臺灣博碩士論文加值系統

一張影像的邊緣可以提供給其他較複雜的程序，像是物體識別，相當重要的資訊，所以，影像邊緣偵測是影像處理中基本且重要的技術。有很多傳統的方法是用影像與遮罩做旋積，可以算出影像中每個像素的梯度值，再根據梯度值來找出該影像的邊緣。傳統的邊緣偵測方法很容易受到雜訊的干擾，因此這篇論文主要探討利用數學結構法來偵測一張受雜訊干擾影像的邊緣。數學結構法提供了抗雜訊的邊緣偵測法，近年來也有越來越多基於數學結構法的邊緣偵測法被提出。
我們分析在數學結構法中用兩種不同結構元素輸出的影像優缺點，將這兩張影像加起來得到一張灰階圖，再根據不同的門檻值將灰階圖轉成二值圖。然後，我們提出一個合作賽局理論架構來適當地選出兩張二值圖。最後，我們用提出的方法將這兩張影像結合出一張影像，在結合的過程中，我們偵測出比傳統方法更為完整的邊緣，還把影像中孤立的邊緣點去除，來減少錯誤的資訊。

Chapter 1 Introduction…………………………………………………1

Chapter 2 Related Works…………………………………………………3

2.1 Berkeley Segmentation Dataset and Benchmark……………3

2.2 Classic Operator……………………………………………………4

2.3 Mathematical morphology…………………………………………6

Chapter 3 Proposed Methods…………………………………………11

3.1 The effect of structure element on edge detection…11

3.2 Choosing threshold with cooperative game………………12

3.3 Image fusion with proposed algorithm……………………17

Chapter 4 Experiment Results………………………………………20

Chapter 5 Conclusions…………………………………………………32

Reference……………………………………………………………………33

[1]Rafael Gonzalez and Richard Woods, “Digital Image Processing”2nd ed., Prentice Hall, 2002
[2] Chen juan, Chen qian-hui, Shi lu-huan. “Edge Detection Technology in Image Tracking” Chinese Journal of Optics and Applied Optics.Vol.2. No.1, pp.46-53, 2009
[3] Wang Hui-feng, Zhan Gui-li , Lu Xiao-ming. “Research and application of edge detection operator based on mathematical morphology” Computer Engineering and Applications, pp. 223 – 226, 2009
[4] Dong Wei, Wang Tong. “Research on Edge Detection Based on Mathematical Morphology Algorithm”Optoelectronics and Image Processing (ICOIP), Vol. 2,issue.11-12, pp. 211 – 213, 2010
[5]ShiHu Zhu. “Edge Detection Based on Multi-structure Elements Morphology and Image Fusion” Computing, Control and Industrial Engineering (CCIE), 2011 IEEE 2nd International Conference, Vol. 2,issue.20-21, pp. 406 – 409, 2011
[6]Hailong Huang, Hong Wang, Fan Guo,Jinfeng Zhang.“A Gray-scale Image Edge Detection Algorithm Based on Mathematical Morphology” Measuring Technology and Mechatronics Automation (ICMTMA), 2011 Third International Conference, Vol. 1 ,issue.6-7, pp. 46 – 52, 2011
[7]Chen Jinlong, Zhang Bin, Qi Yingjian, Jin Fei, Si Xuan. “Image Edge Detection Method Based on Multi-Structure and Multi-Scale Mathematical Morphology” Multimedia Technology (ICMT), 2010 International Conference, issue.29-31, pp. 1 – 4, 2010
[8]David Martin et al., “A Database of Human Segmented Natural Images and Its Application to Evaluating Segmentation Algorithms and Measuring Ecological Statistics” Computer Vision, 2001. ICCV 2001.Proceedings. Eighth IEEE International Conference, Vol. 2, pp. 416–423, 2001,
[9]Komati,Salles, Sarcinelli-Filho.“KSS: Using Region and Edge Maps to Detect Image Boundaries” Computing in Science &; Engineering, Vol. 13 , issue.3, pp. 46 – 52, 2010
[10]YanyanCao, HengjunYue, JianWu, JianmingChen, Zhiming Cui. “Vehicle Edge Detection Based on Multi-structure Mathematical Morphology” Intelligent Information Technology and Security Informatics (IITSI), 2010 Third International Symposium, issue.2-4, pp. 258 – 262, 2010
[11]XuGuo-bao, Wang Ji, Zhao Gui-yan, Yin Yi-xin,Xie Shi-yi.“Adaptive algorithm of edge detection based on mathematical morphology” Journal of Computer Applications, pp. 997 – 999, 2009