Harris partnered with a railroad company to apply its deep learning technology on LiDAR data for asset inventory. The railroad company needed to find a variety of 3D objects (signals, crossings, boxes, poles). Applying deep learning to LiDAR data for 3D feature extraction can be very complex. However, Harris deep learning technology has proven to be very successful, and in this case extracted the features the railroad company was looking for with 90%+ accuracy.

MEGA is used by utility companies as a dependable and cost-effective solution to augment manual Transmission & Distribution inspections. In one recent customer engagement, Harris Geospatial successfully deployed MEGA to identify specific anomalies on distribution poles that, had they gone undetected, might have disrupted service. MEGA was also used to locate poles and insulators, as well as identify anomalies.

The modern battlefield pose great challenges for even the most advanced militaries and intelligence organizations. Harris Geospatial has experience implementing solutions that use deep learning technology to address a wide range of GEOINT problems. Whether it’s navigating an armed conflict in an urban area, or sorting through an information minefield, MEGA will identify patterns, links, and anomalies to provide actionable information.

Harris Geosptial has partnered with EdgeData to bring imagery expertise and deep learning to wind turbine inspections using Unmanned Aerial Systems (UAS). By using UAS, the risk of climbing a turbine is mitigated, and costs and inspection times are greatly reduced. MEGA is used to discern damage to turbine blades from lightning strikes that require repair verses dirt, paint chips, and damage from bird strikes that does not need to be repaired. Harris Geospatial’s solution can be repeatedly applied to produce correct damage assessment accuracies of greater than 95%.