Codecs
A video codec is a mechanism that takes analog data, like light, color, and sound, and compresses (the “co” part of codec) that data into information the computer can understand. From there, the video is decompressed (the “dec” part of codec) so it can be played on our screens. Codecs are an essential part of all digital video recording and IP surveillance cameras.
H.264
The main codec used today is the H.264, or mpeg-4 AVC, codec. H.264 came out in 2003, and companies such as Netflix, Youtube, Vimeo, and Apple still use it today. It can support up to 8k video, but works better for 1080p and lower resolutions. So many companies use this codec because it is very efficient at reducing video bitrate while maintaining good picture quality.
As 4k video becomes more standard, H.264 struggles to keep up with the information load. That’s where its successor H.265 comes in.
H.265
H.265, or HEVC, came out in 2013 and comes with many cameras today. H.265 uses about HALF the bandwidth that H.264 uses while increasing video quality. In IP surveillance, this saves businesses bandwidth costs and improves overall effectiveness of security systems.
H.265, like H.264, uses mainly two methods to save bandwidth and increase video quality: inter and intra-frame compression.
The idea behind inter-frame compression is that, between some shots, very little changes. Take, for example, a presidential speech; the president stands at a podium and talks. The only thing that moves during that speech is the president’s body. The background and podium stay stationary. Inter-frame compression, then, compresses only what changes from frame to frame using a reference picture to fill in what does not change. This reduces overall bitrate and saves massive amounts of bandwidth.
Intra-frame compression further reduces bitrate. Unlike inter-frame, this method of compression looks at a single frame at a time. H.265 analyses 64x64px blocks of each frame and, based on the colors of just a few of the pixels in that block, fills in the rest of pixels using prediction models. This reduces the total bitrate because only a few pixels in each block need to be compressed.
H.264 only analyzes 16x16px blocks at a time. Among many other things, the increase in pixels analyzed at once plus more advanced prediction models make H.265 much more effective than its predecessor.
Some difficulties with H.265
Though more effective at reducing bitrate and increasing video quality, one of the issues that has prevented widespread adoption of H.265 so far is that it requires much more processing power than H.264. Also, as H.265 compresses more data, it also has higher latency than H.264, which can prove problematic in fast-paced environments where PTZ cameras capture quickly-moving objects. Furthermore, not all platforms support H.265 yet, but the industry is moving toward that.
Going forward
Some companies, like Hikvision, are rapidly adopting and even improving upon H.265. Learn more about Hikvision’s H.265+ codec in our blogpost about Hikvision specific IP surveillance software.
More and more companies are moving to H.265 as it provides high-quality compression and excellent video quality while saving consumers on bandwidth costs. It may soon become the industry standard in compression.