A bill recognition method and system based on mobile phone photography

Abstract

The present invention provides a bill recognition method based on mobile phone photography, which includes the following steps: S1. After learning various types of bills, the intelligent recognition system in the mobile phone stores the key information of each type of bills, and establishes the bill key Information database; S2. Scan various mixed bills into electronic images by taking photos with mobile phones, and upload them to the intelligent recognition system to obtain keywords. For tilted and rotated pictures, the intelligent recognition system automatically recognizes and corrects them; S3. The obtained electronic version The image is compared with the stored key information or keywords according to the scanned information, and the type of the bill is obtained. S4. Perform image processing on the invoices that cannot be recognized or are incorrectly checked by the tax bureau for secondary identification. The present invention does not require manual input, and does not need to arrange bill types, which greatly improves efficiency and accuracy, saves cost and time, and liberates manpower.

Description

technical field [0001] The present invention relates to the technical field of bill recognition methods, in particular to a bill recognition method and system based on mobile phone photography. Background technique [0002] With the implementation of my country's taxation system reform from business tax to VAT, value-added tax is now the most important type of turnover tax in my country. The taxation scope of value-added tax has expanded from covering most of the secondary industry to the secondary and tertiary industries. most of the industries. [0003] Now the collection and management of value-added tax is more stringent. At the same time, the number of value-added tax invoices has increased significantly. Manual entry is too slow and checking the authenticity is very time-consuming, and the efficiency is low and the error rate is high. At the same time, there are also such problems with more types of bills, such as various bank receipts, machine-printed invoices, train ...

AI Technical Summary

Problems solved by technology

In the traditional manual input method, users need to invest a lot of labor and time costs, which not only increases the operating cost, but also makes it difficult to increase the input speed and reduce the error rate, which brings many negative effects on improving the timeliness of business processing and improving the quality of enterprise services

[0004] However, only identifying one type of invoice does not conform to the actual usage situation. Usually, enterprises have multiple invoices that need to be recorded, such as value-added tax invoices, machine-printed invoices, quota invoices, train tickets, bank notes, etc.

Images