Security Dealer & Integrator

OCT 2018

Find news and information for the executive corporate security director, CSO, facility manager and assets protection manager on issues of policy, products, incidents, risk management, threat assessments and preparedness.

Issue link: https://sdi.epubxp.com/i/1041113

Contents of this Issue

Navigation

Page 68 of 78

66 Security Dealer & Integrator / www.SecurityInfoWatch.com October 2018 Enabling Artificial Intelligence Both radar and LiDAR are helping to enable further use of Artificial Intelli- gence and deep learning in perimeter detection systems. e detail and accuracy of LiDAR has enabled one company – Cana- da-based Geo-Plus – to create and integrate AI into its functionality. In fact, its VisionLidar point cloud so- ware includes an object recognition tool to automatically detect and recog- nize the geometry of an object. "Many solution providers have been successfully using artificial intelli- gence to 'learn' virtually any object, making LiDAR devices, in many cases, faster, more accurate and less costly than present day video analytics with network cameras," Surfaro says. Surfaro adds that another signifi- cant feature is that objects can be visu- alized and recognized while maintain- ing privacy in public spaces – a key consideration for clients who must abide by new GDPR rules. "Consid- ering the high costs to 'redact' or blur facial features and vehicle license plates in video clips – oen taking eight hours to process a single minute of surveillance video – LiDAR's costs lower even further." Radar is also being paired with advanced artificial intelligence and deep learning-enabled soware to enable precise object classification and false alarm reduction. "Today's advanced radar systems are built with onboard signal processing and ana- lytics, giving them the ability to work with video surveillance systems to determine whether a moving object is a threat or simply an animal," Zussman explains. "is recognition capability significantly reduces the risk of nui- sance alerts." 3D Mapping What would you do with ten billion points of LiDAR geolocation data in three dimensions? Build a Digital Eleva- tion Model (DEM) or a Digital Surface Model (DSM), of course. Unique to LiDAR, these 3D models are so popular that there are public imaging libraries available to users needing 3D plots of a facility's surrounding areas. "Many facility managers with sprawling campuses, as well as regional airport managers, oen find themselves in a posi- tion of compromising a security design in favor of developing a 'mini- mum-needs' plan," Sur- faro says. "From con- struction sites to mapping flood disaster areas, a LiDAR sensor can be mounted on a security guard's vehicle and come along for the ride, deliver- ing 3D imagery." LiDAR devices produce survey-grade data using systems mounted on air- cra, drones, vehicles, ves- sels, and even backpacks. Both building heights and depressions (grades) of a facility's perimeter are rendered in an illustration coloriz- ing objects according to their height. Using these 3D maps gives integrators a much better idea of where surveil- lance cameras can – or should – be mounted. e 3D map can also be used to pinpoint alarms or location of staff in need of medical attention. Target Markets Finally, because both radar and LiDAR deliver similar results, the tar- get markets for these technologies are virtually identical. Potential monitor- ing opportunities include: • Airports; • Bridges; • Commercial facilities; • Construction sites; • City traffic intersections; • Dams; • Data centers; • Government facilities; • Logistics and distribution centers; • Nuclear power plants; • Residential estates; • K-12 and higher education perimeters; • Utility substations; and • Drone protection and mitigation. ■ Commercial radar can detect targets more than half a mile away. Perimeter Security Photo: Magos

Articles in this issue

Links on this page

Archives of this issue

view archives of Security Dealer & Integrator - OCT 2018