Characterization of a Vision-Based Tool for the Investigation of Geometric Characteristics of Ground-Deposited Volcanic Ash

With the support of public authorities and research institutions, volcanic ash fallout and its impact on the safety of people, infrastructure and services are addressed with the aim of defining protocols and instruments for the reliable and effective handling of related emergencies. Most of the solutions proposed in the literature on ash fallout monitoring suffer from high cost and are demanding in terms of installation and maintenance. The approach suggested in this work is based on the use of a low-cost vision embedded system and a dedicated algorithm which automatically processes acquired frames of ground-deposited volcanic ash in order to estimate the main geometric properties of each particle identified in the work area. A complete characterization of the system is presented, along with a robustness analysis of particle shapes, their orientation and their position in the inspected frame. An accuracy of +/- 40.2 mu m (with a 3 sigma limit) and a resolution of 32.9 mu m (in the worst case), over a framed area of 130 mm by 100 mm, were estimated; these values would fulfill the objectives of the application.

Automated summary

This study proposes a low-cost method for volcanic ash monitoring, using a vision embedded system and a dedicated algorithm to automatically process frames of collected volcanic ash and estimate its geometric properties. The system has high accuracy and resolution, and can meet the objectives of the application.