POSTER Session 3

INNOVATION AND ACCESSIBILITY IN REMOTE SENSING: DEVELOPMENT OF AN APPLICATION FOR THE VISUALIZATION AND CLASSIFICATION OF HYPERSPECTRAL IMAGERY

Remote Sensing (RS) plays a key role in Earth monitoring. The combination of hyperspectral imagery (HSI) with artificial intelligence (AI) algorithms is a valuable tool for the characterization of different surfaces, offering more information than a conventional RGB camera. Nevertheless, classification of HSIs faces challenges due to large data volume and the costly processing, which delays AI models training. The use of AI algorithms reduces dimensionality, optimizing the use of hardware for classification. However, these technologies are mainly accessible to those with programming skills. The presented application overcomes this problem providing users without programming experience with access to visualization, spectral dimensionality reduction and classification of HSI through a user-friendly interface. Initially developed to identify rhodamine concentrations in marine water as a first step to the detection of pollutants, this tool has the potential to develop AI models that differentiate water and pollutants in real time using hyperspectral cameras on board drones. Currently, the application has demonstrated its potential for training AI algorithms and facilitating their convenient visualization and interpretation. A language model is being developed allowing document uploads and querying to enhance it for accessing and interpreting project data. The main goal for the short term is to collaborate with specialized HSI users to evaluate its real impact and make the application accessible online. The implementation of this technology not only democratizes access to advanced Earth observation tools among the scientific community but also enhances research and environmental protection by simplifying the management and analysis of large HSI data volumes.

*Alba Martín Lorenzo, University of Las Palmas de Gran Canaria, Spain, [email protected]

Ámbar Pérez García, University of Las Palmas de Gran Canaria, Spain, [email protected]

Cayetano Guerra Artal, University of Las Palmas de Gran Canaria, Spain, [email protected]

María Dolores Afonso Suárez, University of Las Palmas de Gran Canaria, Spain, [email protected]

José Francisco López Feliciano, University of Las Palmas de Gran Canaria, Spain, [email protected]

Code of Conduct
Privacy Policy