The OCR engine of our software will detect ANY pre learned character (based on License plate design) and extract them from any captured image with very powerful and optimum algorithms.
This system has not any essential prohibitions about different type of plates. For example changing in character type (different country), plate color (with, yellow, red, etc, ...), forward or backward plate mounting, even location and size, do not have Improperinfluence on result of engine.
According to many Real Application output reviewed, rough accuracy of system‘s OCR engine is 97%, therefore obviously this system Gain thetoppositionamong other competitors. On the other hand the License plate reading is independent from vehicle’s speed and up to now used for vehicle with speeds up to 250 km/h.
Iranian License Features
We can note that according of using this engine for Persian license plate it achieve some unique ability, for example some character of plates are so similar so even some times amanwouldhardlyrecognizeit, but system Distinguished themfrom each other with high accuracy. Some of these conditions referred here:
System is not depending of environment light, based on powerful hardware and software design, so it work well in day and night and any weather conditions.
Achieving 97% accuracy in real condition is not easy and convenient!
Tough Situations ANPR
It is a fact that changing OCR accuracy from 89 to 90 % is not at all similar to changing from 95 to 96 % , and second improvement (growth) is very hardest and need complex condition consideration. So this engine for achieving high accuracy (97%) in real condition is able to find and read string of some Defectiveor damaged License plates.
Partial Overlapping Of Vehicles
Multi License Plate Vehicle
License Behind The Front Window
License Behind a Net
Dirty License Plate