SBIR Phase I: Semantic 3D for infrastructure asset modeling, maintenance, and predictive analysis

Award ID: 2344140

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase I project is the on the effective maintenance, construction, resilience and performance of large-scale infrastructure. The technology developed in this work will dramatically improve the productivity, accuracy, and quality of the information products generated by surveyors and engineers charged with assessing, designing, and maintaining infrastructure. For example, the roughly 200M utility poles in US should be surveyed and corresponding solutions engineered approximately every three years for weather robustness, fire mitigation, line capacity, and for future overhead and underground extensions. The technology has broad application in adjacent domains like water, natural gas, mining, oil and gas, power generation, transportation, mapping, and construction. This Small Business Innovation Research (SBIR) Phase I project integrates geometric methods for constructing detailed three-dimensional (3D) models from photographs with semantic methods used to segment and classify features or objects in two-dimensional (2D) photographs. Key challenges include building 2D-3D correspondence across the geometric (3D) and semantic (2D) techniques, improving computational efficiency, and creating user interfaces to interact with and label data. The integrated computer vision capability will be validated on datasets for electric power distribution infrastructure to extract critical features like line attachment points, pole height and inclination, wire gauges, pole deterioration, vegetation intrusion, electrical component types. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Alastair Monk