臺灣博碩士論文加值系統

近年來網際網路與行動通訊已經成為我們生活中的一部分，在網際網路中擁有大量的多媒體訊息，但多媒體訊息如果未經過壓縮的處理，就需要很大的儲存空間，藉此在傳輸的過程中就需要更多的傳輸時間，因此，在現今傳輸速度有限的情況下，怎麼使這些多媒體訊息用最小化的方式來儲存，以方便多媒體在傳輸的過程中更有效率，而「影像壓縮」就變成壓縮技術中的重要一環。

在一張512 x 512大小的圖片可能就有一半以上有重複的顏色，LBG與SOM的影像壓縮過程中，會重複運算已經壓縮過的顏色，在這過程中就浪費了許多時間成本。因此本論文的研究動機是要減少運算影像壓縮過程，並具有高品質的圖片。因而本論文是先將圖像中使用過的顏色分類出來，再將顏色切割區塊，並將區塊內的顏色依照八個區塊來加以分
群。再來依照八個區塊的權重值來分配Codebook的數量。經由實驗的模擬結果顯示，此方法可以在一個簡單架構下執行影像壓縮，並只需要設定一個參數，同時可大量減少壓縮所需要的時間。將此方法與多個著名的影像壓縮演算法來比較，本方法在執行時間上優於LBG、SOM、LazySOM等影像壓縮演算法，且在PSNR的數據上都有高水準的表現，經實驗結果後的數據表現，可以證明本論文的影像壓縮演算法具有高效能與高效率。

Recently, a large number of multimedia transmits in the network. However, the transmission speed and storage space confront with several limitations. Therefore, to provide an efficient and effective image compression technique for multimedia transmission environment in networks is required.

In the past, there are several image compression techniques have been presented. However, the quality and execution time of the image compression techniques still need to be improved. Therefore, this thesis proposes a new effective and efficient image compression method to overcome the limitation and improve the drawback. This work involves many critical steps to perform image compression task, such as select a used color, segment the colors with fixed size, compute the average colors for the segmented blocks, divide eight blocks for the average colors, assign codebook size to eight blocks depending on their weights, and allocate codebook size to the block using the maximum number of colors. For simulations and comparisons with different data clustering methods, there are two measure indicators have been utilized- time cost and PSNR. It is observed that the proposed algorithm outperforms several well-known image compression approaches in image compression

摘 要 I
Abstract III
謝 誌 V
目錄 VII
圖索引 X
表索引 XIII
第1章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 2
1.3 研究流程 4
1.4 研究範圍與限制 5
1.5 論文架構 5
第2章 文獻探討 7
2.1 影像介紹 7
2.2 向量量化 9
2.3 LBG演算法 11
2.4 Self-Organizing Map 演算法 13
2.5 LazySOM演算法 16
2.6 壓縮品質評估準則 19
2.6.1 均方誤差 (Mean Square Error, MSE) 19
2.6.2 峰值訊噪比 (Peak Signal-to-Noise Ratio, PSNR) 20
第3章 研究方法 21
3.1 主要概念 21
3.2 演算法之流程 25
第4章 研究成果 37
4.1 測試資料及環境 37
4.2 參數設定 38
4.3 Lena影像實驗 40
4.4 Peppers影像實驗 43
4.5 Baboon影像實驗 46
4.6 Tiffany影像實驗 49
4.7 Airplane影像實驗 52
4.8 Sailboat on lake影像實驗 55
4.9 Flower影像實驗 58
4.10 Store影像實驗 61
4.11 Slippers影像實驗 64
第5章 結論與未來展望 67
5.1 結論 67
5.2 未來研究方向 69
參考文獻 71
作者簡介 73

[1] Sayood, K., “Introduction to Data Compression,” 2nd Edition, Morgan Kaufmann Publishers, San Francisco, 2000.
[2] Gray, R.M., “Vector Quantization,” IEEE ASSP Magazine, Vol. 1, No. 2, pp. 4-29, 1984.
[3] Gersho, A. and Gray, R.M., “Vector Quantization and Signal Compression,” Kluwer Academic Publishers, Boston, MA, 1992.
[4] Sharma G, Trussell H J, Digital Color Imaging, IEEE Trans. on Image Processing, Vol. 6, No. 7, pp. 901–932, 1997.
[5] Cheng-Fa Tsai and Yu-Jiun Lin, "LISA: Image Compression Scheme Basedon An Asymmetric Hierarchical Self-Organizing Map", Lecture Notes inComputer Science, Vol. 5553, pp. 476–485, 2009.
[6] Cheng-Fa Tsai Ying-Sheng Lin and Jen-Chih Wang, “SEAN: A Simple expanding-tree algorithm based on mean-division for color quantization,” International conference on Machine Learning and Cybernetics, pp. 2658–2663, 2010
[7] Cheng-Fa Tsai and Jiun-Huang Ju, “ELSA: A New Image Compression Using An Expanding-Leaf Segmentation Algorithm,” Lecture Notes in Computer Science, Vol. 5579, pp. 624-633, 2009.
[8] Linde, Y., Buzo, A., Gray, R.M., “An algorithm for vector quantizationdesign,” IEEE Transactions on Commun., Vol. COM-28, pp. 84–95, 1980.
[9] Kohonen, T., “Self-organizing map,” Berlim, 1995.
[10] Kohonen, T., “Self-organizing map,” Proceedings of the IEEE, Vol. 78, No. 9, pp. 1464–1480, 1990
[11] Madeiro, F., Vilar, R.M., Neto, B.G.A., “A Self-Organizing Algorithm for Image Compression,” Proceedings of Vth Brazilian Symposium on Neural Networks, pp. 146–150, 1998.
[12] Amerijckx, C.; Verleysen, M.; Thissen, P.; Legat, J.-D., “ Image Compression by Self-Organized Kohonen Map,” IEEE Transactions on Neural Networks, Vol. 9, No. 3, pp. 503–507, 1998.
[13] Kangas, J., Kohonen, T., “Developments and applications of the self-organizing map and related algorithms,” Mathemathics and Computers in Simulation, Vol. 41, pp. 3–12, 1996.
[14] Barbalho, M., Duarte, A., Neto, D., Costa, A.F., Netto, L.A.,“Hierarchical SOM applied to image compression,” Proceedings of International Joint Conference on Neural Networks, pp. 442–447, 2001.
[15] Cheng-Fa Tsai and Yu-Jiun Lin, "LazySOM: Image Compression Using An Enhanced Self-Organizing Map," Lecture Notes in Computer Science, Vol. 5414, pp. 118–129, 2009.