[ad_1]
As the U.S. Coast Guard (USCG) icebreaker Healy embarks on a trip to the North Pole this summer, it’s capturing images of the Arctic to explore this rapidly changing region. Lincoln Laboratory researchers installed the camera system on the Healey while it was in the Port of Seattle before it began its three-month science mission on July 11. The resulting data set, which will be one of the first, will be used to develop artificial intelligence tools capable of analyzing Arctic imagery.
“These datasets not only help mariners navigate safer and more efficiently, but also provide our nation’s defense with critical maritime domain awareness and an improved understanding of how AI analysis can be applied in this complex and unique environment,” said Joe Kurukari, Lincoln Laboratory’s AI Software Architectures and Algorithms project.
As the planet warms and sea ice melts, Arctic passages are opening up to more traffic, both for military vessels and vessels engaged in illegal fishing. These movements may pose national security challenges to the United States. The opening arctic also leaves questions about how its climate, wildlife and geography are changing.
Today, there are very few Arctic image sets available to study these changes. Aerial images from satellites or aircraft can only provide limited information about the environment. An external camera mounted on a ship can capture more details of different angles of the environment and objects in the scene, such as other ships. These types of images can be used to train artificial intelligence computer vision tools to help the USCG plan and automate analysis of naval missions. According to Kurukari, USCG assets in the Arctic are widespread and could greatly benefit from artificial intelligence tools that can act as force multipliers.
The Healy is the USCG’s largest and most technologically advanced icebreaker. Given its current mission, it was a suitable candidate to be equipped with a new sensor to collect this data set. The laboratory’s research team collaborated with the USCG Research and Development Center to determine the sensor requirements. Together they created the Cold Region Imaging and Surveillance Platform (CRISP).
“Lincoln Lab has an excellent relationship with the Coast Guard, particularly the Center for Research and Development. For more than a decade, we have built relationships that have made the CRISP system possible,” said Amna Greaves, CRISP project leader and assistant leader of the AI Software Architectures and Algorithms Group. “We have strong ties not only because of the USCG veterans who work in the lab and on our team, but also because our technology missions complement each other. Today it’s deploying an infrared sensor in the Arctic, tomorrow it could be piloting four-legged robotic dogs on a rapid response cutter.”
The CRISP system includes a long-wave infrared camera manufactured by Teledyne FLIR (pre-infrared) designed for harsh marine environments. The camera can stabilize in rough seas and image in total darkness, fog and glare. It is paired with a GPS-enabled time-synchronized clock and network video recorder to record both video and still images along with GPS positional data.
The camera is mounted on the forward side of the ship’s flybridge, and the electronics are contained in a rack mounted on the bridge. The system can be operated manually from the bridge or placed in autonomous surveillance mode, in which it slowly rotates back and forth, capturing 15 minutes of video every three hours and a still image every 15 seconds.
“Installing the equipment was a unique and fun experience. As with any good project, our expectations for the installation did not meet reality,” says Michael Emil, the project’s IT systems administrator, who traveled to Seattle for the installation. Working with the ship’s crew, the lab team had to quickly adjust the route to run the cables from the camera to the observation station after discovering that the expected access points were in fact inaccessible. “We had 100 feet of cables made for this project just in case of this type of scenario, which was good because we only had a few inches to spare,” says Emily.
The CRISP project team plans to publicly release the database, expected to be about 4 terabytes in size, after the USCG science mission ends in the fall.
The goal of publishing the dataset is to enable the broader research community to develop better tools for those operating in the Arctic, especially as the region becomes more navigable. “Collecting and publishing data enables faster and greater progress than we can achieve on our own,” Kurukari adds. “It also allows the lab to engage in more advanced AI applications, while others make more progress using the dataset.”
In addition to providing data sets, the lab team plans to provide a baseline object detection model from which others can build their own models. More advanced AI applications planned for development are classifiers for specific objects in scenes and the ability to identify and track objects in images.
In addition to helping USCG missions, the project could create an impactful dataset for researchers trying to apply artificial intelligence to Arctic data to combat climate change, says Paul Metzger, who leads the AI Software Architectures and Algorithms Group.
Metzger adds that the group is honored to be part of this project and is excited to see the progress it has made in applying AI to the new challenges facing the United States: “I am very proud of how our group is addressing the highest priority AI challenges in our nation, from predicting the outbreak of CoVID-19 to the employment of the US command.
Once the data set is available, it will be available for free download on the Lincoln Lab data website.
[ad_2]
Source link
