Ticket identification method, apparatus and storage medium

Abstract

The invention provides a ticket identification method comprising the following steps of: firstly detecting and identifying key fields with regard to the inputted ticket picture by using intelligent identification algorithm, then obtaining the confidence of machine identification results of the key fields; sending the key fields of which the confidence is lower than the threshold value to a crowd sourcing platform to verify; sending a same key field to a plurality of users to verify by the crowd sourcing platform, and obtaining verification results from a plurality of users for the same key field; and finally outputting the identification results of the ticket picture. The invention further provides a ticket identification apparatus and a storage medium. According to the ticket identification method, apparatus and storage medium, accuracy of ticket identification can be improved, so that a file can be quickly constructed.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a bill identification method, device and storage medium. Background technique [0002] Large enterprises, institutions, medical examinations in hospitals, and the insurance industry all have a large amount of bills that need to be collected, entered, and archived electronically. At present, the degree of digital management of bills in my country is still relatively low, and the methods of manual entry and manual filing that are often used are labor-intensive, low-efficiency, costly, and error-prone. Although it is currently possible to use machine learning methods to identify bills, the recognition accuracy is not high, which will cause errors in various information on bills, making it impossible to quickly file and improve work efficiency. Contents of the invention [0003] In view of the above content, it is necessary to provide a bill identification method, device and...

AI Technical Summary

Problems solved by technology

At present, the degree of digital management of bills in my country is still relatively low, and the methods of manual entry and manual filing that are often used are labor-intensive, low-efficiency, costly, and error-prone

Although machine learning methods can be used to identify bills at present, the recognition accuracy is not high, which will cause errors in various information of bills, making it impossible to quickly file and improve work efficiency

Images