The License plate recognition analytics. The camera placement requirements

The License plate recognition analytics is a technology effectively used across various spheres. It has a wide range of applications, for example, controlling access to a territory, managing parking lots, optimizing traffic, improving security service operations, and much more.

For this type of analytics to be performed, an IP videocamera with relevant specifications should be properly placed and set up, taking into account the location and operating conditions in ways to receive license plate images of the required quality.

In this guide, you will find the requirements for the main specifications of the IP videocamera, the placement of the camera, and the image requirements to ensure accurate performance of the License plate recognition analytics type.

Videocamera specifications

When selecting an IP videocamera, you should take into consideration the requirements for captured images, the camera specifications, the installation requirements, the control zone peculiarities, and the location and operating conditions.

The key camera settings (maximum exposure, maximum gain, wide dynamic range (WDR) technology, color correction, contrast, and the depth of field) depend on the selected alaytics type and should be set for each camera individually.

For the License plate recognition analytics, an IP videocamera should have the following specifications:

Specifications

Measurement

Matrix

It is a key component of a camera that defines image quality. The value of this component reflects the size of the matrix diagonal in inches.

1/3 inches and larger.
With a high degree of sensitivity to light.

Lens-aperture ratio (light sensitivity)

Plays a crucial role in overall image quality. Lens aperture is the maximum light capacity of a lens, which determines how much light can pass through to the camera sensor. The F-number means the maximum size of the aperture. The smaller the F-number, the more light the lens can gather.

Recommended F1.4 and better.
It should ensure contrast and image clarity during the dark time.

Lens aperture

With this camera component, you can set the amount of light that goes through the lens and control the focus depth. Located within the lens itself, the aperture regulates light intake by altering the size of the opening.

P-Iris technology, integrated into the lens aperture, enhances the image depth of field. The greater the depth of field, the larger part of the frame will be clear. This ensures the license plate image is sharp and clearly visible in cases where the camera vision control area covers a wide range of illumination and distances.
If the camera vision control area is constantly illuminated, it is possible to use a videocamera with a manual or fixed setting of the lens aperture.

Resolution

Camera resolution refers to the measure of the number of pixels that results in sharpness and more detailed images. The unit of measurement for resolution is the pixel.

To ensure license plate recognition, high-resolution cameras aren’t required. Redundancy of resolution increases the server load without significant recognition improvement and may lead to negative results because matrices with higher results have a lower lens aperture ratio.

To calculate the required camera resolution, use the following formula, where:

the driveway width (meters) / the width of the license plate (meters) * the recommended size of the license plate image for recognition (width in pixels) – 300 pixels

An example of the calculations of the camera resolution required under specific conditions:

The driveway width is 2 meters.
The license plate width is 520 mm (0.52 meters).
The recommended size of the license plate image for recognition (width in pixels) is 300 pixels.

(2 m / 0,52 m) * 300 pixels = 1154 pixels

As a result, the required IP camera HD resolution in this case should be 1280х720 pixels.

Focal length

It is one of the key specifications that determines the camera field of view (vision angle), measured in millimeters.

The focal length of the camera lens should correspond to the distance between the camera mounting placement and the anticipated area for license plate recognition, so that the recommended license plate size image in the frame is 150–300 pixels.

To calculate the required focal length, you may use the Lens Calculator.

Shutter speed (exposure time)

It is the time during which the camera shutter is open and light goes through to the matrix. Additionally, the sutter speed influences the sharpness of the final image. The sutter speed unit of measurement is seconds, or second fractions. A larger denominator means a shorter exposure time. For example, a shutter speed of 1/1000 seconds is faster than 1/100 seconds.

The recommended shutter speed is 1/500 seconds.
The shutter speed can be set to 1/250 seconds in cases of imposed speed limits with archificail elevations (for example, speed bumps), architectural forms (for example, parking bollards, concrete semi-sphere bollards), traffic signs (for example, the Stop sign), road directional markings (for example, the Stop line), and proper lighting conditions.
For a license plate image to be sharp and clear, the videocamera specification should include a fixed shutter speed setting.

Lens with infrared correction

This type of lens is used in video surveillance cameras that support day and night modes.

When an infrared camera spotlight is used, select a camera lens with infrared correction (compensation for IR light). This type of lens is usually marked with IR.

Lens Calculator calculations example

To define the required camera specifications, in addition to the list mentioned above, you may use the Lens Calculator. On the opened service page, specify the matrix size, the focal length, the distance from the camera placement to the object of recognition, the mounting height, and so on.

As a result, you’ll receive the relevant camera specifications required for license plate recognition in the specified operating conditions, as well as the geometric placement of the camera (the mounting height and distance from the object of recognition).

image 18.png — Lens Calulator calculations

In the image below, you can find the calculation results from Lens Calculator. The target resolution of the license plate image is 381.4 pixels. This complies with the image requirements of 150–300 pixels in width.

General requirements for camera images

Effective and accurate performance of the License place recognition analytics depends on the quality of images received from an IP camera. For accurate license plate recognition, the image received from the camera should comply with the requirements listed below.

The recommended size of a license plate image in the frame is 150–300 pixels in width.

Рекомендуемая ширина (1).png — An example of a license plate image recommended size

image 18 (2).png — A license plate image with insufficient pixels

The license plate image should have minimal geometric distortions. To ensure high image quality with minimal geometric distortions, place the camera according to the operating conditions and the requirements listed in the Camera installation angle section.

image 18 (3).png — Geometric distortions of a license plate

The orientation of a license plate in the frame should obey the “one line” rule, meaning an imaginary horizontal line must intersect the first and last character of the car license plate.

The license plate image shouldn’t be overlit by external lights. Overexposure can be caused by the headlights of a car, for example, or the built-in infrared illumination of a camera.

image 22.png — An example of an overlit image

image 22 (1).png — Overlit of the license plate by headlights

The license plate should be sharp and clearly visible.

Group 5.png — An example of an unclear image

The license plate should be clean.

image 6 (1).png — An example of a dirty license plate

The requirements for camera placement and engineering infrastructure

Distance

A video camera should be placed in a way that it captures license plates at a distance corresponding to the anticipated speed of cars, meaning higher speeds require a longer distance for image capture. Otherwise, the system won’t be able to recognize a license plate before a car leaves the camera’s field of view. On top of that, the camera depth of field and the range of the infrared illumination should be taken into consideration.

Mounting height

An IP video camera shouldn’t be pointed directly at the light source (for example, the sun, lighting lamps, and so on). To exclude headlight interference during the night, we recommend mounting IP video cameras above the level of the headlights, taking into consideration the types of vehicles. For example, the headlights of heavy-duty vehicles are placed higher than those of light motor cars.

Complying with this rule ensures the captured image of the license plate isn’t overlit with external light sources.

Camera mounting angles

To comply with image requirements, IP cameras should be mounted at optimal angles.

Minimize the angle between the camera and the car’s direction movement so that the camera’s view axis falls perpendicularly on the license plate.

Vertical mounting angles of a camera:

Recommended: up to 18-20°
Maximum allowed: 30°

Horizontal mounting angles of a camera:

Recommended: up to 5-10°
Maximum allowed: 20°

The acceptable angular inclination (tilt) of a license plate is no more than 5° from the horizontal plane.

Should the requirements for IP video camera placement be disregarded, the license plate image will be geometrically distorted. This may result in the failure of license plate recognition due to incomplience with the captured image requirements:

The license plate image should have minimal geometric distortions.
The orientation of a license plate in the frame should adhere to the “one line” rule.

Optimal and maximum allowed mounting angles

5 horizontal plane (2).png — Acceptable angular inclination of a license plate

Illumination

The illumination of the license plate conrole zone should be as follows:

50 lux and more. For more details on illumination ranges, see the Common range of objects illumination page.
20 lux and more if a camera with a large matrix, a relevant lens aperture ratio, and a P-Iris lens is used.

Artificial illumination of the control zone is required in the following cases:

At night.
During the day, if natural light illumination is absent, for example, in underground parking.

Light and infrared projectors can help illuminate the camera field of view. The infrared light is invisible to human eyes and doesn’t blind drivers.

If you use an external light, place it near the video camera because license plates reflect light back to the source. Placement of the light source near the camera will ensure better vision of license plates.

Special cameras have built-in infrared illumination.

Adjustments of vehicles speed and directions

If adjustment of the vehicle's speed and direction is required, the following methods can be applied:

Artificial elevations, for example, speed bumps.
Architectural forms, for example, parking bollards, concrete semi-sphere bollards.
Traffic signs, for example, the Stop sign.
Road directional markings, for example, the Stop line.

image 6 (5).png — Vehicle speed adjustment with a speed bump

Visibility of a license plate

The video camera should be placed in such a way that it can capture the license plates of vehicles following one after another, so that the front car doesn’t hide the license plate of the vehicle behind.

Group 6 (1).png — An example of a partially hidden license plate

Separate control of entrance and exit passages

We recommend placing separate cameras to control entrance and exit passages. If one camera is used for both directions, the vehicles moving in different directions can overlap each other.

Moreover, if the roadway width is enough, we recommend separating the lanes with the road directional markings and architectural forms (for example, parking bollards or concrete semi-sphere bollards).

Camera settings

The recommended exposure depends on the camera orientation and vehicles speed.
Limit the maximum exposure of the camera to ensure the captured image is clear.
Limit the maximum gain of the camera to avoid overlit of a captured image of the license plate.
The camera view should be free from high-contrast objects (billboards, trees, grill fences). If not possible, use a camera setting to mask an object.