Logo LVSN
EnglishAccueil
A proposPersonnesRecherchePublicationsEvenementsProfil
A propos
Publications

 

 

 

 

REPARTI

MIVIM

Bayesian Classification and Unsupervised Learning for Isolating Weeds in Row Crops


François-Michel De Rainville, Audrey Durand, Félix-Antoine Fortin, Kevin Tanguy, Xavier Maldague, Bernard Panneton and Marie-Josée Simard

En savoir plus...

Abstract - This paper presents a weed/crop classification method using computer vision and morphological analysis. Subsequent supervised and unsupervised learning methods are applied to extract dominant morphological characteristics of weeds present in corn and soybean fields. The novelty of the presented technique resides in the feature extraction process that is based on spatial localization of vegetation in fields. Features from the weed leaf area distribution are extracted from the cultivation inter-rows, then features from the crop are inferred from the mixture model equation. Those extracted features are then passed to a naive bayesian classifier and a gaussian mixture clustering algorithm to discriminate weed from crop plant. The presented technique correctly classifies an average of 94% of corn and soybean plants and 85% of the weed (multiple species) without any prior knowledge on the species present in the field.

download documentdownload document

Bibtex:

@article{Rainville946,
    author    = { François-Michel De Rainville and Audrey Durand and Félix-Antoine Fortin and Kevin Tanguy and Xavier Maldague and Bernard Panneton and Marie-Josée Simard },
    title     = { Bayesian Classification and Unsupervised Learning for Isolating Weeds in Row Crops },
    volume    = { 17 },
    number    = { 2 },
    pages     = { 401-414 },
    year      = { 2014 },
    journal   = { Pattern Analysis & Applications },
    web       = { http://dx.doi.org/10.1007/s10044-012-0307-5 }
}

Dernière modification: Oct 29 2012 10:00AM par fmder1

     
   
   

©2002-. Laboratoire de Vision et Systèmes Numériques. Tous droits réservés