For those who want strategic positioning and differentiation from the rest of the market, there is a solution that combines high-tech hardware with advanced artificial intelligence. So let me introduce you to Her Majesty, Targa 960, which through its powerful deep learning neural networks offers extraordinary vehicle recognition capabilities with uncompromising accuracy.
Multi-video analysis
This camera represents the top of the line of Selea products. It integrates as many as 14 AI algorithms on board the camera that work simultaneously to provide, on dual lanes, as many as 13 different types of information. The camera is trained to recognize, with high accuracy, more than 400 makes, more than 9000 models, 16 shades of color, 35 types of “subjects” (cars, trucks, people, motorcycles, bicycles, ambulances, etc…), more than 50 nationalities, instant speed estimation, report vehicle in wrong way, stationary vehicle, smoke and fire, helmet (or no-helmet) presence detection on motorcycle/moped drivers.
See camera performance > (click here)
OCR sensor
The camera is equipped with a 5-Mpx optical sensor with a 12-40 mm motorized lens for license plate character recognition: a useful function in the area of reporting uninsured and unreviewed vehicles, stolen vehicles, or vehicles under administrative detention, which are useful in investigations for effective offense prevention and suppression.
Context sensor
In addition to the license plate sensor, the camera is equipped with a highly sensitive 8Mpx (4K) Sony Ultra Night Vision color sensor, which can capture clear color images of moving vehicles, without the ghost effect on moving vehicles.
Subject Privacy
Whether you want to comply with the GDPR on privacy or you need to show the images to third parties, the “Subject Privacy” function allows you to blur all those subjects, still or moving, that have nothing to do with the vehicle that is the object of the violation or recognition. The peculiarity of the Selea algorithm is that it blurs only the sensitive elements (people, motorcycles, bicycles and moving or stationary vehicles) leaving the rest of the context image unchanged.
Accuracy
Selea not only states the types of algorithms integrated but also states their recognition accuracy, which is:
Of 99.8% on the reading of license plates in all weather conditions, night and day.
Of 98% on reading ADR (Kemler-ONU) tables of vehicles carrying dangerous goods.
Of 95% on recognition of estimated speed, direction of travel, make, model, color, vehicle type, and helmet presence.
Fog-Fighter
The camera is equipped with a powerful IR illuminator that enables license plate recognition even at night in dense haze.
Memory
As standard, the camera comes with 16GB micro SD SLC (Hight Endurance from -40° to + 85°) memory expandable up to 1TB
UNI 107772 Certification
UNI10772:2016 Class A accredited camera with 100% reading accuracy on rear, two-line, motorcycle, and moped license plates.
Privacy by Design Certification
ISDP 10003 certified Camera and designed according to privacy by design and privacy by default criteria to guarantee and comply with security and privacy regulations.
NDAA Compliance
100% NDAA Compliant Camera for Cyber Security
