A qr code position detection pattern grouping method

Abstract

The invention provides a QR code position detection pattern grouping method, which comprises the steps of extracting three position detection patterns acquired by scanning; acquiring the number of black and white alternating times of two positioning patterns among the three position detection patterns, judging whether an absolute value of the difference of the number of alternating times is less than a first limit value or not, determining the version of a QR code formed by the three position detection patterns, and acquiring the dimension of the QR code; recording center point coordinates of the three position detection patterns and the dimension of the QR code; and grouping the position detection patterns accurately so as to improve the accuracy.

Description

technical field [0001] The invention relates to a method for grouping QR code position detection graphics. Background technique [0002] At present, there are many types of two-dimensional codes on the market, but there are only three most common ones, namely: pdf417, datamatrix and QR code, among which QR code is the most widely used two-dimensional code in civilian use. The biggest feature of the QR code is that there are three position detection patterns on the upper left, upper right and lower left of the code, and each position detection pattern can find a fixed 1:1:3:1:1 ratio relationship whether it is scanned horizontally or vertically; And it has a strong anti-interference ability, even partial defects can be accurately located. However, the actual situation is that due to the printing quality, lighting conditions and camera imaging, it is necessary to relax the requirements of its proportional relationship, and it must be determined that there is a scanning line t...

AI Technical Summary

Problems solved by technology

However, the actual situation is that due to the printing quality, lighting conditions and camera imaging, it is necessary to relax the requirements of its proportional relationship, and it must be determined that there is a scanning line that meets the requirements as a possible position detection pattern, otherwise the incomplete position detection pattern cannot be processed. In the case of QR code rotation, this will bring a serious problem. After scanning the image, a lot of pseudo position detection graphics will be obtained. If our goal is to search for an unknown number of QR codes in large-format images, we need Separating QR code location detection patterns from false objects and grouping them correctly

Traditional multi-QR code decoding ideas are divided into two types: one is to combine the found position detection graphics for trial-and-error decoding, which consumes a lot of time and is not feasible

The other is to use the positional relationship between the found position detection patterns to determine a QR code. This method has very low tolerance to pseudo position detection patterns and cannot meet actual needs.

Images