Classification of Bathymetry Grids �Using Open Source Tools
Google Ocean
Kurt Schwehr, Jamie Adams, Jenifer Austin Foulkes
http://earth.google.com/ocean
http://maps.google.com/ocean
http://schwehr.org/blog
Classifications
What is a hole?
What is the boundary?
What artifacts?
Source & relationship
Open Source Software
proj
gdal
qgis
grass
gmt
mbsystem
...
python
ipython notebook
lxml
numpy
scipy measurements
shapely
opencv
...
A call for open data formats and more release data with open formats
Bathymetry
Lidar
...
GSHHS & other shorelines
Raw tide records
...
SAIC's Generic Sensor Format (GSF) library is NOT currently licensed as open source software!
Abstract Title: Classification of Bathymetry Grids Using Open Source Tools
is part of the Paper Session:Advances and Challenges in Digital Elevation Models I (Overview)
Author(s):Kurt Schwehr, PhD* - Google, Jamie Adams - Google, Jenifer Austin Foulkes - Google
Abstract:
Creating global synthesis views of the Earth's bathymetry is a challenge complicated the process of merging data products from diverse sensor platforms with a wide range of data artifact classes. Processing large numbers of gridded bathymetry DEMs requires being able to automatically classify the input DEMs based on the surveying and gridding techniques used and the resulting artifacts. The platform type and details of techniques used are not detailed in a machine readable form within the ISO XML metadata contained in Bathymetry Attributed Grids (BAGs). We demonstrate the results of processed NOAA NGDC's archive of BAGs using Open Source tools to identify the quantity and morphology of data gaps using the Python SciPy library's image processing routines. Once grids have been classified and referenced to the same vertical datum using the Geospatial Data Abstraction Library (GDAL), the grids can be hole filled and merged based on project specific requirements. We will discuss the general classes of artifacts that can be found and propose how each class might be handled to produce a more continuous surface. We show how to use IPython Notebooks and QGIS to assist with quality checking BAGs insure the archived grids represent the quality of the sensor platform and acquisition strategy. We will conclude with suggested strategies for data acquisition and gridding that are more likely to produce DEMs that blend well with large global scale projects such as Google Ocean.
Keywords: