Next generation video surveillance
for governments and enterprises
Looking Glass is a comprehensive solution for all your security needs. It enables instant search across the big data of CCTV videos and real time crime detection, using the most advanced machine learning algorithm and our state-of-the-art Neural Network technology. Looking Glass is a collaboration environment for unlimited users from different organizations, delivering the next generation video surveillance service to governments and enterprises with millions of users and millions of cameras.
All-in-One Analytics Platform Combining
Private Cloud, AI and Big Data Technologies
Looking Glass enables a truly global search of video content objects through thousands of locations. Unlike conventional video management systems (VMS), Looking Glass is based on next generation technologies, widely adopted today by IT giants such as Google and Amazon
Interactive Maps and Plans
Allow the user to organize, analyze and manage all cameras and sensors across different locations
Video and Audio Analytics with Infinite Scalability
Unlimited global search engine across thousands of locations from any device, including unique filters, can identify humans, vehicles and events.
Over 20 different modules allow fast and accurate access and alerts for many event types and intelligence or evidence gathering.
Unlimited number of cameras across unlimited locations cover over 1M events per minute and allow average event response time of around 20 sec, based on average event occurrence of every 65 sec.
Geo-information System (GIS) interactive maps, with the option to organize, analyze and manage all cameras and security sensors across different locations and floor plans.
Real Time AI based Alert Mechanism includes: Face-Track and License Plate track, Side-Track, Top-Track, Smoke-Track, Rail-Track and Audio-Alarm.
Visual tools for video content allow to search, find or alert in an object and/or vehicle with specific behavior, by applying visual tool to a pre-defined frame, such as distance, shape, angel, cross line and more.
Deep Learning methodology optimizes Video and Audio Content Search in the most intuitive way, minimizing long hours of video review.
Web-based fully featured UI Linux, Windows, Android, iOS apps. Allows easy and intuitive search and management of up to 50 different locations (around 500 cameras) per analyst.
Real Time High Performance Response
Looking Glass Video alarm system provides security with equipment able to identify issues in areas with various environmental challenges. Using Looking Glass platform allows one person to monitor 50 different locations (around 500 cameras), providing average event response time of around 20 sec, based on average event occurrence of every 65 sec. Looking Glass is camera agnostic and integrates well with any existing IP cameras or analogue cameras via h.264 converters. In addition, Looking Glass is capable of integrating on top of existing VMS, using second streams of IP cameras.
Looking Glass monitors both indoor & outdoor cameras in different light conditions and various camera network connections providing enhanced level of security and intelligence with a simple methodology and a brilliant user interface.
Artificial Intelligence
Person identification using Face recognition video analytics module, Provides face detection, tracking, biometric features extraction, best angle or view detection with 99% accuracy.
Vehicle and license plate recognition modules of up to 30 km/h and up to 200 km/h including vehicle feature detection with AI (vehicle color and type)
Early smoke detection video analytics module. Works for cameras in the visible band and the infrared band. The module is intended for indoor use only
Object Recognition, Object Detection and tracking using 3D spatial calibration recognize motion, object stopping, object loitering, object abandoned or removed. Filter rain, snow, shadow, insects and birds.
Audio Recognition module Generates alerts upon noise, shouting, cry, glass breaking and gunshot.
Tamper Alarm and Intelligent video motion detection module based on multiple scale dynamic textures modeling. Recognize if image is too light, too dark, out of focus. Detect blocking artifacts and displaces camera.
Technology
What is Deep Learning
Deep learning (also known as deep structured learning or hierarchical learning) is part of a broader family of machine learning methods based on learning data representations, as opposed to task-specific algorithms. It works better with more data, bigger models and more computation.
Deep Learning methods aim to learn feature hierarchies and automatically learning features at multiple levels of abstraction. It allow a system to learn complex functions mapping the input to the output directly from data, without depending completely on human-crafted features.
Looking Glass is using Deep Learning methodology in order to optimize Video and Audio Content Search in the most intuitive way, minimizing long hours of video review.