多普勒天气雷达产品终端及其回波三维显示方法的设计
|
摘要
随着多普勒天气雷达的应用越来越广泛,业务应用和理论研究的快速发展,雷达资料的分析处理水平也得到了很大的提高。本文设计的多普勒天气雷达产品终端系统既满足了业务上的需求,实现了雷达回波实时显示、雷达远程控制、任意两点测距、鼠标联动、基数据放大、游标读值、产品动画、基数据备份、地图编辑、危险天气警报、产品交互(VAD、弱回波区、多层CAPPI、垂直剖面、等值线)等功能,丰富了气象算法、美化了系统界面、简化了操作步骤;同时也满足科研的需求,对基数据进行标准化处理以兼容不同的数据格式,对标准格式的数据进行了多种预处理以提高数据质量,对气象算法进行动态链接库封装并用插件技术进行组装以提高气象算法增减和升级的便利性。整套系统采用多线程、多窗口、多任务、交互式、模块化、插件化等技术,具有较好的兼容性、可移植性、可扩展性,系统运行稳定,操作简便。
目前我国大部分雷达气象产品还是以二维的形式来表示三维结构,其最具代表性的产品就是RHI,WER,多层CAPPI。这些产品虽然可以部分的解决对回波三维结构的分析难题,但还是不够细致,不够方便。光线投射(Ray-Casting)算法能产生三维数据场的整体图像,包括每一个细节,其原理简单、易于实现、可以生成高质量的显示图像,但计算时间较长。本文通过一系列的方法来加速光速投射算法,例如:确定体数据在屏幕上的投影面积来确定投射光线的数量,并采用间隔投射再插值的办法来减少光线投射的次数;通过计算投射光线与体数据的交点来确定初始采样点,使用基于空间跳跃技术的Proximity Clouds算法来确定体数据中的采样点,并设定不透明度阈值来提前结束采样,从而有效减少了投射光线上的合成计算。结合其他一些加速操作,有效地提高了绘制速度,并且保证了较高好的图像质量。
With the more and more broad application of Doppler weather radar, and rapid development of radar forecast operational application and theory study, the level of Doppler data analysis has increased greatly. To fulfill the operational application and the scientific research requirements, the Doppler weather radar production terminal is designed in this paper. The terminal realizes many functions such as radar echoes real-time display, radar remote control, random two-point ranging, mouse linkage, raw data magnification, cursor valuation, product animation, raw data backup, map edit, dangerous weather alerting, interactive product(VAD, WER, M-CAPPI, VCS, Contour) and others, expands meteorology algorithms, beautifies system's GUI, simplifies operational steps, all of these meet the needs of operational application. The terminal also has many preferable designs to fulfill the scientific research requirements, such as standardization of raw data to integrate different raw data formats, many pre-processing methods are used to control the data quality, meteorology algorithms are encapsulated with dynamic link library and assembled with plug-in technique to promote the convenience of meteorology algorithms' adding, reducing and upgrading. Multi-thread, multi-window, multi-task, interaction, modularization, plug-in, and other techniques are adopted in the whole terminal system. The system can be expanded and transplanted. Besides, it's a convenient system which has better compatibility, stability and convenient.
Most radar products are displayed in two-dimension at home which can't exhibit radar echoes' three-dimension structure in focus, the most representational products are RHI(Range Height Indicator), WER(Weak Echo Region), M-CAPPI(Multi-Constant Altitude Plan Position Indicator). These products partly resolved the difficult of analyzing echoes' 3-D structure, but it's not precise or convenient. Ray-Casting algorithm can get the whole image of 3-D data with every details, it has some other advantages, like simple principle, easy realization and high quality image, but its calculation is too large. A series of methods are adopted in this paper to accelerate ray-casting algorithm. Confirming the stared out number of ray by calculating data projective aero on screen, interpolating after interval stares out rays are used to reduce the frequency of staring out rays. Confirming the start point of re-sample by calculate point of intersection of ray and data set, confirming the re-sample points in the data set by proximity clouds algorithm based on space leaping, finishing re-sample by setting opacity threshold, all of above are used to reduce re-sample calculation of ray. Other accelerated techniques are also adopted to improve the rendering speed and maintain high image quality.
目前我国大部分雷达气象产品还是以二维的形式来表示三维结构,其最具代表性的产品就是RHI,WER,多层CAPPI。这些产品虽然可以部分的解决对回波三维结构的分析难题,但还是不够细致,不够方便。光线投射(Ray-Casting)算法能产生三维数据场的整体图像,包括每一个细节,其原理简单、易于实现、可以生成高质量的显示图像,但计算时间较长。本文通过一系列的方法来加速光速投射算法,例如:确定体数据在屏幕上的投影面积来确定投射光线的数量,并采用间隔投射再插值的办法来减少光线投射的次数;通过计算投射光线与体数据的交点来确定初始采样点,使用基于空间跳跃技术的Proximity Clouds算法来确定体数据中的采样点,并设定不透明度阈值来提前结束采样,从而有效减少了投射光线上的合成计算。结合其他一些加速操作,有效地提高了绘制速度,并且保证了较高好的图像质量。
With the more and more broad application of Doppler weather radar, and rapid development of radar forecast operational application and theory study, the level of Doppler data analysis has increased greatly. To fulfill the operational application and the scientific research requirements, the Doppler weather radar production terminal is designed in this paper. The terminal realizes many functions such as radar echoes real-time display, radar remote control, random two-point ranging, mouse linkage, raw data magnification, cursor valuation, product animation, raw data backup, map edit, dangerous weather alerting, interactive product(VAD, WER, M-CAPPI, VCS, Contour) and others, expands meteorology algorithms, beautifies system's GUI, simplifies operational steps, all of these meet the needs of operational application. The terminal also has many preferable designs to fulfill the scientific research requirements, such as standardization of raw data to integrate different raw data formats, many pre-processing methods are used to control the data quality, meteorology algorithms are encapsulated with dynamic link library and assembled with plug-in technique to promote the convenience of meteorology algorithms' adding, reducing and upgrading. Multi-thread, multi-window, multi-task, interaction, modularization, plug-in, and other techniques are adopted in the whole terminal system. The system can be expanded and transplanted. Besides, it's a convenient system which has better compatibility, stability and convenient.
Most radar products are displayed in two-dimension at home which can't exhibit radar echoes' three-dimension structure in focus, the most representational products are RHI(Range Height Indicator), WER(Weak Echo Region), M-CAPPI(Multi-Constant Altitude Plan Position Indicator). These products partly resolved the difficult of analyzing echoes' 3-D structure, but it's not precise or convenient. Ray-Casting algorithm can get the whole image of 3-D data with every details, it has some other advantages, like simple principle, easy realization and high quality image, but its calculation is too large. A series of methods are adopted in this paper to accelerate ray-casting algorithm. Confirming the stared out number of ray by calculating data projective aero on screen, interpolating after interval stares out rays are used to reduce the frequency of staring out rays. Confirming the start point of re-sample by calculate point of intersection of ray and data set, confirming the re-sample points in the data set by proximity clouds algorithm based on space leaping, finishing re-sample by setting opacity threshold, all of above are used to reduce re-sample calculation of ray. Other accelerated techniques are also adopted to improve the rendering speed and maintain high image quality.
引文
[1]张培昌,杜秉玉,戴铁王编著.雷达气象学[M].北京:气象出版社,2001.
[2]朱乾根等.天气学原理与方法,第三版[M].北京:气象出版社,2000.
[3]俞小鼎,姚秀萍,熊廷南等.多普勒天气雷达原理与业务应用[M].北京:气象出版社,2006.1.
[4]胡胜.多普勒天气雷达产品软件的开发与设计[D].南京气象学院硕士研究生毕业论文,2000.
[5]胡胜.多普勒雷达资料业务应用研究[D].南京信息工程大学理学博士学位论文,2006.
[6]胡明宝,高太长,汤达章,多普勒天气雷达资料分析与应用[M].北京:解放军出版社.
[7]沃伟峰.多普勒天气雷达二次产品开发平台的设计和实现[D].南京气象学院硕士毕业论文,2003.
[8]吴蕾.多普勒天气雷达二次产品处理模块的工程设计[D].南京气象学院硕士毕业论文,2003.
[9]刘淑媛,孙健,郭卫东等.多普勒雷达数据处理显示系统[J].气象,30(7):44-47.
[10]叶小玲,吉微,李玮等.多普勒雷达显示系统的设计与实现[J].微计算机信息,2007,23(8-1):305-306.
[11]牛丽莉.多普勒天气雷达二次产品处理系统界面设计[D].南京信息工程大学硕士毕业论文,2006.
[12]Hasler A F,Pierce H,Morris K R.Meteorological data field "in perspective"[J].Bull Amer Meteor Soc,1985,66:795-801.
[13]Hibbard W.Computer generated imagery for 4-D meteorological data[J].Bull Amer Meteor Soc,1986,67:1362-1369.
[14]Hibbard W,Santaek D.Visualizing large data set in the earth science.IEEE computer,1989,22(8):53-57.
[15]James A S,Thomas V P.Visualizing meteorological data[J].Bull Amer Meteor Soc,1990,71:1012-1020.
[16]Gilbert Camara,Lubia Vinhas,et al.Multidimensional data visualization in meteorology,sixth workshop.Proceedings on Meteorological Operational system ECM WF,November 1997,4:115-118.
[17]Philip C Chen.Portable and Desktop 3D Weather data Visualization systems in UNIX and non-UNIX Environments.proceedings on Meteorological Operational system of ECM WF,November 1997,7:195-201.
[18]William Ribarsky.The times they are a-changing:PC graphics moves[J].IEEE-Computer Graphics and Application,1998 May/June,6:20-25.
[19]Wang Hongqing,Kai-Hon Lau,Wai-Man Chan,et al.A PC-based Visualization System for Coastal Ocean and Atmospheric Modeling.proceedings of 6~(th)International Estuarine and Coastal Modeling Conference.New Orlean USA/1999,1233-1243.
[20]王洪庆,张焱,陶祖钰等.五维大型复杂数据集计算机可视化[J].自然科学进展,1998,8(6):742-747.
[21]王洪庆,张焱,陶祖钰.黄海气旋数值模拟的可视化[J].应用气象学报,2000,11(3):280-286.
[22]唐泽圣等.三维数据场可视化[M].北京:清华大学出版社.1999.
[23]Rajon DA,Bolch WE.Marching cube algorithm:review and tri-linear interpolation adaptation for image-based dosimetric models[J].Computerized Medical Imaging and Graphics,2003,27(5):411-435.
[24]Doi A,Koide A.An efficient method of triangulating equi-valued surfaces by using tetrahedral cells[J].IEICE Transactions,1991,74(1):214-224.
[25]Cline H,Lorensen W.Two algorithms for three-dimensional reconstruction of tomograms[J].Medical Physics,1988,15(3):320-327.
[26]Levoy M.Efficient ray tracing of volume data[J].ACM Transactions on Graphics,1990,9(3):245-261.
[27]Parker S,Parker M,Livnat Y,et al.Interactive ray tracing of volume visualization[J].IEEE Transactions on Visualization and Computer Graphics,1999,5(3):238-250.
[28]Westover L.Footprint evaluation for volume rending[J].Computer Graphics,1990,24(4):309-318.
[29]Lacroute P,Levoy M.Fast volume rendering using a shear-Warp factorization of the viewing transformation[C].Computer Graphics,Proceedings,1994:451-458.
[30]Schmidt AEF,Gattass M,Carvalho PCP.Combined 3D visualization of volume data and polygonal models using a Shear-Warp algorithm[J].Computers & Graphics,2000,24(4):583-601.
[31]Shiaofen Fang,Rajagopalan Srinivasan,Su Huang,et al.Deformable Volume Rendering by 3D texture mapping and octree encoding[J].IEEE Visualization,1996:73-80.
[32]周红平,沙雪松,张光峰.批次处理解决气象雷达距离模糊[J].现代雷达,2006,28(11):20-21。
[33]曹俊武.VAD技术在3830多普勒天气雷达中的应用[J].现代电子,2000,72(3):9-13.
[34]李杰,顾松山,胡中兵等.VAD技术三种仰角策略反演风场的比较和应用[J].南京气象学院学报,2006,29(2):220-227。
[35]胡志群,夏文梅,杨昌年等.用改善的EVAD技术和变分法计算大气垂直速度[J].南京气象学院学报,2005,28(3):344-350.
[36]刘式达,辛国君.大气中尺度气旋的三维螺旋结构理论[J].气象学报,2000,58(2):151-158.
[37]陶岚.多普勒天气雷达中的中尺度气旋识别(D].南京信息工程大学理学硕士学位论文,2006.
[38]杨引明.风暴的多普勒雷达自动识别、跟踪、预报及冰雹探测[D].南京气象学院硕士学位论文,1998。
[39]陈章法.基于多普勒天气雷达的风暴识别与设计[D].南京气象学院硕士毕业论文,2005.
[40]徐春芳。多普勒天气雷达的风暴单体自动识别算法及工程设计[D].南京信息工程大学理学硕士学位论文,2006.
[41]胡玲.超级单体雹暴的多普勒雷达特征分析[A].见:中国气象局北京城市研究所编.中美强对流天气临近预报技术国际研讨会文集[C].北京:气象出版社,2004,328.
[42]吴林林.新一代天气雷达冰雹探测算法及在业务中的应用[J].气象,2006,32(1):51-55.
[43]倪允琪,周秀骥。中国长江中下游梅雨锋暴雨形成机理以及监测与预测理论和方法研究[J].气象学报,2004,62(5):647-661.
[44]夏文梅,张亚萍,汤达章.暴雨多普勒天气雷达资料的分析[J].南京气象学院学报,2002,25(6):787-794.
[45]Rinehart R E,Garvey E T.Three-dimensional storm motion detection by conventional weather radar[J].Nature,1978,273:287-289.
[46]陈明轩,王迎春,俞小鼎.交叉相关外推算法的改进及其在对流临近预报中的应用[J].应用气象学报,2007,18(5):690-700.
[47]严红梅.用相关法估测CAPPI等高面上的水平流场[D].南京信息工程大学理学硕士学位论文,2006.5.
[48]Dave Shreiner,Mason Woo,Jackie Neider,et al.邓郑祥译.OpenGL编程指南[M],第四版.北京,人民邮电出版社,2005.
[49]全海英.基于微机的CT图像序列快速直接体绘制方法研究[D].中国科学院博士学位研究生学位论文,2001.
[50]晏靖靖.三维地震数据场直接体绘制的方法研究[D].南京理工大学硕士学位论文,2004.
[51]金朝阳,王建中.自适应光线投射直接体绘制算法及实现[J].中国医学影像技术,2005,21(4):634-638.
[52]Cohen D,Sheffer Z.Proximity clouds,an acceleration technique for 3D grid traversal[J].The Visual Computer,1994,11(1):27-38.
[53]Semwal S K,Kvamstrom H.Directed safe zones and the dual extent algorithms for efficient grid traversal during ray tracing[A].In:Proceedings of Graphics Interface'97[C],Ke-lowna,Canada,1997:68-75.
[54]胡英,王怀志,徐心和.基于光线投射技术的医学图像三维重建[J].东北大学学报(自然科学版),2004,25(1):1-4.
[55]谭浩强.C程序设计[M],第二版.北京:清华大学出版社,1999.
[56]郑莉,董渊.C++语言程序设计[M],第2版.北京:清华大学出版社,2001.
[57]Michael J.Young,VisualC++6从入门到精通[M],北京:电子工业出版社,1999.
[58]David J.Kruglinski,Scot Wingo,George Shepherd,Visual C++6.0技术内幕[M],第五版.北京:北京希望电子出版社,1999.
[59]王育坚编著.Visual C++一面向对象编程教材[M].北京:清华大学出版社,2003.
[2]朱乾根等.天气学原理与方法,第三版[M].北京:气象出版社,2000.
[3]俞小鼎,姚秀萍,熊廷南等.多普勒天气雷达原理与业务应用[M].北京:气象出版社,2006.1.
[4]胡胜.多普勒天气雷达产品软件的开发与设计[D].南京气象学院硕士研究生毕业论文,2000.
[5]胡胜.多普勒雷达资料业务应用研究[D].南京信息工程大学理学博士学位论文,2006.
[6]胡明宝,高太长,汤达章,多普勒天气雷达资料分析与应用[M].北京:解放军出版社.
[7]沃伟峰.多普勒天气雷达二次产品开发平台的设计和实现[D].南京气象学院硕士毕业论文,2003.
[8]吴蕾.多普勒天气雷达二次产品处理模块的工程设计[D].南京气象学院硕士毕业论文,2003.
[9]刘淑媛,孙健,郭卫东等.多普勒雷达数据处理显示系统[J].气象,30(7):44-47.
[10]叶小玲,吉微,李玮等.多普勒雷达显示系统的设计与实现[J].微计算机信息,2007,23(8-1):305-306.
[11]牛丽莉.多普勒天气雷达二次产品处理系统界面设计[D].南京信息工程大学硕士毕业论文,2006.
[12]Hasler A F,Pierce H,Morris K R.Meteorological data field "in perspective"[J].Bull Amer Meteor Soc,1985,66:795-801.
[13]Hibbard W.Computer generated imagery for 4-D meteorological data[J].Bull Amer Meteor Soc,1986,67:1362-1369.
[14]Hibbard W,Santaek D.Visualizing large data set in the earth science.IEEE computer,1989,22(8):53-57.
[15]James A S,Thomas V P.Visualizing meteorological data[J].Bull Amer Meteor Soc,1990,71:1012-1020.
[16]Gilbert Camara,Lubia Vinhas,et al.Multidimensional data visualization in meteorology,sixth workshop.Proceedings on Meteorological Operational system ECM WF,November 1997,4:115-118.
[17]Philip C Chen.Portable and Desktop 3D Weather data Visualization systems in UNIX and non-UNIX Environments.proceedings on Meteorological Operational system of ECM WF,November 1997,7:195-201.
[18]William Ribarsky.The times they are a-changing:PC graphics moves[J].IEEE-Computer Graphics and Application,1998 May/June,6:20-25.
[19]Wang Hongqing,Kai-Hon Lau,Wai-Man Chan,et al.A PC-based Visualization System for Coastal Ocean and Atmospheric Modeling.proceedings of 6~(th)International Estuarine and Coastal Modeling Conference.New Orlean USA/1999,1233-1243.
[20]王洪庆,张焱,陶祖钰等.五维大型复杂数据集计算机可视化[J].自然科学进展,1998,8(6):742-747.
[21]王洪庆,张焱,陶祖钰.黄海气旋数值模拟的可视化[J].应用气象学报,2000,11(3):280-286.
[22]唐泽圣等.三维数据场可视化[M].北京:清华大学出版社.1999.
[23]Rajon DA,Bolch WE.Marching cube algorithm:review and tri-linear interpolation adaptation for image-based dosimetric models[J].Computerized Medical Imaging and Graphics,2003,27(5):411-435.
[24]Doi A,Koide A.An efficient method of triangulating equi-valued surfaces by using tetrahedral cells[J].IEICE Transactions,1991,74(1):214-224.
[25]Cline H,Lorensen W.Two algorithms for three-dimensional reconstruction of tomograms[J].Medical Physics,1988,15(3):320-327.
[26]Levoy M.Efficient ray tracing of volume data[J].ACM Transactions on Graphics,1990,9(3):245-261.
[27]Parker S,Parker M,Livnat Y,et al.Interactive ray tracing of volume visualization[J].IEEE Transactions on Visualization and Computer Graphics,1999,5(3):238-250.
[28]Westover L.Footprint evaluation for volume rending[J].Computer Graphics,1990,24(4):309-318.
[29]Lacroute P,Levoy M.Fast volume rendering using a shear-Warp factorization of the viewing transformation[C].Computer Graphics,Proceedings,1994:451-458.
[30]Schmidt AEF,Gattass M,Carvalho PCP.Combined 3D visualization of volume data and polygonal models using a Shear-Warp algorithm[J].Computers & Graphics,2000,24(4):583-601.
[31]Shiaofen Fang,Rajagopalan Srinivasan,Su Huang,et al.Deformable Volume Rendering by 3D texture mapping and octree encoding[J].IEEE Visualization,1996:73-80.
[32]周红平,沙雪松,张光峰.批次处理解决气象雷达距离模糊[J].现代雷达,2006,28(11):20-21。
[33]曹俊武.VAD技术在3830多普勒天气雷达中的应用[J].现代电子,2000,72(3):9-13.
[34]李杰,顾松山,胡中兵等.VAD技术三种仰角策略反演风场的比较和应用[J].南京气象学院学报,2006,29(2):220-227。
[35]胡志群,夏文梅,杨昌年等.用改善的EVAD技术和变分法计算大气垂直速度[J].南京气象学院学报,2005,28(3):344-350.
[36]刘式达,辛国君.大气中尺度气旋的三维螺旋结构理论[J].气象学报,2000,58(2):151-158.
[37]陶岚.多普勒天气雷达中的中尺度气旋识别(D].南京信息工程大学理学硕士学位论文,2006.
[38]杨引明.风暴的多普勒雷达自动识别、跟踪、预报及冰雹探测[D].南京气象学院硕士学位论文,1998。
[39]陈章法.基于多普勒天气雷达的风暴识别与设计[D].南京气象学院硕士毕业论文,2005.
[40]徐春芳。多普勒天气雷达的风暴单体自动识别算法及工程设计[D].南京信息工程大学理学硕士学位论文,2006.
[41]胡玲.超级单体雹暴的多普勒雷达特征分析[A].见:中国气象局北京城市研究所编.中美强对流天气临近预报技术国际研讨会文集[C].北京:气象出版社,2004,328.
[42]吴林林.新一代天气雷达冰雹探测算法及在业务中的应用[J].气象,2006,32(1):51-55.
[43]倪允琪,周秀骥。中国长江中下游梅雨锋暴雨形成机理以及监测与预测理论和方法研究[J].气象学报,2004,62(5):647-661.
[44]夏文梅,张亚萍,汤达章.暴雨多普勒天气雷达资料的分析[J].南京气象学院学报,2002,25(6):787-794.
[45]Rinehart R E,Garvey E T.Three-dimensional storm motion detection by conventional weather radar[J].Nature,1978,273:287-289.
[46]陈明轩,王迎春,俞小鼎.交叉相关外推算法的改进及其在对流临近预报中的应用[J].应用气象学报,2007,18(5):690-700.
[47]严红梅.用相关法估测CAPPI等高面上的水平流场[D].南京信息工程大学理学硕士学位论文,2006.5.
[48]Dave Shreiner,Mason Woo,Jackie Neider,et al.邓郑祥译.OpenGL编程指南[M],第四版.北京,人民邮电出版社,2005.
[49]全海英.基于微机的CT图像序列快速直接体绘制方法研究[D].中国科学院博士学位研究生学位论文,2001.
[50]晏靖靖.三维地震数据场直接体绘制的方法研究[D].南京理工大学硕士学位论文,2004.
[51]金朝阳,王建中.自适应光线投射直接体绘制算法及实现[J].中国医学影像技术,2005,21(4):634-638.
[52]Cohen D,Sheffer Z.Proximity clouds,an acceleration technique for 3D grid traversal[J].The Visual Computer,1994,11(1):27-38.
[53]Semwal S K,Kvamstrom H.Directed safe zones and the dual extent algorithms for efficient grid traversal during ray tracing[A].In:Proceedings of Graphics Interface'97[C],Ke-lowna,Canada,1997:68-75.
[54]胡英,王怀志,徐心和.基于光线投射技术的医学图像三维重建[J].东北大学学报(自然科学版),2004,25(1):1-4.
[55]谭浩强.C程序设计[M],第二版.北京:清华大学出版社,1999.
[56]郑莉,董渊.C++语言程序设计[M],第2版.北京:清华大学出版社,2001.
[57]Michael J.Young,VisualC++6从入门到精通[M],北京:电子工业出版社,1999.
[58]David J.Kruglinski,Scot Wingo,George Shepherd,Visual C++6.0技术内幕[M],第五版.北京:北京希望电子出版社,1999.
[59]王育坚编著.Visual C++一面向对象编程教材[M].北京:清华大学出版社,2003.