Decoding apparatus using tool information for constructing a decoding algorithm Number:7,194,136 from the United States Patent and Trademark Office (PTO) owispatent A coding and decoding apparatus is constructed so that the coding side transmits coded data together with identifying information for identifying the device of decoding the coded data, and the decoding side is capable of storing a number of decoding schemes so as to perform decoding based on one of the previously stored schemes. The apparatus further has devices for storing the received tools and tool-correspondent information which numerically represents the capacities of the tools so that it can make a comparison between the decoding capacity and the processing capacities of the tools to determine the possibility of the operations of the received tools. Further, a set of the tools are hierarchized so that the coded data produced by the n-ranked tool can be decoded by the (n+1)-ranked tool. Alternatively, the tools are defined in a hierarchical manner so that the decoding tools installed in the decoding apparatus will be able to assure the minimum quality and the requested decoding process can be performed by the received decoding tool. Further, the identification code of the decoding scheme used can be transmitted as required so that the decoding scheme can be expanded by transmitting the differential information from the basic decoding scheme.<H constructing, information, Decoding, appara, 7194136.html, Patent, pto, 7194136

Title: Decoding apparatus using tool information for constructing a decoding algorithm

Abstract: A coding and decoding apparatus is constructed so that the coding side transmits coded data together with identifying information for identifying the device of decoding the coded data, and the decoding side is capable of storing a number of decoding schemes so as to perform decoding based on one of the previously stored schemes. The apparatus further has devices for storing the received tools and tool-correspondent information which numerically represents the capacities of the tools so that it can make a comparison between the decoding capacity and the processing capacities of the tools to determine the possibility of the operations of the received tools. Further, a set of the tools are hierarchized so that the coded data produced by the n-ranked tool can be decoded by the (n+1)-ranked tool. Alternatively, the tools are defined in a hierarchical manner so that the decoding tools installed in the decoding apparatus will be able to assure the minimum quality and the requested decoding process can be performed by the received decoding tool. Further, the identification code of the decoding scheme used can be transmitted as required so that the decoding scheme can be expanded by transmitting the differential information from the basic decoding scheme.

Patent Number: 7,194,136 Issued on 03/20/2007 to Makiyama, et al.

Inventors: Makiyama; Takeshi (Higashihiroshima, JP), Sato; Seiji (Toride, JP), Koizumi; Noritaka (Chiba, JP), Uchiumi; Tadashi (Chiba, JP)
Assignee: Sharp Kabushiki Kaisha (Osaka, JP)

Appl. No.: 10/728,866

Filed: December 8, 2003

Related U.S. Patent Documents


Application Number	Filing Date	Patent Number	Issue Date
09407880	Sep., 1999	6687409
08727787	Nov., 1999	5987181

Foreign Application Priority Data


Oct 12, 1995 [JP]			7-264127
Oct 25, 1995 [JP]			7-277463
Nov 28, 1995 [JP]			7-308915
Nov 29, 1995 [JP]			7-310667

Current U.S. Class: 382/239 ; 382/233; 382/251

Current International Class: G06K 9/36 (20060101)

Field of Search: 382/232-233,236,239-240,248,250-251 375/240.02-240.03,240.11-240.25 348/231.6 713/189 358/426.01-426.02,426.12-426.14

References Cited [Referenced By]

U.S. Patent Documents


5111294	May 1992	Asai et al.
5138459	August 1992	Roberts et al.
5198900	March 1993	Tsukagoshi
5235419	August 1993	Krause
5260783	November 1993	Dixit
5371734	December 1994	Fischer
5376968	December 1994	Wu et al.
5398277	March 1995	Martin, Jr.
5502497	March 1996	Yamaashi et al.
5521717	May 1996	Maeda
5619438	April 1997	Farley
5640198	June 1997	Makiyama et al.
5689346	November 1997	Noda
5692012	November 1997	Virtamo
5802315	September 1998	Uchiumi et al.
5881244	March 1999	Uchiumi et al.
5987181	November 1999	Makiyama
6310981	October 2001	Makiyama et al.
6636970	October 2003	Akiyama et al.

Foreign Patent Documents


0577337	Jan., 1994	EP
4-8064	Apr., 1990	JP
04-008064	Jan., 1992	JP
5-316369	Nov., 1993	JP
5-316369	Nov., 1993	JP
7-203211	Apr., 1995	JP
WO 96/02895	Feb., 1996	WO

Primary Examiner: Sherali; Ishrat
Attorney, Agent or Firm: Birch, Stewart, Kolasch & Birch, LLP

Parent Case Text

This application is a Divisional of co-pending U.S. Pat. application Ser. No. 09/407,880, filed on Sep. 29, 1999 and issued as U.S. Pat. No. 6,687,409 on Feb. 3, 2004, which is a Divisional of prior U.S. Pat. application Ser. No. 08/727,787, filed on Oct. 8, 1996 and issued as U.S. Pat. No. 5,987,181 on Nov. 16, 1999, the entire contents of which are hereby incorporated by reference and for which priority is claimed under 35 U.S.C. .sctn. 120; and this application claims priority of Application Nos. 7-264127 filed in Japan on Oct. 12, 1995; 7-277463 filed in Japan on Oct. 25, 1995; 7-308915 filed in Japan on Nov. 28, 1995; and 7-310667 filed in Japan on Nov. 29, 1995 under 35 U.S.C. .sctn. 119.

Claims

What is claimed is:

1. An image data coding apparatus comprising: a motion compensator; a transformer; a quantizer comprising at least two different quantizing tools; an inverse quantizer comprising at least two different inverse quantizing tools; and an inverse transformer; said image data coding apparatus transmitting information indicating tools constituting a decoding algorithm for decoding a coded image data including information indicating an inverse quantizing tool for inverse quantizing the coded image data.

2. The image data coding apparatus of claim 1 wherein said at least two different inverse quantizing tools comprise a first inverse quantizing tool having a first processing capability and a second inverse quantizing tool having a second processing capability different than the first processing capability.

3. The image data coding apparatus of claim 1 wherein said information indicating an inverse quantizing tool comprises information identifying an inverse quantizing tool.

4. The image data coding apparatus of claim 1 wherein said information indicating an inverse quantizing tool comprises information indicating a processing capability of an inverse quantizing tool.

5. The image data coding apparatus of claim 2 wherein said information indicating an inverse quantizing tool comprises information specifying said first inverse quantizing tool or said second inverse quantizing tool.

6. An image data decoding apparatus comprising: a motion compensator; an inverse quantizer comprising at least two different inverse quantizing tools; and an inverse transformer; said image data decoding apparatus receiving information indicating tools constituting a decoding algorithm for decoding a coded image data including information indicating an inverse quantizing tool for inverse quantizing the coded image data.

7. The image data decoding apparatus of claim 6 wherein said at least two different inverse quantizing tools comprise a first inverse quantizing tool having a first processing capability and a second inverse quantizing tool having a second processing capability different than the first processing capability.

8. The image data decoding apparatus of claim 6 wherein said information indicating an inverse quantizing tool comprises information identifying an inverse quantizing tool.

9. The image data decoding apparatus of claim 6 wherein said information indicating an inverse quantizing tool comprises information indicating a processing capability of an inverse quantizing tool.

10. The image data decoding apparatus of claim 7 wherein said information indicating an inverse quantizing tool comprises information specifying said first inverse quantizing tool or said second inverse quantizing tool.

11. The image data coding apparatus of claim 2 wherein said information indicating an inverse quantizing tool comprises information identifying an inverse quantizing tool.

12. The image data decoding apparatus of claim 7 wherein said information indicating an inverse quantizing tool comprises information identifying an inverse quantizing tool.

Description

BACKGROUND OF THE INVENTION

(1) Field of the Invention

The present invention relates to a coding and decoding apparatus for coding information such as an image signal etc. to transmit it and decoding the coded data. More detailedly, the present invention relates to a coding and decoding apparatus which enables communication between coding and decoding tools having different processing capacities and in which the coding apparatus transmits not only the coded data but also coding information for the construction of a decoding scheme as the means of decoding the coded data and the decoding apparatus receives the coding information together with coded data and reconstructs the decoding scheme based on the coding information so as to decode the received coded data. Further, the present invention is directed to a coding and decoding technology for performing the communication in a coding and decoding apparatus between the transmitting and receiving devices having different capacities in the case where an algorithm includes various coding and decoding tools such as near-future image coding schemes represented by the MPEG4 etc., and more particularly relates to a coding and decoding apparatus which enables simultaneous transmission of coded data and tool information for constructing the algorithm for decoding the coded data in order to realize a hierarchical coding and decoding operation.

(2) Description of the Background Art

In recent years, a wide spread of ISDN (Integrated Services Digital Network) has realized image communication services as a new communication service. Examples of the services include the video phone and video conference system, etc. On the other hand, the development of the mobile communication networks represented by the PHS and the FPLMTS, accelerates demands for further improvement and variations of the services and portability of the devices.

In general, in the case where image information as in the video phone or video conference system is transmitted, the amount of image information is very large. However, due to the line speed used for the transmission and the cost problem, the image information to be transmitted needs to be compressed and coded so that the amount of information can be reduced.

As to the coding schemes for compressing image information, JPEG (Joint Photographic coding Experts Group) has already been standardized internationally for a still image coding system, H.261 for a motion picture coding scheme, and MPEG1 (Moving Picture Coding Experts Group) and MPEG2 for motion picture coding schemes. Further, MPEG4 is now being standardized as a coding scheme of very low-bit rate of 64 kbps or below.

In the current coding schemes such as JPEG, H.261, MPEG1, MPEG2, coding is performed following the specified algorithm. However, the MPEG4 is planned to flexibly deal with various applications and encode each of the applications in its optimal scheme. For this purpose the MPEG4 needs to have many tools (such as transformer, quantizer, inverse transformer, inverse quantizer, etc.) for its coder so that a suitable combination of them will be selected to perform coding.

FIG. 1A is a conceptual view showing the structure of a coding data stream which is formed by coding (compressing) image data based on the H.261 scheme. Each piece of the coded data such as motion vector information, DCT-coefficient, quantization step, etc., shown in FIG. 1A is image data which has been coded (compressed) based on a fixed coding algorithm in the coder, while the decoder has a decoding algorithm fixed corresponding to the coding algorithm so that the received pieces of the coded data will be decoded.

FIG. 1B is a conceptual view showing the structure of a coding data stream which is formed by coding (compressing) image data based on a coding scheme such as MPEG4 etc. whose algorithm is flexible. The coding data stream as shown in FIG. 1B is composed of coded (compressed) image data such as motion vector information 2, transform coefficient 4, motion vector information 6, transform coefficient 8 and quantization step 10 etc., and tool information such as motion compensation tool 1, inverse transform tool 3, motion compensation tool 5, inverse transform tool 7 and quantizing tool 9, etc., for decoding respective image data. FIG. 1B illustrates the details of the motion vector information, DCT-coefficient and quantization step at the leading end of the coding data stream of FIG. 1A. In this case, each piece of the tool information such as motion compensation 1 etc., is allowed to be selected from a number of types of the tool information so that it is possible to freely select a desired combination of the tool information. Accordingly, the coder transmits the tool information which has been used for coding as well as the image data to the decoder. The decoder, upon the decoding of the image data received, will decode the coded image data using the tool information transmitted from the coder.

FIG. 1C is a block diagram showing an example of a conventional coding and decoding apparatus based on H.261. This coding and decoding apparatus is composed of a controller 6a for controlling the entire apparatus, a coder 7a for coding based on H.261, and a decoder 8a for decoding the information which has been coded based on H.261, and a tool storage section 9a consisting of memories for storing tool information.

These coding and decoding processes can be realized by a dedicated hardware device with software installed therein or by an appropriate program executed in a general-purpose processor with a compiler.

First, description will be made of a method using a dedicated hardware device with software installed therein. FIG. 2 is a block diagram showing the configuration of coder 7a of FIG. 1C for yielding the coded data shown in FIG. 1A, based on H.261. In FIG. 2, the coder is composed of: a coding controller 11 for the control of coding; a transformer 12 for performing the DCT; a quantizer 13 for quantizing the coefficients transformed by transformer 12; an inverse quantizer 14 for performing inverse quantization of the coefficients quantized in quantizer 13; an inverse transformer 15 for performing the inverse DCT; a memory 16; and a loop filter 17. Here, memory 16 has the function of causing a variable delay for motion compensation, used when the inter-frame prediction for motion compensation is performed. Filter 17 is the loop filter capable of performing the on/off operation for each of macro blocks.

When the coding algorithm for generating the coding data stream shown in FIG. 1A is executed by the dedicated hardware device with software, the tool functions constituting the algorithm are carried out by software and the dedicated hardware components as shown in FIG. 2, namely, coding controller 11, transformer 12, quantizer 13, inverse quantizer 14, inverse transformer 15, memory 16 having the function of causing a variable delay for motion compensation, and loop filter 17. FIG. 3 is a block diagram showing the configuration of decoder 8a shown in FIG. 1C for decoding the coded data based on H.261. This decoder commonly has the constituents of the coder shown in FIG. 2, and the same components as those in the coder of FIG. 2 are designated at the same reference numerals. Specifically, in FIG. 3, a reference numerals 14 designates an inverse quantizer, 15 an inverse transformer, 16 a memory having the function of causing a variable delay for motion compensation, and 17 a loop filter.

The coded data by the coder shown in FIG. 2 is inverse quantized by inverse quantizer 14, and the signal is then made to undergo the inverse DCT in inverse transformer 15. Here, memory 16 and loop filter 17 are used when the motion compensated prediction coding data is decoded.

When several kinds of algorithms need to be processed using the scheme which performs the coding operation based on a fixed algorithm such as H.261 etc. as stated above, an individual hardware device with software is needed to execute each of the algorithms. FIG. 4 is block diagram showing the structure of a coder which codes the signal of a motion picture based on H.261 and the signal of a still image based on JPEG. For example, when a motion picture is coded based on H.261 and a still image is coded based on JPEG, the coder should have the configuration as shown in FIG. 4, which includes two individual coders, namely a H.261 coder 20 and a JPEG coder 21. In FIG. 4, H.261 coder 20 and JPEG coder 21 receive the motion picture data and the still image data respectively to output coded data of compressed data.

When the algorithm for generating the coded data shown in FIG. 1B is executed by a dedicated hardware device with software, a coder for executing this algorithm is realized by the one shown in FIG. 2 in which the circuit block designated at 18 is configured by the configuration shown in FIG. 5. In this case, the coder has plural types for each of the tools, or, transformer 12, quantizer 13, inverse quantizer 14, and inverse transformer 15. In this configuration, one desired type is selected for each of the tools (one type from transformer tools A to X, one type from quantizer tools A to X, one type from inverse quantizer tools A to X and one type from inverse transformer tools A to X) to perform a coding process.

The decoder for decoding the coding data stream shown in FIG. 1B is realized in a decoder shown in FIG. 3 in which the circuit block designated at 19 is replaced by a circuit block 22 in FIG. 5. In this case, the decoder has plural types for each of the tools, or, inverse quantizer 14, and inverse transformer 15. In this configuration, one desired type is selected fore ach of the tools (one type from inverse quantizer tools A to X and one type from inverse transformer tools A to X) to perform a decoding process.

In this decoding process, each piece of the tool information shown in FIG. 1B, for motion compensation tool 1, inverse transforming tool 3, motion compensation tool 5, inverse transforming tool 7 and quantizing tool 9 is sent to a controller 23, and each piece of the image data, which follows the corresponding tool information, specifically, of motion vector information 2, transform coefficient 4, motion vector information 6 and transform coefficient 8, is sent to the corresponding tools where each image data is processed. At the time, controller 23 selects one of the tools (one from inverse quantizing tools A to X and one from inverse transforming tools A to X shown in FIG. 5) based on the corresponding tool information. in this way, each piece of the image data is processed through the tool selected by controller 23 and is decoded.

However, this method needs a dedicated device with software for each of the tools, thus the scale of the decoder tends to become large. To make matters worse, if the decoder receives the data which has been processed by a tool that is not provided for the decoder, it is impossible to decode the data itself. To solve this problem, a way that can be considered is one in which parts received should be compiled to prepare a processing program and the data should be decoded by a general-purpose processor.

Next, description will be made of a method of achieving the decoding process by executing a suitable program using a general-purpose processor with a compiler. Now, referring to FIG. 6, description will be made of a case where the coding data stream having the structure shown in FIG. 1B is decoded. FIG. 6 is a block diagram showing the structure of a decoder composed of a general-purpose processor 24 and a compiler 25. When all the tool information as shown in FIG. 1B, which includes a motion compensation tool 1, inverse transforming tool 3, motion compensation tool 5, inverse transforming tool 7 and quantizing tool 9, etc., is given to compiler 25, the compiler will prepare a processing program for controlling the operation of general-purpose processor 24. Each piece of the image data, which follows the corresponding tool information, specifically, motion vector information 2, transform coefficient 4, motion vector information 6, transform coefficient 8, quantization step 10, is given to general-purpose processor 24. Then, general-purpose processor 24 processes, with the processing program prepared by the compiler 25, the coded image data following the tool information so as to decode it for producing its decoded data.

In the case where the capacity of the decoding apparatus for processing a certain algorithm is lower than the total processing capacity for all the tools constituting the algorithm requested by the coder side, even if the tools transmitted from the coder is stored at the decoder side, the received data cannot be decoded exactly due to the inferior processing capacity of the decoder side. Thus the memory in the tool storage is also used just in vain.

Also in the conventional coding and decoding apparatus, when the tools which were used in the coder side are compared to the tools which are stored at the decoder side, the tools themselves should be compared to each other; this process required a very long period of time.

In the case where a new algorithm is used to decode the coded information, even if the tools for the algorithm are equivalent to those which have been previously stored, the decoder should receive the tools once again; this process also considerably lengthened the transmission/reception time.

In this way, when a video signal etc. is coded, the coding tools having suitable coding capacities to the quality of the reproduction image required by the decoder side, are selected to perform the coding operation. When the thus obtained coded data is decoded, it is necessary that the decoder should use decoding tools having decoding capacities (i.e. processing capacities) which correspond to the coding capacities (i.e. processing capacities) for the coding tools which were used for the coding operation. Processing capacity indicates a resource that is necessary for coding, decoding or both. For example, coding capacity may be expressed as processing capacity for coding. If these tools on the decoder side do not have the processing capacities for the tools on the coder side, the coded data cannot be decoded, thus making it impossible to establish the communication.

An example of the algorithm for the frame predictive coding will hereinbelow be described. Frame predictive coding shall mean inter-frame predictive coding, intra-frame predictive coding, or both as the context requires. Inter-frame predictive coding refers to any technique for data compression in which a subsequent frame, or a portion thereof, is encoded as differential data with respect to an earlier reference frame. Pixel data may be expressed as such differential data if inter-frame predictive coding is used. Illustratively, description will be made of the influence on the communication when the processing capacities for the frame predictive coding tools are not in agreement with those for the frame predictive decoding tools. The frame predictive coding may be considered as improving the quality of a display image on the decoder side since inter-frame predictive coding is an image data processing technology. Based on the data of the pixels directly obtained by sampling the video signal on the coder side, the pixel data for the display pixels on the decoder side are defined more minutely than the sampling pixels of the coder side since the pixels on the decoder side are predictively interpolated.

FIGS. 7A through 7C are conceptual diagrams for illustrating pixel data arrangements resulting from frame predictive coding. FIGS. 7A, 7B and 7C show the arrangements of pixel data (a.sub.1 to d.sub.1, a.sub.2 to i.sub.2, a.sub.4 to Y.sub.4) produced based on the image data A to D which is directly obtained by coding the video signal inputted from a visual sensor such as a camera etc., by means of the frame predictive coding tools of sampling the data per single pixel, per 1/2 pixel and per 1/4 pixel, respectively. This pixel data is transmitted as the coded data obtained by the frame predictive coding, from the coder side to the decoder side. Here, in each of the frame predictive coding tools, the arithmetic operation for each piece of the pixel data is made based on the calculating formulae shown in Table 1.

TABLE-US-00001 TABLE 1 Pixel data based Pixel data based Operation on the sampling on the sampling formula per 1/2 pixel per 1/4 pixel A a.sub.2 a.sub.4 B c.sub.2 e.sub.4 C g.sub.2 u.sub.4 D i.sub.2 y.sub.4 (A + B)/2 b.sub.2 b.sub.4, c.sub.4, d.sub.4 (A + C)/2 d.sub.2 f.sub.4, k.sub.4, p.sub.4 (B + D)/2 f.sub.2 j.sub.4, o.sub.4, t.sub.4 (C + D)/2 h.sub.1 v.sub.4, w.sub.4, x.sub.4 (A + B + C + D)/4 e.sub.2 g.sub.4, h.sub.4, i.sub.4, l.sub.4, m.sub.4, n.sub.4, g.sub.4, r.sub.4, s.sub.4

In FIG. 7A, pixel data a.sub.1 to d.sub.1 for the pixels indicated by `+` is the pixel data (corresponding to the (n+1)-ranked coded data when the pixel data obtained by the aftermentioned frame predictive coding tool of sampling per 1/2 pixel is assumed as the n-ranked coded data) produced by the frame predictive coding tool (corresponding to the (n+1)-ranked coding tool when the aftermentioned frame predictive coding tool of sampling per 1/2 pixel is assumed as the n-ranked coding tool) of sampling per single pixel. In this case, pixel data a.sub.1 to d.sub.1 obtained by the frame predictive coding is equivalent to image data A to D which is directly obtained by coding the video signal.

In FIG. 7B, pixel data a.sub.2 to i.sub.2 for the pixels indicated by `+` and `.smallcircle.` is the pixel data (corresponding to the (n+1)-ranked coded data when the pixel data obtained by the aftermentioned frame predictive coding tool of sampling per 1/4 pixel is assumed as the n-ranked coded data) produced by the frame predictive coding tool (corresponding to the (n+1)-ranked coding tool when the aftermentioned frame predictive coding tool of sampling per 1/4 pixel is assumed as the n-ranked coding tool) of sampling per 1/2 pixel. Of these, the pixel data for the pixels indicated by `+` is equivalent to the pixel data obtained by the frame predictive coding tool of sampling per single pixel, while the pixel data for the pixels indicated by `.smallcircle.` is the interpolated pixel data which has been predicated based on image data A to D.

In FIG. 7C, pixel data a.sub.4 to y.sub.4 for the pixels indicated by `+`, `.smallcircle.`, `.DELTA.` is the pixel data produced by the frame predictive coding tool of sampling per 1/4 pixel. Of these, the pixel data for the pixels indicated by `+` and `.smallcircle.` is equivalent to the pixel data obtained by the frame predictive coding tool of sampling per 1/2 pixel. The pixel data for the pixels indicated by `+` is equivalent to the pixel data obtained by the frame predictive coding tool of sampling per single pixel, while the pixel data for the pixels indicated by `.smallcircle.` and `.DELTA.` is the interpolated pixel data which has been predicated based on image data A to D.

As understood from FIGS. 7A to 7C, pixel data a.sub.2 to i.sub.2 obtained by the frame predictive coding per 1/2 pixel sampling, includes pixel data a.sub.1 to d.sub.1 (image data A to D) obtained by the frame predictive coding per single pixel sampling, therefore the pixel data is distributed four times as dense as that of sampling pixels of the video signal. Pixel data a.sub.4 to y.sub.4 obtained by the frame predictive coding per 1/4 pixel sampling, includes the pixel data obtained by the frame predictive coding per single pixel sampling and per 1/2 pixel sampling, therefore the pixel data is distributed sixteen times as dense as that of the sampling pixels of the video signal.

In this way, the pixel data obtained by the inter-frame predictive coding tools for producing the pixel data of high-density display pixels, hierarchically includes the pixel data obtained by the frame predictive coding tools for producing the pixel data of the lower density display pixels. For instance, pixel data a.sub.4 to y.sub.4 obtained by the frame predictive coding tool of sampling per 1/4 pixels hierarchically includes pixel data a.sub.2 to i.sub.2 as well as pixel data a.sub.1 to d.sub.1.

FIGS. 8A to 8C are illustrations explaining the effects on the image display by the inter-frame predictive coding and showing the display images of a pattern TA obtained by decoding the coded data based on the inter-frame predictive coding per single pixel sampling, per 1/2 pixel sampling and per 1/4 pixel sampling, respectively. In FIGS. 8A to 8C, `.smallcircle.` and `.cndot.` represent pixels of `light` and `dark` display states, respectively when the coded data obtained by subjecting the video signal of pattern TA as a subject to the inter-frame predictive coding is decoded. Here, in FIGS. 8A to 8C, pattern TA as a subject is assumed to move in the lower right direction, and for easy understanding of the positional relation between pattern TA and the pixels, display images are laid over the pixels, for reference.

Referring to FIGS. 8A through 8C, according to the frame predictive coding per single pixel, pattern TA is represented wit three pixels before movement and with one pixel after movement. On the other hand, according to the frame predictive coding per 1/2 pixel, pattern TA is represented with six pixels before movement and with three pixels after movement. Further, according to the frame predictive coding per 1/4 pixel, pattern TA is represented with fifteen pixels before movement and with ten pixels after movement. In this way, as the dividing number of the pixels in the frame predictive coding is increased so that the density of the display pixels at the decoder side is increased, it becomes possible to reproduce an image of high quality with high precision.

Next, description will be made of a means for practicing the decoding scheme by the combination of individual functional tools (functional modules) independent of one another in the coder described above.

FIG. 9 shows an example of a coding data stream to be used when the coded data based on H.261 is transmitted to a device which does not have the decoding function based on H.261. As stated above, since it is assumed that the coding scheme is not invariant and the combination of the functional tools in the coder can be freely selected, it is necessary to transmit the information of the type of the coding scheme based on which the signal was coded and the types of the functional tools used in the coding process (this information will hereinbelow be referred to as coding information), together with the coded data. In FIG. 9, the data stream includes: coding information composed of motion compensation tool 112a, inverse transforming tool 112b, quantizing tool 112c and decoding scheme constructing information 111; and coded data of motion vector information 113a, transform coefficient 113b and quantization step 113c, which follow the corresponding coding information. The aforementioned each of the functional tool 112a to 112c designate the orders of decoding corresponding coded data 113a to 113c, and may contain operation specifications in some cases, may just indicate the identifying numbers of the functional tools in the other cases. Decoding scheme constructing information 111 specifies the functional tools to be used and the methods of using the resultant outputs from the tools, and other information. In the case shown in FIG. 9, the result after the motion compensation is used to handle the data of a certain image block decoded right before, for instance. That is, this result indicates the information relating to the order of procedure of the coding scheme H.261 in this case. The device on the decoding side, which has received the coding data stream shown in FIG. 9, is able to construct the decoding scheme by interpreting the decoding scheme constructing information, the motion compensation tool, the inverse transforming tool and the quantizing tool so that it can exactly decode the decoded data that follows.

As stated above, the coding information may contain the processing order of the tools and how to use the result obtained from each tool etc., so that the decoder will be able to decode the received coded data even if the signal which requires tools or is based on a decoding scheme that is not provided on the decoder side, is received. In order to improve the efficiency in the use of the line, however, it is preferable to use a decoding scheme which is able to work with a less amount of data to be transmitted such as the specifications on the construction of the decoding scheme and the tool information. In practice, since the purpose of the usage and the required quality will be determined to a certain degree depending on the coding and decoding apparatus, it is realistic that each coding and decoding apparatus has a number of coding and decoding schemes, in advance, which are expected to be used more frequently.

FIG. 10 shows an example of the coding data stream which can be used for the communication between two devices both having some coding and decoding schemes which are expected to be used more frequently. For the coding information, the same decoding scheme incorporated in the decoder is called up by transmitting a predetermined identification code 121a so that the coded data received can be decoded. Comparing with the example of FIG. 9, since this method will not need the transmission of the information on functional tools and the decoding scheme constructing information, it is possible to drastically reduce the transmitted amount of data and therefore the improvement of the efficiency in the use of the communication line can be expected.

However, if the divided number of pixels in the frame predictive coding is dissimilar (or the frame predictive coding tools are different), the structure of the coded data becomes quite different and thus it becomes impossible to interchange the coded data. For this reason, in accordance with the conventional coding and decoding system (method and devices), the decoding side needs to perform its decoding operation using a decoding tool suitable to the structure of the coded data. That is, the decoding tool should have the decoding capacity in one-to-one correspondence to the coding capacity for the coding tool to perform the decoding operation. Therefore, when the processing capacity for the decoding tool is not in agreement with that for the coding tool, it is totally impossible to decode the coded data received.

When the data which is coded using an algorithm provided with various tools (represented by MPEG4, for example) is attempted to be decoded by the device just having a single algorithm such as MPEG1, the decoding side needs additional hardware and/or software for operating the algorithm (coding tools) used in the coding. Therefore, the device is increased in size and cost.

As seen also in the H.261 coding scheme etc., the detailed specifications of the coding scheme is usually switched depending on which is more important, the efficiency of coding or the quality of image, or depending upon the nature etc., of the input image. Further, the usage will be limited if the system has only limited types of coding schemes previously equipped, as stated above. Therefore, it becomes necessary to change over the coding scheme in accordance with the usage. In this case, if the coder side tries to transmit data coded based on the scheme that is not equipped on the decoder side, the coding information should be simultaneously transmitted, as already mentioned above. At this moment, in accordance with the aforementioned method, all the coding information, as shown in FIG. 9, including the information of the functional tools used in the decoding scheme previously provided in the decoder side, needs to be transmitted regardless of whether the difference of the decoding scheme from the coding scheme is small or great. That is, even when the coding scheme is not much different from the decoding scheme that is previously provided, the communication may require a large transmission rate, thus possibly reducing the efficiency in the use of the line. In practice, however, since there are some functional tools which can be commonly used with little dependence on the difference in coding schemes, such as the transform coding in the motion picture, etc., it is possible to develop different kinds of coding schemes by adding other functional tools to the basic functional tools as such.

Further, in recent years, it has become possible to download the tools for JPEG and MPEG1 stated above, on the personal computer communications network etc., and receive image signal and decode it based on the downloaded tools. Therefore, it can be guessed readily that in the near-future video communications, the communication will be able to be performed by downloading the tools for coding and decoding. However, in the aforementioned coding and decoding system of the conventional scheme, the communication can be performed based only on the limited kinds of coding and decoding algorithms. In the case of the next generation image coding scheme (such as MPEG4) which can flexibly deal with various applications and can code the signal in the most suitable manner to each of the applications, if several kinds of algorithms are tried to be processed by a scheme which performs coding with a fixed algorithm such as JPEG, H.261, MPEG1, MPEG2, etc., it becomes necessary to provide hardware and/or software for executing each algorithm. In this way, it is preferable that all the various kinds of algorithms are provided for both the transmitting and receiving sides. However, if all the tools are provided to deal with all the algorithms, the hardware and software becomes bulky, and the apparatus will increase in cost and inevitably becomes large. On the other hand, if the apparatus is reduced in cost and size and therefore the apparatus does not have adequate capacities, the risk of the failure to perform communications becomes high.

In the coding and decoding apparatus which does not have the above capacity, the decoder will download the tools for the required algorithm so as to be able to flexibly deal with the various kinds of applications and decode the signal. In such a coding and decoding apparatus which downloads the tools for the algorithm and is able to store the tools previously used, if the tools stored are not the ones which are required for the next communication, the required tools must be downloaded again before the transmission of the coded data. Therefore, the delay before the start of transmission to the decoding of the coded data becomes long.

In the above coding and decoding apparatus which is able to store the tools previously used, if the coding and decoding tools are provided in such a hierarchical manner that the tools for high quality is provided at the lower rank and the tools for assuring minimum quality which are not replaceable with other tools are provided at the higher rank, it becomes possible to decode the signal using those tools for minimum quality even if the capacity of the decoding apparatus is different from that of the coding apparatus. In this case, the delay before the start of transmission due to the downloading of the tool information can be eliminated, however it is impossible to decode the signal with the anticipated quality. In this case, when the signal is decoded with the anticipated quality, it is necessary to previously download the tools for the anticipated quality. Therefore, the situation is quite similar to the case where the tools are not provided in the hierarchical manner. That is, the delay before the start of transmission to the decoding of the coded data becomes long and therefore it is impossible to make use of the merit from the hierarchical structure of the tools.

SUMMARY OF THE INVENTION

The present invention has been achieved in view of the above problems. It is therefore an object of the invention to provide a coding and decoding apparatus which can select the most suitable tools based on the comparison with the tool-correspondent information and can perform the decoding operation efficiently and quickly when the received coded data is decoded using the tools simultaneously received.

It is another object of the invention to provide a coding and decoding apparatus which is able to decode the coded data even if the coding capacity of the coding tools on the coding side is not in agreement with the decoding capacity of the decoding tools on the decoding side so that the apparatus on the decoding side can be reduced in size and price.

It is a further object of the invention to provide a coding and decoding apparatus in which, by transmitting only the differential information as to the functional tool and the decoding scheme which are already prepared in the decoding apparatus when the decoding scheme is transmitted, it is possible to designate and identify the necessary decoding scheme with fewer steps, so that the efficiency in the use of the communication line will not be deteriorated critically.

It is still another object of the invention to provide a coding and decoding apparatus of the next-generation image coding scheme, in which the coding apparatus on the transmitting side simultaneously transmits the decoding tool information and the coded data when the decoding apparatus on the receiving side has no decoding tools which are requested by the transmitting side, thus making it possible to save the time which would be required for downloading only the tool information and thereby shorten the time before the start of the transmission of the coded data, compared with the configuration which initially transmits tool information only, and then downloads this information and thereafter starts decoding by using the thus downloaded decoding tools.

In order to achieve the above object of the invention, the gist of the invention can be represented as follows:

In a coding and decoding apparatus of a first aspect of the invention where the coder transmits coded data together with identifying information for identifying the means of decoding the coded data, and the decoder is capable of storing a plurality of decoding schemes so as to perform decoding based on one of the previously stored schemes, in order that the coded data and the information of the tools which constitute the algorithm as the means of decoding the coded data can simultaneously be transmitted, the received tools can be reconstructed into the algorithm and the received coded data can be decoded based on the algorithm. The coding and decoding apparatus comprises: a tool storage for storing tools; a tool-correspondent information storage for storing the information corresponding to the tools; a comparing device for comparing the tool-correspondent information received with the information stored in the tool-correspondent information storage; and a selection controller for selecting the optimal tool from the tool storage based on the result from the comparing device to perform the processing with the selected one. The apparatus is constructed such that the coded data, tool information and tool-correspondent information are all transmitted simultaneously, and the coded data is decoded using the tool selected based on the received tool-correspondent information.

A second aspect of the invention resides in a coding and decoding apparatus having the first feature and is characterized in that the tool-correspondent information comprises the processing capacity of each tool, and the processing capacity of the received tool is compared to a decoding capacity stored in the tool-correspondent information storage so that the tools whose capacities fall within the range of the decoding capacity are selected.

A third aspect of the invention resides in a coding and decoding apparatus having the second feature and is characterized in that the processing capacity of the tool is numerically represented and transmitted.

A fourth aspect of the invention resides in a coding and decoding apparatus having the second feature and is characterized in that the tool-correspondent information storage includes a decoding capacity storage section for setting up a decoding capacity of the decoding apparatus and storing it and a coding capacity storage section for storing each of coding capacities of the tools transmitted from the coding apparatus, and the comparator comprises a capacity comparator which compares the coding capacity with the decoding capacity so as to judge whether the transmitted tool is processible.

A fifth aspect of the invention resides in a coding and decoding apparatus having the first feature and is characterized in that the tool-correspondent information comprises keys unique to different tools, and received keys are compared to the keys stored in the tool-correspondent storage so as to select the corresponding tools and operate the selected tools.

A sixth aspect of the invention resides in a coding and decoding apparatus having the first feature and further comprises a response controller for requesting the coding apparatus on the opposite side to transmit the tool information only when tool information is required.

Next, in a coding and decoding apparatus of a seventh aspect of the invention where the coder transits coded data together with identifying information for identifying the means of decoding the coded data, and the decoder is capable of storing a plurality of decoding schemes so as to perform decoding based on one of the previously stored schemes, the apparatus is characterized in that n-ranked (n: a positive integer) coded data which is produced using an n-ranked coding tool and decoded using an n-ranked decoding tool has a hierarchical structure which includes (n+1)-ranked coded data which is produced using an (n+1)-ranked coding tool and decoded using a (n+1)-ranked decoding tool, the coder having an n-ranked coding tool is composed of: a coder which produces the n-ranked coded data using the n-ranked coding tool; and an identifier adder which attaches N-ranked identifiers (N: a positive integer satisfying N.gtoreq.n) to N-ranked coded data which is included in the n-ranked coded data but is other than (N+1)-ranked coded data included in the N-ranked coded data, and the decoder having an m-ranked (m is a positive integer satisfying m>n) decoding tool is composed of: a data reconstructer which extracts the N-ranked coded data which is attached with the N-ranked identifiers where N.gtoreq.m, from the n-ranked coded data; and a decoder which decodes the m-ranked decoded data using the m-th decoding tool.

An eighth aspect of the invention resides in a coding and decoding apparatus having the feature of the seventh aspect and is characterized in that the coding tool is an frame predictive coding tool and the decoding tool is an frame predictive decoding tool.

Further, in a coding and decoding apparatus of a ninth aspect of the invention where the coding side transmits coded data together with identifying information for identifying the means of decoding the coded data, and the decoding side is capable of storing a plurality of decoding schemes so as to perform decoding based on one of the previously stored schemes, the coding and decoding apparatus is characterized in that when the coded data and the coding information which includes a decoding scheme as the means of decoding the coded data and functional tools constituting the decoding scheme are simultaneously transmitted, the decoding side receives the coding information and reconstructs the decoding scheme based on the coding information received, and the received coded data is decoded based on the reconstructed decoding scheme, an identification code of a previously defined basic decoding scheme and the differential information from the basic decoding scheme are transmitted as the coding information from the coding side so that the decoding side will recognize the decoding scheme required therefor.

A tenth aspect of the invention resides in a coding and decoding apparatus having the ninth feature and is characterized in that the coding apparatus comprises: a database of coding schemes for storing plural kinds of coding schemes and functional tools which constitute the coding schemes; a coding scheme selector for selecting the coding scheme based on input data; a coding section for performing a coding process of the input data in conformity with the determined coding scheme; and a coding controller for controlling each section.

An eleventh aspect of the invention resides in a coding and decoding apparatus having the ninth feature and is characterized in that the decoding apparatus comprises: a database of decoding schemes for storing plural kinds of decoding schemes and functional tools which constitute the decoding schemes; a decoding scheme constructing section for reconstructing the decoding scheme in accordance with the received coding information; a decoding section for performing a decoding process of the received data in conformity with the reconstructed decoding scheme; and a decoding controller for controlling each section.

A twelfth aspect of the invention resides in a coding and decoding apparatus having any one of the ninth through eleventh features and is characterized in that the identification code of a basic decoding scheme and the information that one or some kinds of functional tools will be added to the basic decoding scheme, are transmitted as the coding information, so that the decoding scheme incorporated in the decoding side can be expanded for use.

A thirteenth aspect of the invention resides in a coding and decoding apparatus having any one of the ninth through eleventh features and is characterized in that the identification code of a basic decoding scheme and the information that one or some kinds of functional tools will not be used, are transmitted as the coding information so that the decoding scheme incorporated in the decoding apparatus can be simplified for use.

A fourteenth aspect of the invention resides in a coding and decoding apparatus heaving any one of the ninth through eleventh features and is characterized in that the identification code of a basic decoding scheme and the information that one or some kinds of functional tools will be replaced with another or others, are transmitted as the coding information so that the decoding scheme incorporated in the decoding apparatus can be modified for use.

A fifteenth aspect of the invention resides in a coding and decoding apparatus having any one of the ninth and tenth features and is characterized in that when the coding information is transmitted, if there are a number of combinations of selectable coding information, the combination which minimizes the transmitted amount of information will be selected for transmission.

Next, in a coding and decoding apparatus of a sixteenth aspect of the invention where the coding side transmits coded data together with identifying information for identifying the means of decoding the coded data, and the decoding side is capable of storing a plurality of decoding schemes so as to perform decoding based on one of the previously stored schemes, the coding and decoding apparatus is characterized in that: before transmitting the coded data to the decoding apparatus, the coding apparatus transmits the tools constituting an algorithm as the means of decoding the coded data, and the decoding apparatus reconstructs the algorithm using the tools so as to decode the received coded data based on the algorithm and stores the tools therein; when the decoding apparatus receives the coded data which has been coded by the same tools, the decoding side decodes the coded data using the tools previously stored and the tools are defined in a hierarchical manner so that in place of a tool for a certain rank, the higher-ranked tool can be used to secure the minimum quality of the operation; and the coding apparatus on the transmitting side simultaneously transmits the decoding tool information and the coded data if the decoding apparatus on the receiving side does not have the decoding tool requested by the transmitting side.

A seventeenth aspect of the invention resides in a coding and decoding apparatus having the sixteenth feature and is characterized in that when the decoding apparatus on the receiving side has no decoding tool requested by the coding apparatus on the transmitting side, the transmitting side temporarily changes the coding scheme using the coding tool that is in conformity with the decoding tool present on the receiving side.

An eighteenth aspect of the invention resides in a coding and decoding apparatus having the sixteenth feature and is characterized in that when the decoding apparatus on the receiving side has no decoding tool requested by the coding apparatus on the transmitting side, the receiving side, whilst downloading the decoded tool transmitted from the transmitting side to construct the requested decoding tool, temporarily decodes the coded data using a substitutable higher-ranked tool which is lowered in quality but still is able to perform decoding.

A nineteenth aspect of the invention resides in a coding and decoding apparatus having the eighteenth feature and is characterized in that after the decoding tool requested has become prepared, the receiving side starts the decode operation using the requested decoding tool.

According to the invention, the most suitable tools are selected based on the comparison between the tool-correspondent information stored in the tool-correspondent information storage and the received tool-correspondent information. Thus, it becomes possible to perform the decoding operation efficiently and quickly. Particularly, an effective decoding operation is attained because the processing capacities for the tools are used as the tool-correspondent information and the tools, whose capacities fall within a permissible range for the decoding operation, are then selected. In this case, by transmitting the processing capacity which is numerically represented, it becomes possible to efficiently make a quick comparison in capacities.

Further, it becomes possible to perform a quick selection of the tools because, by making the tool-correspondent information easy to compare, the tools themselves do not need to be compared. Especially, the comparison and selection of the tools in the tool storage can be performed quickly and effectively by allotting a unique key to each tool as the tool-correspondent information.

Since the response controller is provided, only when there is a necessity for a tool, it is possible to cause the coding apparatus on the other side to transmit the information of the tool. Thus, it is possible to shorten the time for transmission and reception and thereby perform the process efficiently.

Further, since the coding and decoding tools are defined hierarchically, and the data which conforms to the processing capacities of the thus hierarchically defined tools is attached with headers for identifying the processing capacity of the tool on the coding side, the data can be reconstructed so as to have a data structure which conforms to the processing capacity of the decoding side and can be decoded. As a result, even if the processing capacities of the coding and decoding tools are not compatible, it is possible to decode the coded data.

It becomes no more necessary for the decoding side to have all of various tools in order to deal with the data which is coded by such an algorithm having various coding and decoding tools as a next-generation coding standard represented by MPEG4. Therefore, it is possible to reduce the cost of the apparatus.

Further, since it is possible to alleviate the restriction on the hardware of the decoding apparatus, it becomes possible for a simple low-cost apparatus to decode the data coded based on the next-generation coding standard.

In accordance with the scheme described above, the coding information, within the coding data stream transmitted from the coding apparatus, used for decoding the coded data is made up of only the identification code of the basic decoding scheme and the differential information with respect to the basic decoding scheme. Expanded decoding schemes from the basic decoding scheme can be represented by the differential information. That expansion of the basic decoding scheme is formed by designating the functional tools to be added to, canceled from or replaced in the basic decoding scheme. Further, if there are two or more methods in designating the differential information, the designating method which requires less amount of information is selected so that the amount of data to be transmitted becomes minimum.

Further, in the coding and decoding apparatus described above, even if the receiving side has not tool having requested quality at the time of the first transmission, the transmitting side sta