A comparative study of fine-grained classification methods in the context of the LifeCLEF plant identification challenge 2015

Julien Champ (1) , Titouan Lorieul (1) , Maximilien Servajean (2) , Alexis Joly (1)

1 ZENITH - Scientific Data Management
2 ADVANSE - ADVanced Analytics for data SciencE

This paper describes the participation of Inria to the plant identification task of the LifeCLEF 2015 challenge. The aim of the task was to produce a list of relevant species for a large set of plant observations related to 1000 species of trees, herbs and ferns living in Western Europe. Each plant observation contained several annotated pictures with organ/view tags: Flower, Leaf, Fruit, Stem, Branch, Entire, Scan (exclusively of leaf). To address this challenge, we experimented two popular families of classification techniques, i.e. convolutional neural networks (CNN) on one side and fisher vectors-based discriminant models on the other side. Our results show that the CNN approach achieves much better performance than the fisher vectors. Beyond, we show that the fusion of both techniques, based on a Bayesian inference using the confusion matrix of each classifier, did not improve the results of the CNN alone.

Mots clés: citizen-science; fine-grained classification; benchmark; evaluation; bark; species identification; fruit; flower; leaf; leaves; Plant; LifeCLEF; Species; collection; Images; retrieval; Stem; Branch

Domaines: Informatique [cs]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Julien Champ, Titouan Lorieul, Maximilien Servajean, Alexis Joly. A comparative study of fine-grained classification methods in the context of the LifeCLEF plant identification challenge 2015. CLEF 2015 - Working Notes of the Conference and Labs of the Evaluation Forum, Sep 2015, Toulouse, France. ⟨hal-01182788⟩