次頻/向量量化編碼之位元分配

Abstract

次頻影像編碼在最近已被証實在低位元率是一種有效的高品質編碼技術 ,次頻編碼的基本原理是把影像信號編碼成高低頻帶做傳輸,藉不同頻帶有 不同變異數而達到壓縮的效果.而向量量化器在視訊通訊□是一種高壓縮 的編碼器.我們把向量量化器應用在次頻編碼器後,將會得到更高效率的壓 縮方法.在本論文□,我們設計一次頻/向量量化編碼系統,並針對此系統討 論位元分配問題.我們分析兩種數學方法 ----- 最佳純量量化方法及 Marginal方法,並藉此比較此兩種方法以得到最佳的位元分配. Subband image coding has recently been shown to be an effective technique for high quality coding at low bit rates. The basic idaea of subband coding is to split up the frequency band of the signal. A vector quantization (VQ) scheme is proposed for image coding to achieve high compression ratios in visual communation application. The combination of subband coding and vector quantization can provide a powerful method for compression image. In this thesis, we design a subband vecor quantization system and formulate the bit allocation problem as an optimization problem. We analyse the optimal bit alocation in the subband vector quantization scheme. To get the best performance and bit rate, optimal scalar quantizer method and marginal method are analyzed.