CN101420598A - Video monitoring information interaction system based on Symbian platform - Google Patents

Abstract

The invention relates to a Symbian-platform-based video monitoring information interaction system, the whole system is formed by connecting a plurality of wireless monitoring points through a mobile communication circuit, a monitoring part is modular and mobile, data is transmitted by relying on an encrypted circuit and a protocol, and the system enhances the flexibility and the scalability of system construction, increases the reliability of the system and has higher performance-price ratio. The system is formed by connecting a plurality of wireless monitoring points through the mobile communication circuit, information of all the monitoring points collected by a video collector becomes wireless signals which can be transmitted at a Symbian platform by coding through a coder server, and the wireless signals are sent to a mobile terminal through a monitor signal transponder server.

Description

A kind of based on the video monitoring information interaction system under the Symbian platform

One, technical field:

The present invention relates to a kind of video monitoring information interaction system, especially relate to a kind of can use the Symbian platform realize the control point to portable terminal based on the wireless video monitoring information interaction system under the Symbian platform.

Two, background technology:

With the cordless communication network is carrier, and integrated data and voice, image are transmitted in one, realizes multi-functional, multimedia, many unit collaborative commander, and has the social emergency system of safe and reliable data protection and disaster recovery mechanism.To warning in advance, hazard weather, great communicable disease, public's accident, personnel deploy to ensure effective monitoring and control of illegal activities, mobile communication means such as disaster monitors, the condition of a disaster evaluation provide secure data information service timely and effectively, routine work provides the secure interactive means that move to Police Command Center.

Because existing wireless video monitoring system can only be realized the centralized on duty monitoring of control point to Surveillance center (point toCenter), the supervisory control system control point of this pattern is fixed, Surveillance center is huge, report untimelyly etc. for accident (the condition of a disaster) image, shortcoming has seriously influenced relevant department to the monitoring of disaster/accident, a series of activities of handling, dealing with problems arising from an accident.

(1) Symbian platform technology

The Symbian platform is the mobile phone operation platform of one of main flow the most in the market, it has real-time, multitask, low in energy consumption, characteristics such as EMS memory occupation is few, be fit to very much mobile device uses such as mobile phone, it also supports GPRS, blue bud, SyncML and 3G technology simultaneously.The most key it is a standardized open platform, and anyone can be an equipment development software of supporting Symbian.Different with the Microsoft product is, Symbian is with the current techique of mobile device, the kernel of operating system just, separate with graphical user interface technology, the platform that can well adapt to the different modes input, also can make manufacturer meet personalized trend for the friendly more operation interface of the production of oneself, this also is the main cause that the user can see the symbian system of different appearance.Begun in vogue on the internet for the java program of this platform development now.The user can be by installing these softwares, extended mobile phone function.

The Symbian system mainly is made up of following hardware components, central processing unit, ROM, RAM, IO equipment and power supply.Each performs its own functions for each hardware formation, guarantees the operation of system.General 32 bit processors that adopt of Symbian system, system's operation and data operation all rely on processor to finish.Solidify every function that Symbian system and equipment carry in the ROM.RAM then is in order to the essential data of the program of depositing current active and system operation, and all kinds of provisional swap file, perhaps as the WAP buffer memory etc., is responsible for depositing some data of user in addition.IO equipment comprises general control appliance, as keyboard, and touch-screen, extension storage card, blue tooth interface etc.Power supply then is battery or external power supply.

(2) KMVC video coding algorithm

KMVC video coding algorithm (KingTone Mobile Video coder) is my company low compression algorithm of decreasing of the video under low transmission rate with reference to the coding experience independent research of domestic AVS video encoding standard and MPEG-2 video encoding standard.

The MPEG-2 video encoding standard (be called ITU-T again H.262[2]) had the history in 10 years, expand by MPEG-1, support interlacing scan.Use very extensively, almost be used for all digital television systems, be fit to SD and HDTV (High-Definition Television), be fit to various media delivery, comprise satellite, wired, ground etc., can both transmit effectively.Yet technology can only provide less transmission rate, even DVB-T for similar xDSL, UMTS (general mobile system), does not also have enough frequency ranges to use, and the program that provides is very limited, along with the introducing of HDTV (High-Definition Television), presses for the appearance of high compression ratio technology.

AVS video standard (GB/T 20090.2) is based on China's autonomous innovation technology and the constructed standard of international public technology, mainly, have following characteristics towards high definition and High-Quality Digital Television broadcasting, Web TV, digital storage media and other related application:

(1) performance height, code efficiency are more than 2 times of MPEG-2, are in same level with H.264 code efficiency;

(2) complexity is low, and algorithm complex is obviously lower than H.264, and software and hardware realizes that H.264 cost all is lower than;

(3) China grasps main intellectual property, and the license pattern is simple, and expense is low.

The basic principle of KMVC encryption algorithm is: utilize the correlation of time statistics, exploitation inter prediction algorithm; Utilize the local and start frame contrast of picture, the size of compression and minimizing transmission frame adopts the localized variation generating mode, is implemented in abominable network and the application in the terminal environments.

The KMVC encryption algorithm by the video coding layer (Video Coding Layer, VCL), network abstraction layer (Network Abstraction Layer, NAL), image relatively layer (Image Comparison Layer, ICL) three parts are formed.Comprise VCL encoder and VCL decoder among the VCL, major function is video data compaction coding and decoding, and it comprises compression unit such as motion compensation, transition coding, entropy coding.The ICL major function is that the image with different frame compares, and distinguishes different frame blocks and imports VCL into and encode.Comprise unit such as image segmentation, block comparison, block selection.NAL then is used to VCL that a unified interface that has nothing to do with network is provided, it transmits it after being responsible for video data encapsulated packing in network, its adopt system-data format, comprise the video data of the header packet information of single byte, a plurality of bytes and framing, logic channel signaling, timing information, sequence conclusion signal etc.Comprise storage mark and type code in the packet header.Storage mark is used to indicate current data not belong to by the frame of reference.Type code is used for the type of indicating image data.VCL can transmit the coding parameter of adjusting by current network condition.

KVMC algorithm protocol cataloged procedure

What the KMVC encoder adopted is the mixed code approach of conversion and prediction.The frame of input or field are that unit is handled by encoder with the macro block.At first, enter ICL and carry out the comparison of change dynamics piece, interior frame by frame or inter prediction encoding then method is handled.If employing intraframe predictive coding, its predicted value PRED are drawn after motion-compensated by image.In order to improve precision of prediction, thereby improve compression ratio, actual reference picture can be in the past or following (referring on the display order) coding and decoding rebuild and the frame of filtering in select.After predicted value PRED and current block subtract each other, produce a residual block and produce one group of conversion coefficient X after quantizing through VCL, after reordering again through entropy coding, form a compressed code flow with required some side informations (as predictive mode quantization parameter, motion vector etc.) of decoding, use for transmission and storage through NAL (network self-adapting layer).For the reference picture of further prediction usefulness is provided, encoder design the function of reconstructed image.Therefore the data and the predicted value addition that residual image are obtained after inverse quantization, inverse transformation obtain the frame of non-filtered.In order to remove the noise that produces in the coding and decoding loop, improve the picture quality of reference frame, thereby improve the compressed image performance, be provided with a loop filter, filtered output is reconstructed image, can be used as reference picture.

By the compression bit stream after compression of NAL output of encoder.One group of conversion coefficient X after the entropy decoding obtains quantizing through block integration, motion compensation, ordering, obtains residual error data again.The header that utilization decodes from this bit stream, decoder just produce a prediction piece PRED, and it comprises that with encoder the original PRED among the ICL is identical.Behind PRED and residual error data addition that this decoder produces, just produce the frame of non-filtered, after filtering, just obtain the last decoding output image of rebuilding at last again.

The characteristics of KMVC encryption algorithm

The KMVC encryption algorithm is to adopt dct transform to encode to add the differential coding of DPCM, i.e. hybrid coding structure.Simultaneously, the KMVC encryption algorithm has been introduced new coded system under the framework of hybrid coding, improved code efficiency, and more closing to reality is used.The KMVC encryption algorithm does not have loaded down with trivial details option, and it has than better compression performance H.263++, has the ability that adapts to multiple channel again.The application target of KMVC encryption algorithm is extensive, can satisfy the Video Applications of various different rates, different occasions, has the disposal ability of anti-preferably error code and anti-packet loss.KMVC encryption algorithm coding basic structure is improved in a lot of links, now is listed below:

1, Xian Jin ICL

Take the lead in having adopted the ICL unit in the KMVC encryption algorithm, realize that for block long-time static in the picture once transmission is repeatedly used.And every two field picture is carried out dividing processing, the block that changes in the picture is carried out coding transmission, reduced the coding transmission amount, realize the low-loss compression of picture.

2, multiple better estimation:

The one, high accuracy is estimated

In the KMVC encryption algorithm, then further adopt 1/4 pixel even 1/8 pixel motion estimation.The displacement that is real motion vector may be a base unit with 1/4 even 1/8 pixel.Obviously, the precision of motion vector displacement is high more, and then the interframe residual error is more little, and transmission code rate is low more, and promptly compression ratio is high more.In the KMVC encryption algorithm, adopted the interpolation of 6 rank FIR filters to obtain the value of 1/2 location of pixels.After 1/2 pixel value obtains, 1/4 pixel value can obtain by linear interpolation, for the video format of 4:1:1, therefore 1/4 pixel precision of luminance signal need carry out the interpolative operation of 1/8 pixel to carrier chrominance signal corresponding to 1/8 pixel motion vector of chrominance section.In theory, if being doubled, the precision of motion compensation (for example brings up to 1/2 pixel precision) from whole pixel precision, the coding gain that 0.3bit/Sample can be arranged, but actual verification is found after the motion vector precision surpasses 1/8 pixel, system has not just obviously gained basically, therefore, and in the KMVC encryption algorithm, only adopt the motion vector pattern of 1/4 pixel precision, rather than adopted the precision of 1/8 pixel.

The 2nd, many macro-block partition modes are estimated

In the predictive mode of KMVC encryption algorithm, a macro block (MB) can be divided into the size of 7 kinds of different modes, this multimodal flexible, trickle macroblock partitions, the shape that more suits the actual motion object in the image, so, in each macro block, can include 1,2,4,8 or 16 motion vector.

The 3rd, the multi-parameter frame is estimated

In the KMVC encryption algorithm, can adopt the estimation of a plurality of parameter frames, promptly in the buffer memory of encoder, there are a plurality of parameter frames that just encoded, encoder from wherein select one provide better coding effect as parameter frame, and point out it is that a frame is used to prediction, so just can obtain than only using the better coding effect of a frame that has just encoded as predictive frame.

3, more accurate infra-frame prediction

In the KMVC encryption algorithm, the different weights of all available 17 pixels of before having encoded the most approaching of each pixel in each 4 * 4 and carry out infra-frame prediction.

4, unified VLC

About entropy coding two kinds of methods are arranged in the KMVC encryption algorithm.

Unified VLC (being UVLC:Universal VLC).UVLC uses an identical code table to encode, and decoder is easy to discern the prefix of code word, and UVLC can obtain synchronously heavy when the bit mistake takes place fast.

Hold adaptive binary arithmetic coding:

(CABAC:Context Adaptive Binary Arithmetic Coding) its coding efficiency is better slightly than UVLC, but complexity is higher.

Three, summary of the invention:

The object of the present invention is to provide a kind of based on the video monitoring information interaction system under the Symbian platform, its whole system is formed by connecting by the mobile communication circuit by several wireless monitor points, monitor portion modularization, mobile, data are by encrypting circuit and protocol transmission, system enhancement set up the flexibility and the extendibility of system, increase the reliability of system, had high cost performance.

For achieving the above object, the technical solution used in the present invention is:

A kind of based on the video monitoring information interaction system under the Symbian platform, it is characterized in that: system is formed by connecting by the mobile communication circuit by several wireless monitor points, with all control points by the information of video collector collection by the encoder server coding become can, the wireless signal of Symbian platform transmission, send to portable terminal by the pilot signal forwarding server again, the workflow of system is:

1. initialization;

2. video image acquisition;

3. use KMVC encryption algorithm agreement to encode;

4. transmit: transmit by wireless mode;

5. support the mobile terminal request server data of Symbian;

6. data are carried out buffered;

7. use KMVC encryption algorithm agreement to decode;

8.YUV the output of the image of form is play;

9. request next frame view data repeating step (5) arrives (8).

The collection point is used the wireless monitor collecting device of conventional monitor video head or band Symbian coding chip support CDMA 1X to carry out the collection of image and is collected, for the image that the monitor video head of routine is gathered, will transfer to video acquisition/encoder server and use the KMVC encryption algorithm to encode.

Compared with prior art, the advantage and the effect that have of the present invention is as follows:

1), the present invention to greatest extent utilized the chip functions of existing Symbian platform mobile phone terminal, radio communication service is provided, computing technique is combined with telephony.

2), by adopting improved KMVC intraframe predictive coding algorithm to overcome the technical barrier of the smooth transmission of video under the common wireless communications environment on the Symbian platform, realized high quality images transmission under the low code stream;

3), the interframe timestamp technology by uniqueness overcome the unstable shortcoming of easily interrupting of wireless network, strengthens the processing to error code and packet loss, realized the video flowing transmission of adaptive network situation, solved network and interrupted the image transmission disruption brought;

4), the motion compensation technique that adopts 1/4 sampling precision on the Symbian platform and interframe movement Predicting Technique than the realization under the narrow bandwidth image compression and the transmission of better quality.

5), with cordless communication network as carrying platform, with video monitoring, information interaction is core, integrate functions such as video acquisition, high-definition image transmission, and carry out the application system of wireless information interaction by various auxiliary systems, this system in conjunction with existing communication operator mobile communications network cover a wide range, ubiquitous characteristic, can utilize various advanced technology means to solve information problem accurate and that in time issue, and make a response fast.

Four, description of drawings:

Fig. 1 is the composition schematic diagram of KMVC encryption algorithm encoder;

Fig. 2 is the composition schematic diagram of KMVC encryption algorithm decoder;

Fig. 3 is the prediction diagram under the PB frame pattern;

Fig. 4 is forward prediction and bi-directional predicted figure in the B piece.

Five, embodiment:

The present invention becomes the information of all control points coding can be at the wireless signal of Symbian platform transmission, and can monitor in real time at portable terminal, system comprises support video acquisition/encoder server, the pilot signal forwarding server, portable terminal, its workflow is:

(1) initialization;

(2) video image acquisition;

(3) use KMVC encryption algorithm agreement to encode;

(4) transmit: transmit by wireless mode (CDMA 1X);

(5) the mobile terminal request server data of support Symbian;

(6) data are carried out buffered;

(7) use KMVC encryption algorithm agreement to decode;

(8) output of the image of yuv format is play

(9) request next frame view data repeating step (5) to (8).

The collection point can be used the wireless monitor collecting device of conventional monitor video head or band Symbian coding chip support CDMA1X to carry out the collection of image and be collected, for the image that the monitor video head of routine is gathered, will transfer to video acquisition/encoder server and use KMVC encryption algorithm agreement to encode.

The composition of KMVC encryption algorithm encoder as shown in Figure 1, the composition of KMVC encryption algorithm decoder is as shown in Figure 2.

One, KMVC encryption algorithm key property

1.KMVC encryption algorithm C coding method relatively in the past is an example with MPEG-2, improves code efficiency aspect the picture material prediction, the main feature of improving picture quality is as follows:

● dynamic and static blocks is divided: the KMVC encryption algorithm carries out the division of self adaptation block with static state in the different frame image and dynamic image, selects to produce the block of main variation, carries out encoding compression, has reduced the workload of compressed encoding, improves code efficiency.

● variable size block motion compensation: select motion compensation size and shape more flexible than former standard, minimum brightness motion compensation block may diminish to 4 * 4.

● 1/4 sampling precision motion compensation: maximum 1/2 precision motion compensation of former standard, 1/4 sampling precision motion compensation first appears at the senior class part of MPEG-4 second portion, but the KMVC encryption algorithm has significantly reduced the complexity that interpolation is handled.

● motion vector can be crossed over image boundary: in the former standard, motion vector is limited in the inside of coded reference image.

● the motion compensation of many reference pictures: in the standard before MPEG-2 reaches, the P frame only uses a frame, and the B frame only uses two two field pictures to predict.The KMVC encryption algorithm uses high vision to select technology, can predict with a large amount of image of having encoded in the past and be retained in buffering area, has improved code efficiency greatly.

● weight estimation: the KMVC encryption algorithm is employed new technology, and allows the weighted motion compensated prediction and is offset a certain amount of.This technology greatly improves code efficiency in the scene of being fade-in fade-out, and this technology also can be used for other multiple uses.

● improve " skipping " and " directly " motion and infer: in the former standard, can not there be motion in predictive-coded picture " skipping " district.When coding had the image of global motion, this restriction was very harmful.The KMVC encryption algorithm adopts estimation method to the motion in " skipping " district.B two field picture to two predictions adopts senior motion forecast method, is called " directly " motion compensation, further improves code efficiency.

● the intraframe coding straight space is predicted: being extrapolated in the coded image edge is applied to the prediction of present frame intra-coded picture.

● the circulation block elimination effect filter: there is blocking effect in block-based video coding in image, is mainly derived from prediction and remaining coding.Self adaptation block-eliminating effect filtering technology is very famous technology, can effectively eliminate blocking effect, improves the subjectivity and the objective quality of video.

2. except that improving Forecasting Methodology, other characteristics of improving code efficiency are as follows:

● the fritter conversion: the piece of former canonical transformation all is 8 * 8, and the KMVC encryption algorithm mainly uses 4 * 4 conversion, makes encoder represent that signal local adaptation property is better, is more suitable for predictive coding, reduces " bell " effect.Image boundary needs the fritter conversion in addition.

● the conversion of classification piece: the KMVC encryption algorithm uses the fritter conversion usually, but some signal comprises enough correlations, requires to represent with bulk, and the KMVC encryption algorithm has dual mode to realize.Low-frequency chrominance signal can be with 8 * 8; To intraframe coding, can use special type of coding, low-frequency brightness signal can be with 16 * 16.

● the long conversion of short word: the conversion that standard is used before all requires 32 bit arithmetics, and the KMVC encryption algorithm only uses 16 bit arithmetics.

● mate inverse transformation fully: the error that has certain tolerance limit before all between standard inverse transformation and the conversion, therefore, each decoder outputting video signal is all inequality, produces little drift, the final quality that influences image, KMVC encryption algorithm have realized coupling fully.

● based on contextual entropy coding: the KMVC encryption algorithm uses two kinds of entropy coding methods, CAVLC (Variable Length Code of context-adaptive) and CABAC (context adaptive binary arithmetic coding), and two kinds all are based on contextual entropy coding.

3.KMVC encryption algorithm has powerful error correction and variety of network environments operating flexibility, key property is as follows:

● the parameter set structure: KMVC encryption algorithm parameter set structural design powerful, effectively transmit header information.In the former standard, if several key messages are lost, possible decoder produces serious decoding error.The KMVC encryption algorithm adopts very flexible, special mode, and the separate processes key message can reliably transmit under various environment.

● NAL unit syntactic structure: each syntactic structure in the KMVC encryption algorithm is placed in the unit that is called NAL, and former standard adopts mandatory specific bit stream interface.NAL unit syntactic structure allows customization very freely, almost is fit to all network interfaces.

● flexibly as the bar size: in MPEG-2, stipulated strict picture bar structure, the header data amount is big, reduces forecasting efficiency, and code efficiency is low.Can adopt very flexibly as the bar size at the KMVC encryption algorithm.

● flexible macro-block ordering (FMO): the KMVC encryption algorithm can be called image area again with image division for picture bar group, and each can independently be decoded as bar.FMO has very strong anti-loss of data ability by the relation between the managing image district.

● sort as bar arbitrarily: because each almost can independently be decoded as bar, so can send in any order and receive as bar, in using in real time, can improve time-delay characteristics end to end, be particularly suitable for reception order and sending order can not corresponding network in, as the application of use INTERNET procotol.

● redundant image: for improving the ability of anti-loss of data, comprise a kind of new ability in the design of KMVC encryption algorithm, the redundancy that allows encoder to send image area is represented, still can be correctly decoded when the principal representation of image area is lost.

● data are divided: the importance difference of the coded message in the video flowing, some information (as motion vector, information of forecasting etc.) is more even more important than other information.The KMVC encryption algorithm can will be divided into 3 parts as the bar grammer according to each category as the bar syntactic element, separately transmit.

Two, network layer

The form of NAL standard video data mainly provides header information, to be fit to the transmission and the storage of various medium.NAL supports diverse network, comprising:

● the real-time wired and wireless Internet service of any use RTP/IP agreement.

● as MP4 file storage and multimedia information file service.

● the MPEG-2 system.

● other nets.

1.NAL unit

The video flowing of coding is organized into the NAL unit, and video data is placed in the network element and transmits, and individual network element comprises the integral words joint, and first byte is a header information, the data type of indication NAL unit, and all the other are payloads.

Payload data and emulation prevention byte are done interleaving treatment, and emulation prevention byte is the particular value byte, prevents the accidental sync byte pattern that occurs in payload.

NAL stipulates a kind of general form, both has been fit to towards the bag transmission, also is fit to stream and transmits.In fact, the bag transmission is identical with the mode of flow transmission, and difference is that the flow transmission front has increased a start code prefix.

2.NAL the application of unit in byte stream

Transmission NAL will rely on data identification NAL elementary boundary itself simultaneously as orderly successive byte or bit stream.In such application system, KMVC encryption algorithm normalized definition the byte stream form, each front, NAL unit increases the prefix of 3 bytes, i.e. sync byte.In bit stream was used, each image need increase an extra byte as boundary alignment.Also have a kind of optional attribute, in byte stream, increase additional data, be used as and expand the transmission data volume, can realize quick boundary alignment, recover synchronously.

3.NAL the application of unit in transmitting towards bag

In bag transportation protocol system, comprise packet boundary identification byte at similar Internet/RTP in the pack arrangement, in this case, do not need sync byte.

4.VCL and the NAL unit of non-VCL

The NAL unit is divided into two kinds of VCL and non-VCL, VCL NAL unit comprises the video image sample information, non-VCL comprises various relevant additional informations, for example parameter set (header information is applied to a large amount of VCL NAL unit), the additional information that improves performance, timing information etc.

5. parameter set

Parameter set is the information that seldom changes, and is used for the decoding of lot of V CL NAL unit, is divided into two types:

● sequence parameter set acts on a string continuous video image, i.e. video sequence.

● picture parameter set acts on one or more other images in the video sequence.

Sequence and picture parameter set mechanism, reduced the transmission of repetition parameter, each VCL NAL unit comprises a sign, point to relevant picture parameter set, each picture parameter set comprises a sign, points to the content of relevant sequence parameter set, therefore, only, quote a large amount of parameters, significantly reduce the information that each VCL NAL unit repeats to transmit with the pointer information of minority.

Sequence and picture parameter set can send in the past in transmission VCL NAL unit, and repeated to transmit, and improved error correcting capability greatly.Sequence and picture parameter set can also can be used more reliable other " band is outer " channel transfers at " in the band ".

6. memory cell

The NAL unit of one group of specified format is called memory cell, the corresponding image of each memory cell.Each memory cell comprises one group of VCL NAL unit, forms a primary coded picture, and VCL NAL unit is made up of as bar expression video image sampling.The memory cell front can add a prefix, the boundary memory cell, and additional enhancing information (SEI) (as image timing information) also can be placed on the front of primary coded picture.The VCL NAL unit that adds after the primary coded picture, the redundancy that comprises same image is represented, is called redundant coded picture, when primary coded picture loss of data or damage, the decoding of available redundancy coded image.

7. encoded video sequence

An encoded video sequence is made up of a string continuous memory cell, uses same sequence parameter set.Each video sequence can independently be decoded.Beginning of coded sequence is instant refresh of memory cells (IDR).IDR is an I two field picture, and the image of expression back need not be with reference to former image.A NAL unit stream can comprise one or more encoded video sequence.

Three, video coding layer

Adopt transition coding, the hybrid coding of usage space and time prediction.Fig. 3 is the block diagram of the video coding layer of a macro block.Image division becomes piece in a word, first image of a sequence, it is random access point, the typical case is intraframe coding, the prediction of each sampling only utilizes the adjacent sampling in space of having encoded in the frame in the frame, select which neighbouring sample to predict, and how to predict, these additional informations must be sent to the decoder Synchronous Processing simultaneously.Image between the random access point uses interframe encode.

In order to realize the prediction of next piece or next image, encoder comprises a decoder, quantization transform coefficient is carried out inverse quantization identical with decoder decode and inverse transformation process, derive the decoding prediction residue, remaining and the prediction addition of decoding, the result delivers to block elimination effect filter, produces decoded video output.

Image, frame and

An encoded video sequence is made up of continuous coded image, and coded image can be a whole two field picture, also can be a field picture.KMVC encryption algorithm coding is based on the method for expressing of geometrical concept, rather than based on notion regularly.

2.YcbCr chrominance space and 4:2:0 sampling

Human vision property is according to the brightness and the chrominance information difference perception world.The transmission of video can utilize this characteristic to reduce chrominance information and transmit.4:2:0 sampling structure, 8 bit accuracy that the KMVC encryption algorithm adopts at present, the suggestion of high accuracy color and higher bit precision just under discussion.

3. macroblock partitions

Each video frame image or field can be divided into the macro block of fixed size, and macro block is the basic modular unit of decoding, normally the rectangle zone of one 16 * 16 luminance pixel and two 8 * 8 chrominance component pixels.The brightness of all macro blocks and chroma samples are predicted in the space or on the time, and prediction residue is carried out transition coding.

4. as bar and picture bar group

Being made up of macro block as bar, is the subclass of image as bar, comprises picture parameter set, and syntactic element can be analyzed, and image can be by independent decoding.According to the order of macroblock map table regulation, in bit stream, arrange the transmission sequence of macro block, rather than press raster scan order.

Utilization is as the notion of bar group, and the KMVC encryption algorithm is supported flexible macro-block order characteristic (FMO).FMO has changed the mode of image division for picture bar and macro block.Each is a plurality of set of macroblocks as the bar group, and to the table definition of hinting obliquely at as the bar group, this is hinted obliquely at table and specifies in picture parameter set by macro block.Each macro block has a picture bar group id number, and all number constitute macro block to hinting obliquely at table as the bar group as the bar group id.Each is made up of one or more picture bars as the bar group, is a macroblock sequence as bar therefore, and the macro block in the same picture bar group is handled by raster order.

Utilize FMO, image can be divided into many macro block scanning patterns, the pattern that for example interweaves, intersperses pattern, and one or more prospect picture bar groups, residue be as the bar group, or the pattern pattern is hinted obliquely at etc.Each send as bar component supplementary biography, shown in back two kinds, left side macro block to as the bar group to hint obliquely at proof very useful in the coding of paying close attention to local type is used.The right macro block is fit to application such as security type conference system to the proof of hinting obliquely at as the bar group.

No matter whether use FMO, the KMVC encryption algorithm is supported 5 kinds of picture bar type of codings:

I picture bar: the simplest type of coding, all macro blocks be other image in the reference video sequence not.

P picture bar: except I picture bar type of coding, the part macro block of P picture bar can utilize inter prediction, and each prediction piece can use a motion compensated prediction signal at the most.

B picture bar: except P picture bar type of coding, the part macro block of B picture bar can utilize inter prediction, and each prediction piece can use two motion compensated prediction signals.

More than three kinds similar to former standard, mainly be the reference picture difference, all the other two kinds is SP (switch P) SI (switching I) as the bar type, is new type, is used for effectively switching between different bit rate encoding code streams.

SP picture bar: be called and switch P picture bar, can between the different coding image, switch effectively.

SI picture bar: be called and switch I picture bar, allow the macro block of SP picture bar to mate fully, reach at random reading of data and decode and recover wrong purpose.

5. the encoding-decoding process of macro block

The brightness of all macro blocks and chroma samples will be carried out the prediction of space or time, remnants to prediction carry out transition coding, in order to realize transition coding, the prediction residue of each color component will be divided into littler 4 * 4 again, every is utilized integer transform, conversion coefficient is quantized, and is entropy coding at last.

6. adaptive frame/field encoding operation

In interlaced scanned frames, when the object that moves or video camera move, and to compare line by line, two adjacent lines tendencies reduce the correlation of statistics, and this situation should compression be more effective respectively than every.In order to reach high efficiency, the KMVC encryption algorithm has following possibility when coded frame:

● in conjunction with whole frame of two one-tenth,, be called frame pattern as the frame coding.

● encode respectively, be called field mode for two.

● in conjunction with whole frame of two one-tenth, compress as frame.In when coding, divide vertical two adjacent macro blocks to become two field macro blocks to or the frame macro block, encode again.

Every two field picture one of can adaptively selected 3 kinds of patterns be encoded.Select to be called image adaptive frame/field coding (PAFF) between preceding two kinds, when a frame was encoded as two, every was divided into macro block, and coded system is very similar to frame coding mode, mainly contains following exception:

● motion compensation reference field, rather than reference frame.

● " zig-zag type " scan mode difference of conversion coefficient.

● the intensity of macroblock level edge de-blocking filter is not selected " by force " for use, because a row spatially is the capable distance of twice frame.

When development KMVC encryption algorithm standard, it is reported that adopt ITU-601 resolution, the PAFF coding techniques is compared with the frame coding can reduce code check 16% to 20%.

If image is mixed by motor area and non-motor area and forms, non-motor area is efficient coding method with frame pattern, motor area with field mode.Therefore each vertical macroblocks can independently be selected coding (frame/field) pattern to (16 * 32).This coding selects to be called macro block adaptive frame/field coding (MBAFF).Right for the frame pattern macro block, it is capable that each macro block comprises frame, right for the field mode macro block, and top macroblock comprises a top row, and the bottom macro block comprises field, end row.

Handle right each macro block and a frame PAFF pattern similarity of macro block, yet, because field/frame macro block takes place in the MBAFF frame, need to revise method as following purposes to mixing:

● zig-zag type scanning.

● motion-vector prediction.

● prediction on intra-frame prediction mode.

● infra-frame prediction frame sampling precision.

● block elimination effect filter.

● the entropy coding of context model.

Main thought is to keep many Space Consistency as far as possible, and the adjacent standard in the space of MBAFF frame is quite complicated, and the space of talking about below is adjacent all to be meant non-MBAFF frame.

Another important difference of MBAFF and PAFF is: use the MBAFF method, field can not use macro block in another of same frame as motion compensation reference.Like this, PAFF is more effective than the MBAFF coding sometimes, particularly under situations such as quick global motion, conversion scene, image refreshing.

Adopt ITU-601 resolution, the MBAFF coding techniques is compared than PAFF can reduce code check 14% to 16%.

7. infra-frame prediction

According to picture bar type of coding, each macro block can be selected one of several type of codings.All support the two-stage intraframe coding as the bar type, are called INTRA-4 * 4 and INTRA-16 * 16.INTRA-4 * 4 patterns are fit to the represent images detail section based on predicting each 4 * 4 luminance block respectively.And INTRA-16 * 16 patterns are predicted whole 16 * 16 luminance block, are fit to the smoothed image district.To these two kinds of luma prediction types, colourity is predicted separately in addition.Another kind as INTRA-4 * 4 and INTRA-16 * 16 is selected, and the I_PCM type of coding allows encoder simply to skip prediction and transition coding process, directly sends sampled value.I_PCM allows encoder accurately to represent sampling, some perverted images of ordinary representation, and obviously do not increase data volume.

8. inter prediction

● p picture bar inter prediction

Except the intra-frame macro block type of coding, P picture bar macro block uses multiple type of prediction, i.e. the motion compensation encoding type.Motion is for convenience described, and each P type macro block is corresponding to the macroblock partitions of the fixed size of appointment.The division size of luminance block is: 16 * 16,16 * 8,8 * 16,8 * 8.

Grammer allows motion vector to cross over image boundary, and this situation reference frame needs to extrapolate the outer sampled value of image with extrapolation.Utilize the average or direction prediction of contiguous block, use the differential coding calculating kinematical vector.

Many image motion compensation predictions supported in grammer, and a plurality of images of coding all can be used as motion compensation reference in the past, and each motion compensation need be indicated the index of reference picture.Same reference picture index is used in one 8 * 8 motion compensation less than 8 * 8 that are divided into, and this functional requirement Code And Decode utensil has the multiple image buffer, and decoder is synchronous by management control operation information and encoder in the bit stream.

Except the motion compensation macro block mode, the P macro block can use the P_Skip pattern, and this pattern do not have quantized prediction error, without translatory movement vector and reference key parameter.Signal reconstruction and P_16 * 16 macro block (mb) type prediction signal are similar, and the image that adopts multi frame buffer district index 0 is as the reference image.The motion vector and the prediction of 16 * 16 block motion vectors of rebuilding the P_Skip macro block are similar.The P_Skip type of coding is fit to not have to change or fix the zone of motion, as pan etc., high compression ratio is arranged.

● the inter prediction of B picture bar

The notion of the relative P picture of B picture bar bar and former standard class seemingly, but some difference, B picture bar can be used as the reference picture of motion compensated prediction, B picture bar can compute weighted to two compensation prediction values.

9. transform and quantization

Similar with former video encoding standard, the KMVC encryption algorithm utilizes the prediction residue transition coding.Yet the conversion of KMVC encryption algorithm applies on 4 * 4, without discrete cosine transform (DCT), adopts the integer transform with the discrete cosine transform similar characteristic.Because the use integer arithmetic, so inverse transformation does not have error.

10. entropy coding

The KMVC encryption algorithm is supported two kinds of entropy coding methods, and the simplest entropy coding method is: to all syntactic elements, except quantization parameter, use single unlimited extendible codeword table.Needn't be the VLC table of a special use of each syntactic element design like this, only need be according to the statistical property of data, the single codeword table is hinted obliquely in customization.Use the exp-Golomb sign indicating number to set up single code table, have very simple, well-regulated decoding characteristics.

In order effectively to transmit quantized transform coefficients, CAVLC (Variable Length Code of context-adaptive) is an efficient and simple method.In this scheme, can switch according to the syntactic element that has transmitted for the VLC code table of various syntactic elements.Because VLC table design coupling corresponding condition is added up, improved the performance of entropy coding.

Arithmetic coding engine in the KMVC encryption algorithm does not use multiply operation with the estimation of probability correlation etc., and is to use simple displacement and search operation, compares with the CAVLC operation, and the CABAC typical case reduces code check 5%～15%.At interlaced scanning video, effect is more obvious.

11. remove the piece recursive filter

Characteristics based on the compression algorithm of block encoding are to produce visible block structure once in a while, because the ratio of precision at piece edge is inner poor, blocking effect is the common artificial flaw of present compression algorithm.The KMVC encryption algorithm has defined a self adaptation recursive filter, and the intensity of filtering is by several syntactic element controls.

The basic thought of filtering is: the absolute difference at if block edge is relatively big, and the possibility that the artificial flaw of piece occurs is just very big, therefore need carry out respective handling.Yet if the difference amplitude is very big, the error of coded quantization process can not explain that the edge has reflected the actual sample value of source images probably, need not handle.Through Filtering Processing, reduced blocking effect, and the quality of image is unaffected substantially, so subjective quality improves greatly.If not filtering, same subjective quality need have more 5% ~ 10% code check.

12. hypothetical reference decoder model

The benefit of standard is to guarantee that all standard compliant decoders can decode consistent video quality.Realize standard, the description that encryption algorithm only is provided is not enough, in real-time system, specifies and how to present the code stream bit and give decoder, and how the image of decoding shifts out decoder etc. all is very important.Realize this function, must specify and import/go out buffer model and exploitation and the receiver module of realizing haveing nothing to do.Such receiver module is called hypothetical reference model (HRD).Encoder does not allow to produce the bit stream that the hypothetical reference model can not be decoded.If therefore all receivers are all simulated the behavior of hypothetical reference model, can guarantee that the code stream that all decoders solve has good consistency.

Four, class and level and application thereof

Class and the level

A class and a level indicator chalaza, these corresponding points are used to realize the interoperability between the various standard application.Class definition one group coding instrument and algorithm is used to produce conforming bit stream, and level limits the part key parameter of bit stream.

The class that all decoders meet appointment must be supported all characteristics of this class definition, and encoder needn't require to use any specific feature set of such support, but conforming bit stream must be provided, even support such decoder can realize decoding.

The KMVC encryption algorithm has defined 3 classes: base class, main classes and expansion class.

Base class is supported all KMVC encryption algorithm characteristics except that following two groups of characteristics:

(1) B picture bar, weight estimation, CABAC, a coding, image or macro block self adaptation between the coding of frame field is switched.

(2) PI/SI picture bar and cut apart as the bar data.

Main classes is supported first group of characteristic, but main classes is not supported FMO, ASO and redundant image characteristic, and expansion class support all KMVC encryption algorithm characteristics except that CABAC have comprised the class that some these code stream requirement decoders of sign instruction decoding are supported in the sequential parameter.

At the KMVC encryption algorithm, all classes are used identical level definition, but when realizing individually, and it is grade incomplete same that the class of possible each support is supported.The KMVC encryption algorithm has defined 15 levels.

2. the use field of new standard class

The high compression efficiency of KMVC encryption algorithm has expanded the existing application field, comprises following field at least:

(1) interactive video service below the speed 1Mbps, is hanged down delay usually.

Main application is as follows:

(1) interactive video service utilizes the circuit switching based on the ISDN video conference;

(2) entertainment video is used, 50Kbps～80Kbps code check, 0.5 to 2 second medium time delay.

Main application is as follows:

Radio and television such as wired, satellite, ground, DSL;

Video request program by various medium.

(3) streaming media service, 30kbps is to 1.5Mbps, the time delay more than 2 seconds, these are used and mainly utilize base class or expansion class.Wired or wireless operating position is different, mainly uses as follows:

GPP stream utilizes the IP/RTP transmission, and RTSP does the session setting, and the expansion part of 3GPP standard may only be used base class;

Wired INTERNET stream utilizes the IP/RTP transmission, and RTSP does the session setting.

(4) other services mainly are low code checks, with document transfer mode, do not consider the application of time delay.

Claims (2)

1, a kind of based on the video monitoring information interaction system under the Symbian platform, it is characterized in that: system is formed by connecting by the mobile communication circuit by several wireless monitor points, with all control points by the information of video collector collection by the encoder server coding become can, the wireless signal of Symbian platform transmission, send to portable terminal by the pilot signal forwarding server again, the workflow of system is:

(1) initialization;

(2) video image acquisition;

(3) use KMVC encryption algorithm agreement to encode;

(4) transmit: transmit by wireless mode (CDMA1X);

(5) the mobile terminal request server data of support Symbian;

(6) data are carried out buffered;

(7) use KMVC encryption algorithm agreement to decode;

(8) output of the image of yuv format is play;

(9) request next frame view data repeating step (5) to (8).

2, according to claim 1 a kind of based on the video monitoring information interaction system under the Symbian platform, it is characterized in that: the collection point is used the wireless monitor collecting device of conventional monitor video head or band Symbian coding chip support CDMA1X to carry out the collection of image and is collected, for the image that the monitor video head of routine is gathered, will transfer to video acquisition/encoder server and use the KMVC encryption algorithm to encode.

Images