臺灣博碩士論文加值系統

這項研究提出了一種名為向量量化連通切割的輔助切割方法，該方法利用向量量化（Vector Quantization）的同質性向量特點進行彩色影像像素的分群，並在一開始時，使用小波轉換將影像高頻區域去除，再進行低通濾波，使其增加色塊連通法及產生較平滑的區域集合並降低高頻雜訊，使選取效率提升。
最後，利用連通性方法對像素點進行切割分塊，實現完整的切割程序。由於分群演算法會將影像分成許多同質性區塊，因此使用者可以根據需要選擇所需區塊進行分割，並增加二次擷取功能，從而達到較理想的分割效果。

This study proposed an auxiliary segmentation method named vector quantization connected segmentation. Leveraging homogeneous vectors in vector quantization to cluster the pixels of color images, this method first adopts wavelet transform to remove the high-frequency regions of the target image. Subsequently, low-pass filtering is performed to enhance color block connectivity, produce smoother region sets, and reduce high-frequency noise, thereby improving selection efficiency.
Finally, connectedness is used to segment and block pixel points, achieving a comprehensive segmentation process. Because the established clustering algorithm divides the image into numerous homogeneous blocks, users can select the required blocks for segmentation based on their needs. In addition, a secondary extraction function is provided to achieve satisfactory segmentation results.

摘要 i
ABSTRACT ii
致謝 iii
目錄 iv
圖目錄 vi
第一章　緒論 1
1.1 前言 1
1.2 研究動機與目的 4
1.3 研究流程 5
第二章　彩色影像擷取相關知識與技術 6
2.1 彩色影像基礎介紹 6
2.2 LBG演算法介紹 8
2.3 小波轉換去除高頻之方法 11
2.4 中值濾波去除雜訊之方法 12
2.5 連通性的區域使用方法 14
第三章　研究方法 16
3.1 研究概念之彩色像素點的向量量化 16
3.2 物件擷取之中值濾波影響 18
3.3 連通像素分割 21
3.4 小波轉換示例及應用 24
3.5 二次擷取 29
第四章　實驗結果與分析 31
4.1 實驗環境及參數設定 31
4.2 實驗影像測試 33
4.2.1 固定群數改變中值濾波遮罩 33
4.2.2 改變群數觀察影像擷取效果 39
4.2.3 加入小波轉換的擷取效果 47
4.2.4 使用二次擷取的實驗擷取效果 51
第五章　結論與建議 53
5.1 結論 53
參考文獻 54

[1]郭慶銳、許建龍、孫樹森、何雲，基於顏色重心和 K-means 的彩色圖像聚類分割演算法，浙江理工大學學報，第 27 卷，第 4 期，2010。
[2] Zikopoulos, P., & Eaton, C.Understanding Big Data: Analytics for Enterprise Class Hadoop and Streaming Data. New York: McGraw-Hill,2011.
[3]楚曉麗，K-means聚類演算法和人工魚群演算法應用於圖像分割技術，計算 機系統應用，第22卷，第4期，92-94頁，2013。
[4] Xu, R., & Wunsch, D.. Survey of clustering algorithms. IEEE Transactions on Neural Networks, 16(3),2005.
[5] M. C. Yang, X. B. Lu and Q. B. Zhou, “Image Segmentation Algorithm Based on Incomplete K-means Clustering and Category Optimization,” Journal of Computer Applications, vol. 32, no. 1, pp. 248-251,2012.
[6] Ra, S.-W., & Kim, J.-K. Fast mean distance-ordered partial codebook search algorithm for image vector quantization. IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, 40(9),1993.
[7]P. Indyk and R. Motwani, “Approximate nearest neighbors: Towards removing the curse of dimensionality,” in Proc. 30th Annu. ACM Symp. Theory Comput.,1998.
[8] Baek, S.-J., Jeon, B., & Sung, K.-M. A fast encoding algorithm for vector quantization. IEEE Signal Processing Letter, 4(12),1997.
[9]J. Wang, H. T. Shen, J. Song, and J. Ji, “Hashing for similarity search: A survey,” arXiv preprint,2014.
[10]K. Terasawa and Y. Tanaka, “Spherical LSH for approximate nearest neighbor search on unit hypersphere,”in Proc. 10th Int. Conf. Algorithms Data Struct,2007.
[11]A. K. Jain, “Data clustering: 50 years beyond K-means,” Pattern Recognit. Lett., vol. 31, no. 8,2010.
[12]Linde, Y.; Buzo, A.; Gray, R. An Algorithm for Vector Quantizer Design. IEEE Transactions on Communications,1980.
[13]Y. Gong and S. Lazebnik, “Iterative quantization: A procrustean approach to learning binary codes,” in Proc. IEEE Conf. Comput. Vis. Pattern Recog,2011.
[14]J. Brandt, “Transform coding for fast approximate. nearest neighbor search in high dimensions,” in Proc. IEEE Conf. Comput. Vis. Pattern Recog,2010.
[15]T. Ge, K. He, Q. Ke, and J. Sun, “Optimized product quantization,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 36, no. 4,2014.
[16]Karayiannis, N. R., and Pai, P. –I., “Fuzzy vector quantization algorithms and their application in image compression,” IEEE Trans. on Image Processing, Vol.4, No.9,1995.
[17]C. C. Chang, W. C. Wu, and Y. C. Hu, “Lossless recovery of a VQ index table with embedded secret data,” Journal of Visual 59 Communication and Image Representation, Vol.18, No.3,2007.
[18]謝慕賢。「快速且有效消除數位影像中1-99%等級之椒鹽雜訊之中值濾波器」。博士論文，國立臺灣科技大學電子工程系，2013。
[19]I. Haritaoglu, D. Harwood and L. S. Davis, “W4: real-time surveillance of people and their activities,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, No.8,2000.
[20]F. Chang, C-J. Chen and C-J. Lu, “A Linear-Time Component-Labeling Algorithm Using Contour Tracing Technique,” Computer Vision and Image Understanding, vol. 93, no. 2,2004.
[21]Y. Zhou and Z. Gao, "Intelligent Recognition of Medical Motion Image Combining Convolutional Neural Network With Internet of Things," IEEE Access, vol.7,2019.
[22]G. Lou and H. Shi, "Face image recognition based on convolutional neural network," China Communications, vol. 17, no. 2,2020.
[23]A. Kumar and Y. Zhou, "Human Identification Using Finger Images," IEEE Transactions on Image Processing, vol. 21, no. 4,2012.
[24]R. C. Gonzalez and R. E. Woods, Digital image processing, 4????hedition, Pearson Publishers, Inc,2018.
[25]林家瑜，以連通區域標記為基礎的自動景深估測方法之研究。﹝碩士論文。國立屏東商業技術學院﹞臺灣博碩士論文知識加值系統，2014。
[26] 潘佳河、李逸仙、郭其昌。連續畫面間小波轉換視訊壓縮技術。電腦與通訊，2003。
[27] Markome。Dossy Shiobara. What is GIMP’s equivalent of Photoshop’s Median filter,2012.