Invoice identification method and device

Abstract

The embodiment of the invention provides an invoice identification method and an invoice identification device, and the invoice identification method comprises the following steps: building a graph model of a corner point according to the corner point detected from an invoice image; detecting a text area in the invoice image according to the graph model of the corner point and a pre-established graph model template; performing line text detection on the character area according to the character area; and according to the detected line text, identifying invoice elements on the invoice image, sothat more useful information can be extracted from the invoice image, the purpose of rapidly and accurately identifying bill characters is achieved, and finally financial personnel are liberated fromtedious bill auditing work.

Description

technical field [0001] The embodiments of the present application relate to the technical field of data processing, and in particular, to an invoice identification method and an invoice identification device. Background technique [0002] The problem facing financial management is a large amount of data collection and information processing, which requires a lot of manpower and material resources. The biggest pressure in the process is the collection of original bills and information entry. The heavy bill entry and management work consumes manpower and Another waste of time. To this end, the invoice automatic identification solution was launched. By using the invoice identification technology, the information on the value-added tax invoices and other bills is collected in batches, and the structured data is output. Compared with the traditional manual data entry, the number of financial personnel is greatly reduced. workload, improving its work efficiency. [0003] In the ...

AI Technical Summary

Problems solved by technology

[0008] The disadvantages of the above-mentioned first solution are: due to the huge number of Chinese characters, many types of fonts, image deformation, similar characters, rare characters, etc., the classifier’s ability to recognize a single Chinese character is weak; The segmentation and positioning of characters have a great influence, and more useful information cannot be extracted from the invoice image, so as to achieve the purpose of fast and accurate recognition of bill characters, and the recognition accuracy is low

[0009] The disadvantage of the second solution is that while processing the line text, structural information such as tables is not considered, so that the meaning of the obtained text information is not obvious, and the information of each invoice element is mixed together, and more information cannot be extracted from the invoice image. Useful information to achieve the purpose of quickly and accurately identifying invoice characters, and the accuracy of identification is low

Images