標題: 利用光流技術於數位影像平移與轉動運動之估測
DIGITAL IMAGE TRANSLATIONAL AND ROTATIONAL MOTION ESTIMATION USING OPTICAL FLOW TECHNIQUE
作者: 張柏森
Bo-Sen Chang
張志永
Jyh-Yeong Chang
電控工程研究所
關鍵字: 數位影像穩定;運動估測;光流;Digital image stabilization;Motion estimation;Optical flow
公開日期: 2000
摘要: 本論文提出一種以光流(optical flow)技術為基礎的數位影像序列穩定系統,在影像運動的估測上包含了平移運動與轉動運動。系統可分為三個部分:(1) 擾動運動的估測、(2) 估測向量的抽取、(3) 影像的補償修正。首先,在運動估測中我們使用疊代的光流演算法在每二相鄰影像序列中得到一組局部影光流動向量,每一像素皆有一對應的流動向量。其次在估測向量抽取部分,根據運動的模組,決定出最適當的平移運動量與轉動運動量。最後,在影像的修正與補償部分,去除原影像中不必要的擾動成分,使穩定後的影像序列能達到清晰穩定的效果。
This thesis proposes a digital image sequence stabilization system based on optical flow techniques. The system includes three major modules: (1) disturbance motion estimation, (2) motion vector extraction, and (3) image correction and compensation. First, in motion estimation module, we employ the optical flow technique to obtain the local motion vector field. The outputs of optical flow estimation are the velocity vectors corresponding to every pixel in the current image frame. The motion vector extraction module decides the applicable components of translation and rotation motion according the motion model. Finally, the image correction and compensation module integrates the field motion vectors and removes the residue shake motion.
URI: http://140.113.39.130/cdrfb3/record/nctu/#NT890591087
http://hdl.handle.net/11536/67857
Appears in Collections:Thesis