Cookies
O website necessita de alguns cookies e outros recursos semelhantes para funcionar. Caso o permita, o INESC TEC irá utilizar cookies para recolher dados sobre as suas visitas, contribuindo, assim, para estatísticas agregadas que permitem melhorar o nosso serviço. Ver mais
Aceitar Rejeitar
  • Menu
Sobre

Sobre

Tatiana M. Pinho recebeu, em 2018, o grau de doutoramento em Engenharia Electrotécnica e de Computadores pela Universidade de Trás-os-Montes e Alto Douro (UTAD), Portugal, e INESC TEC, ao abrigo de uma bolsa da Fundação para a Ciência e a Tecnologia (FCT). Em 2011, licenciou-se em Engenharia de Energias na UTAD, e recebeu o grau de mestre em Engenharia de Energias pela mesma universidade, em 2013. Atualmente é investigadora de pós-doutoramento no INESC TEC e os seus interesses de investigação incluem instrumentação, modelação e controlo adaptativo, em particular Model Predictive Control, aplicado a sistemas agro-florestais. 

Tópicos
de interesse
Detalhes

Detalhes

012
Publicações

2022

Benchmark of Deep Learning and a Proposed HSV Colour Space Models for the Detection and Classification of Greenhouse Tomato

Autores
Moreira, G; Magalhaes, SA; Pinho, T; dos Santos, FN; Cunha, M;

Publicação
AGRONOMY-BASEL

Abstract
The harvesting operation is a recurring task in the production of any crop, thus making it an excellent candidate for automation. In protected horticulture, one of the crops with high added value is tomatoes. However, its robotic harvesting is still far from maturity. That said, the development of an accurate fruit detection system is a crucial step towards achieving fully automated robotic harvesting. Deep Learning (DL) and detection frameworks like Single Shot MultiBox Detector (SSD) or You Only Look Once (YOLO) are more robust and accurate alternatives with better response to highly complex scenarios. The use of DL can be easily used to detect tomatoes, but when their classification is intended, the task becomes harsh, demanding a huge amount of data. Therefore, this paper proposes the use of DL models (SSD MobileNet v2 and YOLOv4) to efficiently detect the tomatoes and compare those systems with a proposed histogram-based HSV colour space model to classify each tomato and determine its ripening stage, through two image datasets acquired. Regarding detection, both models obtained promising results, with the YOLOv4 model standing out with an F1-Score of 85.81%. For classification task the YOLOv4 was again the best model with an Macro F1-Score of 74.16%. The HSV colour space model outperformed the SSD MobileNet v2 model, obtaining results similar to the YOLOv4 model, with a Balanced Accuracy of 68.10%.

2021

Prototyping IoT-Based Virtual Environments: An Approach toward the Sustainable Remote Management of Distributed Mulsemedia Setups

Autores
Adao, T; Pinho, T; Padua, L; Magalhaes, LG; Sousa, JJ; Peres, E;

Publicação
APPLIED SCIENCES-BASEL

Abstract
Business models built upon multimedia/multisensory setups delivering user experiences within disparate contexts—entertainment, tourism, cultural heritage, etc.—usually comprise the installation and in-situ management of both equipment and digital contents. Considering each setup as unique in its purpose, location, layout, equipment and digital contents, monitoring and control operations may add up to a hefty cost over time. Software and hardware agnosticity may be of value to lessen complexity and provide more sustainable management processes and tools. Distributed computing under the Internet of Things (IoT) paradigm may enable management processes capable of providing both remote control and monitoring of multimedia/multisensory experiences made available in different venues. A prototyping software to perform IoT multimedia/multisensory simulations is presented in this paper. It is fully based on virtual environments that enable the remote design, layout, and configuration of each experience in a transparent way, without regard of software and hardware. Furthermore, pipelines to deliver contents may be defined, managed, and updated in a context-aware environment. This software was tested in the laboratory and was proven as a sustainable approach to manage multimedia/multisensory projects. It is currently being field-tested by an international multimedia company for further validation.

2021

Grape Bunch Detection at Different Growth Stages Using Deep Learning Quantized Models

Autores
Aguiar, AS; Magalhaes, SA; dos Santos, FN; Castro, L; Pinho, T; Valente, J; Martins, R; Boaventura Cunha, J;

Publicação
AGRONOMY-BASEL

Abstract
The agricultural sector plays a fundamental role in our society, where it is increasingly important to automate processes, which can generate beneficial impacts in the productivity and quality of products. Perception and computer vision approaches can be fundamental in the implementation of robotics in agriculture. In particular, deep learning can be used for image classification or object detection, endowing machines with the capability to perform operations in the agriculture context. In this work, deep learning was used for the detection of grape bunches in vineyards considering different growth stages: the early stage just after the bloom and the medium stage where the grape bunches present an intermediate development. Two state-of-the-art single-shot multibox models were trained, quantized, and deployed in a low-cost and low-power hardware device, a Tensor Processing Unit. The training input was a novel and publicly available dataset proposed in this work. This dataset contains 1929 images and respective annotations of grape bunches at two different growth stages, captured by different cameras in several illumination conditions. The models were benchmarked and characterized considering the variation of two different parameters: the confidence score and the intersection over union threshold. The results showed that the deployed models could detect grape bunches in images with a medium average precision up to 66.96%. Since this approach uses low resources, a low-cost and low-power hardware device that requires simplified models with 8 bit quantization, the obtained performance was satisfactory. Experiments also demonstrated that the models performed better in identifying grape bunches at the medium growth stage, in comparison with grape bunches present in the vineyard after the bloom, since the second class represents smaller grape bunches, with a color and texture more similar to the surrounding foliage, which complicates their detection.

2021

Hydroponics Monitoring through UV-Vis Spectroscopy and Artificial Intelligence: Quantification of Nitrogen, Phosphorous and Potassium

Autores
Silva, AF; Löfkvist, K; Gilbertsson, M; Os, EV; Franken, G; Balendonck, J; Pinho, TM; Boaventura-Cunha, J; Coelho, L; Jorge, P; Martins, RC;

Publicação
Chemistry Proceedings

Abstract
In hydroponic cultivation, monitoring and quantification of nutrients is of paramount importance. Precision agriculture has an urgent need for measuring fertilization and plant nutrient uptake. Reliable, robust and accurate sensors for measuring nitrogen (N), phosphorus (P) and potassium (K) are regarded as critical in this process. It is vital to understand nutrients’ interference; thusly, a Hoagland fertilizer solution-based orthogonal experimental design was deployed. Concentration ranges were varied in a target analyte-independent style, as follows: [N] = [103.17–554.85] ppm; [P] = [15.06–515.35] ppm; [K] = [113.78–516.45] ppm, by dilution from individual stock solutions. Quantitative results for N and K, and qualitative results for P were obtained.

2020

Workload control and optimised order release: an assessment by simulation

Autores
Fernandes, NO; Thurer, M; Pinho, TM; Torres, P; Carmo Silva, S;

Publicação
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH

Abstract
An important scheduling function of manufacturing systems is controlled order release. While there exists a broad literature on order release, reported release procedures typically use simple sequencing rules and greedy heuristics to determine which jobs to select for release. While this is appealing due to its simplicity, its adequateness has recently been questioned. In response, this study uses an integer linear programming model to select orders for release to the shop floor. Using simulation, we show that optimisation has the potential to improve performance compared to ‘classical’ release based on pool sequencing rules. However, in order to also outperform more powerful pool sequencing rules, load balancing and timing must be considered at release. Existing optimisation-based release methods emphasise load balancing in periods when jobs are on time. In line with recent advances in Workload Control theory, we show that a better percentage tardy performance can be achieved by only emphasising load balancing when many jobs are urgent. However, counterintuitively, emphasising urgency in underload periods leads to higher mean tardiness. Compared to previous literature we further highlight that continuous optimisation-based release outperforms periodic optimisation-based release. This has important implications on how optimised-based release should be designed. © 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group.