吉林一号视频星数据在车辆检测中的可行性
|
摘要
针对新型高分辨率视频卫星数据应用于车辆检测问题,提出了一种包括车辆检测、交通密度计算、车速估算3个主要功能的车辆自动检测的算法。首先,基于自适应背景估计算法,利用背景差分法对每帧图像进行处理,得到静态背景图,再通过帧间差分法提取车辆得到车辆检测结果;其次,对车辆做缓冲分析得到每个缓冲区的车辆数量,利用交通密度公式估计出每帧图像中的车辆在整个道路网中的交通密度值;然后,利用车速估计公式估计出车辆的速度;最后,按上述方法得到整个视频的车辆自动检测分析结果。以"吉林一号"视频星数据为例进行了实验验证,经与其他方法比较证明该算法具有很好的可行性以及适用性。
Aiming at the problem of vehicle detection using a certain high-resolution video satellite data,this paper presents an algorithm for vehicle automatic detection,which includes three aspects:vehicle detection,traffic density calculation and vehicle speed estimation.Firstly,based on the adaptive background estimation algorithm,the background differential method is used to process each image,the static background image is obtained,and the vehicle detection result is obtained by the interframe differential method.Secondly,each vehicle is buffered to get the number of vehicles in each buffer zone,and the traffic density value of the vehicles in each frame is estimated in the whole road network by using the traffic density formula.Then,the speed of the vehicle is estimated by the vehicle speed estimation formula.Finally,the result of the vehicle automatic detection and analysis of the whole video is obtained by the above method.Verified by Jilin No.1 data,the experimental results show that the algorithm has good feasibility and applicability,compared with other methods.
Aiming at the problem of vehicle detection using a certain high-resolution video satellite data,this paper presents an algorithm for vehicle automatic detection,which includes three aspects:vehicle detection,traffic density calculation and vehicle speed estimation.Firstly,based on the adaptive background estimation algorithm,the background differential method is used to process each image,the static background image is obtained,and the vehicle detection result is obtained by the interframe differential method.Secondly,each vehicle is buffered to get the number of vehicles in each buffer zone,and the traffic density value of the vehicles in each frame is estimated in the whole road network by using the traffic density formula.Then,the speed of the vehicle is estimated by the vehicle speed estimation formula.Finally,the result of the vehicle automatic detection and analysis of the whole video is obtained by the above method.Verified by Jilin No.1 data,the experimental results show that the algorithm has good feasibility and applicability,compared with other methods.
引文
[1]李德仁,李清泉,杨必胜,等.3S技术与智能交通[J].武汉大学学报(信息科学版),2008,33(4):331-336.
[2]STEWART J,LUIS F M,PAUL S,et al.A flexible,mobile video camera system and open source video analysis software for road safety and behavioural analysis[J].Transportation Research Record Journal of the Transportation Research Board,2013,2365:90-98.
[3]HUA S,WUA J,XUB L.Real-time traffic congestion detection based on video analysis[J].J.Inf.Comput.Sci,2012,9(10):2907-2914.
[4]杨静.基于视频的交通量参数检测技术研究[D].北京:北方工业大学,2010.
[4]王俊.智能交通信息采集子系统的研究与实现[D].武汉:华中科技大学,2006.
[5]BAS E.Road and traffic analysis from video[D].Istanbul:Koc University,2007.
[6]LARSEN S,KOREN H,SOLBERG R.Traffic monitoring using very high resolution satellite imagery[J].Photogrammetric Engineering&Remote Sensing,2009,75(7):859-869.
[7]GERHARDINGER A,EHRLICH D,PESARESI M.Vehicles detection from very high resulution satellite imagery[J].International Archives of Photogrammetry,Remote Sensing and Spatial Information Scienses,2005,24(3):83-88.
[8]EIKVIL L,AURDAL L,KOREN H.Classification-based vehicle detection in high-resolution satellite images[J].ISPRS Journal of Photogrammetry&Remote Sensing,2009,64(1):65-72.
[9]TAN Q,WEI Q,YANG S,et al.Evaluation of urban road vehicle detection from high resolution remote sensing imagery using object-oriented method[C]∥Urban Remote Sensing Event,2009Joint.IEEE,2009:1-6.
[10]TAN Q,WEI Q,HU J,et al.Road vehicle detection using fuzzy logic rule-based method[C]∥Fuzzy Systems and Knowledge Discovery(FSKD),2010Seventh International Conference on.IEEE,2010:1355-1358.
[11]JIN X,DAVIS C H.Vector-guided vehicle detection from high-resolution satellite imagery[C]∥Geoscience and Remote Sensing Symposium,2004.IGARSS′04.Proceedings.2004IEEE International.IEEE Xplore,2004:1095-1098.
[12]JIN X,DAVIS C H.Vehicle detection from high-resolution satellite imagery using morphological shared-weight neural networks[J].Image&Vision Computing,2007,25(9):1422-1431.
[13]刘韬.国外视频卫星发展研究[J].国际太空,2014(9):50-56.
[14]KOPSIAFTIS G,KARANTZALOS K.Vehicle detection and traffic density monitoring from very high resolution satellite video data[C]∥Geoscience and Remote Sensing Symposium(IGARSS),2015IEEE International.IEEE,2015:1881-1884.
[15]TAO Y,QUAN P,LI J,et al.Real-time multiple objects tracking with occlusion handling in dynamic scenes[J].CVPR,2005,1:970-975.
[2]STEWART J,LUIS F M,PAUL S,et al.A flexible,mobile video camera system and open source video analysis software for road safety and behavioural analysis[J].Transportation Research Record Journal of the Transportation Research Board,2013,2365:90-98.
[3]HUA S,WUA J,XUB L.Real-time traffic congestion detection based on video analysis[J].J.Inf.Comput.Sci,2012,9(10):2907-2914.
[4]杨静.基于视频的交通量参数检测技术研究[D].北京:北方工业大学,2010.
[4]王俊.智能交通信息采集子系统的研究与实现[D].武汉:华中科技大学,2006.
[5]BAS E.Road and traffic analysis from video[D].Istanbul:Koc University,2007.
[6]LARSEN S,KOREN H,SOLBERG R.Traffic monitoring using very high resolution satellite imagery[J].Photogrammetric Engineering&Remote Sensing,2009,75(7):859-869.
[7]GERHARDINGER A,EHRLICH D,PESARESI M.Vehicles detection from very high resulution satellite imagery[J].International Archives of Photogrammetry,Remote Sensing and Spatial Information Scienses,2005,24(3):83-88.
[8]EIKVIL L,AURDAL L,KOREN H.Classification-based vehicle detection in high-resolution satellite images[J].ISPRS Journal of Photogrammetry&Remote Sensing,2009,64(1):65-72.
[9]TAN Q,WEI Q,YANG S,et al.Evaluation of urban road vehicle detection from high resolution remote sensing imagery using object-oriented method[C]∥Urban Remote Sensing Event,2009Joint.IEEE,2009:1-6.
[10]TAN Q,WEI Q,HU J,et al.Road vehicle detection using fuzzy logic rule-based method[C]∥Fuzzy Systems and Knowledge Discovery(FSKD),2010Seventh International Conference on.IEEE,2010:1355-1358.
[11]JIN X,DAVIS C H.Vector-guided vehicle detection from high-resolution satellite imagery[C]∥Geoscience and Remote Sensing Symposium,2004.IGARSS′04.Proceedings.2004IEEE International.IEEE Xplore,2004:1095-1098.
[12]JIN X,DAVIS C H.Vehicle detection from high-resolution satellite imagery using morphological shared-weight neural networks[J].Image&Vision Computing,2007,25(9):1422-1431.
[13]刘韬.国外视频卫星发展研究[J].国际太空,2014(9):50-56.
[14]KOPSIAFTIS G,KARANTZALOS K.Vehicle detection and traffic density monitoring from very high resolution satellite video data[C]∥Geoscience and Remote Sensing Symposium(IGARSS),2015IEEE International.IEEE,2015:1881-1884.
[15]TAO Y,QUAN P,LI J,et al.Real-time multiple objects tracking with occlusion handling in dynamic scenes[J].CVPR,2005,1:970-975.