强对流云体三维结构的可视化方法及其构成特征的提取
|
摘要
天气雷达是探测降水系统的主要手段,是对强对流天气(冰雹、大风、龙卷和暴洪)进行监测和预警的主要工具。
为了以图像的形式从各个视角直观展现雷达基数据文件所蕴含的信息,以辅助气象预报和研究工作,本文主要做了以下几方面的研究和探索。
1.构建并显示雷达反射率图像和径向速度图像
我国新一代天气雷达的优点之一是基数据为数字化的,经过算法的处理可以形成一系列的产品。本文依据雷达基数据产品文件结构,从中读取数据信息并生成反射率图像和径向速度图像。经测试,一组18张图像的总体生成时间小于3秒,图像质量及真实率达到已有产品的水平。
2.实现基于多锥面数据的任意三维剖面
为了使云体的立体结构和回波强度分布可视化,本文汇集各仰角即多锥面的数据信息研究云体的三维剖面图像的生成方法。与已有的剖面图不同的是,本文将剖面置于三维坐标系中,使其携带有丰富的空间位置信息;同时因剖面截线的长度和方位不受任何限制,又使其具有了很强的灵活性。
3.基于三维剖面的云体结构可视化方法
在实现了一个位置上的静止剖面图之后,令剖面的截线沿径向平行移动,随之形成一个剖面系列组,再根据气象工作者的专业知识和经验,确定最佳的截线角度;根据视觉特点,确定最佳的观测角度,从而实现了对所剖云体在最佳剖面角度和最佳观测视角下的云体构成的展现。
4.分析并提取云体三维构成特征
云体在三维构成上有很多特征有待研究、归纳和自动提取,及时恰当地将它们提取出来,将有助于定量地描述云体的三维结构,有助于对天气状况的分析和理解,有助于提高预警预报强对流天气的准确率。本文对此项工作做了初步尝试,定义并提取了反映云体三维构成的能量、厚度、高度及其变化特征。
本文工作均在Visual C++ 6.0软件开发平台下实现,实验测试结果达到预期目标。
The pulse-Doppler radar is widely used in weather investigation, is the main instrument that can detect precipitation system and monitor Severe Convective Weather (including hailstorm, gale, cyclone, violent flood and so on).
In order to exhibit information contained in radar initial data files from various angles of view by means of bitmaps, this dissertation study the following aspects. The productions will be used to assist meteorologic analysis.
1. Build and display the reflectivity images and velocity images of Doppler radar.
One merit of Chinese new generation of weather radar is that the initial data is digital, so the initial data can be processed by appropriate algorithms to create a series of productions. According to the framework of radar data file, this system read reflectivity data and velocity data information from it. Then designed algorithms to build a reflectivity image or a velocity image. The system built eighteen radia bitmaps in less than three seconds, and the image quality is satisfying.
2. Realize three-dimensional section plane of discretionary transversal using radar data in multiple elevations
In order to visualize the three-dimensional conformation of cloud and echo distributing status, this system synthesized every elevation’s data information to realize cloud three-dimensional section plane. Compared with conventional weather section, this section is in a three-dimensional reference frame, so it has spacial position information. Also, in this system, the transversal’s angle and longness can be discretionary.
3. The cloud conformation’s visualization technique based on three-dimensional section plane
When the transversal moves along a radius, both spacial position and content of the section change. This evolvement course is represented through a sequence of section planes. Though the transversal can be discretionary choose an appropriate transversal according to aerography knowledge and experience can get a better observational effect. Further more the system selects an appropriate observing angle according to optic factors to exhibit the cloud conformation as clear as possible.
4. Analyze and pick up cloud three-dimensional conformation characters
There’re a lot of characters in cloud three-dimensional conformation to be studied, induced and picked up automatically. Define appropriate characters and pick up them in time can help to describe cloud three-dimensional conformation quantificationally to analyze and understand the weather status and to improve the veracity of severe convective weather forecast. Primary attempt has been made in this dissertation, some characters about cloud three-dimensional conformation have been defined and picked up. They are energy, thickness, height and their changes.
All the software developments are realized in the Visual C++6.0 platform. In this dissertation, report algorithms’effect by means of bitmaps, and the processing results are satisfying.
为了以图像的形式从各个视角直观展现雷达基数据文件所蕴含的信息,以辅助气象预报和研究工作,本文主要做了以下几方面的研究和探索。
1.构建并显示雷达反射率图像和径向速度图像
我国新一代天气雷达的优点之一是基数据为数字化的,经过算法的处理可以形成一系列的产品。本文依据雷达基数据产品文件结构,从中读取数据信息并生成反射率图像和径向速度图像。经测试,一组18张图像的总体生成时间小于3秒,图像质量及真实率达到已有产品的水平。
2.实现基于多锥面数据的任意三维剖面
为了使云体的立体结构和回波强度分布可视化,本文汇集各仰角即多锥面的数据信息研究云体的三维剖面图像的生成方法。与已有的剖面图不同的是,本文将剖面置于三维坐标系中,使其携带有丰富的空间位置信息;同时因剖面截线的长度和方位不受任何限制,又使其具有了很强的灵活性。
3.基于三维剖面的云体结构可视化方法
在实现了一个位置上的静止剖面图之后,令剖面的截线沿径向平行移动,随之形成一个剖面系列组,再根据气象工作者的专业知识和经验,确定最佳的截线角度;根据视觉特点,确定最佳的观测角度,从而实现了对所剖云体在最佳剖面角度和最佳观测视角下的云体构成的展现。
4.分析并提取云体三维构成特征
云体在三维构成上有很多特征有待研究、归纳和自动提取,及时恰当地将它们提取出来,将有助于定量地描述云体的三维结构,有助于对天气状况的分析和理解,有助于提高预警预报强对流天气的准确率。本文对此项工作做了初步尝试,定义并提取了反映云体三维构成的能量、厚度、高度及其变化特征。
本文工作均在Visual C++ 6.0软件开发平台下实现,实验测试结果达到预期目标。
The pulse-Doppler radar is widely used in weather investigation, is the main instrument that can detect precipitation system and monitor Severe Convective Weather (including hailstorm, gale, cyclone, violent flood and so on).
In order to exhibit information contained in radar initial data files from various angles of view by means of bitmaps, this dissertation study the following aspects. The productions will be used to assist meteorologic analysis.
1. Build and display the reflectivity images and velocity images of Doppler radar.
One merit of Chinese new generation of weather radar is that the initial data is digital, so the initial data can be processed by appropriate algorithms to create a series of productions. According to the framework of radar data file, this system read reflectivity data and velocity data information from it. Then designed algorithms to build a reflectivity image or a velocity image. The system built eighteen radia bitmaps in less than three seconds, and the image quality is satisfying.
2. Realize three-dimensional section plane of discretionary transversal using radar data in multiple elevations
In order to visualize the three-dimensional conformation of cloud and echo distributing status, this system synthesized every elevation’s data information to realize cloud three-dimensional section plane. Compared with conventional weather section, this section is in a three-dimensional reference frame, so it has spacial position information. Also, in this system, the transversal’s angle and longness can be discretionary.
3. The cloud conformation’s visualization technique based on three-dimensional section plane
When the transversal moves along a radius, both spacial position and content of the section change. This evolvement course is represented through a sequence of section planes. Though the transversal can be discretionary choose an appropriate transversal according to aerography knowledge and experience can get a better observational effect. Further more the system selects an appropriate observing angle according to optic factors to exhibit the cloud conformation as clear as possible.
4. Analyze and pick up cloud three-dimensional conformation characters
There’re a lot of characters in cloud three-dimensional conformation to be studied, induced and picked up automatically. Define appropriate characters and pick up them in time can help to describe cloud three-dimensional conformation quantificationally to analyze and understand the weather status and to improve the veracity of severe convective weather forecast. Primary attempt has been made in this dissertation, some characters about cloud three-dimensional conformation have been defined and picked up. They are energy, thickness, height and their changes.
All the software developments are realized in the Visual C++6.0 platform. In this dissertation, report algorithms’effect by means of bitmaps, and the processing results are satisfying.
引文
[1]王映辉,王英杰,王彦君,运动器官组织三维可视化量化分析软件平台的研制,计算机工程与应用,2004,40卷,5期,7-13
[2]高山,三维遥感可视化技术在高盖山隧道工程地质调查中的应用研究,《铁道勘察》,2007年33卷6期,50-54
[3]刘延洲,赵洁,医学高级处理3D可视化技术的临床应用,《延安大学学报:医学科学版》,2007年5卷4期,F0003-F0003
[4]Sharma.A,Bullock.D,Peeta.S,Krogmeier.J,Detection of inclement weather conditions at a signalized intersection using a video image processing algorithm,Source:2006 IEEE 12th Digital Signal Processing Workshop & 4th IEEE Signal Processing Education Workshop (IEEE Cat. No.06EX1488),2006
[5]Siddiqui.Khalid J , Knowledge based weather image processing and classification,International Geoscience and Remote Sensing Symposium (IGARSS), v 4, 1997, p 1944-1946
[6]Simone.G,Morabito.F.C,Farina.A,Radar image fusion by multiscale Kalman filtering,Proceedings of the Third International Conference on Information Fusion (Cat. No.00EX438), 2000, pt. 2, p WED3/10-17 vol.2
[7]Shinozawa.Kazuhiko,Fujii.Masaharu,Sonehara.Noboru,Weather radar image prediction method in local parallel computation,IEEE International Conference on Neural Networks - Conference Proceedings, v 7, 1994, p 4210-4215
[8]辽宁省气象局,辽宁省气象志,辽宁省气象出版社,2002
[9]高安宁,梁科,2002年4月5日罕见的强对流天气成因及预报分析,《广西气象》,2004年25卷1期,4-7
[10]魏锋,杨金虎,高鹏等,2003年7月22日漳县大暴雨强对流天气过程分析,《干旱气象》,2004年22卷2期,28-31
[11]钱苹,李继萍,孙晓冬,昆明机场一次强对流天气过程分析,《四川气象》,2004年24卷1期,26-28
[12]王锡稳,陶健红,刘治国等,“5.26”甘肃局地强对流天气过程综合分析,《高原气象》,2004年23卷6期,815-820
[13]李海燕,巴哈古丽,储鸿等,荤尔勒地区两次强对流天气过程的形势和雷达回波对比分析,《沙漠与绿洲气象》,2007年1卷3期,33-36
[14]毕旭,张小军,宋兰萍等,黄土高原一次强对流天气的多普勒雷达回波特征分析,《山西气象》,2007年5期,29-33
[15]周后福,郑媛媛,邱明燕,基于数值模式和多普勒雷达的强对流天气预报技术,《气象科技》,2007年35卷5期,637-641,I0001
[16]曹玲,窦永祥,杨晓玲等,新一代天气雷达在河西走廊强对流天气预报中的应用,《干旱气象》,2007年25卷4期,30-35
[17]俞小鼎,姚秀萍,熊廷南等,多普勒天气雷达原理与业务应用,气象出版社,2006
[18]何爱军,超声心脏图象的多维多参数功能重建:[博士学位论文],四川:四川大学,2002
[19]朱炜,黄毓瑜,蒋卫安,石油测井三维数据场断层可视化算法研究,《计算机与数字工程》,2005年33卷9期,45-48
[20]祁民,张宝林,梁光河,高密度电法的三维数据场可视化,《地球物理学进展》,2006年21卷3期,981-986
[21]尹爱军,秦树人,椭球体三维数据场可视化模型及实现,《计量学报》,2006年27卷4期,304-308
[22]李冠华,宋卫卫,欧宗瑛,三维数据场体绘制中的预积分分类技术,《计算机工程与应用》,2006年42卷22期,23-26
[23]梁成根,黄云仙,基于粒子系统的天气雷达资料三维显示,电脑知识与技术(学术交流),2007,2卷,8期,518-527
[24]郑芬,孙瑞,杨松福,新一代多普勒天气雷达产品在强对流天气中的应用,《广西气象》,2006年27卷A03期,73-74,82
[25]刘君,肖秀珠,张红梅,闽西一次冰雹过程的雷达产品分析,《气象科技》,2007年35卷B12期,86-88,93
[26]廖玉芳,潘志祥,郭庆,基于单多普勒天气雷达产品的强对流天气预报预警方法,《气象科学》,2006年26卷5期,564-571
[27]赵重安,刘开宇,王世权,数值预报产品和多普勒雷达图在强对流天气中的应用,《民航科技》,2006年6期,35-37
[28]Lohou.F,Campistron.B,Druilhet.A,Doppler radar: an efficient tool for the three-dimensional description of organized cloudy boundary layers,Atmospheric Research, v 50, n 1, Jan. 1999, p 69-74
[29]付晓辉,周建志,熊传辉,新一代天气雷达应用分析资料库系统的研制,《湖北气象》,2004年1期,29-31
[30]张晰莹,郑凯,多普勒天气雷达与711天气雷达应用的对比分析,《黑龙江气象》,2003年1期,33-34
[31]张培昌,杜秉玉,戴铁丕,雷达气象学,北京:气象出版社,2001.1
[32]中国人民解放军总参谋部气象局,多普勒天气雷达资料分析与应用,北京:解放军出版社,2000.1
[33]Rafael C. Gonzalez, Richard E. Woods,数字图像处理(阮秋琦,阮宇智等译),北京:电子工业出版社,2003
[34]王润生,图像理解,长沙:国防科技大学出版社,1995
[35]杨淑莹,VC++图像处理程序设计,北京:清华大学出版社,2006
[36]吕凤军,数字图像处理编程入门,北京:清华大学出版社,1999
[37]Zhe Xu;Thi Thi Zin;Hama.H,Estimation of weather conditions using image colors,Memoirs of the Faculty of Engineering, Osaka City University, v 47, Dec. 2006, p 9-15
[38]Maki.Masayuki,Ohkura.Hiroshi,Miyachi Hideo,Analysis of Doppler radar data by three-dimensional computer graphics,International Conference on Radar Meteorology, 1993, p 350a-350c
[39]王中科,杨玲,敬腊梅,基于气象雷达回波3D重建的Marching Cubes改进算法,计算机工程与设计,2007年9月,第28卷,第17期,4118-4324
[40]Pu Zhang , Hong Shu , Yan Liu , Yang Li , Some techniques for three-dimensional doppler weather radar data processing,International Geoscience and Remote Sensing Symposium (IGARSS), 2006 IEEE International Geoscience and Remote Sensing Symposium, IGARSS, 2006, p 2884-2887
[41]丁金铸,谈国新,基于视频序列中的运动目标检测技术,湖南工程学院学报(自然科学版),2007年17卷03期
[42]杨俊红,马国峰,周兵等,序列图像中运动区域的检测与几何特征提取,河南科学,2007年25卷04期
[43]David J.Kruglinski,Scot Wingo,George Shepherd,Visual C++6.0技术内幕(希望图书创作室译),北京希望电子出版社,2002
[44]孙鑫,VC++深入详解,北京:电子工业出版社,2006
[45]侯俊杰,深入浅出MFC,武汉:华中科技大学出版社,2005
[46]林锐,韩永泉,高质量程序设计指南——C++/C语言,北京:电子工业出版社,2007
[2]高山,三维遥感可视化技术在高盖山隧道工程地质调查中的应用研究,《铁道勘察》,2007年33卷6期,50-54
[3]刘延洲,赵洁,医学高级处理3D可视化技术的临床应用,《延安大学学报:医学科学版》,2007年5卷4期,F0003-F0003
[4]Sharma.A,Bullock.D,Peeta.S,Krogmeier.J,Detection of inclement weather conditions at a signalized intersection using a video image processing algorithm,Source:2006 IEEE 12th Digital Signal Processing Workshop & 4th IEEE Signal Processing Education Workshop (IEEE Cat. No.06EX1488),2006
[5]Siddiqui.Khalid J , Knowledge based weather image processing and classification,International Geoscience and Remote Sensing Symposium (IGARSS), v 4, 1997, p 1944-1946
[6]Simone.G,Morabito.F.C,Farina.A,Radar image fusion by multiscale Kalman filtering,Proceedings of the Third International Conference on Information Fusion (Cat. No.00EX438), 2000, pt. 2, p WED3/10-17 vol.2
[7]Shinozawa.Kazuhiko,Fujii.Masaharu,Sonehara.Noboru,Weather radar image prediction method in local parallel computation,IEEE International Conference on Neural Networks - Conference Proceedings, v 7, 1994, p 4210-4215
[8]辽宁省气象局,辽宁省气象志,辽宁省气象出版社,2002
[9]高安宁,梁科,2002年4月5日罕见的强对流天气成因及预报分析,《广西气象》,2004年25卷1期,4-7
[10]魏锋,杨金虎,高鹏等,2003年7月22日漳县大暴雨强对流天气过程分析,《干旱气象》,2004年22卷2期,28-31
[11]钱苹,李继萍,孙晓冬,昆明机场一次强对流天气过程分析,《四川气象》,2004年24卷1期,26-28
[12]王锡稳,陶健红,刘治国等,“5.26”甘肃局地强对流天气过程综合分析,《高原气象》,2004年23卷6期,815-820
[13]李海燕,巴哈古丽,储鸿等,荤尔勒地区两次强对流天气过程的形势和雷达回波对比分析,《沙漠与绿洲气象》,2007年1卷3期,33-36
[14]毕旭,张小军,宋兰萍等,黄土高原一次强对流天气的多普勒雷达回波特征分析,《山西气象》,2007年5期,29-33
[15]周后福,郑媛媛,邱明燕,基于数值模式和多普勒雷达的强对流天气预报技术,《气象科技》,2007年35卷5期,637-641,I0001
[16]曹玲,窦永祥,杨晓玲等,新一代天气雷达在河西走廊强对流天气预报中的应用,《干旱气象》,2007年25卷4期,30-35
[17]俞小鼎,姚秀萍,熊廷南等,多普勒天气雷达原理与业务应用,气象出版社,2006
[18]何爱军,超声心脏图象的多维多参数功能重建:[博士学位论文],四川:四川大学,2002
[19]朱炜,黄毓瑜,蒋卫安,石油测井三维数据场断层可视化算法研究,《计算机与数字工程》,2005年33卷9期,45-48
[20]祁民,张宝林,梁光河,高密度电法的三维数据场可视化,《地球物理学进展》,2006年21卷3期,981-986
[21]尹爱军,秦树人,椭球体三维数据场可视化模型及实现,《计量学报》,2006年27卷4期,304-308
[22]李冠华,宋卫卫,欧宗瑛,三维数据场体绘制中的预积分分类技术,《计算机工程与应用》,2006年42卷22期,23-26
[23]梁成根,黄云仙,基于粒子系统的天气雷达资料三维显示,电脑知识与技术(学术交流),2007,2卷,8期,518-527
[24]郑芬,孙瑞,杨松福,新一代多普勒天气雷达产品在强对流天气中的应用,《广西气象》,2006年27卷A03期,73-74,82
[25]刘君,肖秀珠,张红梅,闽西一次冰雹过程的雷达产品分析,《气象科技》,2007年35卷B12期,86-88,93
[26]廖玉芳,潘志祥,郭庆,基于单多普勒天气雷达产品的强对流天气预报预警方法,《气象科学》,2006年26卷5期,564-571
[27]赵重安,刘开宇,王世权,数值预报产品和多普勒雷达图在强对流天气中的应用,《民航科技》,2006年6期,35-37
[28]Lohou.F,Campistron.B,Druilhet.A,Doppler radar: an efficient tool for the three-dimensional description of organized cloudy boundary layers,Atmospheric Research, v 50, n 1, Jan. 1999, p 69-74
[29]付晓辉,周建志,熊传辉,新一代天气雷达应用分析资料库系统的研制,《湖北气象》,2004年1期,29-31
[30]张晰莹,郑凯,多普勒天气雷达与711天气雷达应用的对比分析,《黑龙江气象》,2003年1期,33-34
[31]张培昌,杜秉玉,戴铁丕,雷达气象学,北京:气象出版社,2001.1
[32]中国人民解放军总参谋部气象局,多普勒天气雷达资料分析与应用,北京:解放军出版社,2000.1
[33]Rafael C. Gonzalez, Richard E. Woods,数字图像处理(阮秋琦,阮宇智等译),北京:电子工业出版社,2003
[34]王润生,图像理解,长沙:国防科技大学出版社,1995
[35]杨淑莹,VC++图像处理程序设计,北京:清华大学出版社,2006
[36]吕凤军,数字图像处理编程入门,北京:清华大学出版社,1999
[37]Zhe Xu;Thi Thi Zin;Hama.H,Estimation of weather conditions using image colors,Memoirs of the Faculty of Engineering, Osaka City University, v 47, Dec. 2006, p 9-15
[38]Maki.Masayuki,Ohkura.Hiroshi,Miyachi Hideo,Analysis of Doppler radar data by three-dimensional computer graphics,International Conference on Radar Meteorology, 1993, p 350a-350c
[39]王中科,杨玲,敬腊梅,基于气象雷达回波3D重建的Marching Cubes改进算法,计算机工程与设计,2007年9月,第28卷,第17期,4118-4324
[40]Pu Zhang , Hong Shu , Yan Liu , Yang Li , Some techniques for three-dimensional doppler weather radar data processing,International Geoscience and Remote Sensing Symposium (IGARSS), 2006 IEEE International Geoscience and Remote Sensing Symposium, IGARSS, 2006, p 2884-2887
[41]丁金铸,谈国新,基于视频序列中的运动目标检测技术,湖南工程学院学报(自然科学版),2007年17卷03期
[42]杨俊红,马国峰,周兵等,序列图像中运动区域的检测与几何特征提取,河南科学,2007年25卷04期
[43]David J.Kruglinski,Scot Wingo,George Shepherd,Visual C++6.0技术内幕(希望图书创作室译),北京希望电子出版社,2002
[44]孙鑫,VC++深入详解,北京:电子工业出版社,2006
[45]侯俊杰,深入浅出MFC,武汉:华中科技大学出版社,2005
[46]林锐,韩永泉,高质量程序设计指南——C++/C语言,北京:电子工业出版社,2007