Smart Engines introduced in-browser document recognition technologies, which do not require  sending data from the client device. The new solution is suitable for personal devices, Internet of Things (IoT) objects and minimizes the risk of personal data breaches on the Internet. The company's researchers solved the complicated scientific and technical task to develop real-time AI algorithms for full-fledged in-browser recognition. This allowed to offer a high-quality and secure alternative to “cloud” OCR services for users, developers and businesses.

With Smart Engines software, users can quickly extract data and fill out online forms. The document images will not leave the perimeter of the browser. Using SDK from Smart Engines developers can deploy document recognition to any gadgets, including smart IoT devices, without tinkering with native applications. For businesses, in-browser recognition of IDs, passports, barcodes, bank cards and other documents means the development of remote omnichannel customer service without threats to privacy and security.

A web browser is the most universal interface for interaction between a person and a device connected to the Internet. The in-browser operation is an opportunity to provide customers with uniform service standards, regardless of what device the person uses. Today, with the help of Smart Engines technologies, the high-end recognition of ID cards,  driver's licenses, passports, business forms and documents, bank cards and barcodes has arrived to web applications for mobile phones, tablets, laptops, desktop computers, as well as televisions, smart watches, devices in the smart home system and any other smart devices equipped with a camera and browser.

Reliable and quick operation of the in-browser recognition algorithms is achieved through the use of proprietary GreenOCR® technology, based on the results of advanced scientific developments of Smart Engines researchers in the field of low-bit neural network architectures. The use of specialized computer vision algorithms and original 8-and 4-bit integer neural network models, as well as the deep algorithmic and software optimization carried out by the company's engineers, made it possible to ensure high recognition speed even in the browser.

To extract data, users can take photos or recognize a document in a video stream by calling the device's camera within a web page. When using Smart Engines recognition technologies, any WebAssembly-enabled web browser and a camera with a resolution of at least 640x480 should be installed on the user's device. WebAssembly allows you to run the program code directly on a web page and perform all calculations in a browser, using low-level optimization capabilities of the platform on which it runs. Today, WebAssembly is supported by the latest versions of Google Chrome, Mozilla FireFox, Safari, and Microsoft Edge browsers.

“In 2017, at Mobile World Congress we introduced the solution for secure recognition of the ID cards of EU, driving licences of USA, passports and other identity documents in mobile apps in real time, the solution did not transfer images to the services and worked on a user's phone or tablet. Today we are opening a new chapter in document recognition on the Internet. The AI algorithms allow us to securely recognize the ID cards, Drivers Licenses, Passports, Visas  and other documents in web apps in real time. As with mobile applications, our products are completely autonomous, work directly in the user's browser and do not transfer images for processing to external services for computerized or manual processing. From the business point of view, in-browser data recognition is not only a matter of concern for the safety of client data, but also the opportunity to reduce the cost of developing cross-platform applications to build an omnichannel client service,” comments Vladimir Arlazarov, CEO of Smart Engines, PhD.

“Our new paradigm of personal and sensitive data processing on the Internet is designed not only to provide a safe interaction for consumers in already created web applications, but also to pave the way for new digital service channels and a cross-sales organization by connecting IoT devices. It is safe to say, the Smart Engines AI technologies are well prepared for the challenges of the Internet of Things (IoT) era,” comments Dmitry Nikolaev, Technical Director of Smart Engines.

You can add data recognition to your web application for mobile phones, laptops, desktops, smartwatches, and other devices using Smart Code Engine, Smart ID Engine, and Smart Document Engine software products, starting with version 1.4.0. All presented solutions are optimized for SPARC, MIPS, ARM, x86 architectures and are compatible with the operating systems iOS, Android, Linux, Windows and macOS. For easy integration, developers are provided with the SDK, API documentation and embedding examples in various programming languages.

Smart Engines technologies securely recognize data from more than 1,600 types of ID documents from 210 jurisdictions worldwide, business documents and forms in more than 100 languages, as well as bank cards and barcodes. Smart Engines solutions in mobile, desktop and server applications are used by Tinkoff Group, Dukascopy Bank, Emirates NBD, Oman Arab Bank, Raiffeisen Bank, Beeline, BioCollections Worldwide, Blockpass IDN, Caribbean Airlines, iDenfy, NEC, Sumsub, Tessi, Travizory, VerifyMyAge, others. etc.