Real Time Non-rigid Surface Detection Based on Binary Robust Independent Elementary Features

碩士 === 國立勤益科技大學 === 資訊工程系 === 101 === The surface deformation detection of an object has been a very popular research project in recent year, in human vision, we can easily detect location of the target and that scale of the surface、rotation and change of the viewpoint makes the surface deformation;...

Full description

Bibliographic Details
Main Authors: DianYu-Ren, 田育任
Other Authors: Wang Chuin-Mu
Format: Others
Language:zh-TW
Published: 2013
Online Access:http://ndltd.ncl.edu.tw/handle/82996553848695716088
Description
Summary:碩士 === 國立勤益科技大學 === 資訊工程系 === 101 === The surface deformation detection of an object has been a very popular research project in recent year, in human vision, we can easily detect location of the target and that scale of the surface、rotation and change of the viewpoint makes the surface deformation; but in vision of computer is a challenge. In those background of question, we can propose a framework that is the surface deformation which is based on detection method of BRIEF to calculate object surface deformation. But BRIEF calculation has some problem that can’t be rotate and change character , we also propose a useful calculation method to solve the problem, and method proved by experiment can overcome the problem, by the way, it’s very useful. The average operation time every picture in continuous image is 50~80 millisecond In 2.5 GHz computer, let us look back for some related estimation technology of surface deformation, and there are still a few successful project that is surface deformation detection in document.