基于光学小波变换的机器人视觉的研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
目前,在工业生产中,存在着高温、核辐射、毒化气体等极限工作环境,
    对机器人的智能化提出了更高的要求。为了使机器人具有高度的智能,需要配
    备各种传感器,其中视觉传感器是最主要的。视觉系统处理的信息数据量大,
    这样,如何提高图象的处理速度成为许多科技工作者研究的热点课题,也是亟
    待解决的难题。利用4f光学系统可以完成图象傅立叶变换,实现频谱滤波。
    作者在此方面进行了一些研究工作,主要内容如下:
     设计了光学小波变换系统,同时采用光寻址的空间光调制器(LASLM)
    和电寻址的空间光调制器(EASLM)来调制图象信号,LASLM将物镜的成象
    转为相干光的图象,此图象的读入转换过程在几十毫秒内完成。根据需要利用
    计算机将滤波函数输入到EASLM,使输入图象与滤波函数在频率域相乘,最
    后由CCD接收调制后的图象信号,进行后续处理,整个处理过程约为100毫
    秒,使图象的处理速度成倍的提高。
     鉴于光学小波变换系统是以实现机器人视觉的路径跟踪为主要目标,并且
    考虑Haar小波的边界检测功能,Mexicohat小波边缘增强的作用,对整个光路
    信息传递进行了分析。针对LASLM的空间采样和量化,将电子学中的A/D转
    换去假频前置滤波器的设计应用于光路分析中,为结构设计提供理论依据。同
    时对量化噪声的形成进行了分析,确定为高斯噪声。
     用计算机进行了仿真实验,得到了对图象的边缘提取,验证了光学系统设
    计的可行性。同时仿真结果证明,对于低频图象而言,选取合适的小波尺度因
    子,Haar小波和Mexicohat小波均可实现对图象边缘的检测和提取,并且能够
    有效地抑制噪声。
In recent years, in industry production, there has limited working surroundings
     which are high temperature ~. nuclear radiation poisonous gas, it needs more
     requirements to robot. In order to make high intelligent for robot, we should fit
     various sensors on robot, the most important of all is visual sensor. The information
     data of robotic visual system is very large, thus, how to improve the speed of image
     processing has been the studying topics and is a difficult problem to be overcome.
     We could achieve the Fourier transform of image and filter in frequency spectrum by
     using 4f optical system. The author do some studying works in this domain, the
     main contents are the follows:
    
     The author has designed the system of optical wavelet transform, put forward
     using spatial light modulation of light address and spatial light modulation of electric
     address to modulate signals of image at one time. LASLM transform the image,
     which is the object image, into coherent image, this process of read-in and
     transforming could be finished in a few milliseconds. According to the need, we can
     input various functions into EASLM by using computer, and the input image
     multiplied by filter function in the frequency domain, at last, the signal of modulated
     image is incepted by Charge Coupled Device (CCD) and is processed in next steps.
     The time of whole processing is almost 100 milliseconds, the speed of image
     processing could raise a lots times.
    
     The main goal of this system based on optical wavelet transform is path tracks
     of robotic vision, and Haar wavelet has the edge of detection properties, Mexicohat
     wavelet has the edge of enhancement. Aiming at spatial sampling and quantification,
     the whole transfer of light path information has been analyzed. The author puts
     3
    
     forward using the design of A/D transfer antialiasing prefilters, which has been used
    
     in electronics, in the analysis of light path, thus we have theoretic foundation in the
     design of structure, and analyze quantification noises, which is the Gauss noise.
    
     We had gained of the image edge by using computer simulation, it proved
     feasibility of the optical system design. In the same time, for the low frequent image,
    
    
     II
    
    
    
    
    
    
    
    
    
     if we select suitable scale of Haar wavelet (or Mexicohat wavelet), we can detect the
     edge of image, and the noises could be effective restrain.
引文
[1] .F.T.S.Yu,Optics and Information Theory,R.E.Kierger,Malabar,Florida, 1984.
    [2] .Young R. An introduction to nonlinear Fourier Series. NEW YORK, Academic Press,1980,1-150
    [3] .Morlet J. Wave propagation and sampling theory and complex waves.Geophysics,1982,47(2) :222-236.
    [4] .Daubechies I. Orthonormal bases of supported wavelets. Commm. on Pure and Appl.Math.,1998,41(7) :909-996.
    [5] .Burt P J.Adelson E.The Laplacian Pyramid as a compact image code.IEEE Trans on communications,1983,31(4) :532-540.
    [6] .Mallat S. A theory for multi-resolution signal decomposition the wavelet representation.IEEE Trans.on PAMI,1989,11(7) :674-693.
    [7] .Mallat S. Multi-frequency channel decompositions of image sand signal Processing.1989,37(12) :2091-2110.
    [8] .Meyer Y. On delette et operateurs.Paris:Hermann Press. 1990,1-200.
    [9] .Chui C K. An introduction to wavelets. NEW YORK: Academic Press, 1992,1-200.
    [10] .Goswami J C,Chui C K. On solving first-kind integral equations using wavelets on a boundies of computation.1996,65(214) :723-737.
    [11] .沈定刚,戚飞虎,柳振宇.小波变换在旋转不变性目标识别中的应用.模 式识别与人工智能,1995,8(1) :14-20.
    [12] .胡正仪,王延平. 用小波锥分解提出旋转和尺度不变特征. 武汉大学 学报(自然科学版),1994,6:45-50。
    [13] .周杰,彭嘉雄,丁明跃。方向小波变换及其在运动弱目标检测中的应用。 信息与控制,1996,25(1) :21-27。
    [14] .王玉平,蔡元龙. 多尺度B样条小波边缘检测算子.中国科学(A辑), 1995. 25(4) :426-437.
    
    
    [15] .贾天旭,郑南宁.基于Bubble小波多尺度边缘提取.电子学报.1996, 24(4) :117-121.
    [16] .周杰,彭嘉雄,丁明跃.基于小波特征的图象匹配方法.模式识别与人 工智能.1996,9(2) :125-129.
    [17] .何立,王延平. 利用小波变换和约束矩阵进行图象压缩编码.电子学 报.1995,23(4) :20-23.
    [18] .倪林,边肇祺,王文渊.基于小波变换的级间相关并行矢量量化方法.模 式识别与人工智能.1995,8(1) :57-63.
    [19] .倪林,王文渊,边肇祺.小波变换域的尺度自适应矢量量化方法.模式 识别与人工智能.1995,8(3) :188-194
    [20] .钟声,石青云,程民德.基于具有更加频率分辨率的小波变换和自适应 矢量量化技术的图象压缩系统.模式识别与人工智能,1993,6(4) :291-299
    [21] .胡国生,鲍虎军.彩色图象的小波变换编码.自然科学进展-国家重点 实验室通讯,1993,3:26-34
    [22] .朱宇霞,蔡德钧,罗和平. 一种用小波分析的彩色图象压缩法.华中理 工大学学报,1994,22(11) :11-21
    [23] .陶德元,何小海,李书平等. 小波变换及其在图象处理中的应用.四川 大学学报(自然科学版),1994,31(4) :481-488版),1994,31(4) : 481-488
    [24] .王延平,梅雁诚,梅林. 图象的小波锥分解及其矢量量化.武汉大学学 报(自然科学版),1993,6:41-48
    [25] .张平,朱光喜,朱耀庭. 分形用于图象边缘检策的研究.华中理工大学 学报,1993,21(2) :78-102
    [26] .黎洪,全子一. 图象矢量量化-频率敏感自组织特征映射算法.通信学 报。1995,16(2) :59-64
    [27] .肖扬,杜锡.线性相位二维跳耦型数路滤波器组设技及其在图象子波带 编码中的应用.通信学报,1995,16(2) :65-70
    [28] .王卫,蔡德钧,万发贯. 一种多分辨率图象混合编码方案.通信学报, 1995. 16(2) :71-78
    
    
    [29] .薛向阳,陈学青,樊昌信.基于旋转Barnes-wall格的增益-波形矢量量 化器及其在序列图箱序列图象编码中的应用.电子科学学刊,1995,17 (6) :569-76
    [30] .刘政权,童调生,彭榕蓉等.图象拼接的小波加快算法研究.计算技术 及自动化,1994,13(4) :1-4
    [31] .方斌. 一种屏幕图文高效压缩存储系统的实现.电子科学学刊,1995, 17(4) :417-419
    [32] .王玉平,蔡元龙,耿中行.基于小波变换的滤波方法.信息与控制,1996, 25(4) :199-205
    [33] .李建平,杨晓帆,陈延槐等.基于小波变换和神经网络的图象压缩.见: 靳蕃,范俊波.神经网络理论与应用研究’96. 成都:西南交通大学出版 社,1996,436-439
    [34] .李建平,张万苹,陈延槐等. 基于小波和神经网络的图象细化算法.后 勤工程学院学报,1997,13(2) :30-34
    [35] .罗会国,朱耀庭,朱光喜等.信号的一种快速子波分解.电子学报,1994, 22(7) :17-22
    [36] .王桥,胡敏钜.分数布郎运动的小波分析.武汉测绘科技大学学报,1995, 20(1) :66-70
    [37] .肖化铸. 小波分析中信号的时频局部化.电子科技大学学报,1995,24 (3) :328-332
    [38] .耿中行,屈梁生. 小波包原理及其在机械故障诊断中的应用.信号处理, 1994,10(4) :244-249
    [39] .蔡志强,吴雅,熊光荣等.突变性特征故障识别的两种实用小波技术.机 械工程学报,1993,29(6) :58-64
    [40] .李勇,华蕴博. 一个新的脑电分析系统-小波分析理论的运用.生物物 理学报,1996,12(2) :367-370
    [41] .李建平,张万苹,陈延槐等.基于小波理论的平面叶栅优化设计.重庆 大学学报(自然科学版),1997,20(2) :63~68
    [42] .李建平,张万苹,陈延槐等. 基于小波分析的双重孔隙储集层系统的数 学模型求解. 系统科学与数学,1997,17(3)
    
    
    [43] .李建平,张万苹,陈延槐等. 基于小波变换的双重孔隙储集层系统的参 数辨识.系统科学与数学,1997,17(4)
    [44] .李建平,陈延槐,徐问之.小波分析及其应用.后勤工程学院学报,1996, 12(3) :20-25
    [45] .陆建峰,刘克,杨静等.直接三维重建研究.计算机研究与发展,1996, 33(12) :933-939
    [46] .Burns,T.,Feilding,K.,Rogeers,S.,Pinski,S.,Buck,D. Optical Haar wavelet transform.Opt Eng,1992(31) ,1852-1858
    [47] .Pinski,S.D.,Rogers,S.D.,Ruck,S.K.,Welsh,D.M.,Kabrinski,B.M.,Warhola,M.,Q uinn,G.,Oxley,D.Image segementation using optical wavelets.SPIE,Proc, 1702(1992) .11-19
    [48] .Freysz E.Optical wavelet transform of fractal aggregates.Phys Rev Lett. 64(1990) ,7745-7748
    [49] .McAulay,A.,Wang,J.Optical wavelet transform classifier with positive real Fourier transform wavelets.Opt Eng,32(1993) ,1333-1339
    [50] .Sheng Y.,Lu,T.,Roberge,D.Optical N4implementation of a two-dimensional wavelet transform.Opt Eng.,31(1992) ,1859-1864
    [51] .陈鹤鸣,贾小兵,施伟华. 应用计算全息制作的子波匹配滤波器实现光 学子波变换的研究.光学学报, Vol18,No.8 (1998) 1087-1089
    [52] .C.Decusatis,J.Kony,P.Das.Hybird optical implementation of discrete wavelet transforms:a tutorial.Optics & laser Technology,Vo.128,No.2,(1996) 51-58
    [53] .朱传贵,薛鸣球,刘德森,高应俊. 应用平面透镜阵列的多通道图象相 关识别.
    [54] .Dong xue Wang,Ju wei Tai,Yan xin Zhang.Two-dimensional optical wavelet transform in space domain and its performance analysis. Applied Optics.(1994) Vo1. 33,No.23,5271-5274
    [55] .吕铁英,彭嘉雄. 一种基于数学形态学的图象多尺度分析方法研究. 数 据采集与处理, (1998) Vo1. 33,No.2,107-11O
    [56] .汪巧萍,王宏远. 一种基于小波分析的纹理背景下的物体分割算法. 数
    
     据采集与处理,(1998) Vol.13,No.1,12-16
    [57] .Robin N.Strickland,Senior Member.Wavelet transform Methods for object Detection and Recovery.IEEE,and Heell Hahn.IEEE Transactions on image processing.,Vol.6,No.5,(1997) ,724-735
    [58] .Boon Yi Soon,Mohammed S.Akam,and Mohammed A.Karim.Improved feature extraction by use of a joint wavelet transform correlator.Applied Optics,Vol.37,No.5,(1998) ,821-827
    [59] .Wenlu wang,Guofan Jin,Yingbai Yan,Minxian wu.Image feature extraction with the optical Haar wavelet transform.Opt Eng.,(1995) ,1238-1242
    [60] .李育林,傅晓理著.空间光调制器及其应用,国防工业出版社.