The video image digitization is real-time encoding compression in video surveillance applications, which is completely different from the VCD and DVD productions, very large the video codec computing strength, with the original DSP video codec is almost impossible to achieve, compression algorithm optimization technology is not mature, must be dedicated ASIC With certain hardware accelerators, such as the movement of the estimated (ME) DCT module, format conversion, and variable length codecs (VLX) to achieve real-time encoding. To achieve due to the complexity of the image codec, rely on pure hardware too large, but it is almost impossible, requires the help of a powerful processor. This is a dedicated image compression chip also need to download microcode reasons, in the use of MPEG-1/2 encoder chip microcode exists, we say that the hardware compression that chip compression processor. It is also the emergence of compression chip, especially the emergence of the single-chip encoder chip, the degree of complexity of the monitoring system is a significant reduction in costs also decreased, which makes video digital monitoring it possible. General-purpose DSP video processing chip has not been developed to the real-time compression of the video signal, the dedicated video compression chip appear driven the video analog monitor to digital ip video surveillance. Video image processing chips (VIDEO-PROCESSOR) In addition to the processor, the chip also integrates some other IP (INTELLIGENTPROPERTY), modules, such as VIDEOIN VIDEOOUT AUDIOIN AUDIOOUT some image compression hardware acceleration module and so on. Such a chip is called SOC (SYSTEMONCHIP), i.e. monolithic systems. Essentially, the dedicated video compression chip and common video processing chips are all SOC of the differences is that the former with a fixed compression module, fixed processing and fixed micro-code, which requires the network camera system developers to video compression algorithm program. Dedicated video chips are designed for consumer applications, will be applied to monitor the area, there are still some other aspects of the problem: First, consumer chips use a single video encoding, will not require a multi-channel video compression at the same time, while the monitoring systems tend to be multi-channel video compression processing; followed by consumer chip continuous use is generally not a long time, and monitoring systems are often continuous use for a long time; consumer chip applications often need to do post-processing of the video image signal for editing, and the monitoring system is a real-time compression, must also be compressed in real time while some of the necessary processing, such as motion detection, time generator, logo, watermarks, etc., these treatments are pretreatment, these auxiliary functions pairs compressed product, appears to have a very important, these processes are carried out at pre-compression or compression; The more important difference is that, consumer applications less need for dynamic adjustment of coding parameters, and the strong need in the field of application of the monitor, dynamically adjust the coding parameters, such as frame rate, the image quality, the frame structure. Monitoring market size is too small compared with the consumer market, and the demand for ip video surveillance applications is diverse, it is impossible to use a chip strange to meet application needs. Chip manufacturers also impossible to monitor the market to develop a series of chip and chip development cycle is long, the cost price is too high, the market is in constant change, product replacement rate is very rapid, monitoring the market, waiting for integration the algorithm dedicated video compression chip mature product development and certainly not keep up with the development of the market. The once widely used in the monitoring field dedicated video compression chip as W99200AF the Z1510, SM2210, SM2288, IVT15 has not meet the requirements of the ip video surveillance market applications. With the development of IC technology advances and processor technology, the three different technologies of DSP, RISC, VLIW processors gradually unified, The multimedia processor integration of three types of processor technology. The SOC technology development is very fast and has been very strong, VIDEOPROCESSOR. Video processor integrated peripheral circuit the VI unit (VIDEOINUNIT), the the the VO unit (VIDEOOUTUNIT) AI unit (AUDIOINUNIT), the the AO unit (AUDIOOUTUNIT), internal some coprocessor, such as variable-length encoder / decoder VLX , motion prediction, ME, DMA controller, IIC controller, etc. the IP module integration greatly enhances the processing capacity of the VP, the VP is also the name of the DSP. With the continuous improvement of the frequency of the DSP processing power of growing, as well as video encoding technology continues to progress, the development of video compression software on the DSP, real-time video compression possible. Which to develop their own knowledge products, based on general-purpose DSP platform and high efficiency video compression software is very important. We have developed a common DSP capacity is not enough, we have developed a dedicated video compression chip capacity is not enough, but we can develop a general-purpose DSP efficient video compression software, and in accordance with the needs of the ip video surveillance market applications continue to add new features . Monitoring the market closed, the video compression standard software can not limit can absorb the latest achievements of the video encoding, in order to meet the market demand for the first goal to meet user requirements. Dedicated video compression chip is designed for the consumer, the emphasis is compatibility. Therefore, in order to meet the needs of the ip video surveillance market, video processor developed custom compression software has the following characteristics: it can not limit the adoption of the latest standard compression technology, in line with the characteristics of the monitor market; the RMON stressed network functionality, low network bandwidth limitations will lead to low bit rate requirements. Some applications on the network delay requirements is very low, so the video compression software for different applications to different functions depending on the application requirements, either low bit-rate to achieve low latency design goals; during compression, it can dynamically modify the coding parameters, such as image quality, frame rate, and a frame structure according to the application and environment changes; it can easily achieve the video signal is superimposed icon, strings, etc., such as the time information, location information; many monitoring situations need to compress voice, general consumer chip supports compressed sound, the sound of much greater bandwidth than voice bandwidth (voice within 3400Hz bandwidth of the sound from 0 ~ 20KHz). G.729a, G.729, G.723, G.723.1 voice compression can be used to compress stream only 8kbit / s, 6.3 kbit / s, 5.3Kbit / s compressed audio stream minimum of 16Kbit / s generally 32Kbit above, this little difference in storage, but is particularly important in the narrowband transmission.
|