智能纸币清分机系统设计及其实现策略研究
|
摘要
纸币清分机是近年来得到快速发展的金融光机电一体化设备。论文针对纸
币清分机产品的特点,基于数字图像处理、模式识别、神经网络理论及数字信号
处理等先进理论,结合当今在相应领域的最新发展应用技术,提出了清分机系统
的解决方案,其中包括新旧识别算法和 ARM+DSP 双核的系统平台构架的两个
创新点,并得以使系统高性价比的实现。论文主要涉及如下几个方面:
1、介绍清分机系统图像处理算法设计基础。首先,运用数字图像处理的一
些主要方法对纸币的图像特征进行了分析。在分析的基础上,介绍了清
分机系统中图像处理相关知识,为实际系统算法设计奠定理论的基础。
2、介绍了清分机系统中硬件平台中的关键部分的技术优势及应用领域,为
系统平台的硬件选型提供技术支撑。
3、通过分析清分机系统的特点,提出了软硬件平台解决方案,从系统的选
型和工作流程等方面详细论证了系统方案的实现。首先,从系统硬件平
台的关键部分来阐述系统的选型论证。其次,从清分机系统的工作流程
来阐述系统各个核的系统软件的流程。最后,从清分机系统的实现功能
来具体介绍系统各个功能的图像处理算法设计。
4、系统的试验结果及性能分析。在系统平台的实现方案上,基于重复试验
的基础上,对系统的实现方案的具体细节不断的调整,获得系统最佳的
性价比。基于系统结果的分析,得出系统方案的稳定性、高性价比等优
点及系统可以进一步提高改善的速度、新旧清分程度方面。
Bill sorter, which has fast been developing lately, is an optical, mechanical and electrical financial facility. In this paper, against the character of the bill sorter, on the basis of the digital image processing .pattern recognizition,digital signal processing, the neural network theory, and so on, combined with the advanced and the most up to date technology in the relative application field , author makes a thorough study on the key technology, brings forward a new solution for the bill sorter's system, the two new innovations of the classification arithmetic of the new and used bills and ARM+DSP dual kernel system included , and realize it with high rate between performance with value . The paper consists of the following parts:1. The design of the image processing arithmetic. Firstly, With the main study methods of the digital image processing, the bill's image characters are analyzed. Then, on the basis, the relative knowledge to lay down the solid base for the arithmetic design of the image processing applied in the real bill sorter's system is introduced.2. The technology dominance and application field about the key part in the bill sort system to give the technology support for the system solution about the hardware is demonstrated.3. By analyzing of the character of the bill sorter system, a new solution for the system platform about the hardware and software is presented, and the realization in details from the two aspects in the choose of the key device and system work flow, and so on are also demonstrated. Firstly, the reason about choose of the key hardware part in the system platform is given. Secondly, the system's software work flow of the every core in the system plat form is described. Finally, the arithmetic design of every function realization in detail is introduced.4. The test results from the system platform are analysed to evaluate the system's performance. On the basis of the repeating tests on the realized platform , by the adjusting the parameter of the system , the high performance-price ratio are obtained. Some merits such as high stabilization and high performance-price ratio and some promotion such as system speed and new—old sort are realized.
币清分机产品的特点,基于数字图像处理、模式识别、神经网络理论及数字信号
处理等先进理论,结合当今在相应领域的最新发展应用技术,提出了清分机系统
的解决方案,其中包括新旧识别算法和 ARM+DSP 双核的系统平台构架的两个
创新点,并得以使系统高性价比的实现。论文主要涉及如下几个方面:
1、介绍清分机系统图像处理算法设计基础。首先,运用数字图像处理的一
些主要方法对纸币的图像特征进行了分析。在分析的基础上,介绍了清
分机系统中图像处理相关知识,为实际系统算法设计奠定理论的基础。
2、介绍了清分机系统中硬件平台中的关键部分的技术优势及应用领域,为
系统平台的硬件选型提供技术支撑。
3、通过分析清分机系统的特点,提出了软硬件平台解决方案,从系统的选
型和工作流程等方面详细论证了系统方案的实现。首先,从系统硬件平
台的关键部分来阐述系统的选型论证。其次,从清分机系统的工作流程
来阐述系统各个核的系统软件的流程。最后,从清分机系统的实现功能
来具体介绍系统各个功能的图像处理算法设计。
4、系统的试验结果及性能分析。在系统平台的实现方案上,基于重复试验
的基础上,对系统的实现方案的具体细节不断的调整,获得系统最佳的
性价比。基于系统结果的分析,得出系统方案的稳定性、高性价比等优
点及系统可以进一步提高改善的速度、新旧清分程度方面。
Bill sorter, which has fast been developing lately, is an optical, mechanical and electrical financial facility. In this paper, against the character of the bill sorter, on the basis of the digital image processing .pattern recognizition,digital signal processing, the neural network theory, and so on, combined with the advanced and the most up to date technology in the relative application field , author makes a thorough study on the key technology, brings forward a new solution for the bill sorter's system, the two new innovations of the classification arithmetic of the new and used bills and ARM+DSP dual kernel system included , and realize it with high rate between performance with value . The paper consists of the following parts:1. The design of the image processing arithmetic. Firstly, With the main study methods of the digital image processing, the bill's image characters are analyzed. Then, on the basis, the relative knowledge to lay down the solid base for the arithmetic design of the image processing applied in the real bill sorter's system is introduced.2. The technology dominance and application field about the key part in the bill sort system to give the technology support for the system solution about the hardware is demonstrated.3. By analyzing of the character of the bill sorter system, a new solution for the system platform about the hardware and software is presented, and the realization in details from the two aspects in the choose of the key device and system work flow, and so on are also demonstrated. Firstly, the reason about choose of the key hardware part in the system platform is given. Secondly, the system's software work flow of the every core in the system plat form is described. Finally, the arithmetic design of every function realization in detail is introduced.4. The test results from the system platform are analysed to evaluate the system's performance. On the basis of the repeating tests on the realized platform , by the adjusting the parameter of the system , the high performance-price ratio are obtained. Some merits such as high stabilization and high performance-price ratio and some promotion such as system speed and new—old sort are realized.
引文
[l].金融信息:金融电子化
[2].崔明星:哈尔滨工业大学硕士学位论文—纸币图象识别系统在清分机中的应用,馆藏号:Y389560,分类号TP391.4
[3] 刘丁:清华大学硕士学位论文—人民币现金清分机关键技术的研究与实践,馆藏号:Y414183,分类号:TH693
[4] 孔潇,沈兰荪,毋立芳:一种基于特征提取和模板匹配的人脸检测算法,测控技术,2003V0l.22 No.5
[5] Xuejun Liang; Jean, J.S.-N: Mapping of generalized template matching onto reconfigurable computers, Very Large Scale Integration (VLSI) Systems, IEEE Transactions on .Volume: 11 , Issue: 3 , June 2003 Pages:485 - 498
[6]Fumiaki Takeda, Sigeru Omatu, Saizo Onami, Takashi Kadono, Kengo Terada: A Paper Currency Recognition Method by a Small Size Neural Network with Optimized Masks by GA, In IEEE94/NN [996], pages 4243-4246. ga94aTakeda
[7]Toshihisa Kosaka, Sigeru Omatu, Toru Fujinaka: BILL CLASSIFICATION BY USING THE LVQ METHOD, International Joint Conference on Neural Networks, 2001, 5, 53-56 (2001).
[8]Sigeru Omatu, Toru Fujinaka, Toshihisa Kosaka: Italian Lira Classification by LVQ, International Joint Conference on Neural Networks 2000, 3,143-148 (2000).
[9]Toshihisa Kosaka, Sigeru Omatu: Bill Money Classification by Competitive Learning, Restructuring and Power Technologies 2000
[lO] 欧阳松林:“金通”现金清分机的研制,电脑与信息技术,2000年第二期
[11] Massaru Teranishi, Sigeru Omatu, Toshihisa Kosaka: New And Used Bills Class For Cepstrum Patterns , International Joint Conference on Neural Networks, 2000,1,23-25(2000).
[12]D.S.KANG, H.MIYAGI, S.OMATU: Neuro-Fuzzy Classification of The New and Used Bills Using Acoustic Data, 2000 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2000 Conference Proceedings, 2649-2654 (2000).
[13]Massaru Teranishi, Sigeru Omatu, Toshihisa Kosaka: Classification of Bill Fatigue Levels by Feature-Selected Acoutisc Energy Pattern Using Competitive Neural Network, International Joint Conference on Neural Networks 2000,6, 249-252 (2000).
[14]Dongshik Kang, RRaveendran, Sigeru Omatu: Algoritahm to Discriminate between New and Used Bills Using a Neural Network, DEXA Workshop 1998: 367-372
[15] 王晓光专利:实用新型专利说明书,专利号01276136.2,2002年10月16日
[16] 王勇 纸币清分机图像处理算法及控制策略的研究
[17] 顾临怡,谢英俊:发明专利申请公开说明书,专利号95115183.5,1996年10月2日
[18] 风萨雷斯:数字图像处理(第二版),北京:电子工业出版社,2003年3月
[19] 张兆礼,赵春晖,梅晓丹:现代图像处理技术及Matlab实现,北京:人民邮电出版社,2001年11月
[20] A.V奥本海姆R.w.谢弗 J.R.巴克 离散时间信号处理(第二版)西安交通大学出版社
[21]Massengill,T.M.;Wilson,D.M.: Dual-use signal processing for a multiple-sensor system: sonar and optical image processing for underwater obstacle avoidance, Sensors, 2002.Proceedings of IEEE, Volume: 2,12-14 June 2002 Pages: 1364 -1367 vol.2
[22] 张改英,宋建社,李孟芹,赵东涛:Fourier变换与小波分析,西安工业学院学报,第23卷第6期
[23] 张丕壮,韩焱:利用相关系数对信噪比进行估计,兵工学报,第24卷第1期
[24] 李跃华,沈庆宏,高敦堂,李兴国:小波神经网络的毫米波雷达目标—维距离像识别,南京理工大学学报,第26卷第1期
[25] 郑建斌,杨小曼,刘辉,仲红波:遗传小波神经网络用于极谱信号的滤噪,西北大学学报(自然科学版),第32卷第5期
[26] 李永刚,李锦萍:有限长单位冲激响应FIR数字滤波器线性相位分析和结构图,首都师范大学学报(自然科学版),第23卷第5期
[27] 袁嗣杰,陈谷仓,李贵新:准连续波雷达数字信号处理系统,装备指挥技术学院学报,第13卷第1期
[28]TI公司:DSC25TRMV.3-Technical Reference Manual,2001年
[29]TI公司:DSC25GHK-Verl00A,2001年
[30] TI公司:DSC25-SiliconErrata-20020703,2001年
[3l] TI公司:DSC25-FINAL01,2001年
[32] Tl公司:DSC25 data sheet dratt,2001年
[33] TI公司:TMS320VC5409 Data Sheet.Texas Instruments Incorporated.1997
[34] TI公司: TMS320VC54x DSP Programmer's Guide. 1997
[35] TI公司: TMS320VC54x DSP Algorithm Developer's Guide. 1997
[36] Analog Devices.Inc: AD9814-Technical Reference Manual, 1999
[37]Syscan .Inc: SV351A4C-P2-Technical Reference Manual
[38] Motorola, Inc: HCS12COREUG?2002
[39] Motorola, Inc: 9S12A128DGV1,2002
[40] Motorola, Inc: S12VREGV1, 2002
[41] Motorola, Inc: S12PWM8B8CV1, 2002
[42] 耿德根,送建国,马潮,叶勇建:AVR高速嵌入式单片机原理与应用,北京:北京航空航天大学出版社,2001年2月
[43] 谢克明:可编程控制器原理和程序设计,北京:电子工业出版社,2003年4月
[44] 潘松,王国栋:VHDL实用教程,成都:电子科技大学出版社,2000年2月
[45]Smith J. R, Chang S. F. Transform Feature for Texture Classification and Discrimination in Large Image Database. Proc of IEEE Intel Conf on Image Processing, 1994.472-481.
[46]la Li.Image Segmentation And Compression Using Hidden Markov Models. Stanford University. 2000:1117
[47] B. M. Mehtre, M. S. Kankanhalli. Color Matching for Image Retrieval. Patern Recognition Leters 16(3), 1995: 325-331
[48] 吕岳.基于左右轮廓特征的打印体邮政编码数字快速识别.通信学报.18(7)l997:94-96
[49] 张利,朱颖,吴国威.版面分割中游程平滑后的图文特征分类.中国图像图形学报v0l.3,no 10,Oct,1998:827-83l
[50] 张玉,林政国.数字图像模式识别技术及工程实践.人民邮电出版社.2001:341346;401418
[51] 江和平,孙茂印:钞票颜色识别系统的设计,微处理机,2000年11月
[52] 李宏,薛冰,杨英科,左少平,谢虹:智能现金清分系统中图像模块设计,微计算机应用,1999年5月
[53] 周芝梅,刘振宇,韩月秋:基于FPGA的高速实时三通道脉压处理器研究,现代雷达,2003年7月
[54] 张静,索继东:基于FPGA和DSP的雷达信号处理系统的设计,大连海事大学学报,第28卷第2期
[55] 冯涛,曹丹华,吴裕斌:基于DSP的CIS信号采集与处理技术,测控技术,2003 V0l.22 No.8
[56] 陈国鼎,陈更生,邢达:基于TMS320VC33 多处理机图像处理系统,微处理机,2002年11月第四期
[57] 王毅,倪国强,李勇量:基于TMS320VC60 双波段图像高速融合系统,中国图像图形学报,第7卷(A)第10期
[58]Robert Allison, Moshe Eizeman , R.David Tomlinson , James Sharpe , Richard C.Frecker,Jennifer Anderson,Lynne McIlmoyl,"Head and Eye Tracking for Study of The Vor during Natural Head Motion",IEEE,1994
[59] 卓坚鹏,影像传感器基础及应用,全华科技图书,1993
[60] 陈俊达,使用线影像传感器架构之眼球轨迹追踪装置 逢甲大学自动控制工程研究所硕士论文,1999
[61] 许书务,”光传感器接口专题制作”,电子出版社,1989
[62]Eddins, S.L.; Orchard, M.T: Using MATLAB and C in an image processing lab course, Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference , Volume: 1,13-16 Nov. 1994 Pages:515-519 vol.1
[63]Meribout, M.; Hou, K.M: Implementation of low level image processing algorithms on a reconfigurable perception system , Computer Architectures for Machine Perception, 1995. Proceedings. CAMP '95 ,18-20 Sept. 1995 Pages:304 - 312
[64]Zhang, X.; Sexton, G: Automatic pedestrian counting using image processing techniques. Electronics Letters, Volume: 31 , Issue: 11 ,25 May 1995 Pages:863 - 865
[65]Yajima, M.; Higashi, K.; Kondo, H.; Inaba, M.; Kondo, Y.; Ohki, M.; Hashiguchi, S: A system for real time extraction of image outlines, Consumer Electronics, IEEE Transactions on , Volume: 40 , Issue: 4 , Nov 1994 Pages:890 - 896
[66]Pak Chung Wong; Kwong Kwok Wong; Foote, H., Thomas, J.: Global visualization and alignments of whole bacterial genomes, Visualization and Computer Graphics, IEEE Transactions on , Volume: 9 , Issue: 3 , July-Sept. 2003 Pages:361 - 377
[67]Ferrari, R.J.; Rangayyan, R.M.; Desautels, J.E.L.; Borges, R.A.; Frere, A.F: Automatic
Identification of the Pectoral Muscle in Mammograms, Medical Imaging, IEEE Transactions on , Volume: 23 , Issue: 2 , Feb. 2004 Pages:232 - 245
[68]Bin Fang; Hsu, W.; Mong Li Lee: On the accurate counting of tumor cells, Nanobioscience, IEEE Transactions on , Volume: 2 , Issue: 2 , June 2003 . Pages:94-103
[69]Xuejun Liang; Jean, J.S.-N: Mapping of generalized template matching onto reconfigurable computers, Very Large Scale Integration (VLSI) Systems, IEEE Transactions on .Volume: 11 , Issue: 3 , June 2003 Pages:485 - 498
[70] 陈永光,王国柱,撒潮,王杨: 木材表面缺陷边缘形态检测算法的研究,木材加工机械,2003 V0L.14 No.3
[7l] 沈明霞,李秀智,姬长英:水果品质检测中的模糊阈值分割方法,农业机械学报,2003Vol.34 No.05
[72] 俞妍妍,王继成:基于改进的多尺度形态梯度的图像边缘检测,计算机工程与应用,2003V01.39 No.18
[73] 陶劲松,陈港,刘松坡:计算机图像处理技术在纸页分析中的应用,中国造纸
[74] 杜春雷,ARM体系结构与编程,清华大学出版社
[75] 马忠梅,马广云,ARM 嵌入式处理器结构与应用基础,北京航空航天大学出版社
[76] 孔祥营,柏桂枝,嵌入式实时操作系统 VxWorks 及其开发环境Tornado,中国电力出版社
[77] 蔡德聪,工业控制计算机实时操作系统,清华大学出版社
[78]Jean J.Labrosse,MicroC/OS.II The Real—Time Kernel(Second Edition),北京航空航天大学出版社
[79] Rafael C Gonzalez,Richard E.Woods,Steven L.Eddins,Digital Image Processing UsingMATLAB.电子工业出版社
[80] Martin T.Hagan,Howard B.Demuth,Mark Beale,Neural Network Design,机械工业出版社
[2].崔明星:哈尔滨工业大学硕士学位论文—纸币图象识别系统在清分机中的应用,馆藏号:Y389560,分类号TP391.4
[3] 刘丁:清华大学硕士学位论文—人民币现金清分机关键技术的研究与实践,馆藏号:Y414183,分类号:TH693
[4] 孔潇,沈兰荪,毋立芳:一种基于特征提取和模板匹配的人脸检测算法,测控技术,2003V0l.22 No.5
[5] Xuejun Liang; Jean, J.S.-N: Mapping of generalized template matching onto reconfigurable computers, Very Large Scale Integration (VLSI) Systems, IEEE Transactions on .Volume: 11 , Issue: 3 , June 2003 Pages:485 - 498
[6]Fumiaki Takeda, Sigeru Omatu, Saizo Onami, Takashi Kadono, Kengo Terada: A Paper Currency Recognition Method by a Small Size Neural Network with Optimized Masks by GA, In IEEE94/NN [996], pages 4243-4246. ga94aTakeda
[7]Toshihisa Kosaka, Sigeru Omatu, Toru Fujinaka: BILL CLASSIFICATION BY USING THE LVQ METHOD, International Joint Conference on Neural Networks, 2001, 5, 53-56 (2001).
[8]Sigeru Omatu, Toru Fujinaka, Toshihisa Kosaka: Italian Lira Classification by LVQ, International Joint Conference on Neural Networks 2000, 3,143-148 (2000).
[9]Toshihisa Kosaka, Sigeru Omatu: Bill Money Classification by Competitive Learning, Restructuring and Power Technologies 2000
[lO] 欧阳松林:“金通”现金清分机的研制,电脑与信息技术,2000年第二期
[11] Massaru Teranishi, Sigeru Omatu, Toshihisa Kosaka: New And Used Bills Class For Cepstrum Patterns , International Joint Conference on Neural Networks, 2000,1,23-25(2000).
[12]D.S.KANG, H.MIYAGI, S.OMATU: Neuro-Fuzzy Classification of The New and Used Bills Using Acoustic Data, 2000 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2000 Conference Proceedings, 2649-2654 (2000).
[13]Massaru Teranishi, Sigeru Omatu, Toshihisa Kosaka: Classification of Bill Fatigue Levels by Feature-Selected Acoutisc Energy Pattern Using Competitive Neural Network, International Joint Conference on Neural Networks 2000,6, 249-252 (2000).
[14]Dongshik Kang, RRaveendran, Sigeru Omatu: Algoritahm to Discriminate between New and Used Bills Using a Neural Network, DEXA Workshop 1998: 367-372
[15] 王晓光专利:实用新型专利说明书,专利号01276136.2,2002年10月16日
[16] 王勇 纸币清分机图像处理算法及控制策略的研究
[17] 顾临怡,谢英俊:发明专利申请公开说明书,专利号95115183.5,1996年10月2日
[18] 风萨雷斯:数字图像处理(第二版),北京:电子工业出版社,2003年3月
[19] 张兆礼,赵春晖,梅晓丹:现代图像处理技术及Matlab实现,北京:人民邮电出版社,2001年11月
[20] A.V奥本海姆R.w.谢弗 J.R.巴克 离散时间信号处理(第二版)西安交通大学出版社
[21]Massengill,T.M.;Wilson,D.M.: Dual-use signal processing for a multiple-sensor system: sonar and optical image processing for underwater obstacle avoidance, Sensors, 2002.Proceedings of IEEE, Volume: 2,12-14 June 2002 Pages: 1364 -1367 vol.2
[22] 张改英,宋建社,李孟芹,赵东涛:Fourier变换与小波分析,西安工业学院学报,第23卷第6期
[23] 张丕壮,韩焱:利用相关系数对信噪比进行估计,兵工学报,第24卷第1期
[24] 李跃华,沈庆宏,高敦堂,李兴国:小波神经网络的毫米波雷达目标—维距离像识别,南京理工大学学报,第26卷第1期
[25] 郑建斌,杨小曼,刘辉,仲红波:遗传小波神经网络用于极谱信号的滤噪,西北大学学报(自然科学版),第32卷第5期
[26] 李永刚,李锦萍:有限长单位冲激响应FIR数字滤波器线性相位分析和结构图,首都师范大学学报(自然科学版),第23卷第5期
[27] 袁嗣杰,陈谷仓,李贵新:准连续波雷达数字信号处理系统,装备指挥技术学院学报,第13卷第1期
[28]TI公司:DSC25TRMV.3-Technical Reference Manual,2001年
[29]TI公司:DSC25GHK-Verl00A,2001年
[30] TI公司:DSC25-SiliconErrata-20020703,2001年
[3l] TI公司:DSC25-FINAL01,2001年
[32] Tl公司:DSC25 data sheet dratt,2001年
[33] TI公司:TMS320VC5409 Data Sheet.Texas Instruments Incorporated.1997
[34] TI公司: TMS320VC54x DSP Programmer's Guide. 1997
[35] TI公司: TMS320VC54x DSP Algorithm Developer's Guide. 1997
[36] Analog Devices.Inc: AD9814-Technical Reference Manual, 1999
[37]Syscan .Inc: SV351A4C-P2-Technical Reference Manual
[38] Motorola, Inc: HCS12COREUG?2002
[39] Motorola, Inc: 9S12A128DGV1,2002
[40] Motorola, Inc: S12VREGV1, 2002
[41] Motorola, Inc: S12PWM8B8CV1, 2002
[42] 耿德根,送建国,马潮,叶勇建:AVR高速嵌入式单片机原理与应用,北京:北京航空航天大学出版社,2001年2月
[43] 谢克明:可编程控制器原理和程序设计,北京:电子工业出版社,2003年4月
[44] 潘松,王国栋:VHDL实用教程,成都:电子科技大学出版社,2000年2月
[45]Smith J. R, Chang S. F. Transform Feature for Texture Classification and Discrimination in Large Image Database. Proc of IEEE Intel Conf on Image Processing, 1994.472-481.
[46]la Li.Image Segmentation And Compression Using Hidden Markov Models. Stanford University. 2000:1117
[47] B. M. Mehtre, M. S. Kankanhalli. Color Matching for Image Retrieval. Patern Recognition Leters 16(3), 1995: 325-331
[48] 吕岳.基于左右轮廓特征的打印体邮政编码数字快速识别.通信学报.18(7)l997:94-96
[49] 张利,朱颖,吴国威.版面分割中游程平滑后的图文特征分类.中国图像图形学报v0l.3,no 10,Oct,1998:827-83l
[50] 张玉,林政国.数字图像模式识别技术及工程实践.人民邮电出版社.2001:341346;401418
[51] 江和平,孙茂印:钞票颜色识别系统的设计,微处理机,2000年11月
[52] 李宏,薛冰,杨英科,左少平,谢虹:智能现金清分系统中图像模块设计,微计算机应用,1999年5月
[53] 周芝梅,刘振宇,韩月秋:基于FPGA的高速实时三通道脉压处理器研究,现代雷达,2003年7月
[54] 张静,索继东:基于FPGA和DSP的雷达信号处理系统的设计,大连海事大学学报,第28卷第2期
[55] 冯涛,曹丹华,吴裕斌:基于DSP的CIS信号采集与处理技术,测控技术,2003 V0l.22 No.8
[56] 陈国鼎,陈更生,邢达:基于TMS320VC33 多处理机图像处理系统,微处理机,2002年11月第四期
[57] 王毅,倪国强,李勇量:基于TMS320VC60 双波段图像高速融合系统,中国图像图形学报,第7卷(A)第10期
[58]Robert Allison, Moshe Eizeman , R.David Tomlinson , James Sharpe , Richard C.Frecker,Jennifer Anderson,Lynne McIlmoyl,"Head and Eye Tracking for Study of The Vor during Natural Head Motion",IEEE,1994
[59] 卓坚鹏,影像传感器基础及应用,全华科技图书,1993
[60] 陈俊达,使用线影像传感器架构之眼球轨迹追踪装置 逢甲大学自动控制工程研究所硕士论文,1999
[61] 许书务,”光传感器接口专题制作”,电子出版社,1989
[62]Eddins, S.L.; Orchard, M.T: Using MATLAB and C in an image processing lab course, Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference , Volume: 1,13-16 Nov. 1994 Pages:515-519 vol.1
[63]Meribout, M.; Hou, K.M: Implementation of low level image processing algorithms on a reconfigurable perception system , Computer Architectures for Machine Perception, 1995. Proceedings. CAMP '95 ,18-20 Sept. 1995 Pages:304 - 312
[64]Zhang, X.; Sexton, G: Automatic pedestrian counting using image processing techniques. Electronics Letters, Volume: 31 , Issue: 11 ,25 May 1995 Pages:863 - 865
[65]Yajima, M.; Higashi, K.; Kondo, H.; Inaba, M.; Kondo, Y.; Ohki, M.; Hashiguchi, S: A system for real time extraction of image outlines, Consumer Electronics, IEEE Transactions on , Volume: 40 , Issue: 4 , Nov 1994 Pages:890 - 896
[66]Pak Chung Wong; Kwong Kwok Wong; Foote, H., Thomas, J.: Global visualization and alignments of whole bacterial genomes, Visualization and Computer Graphics, IEEE Transactions on , Volume: 9 , Issue: 3 , July-Sept. 2003 Pages:361 - 377
[67]Ferrari, R.J.; Rangayyan, R.M.; Desautels, J.E.L.; Borges, R.A.; Frere, A.F: Automatic
Identification of the Pectoral Muscle in Mammograms, Medical Imaging, IEEE Transactions on , Volume: 23 , Issue: 2 , Feb. 2004 Pages:232 - 245
[68]Bin Fang; Hsu, W.; Mong Li Lee: On the accurate counting of tumor cells, Nanobioscience, IEEE Transactions on , Volume: 2 , Issue: 2 , June 2003 . Pages:94-103
[69]Xuejun Liang; Jean, J.S.-N: Mapping of generalized template matching onto reconfigurable computers, Very Large Scale Integration (VLSI) Systems, IEEE Transactions on .Volume: 11 , Issue: 3 , June 2003 Pages:485 - 498
[70] 陈永光,王国柱,撒潮,王杨: 木材表面缺陷边缘形态检测算法的研究,木材加工机械,2003 V0L.14 No.3
[7l] 沈明霞,李秀智,姬长英:水果品质检测中的模糊阈值分割方法,农业机械学报,2003Vol.34 No.05
[72] 俞妍妍,王继成:基于改进的多尺度形态梯度的图像边缘检测,计算机工程与应用,2003V01.39 No.18
[73] 陶劲松,陈港,刘松坡:计算机图像处理技术在纸页分析中的应用,中国造纸
[74] 杜春雷,ARM体系结构与编程,清华大学出版社
[75] 马忠梅,马广云,ARM 嵌入式处理器结构与应用基础,北京航空航天大学出版社
[76] 孔祥营,柏桂枝,嵌入式实时操作系统 VxWorks 及其开发环境Tornado,中国电力出版社
[77] 蔡德聪,工业控制计算机实时操作系统,清华大学出版社
[78]Jean J.Labrosse,MicroC/OS.II The Real—Time Kernel(Second Edition),北京航空航天大学出版社
[79] Rafael C Gonzalez,Richard E.Woods,Steven L.Eddins,Digital Image Processing UsingMATLAB.电子工业出版社
[80] Martin T.Hagan,Howard B.Demuth,Mark Beale,Neural Network Design,机械工业出版社