We know that your ALPR is mission-critical. We’re here to help you succeed. Not convinced? Take us for a free test drive.
Upload an image (up to 3MB) and we’ll decode the license plate.
We love Amazon (especially Prime!) and have a ton of respect for Amazon Rekognition. Amazon Rekognition lets you identify objects, people, text, scenes, and so forth in both images and videos. It’s powerful. Yet, powerfulness and breadth of scope may be good for some use general OCR-type use cases, but not others.
In the world of automatic license plate recognition (ALPR), having a specific ALPR engine to do the work is much preferred. The benefits include much faster deployment, significantly higher accuracy, and stronger support.
Our APIs make it easy to incorporate ALPR metadata into your workflow and systems.
Get software that’s tuned to decode license plates, vehicle make, model, color, and more!
We eat license plates for lunch and have a passion to help others embrace our tech.
A general, important concern in using Amazon Rekognition for license plate recognition is that it decodes every single character in the image. Every single character. It’s like having a toddler spell out the ABCs of whatever she can read in an image.
Take for example the two images below. The one of the left is the decoded license plate from Plate Recognizer and the one on the right is from Amazon Rekognition.
So in terms of integrating the results into your system:
It can take a considerable amount of effort to map the Amazon Rekognition results to the data elements you actually care about.
On the other hand, with Plate Recognizer, you get the full documentation of each JSON Response. No need to guess. Each parameter such as the license plate or the region is clearly spelled out.
To make things even easier, you get the API response in 8 different programming languages! Just copy and paste and you’re on your way to done!
Bumper Stickers, Phone Numbers. Because Amazon Rekognition decodes every single character in the photo, there is a high occurence of false positives. This issue is compounded when street signs, retail stores and other things in the background.
Plate Recognizer, on the other hand, is specially tuned to focus on the license plate and ignore bumper stickers and phone numbers on the vehicle.
Double, Triple Stacked Characters. We find that AWS Rekognition struggles with double- and triple-stacked characters that are quite common in license plates in the USA and other countries.
Icons. License plates with icons can also be a struggle for Rekognition. That’s because the icon may overlap with the characters on the plate.
Two Lines of Text. License plates with two lines of text are very common on motorcycles in many countries throughout the world. Again, AWS Recknogition has a tough time decoding the plate.
Tough Characters. Some license plates contain tough characters or characters of different sizes. This is easy for the human eye to distinguish but tough for an OCR engine such as Amazon Rekognition. Plate Recognizer, on the other hand, accurately decodes the plate.
Whether you are using ALPR to monitor cars in your apartment complex or need it to reward your loyal customers in the drive-thru lane, we’re here to help.
Our team of technical support specialists are located throughout the globe to assist. And, when needed, we escalate issues straight to our engineers.
Get a Free Trial of Snapshot or Stream! See our Pricing plans.
The Plate Recognizer team has been second to none as we developed our product. They are our partners for the long term!
Plate Rec is highly responsive to our business and technical needs.
Plate Rec ALPR is powerful and accurate. We are thankful for a fruitful partnership.
The Plate Rec team is always there to support us whenever we need it.
We selected Plate Rec after some rigorous benchmarking — and never looked back!
Plate Rec is committed to our success and is super-easy to work with!
* DISCLAIMERS. Cost Comparison analysis is baesd on published website pricing as of January 19, 2021. Since OpenALPR provides other products (some with hardware included), we only analyzed CarCheck API prices. Accuracy Comparison analysis is based on the latest versions of ALPR software (Snapshot Cloud and CarCheck API) as of January 19, 2021. Images were randomly selected across a variety of countries and are not intended to reflect the actual images used for testing by customers. For details on the methodoly on Angle, Blur Distance, Lightness Level and Pixel Plate Width analysis, pleaes refer to the Methodology section of ALPR Results. Plate Recognizer does not claim to have a better ALPR engine, a cheaper ALPR software or better customer support compared with any ALPR software vendor. The analysis on this page is for illustration purposes only. We encourage all customers to perform their own tests in the areas of ALPR costs, accuracy, and customer support.
© 2023 Plate Recognizer, a subsidiary of ParkPow, Inc. All rights reserved.
Made with ❤ from Silicon Valley & Budapest.