduced and can be performed at high speed.

In the video signal synthesizing apparatus and the video signal synthesizing method according to another aspect of the present invention, a conversion that specifies the position of a macroblock for synthesizing image data based on the position information input by the position information conversion unit. Generating macro block positions. Further, in the video signal synthesizing apparatus and the video signal synthesizing method according to the present invention, of the compressed data input by the decoder, only the macroblock specified by the above-described converted macroblock position is partially decoded to obtain a partially decoded image. Output to the synthesis unit. Thereby, in the video signal synthesizing apparatus and the video signal synthesizing method of the present invention,
The decoding processing of the compressed data can be appropriately reduced, and the video signal synthesizing processing in the device can be performed at high speed.

Since the video signal synthesizing method of the present invention can be made into a computer program, the video signal synthesizing method of the present invention can be recorded on a computer-executable recording medium and executed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a method of detecting a reference macroblock position in a reference MB detection unit shown in FIG. 1;

FIG. 3 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a second embodiment of the present invention;

FIG. 4 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a third embodiment of the present invention;

FIG. 5 is a block diagram showing a configuration of a video signal synthesizing apparatus according to a fourth embodiment of the present invention.

FIG. 6 is a block diagram showing a configuration of a conventional video signal synthesizing apparatus.

[Explanation of symbols]

 1, 16, 23 Decoder 2, 17, 24 Synthesizing unit 3 Reference MB detecting unit 4, 18, 25 Position information converting unit 5 Subtractor 6, 19, 26 DCT unit 7 DCT coefficient replacing unit 8, 14, 20, 27 Quantum Transformation unit 9, 22, 29 VLC unit 10 Local decoder 11 Motion detection unit 15, 21, 28 Quantized DCT coefficient replacement unit

Continued on the front page F-term (reference) 5C053 FA14 FA24 GA10 GB07 GB23 GB32 GB36 GB37 GB38 HA33 LA11 5C059 KK15 KK37 MA00 MA05 MA23 MC11 MC26 ME01 NN01 NN28 PP05 PP06 PP07 RC19 RF21 SS20 SS26 UA02 UA05 UA39

Claims (18)

[Claims] 1. A video signal synthesizing device for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, comprising: a decoder for decoding compressed data and decoding the decoded image into a decoded image; A combining unit that combines image data with the decoded image from the decoder to generate a combined image; and detects a reference position of a macroblock for performing motion compensation in the decoder and sets the reference position as a reference macroblock position. A reference MB detection unit, and a position for generating a converted macroblock position obtained by converting the position information into a macroblock position to be subjected to partial encoding on the composite image using a reference macroblock position from the reference MB detection unit. An information converting unit, connected to the synthesizing unit, and according to the type of the synthesized image,
A subtractor that outputs pixel data or difference data of the composite image as output data, and among macroblocks of output data from the subtractor,
Performing a DCT operation on the macroblock specified by the converted macroblock position from the position information conversion unit,
A DCT unit for obtaining a partial DCT coefficient; and a DCT coefficient of a macroblock specified by a transform macroblock position from the position information transform unit among DCT coefficients of a decoded image input from the decoder.
A DCT coefficient replacement unit that replaces the partial DCT coefficient from the unit to obtain a combined DCT coefficient; a quantization unit that quantizes the combined DCT coefficient from the DCT coefficient replacement unit to obtain a quantized DCT coefficient; Again decodes the quantized DCT coefficients of the composite image from the first reference image to generate a first reference image that has not been subjected to motion compensation for detecting a motion vector of the composite image, and adds a motion to the first reference image. A local decoder that performs compensation and generates a second reference image for obtaining difference data of the composite image; and a motion vector of the composite image according to the type of the composite image from the position information conversion unit. A motion detection unit that detects a macroblock specified by a transform macroblock position; a quantized DCT coefficient from the quantization unit; and a motion from the motion detection unit. VLC for applying a variable length code to side information including a vector and outputting compressed data of the composite image
A video signal synthesizing device, comprising: 2. A video signal synthesizing apparatus for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, comprising: a decoder for decoding compressed data and decoding the decoded data into a decoded image; A combining unit that combines image data with the decoded image from the decoder to generate a combined image; and detects a reference position of a macroblock for performing motion compensation in the decoder and sets the reference position as a reference macroblock position. A reference MB detection unit, and a position for generating a converted macroblock position obtained by converting the position information into a macroblock position to be subjected to partial encoding on the composite image using a reference macroblock position from the reference MB detection unit. An information conversion unit, connected to the synthesis unit, and according to a type of the synthesized image,
A subtractor that outputs pixel data or difference data of the composite image as output data, and among macroblocks of output data from the subtractor,
Performing a DCT operation on the macroblock specified by the converted macroblock position from the position information conversion unit,
A DCT unit for obtaining a partial DCT coefficient; and a quantization unit for performing a quantization operation on the partial DCT coefficient from the DCT unit using a quantization matrix of the decoded image input from the decoder to obtain a partially quantized DCT coefficient. Among the quantized DCT coefficients of the decoded image input from the decoder, the quantized DCT coefficients of the macroblock designated by the transformed macroblock position from the position information transforming unit, and the partially quantized DCT from the quantizing unit. Replace with coefficients,
A quantized DCT coefficient replacement unit that sets a quantized DCT coefficient; and a motion compensation for detecting a motion vector of the synthesized image by decoding again the quantized DCT coefficient of the synthesized image from the quantized DCT coefficient replacement unit. A local decoder that generates a first reference image that has not been subjected to motion, performs motion compensation on the first reference image, and generates a second reference image for obtaining difference data of the synthesized image; A motion detection unit that detects a motion vector of the synthesized image according to the type of the image for a macroblock specified by a converted macroblock position from the position information conversion unit; and a quantization unit that performs quantization from the quantized DCT coefficient replacement unit. A VL that performs variable-length coding on side information including a DCT coefficient and a motion vector from the motion detection unit, and outputs compressed data of the composite image. Video signal synthesizing apparatus characterized by comprising: a part, the. 3. The compressed data is compressed by a compression format including MPEG, H.261, and H.263 formats using a hybrid coding method combining intra-frame coding and inter-frame predictive coding. The video signal synthesizing device according to claim 1, wherein: 4. A video signal synthesizing apparatus for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein the decoder decodes the compressed data and decodes the decoded data into a decoded image; A combining unit that combines image data with the decoded image from the decoder to generate a combined image; and a converted macroblock that converts the position information into a macroblock position that performs partial encoding on the combined image. A position information conversion unit that generates a position, and a macroblock of pixel data from the synthesis unit.
Performing a DCT operation on the macroblock specified by the converted macroblock position from the position information conversion unit,
A DCT unit for obtaining a partial DCT coefficient; and a quantization unit for performing a quantization operation on the partial DCT coefficient from the DCT unit using a quantization matrix of the decoded image input from the decoder to obtain a partially quantized DCT coefficient. Among the quantized DCT coefficients of the decoded image input from the decoder, the quantized DCT coefficients of the macroblock designated by the transformed macroblock position from the position information transforming unit, and the partially quantized DCT from the quantizing unit. Replace with coefficients,
A quantized DCT coefficient replacement unit that is a quantized DCT coefficient; and a variable length code is applied to the quantized DCT coefficient from the quantized DCT coefficient replacement unit and side information including the quantization matrix to compress the synthesized image. A video signal synthesizing device, comprising: a VLC unit that outputs data. 5. A video signal synthesizing device for synthesizing another video signal with a compressed video signal and compressing the video signal again, the macroblock performing partial encoding on a synthesized image for generating input position information A position information conversion unit for generating a converted macroblock position converted to the position of, and variable length decoding of the compressed data, and among the obtained quantized DCT coefficients, designated by the converted macroblock position from the position information conversion unit. A decoder that performs an inverse quantization operation and an IDCT operation on the quantized DCT coefficient of the macroblock generated to generate a partially decoded image that is partially decoded, and, based on the position information, converts the image into a partially decoded image from the decoder. A synthesizing unit for synthesizing data to generate a partially synthesized image; and a macro block of pixel data from the synthesizing unit.
Performing a DCT operation on the macroblock specified by the converted macroblock position from the position information conversion unit,
A DCT unit for obtaining a partial DCT coefficient; and a quantization unit for performing a quantization operation on the partial DCT coefficient from the DCT unit using a quantization matrix of the decoded image input from the decoder to obtain a partially quantized DCT coefficient. Among the quantized DCT coefficients of the decoded image input from the decoder, the quantized DCT coefficients of the macroblock designated by the transformed macroblock position from the position information transforming unit, and the partially quantized DCT from the quantizing unit. Replace with coefficients,
A quantized DCT coefficient replacement unit that is a quantized DCT coefficient; and a variable length code is applied to the quantized DCT coefficient from the quantized DCT coefficient replacement unit and side information including the quantization matrix to compress the synthesized image. A video signal synthesizing device, comprising: a VLC unit that outputs data. 6. The compressed data according to claim 4, wherein the compressed data is compressed by a compression format including JPEG using only intra-frame encoding and an SD-VCR format. Video signal synthesizer. 7. A video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein a decoding step of decoding the compressed data and decoding the decoded data into a decoded image. A combining step of combining image data with the decoded image based on the position information to generate a combined image; detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the decoding step; A reference MB detection step as a position; a converted macroblock position obtained by converting the position information into a macroblock position to be partially encoded on the composite image using the reference macroblock position detected in the reference MB detection step. A position information conversion step of generating, according to the type of the synthesized image generated in the synthesis step,
Outputting pixel data or difference data of the composite image as output data, and among macroblocks of the output data, a macroblock designated by the converted macroblock position
Performing a CT operation to obtain a partial DCT coefficient; and obtaining, from the DCT coefficients of the decoded image decoded in the decoding step, a DCT coefficient of a macroblock designated by the transformed macroblock position in the DCT operation step. D, which is replaced with the obtained partial DCT coefficient and becomes a combined DCT coefficient.
CT coefficient replacement step, synthesized DCT obtained in the DCT coefficient replacement step
A quantization step of quantizing the coefficients to obtain a quantized DCT coefficient; a quantized DCT of the composite image obtained in the quantization step
Decoding the coefficients again to generate a first reference image that has not been subjected to motion compensation for detecting a motion vector of the composite image, and performs motion compensation on the first reference image;
Generating a second reference image for obtaining the difference data of the composite image, according to a type of the composite image generated in the composite step,
A motion detection step of detecting a motion vector of the synthesized image for a macroblock specified at the position of the converted macroblock, a quantized DCT coefficient obtained in the quantization step, and a motion vector obtained in the motion detection step A variable-length code step of applying a variable-length code to the side information including: generating compressed data of the composite image. 8. A video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein a decoding step of decoding the compressed data and decoding the decoded data into a decoded image. A combining step of combining image data with the decoded image based on the position information to generate a combined image; detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the decoding step; A reference MB detection step as a position; a converted macroblock position obtained by converting the position information into a macroblock position to be partially encoded on the composite image using the reference macroblock position detected in the reference MB detection step. A position information converting step of generating, according to the type of the synthesized image generated in the synthesizing step,
Outputting pixel data or difference data of the composite image as output data, and among macroblocks of the output data, a macroblock designated by the converted macroblock position
A DCT operation step of performing a CT operation to obtain a partial DCT coefficient; a partial D obtained in the DCT operation step using a quantization matrix of the decoded image decoded in the decoding step
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
Replacing the quantized DCT coefficients as CT coefficients, decoding the quantized DCT coefficients of the synthesized image obtained in the quantized DCT coefficient replacement step again, and performing motion compensation for detecting a motion vector of the synthesized image. Generating a first reference image that is not present, and performing motion compensation on the first reference image to generate a second reference image for obtaining difference data of the composite image; Depending on the type of synthesized image
A motion detection step of detecting a motion vector of the synthesized image for a macroblock designated by the transformed macroblock position; a quantized DCT coefficient obtained in the quantized DCT coefficient replacement step; A variable length code step of applying a variable length code to the side information including the motion vector to generate compressed data of the synthesized image. 9. The compressed data compressed by a compression format including MPEG, H.261, and H.263 formats using a hybrid coding method combining intra-frame coding and inter-frame predictive coding. The video signal synthesizing method according to claim 7, wherein: 10. A video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein a decoding step of decoding the compressed data and decoding the decoded data into a decoded image. A combining step of combining image data with the decoded image based on the position information to generate a combined image, and a converted macroblock position obtained by converting the position information into a macroblock position for performing partial encoding on the combined image. Performing a DCT operation on a macroblock designated by the converted macroblock position, out of the macroblocks of the pixel data of the synthesized image generated in the synthesis step,
A DCT calculation step for obtaining a partial DCT coefficient;
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
A variable length code is added to side information including a quantized DCT coefficient replacing step as a CT coefficient, a quantized DCT coefficient obtained in the quantized DCT coefficient replacing step, and a quantization matrix of a decoded image decoded in the decoding step. A variable-length code step of generating compressed data of the composite image. 11. A video signal synthesizing method for synthesizing another video signal with a compressed video signal and compressing the compressed video signal again, wherein the macroblock performs partial encoding on a synthesized image for generating input position information. A position information conversion step of generating a converted macroblock position converted to the position of, the variable length decoding of the compressed data, and a quantization DCT of the macroblock designated by the conversion macroblock position among the obtained quantized DCT coefficients. Inverse quantization operation and ID for coefficients
A decoding step of performing a CT operation to generate a partially decoded partially decoded image; and a combining step of combining the partially decoded image decoded in the decoding step with image data based on the position information to generate a partially combined image. A DCT calculation step of performing a DCT calculation on a macroblock specified at the converted macroblock position among macroblocks of pixel data of the partial composite image generated in the synthesis step to obtain a partial DCT coefficient; Using the quantization matrix of the decoded image, the part D obtained in the DCT operation step
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
A variable length code is added to a quantized DCT coefficient replacing step as a CT coefficient, a quantized DCT coefficient obtained in the quantized DCT coefficient replacing step, and side information including a quantization matrix of a decoded image decoded in the decoding step. A variable-length code step of generating compressed data of the composite image. 12. The compressed data according to claim 10, wherein the compressed data is compressed by a compression format including JPEG using only intra-frame encoding and an SD-VCR format. Video signal synthesis method. A decoding step of decoding the compressed data to decode the decoded image into a decoded image; synthesizing the decoded image decoded in the decoding step based on the position information to generate a synthesized image; A reference MB detection step of detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the step, and setting the reference position as a reference macroblock position; A position information conversion step of generating a converted macroblock position obtained by converting information into a macroblock position for performing partial encoding on the composite image, according to a type of the composite image generated in the composite step,
Outputting pixel data or difference data of the composite image as output data, and among macroblocks of the output data, a macroblock designated by the converted macroblock position
Performing a CT operation to obtain a partial DCT coefficient; and obtaining, from the DCT coefficients of the decoded image decoded in the decoding step, a DCT coefficient of a macroblock designated by the transformed macroblock position in the DCT operation step. D, which is replaced with the obtained partial DCT coefficient and becomes a combined DCT coefficient.
CT coefficient replacement step, synthesized DCT obtained in the DCT coefficient replacement step
A quantization step of quantizing the coefficients to obtain a quantized DCT coefficient; a quantized DCT of the composite image obtained in the quantization step
Decoding the coefficients again to generate a first reference image that has not been subjected to motion compensation for detecting a motion vector of the composite image, and performs motion compensation on the first reference image;
Generating a second reference image for obtaining the difference data of the composite image, according to a type of the composite image generated in the composite step,
A motion detecting step of detecting a motion vector of the synthesized image for a macroblock specified at the position of the converted macroblock; a quantized DCT coefficient obtained in the quantization step; and a motion vector obtained in the motion detecting step. A variable-length code step of applying a variable-length code to the side information including: and generating compressed data of the synthesized image. 14. A decoding step of decoding the compressed data to decode the decoded image into a decoded image; a combining step of combining image data with the decoded image decoded in the decoding step based on position information to generate a combined image; A reference MB detection step for detecting a reference position of a macroblock for performing motion compensation on the decoded image decoded in the step, and setting the reference position as a reference macroblock position; using the reference macroblock position detected in the reference MB detection step, A position information conversion step of generating a converted macroblock position in which information is converted into a macroblock position for performing partial encoding on the synthesized image,
Outputting pixel data or difference data of the composite image as output data, and among macroblocks of the output data, a macroblock designated by the converted macroblock position
A DCT operation step of performing a CT operation to obtain a partial DCT coefficient; a partial D obtained in the DCT operation step using a quantization matrix of the decoded image decoded in the decoding step
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
Replacing the quantized DCT coefficients as CT coefficients, decoding the quantized DCT coefficients of the synthesized image obtained in the quantized DCT coefficient replacement step again, and performing motion compensation for detecting a motion vector of the synthesized image. Generating a first reference image that is not present, and performing motion compensation on the first reference image to generate a second reference image for obtaining difference data of the composite image; Depending on the type of synthesized image
A motion detection step of detecting a motion vector of the synthesized image for a macroblock specified at the position of the converted macroblock, a quantized DCT coefficient obtained in the quantized DCT coefficient replacement step, and a motion vector obtained in the motion detection step. A variable length code step of applying a variable length code to the side information including the motion vector to generate compressed data of the synthesized image. 15. The compressed data may be MPEG, H.261, or H.263 using a hybrid coding scheme combining intra-frame coding and inter-frame predictive coding.
14. The image processing apparatus according to claim 13, wherein the data is compressed by a compression format including the format.
A recording medium recording the video signal synthesizing program according to 4. 16. A decoding step of decoding compressed data and decoding the decoded data into a decoded image; a combining step of combining image data with the decoded image decoded in the decoding step based on position information to generate a combined image; A position information conversion step of generating a converted macroblock position obtained by converting information into a macroblock position for performing partial encoding on the composite image, among the macroblocks of pixel data of the composite image generated in the synthesis step, Performing a DCT operation on the macroblock specified at the transform macroblock position,
A DCT calculation step for obtaining a partial DCT coefficient;
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
A variable length code is added to a quantized DCT coefficient replacing step as a CT coefficient, a quantized DCT coefficient obtained in the quantized DCT coefficient replacing step, and side information including a quantization matrix of a decoded image decoded in the decoding step. And a variable-length code step of generating compressed data of the composite image. 17. A position information conversion step of generating a converted macroblock position converted to a position of a macroblock to be subjected to partial encoding on a synthesized image for generating input position information. Of the quantized DCT coefficients obtained, an inverse quantization operation and ID
A decoding step of performing a CT operation to generate a partially decoded partially decoded image; and a combining step of combining the partially decoded image decoded in the decoding step with image data based on the position information to generate a partially combined image. A DCT calculation step of performing a DCT calculation on a macroblock specified at the converted macroblock position among macroblocks of pixel data of the partial composite image generated in the synthesis step to obtain a partial DCT coefficient; Using the quantization matrix of the decoded image, the part D obtained in the DCT operation step
Performing a quantization operation on the CT coefficients to obtain a partially quantized DCT coefficient, and among the quantized DCT coefficients of the decoded image decoded in the decoding step, the quantization of the macroblock designated by the transform macroblock position The quantized DCT coefficient is replaced with the partially quantized DCT coefficient obtained in the quantization step, and the quantization D
A variable length code is added to a quantized DCT coefficient replacing step as a CT coefficient, a quantized DCT coefficient obtained in the quantized DCT coefficient replacing step, and side information including a quantization matrix of a decoded image decoded in the decoding step. And a variable-length code step of generating compressed data of the composite image. 18. The compressed data according to claim 16, wherein the compressed data is compressed by a compression format including JPEG using only intra-frame encoding and an SD-VCR format. A recording medium on which a video signal synthesizing program is recorded.