摘要
自二十世纪九十年代TFT-LCD开始正式量产以来,其产品被广泛应用于数码相机、车载显示器、笔记本电脑和液晶显示等几乎所有的显示器领域。TFT-LCD自动光学检测设备是一个集成了光、机、电、自动化等技术的精密机器视觉识别系统,在液晶显示屏缺陷检测中,需要一种调焦装置实现高精度非接触的自动对焦来获取清晰玻璃基板图片。本文针对TFT-LCD面板尺寸大、厚度薄、光透过率高的特性,设计了基于激光三角测距法的自动对焦系统。主要包含以下内容：
(1)对自动对焦技术、激光测距的发展现状进行了研究,比较了各种对焦方法和激光测距方法的优缺点,提出了基于激光三角测距法的自动对焦系统来实现大尺寸玻璃基板的非接触快速对焦。
(2)根据TFT-LCD面板光学检测设计的要求,提出采用三维立体机构的机械平台,压电陶瓷主要驱动6个不同倍数的显微物镜上下移动来实现微小行程调焦,具有调焦负载小,调焦速度快,精度高的特点。
(3)分析了激光三角法原理,在其理论基础上提出了自动对焦系统光学回路设计方案,并从原理上进行了论证,根据光斑图像的中心位置和离焦量之间的相关性实现快速对焦,该方法具有调焦方向明确,准确度高的优点。
(4)给出了自动对焦系统的控制方案,通过分析TFT-LCD面扳反射的能量对光斑图像的影响,提出在焦平面附近调节相机快门和增益以消除鬼影,比较了几种常用的调焦评价函数,提出了根据光斑中心位置误差允许范围作为调焦评价的依据,具有调焦准确、精度高的优点。
(5)软件部分设计,自主研发的显微镜自动对焦系统软件,通过1394图像采集卡实现了激光光斑图像的采集处理,通过动念函数的调用实现了三轴电机的运动控制,压电陶瓷的微调焦控制,通过手柄设计快速实现显微镜不同镜头的控制。
(6)实验结果表明：在行程100um的自动对焦范围内,5X物镜下,自动对焦时间为0.237s,重复定位精度为±1.97um。50X物镜下，自动对焦时间为0.29s,重复定位精度为±0.26um。该系统稳定性好,对焦精度高,抗干扰能力强，基本满足大尺寸玻璃基板光学检测的需要。
Since the 1990s, People started to mass produce TFT-LCD and the products have been widely used in almost all display areas,such as digital cameras, automotive displays, notebook computers, LCD monitors and so on. The TFT-LCD Automatic Optic Inspection is a precision machinery visual recognition system which integrates the light,electronics, automation and so on. The TFT-LCD defect detection need one kind of focusing mechanism to realize high accuracy non-contact automatic focusing to get the clear glass substrate images. In view of the TFT-LCD panel have the characteristics of large size,thin depth and high light transmittance,people established a autofocus system based on the Laser-triangulation. Primarily contains the following content:
Firstly, this article makes research to the current development of auto focus technology and laser range finder, compares the advantages and disadvantages of all kinds of focusing methods and laser range methods, put forth the auto-focus system based on laser triangulation to realize the large size glass substrate non-contact quick focusing.
Secondly, According to the requirements of TFT-LCD optical detection, using the three-dimensional organization mechanical platform and the piezoelectric ceramic mainly to driver six different multiples micro objective lens moving up and down to achieve the small stroke focusing, with the characteristics of small focusing load, fast focusing speed and high accuracy.
Thirdly, it analyses the principle of laser triangulation method and proposed the auto-focus system optical circuit design program base on the theoretical principle, and has a demonstration in principle, according to the center of focal spot and the correlation of between the defocusing amount to achieve the quick focusing, this method has the advantages of clearly focusing direction and high accuracy.
Fourthly, giving the control scheme of auto-focus system, It advances the methods about adjusting the shutter and gain around the focal plane to eliminate the ghost according to analyse the effect of reflection energy of TFT-LCD to the light spot image,compared with several kinds of commonly used focusing evaluation functions, proposing according to the the range of allowable error of light spot center as the basis of the focusing evaluation,it has the merit of accurately focusing and high precision.
Fifthly, in the software design part, the independent research microscope autofocus system software adopts the 1394 image acquisition card to achieve the acquisition and processing of the laser spot image, according to call the dynamic function to realize the motion control of three-axis electric motor and the micro focusing control of ceramics piezoelectric, through the handle design quickly realize the different lens control of microscope.
Finally, the experimental results indicate that within the 100um distance auto-focus and under the five times objective lens, the autofocus time is 0.237 seconds, the repetitive positioning accuracy is±1.97um.Under the conditions of fifty times objective lens, the autofocus time is 0.29 seconds, the repetitive positioning accuracy is±0.26um. It can satisfy the system requirements of non-contact, online, real time, higher precision and rapid speed, as well as strong anti-jamming and stabilization.
引文
[1]Jae-Hyuck Woo, Jae-Goo Lee, Yong-Hyun Jun, et al. A Line Inversion-Based Stepwise Data Driving for Low-Power Mobile TFT-LCDs[J].IEEE Transactions on Consumer Electronic,2010.5, pp: 1002-1009.
[2]Chien-Huang Tsai, Jia-Hui Wang.Cross-Self-Bias Rail-to-Rail Column Buffer for 640-Channel 6-Bit TFT-LCD[J].IEEE display Technology,2011, pp:235-241.
[3]Ping-Feng Pai, Kuo-Ping, Shun-Ling Yang, et al. Diagnosing faulty in TFT-LCD manufacturing processes by support vector machines with principal components analysis[C].IEEE Computer Society,2009.247, pp:413-417.
[4]田英良,张磊,黛琳等TFT-LCD基板玻璃化学组成的发展状况与展望[J].硅酸盐通报,2010,29(6).1348-1352.
[5]丁沭沂,李驰.平扳显示技术的现在与展望[J].电脑知识与技术,2010,34(6).9854-9856.
[6]汀敏,凌安海,万异等TFT-LCD彩色滤光片技术及其产业发展现状[J].现代显示,2009,106(1).22-26.
[7]成田利治,高谷辰弥,笛古雅博.无碱玻璃基扳[P].中国专利：CN1590330,2005.
[8]Middleton,R.J.C.; Macfarlane,D,G.; Robertson,D.A.Range autofocus for linearly frequency-modulated continuous wave radar[J].IEEE IET Radar,Sonar&Nacigation,2011, pp: 288-289.
[9]Chen-Song Chiang, Yu-Jen Chen, Ming-Han Hsieh, et al. The study of auto-focus system for biomedical digital microscope [J].IEEE 2009, pp:1-4.
[10]Fu,G; Corradi,P.;Menciassi,A.; Dario,P.An Integrated Triangulation Laser Scanner for Obstace Detection of Miniature Mobile Robots in Indoor Environment[J]. IEEE/ASME TRANSACTIONS ON MECHATRONICS,2011, pp:778-783.
[11]Lee,Hau-Wei;Chen,Chieh-Li,Chien-Hung.Development of an optical three-dimensional laser tracker using dual modulated diodes and a signal detector[J].IEEE Review of Scientific Instruments,2011, pp:035101-035101-13.
[12]Dimitris Aristos, Spyros Tzasfestas.Simultaneous Object Recognition and Position Tracking for Robotic Application[C].IEEE, Proceedings of the 2009 IEEE International Conference on Mechatronics.Malaga, Spain, April 2009,1-7.
[13]Robert T.McKeon, Patrick J.Flynn.Three-Dimensional Facial Imaging Using a Static Light Screen(SLS) and a Dynamic Subject[J].IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL.59, No.4,2010. pp:774-783.
[14]Chih—Yung Chen, Rey—Chue Hwang, Yu-Ju Chen.A passive auto—focus camera control system [J].Applied Soft Computing(S1568—4946),2010,10(1):296-303.
[15]赵志彬,刘晶红.基于图像处理的航空成像设备自动调焦设计[J].液晶与显示,2010,25(6)：863-868.
[16]毛邦福.显微镜自动对焦系统的设计与优化[D].浙江：浙江大学,2006.
[17]王虎.自动对焦原理及方法[J].科技信息,2008,(13)：33.
[18]Warwick Clegg, David F.L.Jenkins.James F.C.Windmill.A Scanning Laser Microscope System to Observe Static and Dynamic Magnetic Domain Behavior[J].IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL.51, NO.1, FEBRUARY 2002:10-13.
[19]Liu, W. D. Ye, L. M. Liu, K. X. Micro-nano scale ripples on metallic glass induced by laser pulse [J]. IEEE Journal of Applied Physics,2010,043109-043105-5.
[20]Acosta, Juan Camilo; Hwang, Gilgueng, et al.A tuning fork based wide range mechanical characterization tool with nanorobotic manipulators inside a scanning electron microscope[J]. IEEE.Review of Scientific Instruments,2011,035116.
[21]Shanthraj, P.; Rezvanian, O.;Zikry, M. A. Electrothermomechanical Finite-Element Modeling of Metal Microcontacts in MEMS [J].IEEE.Journal of Micro electro mechanical Systems.2011,1-12.
[22]AF-1 自动对焦显微镜[EB/OL].http://www.photroland.com/AF-1.htm.
[23]LasertracTM UTOFOCUS[EB/OL].http://www.axsys.com.
[24]CHR SENSORS MODULAR OPTICAL PENS[EB/OL].Http://www.stilsa.com.
[25]ATF5TM Lazer Autofocus&Tracking System[EB/OL].www.wegudevice.com.
[26]周宇,袁艳,胡煌军,张修宝.数字对焦光场成像清晰度评价方法研究[J].光子学报,2010,39(6)：1094-1098.
[27]郑佳,周建忠,何大华基于视频图象处理的快速对焦方法研究[J].光学与光电工程,2010,8(3)：57-6.
[28]陈晓彬.基于小波的自动对焦算法研究[J].通信技术,2010,6(43)：210-212.
[29]苗立刚,轩波,彭思龙.显微镜的快速自动对焦方法[J].光电子·激光,2007,18(1)：9-10.
[30]肖相国,王忠厚,孙传东,白加光.基于光场擞像技术的对焦测距方法的研究[J].光子学报,2008,37(12)：2539-2543.
[31]梁敏华,吴志勇,陈涛.采用最大灰度梯度法实现经纬仪自动调焦控制[J].光学精密工程,2009,17(12)：3016-3021.
[32]符阿迪,踢文娟,郁梅,蒋刚毅.基于Hadamard变换的数码显微成像自动对焦新方法[J].光电子·激光,2008,19(12)：1690-1693.
[33]倪军,袁家虎,吴饮章.基于边缘特征的光学图像清晰度判定[J].中国激光,2009,36(1)：172-176.
[34]李衡,黄永梅,安涛.基于图象消晰度评价的跟踪望远镜自动调焦算法研究[J].光学仪器,2010, 32(5)：69-73.
[35]陈国金,朱妙芬,施浒立,裘晓光.基于最小各向同性小波滤波的图像清晰度识别[J].光学学报,2008,37(2)：395-398.
[36]常雷,袁艳,张修宝,周志良.基于掩膜的光场成像仿真及采样研究[J].光学学报,2010,39(6)：1104－1110.
[37]轩波,苗立刚,彭思龙.显微镜下两层物体的快速自动对焦算法[J].计算机应用,2007,27(1)：143-148.
[38]吴林涛,蔡盛,乔颜峰.经纬仪分离透镜法自动调焦模型建立与分析[J].激光与红外,2008,38(12)：1241－1244.
[39]孙时珍.基于星间激光测距的三星时差定位系统标校方法[J].计测技术,2010,30(6)：25-28.
[40]邓志红,刘明阳,付梦印.一种改进的视觉传感器与激光测距雷达特征匹配点提取算法[J].光学技术,2010,36(1)：43-47.
[41]刘洞波,刘国荣,喻妙华.一种基于激光测距的机器人地图创建方法[J].光电子·激光,2010,21 (2)： 261-26.
[42]Thomas Layton, Eddie L.Jacobs, Steven T.Griffin.3D Laser Scanner and Profiler for Application to Border Security[J].SAS2009-IEEE Sensors Applications Symposium:9781-4244-2787-1.
[43]陈奕,郭颖,杨俊等脉冲式高精度激光测距技术研究[J].红外,2010,36(6)：1-4.
[44]Yu, Xiaomin;Liao Yang;He, Fei, et al.Tuning etch selectivity of fused silica irradiated by femtosecond laser pulses by controlling polarization of the writing pulses[J].IEEE.Journal of Applied Physics,2011,035114.
[45]周小珊,李岩.相位激光测距与外差干涉相结合的绝对距离测量研究[J].应用激光,2010,31(6)：]013-1017.
[46]SALVADE Y, SCHUHLER N. High-accuracy abso-lute distance measurement using frequency comb referenced multiwavelength source [J]. Applied Optics.2008,47(14):2715-2720.
[47]陈念年,张佳成,范勇,等.一种高精度单点激光三角测距方法[J].计算机测量与控制,2010,18(5)：984-986
[48]F.K.Cui, Z.B.Song, X.Q.Wang, et al.Study on Laser Triangulation Measurement Principle of Three Dimensional Surface Roughness [J].Advanced, Materials Research Vol.136,2010, pp:91-94.
[49]TFT-LCD G6检测设备[EB/OL]. http://www.3i-systems.com/ch/Products.htm
[50]胡长德,赵芙蓉.大行程纳米级乐电微动工作台的设计与试验研究[J].传感技术学报,2009,6：803-807.
[51]卢礼华,梁迎春等.传统控制器实现有摩擦系统的大行程纳米定位[J].纳米技术与精密工程,2010,1：1-6.
[52]林小倩.基于CMOS单点激光三角法测距系统研究[D].浙江：浙江大学,2006.
[53]潘健华.大行程纳米级精密对焦控制系统研究[D].广州：广东工业大学,2010
[54]肖磊,程良伦,范富明TFT-LCD面板反射的能量对光斑图像的影响[J].微型机与应用,2010,29(21)：37-41.
[55]范富明,程良伦,王晓芬等.一种新型光学快速自动聚焦系统[J].光电工程,2010,37(5)：127-132.
[56]姜立军,李哲林,熊忐勇.一种CCD相机快门的自动控制算法[J].光电工程,2010,37(2),PP：141-145.
[57]关澈,王延杰.CCD相机实时自动调光系统.光学精密工程[J],2008,16(2),PP：358-366.
[58]A. Rossi, M.Diani, G.Corsini. Temporal statistics de-ghosting for adaptive non-uniformity correction in infrared focal plane arrays[J]. ELECTRONICS LETTERS 4th March 2010,348-349.
[59]Ai-Mei Huang, Truong Nguyen. Correlation-Based Motion Vector Processing With Adaptive Interpolation Scheme for Motion-Compensated Frame Interpolation[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL.18, NO.4, APRIL.2009,740-752.
[60]Cucchiara R, Grana C, Piccardi M.et al.Detecting Moving Object, Ghost, and Shadows in Video Streams[J].IEEE TEANSON Pattern Analysis and Machine Intelligence,2003,25(10):1337-1342.
[61]Y.P.Guan.Spatio-temporal motion-based foreground segmentation and shadow suppression[C].IEEE, The Institution of Engineering and Technology 2009, IET Comput. Vision,2010, Vol.4, 1ss.1, pp.50-60.
[62]孔兵,王昭,谭玉山.基于圆拟合的激光光斑中心检测算法[J].红外与激光工程,2002,31(3)：275-279.
[63]肖锋钢,刘建国,曾淙泳,刘增东.一种新型提高光斑图像质心精度的去噪方法[J].计算机应用研究,2008,25(12)：3683-3687.
[64]章毓晋图像工程(上册)图像处理(第二版)[M]北京：清华大学出版社,1999.
[65]张震,刘天立,张斌,张淑静,等一种变物距的自动对焦方法[J].机电一体化,2010,4：66-75.
[66]官光勇,何文忠,高旭辉.红外系统中自动调焦爬山搜索算法的优化设计[J].激光与红外,2007,37(11)：1213-1215.
[67]郭军,曾文澜,谢铁邦.基于熵函数的快速自动聚焦方法[J].计量技术,2003,(11)：30-32.
[68]吴小翠,郭斯羽,易国基于边界梯度的自动调焦评价函数[J].计算机系统应用,2010,19(3)：45-48.