利用无人机影像和广义迭代重加权最小二乘法的三维重建

为了提高基于无人机影像重建三维场景的效率,提出了应用广义迭代重加权最小二乘法的全局式三维重建方法。首先,介绍了广义迭代重加权最小二乘法的基本思想和具体步骤;接着,将全局旋转矩阵的求解转化为李代数中旋转向量的求解问题,利用结合范数优化和迭代重加权最小二乘的方法求出全局旋转最优解;然后,根据极线几何约束条件,解算相机在全局坐标系下的相对平移方向向量,在广义迭代重加权最小二乘架构下,利用二次规划法求出相机全局位置最优解;最后,对影像位姿参数和重构三维点进行光束法平差优化。实验结果表明该方法在提高重建效率的同时,能够更真实地恢复场景的几何形态。     

Regression model for generating time series of daily precipitation amounts for climate change impact studies

The precipitation amounts on wet days at De Bilt (the Netherlands) are linked to temperature and surface air pressure through advanced regression techniques. Temperature is chosen as a covariate to use the model for generating synthetic time series of daily precipitation in a CO₂ induced warmer climate. The precipitation-temperature dependence can partly be ascribed to the phenomenon that warmer air can contain more moisture. Spline functions are introduced to reproduce the non-monotonous change of the mean daily precipitation amount with temperature. Because the model is non-linear and the variance of the errors depends on the expected response, an iteratively reweighted least-squares technique is needed to estimate the regression coefficients. A representative rainfall sequence for the situation of a systematic temperature rise is obtained by multiplying the precipitation amounts in the observed record with a temperature dependent factor based on a fitted regression model. For a temperature change of 3°C (reasonable guess for a doubled CO₂ climate according to the present-day general circulation models) this results in an increase in the annual average amount of 9% (20% in winter and 4% in summer). An extended model with both temperature and surface air pressure is presented which makes it possible to study the additional effects of a potential systematic change in surface air pressure on precipitation.     

Object-Oriented Island Land Cover Change Detection by Iteratively Reweighted Multivariate Statistical Analysis

ABSTRACT

In studies of changes in island geographic information, a key topic is island land cover change detection using multitemporal remote sensing images; such information can provide good reference for island exploitation, marine resource conservation, marine economic development, and national defense building. This study proposes a fast and easy-to-implement algorithm of object-oriented change detection based on iteratively reweighted multivariate statistical analysis with high-resolution remote sensing images and vector data, improving effects of normal multivariate alternative detection by combining object-oriented technology with probability density function of chi-square distribution. This method builds vectors in feature space with spectral, texture, and spatial structure information of objects segmented based on boundaries and categories of features in vector data. By feature optimization and iterative extension to multivariate statistical analysis for change detection, change regions can be extracted. Experiments on Zhaoshu Island area show that the proposed method is highly effective and accurate in detection of change regions.     

相关观测的L1范数最小化方法的比较分析

在抵御粗差影响方面,L_1范数最小化方法比最小二乘更具可靠性。求解L_1范数最小化问题,主要有选权迭代法和线性规划法两种方法。针对相关观测,通常采用权阵的对角线元素来构造L_1范数最小化问题的目标函数,这种处理方法容易忽略观测值之间的相关性。如果采用Cholesky分解消去观测值之间的相关性,则容易造成粗差的转移,进而影响抗差功效。本文对上述两种方法进行了比较分析,数值实验结果表明将相关观测转换为独立等权观测,有利于增强线性规划的稳健性,而在探测粗差方面则具有等价性。由于基于选权迭代的方法收敛性较差,故不适合求解L_1范数最小化问题。     

基于选权迭代估计与非监督分类的多光谱图像变化检测

针对多光谱图像的变化检测问题,提出了一种基于选权迭代估计(iterative estimation with weight selection,IEWS)与非监督分类(unsupervised classification,UC)的多光谱图像变化检测方法。借鉴IEWS的思想,并以类似于迭代加权多元变化检测(iteratively reweighted multivariate alteration detection,IRMAD)的迭代模式进行回归估计,得到初步的变化检测结果;并通过对初始变化信息的UC处理,以及对不同类别的IEWS,得到最终的变化检测结果。利用该方法对TM图像进行了实验,结果表明:所得到的变化信息在空间位置上同该区域相应时间段内土地利用/覆盖的变化情况具有很好的一致性;同时与多元变化检测及IRMAD方法变化检测的结果相比较,表明该方法对相对较小的变化信息具有更好的变化检测能力。     

本质矩阵优化分解的相对位姿估计

针对从影像恢复摄像机相对位姿的问题,提出了一种基于李群表示的本质矩阵快速分解的位姿估计算法。通过加权最小二乘方法优化了本质矩阵;利用本质矩阵和平移向量的关系求出了平移向量;由本质矩阵和位姿参数的等式关系建立目标函数,基于姿态的李群表示推导了旋转矩阵迭代估计过程;优化了唯一解确定的约束条件,避免了特征点的三维重建。仿真实验和真实图像实验表明提出的算法精度和鲁棒性均优于传统算法,算法效率得到明显提高。提出的算法避免了矩阵奇异值分解运算和大量的矩阵计算,而且只需对两组解进行唯一解确定,能够实现相对位姿的快速高精度估计。