基于Hu相关滤波的光学卫星视频点目标跟踪
|
摘要
针对光学卫星视频运动目标跟踪问题进行研究,提出一种鲁棒的特征描述和匹配跟踪方法。引入相关滤波的思想,首先利用样本集的Hu不变矩和中值滤波,建立目标的跟踪模板并进行目标特征描述。然后,将目标跟踪的判断区域降维处理,建立判断区域的Hu置信模型。利用FFT推导了快速相关法,进而通过求得跟踪置信图最大值实现目标跟踪。针对跟踪轨迹采用卡尔曼滤波辅助和优化跟踪处理,提高算法的鲁棒性。试验数据采用SkySat和吉林一号拍摄的视频各两段,对5个点目标进行跟踪试验,跟踪精度优于90%,跟踪过程目标不丢失,且轨迹平滑。针对13×13的判断区域,与一般相关性方法相比,处理速度可提升约5倍。可为光学卫星视频点目标实现快速可靠跟踪提供技术基础。
Aiming at the moving object tracking problem of satellite video, a robust method of feature description and matching based point-target tracking is proposed. The idea of correlation filter is introduced. Firstly, Hu invariant moments and median filtering of the sample set are used to establish the tracking template of the target and describe the target features. Then, the dimension of the decision area of target tracking is reduced and the Hu confidence model of the judgment area is established. Finally, the target tracking is achieved by finding the maximum value of the tracking confidence map by using the fast FFT derivation correlation method. Moreover, the robustness of the algorithm is improved by using Kalman filtering to assist and optimize the tracking process. The experimental video data are taken from SkySat-1 and Jinlin-1 satellites. The tracking experiments of 5 point targets show that the proposed method achieves good results in the satellite video, and that the accuracy is better than 90%. The target is not lost in the whole tracking process and the tracking trajectory is smooth. In addition, for the 13×13 judgment area, the processing speed of the confidence map is about 5 times higher than the conventional correlation method. This method can provide a technical basis for achieving fast and reliable tracking of point targets in satellite video.
Aiming at the moving object tracking problem of satellite video, a robust method of feature description and matching based point-target tracking is proposed. The idea of correlation filter is introduced. Firstly, Hu invariant moments and median filtering of the sample set are used to establish the tracking template of the target and describe the target features. Then, the dimension of the decision area of target tracking is reduced and the Hu confidence model of the judgment area is established. Finally, the target tracking is achieved by finding the maximum value of the tracking confidence map by using the fast FFT derivation correlation method. Moreover, the robustness of the algorithm is improved by using Kalman filtering to assist and optimize the tracking process. The experimental video data are taken from SkySat-1 and Jinlin-1 satellites. The tracking experiments of 5 point targets show that the proposed method achieves good results in the satellite video, and that the accuracy is better than 90%. The target is not lost in the whole tracking process and the tracking trajectory is smooth. In addition, for the 13×13 judgment area, the processing speed of the confidence map is about 5 times higher than the conventional correlation method. This method can provide a technical basis for achieving fast and reliable tracking of point targets in satellite video.
引文
[1] D’ANGELO P,MATTYUS G,REINARTZ P.Skybox image and video product evaluation[J].International Journal of Image & Data Fusion,2015,7(1):1-16.
[2] 张过.卫星视频处理与应用进展[J].应用科学学报,2016,34(4):361-370.ZHANG G.Satellite video processing and applications[J].Journal of Applied Sciences,2016,34(4):361-370(in Chinese).
[3] 王霞,张过,沈欣,等.顾及像面畸变的卫星视频稳像[J].测绘学报,2016,45(2):194-198.WANG X,ZHANG G,SHEN X,et al.Satellite video stabilization with geometric distortion [J].Acta Geodaetica et Cartographica Sinica,2016,45(2):194-198(in Chinese).
[4] 朱厉洪,回征,任德锋,等.视频成像卫星发展现状与启示[J].卫星应用,2015(10):23-28.ZHU L H,HUI Z,REN D F et al,Development status and inspiration of video imaging satellite [J].Satellite Applications,2015(10):20-28(in Chinese).
[5] 林浩申,刘刚,何兵.自适应空间机动目标跟踪算法[J].中国空间科学技术,2017,37(1):26-32.LIN H S,LIU G,HE B.A localization algorithm for adaptive maneuvering spatial targets[J].Chinese Space Science and Technology,2017,37(1):26-32(in Chinese).
[6] 刘钧彬,谢维信,李良群.一种遮挡检测的核最小二乘视觉跟踪算法[J].信号处理,2017,33(4):505-510.LIU J B,XIE W X,LI L Q.Visual tracking based on kernelized least squares with occlusion detecting[J].Journal of Signal Processing,2017,33(4):505-510(in Chinese).
[7] BOLME D S,BEVERIDGE J R,DRAPER B A,et al.Visual object tracking using adaptive correlation filters[C].Computer Vision and Pattern Recognition.IEEE,2010:2544-2550.
[8] DANELLJAN M,HAGER G,KHAN F S,et al.Discriminative scale space tracking[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016 (99):1-14.
[9] HENRIQUES J F,CASEIRO R,MARTINS P,et al.High-speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2015,37(3):583-596.
[10] XU J,LU Y,LIU J.Robust tracking via weighted spatio-temporal context learning[C].IEEE International Conference on Image Processing,2014:413-416.
[11] WU Y,LIM J,YANG M.Object tracking benchmark[J].IEEE Trans.Pattern Anal.Mach.Intell,2015,37(9):1834-1848.
[12] 王春平,王暐,刘江义,等.基于色度饱和度-角度梯度直方图特征的尺度自适应核相关滤波跟踪[J].光学精密工程,2016,24(9):2293-2301.WANG C P,WANG W,LIU J Y,et al.Scale adaptive kernelized correlation filter tracking based on HHS-OG feature[J].Optics and Precision Engineering,2016,24(9):2293-2301(in Chinese).
[13] 金永涛,李旭青,张周威等.基于多特征的高分遥感图像分割算法研究[J].中国空间科学技术,2016,36(6):38-46.JIN Y T,LI X Q,ZHANG Z W,et al.Segmentation of high resolution multi spectral remote sensing image based on multi feature[J].Chinese Space Science and Technology,2016,36(6):38-46(in Chinese).
[14] 孙卜郊,周东华.基于NCC的存在旋转的图像匹配方法[J].传感器与微系统,2008,27(5):43-45.SUN B J,ZHOU D H.Rotated image registration method based on NCC[J].Transducer and Microsystem Technologies,2008,27(5):43-45(in Chinese).
[15] JIANBO S,CARLO T.Good features to track[C].IEEE Conference on Computer Vision and Pattern Recognition,1994:593-600.
[16] CHEN Z W,CHIANG C C,HSIEH Z T.Extending 3D Lucas-Kanade tracking with adaptive templates for head pose estimation[J].Machine Vision and Applications,2010,21(6):889-903.
[17] FORTUN D,BOUTHEMY P,KERVRANN C.Optical flow modeling and computation:a survey[J].Computer Vision & Image Understanding,2015,134:1-21.
[18] HU M.Visual pattern recognition by moment invariants[J].Ire Transactions on Information Theory,1962,8(2):179-187.
[19] 黄敏,马亚琼,朱颢东,等.基于Hu修正不变矩的商标识别[J].华中师范大学学报(自然科学版),2014,48(2):182-185.HUANG M,MA Y Q,ZHU H D.Rrademark recognition based on Hu modified invariant moment[J].Journal of Huazhong Normal University(Natural Sciences),2014,48(2):182-185(in Chinese).
[20] 贺一楠,耿娟,秦军,等.结合Forstner与NCC约束的UAV图像配准研究[J].国土资源遥感,2015,27(1):48-54.HE Y N,GENG J,QIN J.Integration of Forstner and NCC constraint for UAV image registration[J].Remote Sensing For Land & Resources,2015,27(1):48-54(in Chinese).
[21] 宁晓琳,白鑫贝,姜丽伟.一种基于连续星图观测的自旋姿态估计方法[J].中国空间科学技术,2013,33(6):49-57.NING X L,BAI X B,JIANG L W.New method for attitude determination of spin satellite based on continuous star map observation [J].Chinese Space Science and Technology,2013,33(6):49-57(in Chinese).
[2] 张过.卫星视频处理与应用进展[J].应用科学学报,2016,34(4):361-370.ZHANG G.Satellite video processing and applications[J].Journal of Applied Sciences,2016,34(4):361-370(in Chinese).
[3] 王霞,张过,沈欣,等.顾及像面畸变的卫星视频稳像[J].测绘学报,2016,45(2):194-198.WANG X,ZHANG G,SHEN X,et al.Satellite video stabilization with geometric distortion [J].Acta Geodaetica et Cartographica Sinica,2016,45(2):194-198(in Chinese).
[4] 朱厉洪,回征,任德锋,等.视频成像卫星发展现状与启示[J].卫星应用,2015(10):23-28.ZHU L H,HUI Z,REN D F et al,Development status and inspiration of video imaging satellite [J].Satellite Applications,2015(10):20-28(in Chinese).
[5] 林浩申,刘刚,何兵.自适应空间机动目标跟踪算法[J].中国空间科学技术,2017,37(1):26-32.LIN H S,LIU G,HE B.A localization algorithm for adaptive maneuvering spatial targets[J].Chinese Space Science and Technology,2017,37(1):26-32(in Chinese).
[6] 刘钧彬,谢维信,李良群.一种遮挡检测的核最小二乘视觉跟踪算法[J].信号处理,2017,33(4):505-510.LIU J B,XIE W X,LI L Q.Visual tracking based on kernelized least squares with occlusion detecting[J].Journal of Signal Processing,2017,33(4):505-510(in Chinese).
[7] BOLME D S,BEVERIDGE J R,DRAPER B A,et al.Visual object tracking using adaptive correlation filters[C].Computer Vision and Pattern Recognition.IEEE,2010:2544-2550.
[8] DANELLJAN M,HAGER G,KHAN F S,et al.Discriminative scale space tracking[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2016 (99):1-14.
[9] HENRIQUES J F,CASEIRO R,MARTINS P,et al.High-speed tracking with kernelized correlation filters[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2015,37(3):583-596.
[10] XU J,LU Y,LIU J.Robust tracking via weighted spatio-temporal context learning[C].IEEE International Conference on Image Processing,2014:413-416.
[11] WU Y,LIM J,YANG M.Object tracking benchmark[J].IEEE Trans.Pattern Anal.Mach.Intell,2015,37(9):1834-1848.
[12] 王春平,王暐,刘江义,等.基于色度饱和度-角度梯度直方图特征的尺度自适应核相关滤波跟踪[J].光学精密工程,2016,24(9):2293-2301.WANG C P,WANG W,LIU J Y,et al.Scale adaptive kernelized correlation filter tracking based on HHS-OG feature[J].Optics and Precision Engineering,2016,24(9):2293-2301(in Chinese).
[13] 金永涛,李旭青,张周威等.基于多特征的高分遥感图像分割算法研究[J].中国空间科学技术,2016,36(6):38-46.JIN Y T,LI X Q,ZHANG Z W,et al.Segmentation of high resolution multi spectral remote sensing image based on multi feature[J].Chinese Space Science and Technology,2016,36(6):38-46(in Chinese).
[14] 孙卜郊,周东华.基于NCC的存在旋转的图像匹配方法[J].传感器与微系统,2008,27(5):43-45.SUN B J,ZHOU D H.Rotated image registration method based on NCC[J].Transducer and Microsystem Technologies,2008,27(5):43-45(in Chinese).
[15] JIANBO S,CARLO T.Good features to track[C].IEEE Conference on Computer Vision and Pattern Recognition,1994:593-600.
[16] CHEN Z W,CHIANG C C,HSIEH Z T.Extending 3D Lucas-Kanade tracking with adaptive templates for head pose estimation[J].Machine Vision and Applications,2010,21(6):889-903.
[17] FORTUN D,BOUTHEMY P,KERVRANN C.Optical flow modeling and computation:a survey[J].Computer Vision & Image Understanding,2015,134:1-21.
[18] HU M.Visual pattern recognition by moment invariants[J].Ire Transactions on Information Theory,1962,8(2):179-187.
[19] 黄敏,马亚琼,朱颢东,等.基于Hu修正不变矩的商标识别[J].华中师范大学学报(自然科学版),2014,48(2):182-185.HUANG M,MA Y Q,ZHU H D.Rrademark recognition based on Hu modified invariant moment[J].Journal of Huazhong Normal University(Natural Sciences),2014,48(2):182-185(in Chinese).
[20] 贺一楠,耿娟,秦军,等.结合Forstner与NCC约束的UAV图像配准研究[J].国土资源遥感,2015,27(1):48-54.HE Y N,GENG J,QIN J.Integration of Forstner and NCC constraint for UAV image registration[J].Remote Sensing For Land & Resources,2015,27(1):48-54(in Chinese).
[21] 宁晓琳,白鑫贝,姜丽伟.一种基于连续星图观测的自旋姿态估计方法[J].中国空间科学技术,2013,33(6):49-57.NING X L,BAI X B,JIANG L W.New method for attitude determination of spin satellite based on continuous star map observation [J].Chinese Space Science and Technology,2013,33(6):49-57(in Chinese).
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |