CARE-ON – Can we Rely on Our Network
CARE-ON is a research project conducted and funded by IGG – CNR with the aim of investigating the variability associated with human error in the manual identification of geometric features from visual media. Specifically, the chosen case study concerns the identification of boulders and craters using images of extraterrestrial surfaces, a context of fundamental importance for planetary sciences, where unfortunately on-site observations are not possible except in rare cases.
Although machine learning algorithms are continually advancing in automating these tasks, human identification and image interpretation skills remain fundamental for their training, accuracy control, and validation of results. Manual methods, on the other hand, inevitably introduce a component of subjective variability due to different individual sensitivity, fatigue, and observer expertise/experience.
CARE-ON aims to explore and mitigate these uncertainties by assessing the impact of the human factor on this process and developing guidelines that maximize identification reliability and accuracy.
The project also experiments for the first time with the possibility of involving non-professional volunteers in these identification procedures, subjecting the reliability of results obtained by expert researchers, students, and amateurs to comparative analyses.
CARE-ON shares the results openly, encouraging the adoption of protocols and guidelines, and promoting an inclusive approach to planetary science research.
Main objectives:
1) Define best practices for the visual identification of boulders and craters
2) Estimate operator-related variability;
3) Test citizen science approaches to the activity.