Data detection method and device, computer equipment and storage medium

Abstract

The invention discloses a data detection method and a device, a computer device and a storage medium. The method comprises the steps of obtaining a specified certificate picture comprising a face portrait; correcting the specified certificate picture to be placed according to a preset standard posture so as to obtain an intermediate picture; inputting the intermediate picture into a preset human face feature point detection model for calculation to obtain a plurality of human face feature points; calculating clustering center positions of the plurality of face feature points, and generating aminimum bounding rectangle of the plurality of face feature points; calling the standard certificate picture from a preset database; scaling the standard certificate picture in equal proportion to obtain a scaled picture; overlapping the reference center position in the zoomed picture with the clustering center position in the intermediate picture so as to obtain an overlapped part in the intermediate picture; and marking the overlapped part as a certificate body of the specified certificate picture. Therefore, the data detection accuracy is improved.

Description

technical field [0001] The present application relates to the field of computers, in particular to a data detection method, device, computer equipment and storage medium. Background technique [0002] ID picture analysis is very common in production and life in today's society. For example, when handling a certain business, it is necessary to take a picture of the ID and upload the picture of the ID. If the document body in the document picture can be accurately detected (because the generally taken document picture, the viewfinder range will be larger than the document body, so the document body will generally not fill the entire document picture), it can improve the subsequent analysis of the document picture . The traditional document body detection method is generally based on pixel point detection to obtain the coordinates of the document body (such as the coordinates of the four vertices of the matrix document body). This method consumes a lot of computing resources a...

AI Technical Summary

Problems solved by technology

The traditional document body detection method is generally based on pixel point detection to obtain the coordinates of the document body (such as the coordinates of the four vertices of the matrix document body). This method consumes a lot of computing resources and is accurate in identifying fuzzy documents. The defect of low rate (due to the vagueness of the certificate, the border of the certificate body is not clear, so the body of the certificate recognized by the traditional method will have a large deviation)

Images