AI Inspection Model Library

InFlightAI’s Analog Gauge Reading model uses computer vision AI to automatically read analog pressure, temperature and flow gauges, as well as tap counters. Trained on thousands of gauges across hundreds of industrial types, it accurately interprets readings from varied angles using parallax correction and supports complex configurations, including multi-needle and side, bottom and center-pivot designs.

A neural network locates each gauge face in the frame, then matches it against a stored template to read needle position. Parallax correction accounts for the angle of capture, so readings hold up when the approach isn’t head-on. An existing template library covers most common industrial gauge faces, so those read from day one; faces outside the library are configured for your site during onboarding.

A gauge drifting out of range is flagged on the mision where it happens rather than at the next manual read. Readings arrive as logged data, so trend lines build themselves instead of living on clipboards, and one configured template covers every gauge of that type across the site.

How It Works

Analog Gauge Reading returns digitized gauge output. It reads pressure, temperature and flow gauges as well as tap counters, and handles multi-needle dials along with side-, bottom- and center-pivot needle configurations. Up to four gauges can be read from a single image.

Analog Gauge Reading is an asset maintenance model, currently deployed across industries like Electric Utilities, Oil & Gas, Manufacturing, Logistics, Power Generation, Nuclear and Mining. Detections are returned after the flight has landed, once imagery and waypoint data have been synced to the platform. The model is ready from day one, with no site-specific configuration required.