The service converts data from a photograph or scan of the main page of the passport into text format with an accuracy of 96% and
Thanks to the service, entering passport data takesup to 2 seconds - this is 7-8 times faster than manual input. When creating it, scientists used computer vision algorithms. It finds the necessary fields and information in the image, and then translates it into text. Using the service, you can process any number of documents, there are no restrictions. Data can be stored on the company’s servers, without violating the policy for working with personal data.
The problem with digitizing data is veryrelevant for companies where it is necessary to work with a large amount of information on paper. In the future, we plan to expand our line of algorithms not only within the framework of document digitization, but also within the framework of digitization of all kinds of standardized forms in various industries. Thus, we will help many companies process huge arrays faster than they do now.
Ramil Kuleev, director of the Institute of Artificial Intelligence at Innopolis University
According to the developers, the main advantagefor customers is that they can use this solution without restrictions and additional payments, while other companies usually charge a certain commission for each recognized document. In the future, the service is planned to be used not only for passports, but also for other documents: SNILS, driver's license, TCP and STS.
To read Further
Scientists have been unable to catch Rambo's fox for three years. It prevents rare animals from being released into the forest.
Crypto Artists Will Explain How To Create NFT Art And Collect Crypto Art
See a heavy attack drone that carries a ton of weapons