Kapsch Releases AI-Based Upgrade Of ANPR Software

Kapsch TrafficCom announced a “major update” of its Automatic Number Plate Recognition (ANPR) software. “The latest release features a reworked architecture that includes various deep learning elements, resulting in a significant performance boost. With the new ANPR software, an artificial intelligence software was trained to accurately and reliably identify number plates with hundreds of thousands of images in a [manner compliant with the European Union’s General Data Protection Regulation (GDPR)].” Kapsch’s Julia Azfar said the software upgrade was the product of a continuous improvement process that involves collaboration with partners and suppliers “to combine the latest technologies on the market with [Kapsch’s] own expertise in image processing and in artificial intelligence.” The new recognition software was trained through a process of manual image annotation. Kapsch obtained annotation workers in conjunction with its Vienna pilot project to introduce disadvantaged people to the job market or prepare them for new occupations.

(Editor’s Note: GDPR, a binding regulation rooted in EU privacy and of human rights law, regulates the electronic collection and use of personal data and protects individual rights of privacy and data control afforded residents of the EU and the European Economic Area.)