传统遥感卫星图像处理采用在地面进行目标检测和跟踪处理的模式，卫星将拍摄到的图像数据下传至地面数据处理中心，由地面数据处理系统对接收到的遥感图像数据进行目标检测和跟踪。然而，随着遥感图像分辨率的不断提高，需要下传的数据量增大，导致星地数据传输的时间大大增加，最终造成目标检测和跟踪的时效性降低。针对上述问题，提出一种基于多特征量判别的Canny边缘检测和联合概率数据关联的在轨海上多运动舰船目标检测和跟踪方法。将该方法利用中国科学院微小卫星创新研究院的高分微纳卫星实测数据在模拟星载的嵌入式开发平台上进行验证，结果表明该方法能够在轨对海上多运动舰船目标进行快速、准确的检测和跟踪。
The traditional method of remote sensing satellite image processing utilizes target detection and target tracking on the ground. The satellite transmits the image data to the data processing center on the ground, and the ground data processing system implements target detection and tracking based on the received remote sensing image data. However, with the spatial resolution increase of remote sensing images, the amount of data needed to be transmitted increases greatly. This results in a significant increase in ground-to-satellite data transmission time. Ultimately the efficiency of target detection and tracking reduces. In order to solve the problem, we propose a method based on the multi-feature Canny edge detection algorithm and the joint probability data association algorithm for moving multi-ship detection and tracking by on-orbit satellite. The proposed method is validated by using real data of an optical remote sensing satellite on satellite-based embedded development platform. The results show that the proposed method detects and tracks the multiple ships quickly and accurately.
[1] Ip F, Dohm J M, Baker V R, et al. Flood detection and monitoring with the autonomous sciencecraft experiment onboard E0-1[J].Remote Sensing of Environment, 2006, 101(4):463-481.
[2] 刘小宁. 星上SAR实时成像处理关键技术研究[D]. 北京:北京理工大学, 2016.
[3] 徐伟, 朴永杰. 从Pleiades剖析新一代高性能小卫星技术发展[J]. 中国光学, 2013, 6(1):9-19.
[4] 徐凯. 基于DSP图像处理系统的多目标跟踪关键技术研究及实现[D]. 成都:电子科技大学, 2018.
[5] 高建伟, 李磊, 姚睿, 等. 基于卡尔曼滤波的弱小目标实时检测与跟踪[J]. 计算机工程, 2012, 38(2):4-7.
[6] Tian J, Cui W, Xia X G, et al. A new motion parameter estimation algorithm based on SDFC-LVT[J]. IEEE Transactions on Aerospace and Electronic Systems, 2016, 52(5):2331-2346.
[7] Sittler R W. An optimal data association problem in surveillance theory[J]. IEEE Transactions on Military Electronics, 2007, 8(2):125-139.
[8] 鹿强, 吴琳, 陈昭, 等. 海上目标多源轨迹数据关联综述[J]. 地球信息科学学报, 2018, 20(5):571-581.
[9] 刘勇, 姚力波, 吴昱舟, 等. 静轨光学卫星与自动识别系统的目标点迹关联与误差校正[J]. 电子与信息学报, 2018, 40(7):27-33.
[10] 苏骏, 李亚安, 陈晓, 等双观测站水下纯方位多目标跟踪的数据关联算法[J]. 水下无人系统学报, 2018, 26(2):115-121.
[11] 郭新民, 吕鹏. 一种新的近程多目标轨迹参数测量方法[J]. 雷达科学与技术, 2017, 15(6):661-665.
[12] 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.
[13] 袁益琴, 何国金, 王桂周, 等. 背景差分与帧间差分相融合的遥感卫星视频运动车辆检测方法[J]. 中国科学院大学学报, 2018, 35(1):50-58.
[14] 王保云, 杨昆. 基于最佳阈值分割的舰船目标检测方法[J]. 云南师范大学学报(自然科学版), 2014, 34(4):55-60.
[15] 艾加秋, 齐向阳. 一种基于独立联合K-分布CFAR的舰船检测算法[J]. 中国科学院研究生院学报, 2010, 27(5):651-656.
[16] Tang J, Deng C, Huang G B, et al. Compressed-domain ship detection on spaceborne optical image using deep neural network and extreme learning machine[J]. IEEE Transactions on Geoscience & Remote Sensing, 2015, 53(3):1174-1185.
[17] Canny J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6):679-698.
[18] 蓝贵文, 吴昊铮, 张强, 等. 一种基于CUDA的改进Canny边缘检测算法[J]. 桂林理工大学学报, 2018, 38(3):494-500.
[19] Bar-shalom Y. Multitarget-multisensor tracking:applications and advances[M]. London:Artech House, 2000:77-231.
[20] 何友, 修建娟, 关欣. 雷达数据处理及应用[M]. 北京:电子工业出版社, 2013:111-133.
[21] Daum F. Multitarget-multisensor tracking:principles and techniques[J]. IEEE Aerospace and Electronic Systems Magazine, 1996, 11(2):41-44.
[22] Blackman S S. Multiple hypothesis tracking for multiple target tracking[J]. IEEE Aerospace and Electronic Systems Magazine, 2004, 19(1):5-18.
[23] Cox I J. A review of statistical data association techniques for motion correspondence[J]. International Journal of Computer Vision, 1993, 10(1):53-66.
[24] Oh S, Russell S, Sastry S. Markov chain Monte Carlo data association for multi-target tracking[J]. IEEE Transactions on Automatic Control, 2009, 54(3):481-497.
[25] Cong S, Hong L. Computational complexity analysis for multiple hypothesis tracking[C]//Proceedings of the 36th IEEE Conference on Decision and Control, San Diego, CA, 1997, (5):4991-4996.
[26] 闸旋. CPU+GPU单机异构环境下遥感数据并行处理技术研究[D]. 郑州:中国人民解放军信息工程大学, 2013.
[27] 顾久祥, 杨仁忠, 石璐, 等. 基于GPU的HJ-1C实时成像处理技术[J]. 中国科学院大学学报, 2014, 31(5):708-713.
[28] Plaza A, Du Q, Chang Y L, et al. High performance computing for hyperspectral remote sensing[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2011, 4(3):528-544.
