计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240100031-9.doi: 10.11896/jsjkx.240100031
金鑫豪, 池凯凯
JIN Xinhao, CHI Kaikai
摘要： 数字化浪潮下,企业日益依赖机器人流程自动化(Robot Process Automation,RPA)技术来降低成本、提高效率,以保持竞争力。但流程中部分环节面临汉字点选验证码识别的难题,限制了自动化水平的进一步提高。现有研究方案存在数据集制作难度大、模型泛化性能差、模型复杂度与性能之间不平衡等问题。为此,提出一种数据集制作成本低、模型泛化性能好且轻量化的汉字点选验证码识别方法。具体而言:首先采用经过针对性改进的YOLOv8-n显著轻量化汉字检测模型,然后对汉字图片进行分割、矫正等预处理操作,接着采用泛化性强的PaddleOCR模型进行汉字识别,降低了场景迁移的成本,并通过识别概率矩阵得到最佳匹配结果,进一步提高了准确率。此外,设计了一种半自动的汉字检测数据集构建流程并公开了数据集。该研究旨在推动汉字点选验证码的自动识别技术的发展,促进企业流程自动化水平的提升。
中图分类号:
| [1]ENRÍQUEZ J G,JIMÉNEZ-RAMÍREZ A,DOMÍNGUEZ-MAYO F J,et al.Robotic process automation:a scientific and industrial systematic mapping tudy[J].IEEE Access,2020,8:39113-39129. [2]LIANG Y.Research on the Application of Financial RobotProcess Automation Based on Machine Learning[C]//2021 International Conference on Electronic Information Technology and Smart Agriculture(ICEITSA).IEEE,2021:474-477. [3]HANDOKO B L,LINDAWATI A S L,MUSTAPHA M.Robotic process automation in audit 4.0[C]//The 2021 12th International Conference on E-business,Management and Econo-mics.2021:128-132. [4]BRANDSTATTER C,TSCHANDL M,MITTERBACK C.AGeneric Process Model for the Introduction of Robotic Process Automation in Financial Accounting[C]//Proceedings of the 2023 9th International Conference on Computer Technology Applications.2023:12-18. [5]AHMET UNAL M,BOLUKBAS O.The Acquirements of Digitalization with RPA(Robotic Process Automation) Technology in the Vakif Participation Bank[C]//Proceedings of the 4th International Conference on Information Science and Systems.2021:68-73. [6]LIU S,QI X,LI H.Practice of Robot Process Automation in Power Grid Dispatching Report[C]//Proceedings of the 2022 4th International Conference on Robotics,Intelligent Control and Artificial Intelligence.2022:212-216. [7]RATIA M,MYLLÄRNIEMI J,HELANDER N.Roboticprocess automation-creating value by digitalizing work in the private healthcare?[C]//Proceedings of the 22nd International Academic Mindtrek Conference.2018:222-227. [8]SGANDERLA R B,FANTINATO M,THOM L H.RoboticProcess Automation in Latin American Organizations:Survey and Evaluation of the Current State of Technology Adoption[C]//Proceedings of the XIX Brazilian Symposium on Information Systems.2023:459-467. [9]KEDZIORA D,HYRYNSALMI S.Turning Robotic ProcessAutomation onto Intelligent Automation with Machine Learning[C]//Proceedings of the 11th International Conference on Communities and Technologies.2023:1-5. [10]KHOLIYA P S,KAPOOR A,RANA M,et al.Intelligentprocess automation:The future of digital transformation[C]//10th International Conference on System Modeling & Advancement in Research Trends(SMART 2021).IEEE,2021:185-190. [11]ROTHER C,KOLMOGOROV V,BLAKE A.“GrabCut” interactive foreground extraction using iterated graph cuts[J].ACM Transactions on Graphics(TOG),2004,23(3):309-314. [12]DU Y,LI C,GUO R,et al.Pp-ocr:A practical ultra lightweight ocr system[J].arXiv:2009.09941,2020. [13]BOSTIK O,KLECKA J.Recognition of CAPTCHA characters by supervised machine learning algorithms[J].IFAC-Papers OnLine,2018,51(6):208-213. [14]SACHDEV S.Breaking captcha characters using multi-tasklearning cnn and svm[C]//4th International Conference on Computational Intelligence and Networks(CINE 2020).IEEE,2020:1-6. [15]ZHANG N,EBRAHIMI M,LI W,et al.Counteracting darkWeb text-based CAPTCHA with generative adversarial learning for proactive cyber threat intelligence[J].ACM Transactions on Management Information Systems(TMIS),2022,13(2):1-21. [16]THOBHANI A,GAO M,HAWBANI A,et al.CAPTCHA re-cognition using deep learning with attached binary images[J].Electronics,2020,9(9):1522. [17]WU X,DAI S,GUO Y,et al.A machine learning attack againstvariable-length Chinese character CAPTCHAs[J].Applied Intelligence,2019,49:1548-1565. [18]LUAN S,CHEN C,ZHANG B,et al.Gabor convolutional net-works[J].IEEE Transactions on Image Processing,2018,27(9):4357-4366. [19]WANG J,QIN J,XIANG X,et al.CAPTCHA recognition based on deep convolutional neural network[J].Math.Biosci.Eng,2019,16(5):5851-5861. [20]ZHANG X,LIU X,SARKODIE-GYAN T,et al.Development of a character CAPTCHA recognition system for the visually impaired community using deep learning[J].Machine Vision and Applications,2021,32:1-19. [21]BI X,LIU X.Chinese Character Captcha Sequential SelectionSystem Based on Convolutional Neural Network[C]//International Conference on Computer Vision,Image and Deep Lear-ning(CVIDL 2020).IEEE,2020:554-559. [22]GIRSHICK R.Fast r-cnn[C]//Proceedings of the IEEE International Conference on Computer Vision.2015:1440-1448. [23]CAVNAR W B,TRENKLEJ M.N-gram-based text categorization[C]//3rd Annual Symposium on Document Analysis and Information Retrieval(SDAIR-94).1994:161175. [24]HU J.Research on Security of Chinese Point-and-Click CAP-TCHA[D].Xi'an:Xidian University,2018. [25]YOU X.Research on Chinese Character Captcha RecognitionBased on YOLO V2 [D].Chengdu:Chengdu University of Technology,2019. [26]REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:779-788. [27]LIU W,ANGUELOV D,ERHAND,et al.Ssd:Single shotmultibox detector[C]//Computer Vision－ECCV 2016:14th European Conference,Amsterdam,The Netherlands,Part I 14.Springer International Publishing,2016:21-37. [28]WANG C Y,BOCHKOVSKIY A,LIAOH Y M.YOLOv7:Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of the IEEE/CVF Confe-rence on Computer Vision and Pattern Recognition.2023:7464-7475. [29]GE Z,LIU S,WANG F,et al.Yolox:Exceeding yolo series in 2021[J].arXiv:2107.08430,2021. [30]XU S,WANG X,LV W,et al.PP-YOLOE:An evolved version of YOLO[J].arXiv:2203.16250,2022. [31]LI C,LI L,JIANG H,et al.YOLOv6:A single-stage object detection framework for industrial applications[J].arXiv:2209.02976,2022. [32]CHOLLET F.Xception:Deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:1251-1258. [33]HOWARD A G,ZHU M,CHEN B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[J].arXiv:1704.04861,2017. [34]ZHANG X,ZHOU X,LIN M,et al.Shufflenet:An extremelyefficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2018:6848-6856. [35]KOONCE B.EfficientNet[M]//Convolutional Neural Networks with Swift for Tensorflow:Image Recognition and Dataset Categorization.2021:109-123. [36]HAN K,WANG Y,TIAN Q,et al.Ghostnet:More featuresfrom cheap operations[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:1580-1589. [37]REYNOLDSD A.Gaussian mixture models[J].Encyclopedia of Biometrics,2009,741:659-663. [38]ZHOU X,YAO C,WEN H,et al.East:an efficient and accurate scene text detector[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:5551-5560. [39]SHI B,BAI X,YAO C.An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(11):2298-2304. |
|
||
首页