Implementation of Video Compression Using APRS and DWT

Video compression involves image compression and audio compression. Compression basically removes or reduces the redundant data, but retain the quality of the video. Video compression leads to reduction in storage and easier transmission, which leads to lesser hardware and reduced cost. It also reduces the time required for videos to be sent over internet or downloaded. Video compression can be achieved by using different transforms like FT, DCT, and DWT etc. In this project video compression is done using Discrete Wavelet Transform (DWT). DCT maps a time domain signals to a frequency domain representation. To calculate the DCT of an entire video frame takes an unacceptable amount of time. The only solution is to partition the frame into small blocks and then apply the DCT to each block. However, this leads to degradation in picture quality. DWT offers an advantage that, it can be applied to entire frame without partition. DWT is another transform that maps time domain signals to frequency domain representations. There are three main parts found in a standard video compression algorithm: Motion Compensated Prediction (MCP), Transform Coding (DWT/DCT), and Group of pictures (GOP).
There are several motion estimation algorithms amongst which Adaptive Rood Pattern Search (ARPS) is used in this project. The main advantage of ARPS is that computation time is less. The implementation is done using MATLAB 7.5 version. The audio and video from the input are treated separately. Firstly both the video and audio is divided into frames. The video frames are then converted into images. The group of frames undergo block matching which consist of motion prediction and motion estimation, to judge the movement of objects in a scene. Then DWT is applied which generate coefficients and then actual compression is achieved by eliminating the redundant coefficients by choosing appropriate threshold. This threshold is selected so as to achieve good compression ratio. The audio frames undergo decomposition using DWT. Here the compression is achieved by non-linear quantization of the coefficients obtained. The compressed video and audio are stored and if required transmitted. When the video need to be played, compressed audio and video are reconstructed using Inverse DWT (IDWT). Finally both are combined and played.