CN1783304A - Treating play back bit stream - Google Patents

Abstract

A device reproduces a bitstream signal from an optical disc and controls the bitstream to ensure that only required bitstream data is coupled for MPEG decoding. The converter is repositioned to obtain the required bitstream data before the decoding of the previous MPEG image is completed. Bit stream data is read out prior to buffer storage to select required data for storage and reject unnecessary data. Buffer storage is reallocated for trick-play operations and accessed randomly to facilitate trick-play picture selection. Controls MPEG picture decoding and storage for frame decoding within a field period. The decoded image is stored and read out substantially simultaneously during the field period.

Description

The processing of play back bit stream

This case be that May 6, application number in 1998 are 98814153.1 the applying date, denomination of invention divides an application for the application for a patent for invention of " processing of play back bit stream ".

Technical field

The present invention relates to from the media reproducing digitally encoded signal, relate in particular to identification and processing with multiplexed form replay data.

Background technology

For example utilize the MPEG compression protocol adopt digital compression the Voice ﹠ Video signal record CD to be introduced as sound and picture quality that the consumer provides as broad as long with original material basically.Yet consumption user expects that the feature class that this digital video disk or DVD provide is similar to their analog video recorder or the feature of VCR.For example, VCR can reproduce forward or backwards with the speed that is different from writing speed.The playback features of this non-standard speed is also referred to as trickplay modes.Because image is divided into the compression graded characteristics with different compression level groups, the vision signal of mpeg encoded can not easily provide the special play-back characteristic.These groups are called the decoding sequentially of image sets or GOP and requirement.The detailed description of MPEG 2 standards is announced as ISO/IEC standard 13818-2.Briefly, MPEG 2 signal flows can comprise three kinds of image types with different content compression level.Intra-coded frame or I frame have compression minimum in three types, and can not decipher with reference to any other frame.Predictive frame or P frame are with reference to previous I frame or the compression of P frame, and acquisition is than the compression of intra-coded frame higher level.The mpeg frame of the third type is called alternating binary coding frame or B frame, can compress according to previous frame or the prediction of frame subsequently.The alternating binary coding frame has the compression of highest level.Three types mpeg frame is arranged among image sets or the GOP.GOP for example can comprise 12 frames that arrangement is described as Figure 1A.Because have only intra-coded frame to decipher, so each GOP can only decipher after the I frame coding with reference to any other frame.First predictive frame or P frame can be deciphered and store according to the modification of the previous I frame of storing.P frame subsequently can be by the previous P frame prediction of storage.The prediction of P frame is represented with crooked solid arrow in Figure 1A.At last, alternating binary coding frame or B frame can be by means of previous frame and/or frames subsequently, and for example the prediction of Cun Chu I frame and P frame is deciphered.In Figure 1A, described by the prediction of adjacent storage frame and deciphered the B frame with crooked dotted arrow.

Patented claim EP-A-0 696 798 discloses a kind of method for recording data and equipment, data medium and method of reproducing data and equipment.Application EP-A-0 696 798 lectures and write down mpeg signal on the various types of disc medium, discloses the record format that has in each sector of recording medium with the additional information of separating record as the data of subcode in addition.These subcodes provide the information relevant with every sector useful load data and are used in the playback duration control data to be reproduced.At United States Patent (USP) 5,535, in 008, " skipped mode " that disclose the MPEG record data reproduced, and for example utilizes CD-ROM.United States Patent (USP) 5,535,008 lectures the reproduction of a plurality of data be provided at predetermined intervals, and the data that therefore can find next to reproduce by " predetermined space of integral multiple is deducted first fixed value ".United States Patent (USP) 5,535,008 utilizes the mean distance between the I frame transducer to be moved around or rapid inverse repetition rapidly.In EPO application A-737 975, a kind of MPEG recording method that is used for CD is disclosed.The form of record comprises that management area and program data have the program zone of hierarchy.The form that disclosed form and digital universal disc or DVD adopt is similar a bit.At United States Patent (USP) 5,543, a kind of digital playback device is disclosed in 925, be used to reproduce the photographs of digital processing.Digitized image writes on the compact disk, according to expression prerecord sequence or user definition show the data presentation of storing on screen.United States Patent (USP) 5,543,925 have lectured the digitized image for each storage, and image file contains a plurality of son files, and their definition have the same scan image of different resolutions.It points out that these many resolutions patterns of same image reduce the stand-by period that image shows valuably.Apply for disclosing in 0,651 391 the high-speed replay digital coding image at EPO.EPO applies for that 0,651 391 especially discloses two frame memories being used for during " frame returns reproduction " a plurality of GOP of storage and the purposes of three frame memories.Output signal selection is replaced between decoded image memory.Storer selection and output image retention time are in response to the time of decoding image availability and next required image of retrieval.

The graded characteristics that comprises the coded frame of mpeg image group needs the I frame of each GOP and P frame to decipher at forward.Like this, by jump back to effectively one early or previous I frame and decipher this GOP at forward then, the reverse mode feature can be provided.The frame of decoding is stored in frame buffer memory, is used for later backward reading to obtain required reverse program sequence.Figure 1B illustrates under normal speed and the forward playback of the time before moment t0, selects reverse, triple speed degree pattern under the special-effect mode.This trickplay modes starts at moment t0, the decoded and demonstration of I frame I (25) here.Next frame that need decipher is I frame I (13), and like this, transducer is relocated, as arrow J1 indication, to obtain frame I (13).Behind recovery and the decoding I frame I (13), transducer such as arrow J2 indicators track are to obtain and decoded frame P (16).The repetition of this process such as arrow J3, J4 indication.Catch with decoded frame P (22) after, transducer such as arrow Jn describe to move to recover frame I (1).In order to describe scene motion requirement decoding smoothly and to show I, P and possible B frame.Repeat redirect and playing process for previous GOP, therefore, terminations of advancing, back by this record, the while with reverse sequence the video output smoothing describe program material.

Provide visible smooth reproduction to require to retrieve CD in time and from the specific image of storage access at the special-effect mode reproduction period.Provide the navigation of image inlet point (navigation) data to encode though each optical digital disk adopts in each Video Object Unit, they are limited on number, and may obscure image motion inherently in time.For with multiple speed forward and oppositely on the acquisition time level and smooth special-effect mode reproduce, require access and all coded images of decoding.Though can be that cost obtains this performance to sacrifice memory capacity, being chosen as of bit stream analysis and buffering storer be used to improve special-effect mode by effective memory and reproduces chance is provided.

Summary of the invention

A kind of equipment is from the optical disc replay bit stream signal.Controlling this bit stream is coupled with the bit stream data of guaranteeing to have only requirement and is used for MPEG decoding.Transducer is relocated the bit stream data to obtain to require before finishing previous mpeg image decoding.Bit stream data is read out to select the data that need store and to refuse unwanted data before buffer-stored.Buffer-stored be reallocated be used for trick play operation and by random access so that the special play-back image select.Decoding of control mpeg image and storage are so that at the frame coding of a field in the cycle.The image of decoding is stored simultaneously and is read in the cycle in field basically.

In a kind of configuration of the present invention, a kind of equipment is from the media reproducing digitally encoded signal.This equipment comprises: transducer is used for the converting digital coded signal and from wherein producing bit stream; Processor is coupled to receive bit stream, is used for controllably handling bit stream; Storer is coupled to processor, is used for the bitstream information of stores processor; Controller, be coupled with control store and processor, be used to control the identification of bit stream internal information, its middle controller processor controls is with specific sectors type in the identification bit stream, and according to this specific sector identification, the particular sector of controller control store storage identification.

In a kind of method of the present invention, a plurality of beginning codes obtain a beginning code in the data stream of arranging in a plurality of sectors of reproduction period conversion by optical digital disk equipment.This method comprises that search data stream is to locate the step of sectors type specific in a plurality of sectors.Search for this specific sectors type to begin code from one of a plurality of beginning codes location.Test this and begin code to determine whether it is incomplete.Search data stream is to locate second sector of sectors type specific in a plurality of sectors.Second sector of searching for specific sectors type is with second beginning in the beginning type of code location code from a plurality of beginning codes.Determine whether that second beginning code is the remainder of imperfect beginning code.Incomplete beginning code value and remaining beginning code value are made up to form complete beginning code.

Another kind of present device from the media reproducing digitally encoded signal comprises: transducer is used for the converting digital coded signal and from wherein producing bit stream; First memory is coupled to transducer, is used to store this bit stream; Be used to store the second memory of data, controllably be coupled from first memory; Controller, be coupled to control first and second storeies, be used to control the identification of bit stream internal information, its middle controller control first memory is with from particular sector address output bit stream wherein, and this controller control second memory is with the first of storage from the bit stream of specific sevtor address output.

Another kind of present device from the media reproducing digitally encoded signal comprises: transducer is used for the converting digital coded signal and from wherein producing bit stream; Storer is coupled to transducer, is used to store bit stream; Processor is coupled to storer, and the bit stream that is used to handle storage begins code with the MPEG that identification is included in wherein.The bit stream that processor is searched for storage begins code with the MPEG in the bit stream of identification storage, and begins marking code according to MPEG, and the MPEG that processor indicates this sign and storage to discern begins the sevtor address in the code.

The another kind of method of carrying out image decoding and demonstration in the equipment of optical digital disk of reproducing.This method comprises the step of conversion from the digitally encoded signal of CD.Digitally encoded signal is stored in the first memory.Decipher this digitally encoded signal to produce image.This image is stored in the second memory.Coupling is used for showing from the image of second memory.The storage of control second memory and coupling are used for showing basically generation simultaneously.

In another kind of the inventive method, a field store of image is coupled in first field of second memory and from the previous image of second field of described second memory and is used for showing.

In another kind of the inventive method, a field store of image is coupled in first field of described second memory and from the previous image of first field of second memory and is used for showing.The storage of described first field of described second memory and coupling are used for showing that being controlled in field sequentially took place in the cycle.

In another kind of configuration of the present invention, be used for comprising from the equipment of CD media reproduction digitally encoded signal: transducer is used for the converting digital coded signal; Storer is coupled to transducer, is used to store digitally encoded signal; Code translator comes from wherein deciphering image according to digitally encoded signal; Controller is used for storer, wherein, under first operator scheme, controller is according to first sequence, and control store is to read the digitally encoded signal of storage from storer, under second operator scheme, the controller control store is read the digitally encoded signal of storage from storer with second sequence.

In another kind of configuration of the present invention, a kind of equipment is used for from CD media reproduction digitally encoded signal and comprises: the bitstream source of expression digitally encoded signal; Processor is coupled to bit stream, is used to handle this bit stream to extract wherein at least the first and second data types of expression; Storer controllably is coupled to processor to store in first and second data types; Controller, coupling is with the distribution of control store, and wherein under first reproduction mode, the controller allocate memory is stored first data type, and under second reproduction mode, the controller allocate memory is stored second data type.

In another kind of configuration of the present invention, one is used for comprising from the equipment of CD media reproduction digitally encoded signal: transducer is used for the converting digital coded signal; Processor, coupling be with receiving digitally encoded signal, is used for handling and from wherein producing image; Storer is coupled to processor, is used to store this image; Controller, coupling is with control store and processor, and wherein under first pattern, image is stored in the storer, and this image is by double sampling and be stored in the storer under second pattern.

In another kind of configuration of the present invention, avoided unnecessarily handling unwanted sector data.A kind ofly use up method of reproducing data in the CD player control sector of reading, comprise the step of change set of sectors, the sector that these set of sectors comprise the sector of needs processing and do not need to handle by making.Provide the sector of the needs except not needing the sector to handle, and handle the data sector that needs is wherein represented video information with extraction data to data processor.

In another kind of configuration of the present invention, exempted delay the bit stream path of conversion basically from the processing of control change device position.At the playback duration of optical digital disk equipment, a kind of method comprises the step that receives the first and second transducer addresses, is used for coming control change device position according to the playback address.Whether the address of resetting is relatively equated to detect between them with the first transducer address.Whether equate that according to what detect the running transform device is to the reposition of being determined by the second transducer address.

Description of drawings

Figure 1A has described MPEG 2 image sets.

Figure 1B has illustrated the image sets that writes down during the reverse trickplay of playback and triple speed degree.

Fig. 2 is the block scheme of the demonstration optical digital disk player that comprises that the present invention disposes.

Fig. 3 illustrates in greater detail Fig. 2 part and has described various the present invention's configurations.

Fig. 4 illustrates the optical digital disk player of Fig. 2, has comprised other useful configurations for Fig. 2.

Fig. 5 A and 5B are described in the demonstration bit stream before the track buffer.

Fig. 5 C-5D has described the example data in the memory buffer.

Fig. 6 is a process flow diagram, has illustrated that the present invention who is used for through the beginning code that distributes on the sector borders recovers disposes.

Fig. 7 is a chart, and the sequence of the present invention of the reverse trickplay pattern that is used for 3 times of broadcasting speeds (3 times) has been described.

Embodiment

Fig. 2 has described a kind of exemplary block diagram of digital video display.Piece 10 has been described tray, can bear the digital recording CD 14 by motor 12 rotations.Respond each signal data position and determine the length in each hole by 8/16 modulating-coding, digital signal is recorded on the CD 14 as the spiral session that comprises all holes.Record on the CD 14 is read by the pick device 15 of collecting laser light reflected illumination.Laser light reflected is collected by photodetector or optical take-up apparatus.Imaging device for example forms the lens of pick device 15 parts or catoptron by motor 11 servocontrol with drive to follow the session of record.Can be by reorientating the different piece that the imaging device enters record apace.Servocontrol motor 11 and 12 is driven by integrated circuit driving amplifier 20.Pick device 15 is coupled to light preamplifier piece 30, and this piece comprises the driving circuit that is used for laser light emitting device, and prime amplifier provides from the amplification and the equilibrium of the reflected signal of optical take-up apparatus output.Amplification and balanced replay signal from light preamplifier 30 are connected to channel processor piece 40, utilize replay signal to come synchronous pll here, and the 8:16 that this phaselocked loop is utilized to use in the demodulation record modulates.

By means of the Reed Solomon product coding in the piece that is applied in 16 sectors, the mpeg encoded bit stream is encoded and is used for EDC error detection and correction, and each sector includes 2048 bytes with load data here.Like this, proofread and correct by means of the Reed Solomon product of realizing in the ECC memory buffer 45 and 46 of Fig. 4, replay data stream is deinterlaced or backward shift (going to shuffle) and error correction after the 8:16 demodulation.Each buffer storage is as 16 sectors of the replay data stream of array arrangement, so that the required row and column product of deinterleaving and permission is handled.The ECC memory buffer of cascade is introduced the delay that is similar to (2*16*1.4) millisecond for the order bit stream that reproduces, and 2 expression ECC buffers are right here, and the sector of correction is used in 16 expressions, and 1.4 millisecond meters are shown in 1 times of sector cycle under the rotational speed.Like this, the order bit stream of reproduction is delayed and is similar to 45 milliseconds minimum value.

Error correction signal bit stream 41 is coupled to bit stream or machinery/session memory buffer 60A through link handler.The track buffer device comprises the DRAM type of memory and is used to store the replay data amount, will not produce any visible defects when making the data of losing during transducer or pick device 15 are reorientated in decoding.Like this, last output image stream will show as continuous or seamless for the beholder.Bit stream buffering storer 60A is the part of exemplary 16 megabit DRAM storeies.Another exemplary 16 megabit SDRAM storage blocks can be divided so that frame buffer 60C and 60D to be provided, they provide storage for the picture frame of at least two decodings, the video bit stream of compression is stored among the buffer 60B before decoding, audio bit stream in buffer 60E and other data storage at 60F, 60G, 60H.Channel processor 40 also comprises timing control circuit, and this circuit control is write bit stream buffer 60A by link 505.Because variation in playback session address, for example because user-defined playback of video content is selected as " director montage (Directors cut) ", basic (parental) guide or even at user option replacement angle lens (shot angles), data can be write the bit stream buffer off and on.For the ease of the signal of access more promptly and recovery record, CD 14 can be with the speed rotation of the bit stream of the conversion that causes high bit speed and the increase that may intermittently provide.

As has been described, the data stream of record is arranged in the ECC piece of 16 sectors.Each sector has unique sector identification address of the error correction bit protection of being handled by the ECC piece 47 of Fig. 4.Yet, because sevtor address is short and is that the sector is specific that any delay to sector address signal 42 that produces for correction process piece 47 is insignificant.Sector address signal 42 is coupled to provide positional information to servocontrol integrated circuit 50.Integrated circuit 50 drives and control signal for servomotor 11 and 12 provides.Motor 12 rotary CDs 14 and provide servo-controlled rotation with multiple speed.Optical take-up apparatus or transducer 15, in addition, can be controlled to reorientate or jump to another sevtor address apace according to sector address signal 42 location and servocontrol by motor 11 according to the sevtor address request, or the position on the optical disc surface, by I ² C control bus 514 transmits, and is coupled through the parts 54 of Fig. 4.

Digital video display is by the CPU (central processing unit) of piece 500 or CPU, parts 510 controls, and these parts receive reproduction bit stream and the error flag of self-channel IC 40, and steering order is provided to servo IC 50.In addition, CPU 510 receives the user control command from user interface 90, and from the MPEG code translator control function of the MPEG decoder unit 530 of piece 500.System buffer memory 80 is addressed and provides data to CPU 510.For example, buffer 80 can comprise RAM and PROM memory unit.RAM can be used to store the various data of being extracted from bit stream 41 by CPU 510, and for example this data can comprise descrambling or decryption information, bit stream and frame buffer memory management data and navigation data.PROM can for example comprise useful transducer redirect algorithm, and this algorithm is convenient to realize the trick mode operation of a selected velocity in the speed forward or backwards.

In Fig. 3, the mpeg encoded bit stream is coupled to link handler 505, and it can go multiplexer with audio frequency, video and the control information of mpeg encoded from the bit stream of DVD form separately as hardware.On the other hand, the direct memory access of the 510 couples of buffer 60A of CPU by Fig. 3 or the software control of DMA, can the completion bit diffluence multiplexed.Before track buffer device 60A or inner coding stream by microcontroller 510 search with the location with read title and extract navigation data.Below with reference to Fig. 6 useful bit stream search is discussed.

Microcontroller 510 is through I ²C control bus signal 514 is coupled to front end reorientates the next sector that obtains by the special play-back sequence of requests with control or request transducer.Transducer location can be controlled by useful storage sequence or redirect play mode, and this pattern is retrieved with reference to the sevtor address and the GOP sevtor address that are included in the playback that the navigation bag data among each Video Object Unit or the VOBU read.Exemplary sector address and VOBU navigation bag has been described in Fig. 5 A.Yet, after transducer is reorientated, can be from the sector of front end initial retrieval by exemplary microcontroller 510 identifications, because be not that jump instruction is asked.Like this, microcontroller 510 rewrites these unwanted data in track buffer device 60A and guarantee to have only the data of request to appear in the buffer valuably.

After having discerned sevtor address or title, the direct memory access (DMA) of microcontroller 510 control buffer 60A, it separates mpeg data and the data that are stored in other DVD forms in the buffer effectively.Like this, video DMA 515 coupling is used for being stored in exemplary video position buffer 60B compression video bit separately.Similarly, the audio frequency position of compression is read and is stored in the audio buffer 60E from buffer 60A.Sub-image data is also retrieved and is stored in the buffer 60F from track buffer device 60A by DMA.

Compressed video bitstream among the buffer 60B of search video position is with the beginning code by beginning code detector 520 positioning images or higher level.The beginning code signal 512 that detects is coupled to microcontroller 510, and this microcontroller passes through signal 511 and 530 communications of MPEG code translator subsequently, to indicate next image type, balanced device setting and to start decoding.Decoder state signal 513 coupling gets back to that microcontroller 510 is finished decoding with indication and this view data can be used for showing or storage.As will be described, can think the video bit buffer 60B of compression as FIFO or annular buffer, the bit stream of storage is used for MPEG decoding by access sequentially here, yet the random access by buffer 60B can promote trick mode operation valuably.

Video bit stream in the MPEG code translator 530 is handled by variable length decoder 531, and this code translator search bit stream begins code with location amplitude limit and macro block.When obtaining or construct other image for example when the P of GOP and B image, be written to frame buffer 60C and 60D is used for subsequently as fallout predictor from some decoding image of each image sets.Frame buffer 60C and 60D have the memory capacity of at least two frame of video.Audio packet separately is stored among the buffer 60E of audio frequency position, and it is read out and is coupled and is used for the audio coding of piece 110.Behind MPEG or AC3 audio coding, digital audio and video signals produces, and it is coupled to audio post-processor 130 and is used for digital-to-analog conversion and produces various base-band audio signal outputs.The digital video output signal of the decode block of reading from reference frame buffer 60C/D is converted to the raster scanning form by Display Register 580.Yet during trick mode operation, signal source output may be from untapped storer during the trick mode operation and the field store device that reconfigured valuably.Like this, the piece in Display Register 580 can be controlled according to trick mode operation valuably to the raster scanning conversion.Display Register is coupled to scrambler 590, and this scrambler provides the digital and analogue signals conversion and produces the vision signal of base band video component and coding.

Can consider the operation of demonstration video machines illustrated in fig. 2 with reference to Figure 1B, Figure 1B illustrates forward play and reverse trickplay sequence.As mentioned above, the encoding relation that exists in each GOP needs each image sets to begin decoding with forward from I frame or image.Like this, by rebound effectively with conversion early or previous I image and decipher this GOP at forward then, the reverse mode characteristic can be provided.The image of decoding is stored in and is used in the frame buffer memory reading with reverse order subsequently.Yet, comprise that the sequence of B image can further be utilized the useful feature that will describe.In Figure 1B, suppose a certain moment before time t0, for example in I image I (1), the exemplary video player is supposed according to user command under the forward play condition.I, B that connects as arrow line among Figure 1A and the explanation of P frame, decipher each image sets at forward.A moment before time t0, the reverse trick modes of three times of broadcasting speeds of selection, and in the time t0 startup, the decoded and demonstration of I image I (25) here.As mentioned above, the required next image of reverse trickplay decoding is I image I (13), and like this, as arrow J1 indication, transducer is moved to obtain image I (13).Then, the play sequence of indicating among Figure 1B is followed in signal recovery and decoding, is obtained I (13), is obtained P (16), obtained P (19), obtained by arrow J4 acquisition P (22) and by arrow Jn by arrow J3 by arrow J2 by arrow J1 ....As each trickplay modes special requirement, the B image of the insertion shown in Figure 1B is converted, but can be for example by rewriteeing buffer or forbidding abandoning by code translator.For fear of previously described demand, various beneficial method have been used for MPEG code translator, memory buffer control and distribution for additional reverse mode video cache.

Can bit stream 41 or track buffer device 60A in the sector of quoting be that unit finishes determining of view data.Yet, in the DVD data layout, bury and do not retrain and begin simultaneously because mpeg image begins code with sector borders, with the sector section that position as a result that the image of unit begins code comprises previous, a possible non-video sector inevitably.Fig. 5 A illustrates exemplary bit stream 41 parts that comprise Video Object Unit, and this unit comprises audio frequency and video and sub-image data sector.Each sector comprises 2048 payload bytes, has the sevtor address shown in the sector borders shade.54 finish and and then follow the beginning code that is useful on video image B in the sector at video image A shown in Fig. 5 B.Yet video image B begins remainder 65 generations in the sector of code, and the sector 55-64 of insertion comprises subimage and voice data.Explanation is the definite or location of the view data/video sector of unit with the sector in Fig. 5 C, and here, the beginning code that is used for exemplary image A is shown in the sector 2, and the beginning code of next image B is 9 generations in the sector.Equation 1 illustrates the view data position by sector count, because image A 2 beginnings and 9 end in the sector in the sector, image A has the duration of 8 sectors.Fig. 5 C has illustrated unwanted data fragment, describes video data with reference to (video) sector number here.Yet this video sector number may be directly related with address or sector number in reproducing bit stream.In Fig. 5 C, video bit stream illustrates with the exemplary image A that describes, and the image that image A has in byte 1000 startups of video sector 2 begins code.Very clear, 999 bytes of 2 fronts, sector are corresponding to the data from previous image.When with the byte being the unit location, view data may use more detailed processing.The accurate processing of byte may require the more complicated storer control more required than sector level degree of accuracy.Yet, if use byte accurately to handle, whole image data can only be stored in the video bit buffer, like this, these fragments have been eliminated and have avoided the hang-up of MPEG code translator 530.Determine that at the byte exact image that is used for exemplary image A shown in Fig. 5 C, image begins code and begins code byte 500 beginnings of 9 in the sector in byte 1000 beginnings of video sector 2 and image B here.Therefore, by user's formula 2, the size of image A may be calculated 13,835 bytes.Like this, the specified byte that byte exact image address allows microprocessor 510 to point among the exemplary video position buffer 60B, the variable length decoder VLD531 of Fig. 3 byte thus begins decoding.

If being unit with the sector, view data determines that the image fragment that abandons that produced is suspended before or after the MPEG code translator of video bit buffer reading images must prevent owing to the image decoding at needs.Described this image fragment in the demonstration video bit buffer of Fig. 5 D, this illustrates the sector of a plurality of P of comprising and B image, illustrates from the mode of unwanted data previous or later image with the diagonal line shade here.Each video object module unit or VOBU comprise navigation data, the reference of first two back of GOP of the end sevtor address of first I image of this data identification and VOBU or the last sevtor address of P image.In addition, navigation data comprises formerly and the sevtor address of the I image among the VOBU subsequently, therefore, can easily provide the I that only is used for special-effect mode image.Yet,, can avoid the problem that produces by the image fragment if can discern the end byte of required image.Microprocessor 510/A for example ST20 type is configured to hardware search engine valuably, and the data that its search is stored in the track buffer device are stored in the end byte of the I image in the end sector of buffer 60A with the location.Like this, by identification I image, it can be loaded into video bit buffer 60B separately, therefore, has avoided causing the parts of images storage of code translator lockout issue.Example microprocessor 510/A can be used to find out the only beginning code under the I image model, because known the end sector by navigation data.Yet for P, B or a plurality of I image, example microprocessor can not provide actual solution, because must finish the test of each byte of bit stream data, and the operational concentrated use of its expression microprocessor 510.

Begin the location of code before can realizing easily deciphering by a kind of configuration and determine, the LI(link interface) piece 505 of this configuration using Fig. 3 is searched for the beginning code in the bit stream before the track buffer device 60A.This use of LI(link interface) 505 provides early stage pre-service or analysis to the image and/or the audio title that may send to microprocessor 510 valuably.Like this, discerned the title of incoming bit stream before the track buffer device after, the image and the audio frequency that are required by specific special-effect mode can be stored among the exemplary session storer 60A, unwanted picture and/or other data are deleted by being overwritten in the buffer simultaneously.

In first kind of configuration, locate the beginning code, the bit stream among this detector searches machinery/track buffer device 60A or the video bit buffer 60B by using beginning code detector 520.Though this method has an advantage, promptly to begin the design of code detector be known to MPEG, yet detecting device requires continuous data.Therefore, have only data in the video bit buffer, removing DVD and transmit data structure can be searched.Like this, mpeg data in search mechanism/track buffer device may be difficult to realize that can not optimally use storer, example microprocessor 510 may be owing to some interruptions are increased one's load, like this, require to increase by second microprocessor for example 510A be exclusively used in and realize the beginning code detection.

In a kind of useful configuration, be convenient to realize the beginning code detection by the beginning code detector, the beginning code detector before the track buffer device 60A or among search be exclusively used in the bit stream that MPEG begins code.Like this, by providing early stage analysis to MPEG video title in the bit stream valuably, can estimate special play-back image demand and can finish the storage operation that is exclusively used in trick play operation.During trick mode operation, same useful analysis can be applied to the video packet flow before the video bit buffer.For example, under the reverse reproduction pattern, the specific selection of the special play-back between this pre-service permission image is buffered and is used for decoding, and those unwanted pictures were dropped before storage.During trick play operation, this image is selected for example to abandon the B frame and may be similar to the quantity that makes the I that is stored among the buffer 60B of exemplary video position and P image and expand and be twice.Like this, never need to identify in the data that to need data are direct results of pre-service or analysis before buffer-stored, it allows video bit buffer 60B only to store specific image that need or special play-back.Therefore, can store the particular video frequency object element or the VOBU of more special play-back, so that the reproduction of level and smooth special play-back motion.

In a kind of useful configuration, by the storage data of only selecting to use subsequently, during trickplay modes, the memory capacity of track buffer device 60A and video bit buffer 60B increases.For example, in exemplary trickplay modes, the B frame may be not decoded, therefore, do not need to be stored in session or the video bit buffer.Like this, have only the image that needs to be stored, unwanted picture or other data are dropped.For the ease of realize needing and not needing this useful selection between the image, require bit stream or video packet flow pretreated, analyze or search, with positioning sequence _ title, image sets _ title or image _ title before storage.Like this, the analysis of compression position flow or pre-service allow definite MPEG parameter such as time _ code, closure _ image sets and disconnection _ link data to be used for each image sets or GOP.In addition, by the pre-service stream of packets, image _ beginning _ code can be positioned, and allow to handle image _ title like this, next, for example allow to determine time _ reference, image _ coding _ type (I, P and B).Yet as has been described, because DVD is divided into the sector of 2048 bytes with the data of MPEG type, this useful MPEG analyzes and is difficult to.In addition, (4 bytes) is not sector alignment because MPEG begins code, can begin code through sector borders allocation example image.Bit stream before Fig. 5 B explanation track buffer device 60A, video image A 54 finishes and the back has been followed the beginning code that is used for video image B at once in the sector here.Yet the remainder that video image B begins code produces in sector 65, inserts sector 55-64 and comprises subimage and voice data.Fig. 5 C explanation was removed multiplex video sector bit stream before video bit buffer 60B, the beginning code that is used for exemplary image A here is shown in the sector 2, and the beginning code of next image B produces in sector 9.The beginning code that is used for the distribution of image C occurs, and 12 the 2046th byte starts and 13 continue in the sector in the sector for it.Therefore, the remainder of part and the next video sector of beginning code is in a video sector.

In order to analyze the bit stream of the beginning code with distribution, Fig. 6 illustrates a kind of demonstration methods of the present invention.This demonstration methods identification and memory sector type and address, in addition, the beginning code that identification and storage need.The application of the invention partly begins code signing, and this mark is represented to begin code and occurred, beginning code identification and storage allocation or part.The beginning code remainder that occurs in next video sector is identified and recovers to finish the beginning code.The inventive method of Fig. 6 has been described the search and the MPEG that are applied to track buffer bit stream 41 before and has been analyzed.The search bit stream, the sector that is used for needs is the video sector for example, then the beginning code of search distribution.Other non-video sectors of audio frequency, subimage, navigation data etc. can separate with the beginning code that distributes by for example comprising.Like this, searched and the video sector subsequently of bit stream is identified and handles, the non-video sector of current unwanted insertion simultaneously is for example not processed during specific special-effect mode, and can be dropped in storage on the exemplary session buffer 60A or before rewriteeing.Like this, discerned next video sector after, the search integrated data is to locate next beginning code.Yet, because beginning code, part is set up, and the remainder that partly begins code is searched, and the code combination of this remainder and previous video sector is to finish the beginning code simultaneously.

The explanation of the exemplary view of Fig. 6 is used for the bit stream search with sevtor address, image type and the address of identification needs and the inventive method that is used to detect and ressemble the beginning code of distribution.This method is searched for the error correction bit stream here with the need sector specific from location, a plurality of sector in step 10 beginning, and these sectors comprise navigation, audio frequency and video sector, sub-image data sector.Detect the video sector in step 100, here, " deny " and the result form and continue the loop that bit stream is searched for.Therefore similarly, audio sector can detect and its sevtor address is stored in step 105.If step 100 test for " be ", then detect the video sector and this sevtor address is stored in step 101.The video sector of detecting starts another test to detect the beginning code in the video sector in step 200.Step 200 has been described image and has been begun code, yet, may there be various beginning codes, for example sequence-header, GOP title or image header all exist in the video sector, and therefore, any one all may distribute through sector borders.In step 200 " deny " loop that continues to begin in the search video sector code formed.In step 200 " be " expression detects the beginning code, this code starts another test to begin code in step 250 test section.As Fig. 5 B and 5C explanation, when occurring interrupting owing to sector borders and sevtor address, can think simultaneously and determining between generation step 200 and 250 parts of describing and the whole beginning code sequentially, part or incomplete because any beginning code becomes.In step 250 " deny " form and wait for that part begins the loop that code takes place.In addition, in step 250 " deny " also represent the detection of complete beginning code, it in step 255 test to determine whether the being type of needs.The beginning code tester that needs in step 255 for " be " cause type and byte location in step 260 memory sector address.

Beginning code detection in the part of step 250 obtains " be ", it makes this sequence restart to search for bit stream and locatees next video sector to return step 100 by loop.In step 250 " be " also the test of setting up procedure 300 to be to determine whether that being provided with part begins code signing.Part just is set begins code signing up to detecting beginning code first distribution or part.Like this, in step 300 " deny " make that in step 350 part being set begins code signing, in addition, begin the value of code at step 400 storage area.In step 300 " be " expression detect distribution the beginning code remainder and cause reset portion to divide the beginning code signing in step 500.In step 300 " be " the beginning code remainder that detects in step 450 storage also caused.In step 550, the beginning code that begins code value and be combined and distribute from the part of step 400 to improve from its remainder of step 450.At last, in step 575, store improved beginning type of code, byte and sevtor address.Therefore, specific sectors type and address are discerned and stored to the inventive method of description, beginning type of code and byte address in identification and the memory sector, and discern and ressemble the beginning code segment of distribution.Like this, before buffer-stored, can analyze the DVD format bit streams to determine the image type of for example specific mpeg encoded.

According to the known position that image begins and stops in the video bit buffer, can control mpeg image decoding order valuably.Like this, by picture position among the known video bit buffer 60B, for example Fig. 5 C explanation or as the bit stream search of Fig. 6 determined, allow to begin the storer start pointer required random access image during orientation such as the trick mode operation valuably in code detector 520 and the variable-length detecting device 531.Reverse operating in broadcasting speed and/or motion playback slowly requires to reproduce the B frame.By the order of reverse adjacent B image decoding,, can simplify this reverse mode operation valuably according to the memory buffer requirement.Obtain this reverse decoding order valuably by the storer start pointer is set, make it possible to decipher the required image of special-effect mode.In addition, during trickplay modes, by as specific special play-back algorithm require skip or do not read image in the video bit buffer valuably, can simplify the size and the control of memory buffer.During the special play-back buffer memory, by making a plurality of image decodings immediately or as the carrying out of special play-back algorithm special requirement valuably, further optimizing memory size and control.Provide these useful feature requests control carefully read/write function and between them synchronously.

During trick mode operation, during the reverse-play speed operation, require maximum image buffer capacity to be used for reading especially with reverse order with the memory image group.During this special-effect mode, some player function or feature may not be required, can be useless or unavailable.This function or characteristic comprise audio frequency, multilingual, subimage and screen display and all utilize the memory buffer capacity.Like this, during trick mode operation, can redistribute the untapped memory buffer capacity of these functions or feature so that other image storage to be provided.Yet, at some special-effect mode, for example fast during the play mode, may be for the accompanying audio of high-speed retrieval and be corrected and help the spacing of scene location to have a useful requirement.In addition, may require a limited screen display with indication trick play speed and direction.Like this, untapped memory buffer capacity can dynamically reconfigure with the image of being convenient to buffer memory compression under the special-effect mode valuably, the frame fallout predictor and the video display field of decoding.

In the present invention's configuration, SDRAM memory buffer 60E-60H redistributes on function between forward and trickplay modes operation.Distributing to the memory span of audio frequency 60E, subimage 60G during the forward play pattern can use during special play-back so that other compressed image storage to be provided, to increase video bit buffer 60B and to provide other fallout predictor frame to be used for decoding.Similarly, for example memory buffer can be reallocated, may not require the unnecessary compressed image of storage for some trickplay modes, like this, do not use or the output display buffer memory of describing with the 60H that provides as Fig. 3 is provided for the memory buffer capacity that do not require.The output display buffer memory can stored video data frame or field be used for showing.The storer of this dynamic assignment is convenient to output image and as fallout predictor, like this, has simplified the storage administration during the trick mode operation.As has been described, select to start storer by the user and redistribute, yet, discern the useful image prediction that obtains by the requirement of special play-back sequence and/or the use of useful storage from the bit stream analysis of compressed image type, can determine dynamic assignment.

In the useful configuration of another kind, during trick play operation, by the horizontal double sampling of view data to deciphering before buffer-stored, the frame buffer memory capacity can be expanded effectively to be twice.According to the special-effect mode control command of coming self-controller 510, for example the horizontal double sampling that realizes by demonstration piece 62 is carried out flatly average to the right value of adjacent pixels.In Fig. 3, signal S1 represents to be coupled to the full bandwidth data of double sampling device 62, and signal S2 represents the double sampling output data.Like this, the double sampling image comprises approximately half original pixels quantity, therefore, requires half memory span to allow image or the frame of video capacity storage by field.Like this, by the double sampling of level during trick play operation, can be as the other frame buffer memory of use of special play-back algorithm requirement.In addition, in order to increase the special-effect mode memory span, the use of double sampling of the present invention has reduced the data and address bus control of storage manager during the special-effect mode storage access valuably.For example, have only the data of half in the time of half, to transmit, therefore, simplified storer control and management.

The buffer 60H that the double sampling image of level is redistributed from for example 60C, 60D or the present invention reads by double sampling device 62 and recovers.In Fig. 3, signal S3 represents that the reference image data of the double sampling of reading from storer is used for pixel counts and recovers.Double sampling device 62 can each double sampling memory cell twice of addressing, yet this action doubles the data and address bus utilization, and this utilization is reduced during stores processor valuably.Therefore, the double sampling image is by duplicating each pixel value and be resumed and output being used for before MPEG decoding directly coupling MPEG code translator 530 as signal S4.Though this method is twice the capacity register expansion and has reduced the data and address bus utilization, horizontal space resolution has reduced.Yet the reduction of this horizontal resolution occurs in during the trick play operation, but because the image motion speed that increases, people's psycho-visual sensation may be difficult to perceive this reduction.

The block scheme of Fig. 4 illustrates identical functions and the parts label of describing as Fig. 2, yet Fig. 4 comprises the other the present invention's configuration that will explain.

Can think that the exemplary digital video disc player shown in Fig. 2, Fig. 3 and Fig. 4 comprises two parts that are called front-end and back-end.Front-end control CD and transducer, rear end provide MPEG decoding and The whole control.This function is divided the tangible solution that may represent MPEG decoding that coordinate, steady state (SS).Yet, for example during the trick mode operation and special when the time with reverse-play, this processing of usefulness rear end and the division of control, the microprocessor overload that may become.

As has been described, require microcontroller 510 management in the past the incoming bit stream 41 received of termination and from unwanted data, identifying need.In first kind of useful device, bit stream 41 controllably is coupling between the front-end and back-end.As has been described, in the demonstration player of Fig. 2, optical take-up apparatus or transducer 15 can be reorientated.The sevtor address that obtains in the rear end is through I ² C control bus 514 is sent to front end servo-drive system 50 to reorientate transducer 15.Yet, optical take-up apparatus or transducer 15 according to sevtor address by servocontrol, this address by truncation to remove least significant bit (LSB).This address truncation allows to catch the sector in the group or the piece of 16 sectors.Need this grouping so that during writing down by means of the ReedSolomon product coding be applied in 16 effective load datas on the sector and interweave and carry out error correction (ECC).Like this, obtain this information from the ECC of 16 set of sectors of CD, usually, the retrieve data that comprises the sevtor address of needs is in advance or the previous address by the back-end processing request.In addition, converter using radially or tangential motion move with respect to rotary CD, obtaining comprising the session of sector ECC piece, the sevtor address that needs resides in these sectors.Like this, after reorientating, along with CD rotates towards the ECC sector piece that comprises sevtor address requirement or needs, transducer focuses on again and the sector is converted.Therefore, if consider the most bad situation in location of transducer and the sevtor address that needs, may change a hundreds of unwanted sector.Because along with the quantity of the CD radius sector that increases increases, the quantity that does not need the sector of reproduction also increases.In addition, may require complete CD rotation than catching of address morning or previous, it can produce does not need the sector to reproduce.Like this, before taking place, the sevtor address of needs produces a large amount of data that do not need.This bit stream is described as signal 44 in Fig. 4, and comprises and be coupled needs and the unwanted data that are used for ECC piece 45 and 46 error correction.The error correction bit stream is handled output as signal 41 from ECC, and this signal is coupled to the rear end, and microcontroller 510 never needs the data of identification needs in the data here.

Fig. 4 illustrates a kind of configuration of the present invention, and control assembly 45A for example transmission gate or logic function are exported and passed through to data-signal 44 from 8:16 code detuner here, is coupled to Reed Solomon error correction block 45 and 46.Control assembly 45A is by parts 43 control, the function of parts 43 with error correction in piece 47 and as address signal 42 outputs, the current playback sevtor address that recovers and the next data that need of expression for example the sevtor address 53A of the rear end acquisition of image type compare.Can realize this comparison easily by comparer or logic function.Like this, when playback sevtor address 42 equaled the address 53A of rear end requirement, signal 43A allowed demodulating data output to be coupled to error correction buffer piece ECC 45 and 46.Because error correction applications in 16 set of sectors, is carried out the comparison that requires address and actual address, making to comprise needs the ECC of sector sector piece to can be used in Reed Solomon correction.For example, adopt the address of the least significant bit (LSB) of truncation to be convenient to realize that sevtor address relatively.

For example, because the mpeg image of category-B type may take 3 sectors, may require 30 or more sector as I type mpeg image here, so required sevtor address is represented the primary data sector of the image type of needs.In addition, the signal 43A that expression needs and the playback sevtor address equates basically can be considered to represent lock function, and logic state is held up to the address of needs and is changed promptly till requiring another transducer redirect here.The reception of new sevtor address has changed the state of signal 43A, and its stops the data reproduced till the address of new needs occurs in replay signal and is compared device 43 and detects.In other words, signal 44 keeps enabling error correction, and ECC piece 45 and 46 is enabled and output signal 41 is held, and perhaps briefly, CD continues to play till the different transducer locations of request.

The detection playback that can carry out the sector that needs by the sevtor address that compares truncation takes place, to guarantee that error correction buffer 45 and 46 fills up RS and proofreaies and correct required number of sectors.In another embodiment, utilize signal 45B can adopt identical detection to reset and take place, to control or to enable the operation of error correction memory buffer 45 and 46.In another kind of configuration of the present invention, have only the sector of requirement to be enabled through output control part spare 46A.The selection of parts 46A is different with the control that parts 45A and 45B provide, because the data layout that interweaves or shuffle, enables to comprise the ECC piece of required sector.Can finish the detection of the playback sector that needs by address relatively actual playback sevtor address and requirement or that need.Yet because this control function is basically in error correction with utilize memory buffer to finish after going to shuffle, last output signal 41 is delayed the time cycle of at least one ECC piece.Therefore, the output signal of error correction is corresponding to the set of sectors of changing before as the data (address) that occur the needs of identification at ECC buffer input end.Very clear, because buffer delay is known, for example by using the related method thereof of describing as t, this delay can compensated in the control coupling of the signal 43A of parts 46A.Control assembly 46A is described to a tandem type switch block, and it can enable or forbid being provided to the bit stream of rear end.Like this, suitably the signal 43A that regularly comes compensation deals and buffering to postpone can be applied to selectively enable deinterleaving bit stream 41 and be transferred to processing block 500.The use of the previous embodiment of the invention allows only to be coupled to the rear end from the translation data that requires the sector and is used for storage and decoding, like this, has reduced the work load of microprocessor 510.

As has been described, switching signal 31 demodulation in piece 40 is modulated to remove 8:16, and produces output signal 44 and 44A.Signal 44 is coupled and is used for deinterleaving and error correction, and the independent error correction of signal 44A quilt is to produce the playback sevtor address.In the ECC of Fig. 4 memory buffer 45 and 46, finish deinterleaving and error correction.The storage of each buffer according to 16 sectors of the replay data stream of array arrangement so that deinterleaving and/or the row and column product that enables to require handle.The ECC memory buffer of cascade is introduced a delay for the serial bit stream with the reproduction of 1 times rotational speed, it can be approximated to be (2*16*1.4) millisecond that calculates below, here 2 represent ECC buffers 45 and 46, the sector of proofreading and correct is used in 16 expressions, and 1.4 millisecond meters are shown in the sector cycle of 1 times of rotational speed.Like this, the order bit stream of reproduction is delayed approximate 45 milliseconds minimum value.

Handle bit stream 44A with error correction sector identification address at ECC piece 47.Yet because sevtor address is lacked and is that the sector is distinctive, error correction block 47 is introduced an insignificant delay for playback sector address signal 42.

As has been described, error correction bit is flowed through and is gone through an error correction delay.Receive bit stream 41 in the rear end, various here MPEG groupings separate with the DVD data.Video packets is stored among the exemplary buffer 60B to be deciphered by MPEG code translator 530.As mentioned above, code translator 530 transmits signal 513 and deciphers finishing of image to controller 510 to represent each, and it is then caught and wants decoded next image.Therefore, for example be labeled as the image that comprises in the video sector of A when finishing, produce signal 513 by code translator at Fig. 5 A at specific image.The exemplary next image that need be used for deciphering must recover from CD, and therefore, transducer 15 must be repositioned onto and comprise the sevtor address that needs image.Fig. 5 A illustrates bit stream 41 parts that are coupled to buffer 60A, comprises the Video Object Unit of being made up of a plurality of sectors, and each sector comprises video, audio frequency, subimage and navigation data.Because next sevtor address occurs, perhaps as after the MPEG of signal 513 indications deciphers, in track buffer device 60A or before can determine the end of sector A valuably.Therefore, mark among Fig. 5 A " next one " arrow the generation of the approximate timing of the next sevtor address request from microprocessor 510 to front end is shown.The I of the sector requests by postponing to send needs according to interrupt priority level ²The C control bus transmits this address and jump request.

In the useful configuration of another kind, the interrupt priority level that microcontroller 510 interrupts is resequenced between operator scheme.For example, under the forward play pattern, memory addressing and control request and trick mode operation down and special requirement during the reverse-play speed operation be different.During trick mode operation some feature and therefore their storer and the control of MPEG code translator be unwanted.For example, the processing of audio coding and subimage is unwanted during trick mode operation, therefore, address, DCB interrupt priority level can distribute lower priority, and higher priority is given from session and video bit buffer access graph picture.

In time catching of sector of request is particular importance during trick mode operation.Yet, as has been described, form control loop with a plurality of delay units according to the execution of the needs sector acquisition of back-end processing.Fig. 4 illustrates the present invention's configuration, and it has reduced the delay in sector acquisition, and briefly, it allows the last detection that needs the sector to reset appearance to start the new sevtor address that transducer moves to previous reception.Arrow B shown in Fig. 5 A the be positioned bit stream 41 of indicating playback or track buffer device 60A and to the time proximity relation between the sending of the servo control mechanism of the next one of the present invention/end sevtor address.In Fig. 5 A, shown in arrow B after the navigation bag shown in the shade has been read from bit stream, take place in very short time.In image A, the arrow that illustrates " next one " illustrate that signal 513 is similar to seven sectors (decoding is finished) takes place later on.Yet in fact, I and P types of image have comprised than what Fig. 5 A described and have manyed a lot of sectors, therefore, the arrow that sends corresponding to address and jump request " next one " the ratio explanation that takes place late a lot.Like this, catch and/or after useful image/sevtor address was determined and the form knot makes, the next one of the present invention/ends sevtor address was by microcontroller 510 generations at navigation bag.Using next one/ends sevtor address to identify the sevtor address of needs can be in time instruct with the transducer of redirect and separates.Employing is according to first not transducer redirect of needs sevtor address reproduction execution, and the next one/end sevtor address is preloaded in the transducer servo-drive system effectively.Postpone because sevtor address does not experience the tediously long ECC of bit stream 41, last do not need the sector occur from ECC piece 45 and 46 before the running transform device.

In Fig. 4, pass through I ²C control bus 514 communications of control data, this bus is delivered to servo-control system 50 with the playback sevtor address that the next one needs.The next playback sevtor address that needs is produced by microcontroller 510, and this controller is handled from the special play-back of storage, specific speed sequence, playback and navigation data storage or from the address date of useful definite reproduced picture data.Next address is from I ²The C bus is read and is stored in the parts 53.I ²The C data also comprise end of the present invention/last sevtor address, or first unwanted sevtor address.Finish/last sevtor address can be by recovering and the navigation data of storage obtains, however this only provides the predetermined image address of limited quantity, like this, for special-effect mode, adopts the end of useful definite image sector address.Finish/last sevtor address is from I ²The C bus is read and is stored in the parts 52.Last sevtor address can revised before the bus transfer or after reception, and to prevent the needing sector to lose, by for example sevtor address being added a unit count, guaranteeing addressing like this and detecting first does not need the sector.The address 52A of last sevtor address or modification is coupled and is used for comparing with the playback sector address signal 42 of exemplary comparator 51.Like this, when playback sevtor address 42 equaled address 52A, first did not need the sector to be converted and comparer 51 produces control signal 51A.Control signal 51A enables the coupling from parts 53, for example pass through the address date of loading or mobile storage to servo control mechanism, or as describe by exemplary selector switch 54, this switch next sevtor address reorientating that be coupled to servo-drive system and startup transducer 15.As has been described, transducer move to comprise that the next one needs the session of image and when reproducing the image of needs data output signal 41 enabled by parts 43 valuably.

Transducer continues to follow the session that need sector of reproduction by back-end processing.According to from these sector data recovered, produce a pair of new next one and finish sevtor address and process I ²C transmits.These new addresses are received as before and are stored in parts 52 and 53.Yet, before being reset by parts 51 and detect, starting the transducer redirect at new end sevtor address, exemplary selector switch 54 is reset or opens, and prevents to start prematurely and catch new sevtor address.

The sevtor address that needs by the playback sevtor address that more do not postpone and prestrain, above-described convertor controls sequence of the present invention starts the transducer motion, like this, avoid catching the delay in the new play back bit stream, be convenient to strengthen trick mode operation.

As is well known,, determine mpeg image decoding order, therefore, follow the decoding sequence of forward mode operation by the coded image classification.Yet the situation that image begins and stops in image sequence that requires according to predetermined special play-back algorithm and the known video bit buffer by control mpeg image decoding order, can realize trick play operation valuably.Like this, for example Fig. 5 C calculate or as the bit stream search of Fig. 6 determine that the picture position allows to begin the storer start pointer random access image of orientation as requiring valuably in code detector 520 and the variable-length detecting device 531 among the buffer 60B of known video position during trick mode operation.Demonstration video bit buffer shown in Fig. 5 D comprises aforesaid image fragment.Beginning code detector memory pointer such as arrow SCD describe, and this arrow search example video bit buffer begins code with location MPEG.Yet in the 3rd sector of first P image, beginning code detector memory pointer SCD1 indication is from the detection of the beginning code of next but unwanted picture.Therefore, represent, to known byte memory location accurately, avoided the hang-up of unwanted picture and undesirable code translator by guiding beginning code memory pointer valuably as the arrow SCD2 of Fig. 5 D.

In the configuration of the useful special-effect mode of another kind, be eliminated input and output FIFOs, the first-in first-out register of beginning code detector (SCD) 520 and variable length decoder (VLD) 531 from the data that do not need of previous image.Signal 521/532 removing that Fig. 3 describes or the corresponding FIFO that resets are to remove the remaining data from previous decoded operation.The removing of this FIFO or flushing guarantee that SCD and VLD adopt the new data from demonstration position buffer 60B to begin next decoded operation, like this, have eliminated another source of the code translator maloperation that is produced by remaining past data.

Reverse operating at broadcasting speed requires to reproduce the B frame, and under another special-effect mode, according to the B picture order opposite memory buffer request adjacent with decoding, has simplified the optimization reverse mode operation valuably.By be provided with or the control store start pointer with the decoding of the specific image that enables the special-effect mode requirement, reverse valuably decoding order.In another special-effect mode optimization,,, during trick play operation, can simplify buffer memory size and control by skipping or do not read the image of video bit buffer valuably according to the addressing operation that requires as specific special play-back algorithm.By immediately or as special play-back algorithm special requirement, enable the decoding of a plurality of images valuably, can be during special play-back further optimizing memory size and control.The feature request that these that provide are useful control in earnest read/write function and between them synchronously.

In another kind of special-effect mode optimization, the code translator control ability that promotes audio video synchronization or lip synchronization by skip pictures decoding is in favourable increase and being utilized in range of control during the trick mode operation, skipped or do not decipher to allow between 2 and 6 selectable a plurality of images at least.By skipping the B image in each GOP, the trick play operation of the favourable promotion of this image manipulation under six times of broadcasting speeds.

Except for the control of the storer of trick mode operation and distribute require, by for example decipher I or P image in the cycle and writes decode results to show and/or simultaneously the operation basically of memory stores that optimization MPEG deciphers valuably in field.Suppose ability with the memory buffer do not used decoding category-B type image.This category-B type image decoding is referred to as aerial B frame (Bframes-on-the-fly) (BOF).In addition, by will decipher the field write store and simultaneously in the same storer interleave unit read display field, can strengthen trick play operation valuably.Display field can go up image separately from the time.This read-write operation of while basically can be finished in the cycle in display field.Yet the field of decoding needn't rewrite or read with display field conflicts.Owing to can decipher without memory buffer, the B image does not need this read-write operation that interweaves.

In a demonstration player with reverse trickplay decoding, bit stream or track buffer device 60A are used to store the mpeg video bit stream from the compression of medium recovery.The video bit buffer 60B of track buffer device 60A or compression can be used to promote the access of a plurality of independent mpeg images.The special play-back output signal of decoding must conform to the TV signal standards to allow the demonstration of normal TV receiver.The following Example explanation is used for the control sequence of the present invention of DVD player MPEG decoding.Fig. 7 is the chart of explanation the present invention configuration, is used for the reverse trickplay pattern of 3 times of broadcasting speeds of video machines (3 times).This example chart has the I image of expression mpeg encoded and the row of P image, and these images comprise image sets or GOP A, B, C and D.It is not the image that obtains from the film source that each GOP comprises 12.

In this exemplary special play-back sequence, adopting provides reverse order decoding and shows that the MPEG code translator of decoding video and the useful configuration of two frame buffers can realize reverse decoding easily.In this example, have only I image and P image decoded, therefore, have only them on chart, to list.Fig. 7 has illustrated the sequence of 37 coded images, presentation video numbering in the bracket.Right column is labeled as " output field # ", and expression is the time shaft of incremented with the field cycle.First field, the beginning of output field #1 mark trick play reproduction.Each row illustrates the present invention's processing that occurs in the respective field cycle in the chart.Abbreviation below having used among Fig. 7.Frame buffer is numbered as 1 and 2.Image/the frame of the top indication of upper case " D " expression decoding particular column.The process of deciphering image and storing its result is by " D＞1 " description, and this numeral destination frame buffer number is 1 here.Lower case " d " expression is from the demonstration of the field of the frame of particular column.Can select output field to preserve the output signal interleaved sequence.For continuous output field sequence is provided, the very clear every row of chart that requires comprises a field idsplay order " d ".

The sequence of Fig. 7 explanation begins at output field #1, and I image I (37) is decoded here is among the 60C with being stored in frame buffer 1.In decoding I image (37), show for example top field of I frame (37) of a field.Use useful code translator 530 so that the vision signal of decoding and demonstration decoding simultaneously.During output field #2, mpeg image I (25) retrieves from bit stream buffer 60B, deciphers and be stored in frame buffer 2 is among the 60D.Simultaneously, another field is that the bottom field of the I (37) that reads of 60C is shown from frame buffer 1 for example.

During the output field #3 cycle, the action that illustrates one aspect of the invention takes place.During field #3,, repeat the top field of exemplary I (37) by being that 60C reads from frame buffer 1.When the top field of the repetition of I (37) is read, with reference to the image P (28) of I (25) decoding prediction and to be stored in frame buffer 1 be among the 60C.With synchronization timing exactly, it is among the 60C that decoded frame P (28) is written into frame buffer 1.Read the back by image I (37) display field and sequentially deciphering image P (28) on the basis line by line, obtain this operation simultaneously.Calling over and writing of frame buffer 1 is another useful performance that this example decoder and storage management system provide.

When output field #3 finished, the image I of GOP C (25) and P (28) were stored in respectively among frame buffer 1 (60C) and 1 (60D).Yet, early incident and the frame that can decoding produces in proper order of requirement for example frame P (31) and P (34) on these frame express times.The intra-coded picture I (25) that resides in the storer 2 (60D) is used for decoded frame P (28), but does not currently need.Like this,, adopt frame I (37) to rewrite frame memory 2, read again and decoding from video display buffer 60B in order to be provided for the demonstration of output field #4.In order to keep exporting the interlaced fields sequence, the suitable field of frame I (37) is taken out from frame buffer 2 and is used for showing.At output field #5, repeat field #3 finish useful in handle.By read the field of image I (37) from frame buffer 2, obtain output field #5.Simultaneously, image P (31) deciphers with reference to the image P (28) from frame buffer 1, and decode results is stored in the buffer 2.Like this, the first five output field of these exemplary three times of reverse reproductions comprises the image of I image static or that freeze (37).Yet, when output field #5 finishes, adopt the image I (28) and P (31) beginning that are stored in frame buffer 1 and 2 respectively, produce the special play-back output signal.

At output field #6, the image P (34) of prediction is from bit stream buffer 60A or video bit buffer 60B reads, deciphers and the suitable field of not storage shows.Like this, field #6 starts the demonstration of 3 times of velocity reversal's motions.At output field #7, image P (34) is retrieved once more, is deciphered and the field of other selections is used to show.The image P (31) that had before deciphered and be stored in frame buffer 2 is read out and provides respectively output field #8 and #9.

When output field #9 finishes, no longer require memory image P (31), like this, the intra-coded picture I (13) of the GOP B of next front is obtained, decipher and be stored in the frame buffer 2.Output field #10 and #11 read from the frame buffer 1 that comprises predicted picture P (28).In character sensing strip #11, the image P (16) of prediction obtains, deciphers and sequentially be stored in the frame buffer 1 from bit stream buffer 60B.Because two frame buffers comprise the anchor frame of the GOP B of next front, output field #12 obtains in the output field #6 mode identical with #7 with #13.The prediction image P (25) from bit stream buffer 60B, read, decipher and under the situation of not storing suitable field be shown.

Like this, the GOP B that comprises the next front of image I (13), P (16), P (19) and P (22) as the processing that GOP C is described.

Claims (2)

1. one kind is used for comprising from the equipment of CD media reproduction digitally encoded signal:

Transducer is changed described digitally encoded signal;

Storer is coupled to described transducer, is used to store described digitally encoded signal;

Code translator responds described digitally encoded signal from wherein deciphering image;

The controller that is used for described code translator, wherein, in first operator scheme, controller is controlled described code translator to respond first sequence, read the digitally encoded signal of described storage from described storer, and in second operator scheme, described controller is controlled described code translator with second sequence, reads the digitally encoded signal of described storage from described storer.

2. one kind is used for comprising from the equipment of CD media reproduction digitally encoded signal:

Transducer is changed described digitally encoded signal;

Storer is coupled to described transducer, is used to store described digitally encoded signal;

Code translator responds described digitally encoded signal from wherein deciphering image;

The controller that is used for described code translator, wherein, in first operator scheme, controller is controlled described storer with according to first sequence, read the digitally encoded signal of described storage from described storer, and in second operator scheme, described controller is controlled described storer with second sequence, reads the digitally encoded signal of described storage from described storer.

Images