Video Encoder and Encoding

                   An Encoder is a device used to convert analog video signals into digital video signals. Most of them compress the information so it can be stored or transmitted in the minimum space possible. To achieve this it takes advantage of video sequences that have spatial or temporal redundancy. Therefore, eliminating redundant information obtains that encode information more optimal. The spatial redundancy is erased with DCT coefficient coding. To delete temporal redundancy is used the motion compensation prediction, with motion estimation between successive blocks.

The operation method is:

Ø  Signals are separated luma (Y) and chroma (C).

Ø  Find the error of estimation to make the DCT.

Ø  The coefficients are quantified and entropy coded (VLC).

Ø  Coefficients are multiplexed and passed to the buffer. The buffer controls the quality of signal.

Ø  Check that the outflow bit stream of the buffer is not variable, because the signal   is thought to be transmitted on a channel with a steady speed.

Ø  The quantified image is reconstructed for future reference for prediction and motion estimation.

The DCT algorithm and the block quantification can cause visible discontinuities at the edges of the blocks leading to the known “Blocking effect”, because the DCT omits the 0 in the matrix, so may produce imperfections. As a result of, new standard video coding like H.264/MPEG-4 ACV, includes filter algorithms able to decrease that effect.