|本期目录/Table of Contents|

[1]张苏沛,刘 军*,肖澳文,等.基于卷积神经网络的验证码识别[J].武汉工程大学学报,2019,(01):89-92.[doi:10. 3969/j. issn. 1674?2869. 2019. 01. 015]
 ZHANG Supei,LIU Jun*,XIAO Aowen,et al.CAPTCHA Recognition Based on Convolutional Neural Network[J].Journal of Wuhan Institute of Technology,2019,(01):89-92.[doi:10. 3969/j. issn. 1674?2869. 2019. 01. 015]
点击复制

基于卷积神经网络的验证码识别(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
期数:
2019年01期
页码:
89-92
栏目:
机电与信息工程
出版日期:
2019-03-23

文章信息/Info

Title:
CAPTCHA Recognition Based on Convolutional Neural Network
文章编号:
20190115
作者:
张苏沛12刘 军*12肖澳文12杜 壮12
1. 智能机器人湖北省重点实验室(武汉工程大学),湖北 武汉 430205;2. 武汉工程大学计算机科学与工程学院,湖北 武汉 430205
Author(s):
ZHANG Supei12LIU Jun*12XIAO Aowen12DU Zhuang12
1. Hubei Key Laboratory of Intelligent Robot(Wuhan Institute of Technology), Wuhan 430205, China;2. School of Computer Science & Engineering, Wuhan Institute of Technology, Wuhan 430205, China
关键词:
验证码卷积神经网络字符识别学习率
Keywords:
captcha convolutional neural network character recognition learning rate
分类号:
TP317.4
DOI:
10. 3969/j. issn. 1674?2869. 2019. 01. 015
文献标志码:
A
摘要:
针对传统验证码识别受字符分割限制的问题,将卷积神经网络应用到验证码的特征分析和识别中。使用验证码图像整体作为输入,对传统的LeNet-5的网络结构进行改进,构建一种端到端的卷积神经网络对图像由低级到高级逐层提取图像特征,选取ReLU作为激活函数,实现对验证码的识别。实验过程中设置对照组,研究不同因素对识别准确率的影响。测试结果显示,该模型能够进行端到端的识别,避免了字符分割方法流程过多导致的不足,在测试集上达到99%的识别率。结果表明训练次数的增加以及学习率的优化有助于提高卷积神经网络的准确率。
Abstract:
Aiming at the limitations of character segmentation in traditional completely automated public turing test to tell computers and humans apart (CAPTCHA) recognition we proposed an end-to-end convolutional neural network to characterize and identify CAPTCHAs. Firstly, a whole CAPTCHA image was used as an input, and then the convolutional neural network based on LeNet-5 was constructed to extract image features layer by layer from low-level to high-level. Finally, the ReLU function was selected as activation function to perform recognition task of CAPTCHA image. To study the effect of different factors on the recognition accuracy, a control group was provided in the experiments. The testing results show that the proposed methos realized the end-to-end recognition, thus avoiding the insufficiency caused by too many processes of character segmentation method and achieving 99% recognition rate on the test set. It is found that the increase of training times and the optimization of learning rate could improve the accuracy of convolutional neural network.

参考文献/References:

[1] 王斌君, 王靖亚, 杜凯选,等. 验证码技术的攻防对策研究[J]. 计算机应用研究,2013,30(9):2776-2779. [2] 文晓阳, 高能, 夏鲁宁,等. 高效的验证码识别技术与验证码分类思想[J]. 计算机工程, 2009, 35(8):186-188. [3] 杨雄. 基于Python语言和支持向量机的字符验证码识别[J]. 数字技术与应用,2017(4):72-74. [4] 高海昌,樊晔,王伟. 利用旋转归一化和粗匹配算法破解验证码[J]. 西安电子科技大学学报,2012,39(6):78-83. [5] 秦实宏,叶云丽. 复杂光照下的车牌定位方法[J]. 武汉工程大学学报,2015,37(11):69-73. [6] 周志华, 陈世福. 神经网络集成[J]. 计算机学报, 2002, 25(1):1-8. [7] LECUN Y, BOSERB, DENKER J S, et al. Handwritten digit recognition with a back-propagation network[C]// Advances in neural information processing systems. Denver:Morgan Kaufmann Publishers, 1990:396-404. [8] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998,86(11): 2278-2324. [9] CORTES C, VAPNIK V. Support-vector networks[J]. Machine Learning, 1995, 20(3):273-297. [10] 汪家明,卢涛. 多尺度残差深度神经网络的卫星图像超分辨率算法[J]. 武汉工程大学学报,2018,40(4):440-445. [11] WU S, ZHONG S, LIU Y. Deep residual learning for image steganalysis[J]. Multimedia Tools and Applications,2018, 77(9):10437-10453.[12] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[J]. ?IEEE? Conference? on? Computer Vision?and?Pattern?Recognition,2014(9):1-9.[13] 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, 2017,39(11): 2298-2304.[14] 洪汉玉, 王澍, 朱浩,等. 低对比度嵌入型钢坯字符识别方法[J]. 武汉工程大学学报, 2012, 34(12):38-43. [15] 曲之琳,胡晓飞. 基于改进激活函数的卷积神经网络研究[J]. 计算机技术与发展,2017,27(12):77-80.[16] 田娟,李英祥,李彤岩. 激活函数在卷积神经网络中的对比研究[J]. 计算机系统应用,2018,27(7):43-49.

相似文献/References:

[1]汪家明,卢 涛*.多尺度残差深度神经网络的卫星图像超分辨率算法[J].武汉工程大学学报,2018,40(04):440.[doi:10. 3969/j. issn. 1674?2869. 2018. 04. 018]
 WANG Jiaming,LU Tao *. Satellite Imagery Super-Resolution Algorithm via Multi-Scale Residual Deep Neural Network[J].Journal of Wuhan Institute of Technology,2018,40(01):440.[doi:10. 3969/j. issn. 1674?2869. 2018. 04. 018]
[2]肖澳文,刘 军*,张苏沛,等.基于CNN的三维人体姿态估计方法[J].武汉工程大学学报,2019,(02):168.[doi:10. 3969/j. issn. 1674?2869. 2019. 02. 013]
 XIAO Aowen,LIU Jun*,ZHANG Supei,et al.Three-Dimensional Human Pose Estimation Based on Convolution Neural Network[J].Journal of Wuhan Institute of Technology,2019,(01):168.[doi:10. 3969/j. issn. 1674?2869. 2019. 02. 013]
[3]陈希彤,卢 涛*.基于全局深度分离卷积残差网络的高效人脸识别算法[J].武汉工程大学学报,2019,(03):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]
 CHEN Xitong,LU Tao *.Efficient Face Recognition Algorithm Using Global Deep Separable Convolutional and Residual Network[J].Journal of Wuhan Institute of Technology,2019,(01):276.[doi:10. 3969/j. issn. 1674-2869. 2019. 03. 014]
[4]王丽亚,刘昌辉*,蔡敦波,等.基于CNN-BiLSTM网络引入注意力模型的文本情感分析[J].武汉工程大学学报,2019,(04):386.[doi:10. 3969/j. issn. 1674?2869. 2019. 04. 016]
 WANG Liya,LIU Changhui*,CAI Dunbo,et al.Text Sentiment Analysis Based on CNN-BiLSTM Network and Attention Model[J].Journal of Wuhan Institute of Technology,2019,(01):386.[doi:10. 3969/j. issn. 1674?2869. 2019. 04. 016]
[5]熊寒颖,鲁统伟*,闵 峰,等.基于单一神经网络的实时人脸检测[J].武汉工程大学学报,2019,(05):489.[doi:10. 3969/j. issn. 1674?2869. 2019. 05. 015]
 XIONG Hanying,LU Tongwei*,MIN Feng,et al.Real-Time Face Detection Based on Single Neural Network[J].Journal of Wuhan Institute of Technology,2019,(01):489.[doi:10. 3969/j. issn. 1674?2869. 2019. 05. 015]
[6]杜梦星,王彦伟*.基于CNN的突发事件预警系统的设计与实现[J].武汉工程大学学报,2020,42(02):207.[doi:10.19843/j.cnki.CN42-1779/TQ.201910016]
 DU Mengxing,WANG Yanwei*.Design and Implementation of Emergency Warning System Based on Convolution Neural Network[J].Journal of Wuhan Institute of Technology,2020,42(01):207.[doi:10.19843/j.cnki.CN42-1779/TQ.201910016]
[7]江满星,赵彤洲*,吴泽俊.基于目标形状卷积神经网络在舰船分类中的应用[J].武汉工程大学学报,2020,42(02):213.[doi:10.19843/j.cnki.CN42-1779/TQ.201911022]
 JIANG Manxing,ZHAO Tongzhou*,WU Zejun.Application of Convolution Neural Network Based on Target Shape in Ships and Warships Classification[J].Journal of Wuhan Institute of Technology,2020,42(01):213.[doi:10.19843/j.cnki.CN42-1779/TQ.201911022]

备注/Memo

备注/Memo:
收稿日期:2018-07-20基金项目:智能机器人湖北省重点实验室开放基金(HBIR 201802);武汉工程大学第十届研究生教育创新基金作者简介:张苏沛,硕士研究生。E-mail:zhangsupei@wit.edu.cn*通信作者:刘 军,博士,副教授。E-mail:liujun@wit.edu.cn引文格式:张苏沛,刘军,肖澳文,等. 基于卷积神经网络的验证码识别[J]. 武汉工程大学学报,2019,41(1):89-92.
更新日期/Last Update: 2019-02-19