de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Dokument suchen
Krüger, Lars: Model based object classification and localisation in multiocular images. 2007
Inhalt
Introduction
Motivation
Context
State of the Art
About This Thesis
Aim and Scope
Data and Methods Used
Notational Conventions
Section Overview
The Multiocular Object Recognition System
Methodical Framework
System Overview
Camera Calibration
Appearance Based Methods
Multiocular Template Matching
Feature Pose Maps
Segmentation Methods
Multiocular Active Contours
The Contracting Curve Density Algorithm
Closest Point Methods
Object Recognition using Characteristic Local Features
Object Recognition using Gradient Sign Tables
Experiments and Applications
Evaluating the Camera Calibration
Goal of Investigation
Experimental Setup
Pose Repeatability
Calibration Repeatability
Choice of External Calibration Algorithm
Summary
Oil Cap Inspection
Goal of Investigation
Experimental Setup
Pose Estimation Accuracy
Object Classification Performance
Summary
Obtaining the Trajectory of Tubes and Cables
Goal of Investigation
Experimental Setup
Accuracy of Trajectory Estimation
Summary
Closing Remarks
Summary
A-Priori Questions Answered
Additional Insights
Outlook
Calibration
Template Matching
Feature Pose Maps
Gradient Sign Tables
Trajectory of Tubes and Cables
Other Issues
Appendix
Definitions
Image Tuple
Object Recognition Algorithms
Camera Identifier
Mesh, Facet, Vertex
Appearance Based Object Recognition
Segmentation Based Object Recognition Algorithm
Constraint, Soft Constraint, Hard Constraint
Distance Transform
Epipolar Line
Run Length Encoding
Bibliography