Patent Application
20050063473
This invention relates to a technique for transmitting encoded data and, more particularly, to a method and an apparatus for transmitting and receiving coded packets, and to a program therefor.
Recently, as a method for efficiently transmitting moving picture data with a small transmission bandwidth, use of a moving picture compressed encoding system, such as MPEG (Moving Picture Expert Group)-4, has become widespread. In such a compressed encoding system, an input picture frame is processed with an inter-frame prediction by motion compensation, in terms of a rectangular area of a fixed size, termed a macro block, as a unit and the resultant motion vector and prediction residue picture data are then processed with two-dimensional discrete cosine transform (DCT) and quantization to give compressed signal data, which is then coded by variable length coding.
Meanwhile, if, in a variable length coded bit string (bitstream), even one bit of the bitstream is in error, the next following bitstream cannot be decoded correctly. Thus, in the MPEG-4, a unique word (codestring that does not present itself in the encoded data), also termed a resynchronization marker code (Resync Marker), is inserted in encoded data at a suitable interval as an error resilience technique, so that, even on error occurrence, the codeword is correctly recognized by the next resynchronization marker, such as to allow correct decoding as from this code. The resynchronization marker code (Resync Marker) is introduced at a break of the encoded data of the macroblock, and the introducing position thereof is aligned with the byte boundary, such as to form a video packet structure made up by the resynchronization marker code and an optional number of the next following macroblocks of the encoded information data. That is, by this video packet structure, decoding can be re-started, even in case of occurrence of a bit error, as from the next resynchronization marker code. In other words, it is possible to spatially localize the effect of an error.
Alternatively, when a bitstream is transmitted on the packet basis, it is usually transmitted on the video packet basis, so that, even in case of occurrence of a packet loss, decoding can similarly be re-started as from the next resynchronization marker code.
However, there is a problem that, if bit error or packet loss is produced in a burst-like fashion, such that plural video packets cannot be decoded correctly, decoded pictures are deteriorated more severely.
There is also raised a problem that, since inter-frame prediction is used for encoding, picture deterioration, once produced, is propagated to the next following frames.
In the present invention, the input encoded packet data is re-encoded. Reference is made to the description of, for example, the Japanese Patent Kokai Publication No. JP-A-8-111870, as the publication which has disclosed the technique pertinent to the re-encoding. In this Publication, there is disclosed, as a re-encoding method in the cascaded connection re-encoding consisting in re-encoding the picture information obtained on decoding the picture information obtained in turn on encoding followed by information volume compression in the preceding stage, such a configuration consisting in detecting encoded parameters in the encoding of the picture information of the preceding stage, deciding on a parameter conforming to the so detected encoded parameter and re-encoding the decoded picture information by the so determined encoded parameter.
Accordingly, it is a principal object of the present invention to provide an apparatus, a system, a method, and a program, in which error resilience may be improved and in which marked deterioration in picture quality of decoded pictures may be prohibited despite the bit errors or packet losses occurring on the transmission line for the compressed-encoded moving picture data.
It is another object of the present invention to provide an apparatus, a system, a method, and a program, in which the bandwidth taken up in transmitting encoded moving picture packets may be prevented from increasing.
In one aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided an encoded data transmission apparatus, comprising means for receiving compressed-encoded data and for outputting data converted therefrom, in which said means outputs data obtained on decoding the compression-encoded data, re-encoding the resultant data and on selecting part or all of the re-encoded data, or outputs data selected from the compression-encoded data, as duplicated data, and means for controlling the delay so that the compression-encoded data and the duplicated data will be transmitted with temporal spacing from each other a preset time interval, multiplexing the compression-encoded data and the duplicated data and for outputting the resultant multiplexed data.
In another aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided an encoded data receiving apparatus comprising means for receiving multiplexed data sent from the above-defined encoded data transmission apparatus and for demultiplexing the compression-encoded data and the duplicated data from the multiplexed data, and means for reconstructing encoded data, using the duplicated data received, in substitution for the compression-encoded data, and for outputting the so reconstructed encoded data.
In a further aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided an encoded packet transmitting apparatus comprising a converter for being supplied with a packet compression-encoded at a preset compression ratio, decoding data of the packet supplied once, forming a packet compression-encoded from the data of the decoded packet at a compression ratio different from the preset compression ratio, and for outputting the so formed packet, a delay controller for controlling the delay added to the packet compression-encoded at the preset compression ratio (termed a ‘a first class packet’) and/or to the packet formed on re-encoding by the converter (termed ‘a second class packet’) so that the first class packet and the second class packet are transmitted with temporal spacing, corresponding to a preset time interval, in-between, and a multiplexing unit for outputting a multiplexed transmission packet, obtained on multiplexing the first class packet and the second class packet.
The above-defined encoded packet transmitting apparatus according to the present invention may further comprise a transmission decision unit arranged in substitution for the converter or downstream of the converter for deciding on a packet for transmission at an interval of a period of input packets or based on a preset parameter of packet data.
The above-defined encoded packet transmitting apparatus according to the present invention may further comprise a transmission decision unit arranged in substitution for the converter or downstream of the converter, the transmission decision unit performing decoding processing for a case where there is an error in the packet and for a case where there is no error in the packet, comparing the quality of the decoded results and deciding on a packet for transmission based on the results of comparison.
In a further aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided an encoded packet receiving apparatus comprising a decoding/demultiplexing unit for receiving multiplexed transmission packets including first class packets transmitted multiplexed at a time interval from one another and second class packets formulated based on the first class packets, decoding the number of times the multiplexed transmission packets have been received and for demultiplexing the multiplexed transmission packets into the first class packets and the second class packets, based on the number of times of receipt, a detection unit for detecting whether or not there is packet loss or transmission error in the first or second class packets obtained on demultiplexing, and a generating unit for outputting a predetermined class packet, out of the first class packet and the second class packet, which has been received as normally, as reconstructed data. In case the predetermined class packet has not been received as normally and the other class packet has been received as normally, the generating unit outputs, the normally received packet as reconstructed data.
The apparatus for receiving coded packets according to the present invention may further comprise, in place of the demultiplexing unit, a decoder/separator for receiving multiplexed transmission packets including first class packets transmitted multiplexed at a time interval in-between and second class packets formulated based on the first class packets, decoding the number of times the multiplexed transmission packets have been received and for demultiplexing the multiplexed transmission packets into the first class packets and the second class packets, based on the number of times of receipt.
In the encoded data transmitting/receiving apparatus, according to the present invention, the packet is a moving picture packet obtained on compressed-encoding moving picture data and packetizing the moving picture packet.
In a further aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided a coded data transmitting method comprising:
In another aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided an encoded data receiving method comprising:
In another aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided a program for a computer, forming an apparatus for transmitting encoded data, to execute the processing of being supplied with compression-encoded data and outputting data converted therefrom,
In another aspect of the present invention, providing means for accomplishing at least one of the objects of the present invention, there is provided a program for a computer, forming an apparatus for receiving multiplexed data sent out from an apparatus transmitting the encoded data, to execute
The schematics of the present invention are explained first, and preferred embodiments of the invention are then explained.
The present invention provides the function for preventing significant picture quality deterioration of decoded picture even in case of occurrence of bit errors and packet losses on a transmission line of data, resulting from compressed encoding of moving pictures. That is, according to the present invention, in duplicating and transmitting packets including encoded moving pictures, the packets are transmitted at a preset period or at adaptively changed time intervals, in order to take account of burst characteristicsistics of the bit errors and packet losses on the transmission line.
Moreover, according to the present invention, a packet to be duplicated may be determined by having reference to a parameter in the packet including an encoded moving picture to be transmitted which will have marked effects on the picture quality of the decoded picture.
Alternatively, according to the present invention, the packets including encoded moving pictures to be transmitted are decoded and the results of comparison of the picture quality when the packets contain errors and that when the packets are free from errors are compared to each other to decide on the packets to be duplicated.
The present invention provides the function of suppressing the band occupied in transmitting the packets including encoded moving pictures from increasing. In duplicating and transmitting the packets including encoded moving pictures, the packets are once decoded and subsequently re-encoded to give a high compression ratio.
According to the present invention, the packets for duplication may be verified at a preset period or as the parameters which will have marked effects on the picture quality of the decoded pictures, contained in the packets including encoded moving pictures, are taken into account.
According to the present invention, the packets including encoded moving pictures to be transmitted may also be decoded and the results of comparison of the picture quality when the packets contain errors and that when the packets are free from errors may be compared to each other to decide on the packets to be duplicated.
A transmitting apparatus according to a preferred embodiment of the present invention includes a converter (such as 103 of
The delay controller controls the delay added to the aforementioned packet, compression-encoded at the aforementioned preset compression ratio (termed the first class packet) and/or to the aforementioned packet formed on re-encoding by the above-mentioned converter (termed the second class packet) so that the first class packet and the second class packet will be transmitted at a spacing corresponding to the preset time interval. The multiplexing unit multiplexes the first class packet and the second class packet to output the resulting multiplexed transmission packet.
A receiving apparatus according to a preferred embodiment of the present invention includes a demultiplexer (such as 201 of
The transmitting apparatus according to a preferred embodiment of the present invention may include a transmission decision unit (such as 403 of or 804 of
The transmitting apparatus according to a preferred embodiment of the present invention may include a transmission decision unit (such as 703 of
A receiving apparatus according to a preferred embodiment of the present invention may include a decoder/demultiplexer (such as 501 of
A method according to a preferred embodiment of the present invention comprises the following steps:
Step 1: A packet encoded at a preset compression ratio is input and data of the input packet is once decoded. The so decoded packet data is re-encoded at a compression ratio different from the aforementioned preset compression ratio to form a re-coded packet, which then is output.
Step 2: Delay added to a packet compression-encoded at the aforementioned compression ratio (termed ‘first class packet’) and to a re-coded packet (termed ‘second class packet’) is controlled so that the first class packet and the second class packet are spaced apart from each other a preset time interval and transmitted in this spaced-apart condition.
Step 3: A multiplexed transmission packet, obtained on multiplexing the first class packet and the second class packet, is output. The processing and control of these steps 1 to 3 may, of course, be carried out by a program which is run on a computer.
A method according to the present invention is a method for receiving signals transmitted by the above method, and comprises the following steps:
Step 1: The multiplexed transmission packet, including the first class packets, and the second class packets, formulated on the basis of the first class packets, with the first and second packets being transmitted multiplexed at a preset time interval from each other, is received, and is demultiplexed into the first class packet and the second class packet.
Step 2: The first class packets and the second class packets, thus demultiplexed, are checked for possible packet losses or transmission errors.
Step 3: if the first class packet or the second class packet, whichever is the predetermined class packet, is received, such class packet is output as reconstruction data and, if the predetermined class packet is not received as normally and another class packet different from the predetermined class packet is received as normally, such normally received packet is output as reconstruction data. The processing and control of these steps 1 to 3 may, of course, be carried out by a program which is run on a computer.
For more detailed explanation of the preferred embodiments of the present invention, certain examples of execution of the present invention are explained with reference to the drawings.
Referring to
The encoded packet generating unit 101 is supplied with compression-encoded moving picture data 107, to output an encoded packet 108, which is obtained on division to a preset packet unit size.
The first error detection code/packet discrimination number adding unit 102 is supplied with the encoded packet 108, output from the encoded packet generating unit 101, to add an error detection code and a packet discrimination number for detecting bit errors and packet losses on the transmission line of the encoded packets 108, to output a transmission packet 109.
The encoded packet converting unit 103 is supplied with the encoded packet 108, output from the encoded packet generating unit 101, and re-encodes the encoded packet 108 at a compression ratio higher than the compression ratio for the encoded packets 108, to output high compression encoded packet 110.
The second error detection code/packet discrimination number adding unit 104 is supplied with high compression encoded packet 110, output from the encoded packet converting unit 103, to add an error detection code and a packet discrimination number for detecting bit errors and packet losses on the transmission line of the high compression encoded packet 110, to output a high compression transmission packet 111.
The delay adding unit 105 is supplied with the high compression packet 111, output from the second error detection code/packet discrimination number adding unit 104, and adds delay to the high compression transmission packet 111, to output a delayed high compression transmission packet 112.
The transmission packet multiplexing unit 106 multiplexes the transmission packet 109, output from the first error detection code/packet discrimination number adding unit 102, with the delayed high compression transmission packet 112, output from the delay adding unit 105, to transmit a multiplexed transmission packet 113 on a transmission line.
Referring to
The functions of these units may roughly be defined as follows:
The transmission packet demultiplexing unit 201 receives a multiplexed transmission packet 209 from the transmission line and demultiplexes the multiplexed packet into a transmission packet 210 and a delayed high compression transmission packet 211.
The first transmission packet storage unit 202 is supplied with the transmission packet 210 from the transmission packet demultiplexing unit 201 to store the input transmission packet 210 in a first transmission packet receiving buffer, not shown.
The second transmission packet storage unit 203 is supplied with the delayed high compression transmission packet 211, output from the transmission packet demultiplexing unit 201, to store the input delayed high compression transmission packet 211 in a second transmission packet receiving buffer, not shown.
The first encoded packet extraction unit 204 extracts a first encoded packet 214, which is a compression-encoded moving picture data part, from the first transmission packet 212, stored in the first transmission packet storage unit 202.
The first transmission packet error/packet loss detection unit 205 detects packet loss and bit error in the first transmission packet 212, stored in the first transmission packet storage unit 202, to output a first transmission packet detection result 215.
The second encoded packet extraction unit 206 extracts a second encoded packet 216, which is the compression-encoded moving picture data part, from a second transmission packet 213 stored in the second transmission packet storage unit 203.
The second transmission packet error/packet loss detection unit 207 detects packet loss and bit error in the second transmission packet 213, stored in the second transmission packet storage unit 203, to output a second transmission packet detection result 217.
The encoded data generating unit 208 is supplied with the first encoded packet 214, first transmission packet detection result 215, a second encoded packet 216 and the second transmission packet detection result 217 to output reconstructed encoded data 218.
Referring to
Referring to
The first error detection code/packet discrimination number adding unit 102 adds error correction codes and packet discrimination numbers to the encoded packet 108 to output the resulting transmission packet 109, in order to detect bit errors and packet losses on the transmission line of the input encoded packet 108 by the encoded packet receiving apparatus 2. If, in the present embodiment, a bit error detection mechanism is provided e.g. between the packet transmission apparatus 1 and the packet receiving apparatus 2, there is no necessity of adding the error correction code in the first error detection code/packet discrimination number adding unit 102. Moreover, if the information for packet discrimination is included in the encoded packets 108, there is no necessity for adding the packet discrimination number by the first error detection code/packet discrimination number adding unit 102.
The encoded packets 108, generated by the encoded packet generating unit 101, are supplied to the encoded packet converting unit 103. The encoded packet converting unit 103 first decodes the input encoded packet and subsequently re-encodes the packet. In such re-encoding, high compression encoded packet 110, higher in the compression ratio than the encoded packets 108, is generated, by employing a technique of quantizing DCT coefficients, using larger quantization parameters, or by adaptively cutting higher order DCT coefficients.
The second error detection code/packet discrimination number adding unit 104 adds error detection codes and packet discrimination numbers to the high compression encoded packet 110 to output the resulting high compression transmission packet 111, in order to detect bit errors and packet losses on the transmission line of the input compression encoded packets 110, on the encoded packet receiving apparatus 2. If the packet discrimination number of the high compression transmission packet 111 is associated with the original transmission packet 109, as confirmed by the encoded packet receiving apparatus 2 on the receiving side, any suitable discrimination number may be used. A large number which is not usual occurrence may be added as offset to the packet discrimination number appended to the original transmission packet 109, or “1” may be set at a preset bit in the binary representation of the packet discrimination number. Meanwhile, in an RTP packet, transmitted using the UDP (User Datagram Protocol)/IP (Internet Protocol), the sequence number of the RTP header may be used as the packet discrimination number.
If, in the present embodiment, as in the first error detection code/packet discrimination number adding unit 102, it is possible to detect the bit errors and packet losses on the transmission line, there is no necessity for adding the error detection code/packet discrimination number in the second error detection code/packet discrimination number adding unit 104.
The delay adding unit 105 is supplied with the high compression transmission packet 111 to add the delay of a fixed period or the adaptively changed delay to output the delayed high compression transmission packet 112. In the present embodiment, the delay added to the high compression transmission packet 111 is preferably such time which takes the burst characteristics of bit errors and packet losses on the transmission line into account. On the other hand, the delay added by the delay adding unit 105 is set as the data transfer rate (bit rate) or the size of a buffer for storage of the received packets on the side of the encoded packet receiving apparatus 2 are taken into account. It is noted that the delay time which takes burst characteristics of bit errors and packet losses on the transmission line into account is determined, such as by actual measurement, and is set by the delay added by the delay adding unit 105, in such a manner that, on the side of the receiving apparatus, the transmission packet 109 and the high compression transmission packet 112, formed in association with the transmission packet 109 and added by the delay, will be contiguous to one another, without packet losses, or in such a manner that the probability of these packets missing simultaneously will be that low.
The transmission packet multiplexing unit 106 multiplexes the transmission packet 109, output from the first error detection code/packet discrimination number adding unit 102, with the delayed high compression transmission packet 112, output from the delay adding unit 105, to generate the multiplexed transmission packet 113 to a transmit the packets on the transmission line, not shown, on the packet basis, using preset transmission means, not shown. The transmission means comprises a physical layer corresponding to the transmission line, a network layer positioned on the physical layer, and so forth. The transmission line may be wired or wireless, as desired.
In a step 301 of
The transmission packet demultiplexing unit 201 outputs the transmission packet 210 to the first transmission packet storage unit 202, while outputting the delayed high compression transmission packet 211 to the second transmission packet storage unit 203.
In a step 302, the first transmission packet storage unit 202 stores the transmission packet 210, as input, in a first transmission packet receiving buffer, not shown, until a preset time elapses as from the expected arrival time of the n'th packet.
In a step 303, the first transmission packet error/packet loss detection unit 205 verifies whether or not the n'th packet is present in the first transmission packet receiving buffer. If the n'th packet is present in the first transmission packet receiving buffer, the first transmission packet error/packet loss detection unit 205 proceeds to a step 304. If otherwise, the first transmission packet error/packet loss detection unit deems that the n'th packet is the packet loss and proceeds to a step 306.
In a step 304, the first transmission packet error/packet loss detection unit 205 verifies whether or not there is any bit error in the n'th packet in the first transmission packet receiving buffer, using e.g. the error detection code added by the first error detection code/packet discrimination number adding unit 102 of the encoded packet transmission apparatus 1. If there is no bit error in the n'th packet of the first transmission packet receiving buffer, the first transmission packet error/packet loss detection unit proceeds to a step 305. If there is any bit error in the n'th packet of the first transmission packet receiving buffer, the first transmission packet error/packet loss detection unit proceeds to a step 306.
In the step 305, the first encoded packet 214, extracted in the first encoded packet extraction unit 204, is output as reconstructed encoded data 218 from the encoded data generating unit 208.
In a step 306, it is verified whether or not the n'th packet is in a second transmission packet receiving buffer. If the n'th packet is in the second transmission packet receiving buffer, processing transfers to a step 307. If the n'th packet is not in the second transmission packet receiving buffer, the second transmission packet error/packet loss detection unit 207 deems that the n'th packet represents packet loss and does not output the reconstructed encoded data 218.
In a step 307, the second transmission packet error/packet loss detection unit 207 verifies whether or not there is any bit error in the n'th packet in the second transmission packet receiving buffer, using e.g. the error detection code added by the second error detection code/packet discrimination number adding unit 104 of the encoded packet transmission apparatus 1. If there is no bit error in the n'th packet of the second transmission packet receiving buffer, the second transmission packet error/packet loss detection unit proceeds to a step 308. If there is any bit error in the n'th packet of the second transmission packet receiving buffer, the second transmission packet error/packet loss detection unit 207 does not output the reconstructed encoded data 218.
In a step 308, the second encoded packet 216, extracted by the second encoded packet extraction unit 206, is output as reconstructed encoded data 218 from the encoded data generating unit 208.
Referring to
The transmission packet multiplexing unit 106 multiplexes two packet streams, that is, the transmission packets 109 and the delayed high compression transmission packets 112, into one stream, to output a multiplexed transmission packet 113, shown schematically in
That is, the transmission packet multiplexing unit 106 outputs, as multiplexed transmission packets 113, a stream of the packets A, B, C, A′, B′, C′, D, E, D′, E′, . . . . The delay from the packet A to the packet A′ (the time indicated by an arrow in
Referring to
Referring to
Another modification of the first embodiment of the present invention is now explained with reference to
That is, referring to
If in the example shown in
In the present modification, a three-row four-column matrix table (conversion table):
With the multiplexing processing, by this interleaving unit 120, the packets C′ and D′ of the high compression transmission packets 111 are transmitted earlier than the packets C and D of the original packets 109. On the other hand, the packet C of the transmission packets 109, as the origin of this duplicated packet, is equivalent to the packet C of the transmission packets 109, added by the delay. It is noted that the present duplicated packet C′ is derived from the packet C of the of the transmission packets 109. Moreover, the packet F of the transmission packets 109 is transmitted earlier than the packet C. The packets A′, E′, B′ and F′ of the high compression transmission packets 111 correspond to the packets A, E, B, and F of the transmission packets 109, added by delay, respectively. That is, in the multiplexed output by the interleaving method, the packets are shuffled within and across the pre-multiplexing streams, such that the packet sequence in the pre-multiplexing streams is not kept. Even in this interleaving, the delay is introduced in the original transmission packets 109 and in the original high compression transmission packets 111 so that, for example, the packets A and A′ are not missing simultaneously.
A second embodiment of the present invention is hereinafter explained.
Referring to
Referring to
Referring to
Referring to
In a step 601 of
The steps 602 to 608 of
An operational effect of the second embodiment of the present invention is that the band taken up in transmitting the second encoded packets may be prevented from being increased.
The reason is that, as the second encoded packet, not all of the first encoded packets are transmitted, but the first encoded packets are transmitted at a preset period (once every n packets), or that the second encoded packets transmitted are adaptively determined by having reference to the characteristic parameter of the first encoded packet.
Another operational effect of the second embodiment of the present invention is that deterioration of the picture quality of the decoded picture against bit errors or packet losses occurring on a transmission line for the compression encoded moving picture data may be reduced to a minimum.
The reason is that, in determining whether or not the second encoded packet is to be transmitted, such a parameter significantly affecting the picture quality of the decoded picture is referred to as a characteristic quantity.
In the present embodiment, if the encoded packet is of the same packet discrimination number as that of the previously received packet number, the encoded packet receiving apparatus 5 outputs the encoded packet as a delay high compression transmission packet 511 to the second transmission packet storage unit 503. The delay addition by the interleaving method, consisting in shuffling the sequence of packets multiplexed is not applied. The delay addition unit 405 adds the delay so that the packet sequence will be maintained in the respective transmission packets 109 and 112, as shown in
Referring to
Referring to
A third embodiment of the present invention is hereinafter explained.
In
The operational effect of the third embodiment of the present invention is that picture quality deterioration of the decoded picture may be minimized despite bit errors or packet losses occurring on a transmission line of compression-encoded data of the moving pictures.
The reason is that, in determining whether or not the second encoded packet is to be transmitted, the second encoded packet is actually decoded to take the effect on the picture quality into account. A encoded packet receiving apparatus, adapted for receiving the multiplexed transmission packet 713, output from the encoded packet transmission apparatus 7 of the present embodiment, is the same in structure as the encoded packet receiving apparatus 5, explained with reference to
A fourth embodiment of the present invention is hereinafter explained.
In the encoded packet transmitting apparatus 8 of
In
Alternatively, the encoded packet transmission decision unit 804 refers to the characteristic parameters in the high compression encoded packet 811 to decide on the high compression encoded packet 812 to be transmitted adaptively. The encoded packet transmission decision unit 804 refers to a parameter which will affect the picture quality of a decoded picture by bit error or packet loss, such as motion vector, to decide on the high compression encoded packet 812 to be transmitted adaptively.
The operational effect of the fourth embodiment of the present invention consists in the fact that increase in the band to be taken up in transmitting the second encoded packet as well as deterioration in the picture quality may be suppressed to the minimum.
The reason is that, in determining whether or not the second encoded packets (duplicated packets) are to be transmitted, to the first encoded packet 809, packet data are re-encoded so that the compression ratio will be higher than that of the first encoded packet 809, and is also that at a preset period, that is, once every n encoded packets, or by referring to a parameter as a characteristic quantity that will significantly affect the picture quality of the decoded picture, a packet which will contribute to suppression of deterioration of the picture quality of the decoded picture is selected and transmitted.
A fifth embodiment of the present invention is hereinafter explained.
In the encoded packet transmitting apparatus 9 of
In
The operational effect of the fourth embodiment of the present invention consists in the fact that increase in the band to be taken up in transmitting the second encoded packet as well as deterioration in the picture quality may be minimized.
The reason is that, in determining whether or not the second encoded packets (duplicated packets) are to be transmitted, for the first encoded packet 809, the second encoded packet are re-encoded so that the compression ratio will be higher than that of the first encoded packet, and that the encoded packet picture quality comparison/transmission decision unit 904 actually performs the decoding processing to refer to the effect on the picture quality.
It should be noted that the source of the compression-encoded moving picture data, such as 107 of
As explained in detail hereinabove, the present invention gives rise to the following effects:
The first effect of the present invention is that the decoded picture may be prevented from becoming deteriorated in picture quality even despite bit errors and packet losses on a transmission line of data obtained by compression-encoding of moving picture.
The reason is that, in the present invention, the compression-encoded moving picture data are duplicated and transmitted at time intervals which are either constant or are changed adaptively in order to take account of burst characteristics of bit errors or packet losses occurring on a transmission line. By this configuration, the probability of bit errors occurring in both the moving picture packets transmitted and the duplicated packets or the probability of both packets being lost may be lowered.
The second effect of the present invention is that, in transmitting the two sorts of the compression-encoded moving picture packets, it is possible to suppress the bandwidth taken up by the packet being transmitted.
The reason is that, in generating the second encoded packet from the input first encoded packet, according to the present invention, the first encoded packet is first decoded and subsequently re-encoded so that the compression ratio will be higher than that of the first encoded packet.
The present invention further gives rise to such a effect that, in determining whether a duplicated packet of the original packet (second encoded packet) is to be transmitted, the packet is actually decoded to refer to the effect on the picture quality, thus minimizing the deterioration of the picture quality of the decoded picture despite the occurrence of bit errors or packet losses on the transmission line for the compression-encoded moving picture data.
In addition, according to the present invention, the duplicated packets, generated on re-encoding the original packets, are decimated for transmission, thus suppressing the bandwidth taken up in transmitting the two sorts of the moving picture packets.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2002-126599 | Apr 2002 | JP | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP03/04738 | Apr 2003 | US |
| Child | 10972719 | Oct 2004 | US |
