Precise location method of QR code image symbol region at complex background

Abstract

This invention relates to accurate position method under complex background of QR codes image area, which comprises the following steps: a, adopting self-adapting valve binary method to filter focus out effect and fringe information of each module and keep central information; b, finding out the said QR codes image outline through outer fringe an finding out all position detection image according to detected image characteristics; c, processing rotation test on the said image to determine their normal QR codes image relative position; d, determining its QR code image area according to the said position detection image.

Description

technical field [0001] The invention relates to an image processing method of a two-dimensional barcode, in particular to an accurate positioning method for a QR code image symbol area under a complex background, and belongs to the technical field of image processing. Background technique [0002] QR code, also known as quick response code, is a matrix two-dimensional code symbol, which has a specific compression mode for numbers, letters, and Chinese character encoding. QR code has the characteristics of high capacity, high density, strong error correction ability, and high security strength, and is widely used in many fields such as identity verification, security anti-counterfeiting, and e-commerce. The national standard GB / T 18284-2000 records the standard of QR code. As shown in Figure 1, the QR code is a rectangular array symbol composed of a series of dark and light-colored square modules (hereinafter collectively referred to as modules) (the area where the rectangul...

AI Technical Summary

Problems solved by technology

[0020] 1. In the complex background of practical applications, the method of scanning edge points will no longer be applicable, which is quite obvious, as shown in Figure 4, the reason is that there are many interference points in the complex background, and the scanning line cannot distinguish whether the interference points are belongs to the symbol region, thus locating the error

[0021] 2. There are not always 4 vertices in the symbol area of ​​the QR code image. As shown in Figure 5, the scanning line scans out 5 outermost points (5 small dots in the figure), as shown in the figure, and the location is based on this In the symbol area, the information of several light-colored modules is lost

[0022] 3. When actually taking pictures of QR codes for recognition, the symbol area may be rotated greatly due to the shooting angle. At this time, we need to determine the position of each vertex corresponding to the symbol area of ​​the normal QR code image, that is, the rotation type, but the above The method can only locate the symbol area, but cannot judge the rotation type

Images