Environment-friendly, secure, private, and highly precise proprietary OCR software solution for scanning of over 1300 identity document types from 193 countries.
- Works on everything – from server clusters to a cheap smartphone.
- Scans in any conditions – indoors and outdoors, in twilight and sunshine.
- Freedom from limits – accurate results from photos, scans, or live video.
ID Engine recognition is a GDPR and CCPA compliant solution that does NOT transfer personal data of your clients to third-party services for processing. This offline OCR software does NOT require internet connection, and does NOT save or store any private data – all processing is running on the user’s device.
ID Engine is an identity document scanning software that works autonomously on the end-user device: smartphone, tablet, smart camera, terminal, PC, or server.
The advanced recognition technology automatically extracts data in the video stream (video recognition), photos, or scans of a variety of ID documents, including passports, ID cards, driver’s licenses, visas, residence permits, various certificates, and more.
What is it for?
ID Engine is a digital transformation tool that automates ID scanning and recognition. By providing quick and precise document data entry it optimizes customer service for banks, insurance companies, telecom operators, microfinance organizations, travel and real-estate agencies, cybersport and online gaming, marketplaces, brokers, online trading platforms and other e-commerce organizations.
The AI-based identification technology helps to provide convenient, quick, safe remote customer services for banking and insurance clients. The identity document OCR allows optimizing provision of financial, insurance, transport services, helping with age verification in bars and nightclubs, selling tickets or SIM cards through self-service vending kiosks, and much more, without compromising privacy and security.
Green AI Efficiency
The GreenOCR optical recognition technology developed by our specialists allows precise text recognition in 96 languages, including such writing systems as Japanese kanji, katakana, and hiragana. In the process, as part of the Green AI approach, the system has been optimized to reduce the environmental impact. And our Edge OCR and 4D OCR technologies provide fast and autonomous text recognition in the video stream on mobile devices.
Besides precise text recognition ID Engine identifies and prevents document fraud attempts, and protects both the company and the client.
Executive Vice President for Sales and Client Service
VP & CIO
Head of CETIA
Advantages of automatic document recognition ID Engine:
- Scans IDs, passports, driver’s licenses, and more;
- Provides quick data capture and accurate text recognition;
- Protects individuals with regards to their personal data;
- Complies with the personal data processing regulations of the highest privacy standards (such as GDPR – General Data Protection Regulation, and CCPA – California Consumer Privacy Act);
- Conforms to the regulatory requirements (KYC, AML);
- Helps to streamline identity document processing and enhances the user-experience;
- Minimizes the environmental impact being a part of the Green AI.
ID Engine software is used by major banks, financial organizations, government agencies, and transportation companies around the world, to optimize various identity-centric business processes, including ID verification for customer on-boarding, quick user identification, remote identity verification, check-in automation, age verification, recruitment automation (employee registration); follow regulatory compliance (fraud detection and prevention), passport scanning for border control, and more.
ID Engine is distributed as an SDK containing a core library, simple examples, demo-applications, and documentation.
To get started with ID Engine, or with
questions, please get in touch.
Smart Engines is fully committed to provide an answer within 2 working days. However, it is your responsibility that your IT infrastructure does not block our reply or redirect it into your spams. If you haven’t received any answer from us within 2 working days, please check your spams or simply call us.
Smart Engines guarantees that the provided information will not be made public and will be used only internally.
Mobile document recognition
Supported mobile OS
- Android (version 5.1 and higher)
- iOS (version 9 and higher)
- Sailfish OS (version 2.2 and higher)
- Custom versions: upon request
Server and PC document recognition
Supported server and desktop OS
- MS Windows (all officially supported versions).
- Linux kernel based OS, including Ubuntu, CentOS, SUSE, Astra Linux, RED OS, Oracle Linux, Red Hat (RHEL), Debian, and others.
- macOS (all officially supported versions)
- QNX (version 7.0 and higher)
- Solaris (version 11.3 and higher)
- Custom versions: upon request
Identity documents from 210 countries (including all 193 UN states)
Supported document types
More than 1300 different types of identification documents are supported, and includes the following document classes, depending on the country:
- Passports – 210 countries
- Identity cards (ID card) – 106 countries
- Driving Licenses – 97 countries (including USA, Canada, Mexico, Australia, EU, CIS)
- Visas – 37 countries
- Residence permits – 40 countries
- Work permits (USA, Singapore, Russia)
- Birth, death, marriage, divorce certificates – Russia
- Vehicle Registration Certificate – 15 countries
- Social Security Cards – Russia
- Military identity card – Russia
- Refugee documents – Russia
- Temporary identity documents – Russia
- Health Insurance Cards – Russia, Japan, USA
Multilingual text OCR
Currently, the SDK is able to recognize 613 characters from different alphabets.
In addition, the SDK supports Japanese (hiragana, katakana, kanji), Korean (hangul) and Chinese (classic) characters.
Full list of supported languages:
Berber (Latin alphabet)
Chewa (Latin alphabet)
Chibarwe (Latin alphabet)
Chichewa (Latin alphabet)
Comorian (Latin alphabet)
Fiji Hindi (Latin script)
Malay (Latin alphabet)
Mongolian (Cyrillic alphabet)
Somali (Latin alphabet)
Swahili (Latin alphabet)
Tamazight (Latin alphabet)
ID Engine SDK is compatible with the following programming languages:
- C (standard C99 and higher)
- C++ (standard C++ 11 and higher)
- C # (version 6.0 and higher)
- Java (version 1.7 and higher)
- Python (version 3.7.2 and higher)
- PHP (version 5 and higher).
Integration with other programming languages and frameworks (such as Go, Perl, Xamarin, etc.) is ensured by using the API in C.
The SDK already contains examples of using ID Engine in Swift / Objective C / C / C++ / C # / Java / PHP / Python. Additional information and examples of ID Engine integration can be found here
We have created a new generation of optical character recognition (OCR) technologies that have surpassed all quality standards of OCR through the use of our latest achievements in computational intelligence and deep learning. Using the original neural network models, as well as advanced computer vision algorithms, we were able to bring the quality of automatic ID scanning to a new level.
The GreenOCR technology was developed by our company’s scientists and engineers as part of the Green AI. The green technology provides unrivaled recognition quality and speed with minimal energy consumption and low environmental impact.
The use of computational geometry, anytime algorithms and integration of the results made it possible to fully unleash the potential of video stream recognition by extracting important data from IDs both in low and excessive lightning conditions (in the dark, and in bright sunlight) and in conditions of complex scene geometry.
ID security and data safety
Our automated identity document processing solution follows the “rule of three NOs”: NO data transmission, NO data storage, NO internet access, which means that all data processing is running on the user’s end device in its local RAM, without its transfer (confirmed by an independent audit evidence). This “rule of three NOs” ensures a high level of security for all our customers.
During the recognition process, along with data selection, the smart ID recognition system detects the signs that may indicate a digital or physical forgery. ID Engine carries out cross validation of documents for further ID verification. In addition, our patented algorithms perform image and video stream forensics such as checking the holographic protection layer of the document.
To ensure reliability of the result, our SDK follows the AI-based “I extract what I see” approach, which does not involve the use of any dictionaries, and is based on responsible compact networks. In addition to recognition results, the user can access the confidence rates for each data field, get extracted images of photo and signature, get a rectified image of the whole document, get information about glares and other obstructions, and much more.
To be of the highest quality, our real-time ID scanning system is able to receive the recognition results directly from the webcam or the smartphone camera. The video stream ID processing enables our algorithms to automatically determine the clearest shot; inter-frame integration algorithms ensure reliability of results, even in cases when documents have a holographic protection with glares. ID Engine selects and returns the result best suited for further biometric identification.
Our state-of-the-art recognition algorithms based on computational geometry, computer vision and image processing allow our OCR solution to achieve high precision not only in the video stream, but also on single images with low-resolution, projective distortions, or poor lighting conditions.
Web-camera and Document-camera
ID Engine can be used with cameras you already have – laptop web-cameras, office web-cameras and docucams.
The embedded anytime algorithms allow it to achieve recognition quality comparable with the best specialized passport scanners, at 1.5-2 seconds.
We created a system which makes life easy for the client. You don’t have to find the perfect shooting angle or lighting. The advanced algorithms make the system perform all the steps of the ID scanning automatically, including: detection of the document in the image, determining its type, finding the relevant fields, and capturing specific data.
Speed & Performance
The original integer image processing pipeline, which includes 8- and 4-bit deep neural network architectures, allows achieving high recognition speed even on old smartphones by making the most of the available hardware resources.
Using our solution you can build a system which will process hundreds of thousands of documents per hour.
ID Engine is designed for intelligent capture of a variety of id documents, including: passports, ID cards, residence permits, visas, driver’s license, medical insurance cards, vehicle registration certificates and more. The system can be extended to include bank cards and barcodes recognition, and other document types upon request.
On-device OCR / on-premise OCR
ID Engine is an edge OCR system. It can scan and extract the necessary information using low-end devices: thin clients, data collection terminals, portable cash registers, etc., and work in conjunction with robotic process automation (RPA) technologies.
Robotic Processes Automation (RPA)
ID Engine is compatible with almost any operating system, hardware platform, and device. The SDK can be integrated into existing RPA infrastructure with great flexibility.
Test Drive Our Smart Engines
Free demo apps allow you to experience the power of Smart Engines software for intelligent document scanning in a real-world context.
Why not experience the power of Smart Engines for yourself? Our demo apps allow you to test the capabilities of our identity document recognition software on mobile devices in videostream or in a single image (photo, scan).
Simply display any document to the camera in real-time or choose a photo from the gallery, and the app will recognize and capture the necessary data.