(3.231.29.122) 您好!臺灣時間:2021/02/25 22:35
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:::

我願授權國圖
: 
twitterline
研究生:郭克強
研究生(外文):Ko-Chiang Kuo
論文名稱:二階彩色影像壓縮─應用神經網路與量化誤差修正技術
論文名稱(外文):2-way Color Image Compression, Via Neural Network and Quantization error correction
指導教授:王榮華
指導教授(外文):Jung-Hua Wang
學位類別:碩士
校院名稱:國立海洋大學
系所名稱:電機工程學系
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:1999
畢業學年度:87
語文別:英文
論文頁數:70
中文關鍵詞:小波轉換KARHUNEN-LOEVE 轉換向量量化自成長神經網路有限調色盤顏色量化量化誤差修正影像壓縮
外文關鍵詞:Wavelet TransformKARHUNEN-LOEVE transformvector quantizationself-creating neural networklimited color palettecolor quantizationquantization error correctionimage compression
相關次數:
  • 被引用被引用:0
  • 點閱點閱:105
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
本論文在第一部分提出一套影像壓縮系統架構應用於靜態灰階影像壓縮。實現此一系統架構主要分為四階段:第一階段應用離散小波轉換將原始灰階影像轉換成小波係數,由於小波係數除了在低頻部分的係數分佈很廣且擁有較大的能量,大部分高頻係數能量非常小並分佈在原點附近狹小的範圍內。因此在第二階段,即可利用 KLT找出 Level 1 與Level 2 的最主要的成分來代表整個subband的係數,然後成為第三階段─向量量化的輸入。當我們將第三階段輸入向量的維度減少後,在相同的預設壓縮比之下,可以大幅減少神經網路訓練的時間。節省時間的同時,我們並沒有犧牲影像品質。在第三階段裡,吾人利用一個自成長型神經網路,即週期性活動力守恆網路(PVC) 應用於產生一近似最佳化的向量量化器。至於低頻部分的小波係數則完全由純量量化器 (DPCM) 來量化。最後,第四階段利用適應性算數編碼將第三階段之輸出編碼,以便傳輸或儲存。實驗結果顯示,此一影像壓縮系統的確可以達到在高壓縮比時仍保有相當不錯的影像品質。在本論文的第二部分中,我們提出一套彩色影像壓縮系統應用於靜態全彩影像壓縮。首先,利用 PVC 訓練出一相關於影像的有限(256)色盤。接著,我們經由2-way量化器把一張影像切割成許多區塊,並分類成高頻與低頻部分。而PVC量化器針對不同類別的區塊代表顏色個別量化並結合量化誤差修正技術(QEC)將量化誤差減少到最低。實驗結果顯示,此一結合PVC 神經網路之2-way彩色影像壓縮系統的確可以達到在高壓縮比時比其它彩色影像壓縮系統 較好的影像品質。

In the first part of this thesis, we propose a MICS scheme for image compression of monochrome images based on the discrete wavelet transform and PVC algorithm. Implementation of the MICS scheme involves four major steps. Firstly, apply discrete wavelet transform to obtain a set of biorthogonal subbands of input image. The original image is decomposed at different scales using a pyramidal algorithm. Secondly, use the Karhunen-Loeve transform (KLT) to project wavelet coefficients of some subbands onto fewer principal components. Thirdly, apply a self-creating algorithm, namely, the periodical vitality conservation (PVC) to quantize both the output of KLT and the remaining subbands. Finally, the adaptive arithmetic coding is employed to encode the outputs of PVC. Performance comparisons with the embedded wavelet hierarchical image coder (EZW) (Shapiro, 1993), multi-threshold wavelet coder (MTWC) (Wang and Huo, 1997) and Lazar's coder (Lazar et al., 1996) were conducted. All simulation results indicate that quality reconstructed images can be obtained by using the MICS scheme, even at very low bit rate.
The second part of this thesis focuses on color image compression. The proposed CICS scheme first utilizes PVC to design a limited color palette. Then, a 2-way quantization approach is employed to divide all the image blocks into two classes, namely low-frequency and high-frequency classes. The training vectors of two classes can be presented separately and concurrently to two PVC networks for quantization. Furthermore, each training vector is composed of some number of representative colors, depending to which class the block belongs. Finally, when incorporated with QEC, SAQ and the excellent quantization performance delivered by the PVC network, the 2-way approach is shown capable of minimizing the effect of quantization error induced by the high-frequency class. Experimental results have been shown to justy all the claims in this thesis.

CHAPTER 1 INTRODUCTION 2
1.1 MOTIVATION 2
1.2 OUTLINE 4
CHAPTER 2 MONOCHROME IMAGE COMPRESSION SYSTEM 5
2.1 BASICS AND SURVAYS 5
2.2 DISCRETE WAVELET TRANSFORM 10
2.2.1 Wavelet Decomposition 10
2.2.2 Statistical Distribution of the Wavelet Coefficients 15
2.3 KARHUNEN-LOEVE TRANSFORM 18
2.4 TRAINING ALGORITHM FOR CODEBOOK DESIGN 20
2.5 EXPERIMENTAL RESULTS 28
2.5.1 Using Different Wavelets 28
2.5.2 Comparison with Other Coders 28
CHAPTER 3 COLOR IMAGE COMPRESSION SYSTEM 35
3.1 BACKGROUNDS 35
3.2 LIMITED COLOR PALETTE DESIGN 41
3.2.1 Color Quantization Using PVC 42
3.2.2 The Fast Codebook Searching Algorithm 44
3.3 THE TWO-WAY QUANTIZER 46
3.4 VECTOR QUANTIZATION USING PVC ALGORITHM 49
3.4.1 The Representative Colors 49
3.4.2 Quantization Error Correction (QEC) 49
3.4.3 QEC with Successive Approximation Quantization 50
3.5 EXPERIMENTAL RESULTS 52
CHAPTER 4 DISSCUSIONS AND CONCLUSIONS 67
4.1 CONTRIBUTIONS 67
4.2 FUTURE WORKS 68
REFERENCES 69

[1] D. Lazar, A. Averbuchand and M. Israeli, "Image compression using wavelet transform and multiresolution decomposition," IEEE Trans. on Image Processing, vol.5, pp.4-15, 1996.
[2] C. Q. Chen, S. N. Koh and P. Sivaprakasapillai, "Codebook grneration for vector quantization," IEE Electronics Letters, vol.31, pp.522-523, 1995.
[3] Abut, H. Vector Quantization. IEEE reprint Collection. IEEE Press, Piscataway, New Jersey, 1990.
[4] Nasrabadi, N. M. "Image coding using vector quantization: A review," IEEE Trans. Communications, vol.36, pp.957-971, 1988.
[5] A. N. Netravali, and F. W. Mounts, "Ordering techniques for facsimile coding: A review," Proceeding of the IEEE, vol.68, pp.796-807, 1980.
[6] Y. Linde, A. Buzo and R. M. Gray, "An algorithm for vector quantized design," IEEE Trans. Communications, vol.28, pp.84-95, 1980.
[7] Kohonen, T. Self-Organization and Associative Memory, volume 8 of Springer Series in Information Sciences. Springer-Verlag, New York, 1984.
[8] S. C. Ahalt, A. K. Krishnamurthy, P. Chen and D. E. Melton, "Competitive learning algorithms for vector quantization," Neural Networks., vol.3, pp.277-290, 1990.
[9] D. I. Choi and S. H.Park, "Self-creating and organizing neural networks," IEEE Trans. Neural Networks, vol.5, pp.561-575, 1994.
[10] Fritzke, B. "Growing cell structures a self-organizing network for unsupervised and supervised learning," Neural Networks, vol.7, pp.1441-1460, 1994.
[11] J. H. Wang and W. D. Sun, "Self-development competitive learning VQ based on vitality conservation networks," Proc. IEEE International Conf. on Neural Networks, vol.2, pp.885-890, Alaska, USA, 1998.
[12] A. K. Krishnamurthy, S. C. Ahalt, D. E. Melton and P. Chen, "Neural networks for vector quantization of speech and images," IEEE Journal on Selected Areas in Communications, vol.8, pp.1449-1457, 1990.
[13] Shapiro, J. "Embedded image coding using zerotrees of wavelet coefficients," IEEE Trans. on Signal Processing, vol. 41, pp.3445-3462, 1993.
[14] A. Said and W. A. Pearlman, "A new, fast, and efficient image codec based on set partitioning in hierarchical trees," IEEE Trans. on Circuits and Systems ofr video Technology, vol.6, pp-243-250, 1996.
[15] D. Taubman and A. Zakhor, "Multirate 3-D subband coding of video," IEEE Trans. on Image Processing, vol.3, pp.572-588, 1994.
[16] H. J. Wang and C.-C. J. Kuo, "A multi-threshold wavelet coder (MTWC) for high fidelity image compression," International Conference on image processing., vol.1, pp.652-655, Santa Barbara, CA, USA, 1997.
[17] J. H. Witten, R. Neal and J. G. Cleary, "Arithmetic coding for data compression," Comm. ACM, vol.30, pp.520-540, 1987.
[18] M. Antonini, M. Barlaud, P. Mathieu and I. Daubechies, "Image coding using wavelet transform," IEEE Trans. Image Processing, vol.1, pp.205-220, 1992.
[19] A. Cohen, I. Daubechies and J. C. Feauveau, "Biorthogonal bases of compactly supproted wavelets," Commun. Pure Appl. Math, vol.41, pp.485-560, 1992.
[20] M. Vetterli and C. Herley, "Wavelet and filter banks: Relationships and new results," ICASSP'90: International Conference on Acoust., Speech, Signal Processing, vol.3, pp.1723-1726, Albuquerque, NM, 1990.
[21] O'shaughnessy, D. Speech communication: Human and machine, Reading, MA: Addison-Wesley.
[22] Musmann, H. G. "Advances in picture coding," Proceeding of the IEEE, vol.73, pp.523-548, 1985.
[23] Huffman, D. A. "A method for the construction of minimum-redundancy codes," Proc. Inst. Electr. Radio Eng. 40, vol.9, pp.1098-1101, 1952.
[24] G. Houle and E. Dubois, "Quantization of color images for display on graphics terminal," in Proc. IEEE Global Telecomm. Conf., pp.1138-1142, 1986.
[25] Heckbert, P. "Color image quantization for frame buffer display," ACM Comput. Graphics, vol.16, no.3, pp.297-307, July 1982.
[26] Braudaway, G. "A procedure for optimum choice of a small number of colors from a larger color palette for color imaging," in Proc. Electron. Imaging'87, San Francisco, CA, 1987.
[27] Gray, R. "Vector quantization," IEEE ASSP Mag., vol.1, pp.4-29, Apr. 1984.
[28] C. K. Chen, J. S. Chow, "Optimum display of true color images using vector quantization techniques," in Proc. IEEE Visual Signal Processing and communication, pp.5-8, 1991.
[29] R. Balasubramanian and C. A. Bouman, "Sequential scalar quantization of color images," J. Electron. Imaging, vol.3, no.1, pp.45-59, Jan. 1994.
[30] S. C. Pei and C. M. Cheng, "Dependent scalar quantization of color images," IEEE Trans. Circuits Syst. Video Technol., vol.5, pp.124-139, Apr. 1995.
[31] A. Zaccarin and B. Liu, "A novel approach for coding color quantized images," IEEE Trans. Image Processing, vol.2, pp.442-453, Oct. 1993.
[32] S. C. Pei and Y. S. Lo, "Color image compression and limited Display using self-organization Kohonen map," IEEE Trans. on Circuits and syst. for Video Technolog., vol.8, no.2, pp.191-205, Apr. 1998.
[33] J. Makhoul, S. Roucos and H. Gish, "Vector quantization in speech coding," Proc. IEEE, vol.73, no.11, Nov. 1978.

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔