跳到主要內容

臺灣博碩士論文加值系統

(44.210.21.70) 您好!臺灣時間:2022/08/16 17:24
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:邱富源
研究生(外文):Fu-Yuan Chiu
論文名稱:向量量化的影像壓縮法之研究
論文名稱(外文):A Study on Image Compression Using Vector Quantization
指導教授:魏清煌
指導教授(外文):Ching-Huang Wei
學位類別:碩士
校院名稱:國立高雄第一科技大學
系所名稱:電腦與通訊工程所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2002
畢業學年度:90
語文別:英文
論文頁數:69
中文關鍵詞:向量量化影像壓縮
外文關鍵詞:Vector QuantizationImage Compression
相關次數:
  • 被引用被引用:0
  • 點閱點閱:206
  • 評分評分:
  • 下載下載:38
  • 收藏至我的研究室書目清單書目收藏:0
使用向量量化的影像壓縮法之架構裡,編碼端與解碼端需要一個性能良好的碼簿,目前向量量化在碼簿訓練方法大多使用一種簡單又有效的 K平均演算法,在本論文□,我們提出一種改進式 K-平均演算法,它在碼簿訓練步驟中,使用兩個躍進比率值以交替方式來修正碼簿,反覆以較小值與較大值交互更替方式做修正,以提昇收斂速度。
In the image compression using the vector quantization schemes, a good performance codebook is required in both the encoding and the decoding procedures. The K-means algorithm for training the codebook is widely used in the vector quantization, mainly for its simplicity and relatively good performance. In this thesis, we propose an improved K-means algorithm that alternatively uses the bi-scaling values in a codebook updating step for the design of vector quantizer.The small and big scale values are alternatively used in the codebook updating step to speedup the convergence.
List of Abbreviations
List of Figures
List of Tables
Chapter 1 Introduction
1.1 Motivation
1.2 Thesis Organization
Chapter 2 Vector Quantization (VQ)
2.1 Introduction
2.2  VQ Encoder/Decoder Design
2.2.1  Codevector and Codebook
2.2.2  VQ Encoder and Decoder
2.3  Measuring the Performance of Vector Quantizer
2.3.1  Mean-Square-Error (MSE
2.3.2  Signal-to-Noise Ratio (SNR) and Peak Signal-to-Noise Ratio(PSNR)
2.3.3  Bit rate (BR) and Compress Rate (CR)
2.4  Properties of Optimal Quantizer
2.4.1  Nearest Neighbor Condition
2.4.2  Centroid Condition
2.5  The Generalized Lloyd Algorithm
Chapter 3Codebook Training Methods for Vector Quantization
3.1  Introduction
3.2  Initial Codebooks Design
3.2.1   th Method
3.2.2  Splitting Method
3.2.3  Maximum Method
3.3  Modified K-means Algorithm for Training Codebook
3.4  Paliwal’s Algorithm for Training Codebook
3.5  Our Proposed Algorithm for Training Codebook
3.5.1  Calculation of Stepwise-Optimal Scale Values
3.5.2  Bi-Scaling K-mean Algorithm
Chapter 4Simulations Results
4.1  Introduction
4.2  Simulations Results
4.2.1  Comparison of Convergence Behavior
4.2.2  Comparison of Performance
Chapter 5Conclusions and Future Studies
5.1  Conclusions
5.2  Future Studies
Y. Linde, A. Buzo, and R. M. Gray, ”An algorithm for Vector quantization design,”IEEE Trans. Commun.D. Lee, S. Baek, and K. Sung, “Modified k-means algorithm for vector quantizer design,”IEEE Signal Processing Lett.I. Katsavounidis, C. C. J. Kuo, and Z. Zhang, “A new initialization technique for generalized Lloyd iteration,”IEEE Signal Processing Lett.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top