DeepVerge: Classification of Roadside Verge Biodiversity and Conservation Potential

Perrett, Andrew, Pollard, Harry, Barnes, Charlie , Schofield, Mark, Qie, Lan, Bosilj, Petra and Brown, James (2023) DeepVerge: Classification of Roadside Verge Biodiversity and Conservation Potential. Computers, Environment and Urban Systems . ISSN 0198-9715

Documents
DeepVerge: Classification of Roadside Verge Biodiversity and Conservation Potential
Author's accepted manuscript

Request a copy
[img] PDF
Classification_of_Roadside_Verges__Revisions___Andy_is_modifying_.pdf - Whole Document
Restricted to Repository staff only

3MB
Item Type:Article
Item Status:Live Archive

Abstract

Grasslands are increasingly modified by anthropogenic activities and species rich grasslands have become rare habitats in the UK. However, grassy roadside verges often contain conservation priority plant species and should be targeted for protection. Identification of verges with high conservation potential represents a considerable challenge for ecologists, driving the development of automated methods to make up for the shortfall of relevant expertise nationally. Using survey data from 3,900 km of roadside verges alongside publicly available street-view imagery, we present DeepVerge: a deep learning-based method that can automatically survey sections of roadside verge by detecting the presence of positive indicator species. Using images and ground truth survey data from the rural UK county of Lincolnshire, DeepVerge achieved a mean accuracy of 88% and a mean F1 score of 0.82. Such a method may be used by local authorities to identify new local wildlife sites, and aid management and environmental planning in line with legal and government policy obligations, saving thousands of hours of skilled labour

Keywords:convolutional neural network, road verges, biodiversity, remote surveying, Image Processing, ecological survey
Subjects:G Mathematical and Computer Sciences > G740 Computer Vision
C Biological Sciences > C181 Biodiversity
Divisions:College of Science > School of Computer Science
ID Code:54285
Deposited On:06 Apr 2023 15:16

Repository Staff Only: item control page