Date of Award

August 2020

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Division of Agriculture (SAFES)

Committee Member

John D Mueller

Committee Member

Michael T Plumblee

Committee Member

Michael W Marshall

Abstract

Multiple linear regression models were developed to predict sand and clay content along with soil organic matter content from RGB imagery from both commercially available satellite imagery as well as RGB UAV imagery. UAV Imagery was tested at two flight altitudes to determine if lower or higher altitude had an effect on prediction. In cases of sand, clay, and OM content, flight altitudes did not significantly differ in prediction abilities. Satellite imagery was evaluated using data from Planet Labs as well as Google Earth. Regression models were developed to predict sand, clay, and soil organic matter content from these satellite images, which captured fields with bare soil. An alternative to whole field data collection, referred to herein as the point sampling method, was introduced. A survey of currently available neural network and machine learning technologies was performed to establish which of these technologies could benefit the precision agriculture industry. A sample model was trained to detect and classify cotton blooms from low-altitude RGB imagery collected from a DJI Phantom 3 UAV.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.