DM+：一种融合数字编码的可扫描图像生成技术  

DM+:A Scanning Image Generation Technique Based on Digital Coding

作　　者：于剑楠 李辉勇[1] 李青锋 牛建伟[1] 徐明亮[2] 曹倩[3] YU Jian-Nan;LI Hui-Yong;LI Qing-Feng;NIU Jian-Wei;XU Ming-Liang;CAO Qian(School of Computer Science and Engineering, Beihang University, Beijing 100191;School of Information Engineering, Zhengzhou University, Zhengzhou 450001;School of Computer and Information Engineering, Beijing Technology and Business University, Beijing 100048)

机构地区：[1]北京航空航天大学计算机学院,北京100191 [2]郑州大学信息工程学院,郑州450001 [3]北京工商大学计算机与信息工程学院,北京100048

出　　处：《计算机学报》2019年第9期2035-2048,共14页Chinese Journal of Computers

基　　金：国家自然科学基金(61602024,61702018)资助~~

摘　　要：二维码作为一种可扫描的图像信息载体被广泛应用于快速便捷地获取关键信息.然而,为了确保扫描的健壮性,目前大部分二维码采用黑白块构成,外观单调,并且无法将二维码的外观信息和其承载的信息关联,这从一定程度上给二维码的使用和推广带来了诸多不便.该文提出一种融合数字编码的可扫描图像生成方案.首先,提出了一种基于局部自适应的图像亮度调整算法对图像进行预处理,根据图像亮度分布情况进行自适应调整,突出图像色彩,为二维码融合奠定基础;其次,设计了一种基于高斯消元与图像显著性的二维码模块顺序调整算法,通过提取图像的显著性区域获得显著性矩阵,并依此调整二维码模块优先级,使生成的二维码中间结果与待融合的输入图像具有更好的结构相似性;最后,在兼容传统DM码扫码器基础上,提出了一种基于权重矩阵的图像融合算法,保证扫描健壮性的同时,提高二维码视觉效果.实验结果表明该方案生成400像素以下的二维码图片耗时小于1 s,并且保证扫描成功率大于92%的前提下,比现有的艺术二维码具有更好的视觉效果,较好解决了二维码美感和健壮性的冲突问题,拓宽了二维码的应用场景.As a kind of information carrier which can be scanned, two-dimensional code is widely used to obtain key information quickly and conveniently. In daily life, we can see the two-dimensional code in many places. Such as advertising pages, promotional pages, product packaging and so on. However, at present, in order to ensure the scanning robustness of two-dimensional code, most of the two-dimensional codes are formed by black and white blocks, which are monotonous in appearance and can not relate the appearance information of the code to the carried information, which has brought a lot of inconvenience to the use and promotion of two-dimensional code. The reason for this is that the two-dimensional code encodes information in black and white. The larger the difference between black and white blocks, the easier it is to be identified when scanning. However, this means that the visual effect of the two-dimensional code will getting worse. In this paper, we proposed a scannable image fusion scheme based on digital coding. The information is used to generate a two-dimensional code, and by combining any given picture with the generated two-dimensional code, a new two-dimensional code having the scanability of the two-dimensional code and the visual effect of the picture is generated. Firstly, we propose a local adaptive image brightness adjustment algorithm for image preprocessing. According to the image brightness distribution, adaptive adjustment is made to highlight the color of the image, Which lays a foundation for the integration of two-dimensional code. Besides, the adjustment of picture brightness can enhance the visual effect of the picture itself to a certain extent. Secondly, We use a two-dimensional code module sequence alignment algorithm that combines the Gaussian elimination method with the saliency of the image, the saliency matrix is obtained by extracting the salient regions of the image, and the priority of the two-dimensional code module is adjusted accordingly so that the intermediate result of the

关 键 词：DataMatrix码 图像融合 高斯消元 图像显著性 视觉效果 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]