A financial ticket image intelligent recognition system based on deep learning

Facing rapid growth in the issuance of financial tickets (e.g., bills, invoices), traditional manual invoice reimbursement and financial accounting systems are imposing an increasing burden on financial accountants and consuming excessive manpower. To solve this problem, we propose an iterative self-learning framework of Financial Ticket Intelligent Recognition System (FTIRS), which supports iteratively updating and extensibility of the algorithm model, which are the fundamental requirements for a practical financial accounting system. In addition, we designed a simple yet efficient Financial Ticket Faster Detection Network (FTFDNet) and an intelligent data warehouse of financial tickets to strengthen its efficiency and performance. Currently, the system can recognize 482 types of financial tickets and has an automatic iterative optimization mechanism. Thus, with increased application time, the types of tickets supported by the system will increase, and the accuracy of recognition will improve. Experimental results show that the average recognition accuracy of the system is 97.41%, and the average running time for a single ticket is 173.72 ms. The practical value of the system has been verified in business. It can greatly improve the efficiency of financial accounting and reduce the human cost of accounting staff.