Scanning Application And Analysis Of Two-dimensional Code Recognition Algorithm

With the widespread of pattern recognition and scanning applications,the use of QR codes is becoming more and more frequent.In recent years,the technology of QR code recognition has been developed.Especially for the application of two-dimensional code recognition using the mobile terminal devices,the identification and mapping of corresponding image information is realized by scanning the two-dimensional code.For example,if you scan the QR code on the product,you can know the detailed information about the product.If you scan the payment code of the seller,you can make a payment.If you scan the QR code of the bill,you can compare and verify bill information.Two-dimensional codes is convenience to be used in business and so on,so it is important to promote the identification of two-dimensional code in the applications.When the two-dimensional code is actually used in the scanning of the code,it needs to be aligned with the two-dimensional code pattern to be better recognized.However,in real application,scanning a QR code often encounters problems such as the influence of light,the shape of the picture,and the distortion of the picture at the shooting angle,which all decrease the accuracy of code recognition.In this thesis,we take QR code(Quick Response Code as QR code)as an example,performing the analysis on the application of QR code recognition.Two-dimensional code recognition was firstly needed to be pre-processed,including gray-scale processing,filtering processing and binary processing.In this thesis,the weighted average algorithm is used for grayscale processing,while the method of Gaussian filtering is used for filtering and noise reduction.In terms of binary processing,the classic dynamic threshold algorithm,Otsu algorithm,is proposed.In view of the poor processing effect of the algorithm on uneven illumination pictures,local threshold method combined with mathematical morphology method is suggested.It is found that the good binarization results can solve the problem of the loss of details.In terms of two-dimensional code positioning,the unique structural features of the QR code itself are used to locate the position of the graphic by detecting the position.For the inclined QR code,the outline of QR is extracted by Laplace operator combined with the mask operation for the location.Interms of the correction of the two-dimensional code,the Hough transform is used to detect the straight line to determine the size of the inclination angle for the tilted two-dimensional code.Then,the corresponding angle can be rotated for correction.Through these processes,the practical applications of scanning recognition using the two-dimensional code is improved.Using above mentioned method,the two-dimensional code can be recognized more conveniently and quickly.