基于k-Means算法的彩色QR码识别  

Color QR Code Recognition Based on k-Means Algorithm

作　　者：贾丹 尤飞 张庆立 曾志高 JIA Dan;YOU Fei;ZHANG Qingli;ZENG Zhigao(College of Communication and Art Design,University of Shanghai for Science and Technology,Shanghai 200093,China;School of Computer Science,Hunan University of Technology,Zhuzhou Hunan 412007,China)

机构地区：[1]上海理工大学出版印刷与艺术设计学院,上海200093 [2]湖南工业大学计算机学院,湖南株洲412007

出　　处：《包装学报》2017年第5期62-68,共7页Packaging Journal

基　　金：上海理工大学博士启动基金资助项目(1D-13-309-007);模式识别国家重点实验室开放课题基金资助项目(201700009);湖南省教育厅基金资助项目(17C0481)

摘　　要：为了降低彩色QR码解码过程中出现的混叠效应,提高彩色QR码解码的正确率,提出一种基于HSV颜色模型的k-Means聚类算法。为了选择适合彩色QR码的颜色空间模型,通过实验验证了在RGB,Lab,HSV 3个颜色模型下k-Means聚类算法的效果。在HSV颜色模型下,根据等欧氏距离的原则建立彩色编码模块的配色模型,最大程度地减小解码中颜色的混叠效应。彩色QR码解码预处理阶段,利用基于HSV颜色模型的光线补偿的k-Means聚类算法对彩色编码模块进行颜色分离,以提高解码的精度。研究结果表明:在HSV颜色模型下,k-Means聚类效果最好,图像区域分类效果最清晰;所建立的配色模型可以最优地为彩色编码模块配色;基于HSV颜色模型的光线补偿的k-Means聚类算法可以提高彩色QR码解码的正确率。因此,建立合理的配色模型进行彩色编码模块的颜色设置,同时采用基于HSV颜色模型的光线补偿的k-Means聚类算法进行颜色分割,可以大幅度地降低彩色QR码编码模块之间的混叠效应,从而显著提高彩色QR码解码的正确率。In order to reduce aliasing effects in color QR decoding and improve the accuracy of color QR code decoding,a k-Means clustering algorithm based on HSV color space was proposed.Experiments were conducted to verify the effect of k-Means clustering algorithm in three color spaces of RGB,Lab and HSV for choosing the right color space for color QR codes.In HSV color space,the color matching model of color coding module was established according to the principle of equal Euclidean distance,so as to minimize the aliasing effect of color in decoding.In the preprocessing phase of color QR code decoding,k-Means clustering algorithm based on ray compensation in HSV color spaces was used to separate color coding module and to improve the decoding accuracy.The result showed that in HSV color space,the k-Means clustering effect was the best,with the image region classification effect being the clearest.The color matching model could optimally match the color coding module.Therefore k-Means clustering algorithm based on ray compensation could be used in HSV color space to improve the accuracy of color QR decoding.Establishing a reasonable model was to set color encoding modules,and to use k-Means clustering algorithm based on ray compensation in HSV color space to achieve color segmentation.It could greatly reduce the aliasing effect between QR color encoding modules and significantly improve the accuracy of color decoding QR.

关 键 词：彩色QR码 K-MEANS聚类算法 HSV颜色模型 彩色QR码的配色模型 

分 类 号：TP391.44[自动化与计算机技术—计算机应用技术]