학술논문

Comparison of Automatic Geometric Correction Methods for Orthorectified Image Stacks
Document Type
Conference
Source
IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium Geoscience and Remote Sensing Symposium, IGARSS 2023 - 2023 IEEE International. :5926-5929 Jul, 2023
Subject
Aerospace
Components, Circuits, Devices and Systems
Fields, Waves and Electromagnetics
Geoscience
Signal Processing and Analysis
Earth
Data analysis
Computational modeling
Geology
Geoscience and remote sensing
Data models
Computational efficiency
Block adjustment
coregistration
mosaicking
Language
ISSN
2153-7003
Abstract
This work investigates the absolute geometric accuracy, computational efficiency, and trade-off between two common techniques used for improving the geometric consistency of orthorectified image stacks: Coregistration (CR) and Bundle Block Adjustment (BBA). A semi-analytical model is developed and Sentinel-2 data is used to derive realistic model parameters to quantify the differences between these techniques. The study reveals that while BBA shows significantly smaller errors than CR, it also demands substantially more computational resources. This trade-off thus renders the choice of technique highly dependent on the specific application and computational feasibility. By presenting empirical geometric accuracy curves, this paper provides insights that can help inform the selection of georeferencing techniques in the Earth Observation domain.