A Deep Learning Application to Map Weed Spatial Extent from Unmanned Aerial Vehicles Imagery

A - Papers appearing in refereed journals

Fraccaro, P., Butt, J., Edwards, B., Freckleton, R. P., Childs, D. Z., Reusch, K. and Comont, D. 2022. A Deep Learning Application to Map Weed Spatial Extent from Unmanned Aerial Vehicles Imagery. Remote Sensing. 14 (17), p. 4197. https://doi.org/10.3390/rs14174197

AuthorsFraccaro, P., Butt, J., Edwards, B., Freckleton, R. P., Childs, D. Z., Reusch, K. and Comont, D.
Abstract

Weed infestation is a global threat to agricultural productivity, leading to low yields and financial losses. Weed detection, based on applying machine learning to imagery collected by Unmanned Aerial Vehicles (UAV) has shown potential in the past; however, validation on large data-sets (e.g., across a wide number of different fields) remains lacking, with few solutions actually made operational. Here, we demonstrate the feasibility of automatically detecting weeds in winter wheat fields based on deep learning methods applied to UAV data at scale. Focusing on black-grass (the most pernicious weed across northwest Europe), we show high performance (i.e., accuracy above 0.9) and highly statistically significant correlation (i.e., ro > 0.75 and p < 0.00001) between imagery-derived local and global weed maps and out-of-bag field survey data, collected by experts over 31 fields (205 hectares) in the UK. We demonstrate how the developed deep learning model can be made available via an easy-to-use docker container, with results accessible through an interactive dashboard. Using this approach, clickable weed maps can be created and deployed rapidly, allowing the user to explore actual model predictions for each field. This shows the potential for this approach to be used operationally and influence agronomic decision-making in the real world.

KeywordsDeep learning; Image Segmentation; Unmanned Aerial Vehicle; Weed Detection; Black-grass
Year of Publication2022
JournalRemote Sensing
Journal citation14 (17), p. 4197
Digital Object Identifier (DOI)https://doi.org/10.3390/rs14174197
Web address (URL)https://www.mdpi.com/2072-4292/14/17/4197/htm
Open accessPublished as ‘gold’ (paid) open access
FunderInnovate UK
Funder project or codeaiScope - AI data platform for smart crop protection
Publisher's version
Output statusPublished
Publication dates
Online26 Aug 2022
Publication process dates
Accepted17 Aug 2022
Submitted18 Jul 2022
PublisherMDPI
ISSN2072-4292

Permalink - https://repository.rothamsted.ac.uk/item/9889w/a-deep-learning-application-to-map-weed-spatial-extent-from-unmanned-aerial-vehicles-imagery

39 total views
18 total downloads
0 views this month
0 downloads this month
Download files as zip