Object Extraction Based on Evolutionary Morphological Processing

This paper introduces a novel technique for object detection using genetic algorithms and morphological processing. The method employs a kind of object oriented structure element, which is derived by genetic algorithms. The population of morphological filters is iteratively evaluated according to a statistical performance index corresponding to object extraction ability, and evolves into an optimal structuring element using the evolution principles of genetic search. Experimental results of road extraction from high resolution satellite images are presented to illustrate the merit and feasibility of the proposed method.     

用瞬态瑞雷波反演横波速度评价高速公路压碾效果

在高速公路质量检测中 ,路堤土压实度是衡量土方路基重要的质量指标 ,提出了提高压实度计算精度的处理流程 ,应用τ-p变换提取面波 ,获取高信噪比的频散曲线 ,采用瞬态面波反演横波速度 ,提高横波速度计算精度 ,进而计算压实度。用上述方法对商开高速公路振冲压碾效果进行了无损检测 ,其结果与取样分析及开挖结果一致     

动态遗传算法及其在P波初动震源机制解中的应用

利用动态遗传算法对Ｐ波初动震源机制解进行反演计算，并与前人常用的反演计算方法对比，认为遗传算法在Ｐ波初动 震源机制解中的应用是有很有效的。     

Generating basis sets of double differences

When a collection of double differences is used to compute global-positioning-system satellite orbits from a permanent network of receiving stations, linear dependence among the double-differenced observations reduces the number of double differences that contribute new information to the computations. A maximal linearly independent subset of a large collection of double differences contains all the information content of the full set. If r is the number of receivers and s is the number of satellites, the original collection of double differences may have size O(r ² s ²), whereas the linearly independent subset has size no greater than O(rs). Only such a smaller independent subset needs to participate in the more expensive double-precision matrix computations to correctly correlate all double differences, detect cycle slips, resolve ambiguities, and compute satellite orbits and station positions and relative velocities. Dependence among double differences is characterized using vector space methods together with geometric characterizations of Boolean matrices. These characterizations lend themselves to fast, robust algorithms for computing maximal linearly independent sets (bases) of double differences. An algorithm is given for constructing a generating independent set of double differences from the Boolean array of receiving-station/satellite connections. Characterizations of generator equivalence allow alternative generating sets to be identified and selected. An updating algorithm to handle local changes in the satellite–receiver connection matrix is also described. Received: 27 August 1996 / Accepted: 28 January 1999     

特种遗产资源研究基本问题--中国世界遗产与国家公园研究形势与任务

特种遗产是本文对世界遗产与国家公园认定、保护、保存、展示和遗传后代的科学研究所作的理论概括,同时界定了特种遗产与特种遗产资源的概念,讨论了中国的特种遗产———世界遗产与国家公园研究形势,提出了特种遗产资源研究的8个基本问题。     

Genetic algorithms in probabilistic finite element analysis of geotechnical problems

In application to numerical analysis of geotechnical problems, the limit-state surface is usually not known in any closed form. The probability of failure can be assessed via the so-called reliability index. A minimization problem can naturally be formed with an implicit equality constraint defined as the limit-state function and optimization methods can be used for such problems. In this paper, a genetic algorithm is proposed and incorporated into a displacement finite element method to find the Hasofer–Lind reliability index. The probabilistic finite element method is then used to analyse the reliability of classical geotechnical systems. The performance of the genetic algorithm (GA) is compared with simpler probability methods such as the first-order-second-moment Taylor series method. The comparison shows that the GA can produce the results fairly quickly and is applicable to evaluation of the failure performance of geotechnical problems involving a large number of decision variables.     

Bayesian inversion with Markov chains—I. The magnetotelluricone-dimensional case

We use Monte Carlo Markov chains to solve the Bayesian MT inverse problem in layered situations. The domain under study is divided into homogeneous layers, and the model parameters are the conductivity of each layer. We use an a priori distribution of the parameters which favours smooth models. For each layer, the a priori and a posteriori distributions are digitized over a limited set of conductivity values.
  The Markov chain relies on updating the model parameters during successive scanning of the domain under study. For each step of the scanning, the conductivity is updated in one layer given the actual value of the conductivity in the other layers. Thus we designed an ergodic Markov chain, the invariant distribution of which is the a posteriori distribution of the parameters, provided the forward problem is completely solved at each step.
  We have estimated the a posteriori marginal probability distributions from the simulated successive values of the Markov chain. In addition, we give examples of complex magnetotelluric impedance inversion in tabular situations, for both synthetic models and field situations, and discuss the influence of the smoothing parameter.     

A spatial approach to forest‐management optimization: linking GIS and multiple objective genetic algorithms

Forest‐management decision‐support systems are largely monolithic structures. Spatial details are left out during the optimization process and are elaborated during the operational planning. This might produce misleading results and plans that are impossible to implement. In this paper, a forest‐management spatial decision‐support systems is presented, in which spatial formulation needed for wildlife models is included during the optimization process. To this end, a multiple‐objective genetic algorithm is combined with a geographical information system. An online spatial evaluation of the objective functions is made possible. This is illustrated by a pilot study in Kirkhill forest, Aberdeen.     

基于神经网络和滤波技术的贮仓结构参数识别

利用神经网络和Kalman滤波技术,提出了一种直接识别结构物理参数的方法,用Kalman滤波技术训练网络。在贮仓振动台实验的基础上,用贮仓在动载作用下的位移、速度作为网络的输入,激振加速度和响应加速度作为网络的输出。仿真计算表明,本文方法是可行的。     

On the importance of training methods and ensemble aggregation for runoff prediction by means of artificial neural networks

ABSTRACT

Artificial neural networks (ANNs) become widely used for runoff forecasting in numerous studies. Usually classical gradient-based methods are applied in ANN training and a single ANN model is used. To improve the modelling performance, in some papers ensemble aggregation approaches are used whilst in others, novel training methods are proposed. In this study, the usefulness of both concepts is analysed. First, the applicability of a large number of population-based metaheuristics to ANN training for runoff forecasting is tested on data collected from four catchments, namely upper Annapolis (Nova Scotia, Canada), Biala Tarnowska (Poland), upper Allier (France) and Axe Creek (Victoria, Australia). Then, the importance of the search for novel training methods is compared with the importance of the use of a very simple ANN ensemble aggregation approach. It is shown that although some metaheuristics may slightly outperform the classical gradient-based Levenberg-Marquardt algorithm for a specific catchment, none performs better for the majority of the tested ones. One may also point out a few metaheuristics that do not suit ANN training at all. On the other hand, application of even the simplest ensemble aggregation approach clearly improves the results when the ensemble members are trained by any suitable algorithms.

EDITOR D. Koutsoyiannis; ASSOCIATE EDITOR E. Toth