Graduate Thesis Or Dissertation
 

Science Takes Flight: Detection of Black Leg on Turnip Gray Mold on Hemp

Public Deposited

Downloadable Content

Download PDF
https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/h702qf00k

Descriptions

Attribute NameValues
Creator
Abstract
  • Disease detection through traditional techniques such as scouting fields on foot, molecular assays, or morphological identification of plant pathogens is time consuming and costly. Disease diagnosis in the field can be extremely subjective, and largely depends on the experience and knowledge of pathogen identification and disease quantification. This thesis provides an evaluation of remote sensing for assessing disease in agricultural field settings via an unmanned aerial vehicle and machine learning. Case studies are presented on two different fungal diseases important in western Oregon crop production: black leg on turnip (incited by Leptosphaeria spp.) and gray mold on hemp (caused by Botrytis spp.). Both case studies utilized a support vector machine model to classify pixels of digital images collected with a multispectral Micasense RedEdge-M optical sensor. Turnip leaves were imaged at 1.5 m in situ while hemp plant images were collected by an unmanned aerial vehicle with flights at 10 m ex situ. Detection of pixels exemplifying black leg leaf spot symptoms on turnip leaves had an overall accuracy of 97.0% with a model sensitivity of 0.48. The support vector machine model utilized in gray mold detection on hemp incorporated a novel vegetation index, a modified green-red vegetation index along with the triangular greenness index, to identify pixels of diseased hemp inflorescences extracted from background soil and vegetation. The model had an overall accuracy of 95.8% when identifying a hemp inflorescence as diseased or non-diseased. False negatives were found to be high with a sensitivity of 0.70 in the hemp model. Additionally, gray mold disease incidence determined using the support vector machine model was compared with disease assessments collected by scouting on foot and was found to have similar treatment rankings, although the differences in the relative percentages between the two methods were found to be large. The findings of this study provide the foundation for further development of remote sensing techniques for black leg disease assessments in Brassica crops and potential deployment of remote sensing strategies for measuring gray mold in hemp fields.
Contributor
License
Resource Type
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Advisor
Committee Member
Academic Affiliation
Rights Statement
Publisher
Peer Reviewed
Language

Relationships

Parents:

This work has no parents.

In Collection:

Items

Thumbnail Title Date Uploaded Visibility Actions
file details BatesTaylor2021.pdf 2021-06-11 Public Download