Real words or Buzzwords?: H.264 and I frames, P-frames and B-frames – Part 3
This is the 51st article in the “Real Words or Buzzwords?” series from SecurityInfoWatch.com contributor Ray Bernard about how real words can become empty words and stifle technology progress. Access the other parts of this three-part article in the Related Box on the right.
This article continues the discussion about the H.264/H.265 MPEG video standards, including the manufacturer innovations in smart codecs (once a marketing buzzword but now a very meaningful term). The reasons for discussing this complex video topic are presented in the Part 1 and Part 2 articles.
Those reasons include that video compression impacts video system cost and performance. Video compression reduces the burden on video networks and lowers data storage requirements. At the same time, it increases the data processing requirements – computer memory size and processing power.
Additionally, techniques such as motion-based recording can create problems of their own, such as spiking the processing and storage requirements by five times or more when rain, snow or lots of activity cause recording to be continuous on all outdoor cameras. On the other hand, the right AI-based video analytics can detect the difference between meaningless background changes like snow or rain and can eliminate the recording-spike phenomenon. Of course, running those kinds of analytics also increases the computer video processing power requirements.
There is some good financial news when you consider the exponential advancement of computing hardware and software technology and the two technology megatrends that result from it: decreasing technology costs and increasing technology capabilities. This means that shorter computing hardware lifecycles are appropriate for premises-based computing and storage so that you can stay ahead of the general need to increase the use of video and video analytics, and not run out of computing and storage capacity – potentially compromising security operations and investigations effectiveness.
Furthermore, utilizing a true cloud-based video VMS service and cloud-based analytics means having instantly-expanding compute and storage resources can give you a video technology infrastructure that expands and contracts with your actual usage on a minute-to-minute basis.
Evidence of the cost and capabilities megatrends having an impact is the move that Eagle Eye Networks made in early 2019 to cut its subscription prices by up to 45%. Smart cities are finding that a well-designed cloud-based approach to video can reduce the costs and efforts of video deployments for security and city operations by more than half, as Mexico City recently learned.
Now, back to our topic: smart codecs.
Today’s Codecs
As explained in the previous article, the first generation of video codec technology (codec being short for coding/decoding) dealt with improvements in compressing each video image. But they still sent the visually-same image over and over again when the scene the camera was viewing didn’t change. The next generation of codecs began sending partial images – only the parts that changed – in between the full images. This approach can reduce video storage dramatically, sometimes more than 80%.
Today’s video compression standards are very flexible due to the wide range of devices and software that video streams to across a wide variety of devices and networks, including the Internet. Thus, today’s security video technology benefits from the many advances made in video processing for movies, television and internet video streaming. For example, the H. 264 video standard is at version 26 as of June 2019.
The result is that security camera makers can develop smart codecs that optimize video compression to minimize bandwidth and storage requirements while optimizing video quality based on the content and purpose of the video. This is done using the approaches specified in the H.264 and H.265 standards.
Smart Codecs
Originally, codecs set one compression level, one frame rate, and one I-frame interval (defining a group of pictures [GOP] that start with an initial full-image frame followed by a number of image-change frames, as explained in the Part 2 article) that all remained constant for a stream of video. Today’s smart codecs can change any or all of these values depending upon what’s happening in the video scene. Thus, smart codecs now can have:
- Dynamic Compression. Continuously identify the regions of interest (ROI), such as image areas with people and vehicles, and maintain high quality/low compression for those image regions in each frame, applying low quality/high compression for other regions.
- Dynamic I-frames. Increase the number of full-image frames (I-frames) in a video stream when there is activity, which reduces the compression artifacts (image distortion) for the frames in between the I-frames.
- Dynamic FPS. Increase the number of frames sent per second, increasing the smoothness of moving people and objects, and assuring that important actions (such as a shoplifter hiding an item) aren’t missed.
A smart codec with dynamic FPS adjustment allows you to have fewer frames per second when there is no activity, providing the greatest savings in video storage without compromising the video quality of scene activity.
It is important to get the specifics about what any particular smart codec does, and to make sure the codec’s performance is appropriate for the type of scene and the activity in it. Dynamic FPS is important, for example, for street intersections, retail stores, airport checkpoints, transportation terminals, stairwells, doorways and hallways.
For example, Axis Communications created Zipstream and Avigilon created HDSM SmartCodec (HDSM stands for High Definition Stream Management). The Axis and Avigilon smart codecs adjust all three smart codec capabilities listed above. Most camera codecs address only the first two.
It’s also important to discover what specific smart codes settings you should use for a particular camera’s scene, given that the manufacturer’s default settings won’t be suitable for all camera scenes and activities.
Video Processing Requirements
Keep in mind that the better the video compression, the greater the computing power required to decompress the video. While smart codecs help reduce network bandwidth utilization and video storage requirements, they increase the video processing and memory requirements of video servers and sometimes of the devices for video viewing. Add video analytics into the mix, and it is likely time to give serious thought to what your future roadmap should look like for security video technology.
About the Author:
Ray Bernard, PSP CHS-III, is the principal consultant for Ray Bernard Consulting Services (RBCS), a firm that provides security consulting services for public and private facilities (www.go-rbcs.com). In 2018 IFSEC Global listed Ray as #12 in the world’s top 30 Security Thought Leaders. He is the author of the Elsevier book Security Technology Convergence Insights available on Amazon. Mr. Bernard is a Subject Matter Expert Faculty of the Security Executive Council (SEC) and an active member of the ASIS International member councils for Physical Security and IT Security. Follow Ray on Twitter: @RayBernardRBCS.
