About Site Language

Magni eaque corporis laboriosam vitae placeat obcaecati quaerat nam. Perferendis, doloremque repudiandae? Tempore quod illum quae autem reiciendis.

I changed my language, but I’m still seeing resources in the other languages?

Quibusdam, error nobis harum accusamus laborum neque unde! Blanditiis inventore ad voluptatibus sapiente, natus tempore nam ullam quidem.

WHDL - 00013545

Using Machine Learning to Improve Data Collection of Wildland Fires in Forested Areas

Kamden Brothers

Resource Types

Academic (Thesis, Dissertation, Etc.)

Resource Languages

English

Institutional Repository

Description / Abstract

Submitted to the Department of Mathematics & Computer Science in partial fulfillment of the requirements for the degree of Bachelor of Science

Using machine learning algorithms on imagery obtained from small unmanned aircraft systems (sUAS) has been an efficient and accurate way to collect data on postfire forests. This effort applies machine learning to obtain useful information about postfire forests. It uses a mask region-based convolutional neural network (MR-CNN) to as well as a support vector machine (SVM) to tree mortality as well as burn extent. Using machine learning helps automated the process while still having accurate data. Having fast and accurate process to calculate the damage done by a fire helps land managers make a quick and calculated response to aid in forest rehabilitation.