MJPEG, MPEG4 or H.264 which one is better?

1 Januari 2014

Codecs Analysis

MJPEG vs MPEG4 vs H.264

Selecting the proper imaging technology up front will be one of your most important decisions you will face designing and implementing your network video project. When considering compression options for network video solutions, the following video related issues should be reviewed and carefully considered.

Minimum Resolution
Image Quality
Minimum and Maximum Frame Rates (FPS)
Number of Cameras (Scalability)
Storage Requirements
Available Bandwidth
Overall Latency (time difference between live and displayed video)
Budget

The compression logic used in both MJPEG, MPEG4 and H.264 are beyond the scope of this white paper. Suffice it to say that MJPEG, MPEG4 and H.264 compression algorithms are the most common forms of video compression currently used in the network video industry.

MJPEG/JPEG Analysis

Pros

Well Suited For Mega-Pixel Cameras
Better Image Quality (clear individual images)
Resiliency (fast image stream recovery in the event of packet loss)
Interoperability (industry standard compression/decompression available on all PCs)
Decompression on the PC is better (more video streams on PC)
Large Number of Supported Cameras
Lower Latency (better live viewing and more responsive PTZ control)
Graceful degradation (reduced bandwidth does not reduce image quality)
Constant image quality (quality remains constant regardless of image complexity)
Can Perform Live Motion Detection and Image Analysis (at the PC)
Low complexity (for images searches and manipulation)
Can display and record up to 30 FPS

Cons

High bandwidth usage (at frame rates above 10 FPS)
High storage requirements (at frame rates above 10 FPS)
No support for synchronized sound

MPEG4 Analysis

Pros

Higher compression Rates (above 10 FPS)
Can Maintain Constant bit rate (CBR)
Good Streaming Protocol (designed for real-time viewing)
Can Sync Audio and Video (for live and recorded streams)
Can display and record up to 30 FPS

Cons

Image quality is less than MJPEG
Bandwidth does not perform as advertised
Storage issues with many cameras (fragmentation)
Decompression on the PC is quite high (fewer video streams)
Low robustness (can lose video if bandwidth drops)
Higher latency (delayed live viewing and sluggish PTZ control)
Protocol Issues (not designed for recording and analysis)
Few Number of Supported Cameras
Less Resilient (packet loss causes lost video)
Can Not Perform Live Motion Detection and Image Analysis (at the PC)
CBR Mode Issues (image quality suffers with network congestion or movement in scenes)
Licensing restrictions (may need to pay for viewers)
Not Suited For Mega-Pixel Cameras

H.264 Analysis

Pros

Higher compression Rates (above 10 FPS)
Lower storage requirements (at 10 FPS or higher)
Can Maintain Constant bit rate (CBR)
Good Streaming Protocol (designed for real-time viewing)
Can Sync Audio and Video (for live and recorded streams)
Can display and record up to 30 FPS

Cons

Storage issues with many cameras (fragmentation)
Decompression on the PC is quite high (fewer video streams)
Low robustness (can lose video if bandwidth drops)
Higher latency (delayed live viewing and sluggish PTZ control)

Conclusions:

With the advent of H.264, MPEG4 popularity has diminished dramatically.
If you require synchronized audio and video then H.264 is the clear choice.
For most security surveillance and remote monitoring applications the MJPEG compression technology is the best choice.

Baca juga

21 Mei 2017
BOSCH Cloud Services
Bosch announces Cloud-based Services for video monitoring and business support With new Cloud-based Services from Bosch Security Systems, central monitoring stations (CMS) can now take advantage of the growing trend of video monitoring.

10 Mei 2015
BOSCH vs AXIS vs SONY
BOSCH Starlight vs AXIS LightFinder vs SONY ipela, which one is the best?

8 Juli 2012
GVS1000 Delivers Industry Longest Range Active Infrared Imaging System
The Bosch GVS1000 Long Range Imaging System is an integrated day/night, pan/tilt/zoom surveillance solution for critical security applications With 3,280 feet (1 kilometer) of recognition-level imaging and 3,937 feet (1,200 meters) of classification-level imaging.