IDscan aims to be the world’s foremost identity document scanning business.

It can provide an effective and immediate decision on the legitimacy of an identity document such as a passport, permit or driving licence, taken not just from passport scanners, but from webcams and mobile devices like smartphones and tablets.

IDscan’s world-leading document verification technology is used across a wide range of business sectors, including financial institutions, telecoms providers, recruitment and security.

The business challenge

IDscan uses computer vision – the science that gives machines human vision capabilities – to support processing and automatic identification of document types. Computer vision extracts, analyses and understands useful information from a single image or a sequence of images. However, IDscan faced a skills shortage in the market when searching for in-depth knowledge of advanced biometrics techniques, most notably, automated face recognition.

How did the University help?

It was established that the School of Engineering and Technology at the University had the breadth of expertise that IDscan required. Working together, the University team and IDscan were awarded funding of £131,000 from innovate UK for a three-year Knowledge Transfer Partnership (KTP) project to introduce a face recognition capability into IDscan’s new products for automated identify authentication.

Real business benefits

The collaboration has developed a system based on a fusion of machine learning techniques, submitted a patent application and created a research paper on the work. The second stage of the project focuses on the latest ground-breaking technique in machine learning, known as deep learning or deep neural networks (DNN). The project has successfully developed the first prototype of a DNN-based face recognition system, which has been demonstrated by IDscan in several marketing events.