Gabriel Pagin

Mr. Gabriel Pagin

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 July 2025 in Dairy
Verified
Journal Article
Open access
PDF file
RSRafael Vieira de de Sousa
Rafael Vieira de de Sousa
JCJéssica Caetano Dias Campos
Jéssica Caetano Dias Campos
GPGabriel Pagin
Gabriel Pagin
Danilo Florentino PereiraDanilo Florentino Pereira
Danilo Florentino Pereira
ACAline rabello Conceição
Aline rabello Conceição
LMLuciane Silva Martello
Luciane Silva Martello
Infrared thermography (IRT) is explored as a non-invasive method for indirectly measuring parameters related to animal performance and welfare. This study investigates a feature extraction method termed the “thermal signature” (TS), a descriptor vector derived from the temperature matrix of an animal’s body surface, representing the percentage distribution of temperatures within predefined ranges. The TS, combined with environmental data, serves as a predictor attribute for machine learning-based classifier models to assess heat stress levels. The methodology was applied to a dataset collected from two groups of five dairy calves housed in a climate-controlled chamber and exposed to two artificial heat waves over 13 days. Data, including IRT measurements, respiratory rate (RR), rectal temperature (RT), and environmental variables, were collected five times daily (from 6 a.m. to 10 p.m., every four hours). Classifier models were developed using random forest (RF), support vector machine (SVM), artificial neural network (ANN), and K-nearest neighbor (KNN) algorithms. The RF models based on RR achieved the highest accuracies, 94.1% for two heat stress levels and 80.3% for three heat stress levels, using TS configurations with six temperature ranges. The integration of TS with machine learning-based models demonstrates promising results for developing or enhancing classifiers of heat stress levels in dairy calves.

23 September 2025 in Agriengineering
Verified
Journal Article
Open access
PDF file
TSThiago Silva
Thiago Silva
FDFernanda de Fátima da Silva Devechio
Fernanda de Fátima da Silva Devechio
MTMarcos Tavares
Marcos Tavares
JRJamile Raquel Regazzo
Jamile Raquel Regazzo
ESEdson José de Souza Sardinha
Edson José de Souza Sardinha
MBMurilo Mesquita Baesso
Murilo Mesquita Baesso
Nitrogen fertilization is decisive for maize productivity, fertilizer use efficiency, and sustainability, which calls for fast and nondestructive nutritional diagnosis. This study evaluated the classification of maize plant nutritional status from red, green, and blue (RGB) leaf images using texture attributes. A greenhouse experiment was conducted under a completely randomized factorial design with four nitrogen doses, one maize hybrid Pioneer 30F35, and four replicates, at two sampling times corresponding to distinct phenological stages, totaling thirty-two experimental units. Images were processed with the gray-level cooccurrence matrix computed at three distances 1, 3, and 5 pixels and four orientations 0°, 45°, 90°, and 135°, yielding eight texture descriptors that served as inputs to five supervised classifiers: an artificial neural network, a support vector machine, k nearest neighbors, a decision tree, and Naive Bayes. The results indicated that texture descriptors discriminated nitrogen doses with good performance and moderate computational cost, and that homogeneity, dissimilarity, and contrast were the most informative attributes. The artificial neural network showed the most stable performance at both stages, followed by the support vector machine and k nearest neighbors, whereas the decision tree and Naive Bayes were less suitable. Confusion matrices and receiver operating characteristic curves indicated greater separability for omission and excess classes, with D1 standing out, and the patterns were consistent with the chemical analysis. Future work should include field validation, multiple seasons and genotypes, integration with spectral indices and multisensor data, application of model explainability techniques, and assessment of latency and scalability in operational scenarios.

2 May 2024 in OBSERVATÓRIO DE LA ECONOMÍA LATINOAMERICANA
Verified
Journal Article
Open access
PDF file
HSHudson Sérgio de Souza
Hudson Sérgio de Souza
ENEduardo Lima Nunes
Eduardo Lima Nunes
GPGabriel Pagin
Gabriel Pagin
TGTerezinha Guedes
Terezinha Guedes
Fernando Teruhiko HataFernando Teruhiko Hata
Fernando Teruhiko Hata
WNWilliam Mário de Carvalho Nunes
William Mário de Carvalho Nunes
O site 'Sistema de Treinamento em Acuidade Visual' (SiTAV), foi projetado para melhorar a precisão e confiabilidade das estimativas dos avaliadores de severidades de doenças de plantas. Descrevemos os recursos tecnológicos e procedimentos utilizados para o desenvolvimento do sistema e validação demonstrando as melhorias na estimativa de severidade após o treinamento. A programação do SiTAV foi baseada no sistema operacional GNU/Linux, PHP, HTML, CSS, JavaScript e JQuery. Os dados de treinamento foram armazenados no banco de dados MySQL. A ferramenta de codificação fonte é o NetBeans IDE utilizando a orientação a objetos e reutilização de código fonte através das recomendações do W3C para boas práticas de programação e usabilidade do sistema. As edições das imagens foram realizadas através das ferramentas Gimp, Inkscape e Quant. Utilizamos o gerenciador de gráficos PHPlot (DICE Holdings, 2015). Os dados de treinamento armazenados no banco de dados MySQL podem ser exportados para os formatos ".xls", ".txt" e ".csv" para permitir seu uso em programas estatísticos como SAS ou R para identificação de padrões, tendências e possíveis erros nas estimativas de gravidade de doenças não discernível usando os dados isolados. O software é acessado pela internet por smartphone, ‘tablet’ ou computador pessoal. Imagens de folhas doentes no banco de dados são utilizadas para treinamento que fornece a realidade das características do hospedeiro e dos sintomas. A validação do software foi realizada com a colaboração de 73 (setenta e três) acadêmicos do curso de Agronomia da Universidade Estadual de Maringá - UEM, onde ao final do treinamento foram observadas melhorias significativas no processo de avaliação de doenças de plantas, tendo como média geral entre todos os validadores, 71%.

15 July 2025 in Smart Agricultural Technology
Verified
Journal Article
Open access
GPGabriel Pagin
Gabriel Pagin
LMLuciane Silva Martello
Luciane Silva Martello
RTRubens André Tabile
Rubens André Tabile
RSRafael Vieira de de Sousa
Rafael Vieira de de Sousa

29 July 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
GPGabriel Pagin
Gabriel Pagin
LMLuciane Silva Martello
Luciane Silva Martello
JPJuliana Jorge Paschoal
Juliana Jorge Paschoal
DMDanielle Leal Matarim
Danielle Leal Matarim
RTRubens André Tabile
Rubens André Tabile
RSRafael Vieira de de Sousa
Rafael Vieira de de Sousa
Social hierarchy influences resource access in cattle, but its temporal and social-network dimensions remain poorly characterized in Bos indicus feedlots. This study used automated radio-frequency identification (RFID) monitoring to evaluate how social dominance and environmental factors reorganize feeding and drinking behavior in Nelore bulls. A total of 49 Nelore bulls were monitored in three pens across two data-collection phases at the same automated feedlot, using RFID-equipped feed bunks and waterers that generated 58,779 intake events. Social dominance was quantified from video-recorded agonistic interactions using David’s Score. Data were analyzed using mixed models, distributed-lag regressions, and co-occurrence network analysis. Daily DMI did not differ among dominance classes, and no differences were detected in the performance indicators. However, dominance influenced feeding behavior throughout the day: dominant animals consumed more feed per access during the afternoon, whereas subordinate animals consumed more at night (0.902 vs. 0.800 kg/access; p = 0.0002), suggesting temporal compensation. Feed-bunk networks showed strong negative assortativity (-0.89 to -0.99), consistent with cross-hierarchical co-occurrence driven by resource access rather than social affiliation. Water intake increased with temperature (+1.68 L per standard deviation increase in mean daily temperature) and remained elevated on the following day (+0.48 L), whereas DMI as minimally affected by temperature. Feeding behavior differed between the two pens with contrasting stocking densities, whereas DMI, water intake, and productive performance did not. Social hierarchy reorganized feeding timing and social interactions without detectable effects on daily intake or the performance indicators. Continuous RFID monitoring revealed strong cross-hierarchical feed-bunk interactions, nocturnal compensatory feeding by subordinate animals, and greater thermal sensitivity of water intake than DMI, demonstrating the potential of automated behavioral phenotyping to detecting behavioral adaptations relevant to welfare monitoring in tropical Bos indicus feedlots.

1 April 2024 in Contribuciones A Las Ciencias Sociales
Verified
Journal Article
Open access
PDF file
HSHudson Sérgio de Souza
Hudson Sérgio de Souza
ENEduardo Lima Nunes
Eduardo Lima Nunes
GPGabriel Pagin
Gabriel Pagin
TGTerezinha Guedes
Terezinha Guedes
Fernando Teruhiko HataFernando Teruhiko Hata
Fernando Teruhiko Hata
WNWilliam Mário de Carvalho Nunes
William Mário de Carvalho Nunes
The present study analyzed the data compiled after the training of the precision and visual accuracy of the evaluators of both genders in plant diseases, using the SiTAV (Visual Acuity Training System) website. The training data were analyzed through the exploratory analysis, with the aid of the statistical program SAS 9.4. With 52,495 training data being stored in the system's MySQL database identified. The tabulation of training data ranked the evaluations in three classes of accuracy of major hits (underestimated, precise and overestimated), where 30.36% (15.938) training were classified as accurate. The distribution of training data between accuracy classes and five hit categories (excellent, optimal, good, fair and poor) indicated that the level of excellent severity obtained 30.04% (15,772) of training data of the six levels of severity, level 1 (0 to 3% of real severity) reached 33.31% (17,486) of training data. It is concluded that the smaller the lesion size of plant diseases, the greater the accuracy of the evaluator.