Background technology
The captions of stack are displayed on the bottom of display screen usually when spectators watch foreign film.In optic disk or common television program, captions just are recorded or broadcast after being superimposed on the television image.
In another system such as " leading (CAPTAIN) " system's one class, overlapping text is to transmit with the form of character code or dot pattern.
In CD-G (graph technology), can use the subcode recording geometry.Can also use these subcodes to write down the overlapping text data.
The following describes the data layout of CD-G.As shown in FIG. 13A, the data of a two field picture are to be made of the subcode of a byte and the data of 32 bytes.24 of these data bytes are distributed into 6 samples in each passage.Like this, each sample dispensing has 2 bytes in passage L and R.8 byte allocation of all the other of 32 data bytes are given error correcting code (ECC).
Shown in Figure 13 B, the subcode of totally 98 frame pictures is gathered to form a piece from 0 frame to 97 frames.Figure 13 C shows the details of this subcode piece.
Shown in Figure 13 C, the byte of each subcode of this piece is represented by this subcode being divided into eight passages (being called P, Q, R, S, T, U, V and W).The subcode of 0 frame and 1 frame is designated as the synchronous mode of S0 and S1, and various sub-code data item record is in the subcode of all the other 96 frames.Retrieve data is assigned to P and Q passage.Therefore, graph data can be distributed to all the other positions, 576 (6 * 96) by R to W passage.
In this case, the data of a piece are with the frequency transmission of 75Hz, and result, the sub-code data amount of a frame picture waiting for transmission are exactly the individual bytes in 7350 (75 * 98).Be that subcode transmission information rate is the 7.35K byte per second.
Figure 14 shows the transformat that is used for transmitting graphics data by 576 graph data positions in the piece are defined as a packet.
As shown in figure 14, data are made of 96 code elements.Each code element is determined by 6 bit data of each passage from all passages of R to W.Each packet comprises 4 son bags.Therefore, each son comprises 24 code elements, and promptly code element 0 to 23.
Pattern information is distributed in each son bag corresponding to the code element 0 of passage R, S, T three, and project information is distributed in each son bag corresponding to the code element 0 of passage U, V, W three.Following pattern has been determined in the various combinations of pattern information and project information:
Schema entry
000 000 0 patterns
001 000 graphic models
001 001 TV-graphic models
111 000 user models
Give code element 1 with command assignment.With parity information be not that the additional information of pattern, project or command information is distributed to code element 2 to 7.The parity information of the data of code element 0 to 19 is distributed to code element 20 to 23.Like this, substantial graph data can be distributed to 12 code elements from code element 8 to 19.
According to this transformat, graph data can be distributed to similarly the interior binary data of the individual pixel coverage in 72 (6 * 12) of each the son bag that is distributed in a CD-G.The unit transfer rate is exactly 75 (Hz) * 4 (unit), and this equals 300 son bags of per second.Therefore, when hypothesis was distributed to 72 pixels of each son bag with a character, the per second kind can be transmitted 300 characters.
Because a display screen has been determined in 288 (horizontal pixel) * 192 (OK) in a CD-G, so these characters of transmission need 2.56 seconds on a display screen.This time cycle is represented by following formula:
(288/6) * (192/12) ÷ 300=2.56 (second).
When adopting hexadecimal representation, each pixel needs 4 bit data to represent, the result, for one completely character graphics must transmit 4 times (one of each transmission cycle transmission).Therefore, the transmission time is 10.24 seconds, i.e. 4 2.56 second transmission time of multiply by the above.
Yet the captions of the stack in film for example show just as the lyrics in Karaoke (KARAOKE) system preferably with the preceding of video image and then variation.The tediously long transmission time of above display packing makes that changing overlapping text in the time of the abundance that captions advance with video image becomes impossible.
In the ordinary playing process of " dominating " system or CD-G, can be switched on or switched off the demonstration of overlapping text.Yet, also have such problem, promptly according to the resolution of above display packing captions for human eye vision not enough.
Stipulate in the system that " dominating " each fluoroscopic display area is 248 (horizontal pixel) * 192 (OK), and the resolution of synthetic digital television signal is 720 (horizontal pixel) * 480 (OK).Therefore, when comparing, " dominate that " resolution of system is not enough with synthetic digital television signal system.
In addition, when data are represented as binary data, in CD-G only with pixel of a bit representation.For example, this expression way can cause that the tilting zone that causes character forms the flicker of breach and character now as the existing picture of obscuring that becomes very dazzling.The appearance of these phenomenons makes spectators read captions and becomes very difficult and uncomfortable.
Binary picture can be converted to multistage frame, so that by utilizing wave filter for example to reduce the influence of these phenomenons.Yet, must use high-precision wave filter.But high-precision filter makes cost increase and cause the degradation of background image.Therefore, be difficult to think that it is rational using a kind of like this wave filter.
When in CD-G, using 16 systems to express for each pixel, show in order to change caption data, compare when using binary expression, need about four times time of cost.When using hexadecimal representation, this time delay make must high speed conversion overlapping text demonstration very difficult.
Summary of the invention
Therefore, an object of the present invention is to provide a kind of method and apparatus of passing the overlapping text data encoding/decoding that changes overlapping text in time.
A further object of the invention provides a kind of method and apparatus that is used to show the overlapping text data encoding/decoding of high-quality overlapping text.
According to one aspect of the present invention, a kind of captions code device that is used to have the video image transmission system of the incoming bit stream that comprises caption data and vedio data is provided, and described device comprises: the colored question blank device with the address that is used for the access video data; The piece generating apparatus of packing into, be used for by input stream sampling is comprised the piece of packing into of the information of shifting one's position by a frame period from the positional information generation of the relative position of the pixel data of represent video image, the described information of shifting one's position is used for upgrading and is positioned at shifting one's position of this video image, the current address that wherein is used to visit described colored question blank device all can change to next address in each frame period, make can be on the change location indicated by the described information of shifting one's position the different video data of access, the described information of shifting one's position also is used to the length that makes the updating delay of change location and be used to set described delay; Code device, the subtitle data encoding that is used for changing with described colored question blank device is to form pixel data blocks; And multiplexer, be used for forming for the data block that sends usefulness by described pack into piece and described pixel data blocks.
According to another aspect of the present invention, a kind of caption decoding device that is used for video image display systems is provided, described device comprises: the colored question blank device with the address that is used for the access video data; Receiving trap, be used to receive the data block that comprises pack into piece and pixel data blocks, described pixel data blocks is represented caption data, and the described piece of packing into comprises the information of shifting one's position, the described information of shifting one's position is used for upgrading the change location that is positioned at video image, the current address that wherein is used to visit described colored question blank device can change to next address in each frame period, make can be on the change location indicated by the described information of shifting one's position the different video data of access, the described information of shifting one's position also is used to postpone to the renewal of change location and is further used for setting the length of described delay; The information detector of shifting one's position is used for detecting the described information of shifting one's position according to the described piece of packing into; And control device, be used on display screen showing the described caption data and the video image that become with described shift one's position information and described colored question blank device.
According to another aspect of the present invention, a kind of subtitle data encoding device that is used to have the video image transmission system of the incoming bit stream that comprises caption data is provided, described device comprises: a colored question blank device; The module generator of packing into, be used for by the incoming bit stream sampling being generated the piece of packing into that comprises colorwiping information from the positional information of representing video image pixel data relative position at a frame period, described colorwiping information is used for colorwiping is moved forward, each frame period of wherein said colorwiping can move forward, so that according to described colorwiping information the color of caption data is gradually changed, the described information of shifting one's position also is used to the length that postpones colorwiping and be further used for setting described delay; Code device is used for subtitle data encoding so that form the pixel data blocks that changes with described colored question blank device; And multiplexer, be used for the data block that is formed for transmitting by described pack into piece and described pixel data blocks.
According to another aspect of the present invention, a caption decoding device that is used for video display system is provided, this device comprises: a colored question blank device; Receiving trap, be used to receive the data block that comprises pack into piece and pixel data blocks, the described piece group of packing into comprises colorwiping information, described colorwiping information is used to make described colorwiping to move forward, wherein said colorwiping can move forward in each frame period, so that the color of caption data is gradually changed, the length that the described information of shifting one's position also is used to postpone described colorwiping and is further used for setting described delay, and wherein said pixel data blocks is represented caption data; The colorwiping information detector, be used for detecting described colorwiping information according to the described piece of packing into, and control device, be used on display screen, showing described colorwiping and the caption data that changes with described colorwiping information and described colored question blank device.
According to another aspect of the present invention, a kind of captions Code And Decode device that is used to have the transmission of video images and the display system of the incoming bit stream that comprises caption data and vedio data is provided, and this device comprises: the colored question blank device with the address that is used for the access video data; The module generator of packing into, be used for basis by the incoming bit stream sampling being generated the piece of packing into that comprises the information of shifting one's position from the positional information of the relative position of expression video image pixel data at a frame period, the described information of shifting one's position is used to upgrade the change location that is arranged in video image, the current address that wherein is used to visit described colored question blank device can change to next address in each frame period, make that according to by the indicated different video data of change location access of the described information of shifting one's position, the described information of shifting one's position also is used to postpone to the renewal of change location and the length that is further used for setting described delay; Code device is used for the subtitle data encoding that changes with described colored question blank device so that form pixel data blocks; Multiplexer is used for being formed for data block transmitted by described piece and the described pixel data blocks of packing into; Receiving trap is used to receive the device of described data block; The information detector of shifting one's position is used for detecting the described information of shifting one's position according to the piece of packing into that is received; And control device, be used on display screen showing the caption data and the video image that change with described shift one's position information and described colored question blank device.
According to another aspect of the present invention, a kind of subtitle data encoding and decoding device that is used to have the transmission of video images and the display system of the incoming bit stream that comprises caption data is provided, described device comprises: a colored question blank device; The piece generating apparatus of packing into, be used for by the incoming bit stream sampling is comprised the piece of packing into of colorwiping information at a frame period from the positional information generation of the relative position of expression video image pixel data, described colorwiping information is used for colorwiping is moved forward, each frame period of wherein said colorwiping can move forward, so that according to described colorwiping information the caption data color is gradually changed, the described information of shifting one's position also is used to the length that postpones colorwiping and be further used for setting described delay; Code device is used for subtitle data encoding, to form the pixel data blocks that changes with described colored question blank device; Multiplexer is used for the data block that is formed for presenting by described pack into piece and described pixel data blocks; Receiving trap is used to receive described data block; The colorwiping information detector is used for detecting described colorwiping information according to the received piece of packing into; And control device, be used on display screen, showing described colorwiping and the caption data that changes with described colorwiping information and described colored question blank device.
According to another aspect of the present invention, a kind of captions coding method that incoming bit stream therein comprises the video image transmission system of caption data and vedio data that is used for is provided, this method comprises the steps: by at a frame period incoming bit stream sampling being generated the piece of packing into that comprises the information of shifting one's position in the positional information of the relative position of pixel data from the expression video image, the described information of shifting one's position is used for upgrading the change location that is positioned at video image, this generation step is characterised in that, the current address that is used to visit the colored question blank device with the address that is used for the access video data can change to next address in each frame period, make the different video data of access on the change location indicated, the described information of shifting one's position also be used to postpone and be further used for setting the time span of described delay the renewal of this change location by the described information of shifting one's position; To the subtitle data encoding that changes with described colored question blank device, forming pixel data blocks, and be formed for data block transmitted by described message block and the described pixel data blocks of packing into.
According to another aspect of the present invention, a kind of coding/decoding method that is used for video image display systems is provided, described method comprises the steps: to receive the data message piece that comprises pack into piece and pixel data blocks, described pixel data blocks is represented caption data, and the described piece of packing into comprises the information of shifting one's position, the described information of shifting one's position is used for upgrading the change location that is positioned at video image, wherein, the current address that is used to visit the colored question blank device with the address that is used for the access video data can change to next address in each frame period, make the different video data of access on the change location indicated, the length that the described information of shifting one's position also is used to postpone the renewal of this change location and is further used for setting described delay by the described information of shifting one's position; From the described piece of packing into, detect the described information of shifting one's position; And on display screen, show caption data and the video image that changes with described shift one's position information and described colored question blank device.
According to another aspect of the present invention, a kind of subtitle data encoding method that incoming bit stream therein comprises the video image transmission system of caption data that is used for is provided, described method comprises the steps: by generate the piece of packing into that comprises colorwiping information in the positional information of a frame period to incoming bit stream sampling relative position of pixel data from the expression video image, described colorwiping information is used to move forward colorwiping, wherein said colorwiping can move forward in each frame period, according to described colorwiping information the color of caption data being gradually changed, the described information of shifting one's position also is used to the length that postpones colorwiping and be further used for setting described delay; To subtitle data encoding, to form the pixel data blocks that changes with colored question blank device; And the data message piece that is formed for transmitting by described pack into piece and described pixel data blocks.
According to another aspect of the present invention, a kind of caption data coding/decoding method that is used for video display system is provided, described method comprises the steps: to receive the data block that comprises pack into piece and pixel data blocks, the described piece of packing into comprises colorwiping information, described colorwiping information is used to move forward described colorwiping, but each frame period of wherein said colorwiping moves forward, so that the color of caption data is gradually changed, the described information of shifting one's position also is used to postpone colorwiping, and being further used for setting the length of described delay, described pixel data blocks is represented caption data; Detect described colorwiping according to the described piece of packing into; And demonstration changes with described colorwiping information and colored question blank device on display screen described colorwiping and described caption data.
According to another aspect of the present invention, provide a kind of and be used for wherein that incoming bit stream comprises the transmission of video images of caption data and vedio data and the captions Code And Decode method of display system, described method comprises the steps: by at a frame period incoming bit stream sampling being generated the piece of packing into that comprises the information of shifting one's position in the positional information of the relative position of pixel data from represent video image, the described information of shifting one's position is used for upgrading the change location that is positioned at video image, the current address that wherein is used to visit the colored question blank device with the address that is used for the access video data can change to next address in each frame period, make the different video data of access on the change location indicated, the described information of shifting one's position also be used to postpone and be further used for setting the length of described delay the renewal of change location by the described information of shifting one's position; To the subtitle data encoding that changes with described colored question blank device, to form pixel data blocks; Be formed for data block transmitted by described piece and the described pixel data blocks of packing into; Receive described data block; Detect the described information of shifting one's position according to the piece of packing into that is received; And on display screen, show described caption data and the video image that changes with described shift one's position information and described colored question blank device.
According to another aspect of the present invention, a kind of subtitle data encoding and coding/decoding method that incoming bit stream wherein comprises the transmission of video images and the display system of caption data that be used for is provided, described method comprises the steps: by at a frame period incoming bit stream sampling being generated the piece of packing into that comprises colorwiping information in the positional information of the relative position of pixel data from represent video image, described colorwiping information is used to move forward colorwiping, but each frame period of wherein said colorwiping moves forward, according to described colorwiping information the color of caption data being gradually changed, the described information of shifting one's position also is used to the length that postpones colorwiping and be further used for setting described delay; To subtitle data encoding, to form the pixel data blocks that changes with colored question blank device; Be formed for data block transmitted by described piece and the described pixel data blocks of packing into; Receive described data block; Detect described colorwiping information by the piece of packing into that is received, and on display screen, show described colorwiping and the described caption data that changes with described colorwiping information and described colored question blank device.
According to the present invention, because can be used as the function certificate that is stored in the information of shifting one's position in the piece of packing into, the overlapping text data change as time goes by, so the demonstration of caption data (for example display color) can change on request.Further, when pattern conversion is designated as the conversion display color, can as karaoke OK system, finish the colorwiping of the word that once superposes.
Because superposition of data can be shown by a system (for example NTSC/PAL system etc.), so can show high-quality overlapping text.
Embodiment
Fig. 1 shows the calcspar that is used to implement according to an embodiment of the data deciphering device of overlapping text data decoding method of the present invention.
In Fig. 1, a kind of reproducing signal, for example the multiway transmission signal of reading from a kind of data carrier (for example dish) by a servomechanism (not shown) is transfused to a data demoder and demultiplexer 1.Error correcting code (ECC) decoding so that correct a mistake, is separated the multi-path transmission data of importing, so that form video data, caption data and audio data in data decoder and demultiplexer.Video data is offered Video Decoder 3, caption data is offered demoder 7, and audio data is offered audio decoder 11.
Storer 2 is used as memory buffer, workspace and so on so that finish this processing in data decoder and demultiplexer.
Video Decoder also offers mailbox 5 with decoded video data by utilizing storer 4 with the bitstream decoding of the video data of input.
When the output signal of Video Decoder was in compact model, Filtering Processing was finished in mailbox, so that show in vertical direction and be reduced to 3/4 image, made to be that display image is observed on 100% ground on 4: 3 the display screen in the ratio of width to height.In this case, the timing that is equivalent to 1/4 (field) is regulated.Storer 6 is used to carry out timing and regulates.
Mailbox has a data path that is used for directly exporting the vision signal of compact model.
Audio decoder 11 is decoded voice data as memory buffer or like by utilizing storer 12.Decoded voice data is reproduced and be output as a simulated audio signal by means of a digital audio/simulation (D/A) converter 13.
Subtitle decoder 7 will be added to the bitstream decoding of the caption data of wherein supplying with and with the caption data after decoded on the video data of exporting from mailbox 5.
The vision signal that is applied caption data is converted to NTSC, PAL or SECAM-system by a composite coding device 8, then is output and is converted into analog video signal so that show by a D/A converter 10.
The processing of each part discussed above is all by 14 controls of a system controller.Mode display 9 is provided for monitoring user instruction and various other information, makes such information may be displayed on the display unit that is equipped with in this mode display and/or can be superimposed upon on the vision signal.
Subtitle decoder 7 will have common reproducing caption data and therein by the caption data bitstream decoding of multiplex special reproduction caption data.Therefore, subtitle decoder receives the bit stream of multipath transmission caption data and decoded video data and multiplex caption data is being decoded with the sequential of appointment after these data storage are in a code buffer (for example code buffer among Fig. 2 22).Then subtitle decoder can be superimposed upon caption data on the decoded video data according to reproduction mode.
A calcspar of the preferred embodiment of subtitle decoder is shown among Fig. 2.Each shown in this width of cloth figure part all will be illustrated hereinafter.
(1) the word detecting device 20
Bit stream by the caption data of data decoder and demultiplexer 1 output is transfused to word detecting device 20.This word detecting device detects heading message, header error information and error in data information and gives controller 35 with this information transmission.The more new data and the bit map pixel data of the markers (PTSS) of captions demonstration time, display position information (position data), colored question blank (CLUT) are detected, transmit and are stored in the code buffer 22 by this word detecting device.
(2) scheduler 21
The read/write access control of code buffer is finished by scheduler 21.The bandwidth of the read/write access of code buffer is to be determined by the scheduling controlling of storage access, and the scheduling controlling of sort memory access itself depends on the data rate of the data of sending out from data decoder and demultiplexer and shows speed.
For example, when the maximum data rate from the transmission of data decoder and demultiplexer was 20Mbps, these data write code buffer 22 with the speed of 2.5MHz at least when the I/O of this code buffer port is 8 bit parallel mouths.
Data be as from system controller 14 receives decode start signal after with title at caption data multiplex display position information combined, read from code buffer from the function of the sequential of vertical synchronizing signal and horizontal-drive signal derivation.Read operation speed equals the pixel sampling speed of 13.5MHz.For example, if the read/write operation in code buffer changes with sampling clock, as mentioned above, be necessary owing to be at least the speed of 2.5MHz, so the minimum possible speed that satisfies this write operation is 3.375MHz.3.375MHz speed equal 13.5MHz pixel sampling speed 1/4.
In other words, the sequential of 3.375MHz is used for data are write code buffer, and all the other some sequential of pixel sampling speed are used for from the code buffer sense data.Like this, have only designated being used in clock stage to write the stage from four clocks of 13.5MHz clock, and read all the other 3 designated being used in clock stage.
Because the I/O port is 8, so the figure place that can be read by 3 clock stages is positions, 24 (3 * 8).This demonstration of 24 can be finished in real time owing to 6 (6=24 ÷ 4) position can be appointed as stages in these 4 clock stages, so when the data of a pixel by 6 or be less than 6 and any interruption can not occur when constituting.
(3) controller 35
Controller 35 receives the markers (PTSS) of captions demonstration and it is exported to system controller 14 from word detecting device 20.After sending decode start signal by system controller 14, controller 35 starts the decoding of caption data.
When setting common reproduction mode,, just can from code buffer 22, repeatedly read common reproducing caption data in frame unit as long as the byte of repetition time is alignd and caption data is decoded.Repetition time both can be that the frame number of being scheduled to also can be the predetermined lasting time that shows identical caption data.When the repetition time was represented frame number, its value reduced as the function of the subtract pulse that is sent by system controller or the pulse of successively decreasing.The pulse of successively decreasing is sent from system controller according to frame rate in common reproduction process.When the recovery time is zero, just show other caption data.Controller 35 receives this pulse and for the address administration of scheduler 21 run time version impact dampers of successively decreasing, make according to show time scale between video image and captions, set up correct synchronously.Same analysis is applicable to when the repetition time is represented a period of time, yet when the pulse of successively decreasing will reduce to zero the repetition time, the demonstration time tapered to every next frame period.
When the distinctive signal that sends from system controller 14 was identified as " unusually ", controller 35 sent back to this system controller according to the differentiation to this special reproduction pattern with a confirmation signal, indicates this special reproduction pattern and is correctly received already.When this special reproduction pattern was adjusted to n and doubly presents (FF) or n fast and doubly oppositely present (FR) fast, the pulse of successively decreasing was to send with n times of speed (rather than with above-mentioned frame rate).When such FF or FR special reproduction pattern are set, special reproducing caption data with regard to chosen, from code buffer 22, read and decoded.In this case, the repetition time is reduced according to a corresponding n times speed.If this special pattern is set time-out, just do not send the pulse of successively decreasing, same frame just is repeated decoding.Like this, image and captions keep showing synchronously.
Caption data is read out from code buffer and carries out length-changeable decoding in a reverse variable length code (VLC) circuit 23.Detect in an anti-phase stroke (run-length) circuit 24 " EOP (EOP) ", this circuit is also finished the run-length coding of caption data is handled.When the count value of EOP equaled the repetition time, anti-phase stroke circuit sent one and shows that end mark is to controller 35.Respond this demonstration end mark, controller stops to read from code buffer 22 read operation of caption data.EOP is used to check whether caption data was once correctly decoded, because can detect EOP when the bit map of each this caption data of processing.
When word detecting device 20 just detected the EOP of following one page before controller receives the demonstration end mark, controller just sent an impact damper spill over and gives system controller 14, to stop from the transmission of the data of data decoder and demultiplexer 1.
When sending an instruction from system controller, the demonstration starting position of each frame just is updated.
(4) code buffer 22
Code buffer 22 preferably constitute by RAM (random access memory) and no matter this impact damper be from the outside or provide in inside must have the capacity of storing 2 page datas at least.Because most of jumbo high speed access storage devices have only a port that is used to read and write data for every page,, and write another page or leaf by its port simultaneously so this capacity of two pages allows the port reads one page by it.Like this, can avoid the conflict of read/write.Select outside still inner code buffer is a kind of design alternative actually, and it changes with the desired structure of decode system.RAM is best, and this is because it contains the ability of the delay in the compensation video data decoding process, and has satisfied the additional bandwidth with scheduler 21 visits.
In order to compensate the delay in the decode procedure, when caption data is written into code buffer, controller 35 will show that the time mark (PTSS) of time sends to system controller 14.
System controller sends the decoding enabled instruction to the controller 35 in the subtitle decoder 7 after receiving markers.The timing considerations of decoding enabled instruction is to synchronous clock and the processing delay of the mailbox 5 in the PTSS coupling and the delay (about 1) of Video Decoder from system controller.
Because multichannel is compound is to be to finish in the digital coding unit under 0 such condition precedent in the delay between video data, voice data and the caption data, so will consider the decoding delay that these are a series of.
(5) reverse VLC circuit 23
Oppositely VLC circuit 23 finish to the caption data of reading from code buffer carry out length-changeable decoding and with the data of decoding as paired level data and run-length data output.This anti-phase VLC circuit is not realized its function under some state.
(6) the reverse stroke circuit 24
Reverse stroke circuit 24 is finished run-length coding by the level data that generation equals the run-length data amount.Oppositely VLC circuit and reverse stroke circuit have been expanded compressed caption data.
(7) 3: 4 wave filters 25
When the ratio of width to height of monitor is 4: 3,3: 4 wave filters 25 will be made the caption data of level compression and be handled, and its circularity (roundness) is increased to 100%.After caption data is modified, it is superimposed upon on the video data.In this case, controller 35 for example, is used from the H-synchronizing pulse to be equivalent to 90 times that pixel is so fast, sense data from code buffer 22 according to the H synchronizing pulse.
When the ratio of width to height of monitor was 16: 9, signal was walked around 3: 4 wave filters.The X compression signal that is provided by controller 35 is used to determine whether this signal walks around 3: 4 wave filters.
When the bit stream of a plurality of caption datas that use multiple font was transmitted, this bit stream was walked around 3: 4 wave filters.
(8) colored question blank (CLUT) circuit 26
Brightness data Y, chromatism data Cr, Cb and indication background data and the brightness data Y, critical data (key data) K of the mixture ratio between chromatism data Cr, the Cb that select are stored in the colored question blank (CLUT), for example as shown in Figure 6.These data item are defined as 8 by full scale, but as shown in Figure 6, they can be converted into 4 precision.Such CLUT can be loaded on the CLUT circuit 26 from code buffer 22 in advance.
Critical data K gives mixer 34 as mixture ratio from the CLUT circuit transmission.Also may provide a CLUT as shown in figure 12 to be used to the colorwiping that realizing passs in time changes by the highest significant position that utilizes Input Address.
(9) mixer 34
When the on/off superposed signal is " leading to ", mixer 34 will be added to as the caption data that Y, Cb, Cr data are read from the CLUT circuit according to mixture ratio K on the video data as the input of Y, Cb, Cr data.These data item are added on the preposition according to position signalling or according to the customer location that is provided by controller 35 (U position) signal.
When pattern information was designated as attenuation coefficient, the graph data of caption data multiplied each other with speed and this attenuation coefficient of appointment, so that captions are faded in or fade out.
When superposed signal is " breaking ", only be input to the vision signal of mixer so that show.The user can freely set superposed signal.
Offer by controller 35 system controller 14 various data item content and be shown among Fig. 2,3 and 4 by the content that this system controller offers the various data item of this controller.Fig. 4 has also illustrated other circuit of input or the various data item from wherein exporting in above-described decode system.
Secondly, will an example of by 4 coding modes caption data being encoded be described with reference to figure 5A-5C and Fig. 6 in digital coding of the present invention unit.
Shown in Fig. 5 B and 5C difference, caption data is represented as padding data (fill data), and mixture ratio is represented as critical data.Shown in Fig. 5 A, in this example, alphabetical A will be used as caption data and show.Utilizing this alphabetical resulting padding data of single horizontal scanning line scanning shown in Fig. 5 B.
This padding data is set to a grade corresponding to the brightness signal Y of the letter that will show in period T 3.At the period T before the period T 31, T2 and the period T after period T 34, T5, this padding data is set to lowermost level " 0 ".The grade of padding data has been controlled the intensity level of the caption data that is shown.When padding data when " 0 " changes to " EO ", padding data makes intensity level change to very bright level by black level.
Critical data is set to its lowermost level " EO " in period T 3, so that show captions.Critical data is set to its superlative degree " 0 " among period T 1 that separates in period T 3 front and back and the T5.When critical data is set to " 0 " level, because the mixture ratio of background video signal is 100%, so caption signal is decayed to greatest extent.Yet when critical data was set to " EO " level, the mixture ratio of background signal was 0%, and captions just can be clear that by spectators.
Period T 2 and the T4 of critical data between period T 3 and period T 1 and T5 is set to intergrade respectively.In other words, little by little become lowermost level " EO " in period T 2 critical data grades, and little by little become the superlative degree " 0 " by lowermost level " EO " in period T 4 critical data grades by highest " 0 ".
At period T 2 and T4, background video signal is worth corresponding speed decay according to critical data.In this example, the value of critical data is big more, and the rate of decay of background video signal is fast more.Therefore, the value of critical data is more little, and the rate of decay of background video signal is just slow more.
As above illustrated, background video signal is suppressed near letter or character gradually, can more clearly read captions when making the video image of spectators on watching screen.
For example, colored question blank (CLUT) shown in Figure 6 is the table that is used as 4 coding mode forms of benchmark when coding.
When the address when 0H to 7H changes, critical data K by with 00H → 20H → 40H → 60H → ... → EOH → 8 stages storages, and padding data (brightness data Y) is defined as 00H.When the address changed from 8H to FH, critical data was defined as EO, and padding data by with 00H → 20H → 40H → 60H → ... 8 stages storage of → EOH.In each case, chromatism data Cr, Cb are defined as 7FH.
When in cataloged procedure, being benchmark with colored question blank, from sample circuit 64, to export as the data that are encoded corresponding to the address of the data of each sampling timing shown in Fig. 5 B and the 5C, it is described as follows.
Fig. 7 A and 7B are the block schemes that has used the coded system of subtitle data encoding of the present invention unit therein.
In this coded system, be supplied to the video encoding unit 52 that is used for mould/number (A/D) conversion from the vision signal of video camera 51 output, and then before supply multichannel recombiner 58, compressed and divide into groups.
The vision signal of reappearing can also be supplied with use and replace the video disk player of video camera or the video encoding unit of video tape recorder.
Video encoding unit is equipped with rate controller 52a, makes that speed is by 57 controls of subtitle data encoding unit, so that increase the video signal data amount that is encoded when low volume data is encoded.When mass data is encoded, speed is controlled, so that reduce the video signal data amount that is encoded.Like this, speed is controlled, and is invariable with the total amount of the data that keep being encoded.
The sound signal of being gathered by a microphone 53 is transfused to audio coding unit 54.This signal carried out A/D conversion, compression, coding and packet transaction before supplying with multichannel recombiner 58.The sound signal of being reproduced by blattnerphone or similar device (replacement microphone) also can offer audio coding unit.
Be provided for subtitle data encoding unit 57 by the caption data of character generator 55 generations or the caption data of exporting by flying-spot scanner 56.In this subtitle data encoding unit, with colored question blank 71 be benchmark to subtitle data encoding, a such example is shown among Fig. 6.Before supplying with described multichannel recombiner, this caption data is compressed, encodes and divides into groups.
It is compound that the data through grouping that the multichannel recombiner will be provided by video data encoding unit 52, audio data coding unit 54 and subtitle data encoding unit are respectively carried out multichannel.Modulated then through the data based ECC that multichannel is compound by error correction, for example utilize the EFM (8 to 14 modulation) in the multichannel recombiner.Then, will be recorded in that a kind of recording medium (for example CD 91) is gone up or by a channel transfer or broadcast through the compound data of multichannel.
Below with telltale title digital coding unit 57.
Character generator 55 produces the caption data corresponding to the video image of being encoded by video encoding unit 52.This caption data is transfused to the subtitle data encoding unit.This caption data is provided for the contact (a) of switch 61, and is provided for the contact (b) of switch 61 by the critical data that character generator produces.This switch is transformed into contact (b) according to predetermined sequential by contact (a), so that select to wait to supply with the caption data or the critical data of sample circuit 64.Selected data are offered this sample circuit through wave filter 72 and switch 62.
The caption data that is transfused to sample circuit is that benchmark is quantized in this quantifying unit with colored question blank (CLUT) 71, then is encoded in DPCM circuit according to a differential pulse coding modulation code.After this, in stroke coding circuit 66 and variable length code circuit 67, utilize the skew of the code frequency of occurrences to finish compressed encoding.
When caption data is expressed with 8 grades, be that benchmark has obtained 4 bit address corresponding to the brightness value Y and the critical data K of captions with CLUT71, this example is shown among Fig. 6.Also may send colored question blank to decoding device, make that colored question blank is identical in the Code And Decode process.Under the sort of situation, the colored question blank of being made by the module generator 70 of packing into is by multipath transmission, then is recorded or sends decoding circuit to by captions buffering verifiers (SBV) 68.
In this decoding circuit, the colored question blank that is transmitted is transfused to CLUT circuit 26, like this, the data that are transfused to can with the coding once used same colored question blank be that benchmark is decoded.In CLUT71, with the critical data K of the mixing ratio of brightness data Y, chromatism data Cr, Cb and indication and background image respectively by for example maximum 8 stored, as shown in Figure 6.Described brightness data Y, chromatism data Cr, Cb and critical data K form video data.
In coded system, video data, voice data and caption data divide into groups respectively and to carry out multichannel by the multichannel recombiner compound.Each packet all has the additional header data of the attribute information that comprises data, is read out in the decode operation of these additional header data in data decoder and separates.The title of caption data comprises the information of the data of indicating the data of preparing common reproduction mode decoding or preparing to decode by the special reproduction mode.
In subtitle data encoding unit 57, various control signals (common/special play-back, positional information, captions coded message, time code, EOP, higher limit etc.) are added on the bit map of subtitle graphic data by means of captions buffering verifiers (SBV) 68.Data volume at this impact damper to be stored is verified in SBV, so that regulate the data volume that is encoded, this can prevent the overflow or the underflow of data.This control is to finish according to the SBV of the data volume in this impact damper to be stored by the quantized level of control rate controller 52a and sample circuit 64.
Quantized level can change with plurality of sections.The quantization speed of sample circuit also can be regulated according to the quantized level width.Therefore, quantification can be finished with required bit rate.
With reference now to Fig. 8 A and 8B to Figure 12, characteristic of the present invention is described.
Fig. 8 A and 8B illustrate one by pack into a piece and the captions video data piece that pixel data blocks constitutes.Shown in Fig. 8 B, the first three rows of captions video data piece is distributed to the piece of packing under 4 coding modes.Yet the first six row is distributed to the piece (not shown) of packing under 2 coding modes.
Row 4 to 33 is distributed to 4 pixel data blocks under the coding mode.And row 7 to 33 is distributed to 2 pixel data blocks under coding one pattern.Pixel data blocks comprises the bit map data of the caption data that preparation shows on screen.
Pack into piece by following everybody constitute: 10 frame count value of advanced potential (P position) that is used to upgrade the change location of colored question blank, frame data that holding position (H position) that is used for the renewal of the change location of control next frame when identical CLUT information converting continues multiframe is identical with the maintenance of an indication information converting.
Described information converting P position, H position and frame count value bit representation.This information is used for carrying out the colorwiping operation or the caption data of change on the video image precalculated position that is called as change location (change position) shows.
The piece of packing into is formed according to the data of wiping by the piece generating apparatus 70 of packing in the captions coding unit 57 shown in Fig. 7 A and the 7B.
Import the module generator of packing into by the positional information that the control lever 81 of wiping obtains via an adapter 82.The module generator of packing into makes form shown in Figure 9 at the interval of a frame to the data stream sampling of input and with this information.
The information of same of wiping is mixed in a switch 83 by the caption information that character generator 55 sends, and then is exported to a monitor 84 that is used for tentative demonstration, to check this demonstration of wiping.
Fig. 9 is illustrated in the sample content of the data in the piece of packing under 4 coding modes.The one MSB (highest significant position) the expression P position of row 1, and its 2nd MSB (inferior high significance bit) expression H position.The 3rd MSB of row 1 and LSB (least significant bit (LSB)) represent the 10th and the 9th of frame count value respectively.Frame count value is represented the duration that captions show.The 3rd MSB of row 1 and LSB are a MSB and the 2nd MSB positions of frame count value.
The 8th to the 5th and the 4th to first of frame count value of the corresponding four bit representation frame count value of row 2 and row 3.
The colorwiping operation is illustrated with reference to Figure 10. ".On Z-axis, indicate the expression frame position time t0, t1, t3 ... t9, and transverse axis is represented horizontal level.
Below Figure 10 is illustrated.
Originally caption data shows at time t0.Because the H position is set to " 1 ", and frame count value equals 288, after showing subtitle graphic, and colorwiping advance a pixel and postpone about 10 seconds a period of time, i.e. 288 frames.
At time t1 place, (this time was set to time t0 about 10 seconds afterwards), the delay that begins at time t0 finishes.
At time t2, the colorwiping pixel of advancing.Because the H position is set to " 0 ", so do not postpone at time t2.Be set to " 1 " according to the P position, the horizontal level of colorwiping during by time t2 advances to next location of pixels.
At time t3, the colorwiping pixel of advancing again.
At time t4, colorwiping four pixels of advancing.At this moment, based on H position and frame count value, colorwiping keeps 16 frames, about 0.5 second.When the P position was set to " 0 ", this location of pixels was skipped, and therefore, at time t4, colorwiping skips that three pixels are advanced and based on the P position pixel of advancing.
Colorwiping at time t5,16 frames postpones to finish.
At time t6, based on H position and frame count value, an advance pixel and kept 16 frames once more about 0.5 second of colorwiping.
At time t7 place, the colorwiping that begins at time t6 postpones to finish.
At time t8, colorwiping 3 pixels of advancing.Colorwiping postpones about 1 second again, 32 frames.
At time t9 place, the colorwiping that begins at time t8 postpones to finish.
" display color change from left to right.The carrying out and the delay of colorwiping operation are bases, and for example, the information in the piece of packing into is controlled, as shown in Figure 9.
Figure 11 illustrates and be used to realize that the caption data of above operation shows and the colorwiping control circuit.
In Figure 11, the data that are made of pack into piece and pixel data blocks are transfused to register 200 and detect in P position and H position.
Remove pixel counter 204 with horizontal-drive signal (Hsync), and with pack into the number of pixel on the column address calculated level direction of piece of clock CK basis.When the P position is " 1 ", this count value content with register 205 in comparer (COMP) 206 is compared the position of the last frame that this content indication color is wiped.(during A>B), the mark of an indication A>B is input to register 205 via AND gate 207 as gating signal greater than the value B of register 205 as the count value A of pixel counter 204.This gating signal allows the content of register 205 is upgraded.Like this, a new colorwiping address (position) is stored in the register 205.
Controller 35 in subtitle decoder 7 packing into piece from 1 walk to 3 the row times in an enabling signal is offered register 201,205 and 209.
Register 205 is removed and shown initial subtitle graphic according to Displaying timer PTS such timing that conforms to the clock value SCR of this system.
An output signal of AND gate is imported simultaneously the set end S and the register 205 of set-reset flip-floop 203.The signal of input set end makes set-reset flip-floop set.When set-reset flip-floop is output as " 0 ", can upgrade register 205.When the output of set-reset flip-floop 203 is set when " 1 ", register 201 and 205 is under an embargo, and can not upgrade.
4 bit data are gone continuously by register 200 input registers 201 by each, import the triplex rows of the piece of packing into by register 200 at most.When packing block end into, as shown in Figure 9,10 bit data of indication frame count value are stored in the register 201.The block end in case pack into, register 201 is added to frame count value on the frame down counter 202.
When frame count value was set, the vertical synchronizing signal (Vsymc) that frame down counter 202 is used to from next frame was finished countdown.Value at the frame down counter became before 0, and set-reset flip-floop remains on " 1 ", so the content of register 201 and 205 can not be upgraded.When the count value of frame down counter was 0, the borrow output terminal (BR) of frame down counter became " 1 ", and set-reset flip-floop is reset to " 0 ", and like this, register 201 and 205 content just can be upgraded.When frame count value becomes 0, indicate delay or hold period and finish, like this, colorwiping advances and its horizontal level is updated.
When next frame count value in register 201 is set and repeats similar operations, one period of determining by frame count value of the operating delay of colorwiping.As mentioned above, the position of colorwiping is updated in register 205, and the colorwiping of captions advances as a result.
The output signal of register 205 makes can the colored question blank of conversion on the horizontal direction of associated frame.The value of each frame unit's register 205 can be upgraded.The output of register 205 and horizontal-drive signal synchronously are added on the pixel counter 208.When Counter Value became 0, pixel counter 208 was with its borrow output terminal (BR) output " 1 ".This output signal " 1 " offers a colored question blank (CLUT) 210 shown in Figure 12 as most significant address bit (MSB), and its time is arranged in via register 209 provides pixel data blocks such moment.
As shown in figure 12, different colour difference signal Cb, Cr are stored among the CLUT210 according to MSB.Like this, can be according to the chromatic information of the colour difference signal change of in CLUT, storing as the demonstration caption data of the function of CLUT210.
When pixel data is set to 4/voxel model, because MSB is used to control the conversion of CLUT 210, so only utilize 3 lower positions to finish coding.When pixel data is set to 2/voxel model, owing to use MSB like that as mentioned above, so only use LSB to finish coding.When under 2/voxel model, being cell encoding, among four of being transmitted, may having only one can come conversion CLUT 210, and can use remaining three picture encoded as MSB with two pixels.
As mentioned above, because padding data that quantizes and critical data are to be made of predetermined figure place, so the present invention has realized the high-speed demonstration of high-quality captions is not caused the reduction of the display quality of background video image in a simple circuit.
Because colored question blank in the horizontal direction can conversion, so also can be implemented in the colorwiping in the karaoke OK system.This single memorizer control circuit can be used for realizing being used for the colorwiping of control card karaoke system and show a kind of short-cut method of captions in foreign film.
Particularly, according to the present invention, all must send the piece of packing into for each data block that comprises about 1K byte.When finishing colorwiping or prepare to use surpassing the storage access scope of 4 conventional/pixel or 2/pixel coverage, needn't send additional information data.
Although already preferred embodiment of the present invention had been done to set forth and explanation, apparent, can make various modifications under the premise without departing from the spirit and scope of the present invention to those skilled in the art.