Technology Report

May 25, 2026

LOC8

FLAGSTAFF — The drone flight is the easy part. Then someone has to look at the pictures. The Coconino County Sheriff’s Office is using software that looks at every pixel for a shirt color, Arizona’s Family reported. After a drone covers a search area, the office runs the photos through LOC8. If the missing person was last seen in red, the software hunts for red.

Deputy Paul Clifton, the assistant search-and-rescue coordinator, told the station the problem is not the flight. It is the pictures. “As opposed to looking at thousands of photographs with the human eye that fatigues in 10 minutes, we’re leveraging automation tools,” he said. His illustration — “Maybe it reduces the number of photos you have to look at from a thousand down to twenty” — is his estimate.

“I’m looking for red because we believe that that’s what the person was last seen wearing,” Clifton said. He told the station the software accounts for how light reflects off objects. If LOC8 flags something, he said, it gives searchers a GPS location. DroneXL, writing on May 25, named the current package Loc8 G2, from Unmanned Systems Research of Saskatoon, and said Coconino is pairing it with AI-capable Skydio drones.

The product exists because a search once missed a person who was in the photos. Commercial UAV News reported in 2020 that Loc8, pronounced “locate,” was released in 2019 after searchers went through more than 1,500 drone images and did not find the missing person. Weeks later they concluded the subject had been captured in several frames, partly hidden by evergreen, and was not identifiable from the air with the naked eye. Anthony Lockly, co-founder and chief executive of Loc8 LLC, told the trade paper that the industry had focused on the aircraft, not on software to read the pictures.

Unmanned Systems Research describes Loc8 G2 as a second-generation color image-analysis tool. The company says it works at the individual pixel and can put a flagged target on a map. Commercial UAV News said the algorithm scans imagery from drones, manned aircraft, and ground cameras against user-defined color palettes.

Coconino has flown drones since 2018, Arizona’s Family reported. The office now flies Skydio drones with obstacle avoidance from six onboard cameras. Clifton said that sphere of vision helps in canyons where GPS is poor. DroneXL wrote that the program began with two DJI Mavic 2 Enterprise aircraft and later moved to Skydio.

The sheriff’s office is also working with an unnamed company in Scotland to train models on Northern Arizona terrain, Clifton told Arizona’s Family. “Northern Arizona versus Southern Arizona, the context, the environments are very different.” He said the tools can re-analyze video from past flights. He also said the custom-training feature is still being tested.

A documented Loc8 case, separate from Coconino, came out of Moose Lake, Minnesota. The Moose Lake Star Gazette reported in March 2024 that drone operator Earl Bakke used Loc8 on about 1,200 photographs while searching wooded land for Gerald Knapp, 86, who had dementia. Bakke said that within an hour and a half the program identified a grouping of red pixels, and closer analysis determined it was the missing man.

The same idea — drone photographs, then software looking for a color that does not belong — has shown up on other searches. In January 2026, BBC Future described mountain rescuers in Italy’s Piemonte region using AI to sort thousands of drone images after climber Nicola Ivaldo went missing. The software flagged a red helmet. That was a parallel use of drone-photo AI, not Loc8.