An Automatic Reader of Identity Documents

Identity documents automatic reading and verification is an appealing technology for nowadays service industry, since this task is still mostly performed manually,leading to wasteof economic and time resources. In this paper the prototype of anovelautomatic reading system of identity documents is presented. The system has been thought to extract data of the main Italian identity documents from photographs of acceptable quality, like those usually required to online subscribers of various services. The document is first localized inside the photo, and then classified;finally,text recognitionis executed. A synthetic dataset has been used,both for neural networkstraining,and for performance evaluationof the system. The synthetic dataset avoided privacy issueslinked to the use of real photos of real documents, which will be used, instead, for future developments of the system.