Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots
Premise: Weed removal in agriculture is typically achieved using herbicides. The use of autonomous robots to reduce weeds is a promising alternative solution, although their implementation requires the precise detection and identification of crops and weeds to allow an efficient action. Methods: We trained and evaluated an instance segmentation convolutional neural network aimed at segmenting and identifying each plant specimen visible in images produced by agricultural robots. The resulting data set comprised field images on which the outlines of 2489 specimens from two crop species and four weed species were manually drawn. We adjusted the hyperparameters of a mask region‐based convolutional neural network (R‐CNN) to this specific task and evaluated the resulting trained model. Results: The probability of detection using the model was quite good but varied significantly depending on the species and size of the plants. In practice, between 10% and 60% of weeds could be removed without too high of a risk of confusion with crop plants. Furthermore, we show that the segmentation of each plant enabled the determination of precise action points such as the barycenter of the plant surface. Discussion: Instance segmentation opens many possibilities for optimized weed removal actions. Weed electrification, for instance, could benefit from the targeted adjustment of the voltage, frequency, and location of the electrode to the plant. The results of this work will enable the evaluation of this type of weeding approach in the coming months.
Main Authors: | , , , , , |
---|---|
Format: | article biblioteca |
Language: | eng |
Subjects: | H60 - Mauvaises herbes et désherbage, N20 - Machines et matériels agricoles, U30 - Méthodes de recherche, détermination des espèces, automate, agriculture de précision, désherbage, mauvaise herbe, plante de culture, lutte culturale, désherbage mécanique, http://aims.fao.org/aos/agrovoc/c_10354, http://aims.fao.org/aos/agrovoc/c_25680, http://aims.fao.org/aos/agrovoc/c_92363, http://aims.fao.org/aos/agrovoc/c_8345, http://aims.fao.org/aos/agrovoc/c_8347, http://aims.fao.org/aos/agrovoc/c_1972, http://aims.fao.org/aos/agrovoc/c_2020, http://aims.fao.org/aos/agrovoc/c_8346, |
Online Access: | http://agritrop.cirad.fr/597011/ http://agritrop.cirad.fr/597011/1/aps3.11373.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
dig-cirad-fr-597011 |
---|---|
record_format |
koha |
spelling |
dig-cirad-fr-5970112024-01-29T03:10:55Z http://agritrop.cirad.fr/597011/ http://agritrop.cirad.fr/597011/ Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots. Champ Julien, Mora-Fallas Adán, Goeau Hervé, Mata-Montero Erick, Bonnet Pierre, Joly Alexis. 2020. Applications in Plant Sciences, 8 (7), n.spéc. Machine Learning in Plant Biology: From Genomics to Field Studies:e11373, 10 p.https://doi.org/10.1002/aps3.11373 <https://doi.org/10.1002/aps3.11373> Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots Champ, Julien Mora-Fallas, Adán Goeau, Hervé Mata-Montero, Erick Bonnet, Pierre Joly, Alexis eng 2020 Applications in Plant Sciences H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 Premise: Weed removal in agriculture is typically achieved using herbicides. The use of autonomous robots to reduce weeds is a promising alternative solution, although their implementation requires the precise detection and identification of crops and weeds to allow an efficient action. Methods: We trained and evaluated an instance segmentation convolutional neural network aimed at segmenting and identifying each plant specimen visible in images produced by agricultural robots. The resulting data set comprised field images on which the outlines of 2489 specimens from two crop species and four weed species were manually drawn. We adjusted the hyperparameters of a mask region‐based convolutional neural network (R‐CNN) to this specific task and evaluated the resulting trained model. Results: The probability of detection using the model was quite good but varied significantly depending on the species and size of the plants. In practice, between 10% and 60% of weeds could be removed without too high of a risk of confusion with crop plants. Furthermore, we show that the segmentation of each plant enabled the determination of precise action points such as the barycenter of the plant surface. Discussion: Instance segmentation opens many possibilities for optimized weed removal actions. Weed electrification, for instance, could benefit from the targeted adjustment of the voltage, frequency, and location of the electrode to the plant. The results of this work will enable the evaluation of this type of weeding approach in the coming months. article info:eu-repo/semantics/article Journal Article info:eu-repo/semantics/publishedVersion http://agritrop.cirad.fr/597011/1/aps3.11373.pdf text cc_by info:eu-repo/semantics/openAccess https://creativecommons.org/licenses/by/4.0/ https://doi.org/10.1002/aps3.11373 10.1002/aps3.11373 info:eu-repo/semantics/altIdentifier/doi/10.1002/aps3.11373 info:eu-repo/semantics/altIdentifier/purl/https://doi.org/10.1002/aps3.11373 info:eu-repo/semantics/dataset/purl/https://doi.org/10.5281/zenodo.3906500 |
institution |
CIRAD FR |
collection |
DSpace |
country |
Francia |
countrycode |
FR |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-cirad-fr |
tag |
biblioteca |
region |
Europa del Oeste |
libraryname |
Biblioteca del CIRAD Francia |
language |
eng |
topic |
H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 |
spellingShingle |
H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 Champ, Julien Mora-Fallas, Adán Goeau, Hervé Mata-Montero, Erick Bonnet, Pierre Joly, Alexis Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
description |
Premise: Weed removal in agriculture is typically achieved using herbicides. The use of autonomous robots to reduce weeds is a promising alternative solution, although their implementation requires the precise detection and identification of crops and weeds to allow an efficient action. Methods: We trained and evaluated an instance segmentation convolutional neural network aimed at segmenting and identifying each plant specimen visible in images produced by agricultural robots. The resulting data set comprised field images on which the outlines of 2489 specimens from two crop species and four weed species were manually drawn. We adjusted the hyperparameters of a mask region‐based convolutional neural network (R‐CNN) to this specific task and evaluated the resulting trained model. Results: The probability of detection using the model was quite good but varied significantly depending on the species and size of the plants. In practice, between 10% and 60% of weeds could be removed without too high of a risk of confusion with crop plants. Furthermore, we show that the segmentation of each plant enabled the determination of precise action points such as the barycenter of the plant surface. Discussion: Instance segmentation opens many possibilities for optimized weed removal actions. Weed electrification, for instance, could benefit from the targeted adjustment of the voltage, frequency, and location of the electrode to the plant. The results of this work will enable the evaluation of this type of weeding approach in the coming months. |
format |
article |
topic_facet |
H60 - Mauvaises herbes et désherbage N20 - Machines et matériels agricoles U30 - Méthodes de recherche détermination des espèces automate agriculture de précision désherbage mauvaise herbe plante de culture lutte culturale désherbage mécanique http://aims.fao.org/aos/agrovoc/c_10354 http://aims.fao.org/aos/agrovoc/c_25680 http://aims.fao.org/aos/agrovoc/c_92363 http://aims.fao.org/aos/agrovoc/c_8345 http://aims.fao.org/aos/agrovoc/c_8347 http://aims.fao.org/aos/agrovoc/c_1972 http://aims.fao.org/aos/agrovoc/c_2020 http://aims.fao.org/aos/agrovoc/c_8346 |
author |
Champ, Julien Mora-Fallas, Adán Goeau, Hervé Mata-Montero, Erick Bonnet, Pierre Joly, Alexis |
author_facet |
Champ, Julien Mora-Fallas, Adán Goeau, Hervé Mata-Montero, Erick Bonnet, Pierre Joly, Alexis |
author_sort |
Champ, Julien |
title |
Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
title_short |
Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
title_full |
Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
title_fullStr |
Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
title_full_unstemmed |
Instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
title_sort |
instance segmentation for the fine detection of crop and weed plants by precision agricultural robots |
url |
http://agritrop.cirad.fr/597011/ http://agritrop.cirad.fr/597011/1/aps3.11373.pdf |
work_keys_str_mv |
AT champjulien instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots AT morafallasadan instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots AT goeauherve instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots AT matamonteroerick instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots AT bonnetpierre instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots AT jolyalexis instancesegmentationforthefinedetectionofcropandweedplantsbyprecisionagriculturalrobots |
_version_ |
1792500063268765696 |