IE11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Google Pilot Helps State Emergency Office Get Eyes on Fire Aftermath

Using the Silicon Valley giant’s Street View technology, the state’s emergency services agency is piloting a program to more quickly and efficiently calculate wildfire damage through video and artificial intelligence.

Cal OES Google 2.PNG
Facebook
The state’s emergency services office is working on a pilot project involving one of the biggest names in technology, Google, to help with one of the state’s most pernicious problems: wildfires.

Specifically, Google Cloud Public Sector is working with the California Governor’s Office of Emergency Services (Cal OES) and the Western Fire Chiefs Association on a pilot program to use Google’s Street View technology to assess damage in the aftermath of a wildfire.
Cal OES Google 3.PNG
In this Facebook screenshot, Grady Joseph, an assistant director for Cal OES (left), mounts a Google Street View camera atop a vehicle as part of the damage-assessment project that the state and the company are working on together.
Facebook

“Whether it’s private property or public infrastructure damage … we need to catalog that damage and ideally put a cost behind it,” Eli Owen, an assistant director of Cal OES, told Techwire in an interview. “To date, historically, it’s been a very manual process. I send a team out there and they literally put eyes on – in partnership with the county, usually – walking the damage and making damage estimates. They take pictures to catalog it all. But as you can imagine, although we can be Johnny-on-the-spot as soon as the fire stops burning, it’s still a very manual process.”

Owen and fellow Cal OES Assistant Director Grady Joseph have been working with Google to “basically streamline this whole process,” Owen said.

“We’re using things like artificial intelligence and Google’s cameras,” Owen said. “Because Google or we would have driven it (the burned area) before, we’ll have a baseline of what it looked like previously, and then we’ll come back through, post-damage, and we’ll be able to see the delta. And as you’re driving it, you’re doing it much quicker than my (team members) who are walking it. And so you’re taking a manual process and automating it.”

Assessing the damage after a large-scale wildfire is essential in qualifying for federal disaster reimbursement, insurance claims and other elements of a disaster’s aftermath.

“It’s very exciting, very innovative and something that government should be trying to do, particularly in California,” Owen said. “And since we have our partners in Silicon Valley so close to us, we should be trying to think of how to solve intractable problems with them by partnering. It’s an incredible public partnership, tackling one of the more intractable issues that the state’s continuing to face.”

Joseph and others from Cal OES and Google posted a video on Facebook depicting camera-bearing Cal OES vehicles. In the video, Joseph is seen mounting a camera on a tripod atop a Cal OES vehicle, which then cruises slowly through a burned-out area, recording data. The pre- and post-fire imagery is then compared, and damage assessments are calculated.

Using artificial intelligence and algorithms, state authorities can compile damage estimates much more quickly and inexpensively than by sending staff out on foot with cellphone cameras and clipboards.

“Google is, in this instance, the platform to help with the big data,” Owen said.

“The goal is to expedite assistance to different communities to get major programs turned on, and help individuals and communities who’ve been impacted,” said Joseph. “There have been limited pilots over the last couple of years. Now we're really trying to move the needle forward in terms of expediting damage assessment using the most innovative technology.”

“Once the 360° images are uploaded into the cloud, personally identifiable information such as faces, and license plates will be automatically blurred for privacy protection,” Cal OES said in a statement. “After, Google Cloud’s BigQuery and machine learning (BQML) technology will process the data so that state personnel can take action to provide recovery assistance and help to quickly rebuild. Cal OES plans to use several different data sources, including live and satellite imagery from Google Cloud, to get a higher level of accuracy in its damage assessments.”

Google isn’t charging the state for the work on the pilot project.

If the initiative works as planned, the same technology could be used in other states with other problems – flooding in the Southeast, for instance.

“Customer service is key to not only our mission success, but also to the hope of people who lost everything,” Owen said. “Using cutting-edge technology – like algorithms and artificial intelligence – to shorten the damage assessment process, for example, will have a direct impact on our ability to help the survivors in rebuilding and debris removal.”

Mike Daniels, vice president of Google Cloud Public Sector, said the turnaround in assessing damage will be significantly shorter.

“The state of California is piloting Google Cloud technology to help expedite fire recovery assessment from 30 days to just one day,” Daniels said. “This is a significant step forward in terms of helping Americans who have so often been displaced and significantly impacted in other ways during natural disasters.”

While Cal OES teams train with and test the 360° cameras, the manual damage assessment process will continue through the 2021 wildfire season. Upon successful pilot completion, and contract award, the technology will be integrated into the Cal OES system over the next year and expanded to address other natural disasters like floods and earthquakes.
Dennis Noone is Executive Editor of Industry Insider. He is a career journalist, having worked as a reporter and editor at small-town newspapers and major metropolitan dailies in California, Nevada, Texas and Virginia, including as an editor with USA Today in Washington, D.C. He lives in Northern California.