基于X波段导航雷达海面回波图像预处理的研究
|
摘要
X波段导航雷达海面回波数据的采集过程中存在着很多实际问题,包括数据中存在噪声,沿某一扫描方向的数据丢失等。这些问题对后期利用数据来反演各种海态参数存在很大影响,直接影响了参数估计的精度。因此,本文针对以上问题进行了对原始数据的预处理研究。
首先,因为X波段雷达回波图像与传统的图像在空间分布上存在很大的不同,所以,根据海面X波段雷达回波图像数据的时空特点,本文提出了一种新的去噪方法。该方法利用线检测对雷达图像中的射线状噪声进行分离,然后针对这种特殊的噪声形式提出一种阈值分割的方法,最后在极坐标下针对噪声点进行插值修复。实验表明,与传统的均值滤波和中值滤波等方法相比,新方法无论在主观视觉还是信噪比方面,都有了一定改进,图像保留了更多的细节信息,具有较好的去噪效果。
然后,本文研究了一种基于多幅图像的数据修复算法。在所掌握的数据比较充分的时候,为了保证所修复数据的精度,在图像修复过程中可以考虑引入时间相关性的影响。本文研究了传统的运动估计方法,采用在小波域运动估计的方法查找在时间序列上的匹配块,并在修复的过程中引入这些查找到的匹配块的影响。实验表明,该方法对原始数据有了很好的逼近,保证了修复后数据的精度。
最后,本文对原始的图像数据进行了几何变换,还原了真实的扫描空间分布,对非整数网格点使用反距离加权法插值,最终完成了图像的可视化。
There are many problems in the sea surface images data collected by the X-band nautical radar previously, such as the noise and the data lost of rows along certain directions of scanning. These problems directly affect the inversion of many sea-state parameters and reduce the precision of estimation in the next step. Thus, this paper has put forward research on original data pretreatment.
First, for the difference between radar image and traditional image in space distribution structure, a method of denoising is presented according to the space-time characteristics of the X-band radar sea surface images. The noise in radial pattern can be separated by line detection. Considering the special pattern of the noise, a method of threshold segmentation has been put forward. Furthermore, interpolation in polar coordinates has been adopted on these noise points. The results show that, comparing with traditional methods mean filtering and median filtering, this method achieves improvement on both subjective vision and SNR. It also keeps better detail information and has a good denoising effect.
Next, this paper provides a method of image inpainting based on multiple images sequence. In order to guarantee the precision, the influence of correlation in time-domain should be introduced in the process of inpainting when the data available is sufficient. This paper studied on traditional method of motion estimation and found the matching block in time series using the motion estimation in wavelet domain. After that, the influence of matching block would be introduced in the process of inpainting. The results show that, the data after inpainting has a good approximation to the no-damaged data and ensure the data precision.
Last, this paper has made geometric transformations on original image and restored the true scanning space distribution. IDW (Inverse Distance Weighted) has been used to non-integer index and the image visualization would be finally realized.
首先,因为X波段雷达回波图像与传统的图像在空间分布上存在很大的不同,所以,根据海面X波段雷达回波图像数据的时空特点,本文提出了一种新的去噪方法。该方法利用线检测对雷达图像中的射线状噪声进行分离,然后针对这种特殊的噪声形式提出一种阈值分割的方法,最后在极坐标下针对噪声点进行插值修复。实验表明,与传统的均值滤波和中值滤波等方法相比,新方法无论在主观视觉还是信噪比方面,都有了一定改进,图像保留了更多的细节信息,具有较好的去噪效果。
然后,本文研究了一种基于多幅图像的数据修复算法。在所掌握的数据比较充分的时候,为了保证所修复数据的精度,在图像修复过程中可以考虑引入时间相关性的影响。本文研究了传统的运动估计方法,采用在小波域运动估计的方法查找在时间序列上的匹配块,并在修复的过程中引入这些查找到的匹配块的影响。实验表明,该方法对原始数据有了很好的逼近,保证了修复后数据的精度。
最后,本文对原始的图像数据进行了几何变换,还原了真实的扫描空间分布,对非整数网格点使用反距离加权法插值,最终完成了图像的可视化。
There are many problems in the sea surface images data collected by the X-band nautical radar previously, such as the noise and the data lost of rows along certain directions of scanning. These problems directly affect the inversion of many sea-state parameters and reduce the precision of estimation in the next step. Thus, this paper has put forward research on original data pretreatment.
First, for the difference between radar image and traditional image in space distribution structure, a method of denoising is presented according to the space-time characteristics of the X-band radar sea surface images. The noise in radial pattern can be separated by line detection. Considering the special pattern of the noise, a method of threshold segmentation has been put forward. Furthermore, interpolation in polar coordinates has been adopted on these noise points. The results show that, comparing with traditional methods mean filtering and median filtering, this method achieves improvement on both subjective vision and SNR. It also keeps better detail information and has a good denoising effect.
Next, this paper provides a method of image inpainting based on multiple images sequence. In order to guarantee the precision, the influence of correlation in time-domain should be introduced in the process of inpainting when the data available is sufficient. This paper studied on traditional method of motion estimation and found the matching block in time series using the motion estimation in wavelet domain. After that, the influence of matching block would be introduced in the process of inpainting. The results show that, the data after inpainting has a good approximation to the no-damaged data and ensure the data precision.
Last, this paper has made geometric transformations on original image and restored the true scanning space distribution. IDW (Inverse Distance Weighted) has been used to non-integer index and the image visualization would be finally realized.
引文
[1]周蓓,X波段雷达海面流场信息提取技术研究:[硕士学位论文],青岛;中国海洋大学,2008.
[2]Oudshoorn H M, The use of radar in hydrodynamic survey, Coastal Engineering, 1960, 7: 59~76.
[3]Willis T G, Beaumont H, Wave direction measurements using sea surveillance radars, Tech. Memo. TR 118. R. Aircraft Establishment, Farnbough, Great Britain, 1971.
[4]Evmenov V W. Test of the radar method of defining ocean waves elements, J. Fluid Mech. Sov. Res, 1973, 2(5): 141~145.
[5]Mattie M G, Harris D L, A system for using radar to record wave direction, Tech. Rep. 79-1, 50, U.S. Army Corps of Eng, Coast. Eng. Res. Ctr, Fort Belvoir, Va, 1979.
[6]Hoogeboom P, Rosenthal W, Directional wave spectra in radar images, International Geoscience and Remote Sensing Symposium, IEEE Geosci, and Remote Sensing Soc, Munich, June, 1982.
[7]Ziemer F, W. Rosenthal, Carlson H, Measurements of directional wave spectra by ship radar, IAPSO Symposium, General Assembly Int. Assoc. for Phys. Sci. of the Oceans, Hamburg, Federal Republic of Germany, 1983.
[8]Ziemer F, An instrument for the survey of the directionality of the ocean wave field, In Proc. WMO/IOC Workshop on Operational Ocean Monitoring Using Surface Based Radars, Geneva, Switzerland, 1995, 32: 81~87.
[9]Dittmer J, Use of marine radars for real time wave field survey and Speeding up Transmission/Processing, In Proceedings of the WMO/IOC Workshop on operational ocean monitoring using surface based radars, Geneva, 1995, 133~138.
[10]Nieto J C, Reichert K, Dittmer J, Use of nautical radar as a wave monitoring instrument, Coastal Engineering, 1999, 37: 331~342.
[11]Reichert K. X-Band radar as a tool to determine spectral and single wave properties[EB/OL]. http://www.oceanwaves.de, 2004, 9.
[12]Crombie D D, Doppler spectrum of sea echo at 13.56 Mc/s, Nature, 1955, 175: 681~682.
[13]Wright J W, Backscattering from capillary waves with application to sea clutter, IEEE Trans. Antennas Propagat, 1966, AP-14: 749~754.
[14]袁小平,孙中泉,陈凯,等,BMP图像格式解析及其空间域处理的实现,电子与计算机技术,2008,1,85~86.
[15]许冠军,数字图像去噪算法研究:[硕士学位论文],杭州,浙江大学,2006.
[16]郝燕玲,唐艳红,卢志忠,X波段航海雷达图像噪声检测与滤除方法研究,国土资源遥感,2008,76(2):14~17.
[17]Borge J N, Ridriguez G, Hessner K, Inversion of marine radar images for surface wave analysis, journal of atmospheric and oceanic technology, 2004, 21(8): 1291~1300.
[18]Gangeskar R, Ocean current estimated from X-band radar sea surface images [J], IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(4): 783~792.
[19]张文祥,李进华,雷达同频干扰现象分析研究,火控雷达技术,2007,36(2):50~53.
[20]Frost V S, Stiles J A, A model for radar images and its application to adaptive digital filtering of multiplicative noise, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1982, 4(2): 157~166.
[21]夏良正,数字图像处理,南京:东南大学出版社,1999.
[22]李涛,小波理论及其在图像处理中的应用:[硕士学位论文],西安:西安电子科技大学,2001
[23] Seemann J, Senet C M, Wolff U, Nautical X-band radar image processing: monitoring of morphodynamic processes in coastal waters, OCEANS 2000 MTS/IEEE Conference and Exhibition, 2000, 2: 1329~1335.
[24]于晓晶,中值滤波器在图像去噪中的应用,电脑知识与技术,2008,3(8):1795~1796.
[25]马学磊,基于噪声点检测的中值滤波图像去噪算法:[硕士学位论文],西安:西安电子科技大学,2008.
[26]汤国安,张友顺,遥感数字图像处理,北京:科学出版社,2004.
[27]张俊,张志勇,陈学广,等,SAR图像斑点噪声抑制与滤除技术,遥感技术与应用,1998,13(3):44~47.
[28]Gonzalez, R C, Woods R E,著,阮秋琦,等译,数字图像处理,北京:电子工业出版社,2007.
[29]Gonzalez, R C, Woods R E,著,阮秋琦,等译,数字图像处理(Matlab版),北京:电子工业出版社,2007.
[30]徐雨明,文双春,数字图像插值算法的分析与实现,福建电脑,2007,1:91~94.
[31]李新,空间内插方法比较,地球科学进展,2000,15(3):260~265.
[32]刘洁瑜,张德运,基于运动预测的Motion JPEG2000误码掩饰新算法,信号处理,2008,24(3):516~520
[33]邬文隽,图像和视频修复技术研究:[硕士学位论文],上海:上海交通大学,2007.
[34]侯正信,何宇清,许微,一种快速的图像修复算法,中国图象图形学报,2007,12(10),1909~1912.
[35]李翔,基于分解的图像和视频修复算法研究:[硕士学位论文],长沙:中南大学,2008.
[36]谌安军,数字视频修复若干关键技术研究:[博士学位论文],北京,中国科学院电子学研究所,2006.
[37]陈春涛,杨新,孙锟,三维B超医学图像插值重构研究,信号处理,2004,20(6),632~634.
[38]徐超,吴大千,张治国,山东省多年气象要素空间插值方法比较研究,山东大学学报,2008,43(3),1~5.
[39]张磊,涂国防,一种新型小波域运动估计算法,电子与信息学报,2007,29(5),1131~1135.
[40]宋传鸣,王相海,一种新的小波域视频可分级运动估计方案,计算机学报,2006,29(12),2112~2118.
[41]魏玉芬,周凯,柳培林,序列图像的小波域运动估计和补偿技术研究,黑龙江八一农垦大学学报,2008,20(1),78~81.
[42]曾接贤,李斯新,基于序列图像的三维刚体运动估计研究进展,南昌航空工业学院学报,2006,20(3),23~28.
[43]龚涛,丁润涛,基于小波变换的块匹配运动估计方法,天津大学学报,2003,36(3),320~324.
[44]何宇翔,大气波导模拟分析及雷达资料三维插值处理:[硕士学位论文],南京:南京气象学院,2003.
[2]Oudshoorn H M, The use of radar in hydrodynamic survey, Coastal Engineering, 1960, 7: 59~76.
[3]Willis T G, Beaumont H, Wave direction measurements using sea surveillance radars, Tech. Memo. TR 118. R. Aircraft Establishment, Farnbough, Great Britain, 1971.
[4]Evmenov V W. Test of the radar method of defining ocean waves elements, J. Fluid Mech. Sov. Res, 1973, 2(5): 141~145.
[5]Mattie M G, Harris D L, A system for using radar to record wave direction, Tech. Rep. 79-1, 50, U.S. Army Corps of Eng, Coast. Eng. Res. Ctr, Fort Belvoir, Va, 1979.
[6]Hoogeboom P, Rosenthal W, Directional wave spectra in radar images, International Geoscience and Remote Sensing Symposium, IEEE Geosci, and Remote Sensing Soc, Munich, June, 1982.
[7]Ziemer F, W. Rosenthal, Carlson H, Measurements of directional wave spectra by ship radar, IAPSO Symposium, General Assembly Int. Assoc. for Phys. Sci. of the Oceans, Hamburg, Federal Republic of Germany, 1983.
[8]Ziemer F, An instrument for the survey of the directionality of the ocean wave field, In Proc. WMO/IOC Workshop on Operational Ocean Monitoring Using Surface Based Radars, Geneva, Switzerland, 1995, 32: 81~87.
[9]Dittmer J, Use of marine radars for real time wave field survey and Speeding up Transmission/Processing, In Proceedings of the WMO/IOC Workshop on operational ocean monitoring using surface based radars, Geneva, 1995, 133~138.
[10]Nieto J C, Reichert K, Dittmer J, Use of nautical radar as a wave monitoring instrument, Coastal Engineering, 1999, 37: 331~342.
[11]Reichert K. X-Band radar as a tool to determine spectral and single wave properties[EB/OL]. http://www.oceanwaves.de, 2004, 9.
[12]Crombie D D, Doppler spectrum of sea echo at 13.56 Mc/s, Nature, 1955, 175: 681~682.
[13]Wright J W, Backscattering from capillary waves with application to sea clutter, IEEE Trans. Antennas Propagat, 1966, AP-14: 749~754.
[14]袁小平,孙中泉,陈凯,等,BMP图像格式解析及其空间域处理的实现,电子与计算机技术,2008,1,85~86.
[15]许冠军,数字图像去噪算法研究:[硕士学位论文],杭州,浙江大学,2006.
[16]郝燕玲,唐艳红,卢志忠,X波段航海雷达图像噪声检测与滤除方法研究,国土资源遥感,2008,76(2):14~17.
[17]Borge J N, Ridriguez G, Hessner K, Inversion of marine radar images for surface wave analysis, journal of atmospheric and oceanic technology, 2004, 21(8): 1291~1300.
[18]Gangeskar R, Ocean current estimated from X-band radar sea surface images [J], IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(4): 783~792.
[19]张文祥,李进华,雷达同频干扰现象分析研究,火控雷达技术,2007,36(2):50~53.
[20]Frost V S, Stiles J A, A model for radar images and its application to adaptive digital filtering of multiplicative noise, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1982, 4(2): 157~166.
[21]夏良正,数字图像处理,南京:东南大学出版社,1999.
[22]李涛,小波理论及其在图像处理中的应用:[硕士学位论文],西安:西安电子科技大学,2001
[23] Seemann J, Senet C M, Wolff U, Nautical X-band radar image processing: monitoring of morphodynamic processes in coastal waters, OCEANS 2000 MTS/IEEE Conference and Exhibition, 2000, 2: 1329~1335.
[24]于晓晶,中值滤波器在图像去噪中的应用,电脑知识与技术,2008,3(8):1795~1796.
[25]马学磊,基于噪声点检测的中值滤波图像去噪算法:[硕士学位论文],西安:西安电子科技大学,2008.
[26]汤国安,张友顺,遥感数字图像处理,北京:科学出版社,2004.
[27]张俊,张志勇,陈学广,等,SAR图像斑点噪声抑制与滤除技术,遥感技术与应用,1998,13(3):44~47.
[28]Gonzalez, R C, Woods R E,著,阮秋琦,等译,数字图像处理,北京:电子工业出版社,2007.
[29]Gonzalez, R C, Woods R E,著,阮秋琦,等译,数字图像处理(Matlab版),北京:电子工业出版社,2007.
[30]徐雨明,文双春,数字图像插值算法的分析与实现,福建电脑,2007,1:91~94.
[31]李新,空间内插方法比较,地球科学进展,2000,15(3):260~265.
[32]刘洁瑜,张德运,基于运动预测的Motion JPEG2000误码掩饰新算法,信号处理,2008,24(3):516~520
[33]邬文隽,图像和视频修复技术研究:[硕士学位论文],上海:上海交通大学,2007.
[34]侯正信,何宇清,许微,一种快速的图像修复算法,中国图象图形学报,2007,12(10),1909~1912.
[35]李翔,基于分解的图像和视频修复算法研究:[硕士学位论文],长沙:中南大学,2008.
[36]谌安军,数字视频修复若干关键技术研究:[博士学位论文],北京,中国科学院电子学研究所,2006.
[37]陈春涛,杨新,孙锟,三维B超医学图像插值重构研究,信号处理,2004,20(6),632~634.
[38]徐超,吴大千,张治国,山东省多年气象要素空间插值方法比较研究,山东大学学报,2008,43(3),1~5.
[39]张磊,涂国防,一种新型小波域运动估计算法,电子与信息学报,2007,29(5),1131~1135.
[40]宋传鸣,王相海,一种新的小波域视频可分级运动估计方案,计算机学报,2006,29(12),2112~2118.
[41]魏玉芬,周凯,柳培林,序列图像的小波域运动估计和补偿技术研究,黑龙江八一农垦大学学报,2008,20(1),78~81.
[42]曾接贤,李斯新,基于序列图像的三维刚体运动估计研究进展,南昌航空工业学院学报,2006,20(3),23~28.
[43]龚涛,丁润涛,基于小波变换的块匹配运动估计方法,天津大学学报,2003,36(3),320~324.
[44]何宇翔,大气波导模拟分析及雷达资料三维插值处理:[硕士学位论文],南京:南京气象学院,2003.