In the world of license plate recognition, it’s easy for software vendors to make claims of 99% accuracy, blah, blah, blah.
We share in great detail below what Plate Recognizer can (and cannot) do in various camera angle, image resolution, blurriness, and other situations. We encourage you to use our Free Trial to conduct these tests in your specific environment!
approach: We measured the camera angle with a protractor situated under the vehicle. Distance from camera to vehicle was 6 feet for all images taken.
results: Plate Recognizer ALPR results were correct up to a 70 degree angle. It failed at 75 degree angle.
approach: What matters in image resolution is actually the pixel width of the license plate itself. In the original image (2880×2160), the plate is 340×140, or 12% of the original image width. We resized the images below and indicated the plate’s pixel width.
results: Plate Recognizer ALPR results were correct up to 30 pixels of the plate width. It failed at 28 pixels of plate width.
approach: We applied Paint.net‘s Motion Blur feature at a standard 25 degree angle on various blur distances as stated below.
results: Plate Recognizer ALPR results were correct up to a distance of 27. It failed at a blur distance of 30.
approach: We applied Paint.net‘s Lightness levels (under Hue/Saturation Adjustments) to the original image.
results: Plate Recognizer ALPR results were correct up to a Lightness Level of -93.
results: Plate Recognizer ALPR performed well with license plates with stacked characters. This includes stacked characters in the front, middle or end of the license plate.
results: Plate Recognizer ALPR performed well on license plates with a variety of icons–including icons that look like a letter!
results: Plate Recognizer ALPR performed well on license plates with 2 lines of text. It even correctly decoded plates with different fonts between the top and bottom lines.
results: Plate Recognizer ALPR performed well on license plates with tough characters, including script fonts, fonts of different sizes on the same plate, long license plates, plates with smudges, etc.
results: Plate Recognizer ALPR performed well on personalized or vanity license plates, even in situations where some characters are on top of an icon.
results: Plate Recognizer ALPR was able to correctly identify the license plate among the many different bumper stickers and signages on the vehicle. Note that in some cases, the font style and size of the marketing signs are very similar to that of the license plates.
results: Plate Recognizer ALPR correclty decoded reflective and non-reflective license plates.
Don’t take our word for it! We encourage you to test our ALPR in your specific environment!
approach: For some of the above cases, we took images of a single vehicle (black Lexus IS250) to ensure consistency in the analysis. In other cases, we collected images online or from some of our partners (with permission to share).
Date: Images were taken on December 12, 2020. All the analysis was conducted using Snapshot Cloud (version 1.11.0).
© 2021 Plate Recognizer, a subsidiary of ParkPow, Inc. All rights reserved.
Made with ❤ from Silicon Valley & Budapest.