The operation of License Plate Recognition VT in Axxon One isn't compatible with Auto PSIM. Uninstall Axxon One and Auto PSIM, then reinstall Axxon One.

To configure the License Plate Recognition VT detection tool, do the following:

  1. Go to the Detection Tools tab.
  2. Below the required camera, click Create… → Category: Vehicle → License Plate Recognition VT.

By default, the detection tool is enabled and set to recognize vehicle license plates.

If necessary, you can change the detection tool parameters. The list of parameters is given in the table:

ParameterValueDescription
Object features
Check in listsYesThe Check in lists parameter is disabled by default. If you want to use this detection tool to check in lists of vehicle license plates, select the Yes value (see Configuring vehicle license plate recognition based on the created lists)
No
Record objects trackingYesThe metadata of the video stream is recorded to the database by default. To disable the parameter, select the No value
No
Video streamMain streamIf the camera supports multistreaming, select the stream for which detection is needed. Selecting a low-quality video stream reduces the load on the Server
Other
EnableYesThe detection tool is enabled by default. To disable the detection tool, select the No value
No
NameLicense Plate Recognition VTEnter the detection tool name or leave the default name
Decoder modeAutoSelect a processing resource for decoding video streams. When you select a GPU, a stand-alone graphics card takes priority (when decoding with NVIDIA NVDEC chips). If there is no appropriate GPU, the decoding will use the Intel Quick Sync Video technology. Otherwise, CPU resources will be used for decoding
CPU
GPU
HuaweiNPU
TypeLicense Plate Recognition VTName of the detection tool type (non-editable field)

Advanced SDK settings

VodiCTL_VPW_DYNAMIC_ENABLEYes

By default, the license plate recognition dynamics is enabled, that is, tracking is enabled, and license plate recognition is based on a set of frames. To disable tracking and license plate recognition for each frame separately without reference to the previous frames, and the quality can vary from 0% to 100%, select the No value

No
VodiCTL_VPW_DYNAMIC_OUTPUT_
PERIOD		0.5Specify the time period in seconds for which the recognition result must be output. You can use this parameter only if the VodiCTL_VPW_DYNAMIC_WITH_DUPLICATE parameter is enabled.
The value must be in the range [0, 3600]
VodiCTL_VPW_DYNAMIC_OUTPUT_
TIMEOUT		1

Specify the minimum required time of license plate monitoring in seconds before the recognition result is output to the user. You can use this parameter only if the VodiCTL_VPW_DYNAMIC_ENABLE parameter is enabled. In this mode, the vehicle track is monitored, and the user doesn’t receive the license plate recognition result immediately, rather after the time specified in this parameter. In this case, the first recognition result will be replaced by a better one and subsequently output to the user. If you specify the 0 value for this parameter, you will receive the first recognition result. Note that after the time specified in this parameter has elapsed, the monitoring of the license plate track continues until it disappears from the frame.
The value must be in the range [0, 3600]

VodiCTL_VPW_DYNAMIC_WITH_
DUPLICATE		Yes

By default, periodic output of license plate recognition results is enabled. To disable the periodic output of results, select the No value

No
VodiCTL_VPW_IMAGE_BLUR13

Parameter used for internal requirements. We recommend specifying the 13 value.
The value must be in the range [0, 100000]

VodiCTL_VPW_LOG_SETTINGSYes

By default, the parameter is disabled. To enable logging of all recognition parameters to the log file, select the Yes value

No
VodiCTL_VPW_PLATE_FILTER_
RODROPFACTOR		0Specify the coefficient of license plates filtering by image density (second strategy). Type is unsigned. This coefficient is used during image binarization and has optimal values, which are determined by developers using their own test samples. The parameter is considered a service parameter, and its value must be set according to the recommendations of technical support specialists.
The value must be in the range [0, 100000]
VodiCTL_VPW_PLATE_FILTER_
ROFACTOR		95

Specify the coefficient of license plates filtering by the so-called image density—the ratio of white pixels to the total number of pixels (first strategy). Type is unsigned. This coefficient is used during image binarization and has optimal values, which are determined by developers using their own test samples. The parameter is considered a service parameter, and its value must be set according to the recommendations of technical support specialists.
The value must be in the range [0, 100000]

VodiCTL_VPW_PLATE_FILTER_
SYMCOUNT		0

Enable/disable the algorithm of simple license plate filtering by the minimum number of recognized characters on the plates. If the algorithm is enabled (the parameter value is greater than 0), a basic search for characters on the license plate candidate (geometry, proportions) is performed. If fewer characters are detected on the candidate than the value of this parameter, this candidate isn’t considered a license plate. That is, the value of this parameter is the minimum number of characters that must be present on a license plate candidate when the basic algorithm is running. 
The value must be in the range [0, 100000]

VodiCTL_VPW_PLATE_STAR_MAX0

Specify the maximum number of unrecognized characters on the license plate, at which the result is still considered the result of license plate recognition.
The value must be in the range [0, 100000]

VodiCTL_VPW_TEXT_WITH_PUNCTUATION (Advanced) (starting with Detector Pack 3.12)

Yes

Only for Germany and Taiwan.

The parameter divides the characters on the license plate by hyphens into groups that determine the city/area and a group of the unique part of the license plate. By default, recognized license plate characters are displayed without hyphen. If you want to display recognized license plate characters with a hyphen, select the Yes value

No

Advanced settings

Advanced configuration of the detection tool must be performed with the assistance of AxxonSoft technical specialists.

Advanced image analysis

Yes

By default, the parameter is disabled. Under normal conditions, this parameter doesn’t affect the detection quality. Under poor conditions (for example, if the camera characteristics/settings don’t fully meet the requirements, or detection occurs in bad weather), to improve the recognition quality, select the Yes value. This will increase the frame processing time by 20-30% depending on the frame size

No
Algorithm of recognition of distorted LP imageYesBy default, the algorithm of recognition of distorted license plate image is disabled. If you want to recognize license plates that are at a sharp angle to the camera, select the Yes value
No
Contrast threshold40

Specify the image contrast threshold. We recommend increasing the value up to 50-60 on a good quality image, on an image with poor contrast, the value must be reduced.
The value must be in the range [0, 100]

Event timeout0

By default, the license plate recognition event is generated after the license plate track disappears from the frame. It is possible to generate a recognition event either after expiration of the time specified in seconds in the Event timeout field, or after reaching the similarity percentage specified in the LP display quality field. If the value of both parameters is different from 0, the recognition event will be generated when the first of the two conditions is met.
The value must be in the range [0, 3600]

Frame size change1920

The analyzed frames are scaled down to a specified resolution (1920 pixels on the longer side) by default. The following algorithm is used:

  • If the longer side of the source image exceeds the value specified in the Frame size change field, it is divided by two.
  • If the resulting resolution falls below the specified value, the algorithm stops and this resolution will be used further.
  • If the resulting resolution still exceeds the specified value, it is divided by two until it is less than the specified resolution.

For example, the source image resolution is 2048*1536, and the specified value is set to 1000. In this case, the source resolution will be halved two times (512*384), as after the first division, the number of pixels on the longer side exceeds the specified value (1024 > 1000). If detection is performed on a higher resolution stream and detection errors occur, we recommend reducing the compression.


LP advanced search algorithmYes

By default, the parameter is disabled. If license plates of different sizes will be present in the frame, select the Yes value.
This parameter takes into account the specified maximum and minimum width of a license plate and, in some cases, can help to improve the recognition quality and optimize the processor load

No
LP display quality0

The value of the LP display quality parameter must not be less than the value of the Minimum similarity parameter.

By default, the license plate recognition event is generated after the license plate track disappears from the frame. It is possible to generate a recognition event either after expiration of the time specified in seconds in the Event timeout field, or after reaching the similarity percentage specified in the LP display quality field. If the value of both parameters is different from 0, the recognition event will be generated when the first of the two conditions is met.
The value must be in the range [0, 100]

Maximum number of threads

1

Specify the maximum number of recognition threads. If the value is 0, the recognition process takes place in the same thread that starts it.
The value must be in the range [0, 10].

  • If you specify several countries, we recommend setting the value of the parameter to at least 4.
  • Increasing the number of recognition threads reduces possible drops in the frame rate, but at the same time increases the load on the device.
  • The total value of this parameter for all license plate recognition detection tools must not exceed:
    • the number of recognition threads in the license (you can check it with the lsvpwc utility, see the manufacturer's documentation);
    • number of logical processors;
    • 100.


Minimum similarity40

Specify the minimum percentage of similarity between the recognition result and the corresponding license plate template, at which this result can be considered a license plate recognition result. This parameter is used to filter the obtained results by validity.
The value must be in the range [0, 100]

Tracker timeout3

Specify the time in seconds after which the vehicle track is considered lost. The value must be in the range [0, 3600].

  • You can use this parameter to avoid double triggerings that can occurs, for example, when a license plate is recognized when it appears in the frame, then it is overlapped by a visual obstacle and then recognized by the detection tool again.
  • If you specify the timeout longer than the time of possible overlapping of the license plate in the frame, the detection tool triggers only once.


Basic settings
Algorithm for detecting vehicle direction

By LP coordinatesThe By LP coordinates algorithm is used by default. It works as follows: if the coordinates of the vehicle license plate shift from top to bottom (the vehicle moves towards the camera) and if the coordinates of the vehicle license plate shift from bottom to top (the vehicle moves away from the camera).
Select the By LP scale change parameter if in the Surveillance window, the scale of the vehicle license plate increases (the vehicle moves towards the camera) and if the scale of the vehicle license plate decreases (the vehicle moves away from the camera)
By LP scale change
Analysis mode

Standard (Morph)Select the analysis mode of the detection tool: standard or neural network mode
Advanced (DNN)
License typeSearch in archive

Standard license for searching recognized license plates in the archive.

This type of license provides a 30-second delay between the license plate recognition and the corresponding event (see Vehicle number plate recognition and search).


Normal (25 FPS or 6 FPS)The 25 FPS license has a priority. If you don't have the 25 FPS license, the 6 FPS license is used. You cannot use the license plate recognition function if you haven't purchased both licenses
Fast (25 FPS)

The Fast license allows you to process video streams at up to 25 FPS. License plate recognition for vehicles moving at speeds up to 150 km/h. You cannot use the license plate recognition function if you haven't purchased the license

Slow (6 FPS)

The Slow license allows you to process video streams at up to 6 FPS. License plate recognition for vehicles moving at speeds up to 20 km/h. You cannot use the license plate recognition function if you haven't purchased the license


All license types except the standard one (Search in archive) require the hardware key or the software key activation (see Licensing of the software module for License plate recognition (VT)).
You can use a 60-day trial key (see Licensing of the software module for License plate recognition (VT)).
You can use only a hardware key on virtual machines. To use a hardware key, select the Normal (25 FPS or 6 FPS) license type.


Maximum LP width20

Specify the maximum and minimum width of the license plate as a percentage according to the actual size of the license plate in the frame. The value must be in the range [0, 100].

  • Specified vlaues must exclude omissions in license plate recognition, that is, the values must not be smaller than the real size of the license plate in the frame. Otherwise, the license plates won't be recognized.
  • The Minimum LP width parameter affects the CPU load: the higher the value, the lower the load.
  • Reducing the difference between the maximum and minimum width reduces the CPU load.


Minimum LP width5

Minimum number of detection triggers

6Specify the minimum number of triggers to display the object track. The higher the value of this parameter, the longer the time from object detection to displaying its track. A low value of this parameter can result in false alarms.
The value must be in the range [1, 20]
ModeCPUSelect the processor for the operation of the detection tool algorithm (see Hardware requirements for License Plate Recognition VT, Selecting Nvidia GPU when configuring detection tools). 
In the standard mode, the algorithm can only run on CPU.
In the neural network analysis mode, the algorithm can run on any of the supported processors
Intel GPU
Intel NCS (not supported)
Intel HDDL (not supported)
Nvidia GPU 0
Nvidia GPU 1
Nvidia GPU 2
Nvidia GPU 3
NvidiaDLA
Country selection
Country 1 name

Albania, Algeria, Andorra, Argenting, Armenia, Australia, Austria, Azerbaijan, Bahrain, Brunei, Belarus, Belgium, Bolivia, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Canada, Chile, Colombia, Croatia, Czech Republic, Denmark, Ecuador, El Salvador, Estonia, Finland, France, Georgia, Germany, Greece, Guatemala, Honduras, Hong Kong, Hungary, India, Indonesia, Iran, Israel, Italy, Jordan, Korea, Kazakhstan, Kenya, Kyrgyzstan, Latvia, Liechtenstein, Sri Lanka, Lithuania, Luxembourg, Malaysia, Malta, Mexico, Moldova, Monaco, Mongolia, Montenegro, Nicaragua, Nigeria, Norway, Panama, Paraguay, Peru, Philippines, Poland, Portugal, Romania, Russian Federation, Serbia, Singapore, Slovakia, Slovenia, South Africa, Spain, Sweden, Switzerland, Taiwan, Tajikistan, Tunisia, Turkey, Turkmenistan, USSR, Ukraine, United Kingdom of Great Britain and Northen Ireland, Uruguay, USA, Uzbekistan, Viet Nam, Zambia

In the corresponding fields, select from the list one or more countries whose license plates you want to recognize.

  • The more countries you select, the slower the recognition works and the higher the probability of license plate recognition error.
  • To recognize old USSR license plates, you must additionally select the USSR country in the list.


Country 2 name
Country 3 name
Country 4 name
Country 5 name

In the window, specify the area in which license plate detection and recognition will be performed. You can adjust the area by moving the anchor points . 

  • For convenience of configuration, you can "freeze" the frame. Click the button. To cancel the action, click this button again.
  • The detection area is displayed by default. To hide it, click the button. To cancel the action, click this button again.

To save the parameters of the detection tool, click the Apply  button. To cancel the changes, click the Cancel  button.