Character recognition method and server for claim settlement documents

Abstract

The invention discloses a character recognition method and a server of a claim settlement document. The method includes: after the server receives the image of the claim settlement document with characters to be recognized, it divides the area according to the frame line arrangement of the claim settlement document frame format, and obtains one or A plurality of segmented regions; calling a predetermined analysis model to analyze each obtained segmented region, and performing character recognition on each analyzed segmented region by using a predetermined recognition rule, so as to identify characters in each segmented region. In the present invention, considering the impact of the frame format of the claim settlement document on the recognition accuracy, the region is segmented according to the frame line arrangement of the frame format of the claim settlement document before character recognition, and then the character recognition is performed for each segmented area, which avoids the need to perform character recognition on the entire The impact and interference of the frame lines in the document on character recognition when the characters in the image of the claim settlement document are recognized uniformly can effectively improve the recognition accuracy of the characters in the claim settlement document.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a character recognition method and a server for claim settlement documents. Background technique [0002] With the enhancement of public awareness of insurance and the substantial increase in the number of customers who purchase insurance, insurance companies need to process more and more customer claims applications, and insurance company operators need to enter more and more images of claim documents, so that record operations The manpower of personnel is tight, and at the same time, there are often mistakes in recording orders. In order to effectively reduce recording errors and improve recording efficiency, some insurance companies currently introduce OCR (Optical Character Recognition, Optical Character Recognition) technology in the process of recording documents to automatically identify the characters of the image of the claim document and fill it into the corresponding...

AI Technical Summary

Problems solved by technology

[0003] However, the existing schemes that use OCR technology to identify image characters of claim documents only use their own recognition engine to uniformly identify the characters in the entire image of claim documents, without considering the impact of the frame format of claim documents on the recognition accuracy, nor Considering the interference of the frame line in the document on character recognition, the recognition accuracy of the existing recognition scheme is not high, and a lot of manpower and material resources are required for verification

Images