Webinar | Automated Workflows and Machine Learning Techniques for Coastline Extraction

[KelleyArroyo]

Automated Workflows and Machine Learning Techniques for Coastline Extraction

Learn how to use ArcGIS automated workflows and machine learning techniques for coastline extraction.

Due to anthropogenic activities and natural processes—for example, sea level changes, sedimentation, and wave energy—coastlines are changing worldwide. Traditionally, coastlines were manually digitized, which is a time and labor-intensive way. Remote sensing is an excellent alternative to extract coastlines, using satellite imagery. Satellite imagery of visible range can be used for interpretation and easily obtained. But the imageries covering infrared wavelength is best to extract the boundary between land and water. The band ratio technique is easy to calculate and gives highly accurate results in less processing time. The workflow can be applied to any area using multispectral imagery, Landsat 8 or Sentinel-2.

Have Questions? Post any questions you have about this webinar in the "Reply" area below and our Esri Maritime team will get back to you with answers. 

Additional Resources:

Presentation Deck

Map scale and raster resolution 

Story map: Coastline extraction using Landsat-8 (Image Server) as source

Sample Notebook: Coastline extraction using Landsat-8 (Image Server) as source

Esri Hydrographic Office 

2020 GeoAI webinar: Hydrographic Applications of AI and Deep Learning

GeoAI Ship Wrecks demo application

Products on Demand Custom Charts- demo application

Maritime Chart Service ENC- demo application

St. Thomas Bathymetry-demo application

Maritime Esri Community

Contact at: maritime@esri.com

[Felix10546]

Hello. Thank you for providing detailed material.

I encountered an exception when I try to run code #19 in

Sample Notebook: Coastline extraction using Landsat-8 (Image Server) as source

b2_poly = convert_raster_to_feature(dra2, field='Value', output_type='Polygon', simplify=True, output_name=None, gis=gis2)

OutputCatalogPath failed. Error: <built-in method GetOutputCatalogPath of HostedGP object object at 0x0000000038F3BD08> returned NULL without setting an error
{"messageCode": "RA_120212", "message": "ConvertRasterToFeature failed. OutputCatalogPath failed. Error: <built-in method GetOutputCatalogPath of HostedGP object object at 0x0000000038F3BD08> returned NULL without setting an error."}
Failed to execute (ConvertRasterToFeature).
Failed.

Is there any workaround?

[ShivaniPathak]

Hi Felix,

Are you using Python API playground portal as gis in the notebook? If this is the case, currently Python API Playground doesn't has configured RA server.

[Felix10546]

Hi Shivani, thank you for noticing the cause of the problem.

[Aditya_Allamraju]

Hello Kelley,

Thanks a lot for providing all the materials. I am working on a similar study for my graduate course project. I tried to run the code

selected = landsat.filter_by(where="(Category = 1) and (CloudCover<0.03)", time=[datetime(2016, 1, 1), datetime(2019, 12, 31)], geometry=arcgis.geometry.filters.intersects(aoi_geom))

from Sample Notebook: Coastline extraction using Landsat-8 (Image Server) as source

But I got a key error & an attribute error. I understand the error but need some assistance to resolve it as I am new to ArcGIS API for Python.

I would appreciate your help.

Thanks