Innovation at the Speed of Science
With the capability to extract data from video streams of mobile devices and web cameras and directly processing the information on the device in real-time, our technology is quite simply, unrivalled and very secure. Our specialized recognition products and computer vision technologies work across servers, desktops, mobile operating systems and platforms.
By raising the bar in speed and accuracy, our software virtually elminates data entry work and associated human errors. And thus, by requiring fewer resources, it significantly reduces costs. The end results are clear for all to see: optimal business performance and increased profitability.
Reinventing the OCR engine
We have created the next generation of OCR software with higher intelligence to meet the challenges faced by organisations in document recognition and robotic process automation. Our customers trust us to provide the fastest, highest quality, smartest OCR engines available to operate their businesses as efficiently as possible.
Seeing Things Differently
The Smart ID technology is an unique outcome of relentless scientific research at the very highest level conducted in-house by the Smart Engines R&D team.
The identification of faces, digits or letters is just the first step towards achieving the true potential of Computer Vision. Our system moves far, far beyond by separating shapes and objects from backgrounds.
We have created an industry-leading technology stack which incorporates unique OCR technologies, advanced methods, and cutting edge computer vision algorithms.
With the powerful capability to recognise the entire amount of data visible and extract its most complicated parts, our software successfully overcomes traditional OCR challenges to answer myriad computer vision problems.
SmartID technology performs to the highest standards in the most difficult lighting conditions, as well as on old, out-of-date devices.
It also offers the capability to extract photos from identity documents, which is essential for selfie-check and KYC.
ID Fraud Detection
While increasing the service level for clients, by avoiding the necessity to visit the offices of service providers, remote identification brings with it additional risks of fraud and identity theft issues. Fast-growing deep fake technologies have opened up new opportunities for fraud relating to identity spoofing and is forcing producers of remote document verification systems to enhance anti-forgery features for digitial services. Thus, document verification technologies used for user identification, both remotely and face-to-face, are becoming more and more sophisticated.
Our latest 4D-OCR technology for recognition in the video stream allows you to check the protective elements in a passport or ID that change with the document angle and position relative to the line of vision or the illumination angle such as, optical variable devices (OVD) and holograms. Recognition using the camera video mode means that several images are analyzed, so it becomes possible to see the dynamics of visual element changes. Processed using AI, the absence of optical variable devices in an ID instantaneously indicates a fake.
For more heavy duty use, the SmartID.Engine automatic recognition module can be integrated with specialized multispectral scanners (e.g. at passport control points) that captured passport images simultaneously in the visible, infrared and ultraviolet bands. SmartID.Engine AI detection system determines the necessary protection elements and their conformity with the bult-in reference models for each spectral band.
Technology – The System
Our specialized recognition SDK works stand alone across servers, desktops, mobile operating systems and platforms such as x86, ARM, SPARC, MIPS.
Mobile SDK
iOS, Android
Desktop SDK
Windows, Linux, macOS
Server SDK
Linux (Ubuntu, Debian, Red Hat, CentOS, and others), Windows, Solaris
Technology Resources and Whitepapers
Whitepaper
Binarization Algorithms for Documents Recognition
During the 9-year history of International Competition on Document Binarization DIBCO17 held within ICDAR conference, a lot of bold and unconventional algorithms of binarization have been proposed.
The idea of using a mobile device for document scanning and recognition has been worked on since the appearance of the first camera phone. But for quite a long time, the poor quality of a camera on mobile devices and the low performance of mobile processors didn’t allow developing optical character recognition systems precise enough for practical use. Today smartphones and tablets are considered to be one of the best data entry options – both in terms of the camera quality and the processor.
