動態影像追蹤系統之研製; Design and Implementation of Dynamic Image Tracking Systems

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/48427

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/48427

Title:	動態影像追蹤系統之研製;Design and Implementation of Dynamic Image Tracking Systems
Authors:	王孝瑀;Hsiao-Yu Wang
Contributors:	電機工程研究所
Keywords:	模糊控制;移動物體;景深;fuzzy control;depth;motion detection
Date:	2011-07-01
Issue Date:	2012-01-05 14:54:31 (UTC+8)
Abstract:	活動在生活中是不可或缺的，如果沒有活動就是失去了生活。當我們在活動時各種不同的移動又該如何被發現呢? 移動物體的偵測是一個變因很多的困難課題。對於我們人類要辨別出一個移動物體在何方並不困難，因為我們都擁有銳利的雙眼，我們看到的世界都是三維世界並且我們可以透過聰明的大腦來估測每個不同的移動物體大概離我們的方位和距離。但是我們如何縮短機器科學和聰明的人類之間的差異呢?如果可以縮短之間的差異我們就能建構出一套自動化系統有效的偵測到移動物體。所以本文將針對移動中物體-自製機器鼠,由動態的立體視覺攝影機(stereo vision system)擷取影像，提出一套以改良型鑽石演算法為基礎的影像處理追蹤系統，並用模糊控制對即時影像做出正確追蹤。 Motion is the fundamental of life. No life if no motion. How can be the motion discovered? Motion detection is a very difficult problem for solving. It’s not a big problem for human and animals to solve whether there is or not motion in an observed scene because of their natural ability to see in 3D and to analysis the environment using the potential of their brain. Therefore, the problem is that we cannot create a system, which is similar to psycho physiological parameters of human beings vision. What we can do is to shorten the distance between the ability of human and a technical system, which will be able to discover motion in a scene like human beings, approximately. The purpose of this conference is to create functional and faster models for the robotic rat motion detection with modified diamond search using binocular stereo vision. Using binocular stereo vision with improved diamond search algorithm and frame subtraction to detect out the rat motion and track it in automatic fuzzy control.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	358	View/Open

社群 sharing

Loading...