A Combined Method for Vegetation Classification Based on Visible Bands from UAV Images: A Case Study for Invasive Wild Parsnip Plants

Loading...
Thumbnail Image

Authors

Liu, Jingyi

Date

Type

thesis

Language

eng

Keyword

UAV , Wild Parsnip , Random Forest , Image Classification

Research Projects

Organizational Units

Journal Issue

Alternative Title

Abstract

Wild parsnip is an invasive plant that has serious health risks to humans due to the toxin in its sap. Monitoring its presence and spread has been challenging for conservation authorities due to its small size and irregular shape. On traditional remotely-sensed images acquired by satellites or airplanes, wild parsnip cannot be distinguished from other vegetation due to low spatial resolution of the imagery. Unmanned Aerial Vehicles (UAV) can obtain ultra-high resolution (UHR) imagery and have been used for vegetation-related monitoring in both environmental and agricultural applications in recent years. In this study, UAV images captured at Lemoine Point Conservation Area in Kingston, Ontario, are used to test a methodology for distinguishing wild parsnip. The objective of this study is to develop an efficient invasive wild parsnip classification workflow based on UHR digital UAV imagery. The challenge is that all processing options are based only on visible bands information (RGB) of the digital cameras on UAV, which means there is no infrared spectral information to use. The UAV image processing flow included image orientation, digital elevation model (DEM) and digital surface model (DSM) extractions, individual orthophoto production, and orthomosaic generation processing using Simactive’s software. Three vegetation indices and three texture features are calculated and the first two significant features are added to the mosaicked images. The Random Forest algorithm with relevant variables is used as the classifier to distinguish wild parsnip plants from other vegetation types. The optimal image resolution in parsnip analyses are undertaken by comparing accuracy assessments. The results provide an executable workflow to distinguish wild parsnip and show that UAV images, with a simple digital-light camera, are an appropriate and economic resource for small and irregular vegetation detection. The combined method yields reliable and valid outcomes in detecting wild parsnip plants and shows good performance in mapping vegetation.

Description

Citation

Publisher

License

Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
ProQuest PhD and Master's Theses International Dissemination Agreement
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
CC0 1.0 Universal

Journal

Volume

Issue

PubMed ID

External DOI

ISSN

EISSN