The Panoptes project, which uses data and technologies from the Brazil Data Cube project, wins the Reconhe-Ser 2025 Award
The Panoptes project, led by the Federal Court of Accounts (TCU) with the participation of INPE researchers Karine Ferreira, Gilberto Queiroz, Marcos Adami, and Thales Korting, was the winner of the Reconhe-Ser Award 2025, receiving recognition from both the technical committee and the jury. Developed using data and technologies from the Brazil Data Cube, the project stood out among more than 90 initiatives submitted in the External Control category, demonstrating its high impact on the modernization of Brazilian public administration.
The Reconhe-Ser Award is an initiative of the Federal Court of Accounts created to celebrate the work of public servants and the positive impact generated on society and Public Administration. In its 13th edition, held at the TCU Cultural Center, the award recognized dozens of outstanding initiatives, honoring ten projects in the External Control category, five in the Governance and Management category, and two innovative ideas.
More than a symbolic award, Reconhe-Ser values engagement, innovation, and excellence in public service, strengthening institutional motivation and formally recording these contributions in the professional trajectories of public servants. This was highlighted by the award coordination, which emphasized its role in inspiring, recognizing, and driving new achievements within the TCU.
The dual positive evaluation does not only validate methodological rigor; it demonstrates that this project has the potential to help address a long-standing and real problem in Brazilian public administration. The results that support this award stem from Phase 1 – Structuring and proof of concept (2023–2024), during which Panoptes delivered concrete and scalable outcomes:
- Nationwide automation of rural credit and insurance monitoring, integrating satellite imagery and artificial intelligence into INPE services (Brazil Data Cube, WLTS, and WTSS);
- Large-scale detection of irregularities through four proofs of concept that analyzed approximately 240,000 operations and identified indications of irregularities ranging from 0.7% to 26.6%, totaling more than BRL 2.5 billion in suspicious operations;
- Novel indicators of the effectiveness of agricultural policies, revealing, for example, that only 26.8% of the areas financed for pasture restoration showed improvement;
- Capacity building and transparency, with the production of hundreds of hours of open educational content (GitHub, Kaggle, and YouTube), including eight annotated Jupyter Notebook workbooks, enabling replication and improvement by other stakeholders;
- Readiness for scaling, with the capacity to monitor approximately BRL 605 billion in rural credit per agricultural year.
In addition to technical innovation, Phase 1 consolidated an institutional innovation by bringing together 118 participants from 16 public, academic, and civil society organizations, prioritizing open, low-cost, and sustainable technologies. This solid foundation explains the dual recognition received and reinforces the potential of Panoptes to tackle long-standing challenges in external control in Brazil.
The project also received support from the Brazil Data Cube team through PhD candidates Gabriel Sansigolo and Baggio Luiz de Castro e Silva. The work of both was essential for monitoring operations, directly incorporating research that explores the use of satellite image time series for agriculture. Results from this effort have been published, including the study presenting a web service for extracting phenology metrics from large volumes of Earth observation data[1] and the research that develops methods to detect atypical patterns in rural credit areas based on these time series[2]. These contributions were fundamental to the automation and analytical accuracy of the Panoptes project.
The TCU and its institutional partners sincerely acknowledge the invaluable contribution made by the four INPE researchers to the future of external control in Brazil. The dual award reflects this lasting impact and recognizes the genuine commitment they dedicated to Panoptes.
For more information:
The detailed article “Panoptes Project: Monitoring Credit Operations Through Geotechnologies and Artificial Intelligence” describes the methodology, phases, and results of the project.
The Call for Proposals for the Reconhe-Ser Award 2025 (Call SecPessoas No. 1, August 15, 2025) establishes the rules, criteria, and institutional context of the award.
[1] SANSIGOLO, Gabriel et al. A Web Service for Phenology Metrics Extraction from Big Earth Observation Data. In: Proceedings of the XXI Brazilian Symposium on Remote Sensing, 2025, Salvador. Electronic proceedings, Galoá, 2025. Available at:
<https://proceedings.science/sbsr-2025/papers/a-web-service-for-phenology-metrics-extraction-from-big-earth-observation-data?lang=pt-br>. Accessed on May 5, 2025.
[2] SILVA, Baggio Luiz de Castro et al. Detecting Atypical Agricultural Patterns in Rural Credit Applications Based on Satellite Image Time Series. In: Proceedings of the XXI Brazilian Symposium on Remote Sensing, 2025, Salvador. Electronic proceedings, Galoá, 2025. Available at:
<https://proceedings.science/sbsr-2025/trabalhos/detectando-padroes-atipicos-de-agricultura-em-aplicacoes-de-credito-rural-basead?lang=pt-br>. Accessed on May 5, 2025.