Video compression technology is crucial for security camera systems, enabling efficient video encoding and storage. The two predominant standards are H.264 and H.265. While H.264 is currently more widespread, H.265 offers major improvements in compression efficiency. For security applications, is it worth upgrading to H.265, or should existing H.264 systems be maintained? Here is an in-depth technical comparison between these two popular video compression algorithms.
How Video Compression Works
Video files contain massive amounts of data. For example, high definition 1080p raw video at 30 frames per second requires over 1 Gbps data rate. Recording this to storage without any compression would fill up hard drives very quickly.
Video compression aims to reduce the data size of video files to save storage space while preserving quality. This is done by exploiting redundancies between video frames and encoding visual data more efficiently.
Some key techniques used by video compression algorithms include:
- Motion Compensation – Pixels that do not change between adjacent frames are encoded as instructions to copy those parts of the previous frame. This avoids resending duplicate data.
- Transform Coding – The image is divided into blocks which are converted into frequency domain representations using math transforms. This packs energy into fewer transform coefficients.
- Quantization – The transform coefficients are approximations, reducing precision. This limits data but introduces compression noise.
Entropy Coding – Statistical encoding methods are applied so common values use fewer bits. This gives lossless compression.
The compression ratio depends on how aggressively these techniques are applied. Higher quality video has less compression and lower quality video has more compression. The bitrate, frame size, frame rate, and other factors are balanced to meet the target quality within storage constraints.
Introducing the H.264 Standard
H.264, also known as MPEG-4 Advanced Video Coding (AVC), was formally published in 2003 by a consortium of organizations. It was designed to succeed older standards like MPEG-2 and H.263 while providing significantly better compression efficiency.
Some key improvements in H.264 include:
- Variable block sizes – Blocks can adaptively range from 4×4 pixels up to 16×16 pixels to better isolate changes.
- Multiple reference frames – More frames are used as references to improve motion compensation.
- In-loop deblocking filter – Filtering within the encoding loop reduces block artifacts.
- CABAC entropy coding – A more efficient yet complex statistical encoder.
- Spatial predictions – Pixels can be predicted from surrounding blocks, not just temporal frames.
For a given quality level, H.264 can reduce bitrates by 50% compared to MPEG-2 and 30% over H.263. This translates to smaller file sizes and lower required network bandwidth. H.264 quickly became ubiquitous across the video industry, used in applications like Blu-ray, streaming media, video calling, and video surveillance.
H.265 / HEVC Advances Video Compression
While H.264 represented a major leap forward, there were still opportunities for improvement. In 2010, work began on developing a successor. The result was H.265, also known as High Efficiency Video Coding (HEVC). Standardization was completed in 2013.
H.265 builds on H.264 technologies but enhances almost every aspect of the encoding process. Some key improvements include:
- More flexibility in block partitioning – Blocks can be divided into very small sub-blocks down to 4×4 pixels to isolate subtle changes.
- Increased motion compensation precision – More reference frames with finer pixel accuracy improves prediction.
- Better motion vector signaling – Advanced encoding of motion vectors and modes provides gains.
- Improved in-loop filters – Additional deblocking and adaptive filters smooth artifacts.
- Enhanced entropy coding – Context Adaptive Binary Arithmetic Coding (CABAC) is even more efficient.
The net result is H.265 can reach the same quality level as H.264 at around half the bitrate. This translates to 50% better compression efficiency while maintaining subjective visual quality. Alternatively, for a fixed bitrate H.265 can achieve substantially better quality.
H.264 vs H.265 for Security Camera Applications
For video surveillance applications, is it better to use existing H.264 equipment or upgrade to new H.265 capable products? There are number of factors to consider from performance to cost.
Storage Requirements
The improved compression efficiency of H.265 directly translates into reduced storage requirements for security camera video footage. For a fixed video quality level, replacing an H.264 system with H.265 cameras and NVR can cut the storage bitrate by 50%.
This means H.265 can record 2x longer than H.264 with the same storage hardware. Alternatively, it enables maintaining the same recording durations while using half the HDD capacity. For large scale deployments, the storage savings with H.265 are substantial.
Video Quality
At low bitrates below 2 Mbps, H.265 can deliver major gains in video quality over H.264. The improved encoding efficiency and reduction of artifacts is very noticeable, providing cleaner and sharper video, especially for low light footage.
At higher quality bitrates above 4 Mbps, the gains become less pronounced. There is still an advantage to H.265 but the difference is more moderate. For many security applications, H.264 at good bitrates is sufficient so this benefit may be less relevant.
High Resolution Video
H.265 provides more significant benefits for higher resolution video, such as 4K. Encoding such large video frames strains even H.264. The enhanced compression tools in H.265 handle ultra HD formats much more efficiently.
The savings are necessary to make 4K practical for security use, where H.264 bitrates might be prohibitive. This makes H.265 a better choice as cameras continue to transition to higher resolutions.
Latency
An issue with H.265 is that the additional processing complexity adds encoding/decoding latency. In live viewing or video surveillance analytics, delayed video can be detrimental.
However, modern hardware decoders and displays can buffer inputs to minimize latency. For recording and playback, the delay is irrelevant. With good implementations, it should not be a factor.
Hardware Decoding Requirements
To playback and stream H.265 feeds to clients requires more powerful hardware decoding support. Older equipment often lacks dedicated H.265 acceleration.
New NVRs, cameras, and client devices generally include hardware decoding, so this is only an issue with legacy systems. Replacing end points may be necessary to fully leverage H.265.
Cost Tradeoffs
H.265 capable cameras and NVRs remain at a price premium over H.264 products. The costs are gradually declining but upgrading an entire system is still a major investment.
Depending on the scale, sticking with H.264 a bit longer may make financial sense. However, the operational savings in storage continue throughout the life of the deployed equipment.
Forward Compatibility
Due to the substantially better efficiency of H.265, it is likely to remain the dominant format for many years, unlike the quicker transition from MPEG-2 to H.264. This makes H.265 a safer long term investment.
The backwards compatibility of H.265 decoders also ensures H.264 video will be usable into the future. Despite newer formats, H.264 is not likely to become obsolete anytime soon.
Gradual Transition
A gradual transition is possible. New H.265 cameras can integrate into existing H.264 NVRs, although losing some compression gains. NVRs can also decode both formats. Switching subsets of cameras or locations at a time helps control costs.
Conclusion
For many security camera users, H.264 remains a viable and cost-effective option, particularly at HD resolution and moderate bitrates. However, H.265 delivers tangible benefits through better compression, video quality at low bitrates, support for higher resolutions, and dramatically reduced storage requirements.
New H.265 encoding products can integrate into H.264 environments, allowing a staged transition over time. While requiring more powerful decoding hardware, for most applications H.265 represents the optimal video compression choice going forward.