Using sUAS and Multispectral Sensors to Quantify Feral Hog Damage in Forages
Applied Research
Michael Paskewitz
CEA - Agriculture
Melbourne
Abstract
Feral hog populations and their range continue to expand. The USDA estimates feral hog damage to agriculture at around $1.5 billion annually. Accurate damage quantification has proved to be difficult.
The main objectives of this project were to collect whole field imagery of feral hog damage in forages using small Unmanned Aerial System (sUAS) a Matrice 300 RTK (SZ DJI Technology Co., Ltd., Shenzhen China) equipped with a multispectral sensor and to develop a workflow that accurately highlights and efficiently quantifies damaged areas. Mission flights collected imagery using a 5 band Micasense RedEdge sensor (AgEagle Sensor Sysems Inc. Wichita, Kansas 67226) which captured narrow spectral bands of blue, green, red, NIR and Red edge. With these bands, four separate layers were created and compared; the raw Red Edge and NIR bands, a composite RGB layer, and a Normalized Difference Vegetative Index (NDVI). While the RGB layer was easy to visualize, the raw bands and NDVI provided single values for each pixel that facilitated the extraction of quantitative information for comparisons. While the established vegetative indices NDVI provided quantitate values, early attempts were unsuccessful at correlating pixel values to damage; however, disturbed soil turned by feral hogs was distinguishable when looking at the raw Red Edge and NIR imagery.
The team is currently exploring a threshold approach to map this observable damage in an attempt to ultimately develop an index or workflow using a combination of these two bands (Red Edge, NIR) that accurately highlights damage represented as a heat map. The results were used to visualize and quantify the extent of damage across a whole field.
This project demonstrated the potential value of sUAS and multispectral imagery in efficiently quantifying hog damage in forage production systems.
Poster has NOT been presented at any previous NACAA AM/PIC
This poster is being submitted for judging. It will be displayed at the AM/PIC if not selected as a State winner. The abstract will be published in the proceedings.
Click to view Poster
Authors: Michael Paskewitz, Jason Davis
-
Paskewitz, M. CEA - Agriculture, University of Arkansas System Division of Agriculture, Arkansas, 72556
-
Davis, J. Application Technologist, University of Arkansas System Division of Agriculture, Arkansas, 72501