General

How It Works

The tracking and recognition mode decides which networks the module runs. Each network is trained for one task.

Track vehicle

Track vehicle: video goes to vehicle detection with CNN-1, then a tracker. Vehicle boxes go up to feature detection with CNN-2. The tracker sends vehicle boxes to comparison, features enter from above, and rules enter from below.

CNN-1 finds vehicles in the incoming video and sends their boxes on.
CNN-2 identifies features of the detected vehicles, such as color.

Recognize plate

Recognize plate: video goes to plate detection with CNN-3. Plate boxes go to CNN-4, which reads the numbers and sends them to the tracker. The tracker sends boxes to comparison. The vehicle database sends numbers, and rules enter comparison, which creates events.

CNN-3 finds license plates in the incoming vehicle video and sends them to CNN-4 to read the numbers.
CNN-4 reads the numbers and sends them on for analysis.

Track vehicle and recognize plate

Track vehicle and recognize plate: CNN-1 sends vehicle boxes to the tracker and to CNN-2. CNN-3 sends plate boxes to the tracker and to CNN-4. CNN-4 sends numbers to the tracker. CNN-2 sends features to comparison. The vehicle database sends numbers, and rules enter comparison, which creates events.

CNN-1 finds vehicles in the incoming video and sends their boxes on.
CNN-3 finds license plates in the incoming vehicle video and sends them to CNN-4 to read the numbers.
CNN-2 identifies features of the detected vehicles, such as color.
CNN-4 reads the numbers and sends them on for analysis.

Track vehicles with a recognized plate

Track vehicles with a recognized plate: CNN-5 sends boxes to the tracker and plate boxes to CNN-4. CNN-4 sends numbers to the tracker. The tracker sends vehicle boxes to CNN-2 and boxes on to comparison. CNN-2 sends features to comparison. The vehicle database sends numbers, and rules enter comparison, which creates events.

CNN-5 finds vehicles and their license plates in the incoming video. The vehicle box and its plate box go on together as a pair.
CNN-2 identifies features of the detected vehicles, such as color.
CNN-4 reads the numbers and sends them on for analysis.

The last step compares the numbers that were read with the numbers in the vehicle database, checks other configured conditions, or does both. A vehicle in a restricted-entry zone is one such condition. From that result, the module decides whether to create an event. Vehicle type and color are determined when the event is created and are included in it immediately. The vehicle brand can be added later.

The selected mode also determines which settings are available:

Track vehicle	Recognize plate	Track vehicle and recognize plate	Track vehicles with a recognized plate
Region entrance Vehicle type and color recognition Recognize the brand of the vehicle Save tracks Object focal point Line crossing Motion in region Speed control Illegal parking and waiting	Region entrance No match Invalid License plate Region Threshold Minimum plate size Re-recognition Save tracks Line crossing Motion in region	Region entrance No match License plate is missing Invalid License plate Region Threshold Minimum plate size Re-recognition Vehicle type and color recognition Recognize the brand of the vehicle Save tracks Object focal point Line crossing Motion in region Speed control Illegal parking and waiting Traffic Objects	Region entrance No match License plate is missing Invalid License plate Region Threshold Minimum plate size Re-recognition Vehicle type and color recognition Recognize the brand of the vehicle Save tracks Object focal point Line crossing Motion in region Speed control Illegal parking and waiting Traffic Objects

With Vehicle type and color recognition on, Track vehicle, Track vehicle and recognize plate, and Track vehicles with a recognized plate recognize these vehicle types:

Type	Description
Motorcycle	Any powered two-wheeled or three-wheeled vehicle: a motorcycle, tricycle, scooter, or moped.
Car	A sedan, hatchback, station wagon, coupe, SUV, or crossover, with up to 8 seats.
Light-duty truck	A vehicle with a payload of up to 3.5 tons.
Medium-duty truck	A vehicle with a payload from 3.5 to 12 tons.
Heavy-duty truck	A vehicle with a payload over 12 tons.
Tractor-trailer	A tractor with a semitrailer, or a truck hitched to one or more trailers.
Minivan	Passenger vans and minibuses for up to 8 passengers, including minivans and MPVs (multi-purpose vehicles).
Bus	A vehicle for more than 8 passengers, including city buses, intercity buses, double-decker buses, and trolleybuses.
Machinery	Wheeled and tracked tractors for farm or municipal work, harvesters, and construction and road-building equipment: excavators, graders, pavers, cranes, and loaders.
Trailer	Any trailer or semitrailer standing apart from the tractor, for example in a parking lot, or a passenger-car trailer.

Computational Complexity

The algorithm is rated 5 on a 5-point complexity scale. Average CPU usage is about 100%, and the load can vary with video resolution and the processor.

Camera Requirements

To select cameras, use specialized software (for example, IP Video System Design Tool) or a free lens calculator from the camera manufacturer.

Characteristic	Requirement
CMOS sensor size	1/2.8″ or larger
CMOS sensor light sensitivity	0.005 lux or less for a color image with automatic gain control (AGC) on at 30 IRE; 0.0005 lux or less for a black-and-white image with AGC on at 30 IRE
Electronic shutter speed	adjustable from 1/25 to 1/10,000 second, or a wider range
Electronic shutter speed adjustment modes	automatic / adjustment range / manual
Number of video streams	2 or more
Primary stream resolution	1280×720 to 2560×1920 pixels
Secondary stream resolution (optional, for layouts)	320×240 to 704×576 pixels
Video frame rate	15 to 25 frames per second
Lens type	varifocal
Focus adjustment modes	automatic / semi-automatic / manual
Lens focal length	4 mm or longer
Lens speed	f/1.6 or faster
Lens aperture	adjustable
Aperture adjustment modes	automatic / manual
Primary stream bitrate	up to 8 Mbps
Secondary stream bitrate	up to 1 Mbps
Video encoding type	constant (CBR) / variable (VBR)
Video encoding protocol	H.264 HP/MP/BP, H.265
Network protocol support	TCP/IP, ICMP, HTTP, HTTPS, RTP, RTSP, NTP, SNMP, IGMP, IPv4/IPv6
Video transmission protocol	RTP over TCP
ONVIF compatibility (optional, for remote camera control)	ONVIF (PROFILE S, PROFILE G)
Adjustable image parameters	saturation, brightness, contrast, sharpness, AGC, and switching between color and black-and-white
Image parameter adjustment modes	automatic / manual / scheduled
AGC adjustment modes	automatic / manual (the gain can be limited)
Image enhancement	digital noise reduction (DNR) and backlight compensation (BLC/HLC)
Noise reduction adjustment	automatic / manual (noise reduction levels can be adjusted)
Backlight compensation adjustment	automatic / manual (a target zone for backlight compensation can be set)
Built-in illumination	50 meters or more
Illumination modes	automatic / intelligent / manual (range and intensity can be adjusted)

Camera Mount Requirements

Characteristic	Requirement
Camera tilt relative to the horizon	0° to +20°
Camera angle relative to the road direction	−20° to +20°
Camera roll about its axis	−5° to +5°
Image quality at the border of the controlled zone	250 pixels per meter or more
Recommended mounting height in parking lots and parking garages	0.5 to 2 meters above ground level
Recommended mounting height on roads, highways, and intersections	3 to 8 meters above ground level
Illumination of the controlled zone	50 lux or more across the camera scene. The vehicle in view must be lit evenly.