515 Rate this article:
4.6

Applying ENVI Deep Learning for Disaster Response

Challenge

It is imperative to quickly assess damage after a natural disaster to aid rescue efforts and start the rebuilding process. Disaster response organizations usually receive access to satellite, aerial, and UAV imagery to help with this task. However, manually looking at images to assess damage and make other critical decisions is difficult and time consuming. Even with automated analytics, this process can be daunting if there is a large cache of imagery and data.


Haiti after earthquake An image from Haiti after an earthquake struck

Solution

The ENVI Deep Learning module has intuitive tools and workflows that require no programming, enabling users to create accurate deep learning models and algorithms in a matter of minutes. This allows disaster responders to quickly preprocess, label, train, and classify imagery to make critical decisions. The ENVI Deep Learning module is used after a disaster to assess and quantify damage, locate and identify hazards, evaluate access to infrastructure, map flooding and much more.


Training labels The ROI tool within ENVI was used to create training data with just a few clicks.

Benefits

The ENVI Deep Learning module can be used to quickly and accurately extract actionable information, which accelerates response times.  This tool improves awareness which enables relief organizations to make safer and more informed decisions as the situation unfolds. In the aftermath of an event, the ENVI Deep Learning module is also valuable in the support of rebuilding and recovery efforts.


Output classification The deep learning results show the probability of buildings that have been destroyed and are now rubble.


If you have a problem you think can be solved using deep learning, reach out to us at GeospatialInfo@L3Harris.com and one of our experts will discuss it with you.

DISCOVER MORE

ENVI DEEP LEARNING MODULE

AUTOMATE ANALYTICS

Automate analytics with deep learning for fast, accurate results.

Creating Accurate Digital Terrain Models

RECORDED WEBINAR

A technical deep dive into how to create accurate digital terrain models (DTMs)

OBJECT TRACKING & SITUATIONAL AWARENESS

Situational Awareness Webinar

RECORDED WEBINAR

Learn how real-time intelligence complements situational awareness to solve localized threat problems.

 

 

 

 

 

Please login or register to post comments.