Pipeline Threat Detection
Image & Pattern Recognition
Is that object a downed tree limb or an enemy combatant vehicle? See how our client used crowdsourcing to recognize and tag objects with more speed and improved accuracies.
Protecting a portion of the nation’s fuel supply is of critical importance. Using savvy algorithms to help detect potential threats allows the U.S. Government to do this job much more efficiently and securely.
Aerial imagery has become abundant and much less expensive to attain thanks to advances in satellite capabilities and drone technology. Our U.S. government client desired an algorithmic solution to detect and classify objects within a certain range of an energy pipeline.
The algorithm needed to both identify the objects and effectively tag the detections with an appropriate threat confidence level.
The winning algorithmic solutions are able to effectively process tens of thousands of images in greatly condensed time frames. The gains in speed & accuracy have delivered to our client an efficient and secure way to detect and evaluate potential risks.
The algorithmic solution is being used to drive other government research in planetary satellite classification, Mars reconnaissance, Federal disaster response & recovery and beyond.