Identity card information extraction method and device and computer storage medium

Abstract

The invention provides an identity card information extraction method. The method comprises the steps of obtaining a to-be-processed image; detecting a human face in the to-be-processed image; determining the position of an identity card in the to-be-processed image based on the detected face, and obtaining an identity card image; performing binarization processing on the identity card image to obtain a processed image; obtaining a connected region of the processed image through a connected region search algorithm; carrying out character row segmentation according to the communication area, and carrying out character column segmentation according to a projection method; performing character recognition according to the CNN network model; and acquiring identity card information according toa preset identity card integration format and the character recognition result. In addition, the invention further discloses an identity card information extraction device and a computer storage medium.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an ID card information extraction method, device and computer storage medium. Background technique [0002] As an effective management tool for population information, ID cards have been applied to all aspects of social life. At present, most of the personal information entry in the ID card is manually entered. This method is not only time-consuming and inefficient, but also prone to incorrect information entry due to manual input, resulting in unnecessary losses; and if it can be from the perspective of image processing, Let the machine replace the human to identify the information of the image, which will solve the problem of manual entry. [0003] In the existing technology, manual review is usually used. The disadvantage of manual review is that the review is time-consuming, laborious and error-prone. The most critical thing is that the reviewer is exposed to a larg...

AI Technical Summary

Problems solved by technology

At present, most of the personal information entry in the ID card is manually entered. This method is not only time-consuming and inefficient, but also prone to incorrect information entry due to manual input, resulting in unnecessary losses; and if it can be from the perspective of image processing, Let the machine replace the human to identify the information of the image, which will solve the problem of manual entry

[0003] In the existing technology, manual review is usually used. The disadvantage of manual review is that the review is time-consuming, labor-intensive and error-prone. The most important thing is that the auditor has access to a large amount of ID card information, which may easily cause information leakage.

Images