Hernández-Orellana et al. (2026) Methodological Framework for the Capture and Management of a Hyperspectral Image Dataset Using IoT

Identification

• Journal: Lecture notes in networks and systems
• Year: 2026
• Date: 2026-01-01
• Authors: Carlos Hernández-Orellana, David Raimundo Rivas-Lalaleo, Fernando Caicedo-Altamirano, Carlos Bran, José Luis Serrano-Mira, Jesús Barba-Romero
• DOI: 10.1007/978-981-96-9048-0_35

Research Groups

• Instituto Investigación e Innovación en Electrónica, Universidad Don Bosco, Soyapango, El Salvador
• Universidad de las Fuerzas Armadas—ESPE, Sangolquí, Ecuador
• Universidad Castilla la Mancha, Ciudad Real, Spain

Short Summary

This study presents an IoT-based framework for the systematic capture, management, and analysis of hyperspectral image datasets in a controlled experimental setup simulating hydrocarbon contamination. It successfully generated a comprehensive dataset, addressing the scarcity of specialized data for hydrocarbon spill analysis and laying the groundwork for future machine learning model development.

Objective

• To develop and present an IoT-based methodological framework for the collection, management, and analysis of hyperspectral data in a controlled experimental setup simulating hydrocarbon contamination.
• To address the current scarcity of specialized hyperspectral datasets for hydrocarbon spill analysis and establish a foundation for developing robust predictive models using machine learning techniques.

Study Configuration

• Spatial Scale: Controlled experimental setup simulating hydrocarbon contamination (specific dimensions not provided).
• Temporal Scale: Data captured over time to integrate contaminant behavior (specific duration not provided).

Methodology and Data

• Models used: No specific predictive models were used in this study, as the focus was on data capture and management. The framework is designed to support future development of robust predictive models using machine learning techniques.
• Data sources:
  • State-of-the-art hyperspectral camera for image acquisition.
  • IoT principles for data capture, processing, and storage.
  • Cloud-based repository for data management, ensuring accessibility and scalability.

Main Results

• A comprehensive hyperspectral dataset was successfully generated, integrating contaminant behavior over time in a controlled experimental setup.
• An IoT-based framework was established for the systematic capture, processing, and storage of hyperspectral data in a cloud-based repository.
• The framework incorporates ethical and sustainable database design practices to enhance the dataset’s usability for environmental and scientific applications.
• The generated dataset addresses the existing scarcity of specialized hyperspectral datasets crucial for hydrocarbon spill analysis.

Contributions

• Provides an original IoT-based methodological framework for the capture and management of hyperspectral image datasets, specifically tailored for environmental monitoring of hydrocarbon spills.
• Creates a much-needed, specialized hyperspectral dataset for hydrocarbon spill analysis, which was previously scarce in existing literature.
• Establishes a foundational infrastructure for the development of robust predictive models using machine learning techniques for environmental monitoring.
• Integrates ethical and sustainable database design practices, promoting responsible data management in scientific applications.

Funding

• “Arquitectura Hardware Para Clasificación De Reacciones Químicas Con Firmas Hiperespectrales” (Universidad Don Bosco, El Salvador, 2024)
• ESPE-PIJ-08-2022
• ESPE-PIS-09-2022
• UNACH Conv-2022-05
• UPS SISMO-ROSAS
• CYTED—REDTPI4.0 thematic network

Citation

@article{HernándezOrellana2026Methodological,
  author = {Hernández-Orellana, Carlos and Rivas-Lalaleo, David Raimundo and Caicedo-Altamirano, Fernando and Bran, Carlos and Serrano-Mira, José Luis and Barba-Romero, Jesús},
  title = {Methodological Framework for the Capture and Management of a Hyperspectral Image Dataset Using IoT},
  journal = {Lecture notes in networks and systems},
  year = {2026},
  doi = {10.1007/978-981-96-9048-0_35},
  url = {https://doi.org/10.1007/978-981-96-9048-0_35}
}