de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Dokument suchen
Osterloff, Jonas: Computer Vision for Marine Environmental Monitoring. 2018
Inhalt
Abstract
Acknowledgment
Acronyms
Introduction
Motivation
Computer Vision for Underwater Environmental Monitoring
Image acquisition challenges
Annotation challenges
Example
Contributions
Publications
Structure
Notation
Computer Vision
Digital image
Color spaces
RGB color space
CIE L*a*b* and CIE L*u*v color space
HSV color space
HSI color space
l alpha beta color space
YCbCr color space
Feature extraction
Local statistics
Histogram features
Canny edge features
Gabor features
Data-driven features
Feature normalization
Feature selection
Augmentation
Machine learning
Unsupervised machine learning
Supervised machine learning
Deep learning
Training and tuning
Post-processing
Evaluation
Manual annotation of underwater images
Manual annotation
Quality assessment
Gold standard
Correctness
Reproducibility
BIIGLE
Underwater imaging
Digital imaging
Underwater in-situ imaging
Fixed underwater observatories
Underwater wet-lab imaging
Preprocessing for underwater images
Overview of underwater image preprocessing methods
Applied preprocessing methods
Auto white balancing
Fixed white balancing
Histogram equalization
Contrast limited adaptive histogram equalization
Automatic color equalization
Local space average color scaling
White Patch Retinex and Gray World combination
Unsupervised color correction
Automated underwater image preprocessing
Feature space based illumination and color enhancement
l alpha beta underwater color correction
Assessment
Cluster index ranking
Computer vision for in-situ monitoring
Sessile species: Lophelia pertusa
Material
Task 1: color monitoring
Task 2: polyp activity monitoring
Results
Discussion
Conclusion
Mobile species: shrimp
Material
Methods
Results
Discussion
Conclusion
Computer vision for wet-lab experiments
Sessile species: Mesophyllum engelhartii
Material
Methods
Results
Discussion
Conclusion
Data science for environmental monitoring
In-situ: Lophelia pertusa
Material
Methods
Results
Discussion
Conclusion
Wet-lab: Mesophyllum engelhartii
Material
Methods
Results
Discussion
Conclusion
Discussion
Computer vision systems
Data science
Conclusion
Future perspectives
Bibliography