Nagaraju, Bhargav (2016) Study of a Preferable Compression Plane for Video Coding. MTech thesis.
|PDF (Full text is restricted upto 27.04.2020)
Restricted to Repository staff only
Video recordings are promptly increasing for capturing a huge data. As digital media are encouraged day by day to collect more information and also to get an adequate information on any event. For example we have seen web cams are placed everywhere and every day we use to get a large amount of information. So for storing them we require lots of space. For minimize the above problem we are using video compression. We know that video means a set of continuous images occurs one after another at a certain unit time. The images are two dimensional in space normally we use “XY” plane and the occurrence of these planes in other dimension continuously makes a video i.e in “T” direction. As up to now all video coding standards which are available for us use to take video sequence as natural frames (formed by X and Y plane specifies rows and columns of the frame/image) for spatial redundancy and temporal redundancy (i.e redundancy along T direction). So for compression of the video we use redundancy as the major factor. So in this paper we make a new method for developing a video compression (i.e redundancy reduction) where we take the frames in non XY plane (like TX plane and TY plane) or XY plane depending upon the correlation coefficient for better compression. We are exploiting a complete extent of video redundancy in different frames (XY or non XY planes), and proposes an adaptive effective and preferable compression plane for video compression. This a pre-processing step prior to any standard video compression code. The explanation of the scheme is to ignore the natural meaning of a video. In video compression we can use both lossless and lossy compressions (i.e compression technique with or without inter frame prediction), here we can use this proposed method to get better results.
In this thesis we are going to use this proposed method on some standard coding technique to see the variation in the amount in PSNR value and also see the RD curve. We use H.264 standard code for inter frame prediction method to compress the video and we will see the comparison with the original video.
|Effective compression plane, Correlation coefficient; Pre-processing; Fractional pel search; Motion vector estimation
|Engineering and Technology > Electronics and Communication Engineering > Signal Processing
Engineering and Technology > Electronics and Communication Engineering > Image Processing
|Engineering and Technology > Department of Electronics and Communication Engineering
|Mr. Sanat Kumar Behera
|28 Apr 2018 17:51
|28 Apr 2018 17:51
Repository Staff Only: item control page