Weed Identification by Single-Stage and Two-Stage Neural Networks: A Study on the Impact of Image Resizers and Weights Optimization Algorithms

Simple item page
dc.citation.volume13
dc.contributor.authorSaleem MH
dc.contributor.authorVelayudhan KK
dc.contributor.authorPotgieter J
dc.contributor.authorArif K
dc.coverage.spatialSwitzerland
dc.date.available2022
dc.date.available2022-03-11
dc.date.issued25/04/2022
dc.description.abstractThe accurate identification of weeds is an essential step for a site-specific weed management system. In recent years, deep learning (DL) has got rapid advancements to perform complex agricultural tasks. The previous studies emphasized the evaluation of advanced training techniques or modifying the well-known DL models to improve the overall accuracy. In contrast, this research attempted to improve the mean average precision (mAP) for the detection and classification of eight classes of weeds by proposing a novel DL-based methodology. First, a comprehensive analysis of single-stage and two-stage neural networks including Single-shot MultiBox Detector (SSD), You look only Once (YOLO-v4), EfficientDet, CenterNet, RetinaNet, Faster Region-based Convolutional Neural Network (RCNN), and Region-based Fully Convolutional Network (RFCN), has been performed. Next, the effects of image resizing techniques along with four image interpolation methods have been studied. It led to the final stage of the research through optimization of the weights of the best-acquired model by initialization techniques, batch normalization, and DL optimization algorithms. The effectiveness of the proposed work is proven due to a high mAP of 93.44% and validated by the stratified k-fold cross-validation technique. It was 5.8% improved as compared to the results obtained by the default settings of the best-suited DL architecture (Faster RCNN ResNet-101). The presented pipeline would be a baseline study for the research community to explore several tasks such as real-time detection and reducing the computation/training time. All the relevant data including the annotated dataset, configuration files, and inference graph of the final model are provided with this article. Furthermore, the selection of the DeepWeeds dataset shows the robustness/practicality of the study because it contains images collected in a real/complex agricultural environment. Therefore, this research would be a considerable step toward an efficient and automatic weed control system.
dc.description.publication-statusPublished online
dc.format.extent850666 - ?
dc.identifierhttps://www.ncbi.nlm.nih.gov/pubmed/35548295
dc.identifier.citationFront Plant Sci, 2022, 13 pp. 850666 - ?
dc.identifier.doi10.3389/fpls.2022.850666
dc.identifier.elements-id452874
dc.identifier.harvestedMassey_Dark
dc.identifier.issn1664-462X
dc.identifier.urihttps://hdl.handle.net/10179/17053
dc.languageeng
dc.publisherFrontiers Media
dc.relation.isPartOfFront Plant Sci
dc.subjectconvolutional neural network
dc.subjectdeep learning
dc.subjectoptimization algorithms
dc.subjecttransfer learning
dc.subjectweed detection
dc.subject.anzsrc0607 Plant Biology
dc.titleWeed Identification by Single-Stage and Two-Stage Neural Networks: A Study on the Impact of Image Resizers and Weights Optimization Algorithms
dc.typeJournal article
pubs.notesNot known
pubs.organisational-group/Massey University
pubs.organisational-group/Massey University/College of Sciences
pubs.organisational-group/Massey University/College of Sciences/School of Agriculture & Environment
pubs.organisational-group/Massey University/College of Sciences/School of Agriculture & Environment/Agritech
pubs.organisational-group/Massey University/College of Sciences/School of Food and Advanced Technology
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
fpls-13-850666.pdf
Size:
8.75 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Journal Articles