Table image recognition method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence, is applied to the field of intelligent logistics, and discloses a table image recognition method and device, computer equipment and a storage medium. The identification method comprises the following steps: reading a table image, and carrying out semantic segmentation processing on the table image to determine a plurality of table units contained in the table image; extracting a text line image from the table image, determining a target table unit corresponding to the position of the text line image from the plurality of table units, and associating the text line image with the target table unit; performing character recognition processing on the text line image to obtain a character recognition result; and filling a spreadsheet generated based on the target table unit associated with the text line image with the character recognition result. Compared with a traditional mode, the method has the advantages that modes of semantic segmentation processing, text line image recognition, character recognition, automatic content filling and the like are adopted, so that accurate, efficient and low-cost acquisition and storage of table image information are realized.

Description

technical field [0001] The present invention relates to the technical field of form image recognition. More specifically, the present invention can provide a form image recognition method, device, computer equipment and storage medium. Background technique [0002] With the accelerated pace of economic globalization and the rapid development of international trade, the types and quantities of imported and exported commodities are increasing day by day. Among them, filling, reviewing and storing the customs declaration form or acceptance form of each step are indispensable steps. In order to facilitate the storage, sharing and retrieval of useful data, digital management methods are generally adopted, so it is particularly important to speed up the construction of customs information. However, at present, many tasks still need to be completed manually, which inevitably has great disadvantages. Taking the customs report entry work as an example, the conventional manual entry...

AI Technical Summary

Problems solved by technology

However, at present, many tasks still need to be completed manually, which inevitably has great disadvantages.

Taking customs report entry work as an example, the conventional manual entry method has problems such as time-consuming, low efficiency and high cost

Images