.png)
In recent years CCTV software has meant Video Analytics: crunching the visual data streamed from network cameras and performing real-time event detection and post-event analysis. However, when it comes of video analytics, we should consider the following main distinction:
Standard Video Analytics
Advance Video Analytics
DDA Analytics
Face Recognition
Standard Video Analytics
Video Motion Detection
The so-called “Video Motion Detection” (VMD), is a “standard”video analytic function based on the simple analysis of pixel variation. The cameras supporting VMD can detect changes that occur in their field of view, and generate alarms regardless to the object type. Provision-ISR standard video analytics functions
Advanced Video Analytics
Artificial Intelligence
The A.I. is based on a series of algorithms, or mathematical procedures, which work like a series of questions to compare the object seen with hundreds of thousands of stored reference images of objects in different angles, positions and movements. The A.I. asks itself if the observed object moves like the reference images, whether it is approximately the same size height relative to width, if it has the characteristic two arms and two legs, if it moves with similar speed, and if it is vertical instead of horizontal. Many other questions are possible, such as the degree to which the object is reflective, the degree to which it is steady or vibrating, and the smoothness with which it moves. Combining all of the values from the various questions, an overall ranking is derived which gives the A.I. the probability that the object is or is not a human, a vehicle or an object.
Advanced Analytics Functions
DDA Analytics
Based on smart objects recognition technology, DDA VIDEO ANALYTICS allows the system to distinguish between human beings 2-wheeled vehicles and 4-wheeled vehicles.
The system triggers alarms or sends push notifications only when the desired object is detected and by doing so reduces annoying false alarms caused by shadows, light changes, shaking trees, or animals.
Face Detection
The camera detects a face in the scene and sends the face snapshot to the NVR which analyses the face and compares it to the ones included in the existing database in order to recognize the person. Provision-ISR second generation of IP cameras feature great performances. An Eye Sight (V2) camera featuring face detection is able to detect up to 10 faces in 60 milliseconds while a Smart-Face camera can detect up to 30 faces.
IP camera supporting FACE DETECTION:
I4-340IPE-36
DI-380IPE-MVF
DAI-380IPE-MVF
Face Recognition
Until now, Face Detection was performed by the Camera and Face Recognition was performed by the NVR. Provision-ISR Smart-Face series includes All-in-One cameras performing both Face Detection and Face Recognition, taking the load off the NVR and acting as an independent gate operator or attendance clock.
IP camera supporting FACE RECOGNITION:
IP BX-321FR
DW-320FR-MVF2
TW-320FR-MVF2
THE END OF FALSE ALARMS
Unlike "Motion Detection“ (which is based on Pixel changes), Provision-ISR DDA™ VIDEO ANALYTICS is able to distinguish the type of “object" and to activate an alarm only in the case of a truly "suspicious" detection.
This ability significantly reduces all the false alarms caused by rain, shadows, light changes, shacking trees, animals, etc.
DDA - Line Crossing
The user draws a line in the scene and sets both the crossing direction and crossing permissions.
Example:
If vehicles are not allowed to cross the line (but humans are) the system will generate an alarm only to the passage of a vehicle. No Alarm will be generated if a person crosses the line.
DDA – Sterile Area
The user draws an area in the scene and sets access permissions.
Example:
If vehicles are not allowed to access the area (but people are) the system will generate an alarm only when a vehicle enter the limited area.
In addition, the user can configure if entering/exiting the area will trigger an alarm.
DDA – Object Counting
The user positions the camera at a gate entrance/ exit to get a real-time entrance and exit monitoring.
Example:
If the observed vehicle counting area is set at the entrance of a parking the system will allow to
monitor when the parking reach the maximum number of vehicles allowed.
The Face Recognition is part of the Intelligent Video Analytics functions developed by Provision-ISR.
If the “face detection” is managed by the camera; the “face recognition” is managed by the NVR.
Customer Traffic Counting System
L-PALOSSIE
Cost effective, accurate people counting sensor
Vehicle Count Sensor PALOS-PK
More than 99% accurate. Inbound and outbound vehicles are correctly counted. The system ensures accurate counting even when several vehicles pass in opposite directions, side by side in the wrong direction or even bumper to bumper.
People counting management
Ossia VMS combines data from multiple counting cameras placed at the entrance/exit of a premise, providing real-time data on how many people are present in that certain premise at a certain time.
This insight allows the user to monitor visitor flow and occupancy and to take measures if occupancy exceeds the set threshold.
The same feature can be used for vehicle counting as well and find its exemplary application in the parking management field.