InterFolia: an embedded, educational application to identify plant species in the wild - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

InterFolia: an embedded, educational application to identify plant species in the wild

Maxime Langlade
  • Fonction : Auteur
Debaleena Misra
  • Fonction : Auteur
Laure Tougne

Résumé

This paper presents InterFolia, an iOS application dedicated to help the general public identify the plant species around them. The application can recognize up to 214 species of trees and shrubs found in France and western Europe. Its classification module contains three convolutional neural network models that run locally and are each associated with a plant organ (flower/fruit, leaf, or bark). A fusion module is integrated into the application to combine all available information to help improve the overall accuracy of the classification task. The main novelty of In-terFolia is its use of embedded models that allow users to explore nature beyond network-covered zones.
Fichier principal
Vignette du fichier
InterFolia_extended_abstract_CVPPA_2021.pdf (422.04 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03380533 , version 1 (15-10-2021)

Identifiants

  • HAL Id : hal-03380533 , version 1

Citer

Maxime Langlade, Debaleena Misra, Carlos F Crispim-Junior, Laure Tougne. InterFolia: an embedded, educational application to identify plant species in the wild. 7th workshop on Computer Vision in Plant Phenotyping and Agriculture - ICCV Workshop 2021, Oct 2021, Virtual conference, France. ⟨hal-03380533⟩
99 Consultations
112 Téléchargements

Partager

Gmail Facebook X LinkedIn More