低分辨率下纸币的识别与污损检测
|
摘要
纸币图像识别是近年来在模式识别领域较为活跃的一个课题,而且有着很广阔的应用前景,由此技术研制的纸币清分机正在银行等金融系统发挥着越来越大的作用。在进行纸币图像识别时,由于采集到的图像是在低分辨率下扫描得到的,如何在这种低分辨率条件下实现对纸币进行识别,并且识别速度要满足清分机的要求,是纸币识别的核心问题。本文通过对纸币特征进行分析,对纸币的识别过程做了详细研究。在图像的预处理中,由于系统对纸币处理实时性的要求,该过程必须在很短时间内完成,因此我们对图像的边界检测,撇弃了诸如图像二值化,Hough变换等处理手段,而是直接采用分段线性拟合的方法确定图像中纸币的四个边界,以完成纸币的定位和倾斜矫正。在特征提取中,我们对基于方向块的特征提取方法进行了分析,在此基础上针对美元特点,对图像方向块的划分方式做了研究,并提出了基于几何距离的特征提取方法;在分类器设计中,我们采用了LVQ网络对纸币进行学习与分类,并提出了一种具有两层结构的分类算法,第一层首先对输入的特征向量进行粗分类,选定与特征向量匹配距离最大的两类币种,进入第二层分类器;在第二层分类器中,我们通过研究进入该模块两类币种特征块的相关性,重新设计了特征向量,同时对分类器进行改进,最终实现对纸币的分类。在纸币的污损检测方面,提出了一种基于均匀性特征的纸币污损检测方法,首先利用均匀性特征判定待检纸币上可能存在污损的区域,然后利用标准图像进行图像配准,确定这些区域在参考图像上的对应位置,最后逐像素进行比较,最终判定待检图像的污损状况。
Paper currency recognition is a popular issue in recent years, because it has a great future in the market. The paper currency sorter developed by designer plays an important role in bank area. Due to the reason of low resolution of paper currency image and high speed of paper currency sorter, so how to recognize paper currency under these conditions is a key problem. Then in pre-processing stage, for the allowance of the real time demand, we dispose the method of Hough transform and apply linear fitting to determine the border of paper currency to achieve localization and adjustment directly. By the analysis of based directional block feature extraction, we propose a new method of feature extraction based on geometry distance. In the design of classifier, we use the LVQ network to classify the feature and introduce a classifier with two layers. First classifier chooses two classes whose matching distance between it and paper currency is bigger than others from all class. Then in the second stage, we extract some new feature and improve the classifier to generate the last result. In the stage of defect detection for paper currency, we advances a homogeneity based algorithm for the detection of scratch and cracks appearing on paper currency, in which the homogeneity feature of the sensed paper currency image is first constructed to locate the pixels that probably been polluted, the image registration algorithm is subsequently used to overlay the sensed and reference paper currency image. At last, each probably polluted pixel on the sensed image is compared with its corresponding pixel on the reference image to estimate the contamination level
Paper currency recognition is a popular issue in recent years, because it has a great future in the market. The paper currency sorter developed by designer plays an important role in bank area. Due to the reason of low resolution of paper currency image and high speed of paper currency sorter, so how to recognize paper currency under these conditions is a key problem. Then in pre-processing stage, for the allowance of the real time demand, we dispose the method of Hough transform and apply linear fitting to determine the border of paper currency to achieve localization and adjustment directly. By the analysis of based directional block feature extraction, we propose a new method of feature extraction based on geometry distance. In the design of classifier, we use the LVQ network to classify the feature and introduce a classifier with two layers. First classifier chooses two classes whose matching distance between it and paper currency is bigger than others from all class. Then in the second stage, we extract some new feature and improve the classifier to generate the last result. In the stage of defect detection for paper currency, we advances a homogeneity based algorithm for the detection of scratch and cracks appearing on paper currency, in which the homogeneity feature of the sensed paper currency image is first constructed to locate the pixels that probably been polluted, the image registration algorithm is subsequently used to overlay the sensed and reference paper currency image. At last, each probably polluted pixel on the sensed image is compared with its corresponding pixel on the reference image to estimate the contamination level
引文
1 F. Takeda, S. Omatu, S. Onami, T. Kadono, and K. Terada. A Paper Currency Recognition Method by A Small Size Neural Network with Optimized Masks by GA. In World Congress on Computational Intelligence, (Orlando, Florida), pp. 4243 --4246, June 26 - July 2 1994
2 M. Tanaka. Modeling of Mixtures of Principal Component Analysis Model with Genetic Algorithm Proceeding 31st International Symposium on Stochastic Systems Theory and Its Application, pp. 157-162, Yokohama, Japan, 1999
3 崔国勤 等. 基于支持向量机的人脸识别方法,计算机科学,2003 Vol.30 No.4
4 孙即祥. 模式识别中的特征提取与计算机视觉不变量,2001 年 7 月
5 戴葵. 神经网络实现技术,国防科技大学出版社 1998 年,第一版
6 A.Ahmadi, S.Omatu. A High Reliability Method for Classification of Paper Currency Based on Neural Networks Proceeding of The Eighth International Symposium on Artificial Life and Robotics, Vol.2, pp.601-604, Oita, Japan, 2003
7 Sonehara. An application of neural network to image data. Trans. Inst.Syst, Control and Info.Eng, Vol.35, No.1, pp.11-18, 1991,in Japanese
8 S.Amari. The Present and Future of Neural Network. Kyoritsu Press, 1990, in Japanese
9 李继飞. 神经网络模式识别技术在自动售货机中的应用研究,湖南大学硕士学位论文,2001
10 娄晓东. 智能纸币清分机系统设计及其实现策略研究,浙江大学工学硕士论文,2003
11 Aauro Barini, Rossana Gualtieri. A new possibalistic clustering algorithm for line detection in real world image. Pattern Recognition, Vol.32, No.11, Nov.1999
12 J.Illing Worth,J.Kittler. A Survey of the Hough Transform. Computer Vision and Image Process, 44(1988) 87-116
13 V.F. Leavers. Which Hough Transform. Compute Vision and Image Process, 58 (1993)
14 崔明星. 纸币图像识别系统在清分机中的应用,哈尔滨工业大学硕士学位论文,2001
15 刘家锋,刘松波,唐降龙.一种实时纸币识别方法的研究,计算机研究与发展,第 40 卷第 7 期
16 F.Takeda,S.Omatu,T.Inoue,S.Onami, K.Konishi. High speed conveyed bill money recognition with neural network. Trans. Inst. Elect. Eng. Japan, Vol.112-C, no.4, pp. 101 –110, 1992,in Japanese
17 郭娟,林东,戚文雅.基于加权 Fisher 准则的先行鉴别分析与人脸识别 计算机应用 第 26 卷第 5 期
18 边肇琪 等. 模式识别.清华大学出版社,2000 年第二版
19 Fumiaki Takeda, Sigeru Omatu. High speed paper currency recognition by neural networks. IEEE Trans on Neural Network ,1995 , 6 (1) : 73～77
20 HengDa Cheng, Ying Sun. A hierarchical approach to color image segmentation using homogeneity, IEEE Trans on Image Processing, 2000, 9(12):2071~2082
21 Image registration methods: a survey, Barbara Zitová, Jan Flusser. Image and Vision Computing, 2003, 21(11):977–1000
22 吴浩扬 等. 遗传算法的一种特例--正交试验设计法,软件学报, Vol.12,No.1, 2001
23 漆远 等. 基于遗传算法的指纹图匹配及应用.软件学报,2000,11(4):488-493
24 杨启文 等. 遗传算法优化速度的改进,软件学报,Vol.12,No.2, 2001
25 Xi,Yu-geng , Cai Tian-you, Yun Wei-min. Survey on genetic algorithm. Control Theory and Application, 1996,13(6): 696-704 (in Chinese)
26 P. Moerland. A Comparison of Mixture Models for Density Estimation Artifical Neural Networks, 7-10 September 1999, Conference Publication No. 470 IEE
27 F.Takeda, Sigeru Omatu. High Speed Paper Currency Recognition by Neural Networks. IEEE Trans Neural Network, Vol6, No 1,Jan 1995
28 Frosini, M. Gori, P. Priami. A Neural Network-Based Model for Paper Currency Recognition and Verification. IEEE Transactions on Neural Networks 7(6), 1482-1490, 1996
29 R. O. Duda, P. E. Hart and D. G. Stork, Pattern Classification John Wiley & Sons, INC. 2001
30 M.Moerland. A Comparision of Mixture Models for Density Estimation Artificial Neural Networks, 7-10 Septembe 1999, Conference Publication No.470 IEE 1999
31 张远鹏,董海,周文灵. 计算机图像处理技术基础, 北京大学出版社,1996:253-256
32 章毓晋. 图像工程(上册) 图象处理和分析, 北京:清华大学出版社, 1999,145-160
33 F. Takeda, S. Omatu and S. Onami, Mask Optimization by Genetic Algorithm for A Neuro-Pattern Recognition Machine with Masks, Trans. Inst. of Syst. Contr. and Inform. Engineers, 8(5): 196-203, 1999
34 F. Takeda, S. Omatu. Bank Note Recognition System Using Neural Network with Random Masks Proceedings of the World Congress on Neural Networks, Vol.1, pp.241-244, Portland, USA, 1993
35 Holland, J.H. Adaptation in Natural and Artificial Systems. Ann Arbor, MI: University Michigan Press, 1975
36 F. Takeda, S. Omatu, S. Onami, T. Kadono, and K. Terada. A Paper Currency Recognition Method by A Small Size Neural Network with Optimized Masks by GA. In World Congress on Computational Intelligence, (Orlando, Florida), pp. 4243 --4246, June 26 - July 2 199
37 Guo Zhuo-Feng, Design of a Real Time Image Processing System Based on DSP. Optical Technique. 2004, 30(1):2004
38 Zhou Yu-Bin , Zhao Yue-Jin, Liu Ming, Wang Xu-Yu. Real-time Digital Image Stabilization System Based on C6711 DSP. Transaction of Beijing Institute of Technology. 2004, 24(5): 435-438
39 徐婉莹, 刘建军, 黄新生. 基于 CPLD 和 DSP 的高速图像采集技术研究. 电子工程师. 2004, 06
40 Ueda Hirotada, Kato Kanji, Matsushima Hitoshi, Kaneko Kenji, Ejiri Masakazu. Multiprocessor System Utilizing Enhanced DSP's for Image Processing. Proc Int Conf on Systolic Arrays. 1988:611-620
41 F. Takeda, S. Omatu, T. Inoue, and S. Onami. A Structure Reduction of Neural Network with Random Masks and Bill Money Recognition Proceedings of the 2nd International Conference on Fuzzy Logic and Neural Networks, Japan, Vol.2, 1999
42 Liao Chengbin, Zhang Jian. Real-time Night Vision Image Processing Based on TMS320C6201 DSP Hardware and Software. Proceedings of the Second International Symposium on Instrumentation Science and Technology, Jinan, China, 2002. Harbin Institute of Technology. 3(3):490-494
2 M. Tanaka. Modeling of Mixtures of Principal Component Analysis Model with Genetic Algorithm Proceeding 31st International Symposium on Stochastic Systems Theory and Its Application, pp. 157-162, Yokohama, Japan, 1999
3 崔国勤 等. 基于支持向量机的人脸识别方法,计算机科学,2003 Vol.30 No.4
4 孙即祥. 模式识别中的特征提取与计算机视觉不变量,2001 年 7 月
5 戴葵. 神经网络实现技术,国防科技大学出版社 1998 年,第一版
6 A.Ahmadi, S.Omatu. A High Reliability Method for Classification of Paper Currency Based on Neural Networks Proceeding of The Eighth International Symposium on Artificial Life and Robotics, Vol.2, pp.601-604, Oita, Japan, 2003
7 Sonehara. An application of neural network to image data. Trans. Inst.Syst, Control and Info.Eng, Vol.35, No.1, pp.11-18, 1991,in Japanese
8 S.Amari. The Present and Future of Neural Network. Kyoritsu Press, 1990, in Japanese
9 李继飞. 神经网络模式识别技术在自动售货机中的应用研究,湖南大学硕士学位论文,2001
10 娄晓东. 智能纸币清分机系统设计及其实现策略研究,浙江大学工学硕士论文,2003
11 Aauro Barini, Rossana Gualtieri. A new possibalistic clustering algorithm for line detection in real world image. Pattern Recognition, Vol.32, No.11, Nov.1999
12 J.Illing Worth,J.Kittler. A Survey of the Hough Transform. Computer Vision and Image Process, 44(1988) 87-116
13 V.F. Leavers. Which Hough Transform. Compute Vision and Image Process, 58 (1993)
14 崔明星. 纸币图像识别系统在清分机中的应用,哈尔滨工业大学硕士学位论文,2001
15 刘家锋,刘松波,唐降龙.一种实时纸币识别方法的研究,计算机研究与发展,第 40 卷第 7 期
16 F.Takeda,S.Omatu,T.Inoue,S.Onami, K.Konishi. High speed conveyed bill money recognition with neural network. Trans. Inst. Elect. Eng. Japan, Vol.112-C, no.4, pp. 101 –110, 1992,in Japanese
17 郭娟,林东,戚文雅.基于加权 Fisher 准则的先行鉴别分析与人脸识别 计算机应用 第 26 卷第 5 期
18 边肇琪 等. 模式识别.清华大学出版社,2000 年第二版
19 Fumiaki Takeda, Sigeru Omatu. High speed paper currency recognition by neural networks. IEEE Trans on Neural Network ,1995 , 6 (1) : 73～77
20 HengDa Cheng, Ying Sun. A hierarchical approach to color image segmentation using homogeneity, IEEE Trans on Image Processing, 2000, 9(12):2071~2082
21 Image registration methods: a survey, Barbara Zitová, Jan Flusser. Image and Vision Computing, 2003, 21(11):977–1000
22 吴浩扬 等. 遗传算法的一种特例--正交试验设计法,软件学报, Vol.12,No.1, 2001
23 漆远 等. 基于遗传算法的指纹图匹配及应用.软件学报,2000,11(4):488-493
24 杨启文 等. 遗传算法优化速度的改进,软件学报,Vol.12,No.2, 2001
25 Xi,Yu-geng , Cai Tian-you, Yun Wei-min. Survey on genetic algorithm. Control Theory and Application, 1996,13(6): 696-704 (in Chinese)
26 P. Moerland. A Comparison of Mixture Models for Density Estimation Artifical Neural Networks, 7-10 September 1999, Conference Publication No. 470 IEE
27 F.Takeda, Sigeru Omatu. High Speed Paper Currency Recognition by Neural Networks. IEEE Trans Neural Network, Vol6, No 1,Jan 1995
28 Frosini, M. Gori, P. Priami. A Neural Network-Based Model for Paper Currency Recognition and Verification. IEEE Transactions on Neural Networks 7(6), 1482-1490, 1996
29 R. O. Duda, P. E. Hart and D. G. Stork, Pattern Classification John Wiley & Sons, INC. 2001
30 M.Moerland. A Comparision of Mixture Models for Density Estimation Artificial Neural Networks, 7-10 Septembe 1999, Conference Publication No.470 IEE 1999
31 张远鹏,董海,周文灵. 计算机图像处理技术基础, 北京大学出版社,1996:253-256
32 章毓晋. 图像工程(上册) 图象处理和分析, 北京:清华大学出版社, 1999,145-160
33 F. Takeda, S. Omatu and S. Onami, Mask Optimization by Genetic Algorithm for A Neuro-Pattern Recognition Machine with Masks, Trans. Inst. of Syst. Contr. and Inform. Engineers, 8(5): 196-203, 1999
34 F. Takeda, S. Omatu. Bank Note Recognition System Using Neural Network with Random Masks Proceedings of the World Congress on Neural Networks, Vol.1, pp.241-244, Portland, USA, 1993
35 Holland, J.H. Adaptation in Natural and Artificial Systems. Ann Arbor, MI: University Michigan Press, 1975
36 F. Takeda, S. Omatu, S. Onami, T. Kadono, and K. Terada. A Paper Currency Recognition Method by A Small Size Neural Network with Optimized Masks by GA. In World Congress on Computational Intelligence, (Orlando, Florida), pp. 4243 --4246, June 26 - July 2 199
37 Guo Zhuo-Feng, Design of a Real Time Image Processing System Based on DSP. Optical Technique. 2004, 30(1):2004
38 Zhou Yu-Bin , Zhao Yue-Jin, Liu Ming, Wang Xu-Yu. Real-time Digital Image Stabilization System Based on C6711 DSP. Transaction of Beijing Institute of Technology. 2004, 24(5): 435-438
39 徐婉莹, 刘建军, 黄新生. 基于 CPLD 和 DSP 的高速图像采集技术研究. 电子工程师. 2004, 06
40 Ueda Hirotada, Kato Kanji, Matsushima Hitoshi, Kaneko Kenji, Ejiri Masakazu. Multiprocessor System Utilizing Enhanced DSP's for Image Processing. Proc Int Conf on Systolic Arrays. 1988:611-620
41 F. Takeda, S. Omatu, T. Inoue, and S. Onami. A Structure Reduction of Neural Network with Random Masks and Bill Money Recognition Proceedings of the 2nd International Conference on Fuzzy Logic and Neural Networks, Japan, Vol.2, 1999
42 Liao Chengbin, Zhang Jian. Real-time Night Vision Image Processing Based on TMS320C6201 DSP Hardware and Software. Proceedings of the Second International Symposium on Instrumentation Science and Technology, Jinan, China, 2002. Harbin Institute of Technology. 3(3):490-494