Deep learning-based visual recognition of rumex for robotic precision farming

Tsampikos Kounalakis, Georgios A. Triantafyllidis, Lazaros Nalpantidis

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

In this paper we address the problem of recognising the Broad-leaved dock (Rumex obtusifolius L.) in grasslands from high-resolution 2D images. We discuss and present the determining factors for developing and implementing weed visual recognition algorithms using deep learning. This analysis, leads to the formulation of the proposed algorithm. Our implementation exploits Transfer Learning techniques for deep learning-based feature extraction, in combination with a classifier for weed recognition. A prototype robotic platform has been used to make available an image dataset from a dairy farm containing broad-leaved docks. The evaluation of the proposed algorithm on this dataset shows that it outperforms competing weed/plant recognition methods in recognition accuracy, while producing low false-positive rates under real-world operation conditions.

Original languageEnglish
Article number104973
JournalComputers and Electronics in Agriculture
Volume165
Number of pages11
ISSN0168-1699
DOIs
Publication statusPublished - 1 Oct 2019

Keywords

  • Agricultural robotics
  • Deep learning visual recognition
  • Precision farming
  • Weed recognition

Cite this

@article{9119149f0a764aaeb8842ae4777a35e9,
title = "Deep learning-based visual recognition of rumex for robotic precision farming",
abstract = "In this paper we address the problem of recognising the Broad-leaved dock (Rumex obtusifolius L.) in grasslands from high-resolution 2D images. We discuss and present the determining factors for developing and implementing weed visual recognition algorithms using deep learning. This analysis, leads to the formulation of the proposed algorithm. Our implementation exploits Transfer Learning techniques for deep learning-based feature extraction, in combination with a classifier for weed recognition. A prototype robotic platform has been used to make available an image dataset from a dairy farm containing broad-leaved docks. The evaluation of the proposed algorithm on this dataset shows that it outperforms competing weed/plant recognition methods in recognition accuracy, while producing low false-positive rates under real-world operation conditions.",
keywords = "Agricultural robotics, Deep learning visual recognition, Precision farming, Weed recognition",
author = "Tsampikos Kounalakis and Triantafyllidis, {Georgios A.} and Lazaros Nalpantidis",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.compag.2019.104973",
language = "English",
volume = "165",
journal = "Computers and Electronics in Agriculture",
issn = "0168-1699",
publisher = "Elsevier",

}

Deep learning-based visual recognition of rumex for robotic precision farming. / Kounalakis, Tsampikos; Triantafyllidis, Georgios A.; Nalpantidis, Lazaros.

In: Computers and Electronics in Agriculture, Vol. 165, 104973, 01.10.2019.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Deep learning-based visual recognition of rumex for robotic precision farming

AU - Kounalakis, Tsampikos

AU - Triantafyllidis, Georgios A.

AU - Nalpantidis, Lazaros

PY - 2019/10/1

Y1 - 2019/10/1

N2 - In this paper we address the problem of recognising the Broad-leaved dock (Rumex obtusifolius L.) in grasslands from high-resolution 2D images. We discuss and present the determining factors for developing and implementing weed visual recognition algorithms using deep learning. This analysis, leads to the formulation of the proposed algorithm. Our implementation exploits Transfer Learning techniques for deep learning-based feature extraction, in combination with a classifier for weed recognition. A prototype robotic platform has been used to make available an image dataset from a dairy farm containing broad-leaved docks. The evaluation of the proposed algorithm on this dataset shows that it outperforms competing weed/plant recognition methods in recognition accuracy, while producing low false-positive rates under real-world operation conditions.

AB - In this paper we address the problem of recognising the Broad-leaved dock (Rumex obtusifolius L.) in grasslands from high-resolution 2D images. We discuss and present the determining factors for developing and implementing weed visual recognition algorithms using deep learning. This analysis, leads to the formulation of the proposed algorithm. Our implementation exploits Transfer Learning techniques for deep learning-based feature extraction, in combination with a classifier for weed recognition. A prototype robotic platform has been used to make available an image dataset from a dairy farm containing broad-leaved docks. The evaluation of the proposed algorithm on this dataset shows that it outperforms competing weed/plant recognition methods in recognition accuracy, while producing low false-positive rates under real-world operation conditions.

KW - Agricultural robotics

KW - Deep learning visual recognition

KW - Precision farming

KW - Weed recognition

U2 - 10.1016/j.compag.2019.104973

DO - 10.1016/j.compag.2019.104973

M3 - Journal article

VL - 165

JO - Computers and Electronics in Agriculture

JF - Computers and Electronics in Agriculture

SN - 0168-1699

M1 - 104973

ER -