Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials

Extracting phenotypic traits of grape bunch is crucial for accurately monitoring grape quality, health, and yield estimation. This is important for optimising resources, enhancing marketing strategies, and boosting overall agricultural productivity. While most research concentrates on data processin...

Full description

Bibliographic Details
Authors: Ariza-Sentís, Mar, Baja, Hilmy, Vélez, Sergio, van Essen, Rick, Valente, João
Format: article
Status:Published version
Publication Date:2025
Country:España
Institution:Consejo Superior de Investigaciones Científicas (CSIC)
Repository:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/418535
Online Access:http://hdl.handle.net/10261/418535
Access Level:Open access
Keyword:Precision viticulture
Leaf-occlusion
Data acquisition
Object detection
Object tracking
id ES_7a23fdb78de5de4e0461e106bb530968
oai_identifier_str oai:digital.csic.es:10261/418535
network_acronym_str ES
network_name_str España
repository_id_str
spelling Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trialsAriza-Sentís, MarBaja, HilmyVélez, Sergiovan Essen, RickValente, JoãoPrecision viticultureLeaf-occlusionData acquisitionObject detectionObject trackingExtracting phenotypic traits of grape bunch is crucial for accurately monitoring grape quality, health, and yield estimation. This is important for optimising resources, enhancing marketing strategies, and boosting overall agricultural productivity. While most research concentrates on data processing algorithms, this study focused on the preceding step: collecting reliable data. Object detection and tracking enable precise monitoring and quantification of fruit, facilitating agricultural management. This study compares two data acquisition methodologies for grape bunch detection and tracking in a commercial vineyard where leaf removal was not performed: a traditional single-view approach and a multiple-viewing method designed to mitigate fruit occlusion issues. The PointTrack algorithm, trained and validated using MOTS annotations, was employed to evaluate detection and tracking performance through metrics of three trials. The multiple-view method achieved i) higher ratio between tracked and GT detections of 74 % compared to 23 % for the single-view approach and ii) enhanced tracking metrics, with the multiple viewing trials metrics ranging from −1.35 to 3.84 for MOTSA (Multiple Object Tracking and Segmentation Accuracy) and sMOTSA (soft MOTSA), and iii) higher correlation and lower RMSE of grape bunch phenotypic traits (OIV codes 202 and 203) compared to ground truth measurements (R2 = 0.53, RMSE = 19.13). Nonetheless, the multi-view technique was compromised by motion blur due to UAV movements, complicating the tracking process. This study underscores the importance of strategic data acquisition in improving performance for fruit detection and tracking. Future work should extend this methodology to other fruit varieties and environments to validate its broader applicability, enhancing the reliability of yield estimation in precision agriculture.This work has been carried out in the scope of the H2020 FlexiGroBots project, which has been funded by the European Commission in the scope of its H2020 programme (contract number 101017111, https://flexigrobots-h2020.eu/). The authors acknowledge valuable help and contributions from 'Bodegas Terras Gauda, S.A.' and all partners of the project. This work has also been developed in the scope of the H2020 Icaerus project (contract number 101060643, https://icaerus.eu/). Dr Sergio Vélez contract has been supported partially by the Iberdrola Foundation and the European Commission under the Marie Skłodowska-Curie Actions (MSCA) - E4F, part of the Horizon 2020 program (Grant Agreement No 101034297, https://doi.org/10.3030/101034297). We extend our gratitude to the Horizon Smart Droplets project (contract number 101070496, https://smartdroplets.eu/).Peer reviewedElsevierEuropean CommissionAriza-Sentís, Mar [0000-0002-5483-4532]Baja, Hilmy [0000-0002-6995-3817]Valente, João [0000-0002-6241-4124]Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]202620262025info:eu-repo/semantics/articlehttp://purl.org/coar/resource_type/c_6501Publisher's versioninfo:eu-repo/semantics/publishedVersionhttp://hdl.handle.net/10261/418535reponame:DIGITAL.CSIC. Repositorio Institucional del CSICinstname:Consejo Superior de Investigaciones Científicas (CSIC)Inglés#PLACEHOLDER_PARENT_METADATA_VALUE##PLACEHOLDER_PARENT_METADATA_VALUE#info:eu-repo/grantAgreement/EC/H2020/101017111info:eu-repo/grantAgreement/EC/H2020/101034297https://doi.org/10.1016/j.jafr.2025.101736Síinfo:eu-repo/semantics/openAccessoai:digital.csic.es:10261/4185352026-05-22T06:33:51Z
dc.title.none.fl_str_mv Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
title Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
spellingShingle Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
Ariza-Sentís, Mar
Precision viticulture
Leaf-occlusion
Data acquisition
Object detection
Object tracking
title_short Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
title_full Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
title_fullStr Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
title_full_unstemmed Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
title_sort Comparative analysis of single-view and multiple-view data collection strategies for detecting partially-occluded grape bunches: Field trials
dc.creator.none.fl_str_mv Ariza-Sentís, Mar
Baja, Hilmy
Vélez, Sergio
van Essen, Rick
Valente, João
author Ariza-Sentís, Mar
author_facet Ariza-Sentís, Mar
Baja, Hilmy
Vélez, Sergio
van Essen, Rick
Valente, João
author_role author
author2 Baja, Hilmy
Vélez, Sergio
van Essen, Rick
Valente, João
author2_role author
author
author
author
dc.contributor.none.fl_str_mv European Commission
Ariza-Sentís, Mar [0000-0002-5483-4532]
Baja, Hilmy [0000-0002-6995-3817]
Valente, João [0000-0002-6241-4124]
Consejo Superior de Investigaciones Científicas [https://ror.org/02gfc7t72]
dc.subject.none.fl_str_mv Precision viticulture
Leaf-occlusion
Data acquisition
Object detection
Object tracking
topic Precision viticulture
Leaf-occlusion
Data acquisition
Object detection
Object tracking
description Extracting phenotypic traits of grape bunch is crucial for accurately monitoring grape quality, health, and yield estimation. This is important for optimising resources, enhancing marketing strategies, and boosting overall agricultural productivity. While most research concentrates on data processing algorithms, this study focused on the preceding step: collecting reliable data. Object detection and tracking enable precise monitoring and quantification of fruit, facilitating agricultural management. This study compares two data acquisition methodologies for grape bunch detection and tracking in a commercial vineyard where leaf removal was not performed: a traditional single-view approach and a multiple-viewing method designed to mitigate fruit occlusion issues. The PointTrack algorithm, trained and validated using MOTS annotations, was employed to evaluate detection and tracking performance through metrics of three trials. The multiple-view method achieved i) higher ratio between tracked and GT detections of 74 % compared to 23 % for the single-view approach and ii) enhanced tracking metrics, with the multiple viewing trials metrics ranging from −1.35 to 3.84 for MOTSA (Multiple Object Tracking and Segmentation Accuracy) and sMOTSA (soft MOTSA), and iii) higher correlation and lower RMSE of grape bunch phenotypic traits (OIV codes 202 and 203) compared to ground truth measurements (R2 = 0.53, RMSE = 19.13). Nonetheless, the multi-view technique was compromised by motion blur due to UAV movements, complicating the tracking process. This study underscores the importance of strategic data acquisition in improving performance for fruit detection and tracking. Future work should extend this methodology to other fruit varieties and environments to validate its broader applicability, enhancing the reliability of yield estimation in precision agriculture.
publishDate 2025
dc.date.none.fl_str_mv 2025
2026
2026
dc.type.none.fl_str_mv info:eu-repo/semantics/article
http://purl.org/coar/resource_type/c_6501
Publisher's version
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10261/418535
url http://hdl.handle.net/10261/418535
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv #PLACEHOLDER_PARENT_METADATA_VALUE#
#PLACEHOLDER_PARENT_METADATA_VALUE#
info:eu-repo/grantAgreement/EC/H2020/101017111
info:eu-repo/grantAgreement/EC/H2020/101034297
https://doi.org/10.1016/j.jafr.2025.101736
Sí
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:DIGITAL.CSIC. Repositorio Institucional del CSIC
instname:Consejo Superior de Investigaciones Científicas (CSIC)
instname_str Consejo Superior de Investigaciones Científicas (CSIC)
reponame_str DIGITAL.CSIC. Repositorio Institucional del CSIC
collection DIGITAL.CSIC. Repositorio Institucional del CSIC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869411415667769344
score 15,811543

Similar Items