Felipe A. Lopes, Ph.D.
Research Scientist, Remote Sensing Laboratory, Saint Louis University
St. Louis, Missouri, USA
felipealencarlopes@gmail.com | Google Scholar | LinkedIn

Summary

I am a research scientist working on machine learning for Earth observation. I fine-tune geospatial foundation models (Prithvi-EO-2.0, AlphaEarth) and build graph-based and generative models for multi-sensor satellite and UAV time series, applied to forest monitoring and precision agriculture. I led the writing of GENESIS, a $900,000 USDA-NIFA-AFRI award starting in January 2027, on which I am Co-PI. My Ph.D. in Computer Science was on Q-learning for adaptive runtime models in software-defined networking, and I bring over a decade of software engineering practice to research code.

Experience

Research Scientist
Saint Louis University, St. Louis, USA | 06/2024 – present

  • Build and maintain Python and Google Earth Engine pipelines that ingest, align, and quality-check multi-year satellite and UAV time series, and fine-tune geospatial foundation models (Prithvi-EO-2.0, AlphaEarth) on them for forest monitoring and precision agriculture.
  • Developed the Geospatial Time Machine, a graph neural network-based generative model that enhances the spectral and temporal resolution of satellite data.
  • Lead development of ForestTrace, a forest disturbance-to-recovery monitoring platform built on fine-tuned foundation models and multi-sensor data, applied in Ontario boreal forests and California mixed-conifer forests.
  • Led the writing of the GENESIS proposal, funded as a $900,000 USDA-NIFA-AFRI grant starting January 2027, on which I am Co-PI; continue to write proposals to NSF and NASA ROSES.
  • Advise and mentor students through their research projects.

Postdoctoral Fellow
Saint Louis University, St. Louis, USA | 01/2023 – 05/2024

  • Early plant disease detection: developed the full data cleaning and machine learning pipeline, resulting in a generative adversarial network (GAN)-based method that improved classification accuracy to about 80%.

Senior ML Software Engineer
Doity – Events Platform, Remote, Brazil | 12/2022 – 12/2023

  • Recommendation system: managed planning, budgeting, and scheduling, and led the design and implementation of the recommendation and machine learning algorithms (Python, PHP, Laravel, Vue.js).

Founder and Coordinator
Laboratory of Data Engineering and Analysis (LEAD), Arapiraca, Brazil | 03/2021 – 12/2022

  • Co-founded the first data science laboratory in the state of Alagoas, obtaining almost $100k in research grants.
  • Cyber-physical platform for smart irrigation: coordinated development of a low-cost irrigation platform using IoT and machine learning models.

Associate Professor
Federal Institute of Alagoas (IFAL), Arapiraca, Brazil | 03/2018 – 04/2024

  • Won 2 national awards and secured over BRL 300,000 in approved research grants.

Visiting Researcher
Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany | 09/2016 – 03/2017

  • Models at runtime for SDN applications: built Petri net and Markov chain models of network behavior and implemented them in Python within SDN frameworks for experimental evaluation (NOMS 2018).

Research Assistant
Foundation for the Support of Science and Technology of Pernambuco, Recife, Brazil | 03/2013 – 03/2020

  • Adaptive SDN control and DDoS detection: applied Q-learning to runtime models for autonomous SDN policy adaptation (Ph.D. research), and developed machine learning models in Python and PyTorch to detect and mitigate DDoS attacks in SDN and P4-based networks in real time.

Full-Stack Developer
Indra Company, Maceió, Brazil | 03/2012 – 03/2013

  • Designed, developed, and maintained front-end and back-end components of software applications.

Education

  • Ph.D. in Computer Science, Federal University of Pernambuco (UFPE), Brazil, 2015 – 2020. Dissertation: Q-learning for adaptive runtime models in SDN.
  • M.Sc. in Computer Science, Federal University of Pernambuco (UFPE), Brazil, 2013 – 2015.
  • B.Sc. in Information Systems, Federal Institute of Alagoas (IFAL), Brazil, 2008 – 2013.

Selected Publications

  1. M. Rahaman, V. Sagan, F. A. Lopes, H. Alifu, C. Gul, H. Aliakbarpour, et al. Self-supervised learning for soybean disease detection using UAV hyperspectral imagery. Remote Sensing, 17(23), 3928, 2025.
  2. F. A. Lopes, V. Sagan, S. Sarkar, A. Stylianou, F. Esposito. Geospatial time machine: a generative model to enhance spectral–temporal data resolution. IEEE Transactions on Geoscience and Remote Sensing, 63, 1–13, 2025.
  3. F. A. Lopes, V. Sagan, A. Pawar, H. Alifu. SoilSR: A soil-oriented super-resolution method to enhance satellite-based SOC analysis. Optical Sensors, 2025.
  4. J. Skobalski, V. Sagan, H. Alifu, O. Al Akkad, F. A. Lopes, F. Grignola. Bridging the gap between crop breeding and GeoAI: Soybean yield prediction from multispectral UAV images with transfer learning. ISPRS Journal of Photogrammetry and Remote Sensing, 210, 260–281, 2024.
  5. F. A. Lopes, V. Sagan, F. Esposito. PlantPlotGAN: A physics-informed generative adversarial network for plant disease prediction. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024.
  6. F. A. Lopes, P. Tiburcio, R. Bauer, S. Fernandes, M. Zitterbart. Model-based flow delegation for improving SDN infrastructure compatibility. IEEE/IFIP Network Operations and Management Symposium (NOMS), 2018.
  7. F. A. Lopes, M. Santos, R. Fidalgo, S. Fernandes. A software engineering perspective on SDN programmability. IEEE Communications Surveys & Tutorials, 18(2), 1255–1272, 2016.

Full list on Google Scholar.

Professional Service

  • Area Chair, NeurIPS 2026 Position Paper Track.
  • Mentorship Program Chair, LatinX in AI Workshop at NeurIPS 2025.
  • Guest Editor, MDPI Remote Sensing Special Issue on AI for hyperspectral and multimodal plant stress and disease detection (in preparation).
  • Reviewer for journals including the ISPRS Journal of Photogrammetry and Remote Sensing.
  • IEEE Member.

Technical Skills

  • Programming: Python, Java, JavaScript, PHP, Node.js.
  • Machine learning: PyTorch, TensorFlow; geospatial foundation model fine-tuning (Prithvi-EO-2.0, AlphaEarth); Google Earth Engine.
  • Data and systems: Linux/Bash, Git; data pipelines (Kafka, Spark); SQL and NoSQL query design; data cleaning; Terraform.