Bill identification method and device, electronic equipment and computer readable storage medium

Abstract

The invention provides a bill identification method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps: obtaining a target training bill imagewhich comprises a target synthesis bill image and a target real bill image; pre-training a target recognition model according to the target synthetic bill image, wherein the target recognition model comprises a first neural network, a second neural network and a third neural network; according to the target real bill image, carrying out migration training on the pre-trained target identification model; processing the target bill image through the first neural network after migration training to obtain a first target bill in the target bill image; processing the first target bill through the second neural network after migration training to obtain a target textbox in the first target bill; and processing the target textbox through the third neural network after migration training to obtaintarget text information in the target textbox.

Description

technical field [0001] The present disclosure relates to the technical field of image processing, and in particular to a bill recognition method and device, electronic equipment, and a computer-readable storage medium. Background technique [0002] In the insurance industry, due to the variety of insurance types and the endless content of protection, there are many types of insurance bills. In order to speed up the insurance payment process, it is usually necessary to automatically recognize the insurance bills to obtain the bill content, that is, convert the unstructured bill image into structured data to realize the extraction of bill information. At present, there are roughly two types of structured processing of bills: (1) manual entry of bill images. This method requires employees to perform manual operations, which is time-consuming and labor-intensive, with high labor costs, and due to manual entry, it will inevitably bring some human errors; (2) based on OCR (Optica...

AI Technical Summary

Problems solved by technology

This method requires employees to perform manual operations, which is time-consuming and labor-intensive, with high labor costs, and due to manual entry, it will inevitably bring some human errors; (2) based on OCR (Optical Character Recognition, optical character recognition) technology for insurance bills Extract information automatically

However, OCR technology is affected by many factors, resulting in poor recognition effect

For example, OCR technology will be affected by factors including: complex background, artistic fonts, low resolution, non-uniform lighting, image degradation, character deformation, multi-language mixing, complex text line layout, and incomplete characters in the detection box, making the overall OCR technology Poor recognition effect

Images