嵌入式眼标系统的硬件设计
|
摘要
目前的人机交互技术,不管是通过键盘,鼠标,还是摇杆,手柄,都是人主动去适应设备,将多维的外部信息压缩成传统输入设备所能接受的信息流。在计算机领域,技术的发展趋势是向着人性化,自然化的方向迈进,眼标系统正是自然人机交互的一种方式。PC版的眼标系统需要耗费大量的系统资源,影响PC处理其他的工作,嵌入式的眼标系统如果采用多核并行架构,则可以实现眼标系统的实时控制。本文提出了一个嵌入式眼标系统的多核并行架构的具体实现。
本文的主要工作体现在完成了在DSP平台上对部分眼标软件程序的测试,根据测试结果和相关资料,放弃了选用DSP核,而选取了三星高性能的s3c2440芯片作为眼标硬件系统的并行处理器,同时根据眼标系统的需求选择了一系列符合系统需求的元器件,包括恩智浦公司的ISP1564,Altera公司的EP2C15AF484C等等;最后在选择的这些元器件的基础上合理的设计硬件电路。本文为眼标硬件系统设计了6个模块,即一个主处理器模块、三个辅处理器模块,一个总线模块和JTAG下载调试模块,并完成了各个模块的硬件电路的设计。
本文设计的眼标硬件系统,能够实现一秒钟处理10幅以上的图片,实现了眼标系统的实时控制。
The current human-computer interaction technology, whether through the keyboard, mouse or through joystick handle, is all that requires people adapting the equipment, to compress the external multidimensional information into a flow of information that traditional input devices could accept. In the computer field, the technology trend is towards humanity and naturalization. Eye-mouse system is one way of natural human-computer interaction. The eye-mouse system version PC requires a lot of system resources, cause the PC can not handle other work well. If using multi-core parallel architecture, embedded eye-mouse system can achieve real-time control the mouse system. This paper provide a concrete realization of multi-core parallel architecture for eye-mouse system.
The major work of this paper contains: finished the testing of a part of eye-mouse system’s software on DSP platform; according to the test results and related information, this paper gave up selecting DSP core, and selected high performance chip s3c2440 of Samsung as the parallel processor of the eye-mouse system; according to the demand of eye-mouse system,this paper selected a range of components to meet the system requirements in addition, including NXP’s ISP1564, Altera Corporation’s EP2C15AF484C, and so forth. Finally, based on the selected components, this paper designed the hardware reasonably. This paper designed 6 modules for eye-mouse system. These includes a main processor module, three auxiliary processor modules, a bus module and JTAG download and debug module. And completed hardware design for each module.
The eye-mouse hardware system that this paper designed, can achieve handle more than 10 pictures in one second, can realize real-time control the mouse system.
本文的主要工作体现在完成了在DSP平台上对部分眼标软件程序的测试,根据测试结果和相关资料,放弃了选用DSP核,而选取了三星高性能的s3c2440芯片作为眼标硬件系统的并行处理器,同时根据眼标系统的需求选择了一系列符合系统需求的元器件,包括恩智浦公司的ISP1564,Altera公司的EP2C15AF484C等等;最后在选择的这些元器件的基础上合理的设计硬件电路。本文为眼标硬件系统设计了6个模块,即一个主处理器模块、三个辅处理器模块,一个总线模块和JTAG下载调试模块,并完成了各个模块的硬件电路的设计。
本文设计的眼标硬件系统,能够实现一秒钟处理10幅以上的图片,实现了眼标系统的实时控制。
The current human-computer interaction technology, whether through the keyboard, mouse or through joystick handle, is all that requires people adapting the equipment, to compress the external multidimensional information into a flow of information that traditional input devices could accept. In the computer field, the technology trend is towards humanity and naturalization. Eye-mouse system is one way of natural human-computer interaction. The eye-mouse system version PC requires a lot of system resources, cause the PC can not handle other work well. If using multi-core parallel architecture, embedded eye-mouse system can achieve real-time control the mouse system. This paper provide a concrete realization of multi-core parallel architecture for eye-mouse system.
The major work of this paper contains: finished the testing of a part of eye-mouse system’s software on DSP platform; according to the test results and related information, this paper gave up selecting DSP core, and selected high performance chip s3c2440 of Samsung as the parallel processor of the eye-mouse system; according to the demand of eye-mouse system,this paper selected a range of components to meet the system requirements in addition, including NXP’s ISP1564, Altera Corporation’s EP2C15AF484C, and so forth. Finally, based on the selected components, this paper designed the hardware reasonably. This paper designed 6 modules for eye-mouse system. These includes a main processor module, three auxiliary processor modules, a bus module and JTAG download and debug module. And completed hardware design for each module.
The eye-mouse hardware system that this paper designed, can achieve handle more than 10 pictures in one second, can realize real-time control the mouse system.
引文
[1]舒志龙,陆俊,阮秋琦,基于视觉的感知用户界面(VBPUI)研究[J],中国图象图形学报,2000,5(8):683-687
[2]冯成志,沈模卫,视线跟踪技术及其在人机交互中的应用[J],浙江大学学报(理学版),2002,(2):225-232
[3]李茂贞.戴国忠,董上海,多通道界面模型与关建技术[J]:计算机科学,1997,24(1):1-4.
[4]赵新灿,左洪福,任勇军,眼动仪与视线跟踪技术综述[J],计算机工程与应用,2006,(12):118-120
[5]Shupert C,Fuchs AF Developmelxt of conjugate hu—Rlan eye movement[J].Vision Rts.1988.23(5):585—596.
[6]王勇,袁景和,常胜江,张延炘，基于参数提取的人眼注视方向的判别[J],光电子.激光,2001,12(12):1284-1287
[7]吴倩,鼠标时代设计理念,华章,2007(7):161
[8]吴祈宗.运筹学与最优化方法[M].北京:机械工业出版社,2003.
[9]罗兴贤,魏生民,刘雅婧,孟举,基于图像处理的视线方向跟踪研究[J],现代制造工程,2007(01):87-90
[10]尹海荣,屠大维,基于红外电视法的眼睛盯视人机交互技术[J],红外技术,2002,24(4):1-3
[11]林粤江,浅谈人机交互技术[J],黑龙江科技信息,2007,(7):68
[12]邓凯,eyetoy是怎样炼成的[J],程序员:游戏创造,2006(4):68-71
[13]WANG JIAN.”Integration of eye-gaze, voice and manual response in multimodal user interface”.In Proc. Of the IEEE International Conference on System, Man and Cybemetics. IEEE.1995.Piscataway,N.J. P3938-3942
[14]方志刚,人机交互中的视线跟踪技术[J],人体工效学,2006,6(1):43-45
[15]廖卫华,严由伟,徐子清,视线跟踪技术在人机交互中的研究,长春师范学院学报(自然科学版),2005,24(2):130-132
[16] STIEFELHAGEN R, YANG J, Gaze tracking for multimodal human-computer interaction [A], In Proceedings of International Conf. on Acoustics, Speech, and Signal Processing[C], Germany: Munich, 1997
[17] Robinson D.A. A method of measuring eye movements using a scleral search coilin a magnetic field, I.E.E.E. Trans.Biomed.Electron.BME10, 137-140
[18] Y. Ebisawa, K. Kaneko, S. Kojima, T. Ushikubo and T. Miyakawa. Non-invasive eye-gaze position detecting method used on man/machine interface for the disabled. Proc. 4th Symp. Computer-Based Medical System 374-380 (1991).
[19]耿卫东,宋黎明,基于摄像机的无障碍交互技术综述,计算机应用,2007,27(9):2087-2090
[20] Hutchinson T E. White K P. Human-Computer Interaction Using Eye_Gaze Input[A],IEEE.Transaction Systems,Man,and Cybernetics[C],1989.19(6):1527~1533
[21]李卫娜,侯文生,郑小林,彭承琳,基于Electrooculogram的眼动信息识别,仪器仪表学报,2007,28(8):1428-1433
[22]郭晓波,基于视线跟踪技术的眼控鼠标系统研究[D],天津,天津大学,2009
[23] Rein-Lien Hsu, Abdel-Mottaleb M., Jain A.K., Face Detection in color images, IEEE Trans. on Pattern Analysis and Machine Intelligence, 2002, 24(5): 696~706
[24]陈丽华,龚雯,沈建国,基于人脸结构特征的眼睛定位算法,电子工程师,2005,31(9):34-36
[25]左坤隆,刘文耀,朱昊,基于差分投影和梯度向量流场的眼睛特征提取,计算机工程, 2005,31(4):26~28
[26]朱军林,朱秋煜,张婷,证件照片脸部特征精确定位与提取,微计算机信息, 2006,07S, 200~202
[27]张桂林,利用SUSAN算子的特征复合相关跟踪算法,红外与激光工程,2000 ,29( 4),34~37
[28]王宇,嵌入式眼控鼠标系统的硬件架构设计[D].天津,天津大学,2009
[29]徐艳敏,张星明,刘晓红,苏曙,基于色彩聚类的肤色分割方法的研究,广州航海高等学校学报,2004,12(1):24~27
[30]田欣,基于不同色彩空间的肤色模型,西安科技学院学报,2001,21(4): 369~372
[31] Kanade T. Picture processing by computer complex and recognition of human faces [PH.D. Thesis]. Kyoto: Kyoto University, 1973.
[32] Brunelli R, Poggio T. Face recognition: Features versus templates. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993, 15(10): 1042~1052
[33]周长发,精通Visual C++图像处理编程,北京:电子工业出版社,2004,281～286
[34] Rafael C. Gonzalez, Richard E. Woods,数字图像处理(第二版)阮秋琦,阮宇智等译,北京:电子工业出版社,2003,423~427
[35]魏阳杰,董再励等,新的Ostu阈值改进方法的红处小目标检测,光学工程,2004.11:23~26
[2]冯成志,沈模卫,视线跟踪技术及其在人机交互中的应用[J],浙江大学学报(理学版),2002,(2):225-232
[3]李茂贞.戴国忠,董上海,多通道界面模型与关建技术[J]:计算机科学,1997,24(1):1-4.
[4]赵新灿,左洪福,任勇军,眼动仪与视线跟踪技术综述[J],计算机工程与应用,2006,(12):118-120
[5]Shupert C,Fuchs AF Developmelxt of conjugate hu—Rlan eye movement[J].Vision Rts.1988.23(5):585—596.
[6]王勇,袁景和,常胜江,张延炘，基于参数提取的人眼注视方向的判别[J],光电子.激光,2001,12(12):1284-1287
[7]吴倩,鼠标时代设计理念,华章,2007(7):161
[8]吴祈宗.运筹学与最优化方法[M].北京:机械工业出版社,2003.
[9]罗兴贤,魏生民,刘雅婧,孟举,基于图像处理的视线方向跟踪研究[J],现代制造工程,2007(01):87-90
[10]尹海荣,屠大维,基于红外电视法的眼睛盯视人机交互技术[J],红外技术,2002,24(4):1-3
[11]林粤江,浅谈人机交互技术[J],黑龙江科技信息,2007,(7):68
[12]邓凯,eyetoy是怎样炼成的[J],程序员:游戏创造,2006(4):68-71
[13]WANG JIAN.”Integration of eye-gaze, voice and manual response in multimodal user interface”.In Proc. Of the IEEE International Conference on System, Man and Cybemetics. IEEE.1995.Piscataway,N.J. P3938-3942
[14]方志刚,人机交互中的视线跟踪技术[J],人体工效学,2006,6(1):43-45
[15]廖卫华,严由伟,徐子清,视线跟踪技术在人机交互中的研究,长春师范学院学报(自然科学版),2005,24(2):130-132
[16] STIEFELHAGEN R, YANG J, Gaze tracking for multimodal human-computer interaction [A], In Proceedings of International Conf. on Acoustics, Speech, and Signal Processing[C], Germany: Munich, 1997
[17] Robinson D.A. A method of measuring eye movements using a scleral search coilin a magnetic field, I.E.E.E. Trans.Biomed.Electron.BME10, 137-140
[18] Y. Ebisawa, K. Kaneko, S. Kojima, T. Ushikubo and T. Miyakawa. Non-invasive eye-gaze position detecting method used on man/machine interface for the disabled. Proc. 4th Symp. Computer-Based Medical System 374-380 (1991).
[19]耿卫东,宋黎明,基于摄像机的无障碍交互技术综述,计算机应用,2007,27(9):2087-2090
[20] Hutchinson T E. White K P. Human-Computer Interaction Using Eye_Gaze Input[A],IEEE.Transaction Systems,Man,and Cybernetics[C],1989.19(6):1527~1533
[21]李卫娜,侯文生,郑小林,彭承琳,基于Electrooculogram的眼动信息识别,仪器仪表学报,2007,28(8):1428-1433
[22]郭晓波,基于视线跟踪技术的眼控鼠标系统研究[D],天津,天津大学,2009
[23] Rein-Lien Hsu, Abdel-Mottaleb M., Jain A.K., Face Detection in color images, IEEE Trans. on Pattern Analysis and Machine Intelligence, 2002, 24(5): 696~706
[24]陈丽华,龚雯,沈建国,基于人脸结构特征的眼睛定位算法,电子工程师,2005,31(9):34-36
[25]左坤隆,刘文耀,朱昊,基于差分投影和梯度向量流场的眼睛特征提取,计算机工程, 2005,31(4):26~28
[26]朱军林,朱秋煜,张婷,证件照片脸部特征精确定位与提取,微计算机信息, 2006,07S, 200~202
[27]张桂林,利用SUSAN算子的特征复合相关跟踪算法,红外与激光工程,2000 ,29( 4),34~37
[28]王宇,嵌入式眼控鼠标系统的硬件架构设计[D].天津,天津大学,2009
[29]徐艳敏,张星明,刘晓红,苏曙,基于色彩聚类的肤色分割方法的研究,广州航海高等学校学报,2004,12(1):24~27
[30]田欣,基于不同色彩空间的肤色模型,西安科技学院学报,2001,21(4): 369~372
[31] Kanade T. Picture processing by computer complex and recognition of human faces [PH.D. Thesis]. Kyoto: Kyoto University, 1973.
[32] Brunelli R, Poggio T. Face recognition: Features versus templates. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1993, 15(10): 1042~1052
[33]周长发,精通Visual C++图像处理编程,北京:电子工业出版社,2004,281～286
[34] Rafael C. Gonzalez, Richard E. Woods,数字图像处理(第二版)阮秋琦,阮宇智等译,北京:电子工业出版社,2003,423~427
[35]魏阳杰,董再励等,新的Ostu阈值改进方法的红处小目标检测,光学工程,2004.11:23~26