摘要
近年来,二维码因其包含更多的信息存储成为移动终端上较为普遍的编码方式,广泛应用于手机支付、公共交通等场景。在带来高效的信息交互同时,现有的二维码识别技术有一定的局限性,例如只能识别黑白、高分辨率等特定的二维码。深度学习具有很强的表征能力,采用Mask R-CNN算法,对多种类别的二维码进行归类分析,实现基于深度卷积网络的端到端的二维码定位与检测系统,在制作标记相应数据集并训练后,实验获得较好的效果。
Recently, QR Code has become a more common coding method in mobile terminal because it contains more information storage, which iswidely used in mobile payment, public transport and other scenes. At the same time, the existing QR code detection technology has somelimitations, such as only the identification of black and white, high resolution and other specific QR code. Deep learning has strong charac-terization ability, based on the Mask R-CNN algorithm, classifies and analyzes various kinds of QR codes, makes data sets and trains it, re-alizes the end-to-end QR code location and detection system based on deep convolutional network, in the production of markers corre-sponding data sets and training, the experiment has obtained better results.
引文
[1]He Kaiming,Gkioxari,et al.Mask R-CNN[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,05 June 2018:1-1.
[2]Wen Y,Lu Y,Yan J,et al.An Algorithm for License Plate Recognition Applied to Intelligent Transportation System[J].IEEE Transactions on Intelligent Transportation System,2011,12(3):830-845.
[3]Lingling Tong,Xiaoguang Gu,Feng Dai.QR Code Detection Based on Local Features.ICIMCS,14,July 10-12,2014.
[4]屈德涛,金立左.复杂场景中的QR码检测[J].工业控制计算机,2017(1):70-71.
[5]Viola P,Jones M.Rapid Object Detection Using a Boosted Cascade of Simple Features[C].Computer Vision and Pattern Recognition,2001.CVPR 2001.Proceedings of the 2001 IEEE Computer Society Conference on.IEEE,2001,1:I-511-I-518 vol.1.
[6]Girshick R,Donahue J,Darrell T,et al.Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society.Columbus,IEEE,2014:580-587.
[7]R.Girshick.Fast R-CNN.In ICCV,2015.
[8]S.Ren,K.He,R.Girshick,and J.Sun.Faster R-CNN:Towards real-time Object Detection with Region Proposal Networks.In NIPS,2015.
[9]Redmon J,Divvala S,Girshick R,et al.You Only Look Once:Unified,Real-Time Object Detection[C].Conference on Computer Vision and Pattern Recognition(CVPR).Las Vegas:IEEE,2015:779-788.
[10]J.RedmonandA.Farhadi.Yolo9000:Better,Faster,Stronger.In Computer Vision and Pattern Recognition(CVPR),2017 IEEE Conference on,pages 6517-6525.IEEE,2017.
[11]T.-Y.Lin,M.Maire,S.Belongie,J.Hays,P.Perona,D.Ramanan,P.Dolla?r,C.L.Zitnick.Microsoft coco:Common Objects in Context.In European Conference on Computer Vision,740-755.Springer,2014.
[12]T.-Y.Lin,P.Dollar,R.Girshick,K.He,B.Hariharan,S.Belongie.Feature Pyramid Networks for Object Detection.In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2117-2125,2017.
[13]J.Long,E.Shelhamer,T.Darrell.Fully Convolutional Networks for Semantic Segmentation.In CVPR,2015.