CAJ | 학술논문

针对传统的 QR码校正算法在光照、拍摄角度影响下会导致低校正率、高运算量问题，提出一种基于图像特征的QR码校正算法。对图像二值化，在进行行扫描时引入冗余点剔除过程，准确获取条码左上、右上和左下3个顶点坐标，基于边界黑色像素点间隔抽样和斜率偏离度容错处理快速找到第4个顶点，采用逆投影变换完成图像的几何校正。该算法抗光照干扰能力强，具有较高的识别成功率，可以在多种不同光照条件下，对不同拍摄角度方向的图像进行定位校正。实验结果表明，该算法能有效提高 QR码识别成功率，满足其实时性的需求。
침대전통적 QR마교정산법재광조、박섭각도영향하회도치저교정솔、고운산량문제，제출일충기우도상특정적QR마교정산법。대도상이치화，재진행행소묘시인입용여점척제과정，준학획취조마좌상、우상화좌하3개정점좌표，기우변계흑색상소점간격추양화사솔편리도용착처리쾌속조도제4개정점，채용역투영변환완성도상적궤하교정。해산법항광조간우능력강，구유교고적식별성공솔，가이재다충불동광조조건하，대불동박섭각도방향적도상진행정위교정。실험결과표명，해산법능유효제고 QR마식별성공솔，만족기실시성적수구。
Aiming at the problem that the traditional QR code correction algorithms have low correction rate and a large amount of calculation when the images are collected in the insufficient light or at the changing shooting angle,a correction algorithm of QR code based on image characteristic was proposed.Image binarization was used to pretreat the barcode,and the upper left, upper right and lower left three vertices of the quadrilateral were obtained accurately using the method of line scanning and re-dundant point elimination procedure,besides the fourth vertex was got based on the boundary of black pixel interval sampling and the j udgment of slope deviation.Inverse perspective transformation was used to geometrically correct image effectively.The al-gorithm has stronger resistant ability of light interference and higher recognition success rate under a variety of different illumina-tion conditions and at different shooting angle directions.Experimental results show that the method can greatly improve the QR code recognition efficiency,and meet the needs of real-time performance.