Deep learning aided system design method for intelligent reimbursement robot

Abstract

Recently, the massive use of invoices has increased the burden on financial staff. In order to ensure high accuracy, conventional financial reimbursement is mainly manual, which wastes lots of human and material resources. This paper proposes an intelligent reimbursement system based on a deep learning method. The system mainly contains face recognition, invoice identification, and information storage. Face recognition ensures the security in the invoice reimbursement, and invoice identification has acceptable accuracy and operating speed. The experimental results indicate that the proposed system achieves both high accuracy and fast running speed.