UNet Deep Learning Image Classification using the ArcGIS...

 

Full Title: UNet Deep Learning Image Classification using the ArcGIS Pro Image Analyst Extension

Instructor(s)

Amr Abd-Elrahman, University of Florida

This workshop introduces:

  • Pixel- and object-based image classification using traditional (Support Vector Machine and Random Forests) machine learning algorithms
  • Building models for data preparation and experiment with different classification parameters
  • Deep learning semantic segmentation using the UNet Model applied on wetland land cover classification
  • Participants will be briefly introduced to necessary theoretical background information as well as practical implementation. The workshop will present a complete workflow with clear implementation instructions.

Outline:

  • Introduction to image classification concepts and machine learning algorithms
  • Preparing training datasets for the UNet deep learning model
  • Training the UNet Model
  • Image classification and accuracy assessment using the UNet Model

To access the meeting/recording, register here and then visit the ASPRS Online Learning Catalog after your order has been completed.
When
5/20/2021 10:00 AM - 2:15 PM
Eastern Daylight Time
Where
Virtual

A MyASPRS account is required to access services, register for events, and purchase products from ASPRS. You do not have to be an ASPRS member to create a MyASPRS account. If you wish to join ASPRS to receive member benefits, including discounts on most products and services, you will have the option to join once you have created your MyASPRS account.

All ASPRS members have MyASPRS accounts. If you do not remember your username or password, use the links below to reset. An email will be sent to your primary email address in your member profile. If you no longer have access to this email, please contact [email protected] or call 301-493-0290 for assistance.

To register, Sign In with your ASPRS Account or Create Account

Run into any problems with the new member website? Please contact the ASPRS office at (225)-408-4747 or by email at [email protected] to let us know!