机载光电成像系统的图像拼接技术研究
|
摘要
基于多尺度配准的图像拼接技术选取渐进式配准方法,利用空域图像拼接法拼接机载图像,未对图像进行几何校正,导致机载图像几何位置出现偏差,图像拼接效果差。研究新的机载光电成像系统的图像拼接技术,通过确定散乱扫描点同非散乱扫描点间的对应关系校正机载图像;采用SURF算法构建校正后机载图像的尺度空间,确定图像特征点主方向,得到图像扩展点SURF描述符,依据该描述符对比参考图像与待匹配图像关键点相似性,采用RANSAC算法和最小二乘法清除匹配错误的特征点对,获取配准图像间的空间变换关系,实现机载图像的特征匹配和图像配准;依据配准图像空间变换关系,将不同机载图像映射至相同参考坐标系后,采用渐入渐出法融合不同机载图像的重合部分,完成机载图像无缝拼接。实验结果表明,所提技术图像拼接性能优于对比技术,图像拼接效果佳。
The image splicing technology based on multi-scale registration selects the progressive registration method,and the airborne image splicing method is used to splice the airborne image. But the geometric correction of the image is not carried out,resulting in the deviation of the geometric position of the airborne image and the image splicing effect is poor. The new image splicing technology of airborne photoelectric imaging system is studied,and the correlation between scattered scanning point and non-scattered scanning point is determined. The SURF algorithm is used to construct the scale space of the corrected airborne image,determine the main direction of the image feature point,and obtain the image extension point SURF descriptor. According to the descriptor,the similarities between the reference image and the key point of the image to be matched are compared. By means of RANSAC algorithm and least squares method,the spatial transformation relationship between registration images is obtained,and the feature matching and image registration of airborne images are realized. After mapping different airborne images to the same reference coordinate system according to the spatial transformation relationship of registration images,the overlapping parts of different airborne images are fused by the gradient method to complete the seamless splicing of airborne images. The experimental results show that the splicing performance of the proposed technology is better than that of contrast technology and the splicing effect is better.
The image splicing technology based on multi-scale registration selects the progressive registration method,and the airborne image splicing method is used to splice the airborne image. But the geometric correction of the image is not carried out,resulting in the deviation of the geometric position of the airborne image and the image splicing effect is poor. The new image splicing technology of airborne photoelectric imaging system is studied,and the correlation between scattered scanning point and non-scattered scanning point is determined. The SURF algorithm is used to construct the scale space of the corrected airborne image,determine the main direction of the image feature point,and obtain the image extension point SURF descriptor. According to the descriptor,the similarities between the reference image and the key point of the image to be matched are compared. By means of RANSAC algorithm and least squares method,the spatial transformation relationship between registration images is obtained,and the feature matching and image registration of airborne images are realized. After mapping different airborne images to the same reference coordinate system according to the spatial transformation relationship of registration images,the overlapping parts of different airborne images are fused by the gradient method to complete the seamless splicing of airborne images. The experimental results show that the splicing performance of the proposed technology is better than that of contrast technology and the splicing effect is better.
引文
[1]周前飞,刘晶红,熊文卓,等.机载光电成像平台的多目标自主定位系统研究[J].光学学报,2015,35(1):181-195.
[2]骆媛,张科,纪明.增强无人机态势感知的彩色图像融合方法[J].红外与激光工程,2016,45(B05):200-206.
[3]李铁成,刘晶红,周前飞.基于相机内外参数的机载光电平台图像几何校正[J].压电与声光,2016,38(1):32-37.
[4]陈思颖,王晓鲁,张寅超,等.城区机载激光雷达点云数据与航空影像的多尺度配准方法[J].北京理工大学学报,2016,36(2):186-190.
[5]杨红,康登魁,姜昌录,等.可见光光电成像系统整机综合参数校准技术研究[J].应用光学,2015,V36(2):253-258.
[6]佚名.机载光电载荷视轴指向控制技术综述[J].光学精密工程,2018,26(7):1642-1652.
[7]郑小梅,方俊永,张保辉,等.基于变焦镜头的机载相机监控系统设计及试验[J].光学技术,2015,41(6):548-553.
[8] YAHYANEJAD S,RINNER B.A fast and mobile system for registration of low-altitude visual and thermal aerial images using multiple small-scale UAVs[J]. Isprs Journal of Photogrammetry&Remote Sensing,2015,104(Complete):189-202.
[9]刘帅,牛燕雄,刘海月.全向凝视光电成像系统鱼眼透镜的设计[J].激光技术,2017,41(2):296-301.
[10]廖周,邱琪,张雨东.大口径拼接望远镜成像系统的远场特性[J].光电工程,2015,42(2):1-8.
[11]王建威,裴琳琳,谭政,等.机械快门滤光片式多光谱成像仪相对定标方法研究[J].光谱学与光谱分析,2017,37(8):2615-2618.
[12] YANG F,SU D,KAI Z,et al.Mosaicing of airborne Li DAR bathymetry strips based on Monte Carlo matching[J]. Marine Geophysical Research,2017,38(3):303-311.
[13]沈君辉,杨光,陶忠,等.舰载直升机光电系统发展及其关键技术[J].应用光学,2015,36(2):161-170.
[14]王雨曦,亓洪兴,马彦鹏,等.画幅扫描红外成像实时拼接中的光束法平差[J].红外与激光工程,2016,45(1):104002-0104002.
[15]刘洪兴,李宪圣,李晓杰,等.反射式拼接CCD相机渐晕图像校正[J].光电工程,2015,42(2):9-14.
[16] MANIKANDAN S,VARDHINI J P. Enhanced Feature Based Mosaicing Technique for Visually and Geometrically Degraded Airborne Synthetic Aperture Radar Images[J].Sensing&Imaging,2015,16(1):1-19.
[2]骆媛,张科,纪明.增强无人机态势感知的彩色图像融合方法[J].红外与激光工程,2016,45(B05):200-206.
[3]李铁成,刘晶红,周前飞.基于相机内外参数的机载光电平台图像几何校正[J].压电与声光,2016,38(1):32-37.
[4]陈思颖,王晓鲁,张寅超,等.城区机载激光雷达点云数据与航空影像的多尺度配准方法[J].北京理工大学学报,2016,36(2):186-190.
[5]杨红,康登魁,姜昌录,等.可见光光电成像系统整机综合参数校准技术研究[J].应用光学,2015,V36(2):253-258.
[6]佚名.机载光电载荷视轴指向控制技术综述[J].光学精密工程,2018,26(7):1642-1652.
[7]郑小梅,方俊永,张保辉,等.基于变焦镜头的机载相机监控系统设计及试验[J].光学技术,2015,41(6):548-553.
[8] YAHYANEJAD S,RINNER B.A fast and mobile system for registration of low-altitude visual and thermal aerial images using multiple small-scale UAVs[J]. Isprs Journal of Photogrammetry&Remote Sensing,2015,104(Complete):189-202.
[9]刘帅,牛燕雄,刘海月.全向凝视光电成像系统鱼眼透镜的设计[J].激光技术,2017,41(2):296-301.
[10]廖周,邱琪,张雨东.大口径拼接望远镜成像系统的远场特性[J].光电工程,2015,42(2):1-8.
[11]王建威,裴琳琳,谭政,等.机械快门滤光片式多光谱成像仪相对定标方法研究[J].光谱学与光谱分析,2017,37(8):2615-2618.
[12] YANG F,SU D,KAI Z,et al.Mosaicing of airborne Li DAR bathymetry strips based on Monte Carlo matching[J]. Marine Geophysical Research,2017,38(3):303-311.
[13]沈君辉,杨光,陶忠,等.舰载直升机光电系统发展及其关键技术[J].应用光学,2015,36(2):161-170.
[14]王雨曦,亓洪兴,马彦鹏,等.画幅扫描红外成像实时拼接中的光束法平差[J].红外与激光工程,2016,45(1):104002-0104002.
[15]刘洪兴,李宪圣,李晓杰,等.反射式拼接CCD相机渐晕图像校正[J].光电工程,2015,42(2):9-14.
[16] MANIKANDAN S,VARDHINI J P. Enhanced Feature Based Mosaicing Technique for Visually and Geometrically Degraded Airborne Synthetic Aperture Radar Images[J].Sensing&Imaging,2015,16(1):1-19.