With the advent of IP Security Camera Systems and its increased popularity over Analogue CCTV Systems, it has now become inevitable for the Camera Manufacturers/System Integrators to focus more on the data compression to maintain the quality of image as well as save on transmission bandwidth and hard disc storage space.
Its because of this reason we can see Cameras and also DVR’s flooded in market with multiple streams having different types of image compression standards for different applications. An application which requires live monitoring will need signal streams high quality image, whereas for recording purpose or even for viewing in Mobile from a remote location, different streams are required with different compression techniques. There are DVR’s in the market with even five separate streams for five different kinds of applications.
Following are some of the ongoing and upcoming image compression techniques used in IP CCTV/Video industry:-
- JPEG Image Compression (Joint Photographic Expert Group
- M-JPEG Image Compression
- MPEG 1,2, 4 (Part2) (Moving Pictures Expert Group)
- H.264 or MPEG 4 (Part 10)
Broadly there are two concepts which govern most of the Image Compression techniques – Spatial Compression & Temporal Compression. In Spatial Compression – all data compression is done on a single image frame without touching other frames- eg JPEG, Wavelet & M-JPEG, where as Temporal Compression deals with the changes in the previous and subsequent frames during transition- eg MPEG, H.264 etc.
Below are some more details on individual Compression Techniques:-
JPEG Image Compression:
Finds its application in still photographs or a video with very low fps (Frames per second). The data is compressed of individual photographs.
M-JPEG is similar to JPEG, where images are at a very high fps to give an illusion of a running video. The drawback here is that its today not a standard, hence compatibility issues come into picture with different manufacturers.
MPEG 4 (Part2):-
It is most acceptable image compression in industry with its different versions. This technology focuses on the dealing with the change in previous and subsequent frames rather than on each frame itself. It saves only that portion of the frame which is changed from that of its previous state. Hence each frame also called p- frame or predictive frame is actually the composition of previous frame plus the changes in same with that of current frame.
MPEG 4 (Part 10) or H.264:-
Also called Advanced Video Coding (AVC) is more efficient in terms of maintaining the
quality of image as apart from P frames, it also has an I-frame or intra frame which do not refer to any of the precious or future frames but is independent of both of them and because of which it is a frame with complete information. I frames can also be used to implement fast-forward, rewind and other random access functions. Its because of this reason bit rate is 50% lesser than MPEG 4 (Part2) and H.264 proves its utility while saving of storage disc space is the main concern along with maintaining the quality of image.
Apart from the above, Wavelet Image Compression and Fractal Image Compression are two more compression techniques on images not followed as a standard practice.