Hi, I'm Felipe. I'm currently a research scientist specializing in Applied Artificial Intelligence at Saint Louis University (United States), while on leave from my
Associate Professor position at the Federal Institute of Alagoas (Brazil). Formerly, I was the Principal Investigator at
LEAD and a software developer.
I also served as a Visiting Researcher at the Karlsruhe Institute of Technology (Germany), and I completed my doctoral degree in Computer Science at
CIn-UFPE.
When I'm not researching, teaching, or developing, you'll probably find me
surfing or playing soccer.
Here, you can explore my research interests, the technologies I work with, and some of the papers I've published.
Applied Artificial Intelligence, Deep Learning, Geospatial Data Sciente, Models at Runtime, Software-Defined Networking, Software Engineering.
List of research projects that I contribute(d) to.
The "Climate Smart Carbon" project aims to revolutionize carbon management in agriculture by leveraging advanced remote sensing, IoT, and AI technologies. Focused on measuring and enhancing carbon sequestration in crops, the project integrates multi-source data, including hyperspectral imagery and LiDAR, to monitor and model carbon dynamics in real time. By providing farmers with actionable insights on soil health, crop productivity, and greenhouse gas emissions, this initiative seeks to promote sustainable farming practices that mitigate climate change while improving yields. The project is expected to contribute significantly to carbon-smart agriculture strategies and climate resilience on a global scale.
The goal of this project is to propose a system to orchestrate data gathered using sensors, such as hyperspectral and thermal cameras to collect imagery on soybean, sorghum, and orther crops. Preprocessed plant datasets will be then offered to scientistics and farmers in different formats via a web-based system, ready to be processed by deep learning algorithms or consumed by thin clients.
In this project, the aim is to develop and analyze models for the prediction of heart diseases using various patient cardiac attributes and to detect imminent heart diseases using machine learning techniques such as backward elimination algorithm, logistic regression, REFCV, and deep neural networks, both on publicly available datasets and data from popular heart monitors. The results are also assessed using confusion matrices and cross-validation. Early prognosis of cardiovascular diseases can assist in decision-making regarding lifestyle changes for high-risk patients, ultimately reducing long-term complications.
The object to be developed in this project is an intelligent platform to monitor the level of water needed in a given crop and direct an optimized amount of water to the places where there is demand for the resource. The Internet of Things (i.e., Internet of Things - IoT) with its multitude of sensors, actuators and other related technologies (e.g., Machine Learning) can be used for this purpose. Thus, the platform to be developed: (i) will have a set of IoT tools combined to monitor, through drones and/or field sensors, the crop's water needs; (ii) will act in defining the opening or closing of water pumps to direct the amount of water to specific areas of cultivation; (iii) it will have prediction models - based on Machine Learning and/or application of water balance models (e.g., ACQUACROP) - to control the water pumps used in irrigation.
Considering that Machine Learning models have a high dependence on the data that are used in their training processes, this project aims to analyze the performance of these models to perform the prediction of COVID-19 in local contexts of the State of Alagoas, in the northeast of Brazil. For this, databases from the region will be used to evaluate the predictions obtained from the models. In this way, it is expected to obtain statistically valid conclusions to identify which characteristics and algorithms should be taken into account to obtain relevant information that will help public health agents in this and similar states to identify the course of COVID-19.
The objective is to map the open data ecosystem of the chambers, which will support proposals to improve quantitatively and qualitatively the access to information in the municipalities. For this, we will investigate the transparency of Alagoas municipal councils from the collection of empirical evidence. This evidence will form part of a diagnostic report and a transparency ranking.
The SWAMP project developed a high-precision smart irrigation system concept for agriculture. The main idea is to enable the optimizations of irrigation, water distribution, and consumption based on a holistic analysis that collects information from all aspects of the system, including even the natural water cycle and the cumulated knowledge related to growing particular plants. It results to savings for all parties as it detects all the leakages and losses and guarantees better the availability of water in situations where water supply is limited.
The SWAMP project developed a high-precision smart irrigation system concept for agriculture. The main idea is to enable the optimizations of irrigation, water distribution, and consumption based on a holistic analysis that collects information from all aspects of the system, including even the natural water cycle and the cumulated knowledge related to growing particular plants. It results to savings for all parties as it detects all the leakages and losses and guarantees better the availability of water in situations where water supply is limited.
The SWAMP project developed a high-precision smart irrigation system concept for agriculture. The main idea is to enable the optimizations of irrigation, water distribution, and consumption based on a holistic analysis that collects information from all aspects of the system, including even the natural water cycle and the cumulated knowledge related to growing particular plants. It results to savings for all parties as it detects all the leakages and losses and guarantees better the availability of water in situations where water supply is limited.
This project focused on researching and developing methodologies to enhance the implementation of applications within Software-Defined Networks (SDN). The work explored the programmability of network infrastructures, offering novel approaches to improve network performance, flexibility, and security. Through the development of innovative SDN applications, this research contributed to more efficient traffic management, automated network control, and the advancement of network virtualization techniques. The outcomes of this project are applicable in data centers, enterprise networks, and telecommunications infrastructure.
In this project, the focus was on assessing the performance and optimization of cloud computing controllers, which manage the allocation and utilization of virtualized resources. The research evaluated existing cloud control algorithms, identifying key inefficiencies and proposing optimizations to enhance scalability, resource management, and energy efficiency. By optimizing cloud controllers, this project aimed to improve the overall performance of cloud environments, ensuring better service delivery and cost-effective resource utilization for enterprises and cloud providers alike.
List of selected papers published from my research. For a full list, please visit my Google Scholar profile.
A multifaceted benchmarking of GAN architectures on generating synthetic satellite imagery
Kevin Wells, Felipe A. Lopes, Vasit Sagan, Flavio Esposito.
52nd Applied Imagery Pattern Recognition Workshop (AIPR). 2023.
PC2I: Cyber-Physical Platform for Smart Irrigation through the Internet of Things
Jammys G.B. da Silva, Jabes S. Santos, Josefa L.M. dos Santos, Carlos E.B. Santos, Tony Silva, Tamilly K. do Nascimento, Felipe A. Lopes.
Congress of the Brazilian Computer Society. 2021.
Using the RFC 7575 and Models at Runtime for Enabling Autonomic Networking in SDN
FA Lopes.
Congress of the Brazilian Computer Society. 2021.
Model-based flow delegation for improving SDN infrastructure compatibility
FA Lopes, P Tiburcio, R Bauer, S Fernandes, M Zitterbart.
NOMS 2018-2018 IEEE/IFIP Network Operations and Management Symposium, 1-9. 2018.
Automatically Generated Simulations for Predicting Software-Defined Networking Performance
FA Lopes, RR Souza, S Fernandes
IEEE Symposium on Computers and Communications (ISCC), 915-918. 2018.
A Software Engineering Perspective on SDN Programmability
FA Lopes, M Santos, R Fidalgo, S Fernandes.
IEEE Communications Surveys & Tutorials. 2015.
GitHub repositories that I've been working on.
For the full list of repositories, please visit my GitHub profile.
Posts about research findings, articial intelligence, software development, and related stuff.
Papers I've read in my research.
Technologies that I use to work or research.