Novel Video Codec Based on Wavelet-DCT Coefficients and Hybrid Motion Estimation Technique


Author (s):

  1. Dr. D.R.Bhojani, Darshan Institute of Engineering & Technology, Rajkot, Gujarat, INDIA
  2. P.N.Korat, Darshan Institute of Engineering & Technology, Rajkot, Gujarat, INDIA, korat.piyush8@gmail.com

Abstract:

Digital video is present in many applications, including video conferencing, DVDs, and broadcast television. Since digital video requires a prohibitive amount of storage in uncompressed form. There is a limit to the available storage space and hardware resources. The widely established compression standards suffer from the annoying blocking artifacts. The reason for the blocking artifacts is the usages of block based DCT. Therefore it is judicious to use discrete wavelet transform which can be directly applied to the whole image/frame at once. Therefore, this research develops an enhanced MPEG-2 video compression algorithm that explores the advantages of DWT. The compressibility factor, quality of reconstruction of frames; everything depends on the efficiency of the motion estimation algorithm that is used in the standard, therefore hybrid motion estimation technique (H-MET) based on Diamond Search (DS) algorithm and Adaptive Rood Pattern Search (ARPS) algorithm proposed. To measure the visual quality of the video parameters like: Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Structural Similarity Index (SSIM), Mean Absolute Difference (MAD), Blocking Beta, Blurring Beta etc. are used. Experimental results show that the MPEG codec with hybrid transformation and hybrid motion estimation technique (H-MET) provides more compression as well as less computational time as compared to existing MPEG codec.

No of downloads: 24