Two-dimensional code positioning identifying method and apparatus based on two-stage classification

Abstract

A two-dimension code local recognition method and the device based on the secondary classification. The device includes: the coding unit, the module local unit, the firstly secondary classification unit and the secondly secondary classification. The firstly secondary classification unit makes the whole figure with the two-dimension code secondary classification firstly and output the secondary figure to the module local unit; the module local unit locates the position in the figure of every module in the two-dimension code to get the grey value of every module, then to classify secondarily to get the code number of every module and output to the coding unit; the coding unit can get the information stored in the two-dimension code according to the code number. The invention can improve the accurate of reading the two-dimension code.

Description

technical field [0001] The invention relates to image recognition technology, in particular to a two-dimensional barcode recognition method. Background technique [0002] Two-dimensional code is a new coding scheme based on traditional barcode. After error-correcting coding of characters, text, images and other information, the encoded binary code stream is expressed as a two-dimensional barcode symbol. Compared with the traditional barcode, it has the characteristics of high information capacity, strong error correction ability, and does not rely on database support. It is widely used in document reading, logistics, and electronic data exchange. [0003] Quick Response Matrix Code (QR Code) is a matrix two-dimensional code developed by Denso Corporation of Japan in September 1994. It is specially optimized for Chinese character coding and is very suitable for the processing of Chinese character information. figure 1 Shown is a QR code, which is a square, including many squ...

AI Technical Summary

Problems solved by technology

[0009] In some application scenarios (such as reading a QR code on a business card with a mobile phone), the area of ​​the QR code is small and the module density is high, and the focal length and resolution of the acquisition device are limited, so the acquired image is blurred and noisy , The module resolution of the two-dimensional code in the image is low, which brings great difficulties to the image processing of the two-dimensional code

The traditional two-dimensional code recognition technology can no longer meet the demand. We can only use macro lenses or adjustable focal lengths to obtain relatively large pictures at close range and then process them. However, these lenses are more expensive than ordinary fixed-focus lenses. relatively inconvenient

Ordinary fixed-focus lenses can only be supported by complex image processing algorithms, and complex image processing algorithms greatly increase the hardware requirements of the system and increase the processing time

Images