An evaluation of Guided Regularized Random Forest for classification and regression tasks in remote sensing     

《International Journal of Applied Earth Observation and Geoinformation》2020

New Earth observation missions and technologies are delivering large amounts of data. Processing this data requires developing and evaluating novel dimensionality reduction approaches to identify the most informative features for classification and regression tasks. Here we present an exhaustive evaluation of Guided Regularized Random Forest (GRRF), a feature selection method based on Random Forest. GRRF does not require fixing a priori the number of features to be selected or setting a threshold of the feature importance. Moreover, the use of regularization ensures that features selected by GRRF are non-redundant and representative. Our experiments based on various kinds of remote sensing images, show that GRRF selected features provides similar results to those obtained when using all the available features. However, the comparison between GRRF and standard random forest features shows substantial differences: in classification, the mean overall accuracy increases by almost 6% and, in regression, the decrease in RMSE almost reaches 2%. These results demonstrate the potential of GRRF for remote sensing image classification and regression. Especially in the context of increasingly large geodatabases that challenge the application of traditional methods.  相似文献   

Smeared crack approach: back to the original track     

M. Cervera  M. Chiumenti 《国际地质力学数值与分析法杂志》2006,30(12):1173-1199

This paper briefly reviews the formulations used over the last 40 years for the solution of problems involving tensile cracking, with both the discrete and the smeared crack approaches. The paper focuses on the smeared approach, identifying as its main drawbacks the observed mesh‐size and mesh‐bias spurious dependence when the method is applied ‘straightly’. A simple isotropic local damage constitutive model is considered, and the (exponential) softening modulus is regularized according to the material fracture energy and the element size. The continuum and discrete mechanical problems corresponding to both the weak discontinuity (smeared cracks) and the strong discontinuity (discrete cracks) approaches are analysed and the question of propagation of the strain localization band (crack) is identified as the main difficulty to be overcome in the numerical procedure. A tracking technique is used to ensure stability of the solution, attaining the necessary convergence properties of the corresponding discrete finite element formulation. Numerical examples show that the formulation derived is stable and remarkably robust. As a consequence, the results obtained do not suffer from spurious mesh‐size or mesh‐bias dependence, comparing very favourably with those obtained with other fracture and continuum mechanics approaches. Copyright © 2006 John Wiley & Sons, Ltd.  相似文献   

Simulation of adaptive optics for the Vacuum Tower Telescope     

Anita Enmark  Thomas Berkefeld  Oskar von der Lühe  Torben Andersen 《Experimental Astronomy》2006,21(2):87-99

A simulation model of the adaptive optics of the German Vacuum Tower Telescope (VTT), Observatorio del Teide, Tenerife, is presented. The model uses modules from the integrated model of the Euro50 extremely large telescope, and includes submodels of a Shack-Hartmann wavefront sensor, a de-formable mirror, a tip-tilt mirror, high-voltage amplifier low-pass filters, a reconstructor and a controller. We investigate the impact on the closed loop bandwidth of changes in controller configuration and certain system parameters, such as low pass filter bandwidth and camera integration and readout time. Control strategies were tested on simple models before implementation on the full VTT model. Using the models, different control strategies are compared.  相似文献   

Prediction of Stellar Atmospheric Parameters using Instance-Based Machine Learning and Genetic Algorithms   总被引：1，自引：0，他引：1  

J. Federico Ramírez  Olac Fuentes  Ravi K. Gulati 《Experimental Astronomy》2001,12(3):163-178

In this article we present a method for the automated prediction of stellar atmospheric parameters from spectral indices. This method uses a genetic algorithm (GA) for the selection of relevant spectral indices and prototypical stars and predicts their properties, using the k-nearest neighbors method (KNN). We have applied the method to predict the effective temperature, surface gravity, metallicity, luminosity class and spectral class of stars from spectral indices. Our experimental results show that the feature selection performed by the genetic algorithm reduces the running time of KNN up to 92%, and the predictive accuracy error up to 35%. This revised version was published online in July 2006 with corrections to the Cover Date.  相似文献   

用遗传算法解算病态方程   总被引：7，自引：1，他引：6  

曾群意  欧吉坤 《大地测量与地球动力学》2003,23(3):93-97

对应用遗传算法解决病态方程问题进行了探讨。利用拟合法而不是通过法方程求解参数，从而避免了法方程系数求逆，使病态方程的解答有了较好的结果。通过模拟计算并和其他方法进行比较，证明该方法是可行的和有效的。  相似文献   

非线性洋面温度反演     

吴雪宝  邱康睦 《海洋预报》1996,13(4):22-26

本文从辐射方程出发，简述洋面温度与亮度温度之间的非线性关系，进行非线性洋面温度反演的方案设计，用正演算法 计算出非线性洋面温度反演公式的回归系数，对样进行检验。  相似文献   

神经网络反演散射计风场算法的研究   总被引：3，自引：1，他引：3  

宋新改  林明森  蒋兴伟 《海洋学报》2006,28(1):42-46

建立了一个神经网络反演卫星散射计海面风场的B-P算法,给出了一个神经网络反演风场的模型,并利用该反演算法和模型对实际卫星散射计数据进行了海面风场反演试验,对风向的多解性利用圆中数滤波方法进行排除.对神经网络训练和检验数据集分别采用ERS-1/2散射计数据和欧洲中期天气预报(ECMWF)提供的风场作为配准点数据.把反演的风速和风向与CMCD4和ECMWF的风场作了比较,它们吻合得比较好;研究表明神经网络反演海面风场是可行和高效的.  相似文献   

计算数据域特征值的两种算法     

洪忠渝 《中国海洋大学学报(自然科学版)》1994,(Z2)

设计两种计算数据域特征值的算法。此特征值通常是用线性反馈移位寄存器（ＬＦ－ＳＲ）组成的特征分析器得到的．在数据已按字节存放和速度要求不高的场合下，本算法将是十分方便和有用的．  相似文献   

基于分块分形编码的多姿态、表情人脸图像检索     

姬光荣  曲翠璐  赵文仓  廖向东  屈新岳 《中国海洋大学学报(自然科学版)》2004,34(5):826-830

利用图像分形编码中定义域块和最小均方误差一一对应的特点，提出了1种基于分形编码的多姿态、表情的人脸图像检索方法。该方法将待检索图像分割为相同大小的值域块，然后将每一值域块按给定的定义域块进行分形编码，得到最小均方误差，计算该最小均方误差与图像库中最小均方误差的欧氏距离。将待检索图像所有值域块的欧氏距离求平均，此平均欧氏距离较小的几幅图像即为检索出的图像。实验证明该方法能够准确地检索出图像库中存储的同一人的不同姿态、表情的图像。  相似文献   

基于MATLAB的基本矩阵随机采样鲁棒估计   总被引：2，自引：0，他引：2  

高晟丽  陈杰  李海滨 《海洋测绘》2006,26(3):31-33,38

基本矩阵的估算可以分为线性算法、非线性优化迭代算法、鲁棒估计算法三大类,而以鲁棒估计效率、效果最为突出,但是实现比较复杂。本文通过数码相机获取的两张未标定照片,采样互相关系数法和RANSAC算法,在MATLAB中实现了基本矩阵的自动拟合,实验表明拟合精度较高、速度快,为下一步三维场景的自动重建和量测奠定了基础。  相似文献