• 沒有找到結果。

應用類神經網路於多角曲線之研究 張崑麒、鍾翼能

N/A
N/A
Protected

Academic year: 2022

Share "應用類神經網路於多角曲線之研究 張崑麒、鍾翼能"

Copied!
2
0
0

加載中.... (立即查看全文)

全文

(1)

應用類神經網路於多角曲線之研究 張崑麒、鍾翼能

E-mail: [email protected]

摘 要

本論文將利用競爭型Hopfield類神經網路的方法應用於多角曲線及二維影像輪廓辨識上之應用。首先,找尋輪廓上的特徵 點,特徵點的位置所在則包含輪廓上高曲線及轉角的部分,找出特定點後我們以多邊型近似的方法來描述圖形。就個別輪 廓上所擁有的若干特徵運算最佳之轉折點,進而尋多角曲線之近似圖形,降低近似誤差,此法可節省電腦記憶容量,並可 運用於目標追蹤平滑化,股市K線圖或各種曲線之近似圖形。

關鍵詞 : 競爭型Hopfield類神經網路 ; 多邊型近似 ; 目標追蹤平滑化 目錄

封面內頁 簽名頁 授權書.........................iii 中文摘要............

............iv 英文摘要........................v 誌謝.........

................. vi 目錄..........................vii 圖目錄..

.......................ix 第一章 緒論 1.1前言....................

.1 1.2理論應用.................1 1.3研究動機...................2 1.4論文 架構...................3 第二章 類神經網路理論 2.1前言.................

....4 2.2類神經網路簡介................7 2.3神經元介紹.................

.9 2.4類神經網路的分類 ..............11 第三章 Hopfield類神經網路 3.1理論基礎 .........

.........14 3.2 Hopfield模型 ................15 3.3 Lyapunov函數..........

......18 3.4類神經網路設計 ...............19 第四章 競爭型Hopfield類神經網路 4.1前言 ..

..................25 4.2類神經網路模型 ...............26 第五章 模擬結果 5.1 前言....................29 5.2應用CHNN之曲線近似模擬..........29 第六章 結論 與建議...................33 參考文獻........................35 圖目錄 圖2.1人工神經元模型 ..................6 圖3.1 Hopfield模型 .............

......15 圖3.2 Hopfield網路 ...................17 圖3.3人工神經元模型........

..........20 圖3.4循環網路.....................23 圖5.1目標軌道曲線原圖 (取 樣100點)..........30 圖5.2目標軌道曲線近似圖(取樣50點)..........30 圖5.3曲線近似圖(

取樣22點)..............31 圖5.4目標軌道曲線...................31 圖5.5軌道 近似曲線(取樣50點).............32 圖5.6軌道近似曲線(取樣22點).............32 參考文獻

[1] S. Haykin,“Neural Networks: a comprehensive foundation 2nd edition.”Prentice Hall, 1999, pp.664-727.

[2] N. M. Nasrabadi, W. Li,“Object Recognition by a Hopfield Neural Network.”IEEE Trans. SMC, Vol.21, No.6, 1911, pp.1523-1535.

[3] P. N. Suganthan, E. K. Teoh, D. P. Mital,“Programming Hopfield Network for Object Recognition.”in Proc. Of SMC Conf., 3, 1993, pp.114-119.

[4] P. N. Suganthan, E. K. Teoh, D. P. Mital,“Homomorphic ARG Matching by Hopfield Network.”in Proc. IEEE Int. Conf. Industrial Electronics, Vol.1, 1995, pp.161-165.

[5] M. N. Fu, H. Yan,“A Shape Classifier based on Hopfield-Amari Network.”in Proc. IEEE Int. Conf. Neural Network, Vol.1, 1996, pp.558-593.

[6] W. J. Li, T. Lee,“Hopfield Neural Network for Affine Invariant Matching.”IEEE Trans. Neural Networks, Vol.12, No.6, 2001, pp.1400-1410.

[7] W. J. Li, T. Lee,“Object recognition and articulated object learning by accumulative Hopfield matching.”Pattern Recognition, 35, 2002, pp.1933-1948.

[8] D. L. Lee,“Pattern Sequence Recognition Using a Time Vary Hopfield Network.”IEEE Trans. Neural Networks, Vol.13, No.2, 2002, pp.330-342.

(2)

[9] 葉怡成,“類神經網路模式應用與實作.”儒林圖書有限公司, 2002.

[10] 張嘉鍇,“在低解析度六角格子影像上之輪廓辨識.”國立中山大學機械工程研究所碩士論文, 2000.

[11] 李祐魁,“利用次像素在低解析度六角格子上作輪廓辨識.”國立中山大學機械工程研究所碩士論文, 2000.

[12] 黃國源,“類神經網路與圓型辨識.”維科出版社, 2000.

[13] 蕭富介,“類神經網路應用於瓦特I型六連桿組耦點曲線之合成.” 國立中山 大學機械工程研究所碩士論文, 2002.

[14] M. Cooper,“Visual occlusion and the interpretation of ambiguous.”ELLIS HORWOOD,1992.

[15] M. Egmont-Peterson, D. de Ridder, H. Handels,“Image processing with neural networks-s review.”Pattern Recognition, 35, 2002, pp.

2279-2301.

[16] J. S. Lee, C. H. Chen, Y. N. Sun,G. S. Tseng,“Ocluded objects recognition using multiscale features and Hofield neural network.”Pattern Recognition, Vol.30, No.1, 1997, pp.113-122.

[17] 王進德, 蕭大全,“類神經網路與模糊控制理論入門.”全華科技圖書股份有限公司, 1994.

[18] 林昇甫, 洪成安,“神經網路入門與圖樣辨識(修訂第二版).”全華科技圖書股份有限公司, 2002.

[19] M. N. Fu, H. Yan, K. Huang,“A curve band fuction based method to characterize contour shapes. ” Pattern Recognition, Vol. 31, No. 10, 1997, pp.1661-1671.

[20] J. H. Kim, S.H.Yoon, C. W. Lee, K. H. Sohn,“A robust solution for object recognition by mean field annealing techniques.”Pattern Recognition, 34, 2001, pp. 885-902.

[21] Y. Uchiyama, M. Haseyama, H. Kitajima,“Hopfield neural networks for edge dete detection.”ISCAS 2001. Vol.3, 2001, pp. 608-611.

[22] Rosenfield, E. Johnston,“Angle detection on digital curves.”IEEE Trans. Compute. Vol. C-22, 1973, pp. 875-878.

參考文獻

相關文件

Because simultaneous localization, mapping and moving object tracking is a more general process based on the integration of SLAM and moving object tracking, it inherits the

Tekalp, “Frontal-View Face Detection and Facial Feature Extraction Using Color, Shape and Symmetry Based Cost Functions,” Pattern Recognition Letters, vol.. Fujibayashi,

grep - print lines matching a pattern. $ grep [OPTIONS]

Deep learning usually refers to neural network based model.. Shallow – Speech Recognition. ◉

Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp... Annealed

蔣松原,1998,應用 應用 應用 應用模糊理論 模糊理論 模糊理論

F., “A neural network structure for vector quantizers”, IEEE International Sympoisum, Vol. et al., “Error surfaces for multi-layer perceptrons”, IEEE Transactions on

Jones, "Rapid Object Detection Using a Boosted Cascade of Simple Features," IEEE Computer Society Conference on Computer Vision and Pattern Recognition,