In case if you digitize a signal from standard CCTV camera you can get a digital stream with more than 150 Mbps (Mega bits per second), but such a stream contains a lot of redundant information.
All image compression algorithms are divided in two groups namely:
Most CCTV compression algorithms used in CCTV are lossy compressions, because such algorithms offer higher compression ratio (the ratio of the resulting video file size compared with the original file size).
Video compression algorithms are divided in two groups:
- Frame based compression (JPEG, Wavelet, JPEG 2000)
- Stream based compression (MPEG-2, MPEG-4, H.264, MPEG-7)
Usage of stream based compression algorithms enables greater savings on storage space and network bandwidth but as a trade off these algorithms require higher computing performance.
Four of the most common type of compressions widely used in CCTV are:
- Motion JPEG
IP Video System Design Tool can help you to see how choosing of compression methods can affect required network bandwidth and CCTV storage space, so you can clearly see the difference in terms of compression ratio.
Motion JPEG is very popular compression format. MJPEG fits very well for video archives because of its frame based nature.
MPEG4 can be 3 times more efficient in terms of compression ratio in compare with Motion JPEG.
But MPEG4 is a bad choice for systems with frame rate less than 5-6 frames per second.
H.264 can be 50-100% more efficient in compare with MPEG-4
MPEG-4 and H.264 are ideal for CCTV systems with limited but stable bandwidth.
JPEG2000 is similar to JPEG, but uses wavelet transform instead of discreetest-cosine-transform (DCT) of JPEG. JPEG2000 offers a better image quality on higher compression levels. Another great advantage is a possibility to decompress lower resolution representation of the image. This feature is good for motion detection algorithms. However JPEG2000 compression needs way higher CPU performance, than JPEG.
More CCTV Guides