Repository logo
 
Publication

SpectraNet–53: A deep residual learning architecture for predicting soluble solids content with VIS–NIR spectroscopy

dc.contributor.authorA. Martins, J.
dc.contributor.authorGuerra, Rui Manuel Farinha das Neves
dc.contributor.authorPires, R.
dc.contributor.authorAntunes, M.D.
dc.contributor.authorPanagopoulos, T.
dc.contributor.authorBrázio, A.
dc.contributor.authorAfonso, A.M.
dc.contributor.authorSilva, L.
dc.contributor.authorLucas, M.R.
dc.contributor.authorCavaco, A.M.
dc.date.accessioned2023-03-09T16:21:07Z
dc.date.available2023-03-09T16:21:07Z
dc.date.issued2022-04
dc.description.abstractThis work presents a new deep learning architecture, SpectraNet-53, for quantitative analysis of fruit spectra, optimized for predicting Soluble Solids Content (SSC, in Brix). The novelty of this approach resides in being an architecture trainable on a very small dataset, while keeping a performance level on-par or above Partial Least Squares (PLS), a time-proven machine learning method in the field of spectroscopy. SpectraNet-53 performance is assessed by determining the SSC of 616 Citrus sinensi L. Osbeck 'Newhall' oranges, from two Algarve (Portugal) orchards, spanning two consecutive years, and under different edaphoclimatic conditions. This dataset consists of short-wave near-infrared spectroscopic (SW-NIRS) data, and was acquired with a portable spectrometer, in the visible to near infrared region, on-tree and without temperature equalization. SpectraNet-53 results are compared to a similar state-of-the-art architecture, DeepSpectra, as well as PLS, and thoroughly assessed on 15 internal validation sets (where the training and test data were sampled from the same orchard or year) and on 28 external validation sets (training/test data sampled from different orchards/years). SpectraNet-53 was able to achieve better performance than DeepSpectra and PLS in several metrics, and is especially robust to training overfit. For external validation results, on average, SpectraNet-53 was 3.1% better than PLS on RMSEP (1.16 vs. 1.20 Brix), 11.6% better in SDR (1.22 vs. 1.10), and 28.0% better in R2 (0.40 vs. 0.31).pt_PT
dc.description.sponsorshipproject NIBAP ALG-01-0247-FEDER-037303, project OtiCalFrut ALG-010247-FEDER-033652pt_PT
dc.description.versioninfo:eu-repo/semantics/publishedVersionpt_PT
dc.identifier.doi10.1016/j.compag.2022.106945pt_PT
dc.identifier.eissn1872-7107
dc.identifier.urihttp://hdl.handle.net/10400.1/19216
dc.language.isoengpt_PT
dc.peerreviewedyespt_PT
dc.publisherElsevierpt_PT
dc.relationSmart tools for optimizing management in precision agriculture of citrus orchards
dc.relationNot Available
dc.relationCenter for Electronics, Optoelectronics and Telecommunications
dc.relationCenter for Electronics, Optoelectronics and Telecommunications
dc.relationCenter for Electronics, Optoelectronics and Telecommunications
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/pt_PT
dc.subjectDeep learningpt_PT
dc.subjectResidual networkpt_PT
dc.subjectNear-infraredpt_PT
dc.subjectSpectroscopypt_PT
dc.subjectCitruspt_PT
dc.titleSpectraNet–53: A deep residual learning architecture for predicting soluble solids content with VIS–NIR spectroscopypt_PT
dc.typejournal article
dspace.entity.typePublication
oaire.awardTitleSmart tools for optimizing management in precision agriculture of citrus orchards
oaire.awardTitleNot Available
oaire.awardTitleCenter for Electronics, Optoelectronics and Telecommunications
oaire.awardTitleCenter for Electronics, Optoelectronics and Telecommunications
oaire.awardTitleCenter for Electronics, Optoelectronics and Telecommunications
oaire.awardURIinfo:eu-repo/grantAgreement/FCT//SFRH%2FBPD%2F101634%2F2014/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/DL 57%2F2016/DL 57%2F2016%2FCP1361%2FCT0040/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UID%2FMulti%2F00631%2F2019/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F00631%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00631%2F2020/PT
oaire.citation.startPage106945pt_PT
oaire.citation.titleComputers and Electronics in Agriculturept_PT
oaire.citation.volume197pt_PT
oaire.fundingStreamDL 57/2016
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
person.familyNameMartins
person.familyNameGuerra
person.givenNameJaime
person.givenNameRui
person.identifier.ciencia-id3D16-5067-D6BB
person.identifier.orcid0000-0001-9360-0221
person.identifier.orcid0000-0002-8642-5792
person.identifier.scopus-author-id55061238600
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
rcaap.rightsopenAccesspt_PT
rcaap.typearticlept_PT
relation.isAuthorOfPublicationdc16525e-6025-44b8-ac52-425a8e9e6c06
relation.isAuthorOfPublicationeff7071e-e676-465b-bae3-983f522acd98
relation.isAuthorOfPublication.latestForDiscoverydc16525e-6025-44b8-ac52-425a8e9e6c06
relation.isProjectOfPublication521a5b1c-880b-4df5-9130-4c211e088c40
relation.isProjectOfPublicatione2c8dff4-38e1-4590-b6f1-5f5afc7e3a22
relation.isProjectOfPublicationa82b8cf6-bde8-4f87-99a9-9844df16a73e
relation.isProjectOfPublication77b70459-1e8c-4a6c-9856-58860aaddb6b
relation.isProjectOfPublication6c1217d9-1340-45e8-91d8-e75348854f62
relation.isProjectOfPublication.latestForDiscoverya82b8cf6-bde8-4f87-99a9-9844df16a73e

Files

Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
A deep residual learning architecture for predicting soluble.pdf
Size:
8.95 MB
Format:
Adobe Portable Document Format
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
3.46 KB
Format:
Item-specific license agreed upon to submission
Description: