How do video compression standards affect the encoding process? Katharina Geutebruck, managing director of Geutebruck GmbH which manufactures and develops intelligent video security solutions, explains.
It's probably fair to say that few buyers of video security systems are experts in video encoding. And why should they be? It's enough to
|The encoding process is defined by the developer's expertise rather than the video compression standard|
know that data need to be compressed and that there are different ways of doing it. What's important is to be able to recognise good video encoding when you see it.
Whenever a new video standard is introduced, a ripple of uncertainty goes through the security community, and some manufacturers see an easy marketing opportunity among disconcerted buyers. So, if you are currently thinking: "Should I be demanding H.264 or the latest standard?", "It's new, so is it better than previous ones?" or "Does that mean other digital video standards will soon be obsolete?"- then you should read on.
Video compression standard - a set of tools
The first thing to realise about video encoding is that contrary to what you might expect, a standard does not define the specific process which is used to encode the data. And, just because two systems both use the same standard, it does not mean that the resulting video output is of equal quality or similar utility - in fact, it varies very widely.
Just because two systems both use the same standard, it does not mean that the resulting video output is of equal quality or similar utility
Instead, it is much more helpful to think of each standard as a set of tools in a toolbox. Just as a carpenter might use a saw, a plane, a hammer, drills and chisels from his toolbox to make a piece of furniture, the developer uses tools from the video standard toolbox to create an encoding process.
The carpenter is free to decide what he is going to make, which tools to use, how and in what order. Similarly, the developer can encode for multi-media, mobile phone, CCTV or other applications; he can choose which of the video compression tools to use and how to use them. And, just as the utility and quality of the finished piece of furniture depends on the design skills and craftsmanship of the carpenter, so the performance of a video encoding process depends on the specialist knowledge and expertise of the developer.
Differences between video compression standards
|New video compressions standards may not offer the same encoding speed as former versions|
Each new video standard has a different selection of tools. Some tools are the same as in an earlier standard, or are upgraded versions of them. This might equate to the carpenter acquiring a new diamond toothed saw and a power drill. They enable him to work faster, but they rely on the same creativity and skill. Some of the new video compression (or video encoding) tools are intended for specific purposes, so whether they are useful or not depends on the product in view. Just as the carpenter is unlikely to see a soldering iron as a positive addition to his toolbox, the developer may not find mobile phone imaging tools much help in creating high-quality CCTV images. And a tool which delivers greater compression but also reduces image quality and processing speed may be better left in the box!
There are two important lessons that we can draw from all this:
1. The performance of a CCTV surveillance system depends not on which video standard is used but on the skills of the developer.
2. There are no ‘better' video standards - only ‘better' implementations.
Evaluating video standard selection process
So, if you have not mastered the intricacies of H.264 or any other video compression standard, there is no need to feel inadequate. Even for the encoding experts among us, judging a good implementation is about comparing performance in action and about using experience and professional judgement.
Even for the encoding experts among us, judging a good implementation is about comparing performance in action
It's about defining what quality criteria are important for the specific situations and tasks in your video surveillance application, then getting suppliers to demonstrate how their systems perform. How many live screens can they support? Do they maintain clear sharp pictures when there is a lot of movement in the scene? What bandwidth management features are supported? Is there a significant delay or is the encoding fast enough to support normal speed-dome control? Don't take anyone's word for it, ask to see it done - there is no substitute for real life demonstrations with footage and situations similar to your own.