Front. Plant Sci.Frontiers in Plant ScienceFront. Plant Sci.1664-462XFrontiers Media S.A.10.3389/fpls.2025.1664228Plant ScienceCorrectionCorrection: Optimizing Mask R-CNN for enhanced quinoa panicle detection and segmentation in precision agricultureEl AkrouchiManal12*†MhadaManal1†GraciaDachena Romain1HawkesfordMalcolm J.3GérardBruno11College of Agriculture and Environmental Sciences, University Mohammed VI Polytechnic (UM6P), Ben Guerir, Morocco2School of Collective Intelligence, University Mohammed VI Polytechnic (UM6P), Rabat, Morocco3Sustainable Soils and Crops Department, Rothamsted Research, Harpenden, Hertfordshire, United Kingdom
Approved by: Frontiers Editorial Office, Frontiers Media SA, Switzerland
*Correspondence: Manal El Akrouchi, manal.elakrouchi@um6p.ma
†These authors have contributed equally to this work and share first authorship
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
A Correction on Optimizing Mask R-CNN for enhanced quinoa panicle detection and segmentation in precision agriculture By El Akrouchi M, Mhada M, Gracia DR, Hawkesford MJ and Gérard B (2025). Front. Plant Sci. 16:1472688. doi: 10.3389/fpls.2025.1472688Mask R-CNNinstance segmentationquinoaprecision agriculturedeep learningsection-in-acceptanceTechnical Advances in Plant Science
There was a mistake in Figure 3 as published. I was working on two papers simultaneously, this one and another related to citrus (see this link). While preparing the flowcharts for both projects, I inadvertently used the same name for both files, which led to this confusion. The corrected Figure 3 appears below.
Overall process flowchart of quinoa panicles detection and segmentation.
Flowchart depicting a pipeline for image-based quinoa panicle counting using Mask R-CNN models. Sections include protocol design, dataset generation, backbone comparison, segmentation and counting, and final outputs. Backbones tested are ResNet50, ResNet101, ViTDet_b, Swin_b, Mish-based EfficientNet B7 with FPN(LN), and FPN(GN). Steps involve image acquisition, data augmentation, and deep learning model training. The final output involves comparing manual and predicted counts and analyzing results.
The original version of this article has been updated.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.