橢圓幾何參數鑑別之理論模擬及實作; Direct Least Square Fitting of Five Parameters of Ellipse

NCUIR > Engineering College > Graduate Institute of Mechanical Engineering > Electronic Thesis & Dissertation > Item 987654321/2464

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/2464

Title:	橢圓幾何參數鑑別之理論模擬及實作;Direct Least Square Fitting of Five Parameters of Ellipse
Authors:	陳章沂;Chang-Yi Chen
Contributors:	機械工程研究所
Keywords:	橢圓參數鑑別;最小平方差法;least square;fitting;ellipse
Date:	2004-07-01
Issue Date:	2009-09-21 11:47:38 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	在本篇論文中將提出一個快速而強健的方法，來求取橢圓的中心座標、旋轉角、兩軸長，這五個參數。並建構一個新的影像擷取和影像處理平台，以用來實作本文所提出的演算法。橢圓偵測的方法已經有許多人研究過且提出了很多方法，大多是利用全域幾何的特性，也有全域和局域的幾何特性兼用的。Hough Transform(以下簡稱為HT)也常被研究來偵測橢圓。本文示出演算法的推導過程、對演算法的效果進行測試、提出演算法的可能應用方向和實作演算法對實物圖形進行橢圓偵測。本文會簡介影像擷取平台的設計，從平台的特色、硬體架構到系統規劃，並秀出實際所取得的照片。 This paper proposes a new formulism of least square problem that makes use of only five parameters to fit ellipses. These five parameters are one to one mapping of the five geometric characteristic parameters of ellipses. Therefore, traditional least square algorithms can be used to calculate the parameters without any modification. This new approach is the first ellipse identification approach that has definite computation time, so that it is available for real time applications. Furthermore, we also provide the procedure to transform the result parameters to the more informative geometric parameters, the center coordination, the length of two axes, and the inclined angle of the major axis. This innovatory approach in least square identification of ellipses is much more computationally efficient and can easily be implemented recursively.
Appears in Collections:	[Graduate Institute of Mechanical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Size	Format

社群 sharing

Loading...