Jon A. Justo

Dear user,
My personal professional motto is "letting data change today's future to tomorrow's truth". This is why I am convinced that this work and its data contribute with an important value to the open community. First, our data allows scientific exploration and research using a complete set of hundreds of images captured across many countries in the world by the NTNU/HYPSO-1 satellite. Second, our data with its ground-truth labels will be definitely crucial for AI researchers in our field in the need of extensive, consistent, and reliable ground-truth annotations to train their Machine Learning and Deep Learning models demanding the adequate generalisation of such models for unseen new data. Recently, one of my colleagues in the AI field told me: "never mess with the data!". Following this advice, we have put a thorough care on the data which we know is, certainly, the initial constructing brick of the AI models that many of you will be likely using the data for. As a matter of fact, in this work we indeed give an example of a Deep Learning model to solve a Computer Vision and Pattern Recognition problem both for on-ground data processing and for inference at the edge on-board HYPSO-1 platform, whose flexibility allows the update of its Software with the AI Deep Learning model at any time.
In my motto, I mention "tomorrow's truth" since data indeed has the capability to change our future for better. For a better and sustainable future, we must support the development of AI technology for the benefit of society, but in a safe and responsible manner. This intelligent technolgy allows us to automate many processes that end up benefiting our management and care of for instance, our Earth and ocean. This becomes even more important nowadays considering the climate change effects we are seeing, and it also aligns with NTNU's motto "knowledge for a better world".
Finally, I would like to express my grattitude to Professors Tor Arne Johansen at NTNU and Radu Tudor Ionescu at the University of Bucharest (UB) for their valuable feedback and insights while developing this work. I extend this appreciation to my fellow co-author colleagues in the article who have helped shape the article while giving me precious input. Thank you very much to all of you. Also, special thanks to Iuliana Georgescu also at UB, Anum Masood at the Dept. of Circulation and Imaging at NTNU and to the NTNU Small Satellites team who developed the payload of the HYPSO-1 satellite, in operation at the time of this work.
On behalf of my fellow colleagues, we hope this work brings value to your work and insights to you.
Thank you very much for reading these words.

Best regards,
Jon A. Justo
NTNU Small Satellites
Information and Communication Technologies Engineer (BSc, MSc)
Doctoral Researcher on Artificial Intelligence for Ocean Science at the Department of Cybernetics (NTNU)

Joseph Garrett

Dennis D. Langer

Marie B. Henriksen

Radu T. Ionescu

Tor A. Johansen