Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/1911
標題: 高斯球在多面體形貌影像分割之研究
Polyhedral Range Image Segmentation Based on Gauss-Sphere
作者: 余加新
Yu, Chia-Hsin
關鍵字: 形貌影像;影像分割;區域合併;區域相鄰關係圖
出版社: 機械工程學系
摘要: 
電腦視覺系統從3D世界中所取得龐大的影像資訊,要如何轉換成有用的資訊。若直接藉由電腦來闡釋原始影像資訊是有困難的,通常最適宜的方式先做影像切割(Image segmentation)的步驟,將影像分割成一件件的低階元件來做後續的處理,所謂低階元件就是指藉由具有相同特性的像素群體組成的。而本論文目的就是建立一套3D形貌影像切割演算法,切割演算法大致分為初略切割與區域合併兩大部分。首先,使用法向量的角度及原始的深度值作初略的切割,接著做區域合併,區域合併是採用同性質的區域作合併,直到設定的性質在誤差範圍內。文中之實驗部分所用到的輸入影像是用電腦模擬形貌影像的資料,而影像中元件係以多面體的幾何形狀組成,切割的效果以影像來呈現整個演算法的過程及結果。

In this paper we present a region based segmentation algorithm for range image. The algorithm can be viewed as a two-stage process: (1)initial segmentation, (2)region merge. First, we segment surface into a set of regions in the range image by using orientation of the normal vector's angle and original range value. Second, we perform an region merge using the same property region to merge, until setting the property is within error scope. Experimental results show the algorithm performance on images.
URI: http://hdl.handle.net/11455/1911
Appears in Collections:機械工程學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.