The Institution of Engineering and Technology, 2009. — 300 p.In the early 1990s, while MPEG was still working towards its first international standard, I gained my first experience of motion-compensated video compression as a development engineer in the Advanced Product Division of National Transcommunications. I was working on a research project to develop a real-time, motion-compensated video compression codec for television transmission. My first impression of such algorithms was that they would be interesting as research projects but that they were far too complex to be used for real-time commercial television applications. Apart from DCT and motion estimation chips, no dedicated ASICs were available to implement the algorithms. Most of the algorithms had to be implemented using earlier versions of FPGAs, comprising a few hundred logic blocks and flip-flops. Nevertheless, in September 1992, even before MPEG-1 had been promoted to International Standard, we had our first real-time SDTV MPEG-1 codec working. It consisted of a 12U encoder and a 3U decoder, and yes, it was a commercial success. Fast forward to today, one can get an HDTV MPEG-4 (AVC) encoder in a single chip the size of a thumb nail. Video compression has moved not just into television production, broadcast and telecommunications equipment but also into surveillance equipment and a large range of consumer devices: from camcorders, PCs and PVRs to mobile phones. For quite a few years I have been presenting introductory lectures on MPEG-2 video and audio compression and, more recently, also on HDTV and MPEG-4 (AVC). The courses are organised by the Continuing Education Department of Surrey University and are aimed at engineers with some knowledge of broadcast television. One of the questions I am asked on a regular basis is: ‘Are there any books on this subject?’ The answer is yes, there are many books on video compression, but most of them are quite mathematical and targeted towards engineers who are developing compression algorithms and equipment. Other books give a brief introduction to digital video processing and compression, but they do not cover the wider issues of video compression systems. This gave me the idea that I could write one myself, aimed at users of compression equipment rather than engineers developing compression encoders.Introduction Digital video Picture quality assessment Compression principles MPEG video compression standards Non-MPEG compression algorithms Motion estimation Pre-processing High definition television (HDTV) Compression for mobile devices MPEG decoders and post-processing Statistical multiplexing Compression for contribution and distribution Concatenation and transcoding Bit-stream processing Concluding remarksA: Test sequences referred to in this book B: RGB/YUV conversion C: Definition of PSNR D: Discrete cosine transform (DCT) and inverse DCT E: Introduction to wavelet theory F: Comparison between phase correlation and cross-correlation G: Polyphase filter design H: Expected error propagation time I: Derivation of the bit-rate demand model
Чтобы скачать этот файл зарегистрируйтесь и/или войдите на сайт используя форму сверху.