二维条形码识别系统研究与设计
|
摘要
如今是信息产业革命的高潮，人们的生活充斥着各种各样的高新信息技术，而条形码技术则是融入人们生活中最具有代表性的一项，它在人们的日常生活中随处可见，给人们新时代的生活带来了巨大的便利。条形码技术是集数据编码、条码印刷、条码识别、数据采集于一体的全球通用商务语言，其拥有标准化的数据采集手段，可对商品信息进行统一编码和标识。目前条形码技术在我国正处在研究、开发和推广的重要时期，随着中国受加入WTO所带来的经济高速发展的影响，中国条形码技术的研究已迫在眉睫。目前国内二维条码技术的成熟性还远没有一维码技术成熟，其应用范围仅在部分交通运输业和商业等场合。因此，对于二维条码识别系统的研究有重要的价值和深远的意义。
本论文首先介绍主流二维条码数据矩阵的编码原理，然后详细讲述基于图像处理的二维条码识别算法，该算法针对手持设备获取二维码图像的不清晰、难定位等问题有较好的处理效果；接着重点研究图像预处理及条码定位算法，同时在分析对比各种二维条码图像定位算法优劣的基础上，选择使用最优定位算法适用于移动设备的二维条码识别。并最终在Microsoft Visual Studio 2008集成开发环境下将整套二维条形码识别算法实现。
本文围绕提高条码识别率的关键点，重点针对图像预处理及定位环节进行创新和改进。主要研究工作如下：
（1）对于图像二值化处理过程中所使用的阈值在选取上进行多次试验，寻找最优方法获取最优阈值。
（2）针对条码定位误差大，识别效果不理想的问题，选择Sobel算子进行边缘检测，并对处理后所得结果投影定位，然后结合Hough变换精确定位条码。
（3）对条码旋转校正以及条码数据采样算法进行深入地研究，并在数据采样提取时，提出使用3×3邻域投票决定像素值的方法。
（4）实现整套系统，并对多个算法进行验证对比。
Now the information industry revolution climax become more and more popular,people's life is filled with all kinds of high-tech information technology，While the barcode technology is the most representative one that integrated into people's lives, it inthe daily life of people everywhere, it bring people great convenience in the new agelife. Bar code technology is a universal business language integrated data coding, barcode printing, bar code recognition, data collection, It has a standardized datacollection methods, which can uniform coding and marking for Commodityinformation. The bar code technology in our country are in the important period ofresearch, development and promotion, With the influence of rapid economicdevelopment brought by China's accession to the WTO, Chinese bar code technologyresearch has been imminent.
But at present, two dimensional barcode technology is far from a d codetechnology in domestic，and its application only for part of transportation industry andbusiness field. So the research about two dimensional barcode’s identification systemis necessary.
First, this paper introduces the principle of major two-dimensional codes DataMatrix encoding theory, then describes the two-dimensional bar code recognitionalgorithm based on image processing in detail, which has a good processing effect forhandheld devices to get two dimensional barcode image that is not clear and difficultorientation; Then highlights the algorithm of image preprocessing and bar codepositioning, and in contrast the pros and cons of all kinds of two dimensional barcodeimage positioning algorithm choose to the optimal positioning algorithm used inmobile devices of two dimensional barcode identification. And finally complete a setof two-dimensional barcode identification algorithm in Microsoft Visual Studio 2008integrated development environment.
To improve the recognition rate for barcode, this paper make an innovation andimprovement on image’s preprocessing and positioning link. The main completed works are as follows:
(1) For binary image threshold value has carried on many times the test, lookingfor the best methods to gain the most optimum threshold.
(2) To bar code positioning error is big, the result of identify is bad, chooseSobel operator on edge detection and treatment results projection positioning.
(3) Do a thorough research on the barcode rotation correction and bar code datasampling algorithms, and when in data sampling extraction, using the 3×3neighborhood voted to pixel values.
(4) To implement the system, and make verification contrast for multiplealgorithm.
本论文首先介绍主流二维条码数据矩阵的编码原理，然后详细讲述基于图像处理的二维条码识别算法，该算法针对手持设备获取二维码图像的不清晰、难定位等问题有较好的处理效果；接着重点研究图像预处理及条码定位算法，同时在分析对比各种二维条码图像定位算法优劣的基础上，选择使用最优定位算法适用于移动设备的二维条码识别。并最终在Microsoft Visual Studio 2008集成开发环境下将整套二维条形码识别算法实现。
本文围绕提高条码识别率的关键点，重点针对图像预处理及定位环节进行创新和改进。主要研究工作如下：
（1）对于图像二值化处理过程中所使用的阈值在选取上进行多次试验，寻找最优方法获取最优阈值。
（2）针对条码定位误差大，识别效果不理想的问题，选择Sobel算子进行边缘检测，并对处理后所得结果投影定位，然后结合Hough变换精确定位条码。
（3）对条码旋转校正以及条码数据采样算法进行深入地研究，并在数据采样提取时，提出使用3×3邻域投票决定像素值的方法。
（4）实现整套系统，并对多个算法进行验证对比。
Now the information industry revolution climax become more and more popular,people's life is filled with all kinds of high-tech information technology，While the barcode technology is the most representative one that integrated into people's lives, it inthe daily life of people everywhere, it bring people great convenience in the new agelife. Bar code technology is a universal business language integrated data coding, barcode printing, bar code recognition, data collection, It has a standardized datacollection methods, which can uniform coding and marking for Commodityinformation. The bar code technology in our country are in the important period ofresearch, development and promotion, With the influence of rapid economicdevelopment brought by China's accession to the WTO, Chinese bar code technologyresearch has been imminent.
But at present, two dimensional barcode technology is far from a d codetechnology in domestic，and its application only for part of transportation industry andbusiness field. So the research about two dimensional barcode’s identification systemis necessary.
First, this paper introduces the principle of major two-dimensional codes DataMatrix encoding theory, then describes the two-dimensional bar code recognitionalgorithm based on image processing in detail, which has a good processing effect forhandheld devices to get two dimensional barcode image that is not clear and difficultorientation; Then highlights the algorithm of image preprocessing and bar codepositioning, and in contrast the pros and cons of all kinds of two dimensional barcodeimage positioning algorithm choose to the optimal positioning algorithm used inmobile devices of two dimensional barcode identification. And finally complete a setof two-dimensional barcode identification algorithm in Microsoft Visual Studio 2008integrated development environment.
To improve the recognition rate for barcode, this paper make an innovation andimprovement on image’s preprocessing and positioning link. The main completed works are as follows:
(1) For binary image threshold value has carried on many times the test, lookingfor the best methods to gain the most optimum threshold.
(2) To bar code positioning error is big, the result of identify is bad, chooseSobel operator on edge detection and treatment results projection positioning.
(3) Do a thorough research on the barcode rotation correction and bar code datasampling algorithms, and when in data sampling extraction, using the 3×3neighborhood voted to pixel values.
(4) To implement the system, and make verification contrast for multiplealgorithm.
引文
[1] Hiroko Kato and Keng T．Tan，Pervasive 2D Barcodes for Camera PhoneApplications．IEEE Computer Society，2007．
[2] S．Yahyanejad and J．Strom．Removing motion blur from barcode images．InProc．IEEE International Workshop on Mobile Vision (in Conjunction withCVPR’2010)，2010．
[3]薛红．条码技术及商业自动化系统-条码技术[M]．北京：中国轻工业出版社，2010．
[4]周晓伟．二维条码识别技术研究[P]．上海：上海交通大学，2007．
[5]蒋长兵，白丽君．物流自动化识别技术[M]．北京：中国物资出版社，2009．
[6] WeiXu．2D Barcode Localization and Motion Deblurring Using a Flutter ShutterCamera．IEEE，2010．
[7] Y．Ding，S．McCloskey，and J．Yu．Analysis of motion blur with a flutter shuttercamera for non-linear motion．In Proc．11th European Conference on ComputerVision，2010．
[8] R．Fergus，B．Singh，A．Hertzmann，et al．Removing camera shake from asingle photograph．ACM Transaction on Graphics，2006，25(3)：787–794．
[9]中国物品编码中心，中国标准出版社第四编辑室．物流标准汇编．物流信息分类编码卷[M]．北京：中国标准出版社．2010．
[10] L．A．F．Fernandes and M．M．Oliveira．Real-time line detection through animproved hough transform voting scheme．Pattern Recognition，2008，41(1)：299–314．
[11] H．Hu，W．Xu，and Q．Huang．A 2D barcode extraction method based on texturedirection analysis．In Proc．Fifth Int’l Conf．on Image and Graphics，2009，759–762．
[12] H．Kato and K．T．Tan，2D Barcodes forMobile Phones，Proc．2nd Int’lConf．Mobile Technology，Applications and Systems (MTAS 05)，IEE，2005，1–4．
[13]李雅静．数据矩阵二维条码图像识别的算法研究与实现[P]．北京：北京交通大学，2009．
[14] Rafael C．Gonzalez，Richard E．Woods．Digital Image Processing SecondEdition[M]．Beijing：Publishing House of Electronics Industry．2010．
[15] Wu X S，Qiao L Z，Deng J．A new method for bar code localization andrecognition[A]．Proceedings of the 2009 2nd International Congress on Imageand Signal Processing(CISP)[C]．Tianjin：IEEE，2009，(2)：901-906．
[16]赖忠喜．数据矩阵二维条形码技术及在邮政上的应用研究[P]．浙江：浙江工业大学，2009．
[17]徐福泽．二维条码识别系统的图像处理[P]．山东：烟台大学，2008．
[18]万蒨．二维条码的编解码及系统实现[P]．上海：上海交通大学，2007．
[19] Automatic Identification and Data Capture Techniques－数据矩阵Bar CodeSymbology Specification，ISO/IEC 16022，Int’l Organization forStandardization，2006．
[20] K．Q．Wang．Barcode reading from images captured by camera phones，IEEMobility Conference，2005．
[21]乔连芝．基于图像处理方式定位条形码[P]．广州：华南理工大学，2010．
[22]刘大伟．基于边缘检测的图像去噪方法在二维条码识别中的研究与应用
[D]．青岛：中国海洋大学，2008．
[23] D．Chai and F．Hock．Locating and Decoding EAN-13 Barcodes from ImagesCaptured by Digital Cameras[C]．Fifth International Conference on Information，Communucation and Signal Processing(ICICS’2005)，Bangkok，Thailand，2005，1556-1560．
[24]吴海彬，何祖恩，李文锦．基于改进Sobel算子的图像边缘检测及其嵌入式实现[J]．机床与液压，2009，37(10)：133-136．
[25]袁春兰，熊宗龙，周雪花，彭小辉．基于Sobel算子的图像边缘检测研究[J]．激光与红外，2009，39(1)：85-87．
[26] I．Sobel．Camera Models and Machine Perception Stanford AI，1970．
[27]卢惠民，郑志强．一种新的用于检测直线的快速Hough变换[J]．计算机应用，2005，25(10)：2379-2380．
[28] FengXiao Jun，Gormish M J．Locating barcodes using JPEG2000 compresseddata [J]．Processing of the SPIE，IEEE，2005，5960：908-916．
[29] Hsia S C，Chen M H．A cost-effective line-based light-balancing technique usingadaptive processing[J]．IEEE Transactions on Image Processing，2006，15(9)：2719-2729．
[30]高守传，姚领田．Visual C++实践与提高—数字图像处理与工程应用篇
［M］．北京：中国铁道出版社，2006．
[31] R．Adelmann，M．Langheinrich，C．Florkemeier．A Toolkit for Bar-CodeRecognition and Resolving on Camera Phones—Jump Starting the Internet ofThings．Proc．Informatik Workshop Mobile and Embedded Interactive Systems，2006．
[32] H．Ji and C．Liu．Motion blur identification from image gradients．In Proc．IEEEConf．on Computer Vision and Pattern Recognition，2008，1–8．
[33] R．Liu，Z．Li，and J．Jia．Image partial blur detection and classification．InProc．IEEE Conf．on Computer Vision and Pattern Recognition，2008，1–8．
[34] X．Lu，G．Fan，Y．Wang．A robust barcode reading method based on imageanalysis of a hierarchical feature classification．In Proc．2006 IEEE/RSJ Int’lConf．on Intelligent Robots and Systems，2006，3358–3362．
[35] S．McCloskey．Velocity-dependent shutter sequences for motion deblurring．InProc．11th European Conference on Computer Vision，2010．
[36] D．Parikh and G．Jancke．Localization and segmentation of a 2D high capacitycolor barcode．In Proc．2008 IEEE Workshop on Applications of ComputerVision，2008，1–6．
[37] R．Raskar，A．Agrawal，and J．Tumblin．Coded exposure photography：motiondeblurring using fluttered shutter．ACM Transaction on Graphics，2006，25(3)：795–804．
[38] C．Zhang，J．Wang，M．Y．Shi Han，Z．Zhang．Automatic real-time barcodelocalization in complex scenes．In Proc．2006 IEEE Int’l Conf．on ImageProcessing，2006，497–500．
[2] S．Yahyanejad and J．Strom．Removing motion blur from barcode images．InProc．IEEE International Workshop on Mobile Vision (in Conjunction withCVPR’2010)，2010．
[3]薛红．条码技术及商业自动化系统-条码技术[M]．北京：中国轻工业出版社，2010．
[4]周晓伟．二维条码识别技术研究[P]．上海：上海交通大学，2007．
[5]蒋长兵，白丽君．物流自动化识别技术[M]．北京：中国物资出版社，2009．
[6] WeiXu．2D Barcode Localization and Motion Deblurring Using a Flutter ShutterCamera．IEEE，2010．
[7] Y．Ding，S．McCloskey，and J．Yu．Analysis of motion blur with a flutter shuttercamera for non-linear motion．In Proc．11th European Conference on ComputerVision，2010．
[8] R．Fergus，B．Singh，A．Hertzmann，et al．Removing camera shake from asingle photograph．ACM Transaction on Graphics，2006，25(3)：787–794．
[9]中国物品编码中心，中国标准出版社第四编辑室．物流标准汇编．物流信息分类编码卷[M]．北京：中国标准出版社．2010．
[10] L．A．F．Fernandes and M．M．Oliveira．Real-time line detection through animproved hough transform voting scheme．Pattern Recognition，2008，41(1)：299–314．
[11] H．Hu，W．Xu，and Q．Huang．A 2D barcode extraction method based on texturedirection analysis．In Proc．Fifth Int’l Conf．on Image and Graphics，2009，759–762．
[12] H．Kato and K．T．Tan，2D Barcodes forMobile Phones，Proc．2nd Int’lConf．Mobile Technology，Applications and Systems (MTAS 05)，IEE，2005，1–4．
[13]李雅静．数据矩阵二维条码图像识别的算法研究与实现[P]．北京：北京交通大学，2009．
[14] Rafael C．Gonzalez，Richard E．Woods．Digital Image Processing SecondEdition[M]．Beijing：Publishing House of Electronics Industry．2010．
[15] Wu X S，Qiao L Z，Deng J．A new method for bar code localization andrecognition[A]．Proceedings of the 2009 2nd International Congress on Imageand Signal Processing(CISP)[C]．Tianjin：IEEE，2009，(2)：901-906．
[16]赖忠喜．数据矩阵二维条形码技术及在邮政上的应用研究[P]．浙江：浙江工业大学，2009．
[17]徐福泽．二维条码识别系统的图像处理[P]．山东：烟台大学，2008．
[18]万蒨．二维条码的编解码及系统实现[P]．上海：上海交通大学，2007．
[19] Automatic Identification and Data Capture Techniques－数据矩阵Bar CodeSymbology Specification，ISO/IEC 16022，Int’l Organization forStandardization，2006．
[20] K．Q．Wang．Barcode reading from images captured by camera phones，IEEMobility Conference，2005．
[21]乔连芝．基于图像处理方式定位条形码[P]．广州：华南理工大学，2010．
[22]刘大伟．基于边缘检测的图像去噪方法在二维条码识别中的研究与应用
[D]．青岛：中国海洋大学，2008．
[23] D．Chai and F．Hock．Locating and Decoding EAN-13 Barcodes from ImagesCaptured by Digital Cameras[C]．Fifth International Conference on Information，Communucation and Signal Processing(ICICS’2005)，Bangkok，Thailand，2005，1556-1560．
[24]吴海彬，何祖恩，李文锦．基于改进Sobel算子的图像边缘检测及其嵌入式实现[J]．机床与液压，2009，37(10)：133-136．
[25]袁春兰，熊宗龙，周雪花，彭小辉．基于Sobel算子的图像边缘检测研究[J]．激光与红外，2009，39(1)：85-87．
[26] I．Sobel．Camera Models and Machine Perception Stanford AI，1970．
[27]卢惠民，郑志强．一种新的用于检测直线的快速Hough变换[J]．计算机应用，2005，25(10)：2379-2380．
[28] FengXiao Jun，Gormish M J．Locating barcodes using JPEG2000 compresseddata [J]．Processing of the SPIE，IEEE，2005，5960：908-916．
[29] Hsia S C，Chen M H．A cost-effective line-based light-balancing technique usingadaptive processing[J]．IEEE Transactions on Image Processing，2006，15(9)：2719-2729．
[30]高守传，姚领田．Visual C++实践与提高—数字图像处理与工程应用篇
［M］．北京：中国铁道出版社，2006．
[31] R．Adelmann，M．Langheinrich，C．Florkemeier．A Toolkit for Bar-CodeRecognition and Resolving on Camera Phones—Jump Starting the Internet ofThings．Proc．Informatik Workshop Mobile and Embedded Interactive Systems，2006．
[32] H．Ji and C．Liu．Motion blur identification from image gradients．In Proc．IEEEConf．on Computer Vision and Pattern Recognition，2008，1–8．
[33] R．Liu，Z．Li，and J．Jia．Image partial blur detection and classification．InProc．IEEE Conf．on Computer Vision and Pattern Recognition，2008，1–8．
[34] X．Lu，G．Fan，Y．Wang．A robust barcode reading method based on imageanalysis of a hierarchical feature classification．In Proc．2006 IEEE/RSJ Int’lConf．on Intelligent Robots and Systems，2006，3358–3362．
[35] S．McCloskey．Velocity-dependent shutter sequences for motion deblurring．InProc．11th European Conference on Computer Vision，2010．
[36] D．Parikh and G．Jancke．Localization and segmentation of a 2D high capacitycolor barcode．In Proc．2008 IEEE Workshop on Applications of ComputerVision，2008，1–6．
[37] R．Raskar，A．Agrawal，and J．Tumblin．Coded exposure photography：motiondeblurring using fluttered shutter．ACM Transaction on Graphics，2006，25(3)：795–804．
[38] C．Zhang，J．Wang，M．Y．Shi Han，Z．Zhang．Automatic real-time barcodelocalization in complex scenes．In Proc．2006 IEEE Int’l Conf．on ImageProcessing，2006，497–500．