de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
Schliessen
Publizieren
Besondere Sammlungen
Digitalisierungsservice
Hilfe
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Werk suchen
Towards On-The-Fly Image Processing / Alexander Jungmann. [Supervisors: Prof. Dr. Franz-Josef Rammig, Paderborn University ; Prof. Dr. Eyke Hüllermeier, Paderborn University]. Paderborn, 2016
Inhalt
Introduction
On-The-Fly Image Processing
Objectives
Outline and Contributions
Preliminaries
Introduction to Image Processing
Image Manipulation vs. Image Processing
Fundamental Steps in Image Processing
Real-world Application Scenario
Developing Image Processing Solutions
Introduction to On-The-Fly Computing
Principles of Service-Orientation
Service-Oriented Computing
The On-The-Fly Computing Concept
On-The-Fly Composition Process
On-The-Fly Image Processing
Principles of Service-oriented Image Processing
Fundamental Challenges
Adaptivity by Feedback-based Learning
Related Work
Use Cases
Data-Flow and Control-Flow
Data-Flow Graphs as Execution Model
Elementary Net Systems based on Petri Nets
Three Classes of Composed Solutions
Thumbnails for an Online Photo Gallery
Required Functionality
Characteristics
Color-based Segmentation
Concrete Context
Required Functionality
Characteristics
Motion-based Object Detection
Concrete Context
Required Functionality
Characteristics
Summary
Symbolic Service Composition
Knowledge-based Specifications
Body of Knowledge
Service and Request Specification
Specification Example: Thumbnails
Specification Example: Segmentation
Planning-based Service Composition
Composed Services
Body of Rules
Formal Framework
Composition Algorithm
Composition Example: Thumbnails
Shortcomings and Extensions
Exponentially Growing Solution Space
Incorrect Task Definitions
Superfluous Search Paths and Services
Discarding Properties of Visual Data
Outlook: Necessity for Learning
Evaluation
Prototypical Implementation
Concrete Composition Problem
Search Space and Solution Space
Time to Solution
Conclusion
Related Work
Execution and Rating
Service-oriented Architecture for Execution
Key Concepts and Building Blocks
Integration into OTF Image Processing
Problem Domain specific Rating Processes
Preliminary Considerations
Segmentation Use Case
Object Detection Use Case
Evaluation
Segmentation of Color Palette
Motion-based Robot Detection
Motion-based Ball Detection
Conclusion
Adaptive Service Composition
Learning Recommendation System
Reinforcement Learning
Recommendation Model
Learning Process
Combining Composition and Recommendation
Overview and Interactions
Update Step
Evaluation Step
Modified Search Node Selection
Episode Finalization
Evaluation
Segmentation of Color Palette
Motion-based Robot Detection
Motion-based Ball Detection
Conclusion
Related Work
Conclusion and Outlook
Future Work
List of Figures
List of Tables
List of Algorithms
Own Publications
Bibliography
Die detaillierte Suchanfrage erfordert aktiviertes Javascript.