CN100591125C

CN100591125C - System and method for video data compression

Info

Publication number: CN100591125C
Application number: CN200710112219A
Authority: CN
Inventors: 冯汉忠
Original assignee: Via Technologies Inc
Current assignee: Via Technologies Inc
Priority date: 2006-07-18
Filing date: 2007-06-21
Publication date: 2010-02-17
Anticipated expiration: 2027-06-21
Also published as: CN101083765A

Abstract

The present invention relates to a system and method for compressing data in a data area block and storing in a memory, as well as transmitting along data path. Through using the bit in a data word which is not used before, storing the available data effectively and transmitting with lasser transmission cycle, so as to increase the usability of the data bus to other main controller. The present invention is used for a system for storing and transmitting compressed data, the system comprises a main controller and a controlled device connecting with each other through data bus. One of the main controllers is video input interface, which is used to receive video data of external video source and compresses further the video data using compression algorithm of the color difference of two contiguity pixels; the other of the main controller is video display controller which is used to receive compressed video data. Also, the video display controller can be used to decode the compressed videodata, so as to recover the video data to its original type to transmit to the external video display.

Description

The System and method for of video data compression

Technical field

The invention relates to a kind of computer system, this computer system comprises main controller and the controlled device of sharing common data bus, and also comprises memory.More particular words it, the invention relates in order to compressed video data it be stored in the memory more effectively, and more efficiently along data/address bus system for transmitting and method.

Background technology

Fig. 1 is the part calcspar of known integrated circuit (IC, integrated circuit) chip 10 1 examples.Chip 10 comprises by data/address bus 16 an interconnective x main controller 12 and y controlled device 14.Chip 10 also comprises bus arbiter 18, the bus arbitration request that bus arbiter 18 receives from main controller 12, and once allow a main controller 12 control buss 16.When a certain main controller 12 is allowed to bus 16 controlled, any controlled device 14 of this main controller 12 access optionally subsequently.

For example, one in the main controller 12 can be through being designed in order to receive the interface from the video data of external source.This interface can with video Data Transmission to the controlled device 14 one, such as Memory Controller, " the news frame buffer " of this Memory Controller may command external memory devices (framebuffer) in storage of video data.Again, one in the main controller 12 can be video display controller, and it can ask access to be stored in by the video data in the news frame buffer of controlled device (Memory Controller) 14 storage arrangements of being controlled.Video data is captured by Memory Controller 14 subsequently, and deposits to data/address bus 16 and use for video display controller.Video display controller captures video datas from data/address bus 16 subsequently, and video data is fed to the external video display.

It should be noted that the typical storage apparatus organizes with the data width of 16 (half-words) or 32 (words) usually.Yet video data is normally organized according to the data width of pixel wide, and this data width can be 16,18 or 24 bit wides.Each pixel of video data generally includes three colour components: red (R), green (G) and blue (B), these colour components are the three primary colors of light.Usually, each colour component comprises that 6 or 8 data bit are to represent this color.Comprise at each colour component under the situation of 6 positions that pixel comprises 18 positions altogether.Comprise at each colour component under the situation of 8 positions that pixel comprises 24 positions altogether.All be stored in the storage arrangement at 18 of each pixel or 24 video data with 32 bit formats.Any do not use or " not minding " (don ' t care) position is all ignored by video display controller.

Fig. 2 A is the allocation map table that is stored in 8 pixel block of video data in the 32 bit data width memories.In this configuration, each colour component comprises 6 positions, and each pixel has 18 positions.For example, position 17 to 12 can be used for redness (R) composition, and position 11 to 6 can be used for green (G) composition, and 5 to 0 of positions can be used for blueness (B) composition.At this system's meta 31 to 18 for not using or not minding position (X ' s).First word table shows first pixel (pixel " 0 ") that comprises colour component R0, G0 and B0; Second word table shows second pixel (pixel " 1 ") that comprises colour component R1, G1 and B1, and the rest may be inferred.

Fig. 2 B is the allocation map table that is stored in 8 pixel block of video data in the 32 bit data width memories.In this configuration, each colour component comprises 8 positions, and each pixel has 24 positions.For example, position 23 to 16 can be used for redness (R) composition, and position 15 to 8 can be used for green (G) composition, and position 7 to 0 can be used for blueness (B) composition.Position 31 to 24 is not for using the position.Although the known configurations shown in Fig. 2 A and Fig. 2 B represents to store the plain mode of video data, have at least 1/4th not use as can be seen in each word.

In addition, wait in the known stored technology thus may be obvious that, exist many be stored in the memory and along position that data/address bus 16 transmits for not minding " X ".By minimizing the number that does not use the position that is stored in the memory, more storage area can be used for other purposes.In addition, minimize during the transmission not and to use the position also to can be all bus users bigger bus availability is provided, thereby allow chip running more effectively under given speed.

Summary of the invention

The present invention is used for data storing in memory and along the system and the method for data/address bus transmission data.By from data source end Real Time Compression data, data can be sent to memory in the less clock cycle.As is known to the person skilled in the art, compressed data are utilized less storage space significantly, thereby need save the storage area for other.And when from the memory requests data, compressed data can be sent to the requestor in the less clock cycle, thereby increase the availability of data/address bus to the main controller of other shared data bus.After compressed transfer of data finished, the main controller decodable code data of filing a request were to return to unprocessed form with it.For example, disclosed system of the present invention and method can be used for the transmission and the storage of video data, and this video data comprises that at common each word of its reset condition several do not use the position.

Generally speaking, the present invention discloses a kind of a plurality of main controllers and controlled device and system that is connected the data/address bus of these main controllers and controlled device of comprising.In the main controller one is video input interface, and it is in order to receive the video data from external video source.Video input interface also can utilize according to the compression algorithm of aberration between two neighborhood pixels and come compressed video data.When be stored at first in the data word per two neighborhood pixels between difference when surpassing predetermined critical, the described compression algorithm of described video input interface is compressed to described two or more neighborhood pixels in the data word.Another person in the main controller is a video display controller, and it is in order to receive compressed video data.Video display controller also can be in order to the compressed video data of decoding, thereby video data is returned to its primitive form to be used to transfer to the external video display.

One embodiment of video input interface of the present invention can comprise temporary transient video storage device, color change analytical equipment, device for classifying data and buffer.Temporary transient video storage device is in order to receive the video data from external video source.The color change analytical equipment is in order to analyze from first preceding pixel the color change to current pixel.Device for classifying data in order to judge whether color change value by analysis can be used as packed data is compressed to described two or more neighbors that change by analysis of representing the colour component of two above pixels in the single data word.Buffer is in order to store compression and/or unpressed video data.

In addition, the video display controller of the embodiment of the invention is also in conjunction with video input interface.The buffer of video input interface uses in order to stored data are placed on the data/address bus for Memory Controller, this Memory Controller be in order to data storing in external memory storage.Memory Controller will be sent to video display controller from the data of external memory storage.Video display controller can utilize and the algorithm of the used compress technique complementation of the device for classifying data compressed video data of decoding.

In running, these systems can utilize some method to come processing video data.For example, an embodiment of the disclosed method of the present invention comprise calculating from first preceding pixel the color change value to current pixel.The method comprises that also the colour component of analysis first pixel to the second pixel in described video data changes; Whether can be used as packed data with the value that changes according to the described colour component of expression, described two or more neighbors that change by analysis of representing the colour component of two above pixels are compressed in the single data word.

Other system of the present invention, method, feature and advantage will explain that in detail those skilled in the art should fully understand the present invention in the graphic and explanation hereinafter.All systems, method, feature and advantage that spirit according to the present invention is in addition derived all belong to scope of the present invention, and the claim of being enclosed and protecting.

Description of drawings

Fig. 1 is the calcspar of the known main controller/controlled device configuration of explanation integrated circuit (IC) chip.

Fig. 2 A and Fig. 2 B are the chart of the known configurations of the pixel data in the block of video data.

Fig. 3 is the part calcspar according to an embodiment of the computer system of teaching of the present invention.

Fig. 4 is the calcspar of an embodiment of the video input interface shown in Fig. 3.

Fig. 5 is the calcspar of an embodiment of the temporary transient video storage device shown in Fig. 4.

Fig. 6 is the calcspar of first embodiment of the color change analytical equipment shown in Fig. 4.

Fig. 7 is the calcspar of first embodiment of the device for classifying data shown in Fig. 4, and described device for classifying data combines running with the color change analytical equipment of Fig. 6.

Fig. 8 is for illustrating an example chart of the video data that compresses according to the possible compress technique of the embodiment that utilizes Fig. 5 to Fig. 7.

Fig. 9 is the calcspar of second embodiment of the color change analytical equipment shown in Fig. 4.

Figure 10 is for illustrating an example chart of the video data that compresses according to the possible compress technique of the embodiment that utilizes Fig. 9.

Figure 11 is the calcspar of second embodiment of the device for classifying data shown in Fig. 4, and described device for classifying data combines running with the color change analytical equipment of Fig. 9.

Figure 12 is the calcspar of an embodiment of the video display controller shown in Fig. 3.

Figure 13 is the flow chart of an embodiment of the method for explanation processing video data.

[main element label declaration]

10,22: integrated circuit (IC) chip

12,30: main controller

14: controlled device

16,34: data/address bus

18,36: bus arbiter

20: computer system

24: memory

26: the video data capture device

28: video display

30a: video input interface

30b: video display controller

32: Memory Controller

40: temporary transient video storage device

42: the color change analytical equipment

44: device for classifying data

46: buffer

50: color separator

52: the current pixel buffer

54: first preceding pixel buffer

The 56:R buffer

The 58:G buffer

The 60:B buffer

The 62:R-1 buffer

The 64:G-1 buffer

The 66:B-1 buffer

68: line

70,92: color incremental computations device

72,100: the critical value checkout gear

74,102: red increment calculator

76,104: green increment calculator

78,106: blue increment calculator

80: the code adding set

82,86,120: temporary transient buffer

84,118: data are put and the code adding set

88: the flag generation device

90,124: data link

94: previous color increment buffer

96: the first high increment detecting devices

98: the second high increment detecting devices

108: previous red increment buffer

110: previous green increment buffer

112: previous blue increment buffer

122: control device

Embodiment

The invention relates in order to compressed video data and with it and be stored in system and method in the memory more efficiently.Under the situation according to video data compression method of the present invention, data also can be transmitted along data/address bus more efficiently.For example, system of the present invention and method can be used in the computer system, especially have in integrated circuit (IC) chip or processor of shared common data bus.Compressed video data can reduce the number of not minding (not using) position in real time, preferably to utilize previous untapped space in the data block.In view of the above, video data can store in the less clock cycle and transmit along data/address bus more effectively.Thus, reduce the number of clock cycle and can be the total power consumption that other bus user provides bigger bus bandwidth and can reduce system.

In computer processing system, data/address bus, Memory Controller and external memory storage are the public resources that processor and many main controllers and peripheral device are shared.Therefore, each bus user of optimization is useful to the utilization of these public resources.The saving format of video data can not utilize each position promptly to be stored in the memory usually.Therefore, when preparing data storing in memory, compressed video data is favourable in real time.Again, compressed video data block can be transmitted between memory and other main controller by data/address bus in the clock cycle in fewer purpose, and can not take data/address bus for a long time.

Fig. 3 is the part calcspar of computer system 20 1 embodiment of the teaching according to the present invention.Computer system 20 comprises integrated circuit (IC) chip 22, memory 24, video data capture device 26 and video display 28.Memory 24 can comprise read-only memory (ROM, read-only memory) and/or random access memory (RAM, and can also comprise dynamic random access memory (DRAM, dynamic random access memory) random access memory).Memory 24 is in the outside of chip 22, and with chip 22 in the different mode access of high-speed cache.Video data capture device 26 can be digital camera, complementary metal oxide semiconductors (CMOS) (CMOS) imageing sensor or other suitable video source.Video display 28 can be raster scan device, cathode-ray tube display (CRT), LCD (LCD) or other suitable display unit.

Chip 22 comprises by data/address bus 34 interconnective a plurality of main controllers 30 and Memory Controller 32.Chip 22 also comprises the bus arbiter 36 that is connected to data/address bus 34, to be used for the control bus arbitration.Among main controller 30, at least one is video input interface 30a, and at least one is video display controller 30b.Video input interface 30a receives the video data from video data capture device 26, and via data/address bus 34 with video Data Transmission to Memory Controller 32 to be stored in the memory 24.After video display controller 30b sent request, memory 24 was via data/address bus 34 supply video data.Can provide data to video display 28 continuously behind the video display controller 30b receiving video data, video display 28 is subsequently to computer user display video.

Chip 22 also can comprise one or more controlled devices, only illustrates a controlled device in Fig. 3, Memory Controller 32.Although only illustrate a controlled device in this embodiment, should notice that chip 22 can comprise the controlled device of any number.The bus request that bus arbiter 36 receives from main controller 30, and once allow a main controller 30 control buss 34.When main controller 30 when allowing bus 34 controlled, this control main controller 30 is any controlled device of access optionally subsequently, such as Memory Controller 32.

Usually, video data is stored in the block (being called news frame buffer) of memory 24, and news frame buffer is the specific part at memory 24, and it can utilize storage and uniform device framework (UMA, unifiedmemory architecture) to distribute or store.Video input interface 30a receives the video data from video data capture device 26, and utilizes the compression algorithm that can not sacrifice video information to come compressed video data.Video input interface 30a subsequently via data/address bus 34 with compressed video Data Transmission to Memory Controller 32, in the news frame buffer that is stored in memory 24.By the data of compressed video input interface 30a, compare usually with the situation of data uncompressed and can in the clock cycle still less, carry out along the transfer of data of data/address bus 34.

Subsequently, video display controller 30b is from the compressed video data of memory 24 acquisitions, and decoding is compressed video data so that it is returned to primitive form.Video display controller 30b provides video data to video display 28 in the fixing mode of crossfire (constant stream) subsequently.Video display controller 30b can utilize and the reverse algorithm of the used compression algorithm of the video input interface 30a compressed video data of decoding, and data being returned to its primitive form and can not lose video information, thereby keeps original image quality.Therefore, video input interface 30a and video display controller 30b comprise that complementary circuit and/or logic are to carry out compression and decoding processing.Hereinafter will discuss the functional of the details of embodiment of video input interface 30a and video display controller 30b and data compression and decoder module.

Fig. 4 is the calcspar of the embodiment of the video input interface 30a shown in Fig. 3.In this embodiment, video input interface 30a generally includes temporary transient video storage device 40, color change analytical equipment 42, device for classifying data 44 and buffer 46.The video data that temporary transient video storage device 40 receives from video data capture device 26, and each pixel is divided into its corresponding colour component, for example red (R), green (G) and blue (B).Temporary transient video storage device 40 stores the data of current pixel, and also stores the data of first preceding pixel.In video news frame, current pixel may be the pixel that is adjacent to first preceding pixel in same column.In many cases, current pixel may have similar in fact value to first preceding pixel.In another embodiment, a broomrape preceding pixel can compare with the pixel in the adjacent column.In another embodiment, the pixel of a news frame can compare with the last pixel that is right after in the news frame.These alternate embodiment depend primarily on the available storage area amount that stores pixel data.Temporary transient video storage device 40 with current and previous colour component transfer of data to color change analytical equipment 42.Again, temporary transient video storage device 40 is supplied to device for classifying data 44 with the not packed pixel data (for example 24 positions) of current pixel.

Color change analytical equipment 42 receives current and previous color data and detects the variation of each color-values to current pixel from first preceding pixel.In certain embodiments, the changing value of each color is defined by 5 positions, one of them bit representation symbol (plus or minus) and values that four bit representations change.5 changing values of each color are combined into 15 color change value and are supplied to device for classifying data 44.For example, in another embodiment, can use extra position to define each color change.Again, whether color change analytical equipment 42 is analyzed each aberration and can be defined by a limited number of position that is no more than predetermined critical (for example 5 position or still less).If can not, then color change analytical equipment 42 signal provision of will " facing limit " to device for classifying data 44 allows too greatly and not to compress to indicate at least one color change.

In other embodiments, color change analytical equipment 42 can be used the packed data of representing value of chromatism with supply, and wherein the size of packed data is variable.In this regard, color change analytical equipment 42 is also supplied coded message to device for classifying data 44, the particular data size of this coded message indication individual pixel.By utilizing size of data, device for classifying data 44 is packaged data and can not lose effective information suitably.Therefore, Fig. 9 to Figure 10 hereinafter described illustrates that permission is with the embodiment of data compression to the flexibility of variable-size.

24 uncompressed video data signals that device for classifying data 44 receives from temporary transient video storage device 40.Device for classifying data 44 also receives from the difference signal of the expression color change value of color change analytical equipment 42 and faces limited signal and/or coded message.Suitably be in the limit (bit limit) if face limited signal indication color change, then device for classifying data 44 can be with the actual data of the worthwhile work of color change, thereby are stored in the buffer 46 as compressed video data.When device for classifying data 44 received the compressible data of two consecutive periods, these two color change values can be stored in one 32 the word.If the size of the data of expression color change is not to be in the specific size limit, then device for classifying data 44 uses 24 not packed data to be stored in the buffer 46.

In addition, device for classifying data 44 can increase bits of coded, to indicate the compressive state of stored data.Subsequently, when the amount of the video data in the buffer 46 reaches predetermined level, can compressed block of video data be transferred to Memory Controller 32 via data/address bus 34.Hereinafter will the embodiment of temporary transient video storage device 40, color change analytical equipment 42 and device for classifying data 44 of video input interface 30a and the details of running be described referring to Fig. 5 to Fig. 7 and Fig. 9.

Fig. 5 is the calcspar of an embodiment of the temporary transient video storage device 40 shown in Fig. 4.In this embodiment, temporary transient video storage device 40 comprises color separator 50, current pixel buffer 52 and first preceding pixel buffer 54.Current pixel buffer 52 comprises that the R buffer 56 of the red composition that is used to store current pixel, the G buffer 58 and being used to that is used to store the green composition of current pixel store the B buffer 60 of the blue composition of current pixel.Elder generation's preceding pixel buffer 54 comprises that the R-1 buffer 62 of the red composition that is used to store first preceding pixel, the G-1 buffer 64 and being used to that is used to store the green composition of first preceding pixel store the B-1 buffer 66 of the blue composition of first preceding pixel.

The video data that color separator 50 receives from video data capture device 26.Subsequently, color separator 50 is according to the data width of the video data that is imported into and separate each pixel according to colour component.When video data is 18 bit wides (as the situation of Fig. 2 A), represent that the position 17 to 12 of red composition is placed in the R buffer 56, represent that the position 11 to 6 of green composition is placed in the G buffer 58, and represent that the position 5 to 0 of blue composition is placed in the B buffer 60.When video data is 24 bit wides (as the situation of Fig. 2 B), position 23 to 16 is placed in the R buffer 56, and position 15 to 8 is placed in the G buffer 58, and position 7 to 0 is placed in the B buffer 60.Therefore,

buffer

56,58 and 60 must can store at least 8 positions.In this embodiment, the current pixel information in the current pixel buffer 52 exports device for classifying data 44 to along 24 paths.Although some examples herein relate to 24 original video data width, should understand and to carry out some modification and to adapt to other width the present invention.

Buffer

56,58 and 60 can be shift registor, with when receiving new current pixel information, current pixel information is moved in the first preceding pixel buffer 54 from current pixel buffer 52.In this way, first preceding pixel buffer 54 stores current pixel buffer 52 previous stored Pixel Information.For example, when receiving new pixel data or according to clock signal, data can be moved.In another embodiment, remaining in 24 Pixel Information in the first preceding pixel buffer 54 also can transfer to device for classifying data 44 and be used.The value that remains in each of

buffer

56,58,60,62,64 and 66 exports color change analytical equipment 42 to along route 68.

Fig. 6 is the calcspar of first embodiment of the color change analytical equipment 42 shown in Fig. 4.Generally speaking, the color change analytical equipment 42-1 among this embodiment comprises color increment (delta) calculation element 70 and critical value checkout gear 72.Color incremental computations device 70 comprises red increment calculator 74, green increment calculator 76 and blue increment calculator 78.Each increment calculator 74,76 and 78 receives the color component value of current pixel (R, G, B) and the color component value of first preceding pixel (R-1, G-1, B-1) respectively, and calculate difference between this equivalence, note the symbol (plus or minus) and the value of difference.In certain embodiments, each increment calculator 74,76 and 78 capacity can keep only 5 positions.If increment size is higher than five positions, then overflow position or carry digit can be sent to critical value checkout gear 72, face limit checkout gear 72 and judge whether have increment to surpass 5 positions.Perhaps, increment calculator 74,76 and 78 capacity can store the expression colour the highest may the variation (that is, from colourless to panchromatic, or from panchromatic to colourless) value.

Export device for classifying data 44 from the every of each increment calculator 74,76 and 78 to along 15 bit lines.Surpass 5 with upper if each increment calculator can store, then only 1 sign bit and 4 minimal effective dose value positions are transferred into device for classifying data 44.Again, increment calculator 74,76 and 78 can transmit at least one position to critical value checkout gear 72.If each increment calculator can be handled 5 positions, then can send single position to critical value checkout gear 72.As mentioned above, this single position can be overflow position or carry digit.Perhaps, increment calculator also can send the indicating bit of any adequate types, has indicated increment whether to surpass 5 limit.Can handle among the embodiment of maximal increment value at each increment calculator 74,76 and 78, increment calculator can transfer to critical value checkout gear 72 with whole increment size.No matter the understanding of the present invention and the what state of these or other embodiment of dreaming up, critical value checkout gear 72 can receive information and whether surpass 5 limit with the increment size of judging one or more colour components.If arbitrary increment greater than 5 positions, then can not attempted current pixel is compressed in the case usually.Yet in the embodiment of Fig. 9, color change analytical equipment 42 can be handled the situation that increment can surpass 5 positions, hereinafter will be explained in more detail this.

Referring to Fig. 6, after judging whether increment size oversteps the extreme limit, critical value checkout gear 72 is sent to device for classifying data 44 to indicate whether to compress current pixel with " facing limit " signal of 1 once more.For example, when not needing packed data, critical value checkout gear 72 can send the logical zero signal, and when the needs packed data, then sends the logical one signal.Critical value checkout gear 72 can comprise any appropriate combination of logic element and be configured to generation and provide this to face limited signal.

Fig. 7 is the calcspar of an embodiment of the device for classifying data 44 shown in Fig. 4.Device for classifying data 44-1 among this embodiment can combine running with the embodiment of the color change analytical equipment 42-1 of Fig. 6.Device for classifying data 44-1 receives 24 signals from temporary transient video storage device 40, the not packed pixel information of these 24 signal indication current pixels.Perhaps, this unpressed Pixel Information also can directly be received from video data capture device 26.Device for classifying data 44-1 also receives from 15 signals of color change analytical equipment 42-1 and faces limited signal.5 increments of three colour components of these 15 each color incremental computations of signal indication.

According to the embodiment of Fig. 7, device for classifying data 44-1 comprises code adding set (code adding device) the 80 and first temporary transient buffer 82 that is in first branch.Device for classifying data 44-1 comprises the data storing and the code adding set 84 and the second temporary transient buffer 86 that is in second branch.Device for classifying data 44 also comprises flag generation device 88 and data link 90.In first branch, code adding set 80 receives 24 not packed pixel data, and particular code value is placed in one or more positions to indicate this pixel data uncompressed.Code adding set 80 produces one 32 word, comprises 24 valid data and the one or more bits of coded of representing unpressed pixel data.Remaining bit be do not mind the position and can be left in the basket.In this example, a position (for example logical zero) is for being placed in the position 31 to indicate the bits of coded of Pixel Information uncompressed.These 32 words are stored in the first temporary transient buffer 82 subsequently.

The limited signal that faces that flag generation device 88 receives from critical value checkout gear 72.Flag generation device 88 produces flag signals with the response minimum detectable signal, flag signals can illustrate data put with 84 two continuous compressed signals of code adding set will be stored in 32 word where.For example, flag signals can be indicated first group of compressible pixel data is stored in the position, 15 positions, bottom 14 to 0 of this word, and second group of compressible pixel data is stored in the position, 15 positions, top 30 to 16 of this word.If can put upside down this configuration when needing.In this embodiment, flag generation device 88 is reset to zero.Subsequently, in response to the indicated limited signal that faces above, wherein " 0 " expression does not have compression and " 1 " expression compression, and flag generation device 88 produces flag signals as follows: be " 0 " when facing limited signal, flag signals is reset to " 0 " in following one-period; Be " 1 " when facing limited signal, flag signals triggered in following one-period.

According to respect to the above-mentioned logic of facing limited signal and flag signals, data are put with code adding set 84 and operated as follows in second branch: data are put 15 increment signals that receive expression color change to current pixel from first preceding pixel with code adding set 84.This increment signal is actual compression signal used among the present invention and must not carries out extra process, unless this paper clearly mentions.In conjunction with temporary transient buffer 86, data are put with code adding set 84 compressed signal (increment signal) are placed in 16 positions in 16 positions, top or bottom of 32 the word that is kept in the temporary transient buffer 86.In addition, data are put two bits of coded of adding the indication compressive state with code adding set 84.

When flag signals is " 0 ", represent that temporary transient buffer 86 does not comprise valid data.In the case, compressed signal can be stored in 15 positions, bottom of temporary transient buffer 86, and coding place value " 1 " is stored in a position 31.When flag signals is " 1 ", represent that temporary transient buffer 86 is part fillings, and have valid data in the position, position 14 to 0 down, that is, when being " 0 ", flag signals is stored in this place.Next 15 compressed signal is stored in a position 30 to 16 subsequently, and the coding place value " 1 " be stored in a position 15, thereby the indication upper part be effective.Finish if the content of temporary transient buffer 86 stores, and only have following position effective, then code value " 0 " is stored in a position 15, thereby the indication upper part is invalid.Mode is " 11 " as if a position 31 and 15, position according to this, and then going up group all is effective with following group position.For value " 10 ", it is effective only descending the position.For the code adding set 80 in first branch because " 0 " is inserted in the position 31 to indicate unpressed data, so a position 15, position comprises valid data, but in order to discern compressive state when decipher as not minding the position.Again, under the situation of the incompressible pixel of " 0 " in the position 31 on the throne indication, the content of temporary transient buffer 86 is discardable or reset, because data wherein are invalid.

Although this paper utilizes particular value " 0 " or " 1 " to describe flag signals and face limited signal, it should be noted that convertible this equivalence under the situation that does not break away from spirit of the present invention and category.Again, specific bit-position as herein described is arbitrarily and can optionally be changed to any suitable configuration.In addition, these are revised and are not broken away from spirit of the present invention and category.

The first temporary transient buffer 82 and the second temporary transient buffer 86 keep the not compression and the compressed version of current pixel information respectively.When two continuous compressible pixels were stored in wherein, temporary transient buffer 86 also can keep previous packed pixel.Data link 90 is access buffer 82 and 86 according to facing limited signal and flag signals, and related data is supplied to buffer 46.When running, data link 90 transmits data according to explanation hereinafter.

When flag signals for " 0 " and face limited signal when " 0 ", mean that temporary transient buffer 86 does not comprise the effective information from first preceding pixel, and can not compress current pixel, then data link 90 is from the temporary transient unpressed data of buffer 82 accesses, and this transfer of data to buffer 46 is stored being used for.When flag signals for " 0 " and face limited signal when " 1 ", mean that temporary transient buffer 86 does not comprise the effective information from first preceding pixel, and will compress current pixel, then data are put with the first group packed pixel data of code adding set 84 with current pixel and are placed in the temporary transient buffer 86, organize filling remaining bit (if this second continuous group also is compressible) continuously in the hope of utilizing second.Yet in the case, data link 90 does not send any data to buffer 46.

When flag signals for " 1 " and face limited signal when " 0 ", mean that temporary transient buffer 86 comprises the effective information from first preceding pixel, and can not compress current pixel, the word of the part filling in the temporary transient buffer 86 of data link 90 accesses then, and these data are sent to buffer 46.Subsequently, data link 90 also will be sent to buffer 46 from the current word that do not compress of temporary transient buffer 82.Need carry out two transmission in the case.In another embodiment, data link 90 can receive 24 signals optionally, and its expression is from the previous Pixel Information of first preceding pixel buffer 54 shown in Figure 5.In the case, when flag signals and when facing limited signal and being respectively 1 and 0, data link 90 can only transmit the not compressed version of first preceding pixel, and need not send the word from the part filling of temporary transient buffer 86.

When flag signals for " 1 " and face limited signal when " 1 ", mean that temporary transient buffer 86 comprises from the effective information of first preceding pixel and will compress current pixel, then data link 90 pending datas such as grade are put with code adding set 84 current packed data and previous packed data are placed in the word stored in the temporary transient buffer 86.Subsequently, the compression word of the complete filling in the temporary transient buffer 86 of data link 90 accesses and these data are sent to buffer 46.

One embodiment chart of the possible configuration of the video data that Fig. 8 compresses for compression algorithm according to the present invention.The compressible video data of video input interface 30a to be utilizing more not usage space, but not allows video data to keep unprocessed form shown in Fig. 2 A and Fig. 2 B.In this configuration, video data can transmit along data/address bus 34 more effectively, and is stored in more efficiently in the memory 24.

The chart of Fig. 8 is showed an embodiment of the Pixel Information of the video data block that comprises 11 pixels (pixel 0 to 10).In this embodiment, some pixel be unpressed and other pixel be the compression.The red composition of the not compression of symbol R0 remarked pixel 0.Symbol dR2 represents the compressed signal that defined with respect to the increment size (or difference) of the previous red composition in the pixel 1 by the red composition of pixel 2.In this example, for first pixel of video data block, face limited signal and will be reset to 0.Because do not have the reference pixel that can supply increment to compare, first pixel can not be compressed.Therefore pixel 0 is unpressed, and the indication of the coding place value " 0 " in the position 31 pixel is a uncompressed.

Pixel

1 and 2 among this embodiment all has the increment less than 5 positions, therefore allows these pixels of compression.As seen in this embodiment, two continuous compressible pixels are used to form the compression word of complete filling.As described herein, pixel 3 with 4 be the compression and be placed in the word, pixel 5 with 6 be the compression and be placed in the word, and pixel 9 with 10 be the compression and be placed in the word.

In this embodiment, also visible pixels 7 is compressed and be placed in down in the position, position 14 to 0, and wherein dR7 is in the position 14 to 10, and dG7 is in the position 9 to 5, and dB7 then is in the position 4 to 0.Yet,, can transfer to buffer 46 so comprise the compression word of the part filling of packed pixel 7, and the upper part is not for minding the position because be incompressible according to this embodiment pixel 8.Position 31 is effectively for " 1 " with indication lower position (14 to 0), but position 15 is " 0 ", because upper position (30 to 16) is invalid.Proposed as mentioned, data link 90 can be selected one and receive the pixel 7 in the uncompressed state and store this pixel under unpressed situation.

In this embodiment as seen, four compression words that word is complete filling.Therefore, take 11 words in the memory in contrast to requiring original video data is stored in, and it was transmitted along data/address bus in 11 transmission cycles, this embodiment can transmit data storing in 7 words and in 7 cycles, so that delivery time and storage area reduce about 36%.

Fig. 9 is the calcspar of second embodiment of the color change analytical equipment 42 shown in Fig. 4.In this embodiment, color change analytical equipment 42-2 comprises color incremental computations device 92 and previous color increment buffer 94.In addition, color change analytical equipment 42-2 comprises first high increment detecting device 96, second high increment detecting device 98 and the critical value checkout gear 100.Be similar to the color incremental computations device 70 shown in the embodiment of Fig. 6, color incremental computations device 92 comprises red increment calculator 102, green increment calculator 104 and blue increment calculator 106.Previous color increment buffer 94 comprises previous red increment buffer 108, previous green increment buffer 110 and previous blue increment buffer 112, and each is used for temporarily storing previous color change.

Each increment calculator 102,104 of color

incremental computations device

92 and 106 receives the colour (R, G, B) of current pixel and the colour (R-1, G-1, B-1) of first preceding pixel, and calculate difference between these colours, and the symbol (plus or minus) and the value of record difference.In customized configuration about Fig. 9, each increment calculator 102,104 and 106 be in order to the highest may the variation of calculating and keeping the expression colour (that is, from colourless to panchromatic, or from panchromatic to colourless) value.The value of expression color change is transferred to the first high increment detecting device 96.Perhaps, increment calculator 102,104 and 106 can be in order to store nearly 6 positions.In the case, if increment size is higher than 6 positions, then overflow position or carry digit can be sent to critical value checkout gear 100, and whether critical value checkout gear 100 decidables have increment to surpass 6 positions.

Maximum 6 positions from each increment calculator 102,104 and 106 export device for classifying data 44 to along 18 bit lines.Each increment calculator in order to store 6 with upper embodiment in, only a sign bit and five minimal effective dose value positions are transferred into device for classifying data 44 in each color.Again, increment calculator 102,104 and 106 can transmit at least one position to the first high increment detecting device 96 with the necessary bits number of the indivedual color change of indication expression.If when needing, increment calculator 102,104 and 106 can transmit the high increment detecting device 96 of whole increment size to the first.No matter other the understanding of the present invention and the what state of these or other embodiment of dreaming up, the first high increment detecting device 96 can independent increment calculator 102,104 and 106 reception information to judge the highest increment size of three color change.Whether the increment size that the first high increment detecting device 96 also can detect one or more colour component surpasses 6 limit.If any increment greater than six positions, then can not attempted to compress in this embodiment usually.The high increment detecting device 96 increment size of high judgement or signal that the needed bits number of the highest increment size is defined in indication transfers to and faces limit checkout gear 100.

High increment detecting device 96 receives about representing the information of the necessary bits number of each color change, and which color of judging current pixel experiences High variation.If at least one color surpasses 6 limit, then high increment detecting device 96 notices are faced limit checkout gear 100 and current pixel are not carried out compression.High increment detecting device 96 sends the signal of the value of indication maximum color increment.

Increment calculator 102,104 and 106 also can be used as shift registor moves to previous color increment buffer 94 correspondences with the value with the color increment increment buffer 108,110 and 112.Again, the second high increment detecting device 98 can operate similarly with the first high increment detecting device 96, changes with the maximum color of the color change group that detects previous color increment buffer 94.Perhaps, with respect to the highest increment size that is calculated first preceding pixel by the second high increment detecting device 98, the first high increment detecting device 96 can move to this information and the color increment information that moves in the second high increment detecting device 98 simply concurrently.According to these or other embodiment, the second high increment detecting device 98 is sent to critical value checkout gear 100 with the highest increment size of previous color increment.

When critical value checkout gear 100 received the highest increment size of current color increment and previous color increment, it judged that then whether two peak sums are greater than 10.If greater than 10, then for these two contiguous pixels, color change is excessive and can't be stored in 32 words.Yet, and if be 10 or littler, may these two contiguous pixels be stored in the word according to the increment of these two contiguous pixels.Preferably, under the situation of this embodiment, the highest increment of one group of color in the pixel can be set the bits number that is used for this each color increment of group.Therefore, if the highest increment needs 6 positions, then each color increment of pixel is defined by 6 positions.Therefore, the increment combination of two contiguous pixels may be 4 positions for each color increment of first pixel, and for each color increment of second pixel, be 6 positions, or be 5 positions and be 5 positions for second pixel for first pixel, or for first pixel 6 positions and be 4 positions for second pixel.The symbol of these configurations is described by 4/6,5/5 and 6/4 respectively in this article.In these configurations, the position of two contiguous pixels adds up to 30, thereby allows two bits of coded to be used to indicate compressive state.

Figure 10 illustrates an example chart of the Pixel Information of the video data block that comprises 7 pixels (pixel 0 to 6).Show not compressed configuration of three compressed configuration and in this embodiment.First configuration comprises 5/5 distribution, and wherein the color change of first pixel (pixel 0) is defined by 5 positions of each color, and the color change of second pixel is also defined by 5 positions of each color.In 6/4 distribution, the color change of pixel 2 is defined by 4 positions of each color, and the color change of pixel 3 is then defined by 6 positions of each color.In 4/6 distribution, the color change of pixel 4 is defined by 6 positions, and the color change of pixel 5 is defined by 4

positions.Position

31 and 30, position is used to store two allocation of codes that sent by critical value checkout gear 100 in this embodiment.In the case, " 11 " indication 5/5 distributes, and " 10 " indication 6/4 distributes, and " 01 " then indicates 4/6 to distribute.In addition, allocation of codes " 00 " is illustrated in does not have compression under the situation of pixel 6.In the case, position 29 to 24, position is not mind the position.

Figure 11 is the calcspar of another embodiment of the device for classifying data 44 shown in Fig. 4.Device for classifying data 44-2 among this embodiment combines running in order to (for example) with the embodiment of the color change analytical equipment 42-2 of Fig. 9.Device for classifying data 44-2 receives 24 signals from the not packed pixel information of the expression current pixel of temporary transient video storage device 40.Perhaps, unpressed Pixel Information can directly receive from video data capture device 26.Device for classifying data 44-2 also receives 18 signals from color change analytical equipment 42-2, faces limited signal and two allocation of codes.18 signal indications are at three increment sizes of maximum 6 positions of each color incremental computations.If only need 4 or 5 positions to express increment size fully, then 18 signals can comprise the effective value below 18.

Device for classifying data 44-2 comprises the code adding set 114 and the first temporary transient buffer 116 that is in first branch.Device for classifying data 44-2 comprises the data storing and the code adding set 118 and the second temporary transient buffer 120 that is in second branch.Device for classifying data 44 also comprises control device 122 and data link 124.In first branch, code adding set 114 receives 24 not packed pixel data, and particular code value is added one or more positions to indicate this pixel uncompressed.Code adding set 114 produces 32 words of the valid data with 24, to represent unpressed pixel data and two or more bits of coded.Remaining bit be do not mind the position and can ignore.In this embodiment, value " 00 " is the encoded bit pattern of uncompressed for being placed in a

position

31 and 30 with the indication Pixel Information.These 32 words are stored in the first temporary transient buffer 116 subsequently.

Control device 122 receive from high increment detecting device 96 face limited signal and from two allocation of codes of critical value checkout gear 100.Control device 122 produces code signals with response, and code signal is put with two continuous compressed signals of code adding set 118 indications to data and where will be stored in 32 words.Code signal is also indicated the size of each compressed signal, to allow data to put with code adding set 118 compressed signal suitably is placed in this word.For example, it is 18 positions that code signal can be indicated first group of compressible pixel data, and will be stored in the position, 18 positions 17 to 0 of this word, and can to indicate second group of compressible pixel data be in 12 positions and the position, 12 positions 29 to 18 that will be stored in this word.This configuration can optionally be put upside down.In this embodiment, control device 122 is resettable to be zero, is similar to flag generation device 88 shown in Figure 7.Whether in addition, control device 122 also can be in response to facing limited signal, compressible with any one or both that look two neighborhood pixels, and control the storing of two neighborhood pixels.Data put with code adding set 118 packed data is placed in this word with the similar in fact mode of the described mode of Fig. 7.

The first temporary transient buffer 160 and second buffer 120 keep the not compressed version and the compressed version of current pixel information respectively.When two continuous compressible pixels were stored in wherein, temporary transient buffer 120 also can keep previous packed pixel information.Data link 124 is according to from the control signal of control device 122 and access buffer 116 and 120, and the supply related data is to buffer 46.For example, but control device 122 designation data conveyers 124 with the described flag signals of Fig. 7 and the explanation of facing limited signal in fact similarly mode transmit data.Control device 122 can take into account the effective information whether temporary transient buffer 120 comprises from first preceding pixel or current pixel.Whether control device 122 also can take into account first preceding pixel or current pixel compressed, and which kind of compression distribution it utilizes compress.It is that transmission zero, one or two effective words are to buffer 46 that control device 122 also can take into account data link.

Referring to Fig. 9, will an alternate embodiment that allow bigger compression be described once more.Extra previous color increment buffer 94 can be used to store the color change of big array in this embodiment.In the case, extra high increment detecting device also can be in order to corresponding with each extra previous color increment buffer.Again, but critical value checkout gear 100 receiving arraies each the row a plurality of high increment size.Critical value checkout gear 100 is obtained color increment sum to judge transformable allocative decision.And if be less than the useful number of word meta, then the color increment of two above contiguous pixels can be combined in 32 words, thus compressed video data further.For example, the highest increment of contiguous pixels can be represented according to following configuration, can comprise 2/2/5,2/5/2,5/2/2,2/3/4,2/4/3,3/2/4,3/4/2,4/2/3,4/3/2,3/3/3,2/2/2/3,2/2/3/2,2/3/2/2 or 3/2/2/2, wherein each combination and be 9 or littler.Under the situation of using 27 positions with the color increment that defines three colors among this embodiment, 5 extra bits can be used as code bit to define 2 ⁵(32) planting may compressed configuration.Can utilize other compression scheme to come further compressed video data on demand.

In other embodiments of the invention, color change analytical equipment 42 can be compared each pixel with predetermined pixel value, rather than compares with neighborhood pixels.For example, color change analytical equipment 42 can be selected the reference pixel of colorful mean value, and each pixel is compared to find out increment with reference pixel.Therefore, video display controller 30b can recover video data by increment is added into reference pixel.And in other embodiments of the invention, video data can be divided into black/white or GTG, and this structure on video data is decided.

Figure 12 is the calcspar of the embodiment of the video display controller 30b shown in Fig. 3.This embodiment that the understanding of the present invention dreams up and other suitable embodiment are intended to use with the device cooperation of the packed data according to being scheduled to compression algorithm (such as video input interface 30a as herein described used algorithm or method).In this embodiment, video display controller 30b comprises decoder 130, increment buffer 132, recovers buffer 134 and buffer 136.

Decoder 130 receives from the compressed video data of data/address bus 34 and these data of decoding.At first whether decoder 130 code analysis positions are compressed with decision data.For the embodiment of Fig. 4 to 7, decoder 130 can be checked the code bit in position 31 and 15.If this equipotential is " 0X ", then this word is a uncompressed.If this equipotential be " 10 ", then only the position in the position 14 to 0 is effective, and this expression comprises the word of part filling of the packed data of a pixel.If this equipotential is " 11 ", then two pixels are compressed in the word, and wherein position 14 to 0 stores first packed pixel, and position 30 to 26 stores second packed pixel.For the embodiment of Fig. 9, decoder 130 is analyzed

position

31 and 30 to be judged to be the selected compressed configuration of pixel.For the alternate embodiment (wherein 5 positions are used to represent compressive state, and position 26 to 0, position is valid data) of Fig. 9, decoder 130 can be analyzed a position 31 to 27.

When code bit was indicated unpressed word, decoder 130 not packed data transferred to recovery buffer 134.Recover buffer 134 and keep these words, and also these data are sent to buffer 136 and store.If the code bit indication is compressed fully or the word of part compression, then decoder 130 transfers to increment buffer 132 with increment size.If Ya Suo (two pixels are compressed) fully, then increment buffer 132 stores the increment size of each pixel in order.Again, increment buffer 132 transfers to increment size and recovers buffer 134, wherein increment size is added into first preceding pixel, thereby video data is returned to its primitive form.Recover buffer 134 subsequently data recovered is transferred to buffer 136.If a word comprises two packed pixels, then repeat to be added into the process of recovering buffer 134, every group of increment is once.Buffer 136 can be supplied to video display 28 with the video data that recovers with normal crossfire continuously subsequently.

Video display 28 is seen video data by the mode that video data capture device 26 is captured at first with video data.In this regard, video display 28 need not redesign.Although data are recovered by video display controller 30b, along data/address bus 34 transmission and during memory 24 storages, the form of this video data is changed by the performed compression algorithm of video input interface 30a at this video data.Video input interface 30a compressed video data can not lose valid data by this to reduce size.In this way, the video data of compression can utilize less storage space to store, and transmits along data/address bus 34 in the less cycle.Video display controller 30b utilizes subsequently with the decoding algorithm of the compression algorithm contrary of video input interface 30a the video data of compression is changed back its primitive form.

Video input interface 30a of the present invention and video display controller 30b can be established in hardware, software, firmware or its combination.In disclosed embodiment, any person of the counter element of these devices all can be established in the memory stored software or firmware at least in part, and comes executive software or firmware by the instruction of suitable executive system.Perhaps, this logic can be established in the hardware of any combination with appropriate members, such as the discrete logic with the grid that is used for the construction logic function, application-specific integrated circuit (ASIC) (ASIC, application specific integrated circuit) etc.

Figure 13 is the flow chart of an embodiment of the method for explanation processing video data.Determine in 140 in the district, at first calculate the color change value to current pixel from first preceding pixel.This equivalence can be represented the color change of redness, green and blue composition about color.In block 142, whether the value of judging color change is greater than predetermined critical.For example, be used to represent that the bits number of value can be by analysis to judge whether value surpasses critical value.If judge that in block 142 value is not more than critical value, the color change that this indication is less relatively, then flow process enters to block 144.If value is greater than critical value, then flow process enters to block 146.In block 144, the color change that is calculated in the block 140 is elected to be the compressed version of current pixel.In block 146,, do not utilize compression at the color change value when color change during greater than critical value.In the case, select the not compressed version of current pixel.

The How It Works of video input interface 30a and video display controller 30b can comprise any suitable architecture of the various embodiments of process software, functional and/or running.About this point, each function can be the part of module, section or code, and it comprises one or more executable instruction that is used for the construction specified.Also it should be noted that in some alternate embodiment these functions can or can be carried out in fact simultaneously not according to the specified order generation.

What should emphasize is that the foregoing description only is the example of possible embodiment.Under the situation that does not break away from notion of the present invention, principle and teaching, can carry out many changes and modification to the foregoing description.All these modifications and change are intended to be included in herein in category of the present invention and protected by described claim scope.

Claims

1. video data compression system comprises:

Video input interface, in order to receive video data with pixel sequence from external video source, described video input interface further utilizes and compresses described video data at the compression algorithm of the aberration between two neighborhood pixels in the described pixel sequence, wherein when be stored at first in the data word per two neighborhood pixels between difference when surpassing predetermined critical, the described compression algorithm of described video input interface is compressed to described two or more neighborhood pixels in the data word; And

Video display controller, in order to receiving described compressed video data, described video display controller further in order to the described compressed video data of decoding to transfer to the external video display.

2. video data compression according to claim 1 system also comprises the combination of following project:

Memory Controller, described Memory Controller receives from the described compressed video data of described video input interface via data/address bus and is stored in the memory, and described Memory Controller further via described data/address bus will described compressed video data from described memory transfer described video display controller extremely; And

Bus control unit, described bus control unit will be exported to one in described video input interface, video display controller and the Memory Controller to the control signal of described data/address bus.

3. video data compression according to claim 1 system, wherein said compression algorithm further is applied to code value at least one position in the described data word, representing whether described data word is compressed, and represent the correspondence of position, position and a position of described original pixels in the described compressed data word; Wherein said video display controller is further according to the described compressed video data of decoding of the described code value in the described compressed data word.

4. video data compression according to claim 1 system, wherein said compression algorithm further is divided into a plurality of colour components with each pixel of described video data, and the corresponding difference of the colour component of more described two neighbors.

5. video data compression according to claim 1 system, wherein said video input interface further produce the described pixel of indication whether compressed face limited signal.

6. video input interface, it comprises:

Temporary transient video storage device is in order to receive the video data with pixel sequence from external source;

The color change analytical equipment is in order to analyze the colour component variation from first pixel to the second pixel of described pixel sequence;

Whether device for classifying data can be used as packed data in order to the value that changes according to the described colour component of expression, and described two or more neighbors that change by analysis of representing the colour component of two above pixels are compressed in the single data word; And

Buffer is in order to store compressed video data, unpressed video data or both.

7. video input interface according to claim 6, wherein said temporary transient video storage device comprises:

Color separator is divided into a plurality of colour components in order to each pixel with described video data, and wherein said colour component comprises redness, green and blue;

First buffer is in order to the value of each colour component of storing described first pixel; And

Second buffer is in order to the value of each colour component of storing described second pixel.

8. video input interface according to claim 7, wherein said color change analytical equipment comprises:

At least one increment calculator in order to receiving the described colour component from described first buffer and described second buffer, and calculates the color increment between described first pixel and described second pixel at each colour component; And

Whether the critical value checkout gear surpasses predetermined critical in order to judge described color increment.

9. video input interface according to claim 8, wherein said temporary transient video storage device also comprises previous increment buffer, wherein said at least one increment calculator further moves to described color increment described previous increment buffer, and calculate the color increment of each colour component of the 3rd pixel, described color change analytical equipment also comprises:

The first high increment detecting device is in order to the soprano of the described color increment that detects described the 3rd pixel;

The second high increment detecting device is in order to the soprano of the described color increment that detects described second pixel; And

The critical value checkout gear, in order to the described maximum color increment of judging described second pixel and described the 3rd pixel and whether surpass critical value.

10. video input interface according to claim 6, wherein said device for classifying data comprises:

The code adding set is in order at least one position in first word of the not compressed version that first generation code value made an addition to the pixel that comprises video data, to represent that described first word is a uncompressed;

The first temporary transient buffer is used to store described first word;

Data are put and the code adding set, and its compressed version with the described pixel of video data is placed in second word, and second generation code value is made an addition at least one position in described second word, comprise compressed data to represent described second word;

The second temporary transient buffer, it is used to store described second word; And

Data link, it is used to supply the described first temporary transient buffer, described second temporary transient buffer or both contents.

11. video input interface according to claim 10, also comprise the flag generation device, whether described flag generation device comprises the compression word of part filling in order to indicate the described second temporary transient buffer, wherein when described flag generation device indicates the described second temporary transient buffer to comprise the compression word of part filling, described data are put with the code adding set and further the compressed version of another pixel of video data are placed in the different piece of described second word.

12. the method for a processing video data, described method comprises:

Receive video data with pixel sequence from memory;

The colour component of analysis first pixel to the second pixel in described video data changes; With

Whether the value that changes according to the described colour component of expression can be used as packed data, and described two or more neighbors that change by analysis of representing the colour component of two above pixels are compressed in the single data word.

13. whether method according to claim 12, wherein said compression step further make an addition to code value in the described data word, compressed to indicate described data word.

14. method according to claim 12 also comprises:

When the bits number of described variation value surpasses predetermined critical, do not compress described second pixel.

15. method according to claim 14, wherein said compression step also comprise at least one position is made an addition in described second pixel, to represent that described second pixel is a uncompressed.

16. method according to claim 14 also comprises:

Described compression and unpressed video data are transferred to Memory Controller to be stored in the described memory via data/address bus;

With stored compression and unpressed video data via described data/address bus from described memory transfer to video display controller; And

By the decode video data of described compression of described video display controller.

17. method according to claim 12, wherein calculate described variation value also comprise calculate in the expression three primary colors each from described first pixel variation value to described second pixel.