遥感激光图像的移动特征定位跟踪方法
|
摘要
针对SVM方法在遥感激光图像移动特征的定位跟踪方面存在误差率较大、效率较低的弊端,提出新的遥感激光图像特征定位跟踪方法,其基于遥感激光图像特征获取过程中存在碎云杂波遮盖图像主要内容的情况,增加图像中灰度参数,提高图像对比度与画质,采用最小二乘法获取准确的图像特征参数信息,完成遥感激光图像特征提取。采用基于约束领域小基团特征能量的移动特征定位跟踪方法,基于遥感激光图像特征,获取特征点的似然能量与先验能量,运算获取的目标函数是特征点的后验能量,基于后验能量实现遥感激光图像的移动特征定位跟踪。实验结果表明,所提方法在遥感激光图像特征定位跟踪方面存在误差小、效率高的优势。
Aiming at the disadvantages of SVM method in the location and tracking of remote sensing laser image moving features,the new remote sensing laser image feature localization tracking method is proposed. Based on the remote sensing laser image feature acquisition process,there are broken cloud clutter. Covering the main content of the image,increasing the grayscale parameters in the image,improving the image contrast and image quality,using the least squares method to obtain accurate image feature parameter information,and completing the feature extraction of remote sensing laser image. And the moving feature localization tracking method based on the small group feature energy of the constrained domain is used to acquire the likelihood energy and the prior energy of the feature point based on the characteristics of the remote sensing laser image. The objective function obtained by the operation is the posterior energy of the feature point. The moving feature location tracking of remote sensing laser images is realized based on the posterior energy. The experimental results show that the proposed method has the advantages of small error and high efficiency in the feature tracking of remote sensing laser image.
Aiming at the disadvantages of SVM method in the location and tracking of remote sensing laser image moving features,the new remote sensing laser image feature localization tracking method is proposed. Based on the remote sensing laser image feature acquisition process,there are broken cloud clutter. Covering the main content of the image,increasing the grayscale parameters in the image,improving the image contrast and image quality,using the least squares method to obtain accurate image feature parameter information,and completing the feature extraction of remote sensing laser image. And the moving feature localization tracking method based on the small group feature energy of the constrained domain is used to acquire the likelihood energy and the prior energy of the feature point based on the characteristics of the remote sensing laser image. The objective function obtained by the operation is the posterior energy of the feature point. The moving feature location tracking of remote sensing laser images is realized based on the posterior energy. The experimental results show that the proposed method has the advantages of small error and high efficiency in the feature tracking of remote sensing laser image.
引文
[1]尚敏.基于船舶卫星遥感图像的目标特征算法[J].舰船科学技术,2015,37(5):159-162.
[2]RAMASAMY M,LEISHMAN J G.Benchmarking Particle Image Velocimetry with Laser Doppler Velocimetry for Rotor Wake Measurements[J].Aiaa Journal,2015,45(45):2622-2633.
[3]王强,张合新,孟飞,等.激光图像可匹配性评价模型[J].激光与红外,2016,46(11):1408-1413.
[4]盛亮,彭博栋,袁媛,等.表面绝缘混合平面丝阵Z箍缩激光阴影图像诊断[J].物理学报,2014,63(23):225-231.
[5]G M Mikheev,K G Mikheev,T N Mogileva,et al.Laser image recording on detonation nanodiamond films[J].Quantum Electronics,2014,44(44):1-3.
[6]周鑫,孙剑峰,姜鹏,等.基于阈值分割方法的复杂场景激光图像目标提取算法研究[J].光学学报,2014,34(b12):149-157.
[7]范有臣,赵洪利,孙华燕,等.像质评价结合互信息在激光图像配准中的应用[J].光学精密工程,2015,23(10z):660-667.
[8]赵宇枫,徐科.基于约束流体化的激光图像特征三维重建算法[J].机床与液压,2017,45(18):34-40.
[9]蒋理,李维勇.红外激光成像中的弱小目标的准确定位技术研究[J].现代电子技术,2016,39(20):83-86.
[10]崔治,邓曙光,肖卫初.利用HSSIM和残差比阈值的3维激光扫描图像去噪[J].激光技术,2015,39(5):669-673.
[11]丁小康,闫磊,孔建磊,等.基于二维激光与图像的人工林采育目标检测方法[J].林业科学,2015,51(7):129-135.
[12]TRAVELLETTI J,MALET J P,DELACOURT C.Imagebased correlation of Laser Scanning point cloud time series for landslide monitoring[J].International Journal of Applied Earth Observation&Geoinformation,2014,32(1):1-18.
[13]李莉,秦勤,李建.基于粒子群优化的激光光斑图像增强改进算法[J].计算机工程与设计,2014,35(3):963-967.
[14]蔡健荣,刘梦雷,孙力,等.基于改进惯性矩算法的冷鲜猪肉新鲜度激光散斑图像检测[J].农业工程学报,2017,33(7):268-274.
[15]童佳楠.基于锚点建图的半监督分类在遥感图像中的应用[J].现代电子技术,2016,39(22):92-96.
[16]戴路,张晓晖.高重频距离选通水下激光雷达图像噪声特性分析[J].电光与控制,2015,22(10):85-88.
[17]梁文斌,谢跃雷,彭诚诚,等.一种激光标刻图像验证装置的研究与设计[J].电视技术,2015,39(5):49-52.
[18]OVELAND I,HAUGLIN M,GOBAKKEN T,et al.Automatic Estimation of Tree Position and Stem Diameter Using a Moving Terrestrial Laser Scanner[J].Remote Sensing,2017,9(4):350.
[19]徐苗,孙柯,王卓,等.激光散射图像检测桃果实货架期的食用价值[J].食品科学,2016,37(18):244-249.
[2]RAMASAMY M,LEISHMAN J G.Benchmarking Particle Image Velocimetry with Laser Doppler Velocimetry for Rotor Wake Measurements[J].Aiaa Journal,2015,45(45):2622-2633.
[3]王强,张合新,孟飞,等.激光图像可匹配性评价模型[J].激光与红外,2016,46(11):1408-1413.
[4]盛亮,彭博栋,袁媛,等.表面绝缘混合平面丝阵Z箍缩激光阴影图像诊断[J].物理学报,2014,63(23):225-231.
[5]G M Mikheev,K G Mikheev,T N Mogileva,et al.Laser image recording on detonation nanodiamond films[J].Quantum Electronics,2014,44(44):1-3.
[6]周鑫,孙剑峰,姜鹏,等.基于阈值分割方法的复杂场景激光图像目标提取算法研究[J].光学学报,2014,34(b12):149-157.
[7]范有臣,赵洪利,孙华燕,等.像质评价结合互信息在激光图像配准中的应用[J].光学精密工程,2015,23(10z):660-667.
[8]赵宇枫,徐科.基于约束流体化的激光图像特征三维重建算法[J].机床与液压,2017,45(18):34-40.
[9]蒋理,李维勇.红外激光成像中的弱小目标的准确定位技术研究[J].现代电子技术,2016,39(20):83-86.
[10]崔治,邓曙光,肖卫初.利用HSSIM和残差比阈值的3维激光扫描图像去噪[J].激光技术,2015,39(5):669-673.
[11]丁小康,闫磊,孔建磊,等.基于二维激光与图像的人工林采育目标检测方法[J].林业科学,2015,51(7):129-135.
[12]TRAVELLETTI J,MALET J P,DELACOURT C.Imagebased correlation of Laser Scanning point cloud time series for landslide monitoring[J].International Journal of Applied Earth Observation&Geoinformation,2014,32(1):1-18.
[13]李莉,秦勤,李建.基于粒子群优化的激光光斑图像增强改进算法[J].计算机工程与设计,2014,35(3):963-967.
[14]蔡健荣,刘梦雷,孙力,等.基于改进惯性矩算法的冷鲜猪肉新鲜度激光散斑图像检测[J].农业工程学报,2017,33(7):268-274.
[15]童佳楠.基于锚点建图的半监督分类在遥感图像中的应用[J].现代电子技术,2016,39(22):92-96.
[16]戴路,张晓晖.高重频距离选通水下激光雷达图像噪声特性分析[J].电光与控制,2015,22(10):85-88.
[17]梁文斌,谢跃雷,彭诚诚,等.一种激光标刻图像验证装置的研究与设计[J].电视技术,2015,39(5):49-52.
[18]OVELAND I,HAUGLIN M,GOBAKKEN T,et al.Automatic Estimation of Tree Position and Stem Diameter Using a Moving Terrestrial Laser Scanner[J].Remote Sensing,2017,9(4):350.
[19]徐苗,孙柯,王卓,等.激光散射图像检测桃果实货架期的食用价值[J].食品科学,2016,37(18):244-249.