What plant is that? Tests of automated image recognition apps for plant identification on plants from the British flora. Issue 6 (23rd September 2020)

• Record Type: Journal Article
• Authors: Jones, Hamlyn G
• Editors: Atkins, Jeff
• Is Part Of: AoB plants. Volume 12:Issue 6(2020)
• Publication Date: 2020-09-23
• Subjects: Apps -- artificial intelligence -- image recognition -- plant identification -- smartphones; Plants -- Periodicals; Botany -- Periodicals; 580.5
• DOI: 10.1093/aobpla/plaa052
• Languages: English
• ISSNs: 2041-2851

Abstract

Abstract: There has been a recent explosion in development of image recognition technology and its application to automated plant identification, so it is timely to consider its potential for field botany. Nine free apps or websites for automated plant identification and suitable for use on mobile phones or tablet computers in the field were tested on a disparate set of 38 images of plants or parts of plants chosen from the higher plant flora of Britain and Ireland. There were large differences in performance with the best apps identifying >50 % of samples tested to genus or better. Although the accuracy is good for some of the top-rated apps, for any quantitative biodiversity study or for ecological surveys, there remains a need for validation by experts or against conventional floras. Nevertheless, the better-performing apps should be of great value to beginners and amateurs and may usefully stimulate interest in plant identification and nature. Potential uses of automated image recognition plant identification apps are discussed and recommendations made for their future use. Abstract : There is much interest in potential uses of artificial intelligence approaches to the automated identification of unknown plants. This paper reviews and compares a number of free smartphone apps that attempt automatically to identify unknown plants from images taken in natural environments, using images of plants growing wild in Britain. These apps are continually improving but the best ones already show an outstanding success rate identifying up to three quarters of samples to family and half to species. Criteria for the selection of an app for any situation are discussed and highlight the importance of minimizing erroneous identifications.