Research On QR Code Recognition Algorithm

With the advent of the information age,the advance of computer technology and the popularity of smart phones,more and more 2D bar codes are being used in all walks of life.QR code is the first 2D bar code which is encoded directly by non English characters.Because of its fast response and unique coding characteristics,QR codes occupy the leading position in the 2D bar code market of our country.However,in daily life and industrial production,the QR code images often have problems such as uneven illumination,distortion,background complexity and so on.In these cases,the QR code is difficult to be correctly read out.Therefore,the research on QR code recognition technology is of great significance.QR code recognition algorithm includes image preprocessing algorithm,localization algorithm and correction algorithm.The purpose of these algorithms is to extract a clear and regular QR code,which is easy to be decoded.In this paper,the related technology of QR code recognition is studied deeply,and the recognition algorithm is implemented to deal with the QR code image which has uneven illumination and complex background.Firstly,preprocessing algorithm is realized based on the character of QR code symbol.Image preprocessing can be divided into four steps:gray scale transformation,illumination equalization,denoising and binaryzation.In view of gray scale transformation,the YUV linear transformation algorithm is adopted.In view of illumination equalization,the histogram equalization algorithm is adopted.In view of denoising,the wavelet transform algorithm is studied and applied.In view of binaryzation,Otsu algorithm is analyzed and improved.The results show that image preprocessing algorithm can effectively remove the interference information and greatly enhance the contrast.Then,based on mathematical morphology theory and edge detection algorithm,localization algorithm is implemented and improved.The results show that the improved algorithm can effectively remove the background information and obtain the complete QR code symbol.Then,based on least square method and reverse perspective transform,correction algorithm is realized.In view of discontinuity of gray value,bilinear interpolation method is used to optimize the gray value.The results show that the correction algorithm can effectively correct the distortion of the image,recover its true shape.Finally,decoding algorithm is studied,and the accurate information is obtained.Overall,the QR code recognition algorithm,which meets the requirements of real-time and accuracy,has certain practical value.