Text detection method, text detection device, electronic equipment and storage medium

Abstract

The invention provides a text detection method, a text detection device, electronic equipment and a storage medium. The method comprises: firstly obtaining a to-be-detected picture; detecting a targettext area in the picture to be detected, wherein the target text area comprises at least one of a header text area, a seal text area and a layout text area; determining a target text detection modelcorresponding to the target text area according to a preset corresponding relationship between the text area and the text detection model; and detecting the text in the target text area by adopting atarget text detection model to obtain a target text detection box. According to the technical scheme, the target text region in a to-be-detected picture is firstly detected, and by combining the advantages of different text detection models, the text detection model matched with the target text area is adopted to detect the text in the area in a targeted mode, so the integrity of text detection can be improved, and powerful support is provided for subsequent text recognition.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a text detection method, a text detection device, electronic equipment and a storage medium. Background technique [0002] Nowadays, with the development of the economy and the improvement of people's living standards, more and more people choose to purchase medical, commercial, financial and other insurance. Some of these insurance companies have gradually started self-service claims business. For example, in the process of medical claim settlement, users only need to take photos of outpatient or hospital bills and upload them to the insurance company system, and the insurance company salesmen will upload the information on the bill pictures uploaded by users. Entering into the claim settlement system, the whole process requires a lot of manpower to process the bill images uploaded by users, and the error rate of manual data entry is high. [0003] The optical charact...

AI Technical Summary

Problems solved by technology

Some of these insurance companies have slowly started self-service claims business. For example, in the process of medical claim settlement, users only need to take photos of outpatient or hospital bills and upload them to the insurance company system, and the insurance company salesmen will upload the information on the bill pictures uploaded by users. Entering into the claims system, the whole process requires a lot of manpower to process the bill images uploaded by users, and the error rate of manual data entry is high

[0003] The optical character recognition technology (OCR, Optical Character Recognition) can improve the efficiency of receipt entry under certain conditions. However, the pictures uploaded by users are affected by the shooting angle, and there are curved text, oblique text, etc. In addition, due to the complexity of the receipt layout , there are also long text in the header, text in the seal, etc. These situations greatly increase the difficulty of text detection and recognition

For complex layouts, the related text detection technology has the problem of being unable to detect or incomplete detection, because the quality of the detection effect directly affects the final text recognition results, therefore, to solve the problem of text detection in complex layouts, in the optical character recognition technology appears to be particularly important

Images