Bill information extraction method and device, equipment, medium and product

Abstract

The invention discloses a bill information extraction method and device, equipment, a medium and a product, and the method comprises the steps: carrying out the text recognition of a bill image, and obtaining the coordinate information of a textbox of each text region and the text information in the textbox; graph structure data are constructed, each node represents the coordinate information of the single textbox and the feature information of the text information, and the edge between any two nodes represents the feature information of the relative distance information between the textboxes of the two nodes; adopting a preset graph convolutional neural network to carry out multiple times of iterative updating on the feature information of the nodes and the edges in the graph structure data to obtain updated graph structure data; and inputting the feature information of the nodes in the updated graph structure data into a classification network for classification, and determining a classification label corresponding to the text information of each textbox. According to the method and the device, the text information required by the order data can be accurately extracted from the bill image.

Description

technical field [0001] The present application relates to the technical field of e-commerce information, and in particular, to a method for extracting bill information and a corresponding device, computer equipment, computer-readable storage medium, and computer program products. Background technique [0002] For some offline e-commerce scenarios, consumer users of the e-commerce platform may provide bill information corresponding to their e-commerce orders, such as product information or To pay for the remittance payment information corresponding to an order, etc., according to this, it is necessary to obtain order-related data on the basis of pictures. [0003] Obtaining the data corresponding to the order from the picture involves the rapid extraction and structuring of the key information in the transaction order, which helps merchants to quickly collect all transaction orders, quickly assist merchants to complete order information statistics, and help merchants to grasp...

AI Technical Summary

Problems solved by technology

[0004] For image-based bill information processing, the industry generally obtains text information based on image text recognition, and combines conventional rule matching for data structuring. This type of method is relatively primitive, and the processing process is complicated and the degree of intelligence is insufficient. Especially when it is necessary to provide bill information extraction services corresponding to multiple languages ​​in multiple regions, it is not universal, so it needs to be improved

Images