The ULSee infrared Monitor Solution combines the strength of ULSee’s core competencies in computer vision and AI based algorithms with the rising demand for thermal detection products.
The result is a sophisticated thermal detection system with facial / object recognition capabilities and a local / cloud backend. The local / cloud backend provides monitoring and event processing functionalities, even in the most complex scenarios involving multiple camera locations. All pre-programmed events (e.g. individual’s temperature is out of range, a person is not wearing a mask or a face is not recognised) are logged in the system showing the location, time, temperature and image of the individual who triggered the event.
THERMAL DETECTION CAMERA
The ULTI-200 thermal detection camera is a high resolution, high accuracy module, designed for multiple person detection. The IR resolution is 384 x 288 and the temperature readings, within the designated range, are accurate within 0.5 degrees Celsius. The design for multiple person detection is achieved through expanded field of view (FOV) and detection distance. The extended distance range allows for accurate readings between 0.3-4m These technical features enable the system to be less intrusive and disruptive, and to perform well in realistic “real-life” situations.
The RGB camera module is a crucial component of the system, adding computer vision capabilities and providing raw data for facial and object recognition. The software underpinning this technology is part of ULSee’s core competency and has been continuously developed and improved since the formation of the company five years ago. Our technology has been designed to easily interface with other systems on site, through APIs. Facial recognition can provide keyless, touch-less door entry points or enhance safety and security systems. The object recognition capabilities are extremely helpful in preventing false alarms from the thermal detection system (e.g. a person carrying a hot drink in front of the thermal camera).
The A.I. box is a compact and cost-efficient edge computing unit. In complex systems, multiple boxes can be grouped through the local or cloud backend. Both the thermal detection camera and RGB camera are attached to the A.I. box. The A.I. box will then handle the necessary and complex calculations needed for all the technical algorithms which enable the essential features of the overall solution. These features include merging the feed from the thermal detection camera and RGB camera, running facial recognition, running object recognition and allowing for the logging of data which gets passed to the local / cloud backend.
A.I. LOCAL / CLOUD BACKEND
The AI Local/Cloud Back end provides a database (which can be stored locally, within the user’s domain, or remotely) and a platform for user interaction and data analysis. Information is stored in a database which is specific for each customer, for increased security and privacy. The database logs individual events and stores associated image and date / time stamp for historic analysis. The user interface provides easy management and control of multiple locations and camera feeds.