共查询到20条相似文献，搜索用时 531 毫秒
1.
2.
3.
遥感影像的复杂模糊性问题会干扰影像变化检测的结果,可引入区间二型模糊C均值聚类算法解决此问题,但算法参数的随机性导致检测结果不稳定。本文首先利用局部最优解优化萤火虫算法中的候选解,引入可变步长因子,以此自适应寻优区间二型模糊C均值聚类算法的模糊因子;然后结合寻优得到的模糊因子进行区间二型模糊C均值聚类,迭代更新隶属区间提取变化信息;最后通过基于复合梯形法则的加权Karnik-mendel算法降型和解模糊优化聚类中心,依据最大隶属度原则判断变化类型。通过试验验证,本文方法得到更优模糊因子和更精确的聚类中心,具有更好的稳健性,变化检测精度得到提高,检测得到的变化区域更加精细。 相似文献
4.
针对线性化近似法模型误差大,牛顿迭代法和高斯-牛顿法局部收敛等不足,而同伦算法在非线性数据处理方面有独到优势。通过对同伦路径跟踪过程中牛顿迭代终止判据和步长控制策略进行了改进,得到了一种快速、稳定的同伦路径跟踪算法。 相似文献
5.
阐述了GIS网络分析中不确定性选址问题的基本模型及特性。从问题的定义可知其为NP完备类问题。推导了最优解在紧条件的下界算法,并结合广义Powell算法及遗传算法,提出了不确定性选址问题的混合遗传算法,实验证明,在最优解的品质和收敛速度上都达到了比较好的效果。同时,实验的结果从另一个角度证明,如果兼顾收敛速度和解的品质这两个指标,单纯的遗传算法未必比其他搜索算法更优越,采用一些局部搜索性能较好的算法结合遗传算法,可以从两方面改善求解效果。 相似文献
6.
针对现有分布式目标通道最优极化算法存在运算量偏大等问题,以算法的实用性为目的,在理论分析的基础上提出一种求解分布式目标通道最优极化的快速算法.首先将任意通道下的天线接收功率模型统一为同极化通道下的天线接收功率拓展模型,然后对该拓展模型进行变换极化基处理,在此基础上理论确定目标最优极化在(r,θ)平面上的位置区间,从而为简化目标最优极化求解或预判目标最优极化位置等提供理论支撑.为获取目标最优极化,采用区间二分法在目标最优极化位置区间内迭代搜索.通过对比实验对算法运算速度和实现方面予以验证.Abstract： Aiming at the large mount of calculation of the traditional algorithms for polarimetric power optimization of random target in arbitrary channel, a fast algorithm is proposed in this article. Firstly, the function of received power in arbitrary channel is unified as the form of copolar power. Then based on the change of polarimetric basis,the copotar power is analyzed theoretically to obtain the minimum interval of target optimal polarization state in (r,θ)plane, which provides theoretic support for simplifying the process of obtaining optimal polarization states or anticipating their positions. In order to obtain optimal polarization states, the interval dichotomy is used to search in the minimum interval of target optimal polarization state, The experiment results have demonstrated that the proposed algorithm has better performance than the Lagrange multiplier method or the method of traversal search in algorithmic realization and computational speed. 相似文献
7.
在对基本果蝇优化算法的优化流程进行深入分析的基础上,通过改变其随机搜索方向与增加搜索半径调整系数,给出了一种改进的果蝇优化算法(IFOA)。并在IFOA算法的目标函数中引入正则化项,提出了将IFOA算法与Tikhonov正则化方法进行结合以进行病态问题解算的方法。通过实例分析表明:该方法的解算精度要优于遗传算法和单一的Tikhonov正则化方法;在观测值含有粗差时,使用最小二乘法进行求解,其结果与真值的偏差会迅速增大,而此时本文方法的解算结果具有一定的稳健性。与以遗传算法为代表的智能搜索方法相比,本文方法具有参数设置少、计算速度快、寻优过程简单等特点,在病态问题解算中更具有实用性。 相似文献
8.
9.
基于自适应遗传算法和改进BP算法的遥感影像分类 总被引：1，自引：0，他引：1
介绍了采用自适应遗传算法和改进BP算法相结合的混合算法来训练BP网络的方法，即先用自适应遗传算法进行全局训练，再用改进BP算法进行精确训练，以达到加快网络收敛速度和避免陷入局部极小值的目的。结果表明，该算法收敛速度快，分类精度较高。 相似文献
10.
11.
Mudanyal O. Yldz S. Semerci O. Yapar A. Akduman I. 《Geoscience and Remote Sensing Letters, IEEE》2008,5(2):180-184
A microwave imaging method for nondestructive testing of perfectly conducting surfaces beyond a layered media is presented. The method is an adaptation of the surface reconstruction approach by Yapar et al. to the present problem. It is based on the analytical continuation of the measured data to the surface under test through a special representation of the scattered field in terms of Fourier transform and Taylor expansion. Then the problem is reduced to the solution of a nonlinear equation which is solved iteratively via the Newton method and regularization in the least squares sense. Numerical simulations show that defects as small as lambda/500 can be recovered through the presented algorithm. 相似文献
12.
提出了一种用于倒车辅助系统的基于改进Hough变换的车位线识别方法。算法首先采用局部像素分布特征以剔除非直线的干扰像素,然后使用像素直方图自适应地选择与车位线边缘直线相适宜的梯度方向区间,减少参与Hough变换的像素点数的同时减少背景干扰直线边缘的影响,最后通过判断相邻边缘间的像素灰度关系,进一步确认车位线边缘,从而识别出图像中的车位线。相比于传统hough变换和基于梯度方向区间的改进随机Hough变换（Grad-RHT）,本文算法在保持与Grad-RHT运行效率相当的情况下,在各种干扰背景下都取得了更好的车位线边缘直线检测正确率,表现出较好的识别鲁棒性。 相似文献
13.
一种快速、高精度和稳健的影像匹配算法 总被引：5，自引：1，他引：5
在遥感影像应用中,匹配是一项很关键技术,它的优劣直接关系到后续成果的质量.而匹配质量本身的优劣又取决于匹配的速度、精度和可靠性.针对这三个方面的因素,本文探讨一种基于小波变换、遗传算法和最小二乘匹配的快速高精度影像匹配组合算法.小波变换用于减少匹配搜索数据,遗传算法用于优化搜索空间,而最小二乘匹配则用于获得子像素匹配精度.同时为提高遗传搜索的稳定性和匹配结果的可靠性,本文探讨一种改进的自适应算子概率遗传算法.该改进的基本原理是杂交概率依选择的两父代染色体间的Hamming距离而自适应的变化,变异概率则依父代个体的适应度而自适应的变化.实验结果表明:与传统方法相比,本文方法具有明显的性能优势. 相似文献
14.
针对区间数多指标决策问题中评价值的取值概率问题,提出了基于概率分布的区间数多指标灰色关联模型。该模型将区间数距离映射到概率分布函数上,根据灰色关联分析方法的基本思想,计算决策方案与理想方案之间的关联度,通过关联度对决策方案进行排序。最后给出了两个实例分析,验证了该模型计算结果的正确性与有效性。 相似文献
15.
Tao Wang 《地球空间信息科学学报》2008,11(2):103-106
This paper proposes a new algorithm for determining the starting points of contour lines. The new algorithm is based on the interval tree. The result improves the algorithm＇s efficiency remarkably. Further, a new strategy is designed to constrain the direction of threading and the resulting contour bears more meaningful information. 相似文献
16.
数字高程模型是数字地形模拟的重要成果,作为空间数据框架中的主要内容,也成为各种地理信息的载体。在铁路、城轨、公路等线形构筑物的设计和建设中,沿线两边的地形信息是工程设计和建设的基础。本文结合带状区域的特点以及数字高程模型本身的特点,从工程实用的角度出发,围绕带状区域中多块不等采样间隔DEM数据在工程应用中的问题,重点分析了多块不等采样间隔DEM之间拼接的问题,确定了一种不等采样间隔的DEM间无缝拼接的方法,简单介绍了多块不等间隔数字高程模型的裁剪方法和三维可视化方法。 相似文献
17.
18.
The least squares collocation algorithm for estimating gravity anomalies from geodetic data is shown to be an application
of the well known regression equations which provide the mean and covariance of a random vector (gravity anomalies) given
a realization of a correlated random vector (geodetic data). It is also shown that the collocation solution for gravity anomalies
is equivalent to the conventional least-squares-Stokes' function solution when the conventional solution utilizes properly
weighted zero a priori estimates. The mathematical and physical assumptions underlying the least squares collocation estimator
are described. 相似文献
19.
An operational algorithm for computation of terrain correction (or local gravity field modeling) based on application of closed-form solution of the Newton integral in terms of Cartesian coordinates in multi-cylindrical equal-area map projection of the reference ellipsoid is presented. Multi-cylindrical equal-area map projection of the reference ellipsoid has been derived and is described in detail for the first time. Ellipsoidal mass elements with various sizes on the surface of the reference ellipsoid are selected and the gravitational potential and vector of gravitational intensity (i.e. gravitational acceleration) of the mass elements are computed via numerical solution of the Newton integral in terms of geodetic coordinates {,,h}. Four base- edge points of the ellipsoidal mass elements are transformed into a multi-cylindrical equal-area map projection surface to build Cartesian mass elements by associating the height of the corresponding ellipsoidal mass elements to the transformed area elements. Using the closed-form solution of the Newton integral in terms of Cartesian coordinates, the gravitational potential and vector of gravitational intensity of the transformed Cartesian mass elements are computed and compared with those of the numerical solution of the Newton integral for the ellipsoidal mass elements in terms of geodetic coordinates. Numerical tests indicate that the difference between the two computations, i.e. numerical solution of the Newton integral for ellipsoidal mass elements in terms of geodetic coordinates and closed-form solution of the Newton integral in terms of Cartesian coordinates, in a multi-cylindrical equal-area map projection, is less than 1.6×10–8 m2/s2 for a mass element with a cross section area of 10×10 m and a height of 10,000 m. For a mass element with a cross section area of 1×1 km and a height of 10,000 m the difference is less than 1.5×10–4m2/s2. Since 1.5× 10–4 m2/s2 is equivalent to 1.5×10–5m in the vertical direction, it can be concluded that a method for terrain correction (or local gravity field modeling) based on closed-form solution of the Newton integral in terms of Cartesian coordinates of a multi-cylindrical equal-area map projection of the reference ellipsoid has been developed which has the accuracy of terrain correction (or local gravity field modeling) based on the Newton integral in terms of ellipsoidal coordinates.Acknowledgments. This research has been financially supported by the University of Tehran based on grant number 621/4/859. This support is gratefully acknowledged. The authors are also grateful for the comments and corrections made to the initial version of the paper by Dr. S. Petrovic from GFZ Potsdam and the other two anonymous reviewers. Their comments helped to improve the structure of the paper significantly. 相似文献
20.
GPS/INS组合导航非线性系统最优估计算法中,基于统计信息和假设检验理论的多渐消因子自适应滤波算法的应用前提条件是残差向量为高斯白噪声。本文针对观测异常会影响残差向量的数字特性分布,提出了一种神经网络辅助的多重渐消因子自适应SVD-UKF算法。该算法采用神经网络算法削弱观测异常对残差序列高斯白噪声分布特性的影响,利用奇异值分解抑制UKF中先验协方差矩阵负定性变化,同时构造多重渐消因子对预测状态协方差阵进行调整,使得不同的滤波通道具有不同的调节能力,高效地应用于多变量复杂系统。最后利用车载实测数据进行了验证。结果表明,神经网络算法极大削弱了观测粗差对残差序列高斯白噪声分布特性的影响,拓展了多重渐消因子的应用范围,使其能在观测值含有粗差的条件下自适应调节不同滤波通道,消除滤波状态中的异常,提高组合导航解的精度和可靠性。 相似文献