共查询到20条相似文献，搜索用时 109 毫秒
1.
作者提出了一种新的 BP神经网络模型 ,其隐层激活函数采用中心参数可调的 Gaussian函数 ,输出层采用斜度可调的 Sigmoid函数 ,从而神经元具有了更强的信息存储、处理能力。由于采用组合函数 ,将 Gaussian函数良好的局部性和 Sigmoid函数良好的全局性相结合 ,提高了神经网络的收敛速度。几个典型实验的结果表明 ,与传统 BP网络模型相比 ,新网络模型在学习能力和泛化推广能力方面都有明显提高 相似文献
2.
基于第三代海浪模型SWAN建立中国东海波浪数值模型,模拟1509号“灿鸿”台风,1909号“利奇马”台风和2106号“烟花”台风过境期间的波浪场,并将结果作为多层感知机,随机森林和支持向量回归预报台风浪模型的数据集,同时引入空间性特征对算法输入进行改进,并对比不同模型的训练速度。结果显示,加入空间性特征后,模型预报精度得到提升,其中多层感知机均方根误差降低40.4%,运行速度最快,随机森林均方根误差降低20.8%,运行速度最慢,支持向量回归均方根误差降低8.7%,运行速度介于前两者之间。表明改进后的模型有效提高了预报精度,可以用于针对台风浪的短期预报。 相似文献
3.
4.
基于多小波分析的多聚焦图像融合 总被引：5，自引：0，他引：5
多聚焦图像融合的关键问题是如何保持原始图像的细节。基于多小波分析具有多个分析基函数和产生更多分解子图像的特点,本文提出1种基于多小波分析的多聚焦图像融合方法。根据小波变换域分解信号的特点,分别对细节信号采用基于局部区域能量指导下的融合规则和对粗大信号采用局部均方误差加权均衡的融合规则,最后考虑了焦点区域的一致性调整。通过实验证明,在采用空间频率和清晰度作为客观评价条件之下,该方法比基于单小波分解、塔式分解和其它的融合方法都具有更好的融合效果。 相似文献
5.
本文选取1989-2021年我国台风风暴潮直接经济损失统计数据,依据线性趋势法和Mann-Kendall非参数检验法进行分析,结果表明,32年间我国风暴潮灾害经济损失呈现显著下降趋势,整体呈厚尾分布特征,采用对数化处理后呈显著的正态分布特征。采用Morlet小波变换对我国台风风暴潮直接经济损失的周期变化规律进行分析,t检验结果显示,全域存在准两次高频振荡,1～2年及7～8年的周期振荡,但随时间变化年际周期逐渐缩短为3～5年,说明风暴潮经济损失序列存在高频振荡和多周期嵌套的低频振荡规律。在此基础上,采用Daubechies小波分解分离高频信号和低频信号,均方根误差和信噪比精度分析结果表明,当小波基设置消失矩为7,分解层数为2时,我国台风风暴潮直接经济损失时间序列具有最优分解重构效果。对各分解层进行小波系数平稳性检验和白噪声检验,建立的小波分解-ARMA组合模型的模拟精度和预测精度均优于传统的自回归移动平均模型和Fourier级数拓展模型,证明了小波分解法在我国台风风暴潮经济损失快速评估中具有可靠性和优越性。 相似文献
6.
7.
海洋沉积物工程定名对于开展海洋工程建设具有重要作用,然而海底粉土和黏性土的定名受人为因素影响容易产生误差。使用人工神经网络的方法对黄河口埕岛海域284组细粒土数据进行了训练和学习,得到了只利用沉积物粒径质量分数进行定名的方法。结果表明,使用人工神经网络的方法能够有效地对沉积物进行工程定名。当网络含有5个输入层节点、9个隐藏层节点、3个输出层节点、训练函数为Scaled conjugate gradient时定名准确率最高,检验准确率高达97.7%。训练数据的数量是造成神经网络预测存在误差的重要因素,随着数据量的增加,网络的可靠性和通用程度将越来越高。 相似文献
8.
本文对用于模式分类、函数逼近、参数估计的多层感知器 (MLPs)给出 1个清晰的关于内部行为的解释。作者以单隐层的 MLP为例 ,论述了关于 MLP的内部行为的半线性分析理论。对受训的MLP,将隐层单元的输出分别定义为网络输出的正、负“内部分量”;定义内部分量的连接权重集为给定问题的“内部判别模式”;建立了 MLP和模糊集相结合的新模型 ;分析了 MLP的结构为 N- 2 - 1和N- H- 1 ,给出权重初始化的方法 ;提出了 1种从受训神经 -模糊模型 (NFMs)中提取知识的全新的具有实用价值的方法。 相似文献
9.
以欧洲中期天气预报中心ECMWF(European Centre for Medium Range Weather Forecasts)的ERA5风场数据为真实风速参考值,利用HY-2B卫星散射计L2A数据,使用反向传播神经网络方法对风速进行了反演,分别建立了中高风速、中低风速和全风速反演模型。与基于NSCAT-4地球物理模式函数得到的L2B风速相比,在训练集中,3种网络模型反演风速的均方误差(Mean Square Error,MSE)分别达到了0.18,0.14和0.32 m/s,平均绝对值误差(Mean Absolute Error,MAE)分别达到了0.27,0.24和0.34 m/s;在测试集中,3种网络模型反演风速的均方误差(MSE)分别达到了0.54,0.27和0.46 m/s,平均绝对值误差(MAE)分别达到了0.48,0.35和0.42 m/s。研究结果表明,中高和中低风速模型优于全风速模型,其中中低风速模型反演风速的MSE和MAE最低,中高风速模型反演风速的MSE和MAE下降幅度最大;3种模型都具有良好的泛化能力。 相似文献
10.
针对海洋磁力测量数据中存在的高斯噪声,采用基于交叉证认的小波阈值去噪方法进行噪声去除。首先对磁力数据进行小波分解,利用交叉证认方法自动识别小波分解的信号层与噪声层,然后对噪声层小波系数进行阈值处理,进一步提取噪声层中的有用信息,最后进行小波重构得到去噪信号。同时提出了一种改进阈值处理函数,能够改善传统软、硬阈值函数存在的缺陷,提高重构信号的精度。实验分析表明,当噪声水平小于1.5nT时,相比于传统软、硬阈值函数的小波阈值去噪方法,该方法可更好地去除海洋磁力数据中的高斯噪声,并可较好地保留数据中的有用细节信息,使数据质量得到提高。 相似文献
11.
小波神经网络研究进展及展望 总被引：4，自引：0，他引：4
关于小波分析与人工神经网络结合的研究,近些年来已成为信号处理学科的热点之一,已有大量的研究成果见诸各种学术刊物和会议论文。小波变换具有良好的时频局部性质,神经网络则具有自学习功能和良好的容错能力,小波神经网络（ＷＮＮ）由于较好地结合了两者的优点而具有强大的优势。作者较系统地综述了小波神经网络的研究进展,讨论了小波神经网络的主要模型和算法,并就其存在的一些问题,应用与发展趋势进行了探讨 相似文献
12.
With the launch of SARAL/AltiKa altimeter, efforts have been made to develop wind speed retrieval algorithms. Here we present two algorithms for estimating and validating wind speed from AltiKa. The first method is based on a theoretical Geophysical Model Function (GMF) using forward model simulations for Ka band specifications. The second is the model function developed using the matched database of input and output vectors of Normalized Radar Cross Section (NRCS) from AltiKa and wind speed measurements from concurrent Jason-2 altimeters. Since the NRCS depends on both the surface roughness due to surface wind speed and on mean square slope of the surfaces, the significant wave height is used along with wind speed for model development as an proxy variable. Both the theoretical and empirical GMFs are evaluated for retrieval of wind speed from AltiKa and validated with NDBC buoys data. The empirical model provide wind speed retrieval accuracy of 1.4 m/s. The accuracy of wind retrievals from theoretical model is also in the similar range (1.6 m/s), indicating the sound physical basis applicable for the future altimeters with various incidence angles. The retrieved wind speed is applied for various case studies, bringing out all the regional and global features quite well. 相似文献
13.
利用机器学习的方法,对14个周期HY-2A卫星高度计数据:风速、有效波高和海面高度差值进行训练,探究海况偏差和风速、有效波高之间的关系,创建海况偏差核函数非参数模型(NPSSB),并与参数模型中具有代表性的BM3、BM4模型进行对比。研究表明:(1)核函数NPSSB模型能够很好的反映SSB与U、SWH之间的关系,SSB与U呈二次函数关系,SSB与SWH呈反比例函数关系;(2)核函数NPSSB模型对SSB的模拟能力与训练数据集相关,数据量越多,模拟能力越好;(3)核函数NPSSB模型与BM3、BM4模型都存在0^-0.03 m的差值,随着风速和有效波高的增加,差值的绝对值越大。 相似文献
14.
Mapping wind with high-frequency(HF) radar is still a challenge. The existing second-order spectrum based wind speed extraction method has the problems of short detection distances and low angular resolution for broad-beam HF radar. To solve these problems, we turn to the first-order Bragg spectrum power and propose a space recursion method to map surface wind. One month of radar and buoy data are processed to build a wind spreading function model and a first-order spectrum power model describing the relationship between the maximum of first-order spectrum power and wind speed in different sea states. Based on the theoretical propagation attenuation model, the propagation attenuation is calculated approximately by the wind speed in the previous range cell to compensate for the first-order spectrum in the current range-azimuth cell. By using the compensated first-order spectrum, the final wind speed is extracted in each cell. The first-order spectrum and wind spreading function models are tested using one month of buoy data, which illustrates the applicability of the two models. The final wind vector map demonstrates the potential of the method. 相似文献
15.
Microwave remote sensing is one of the most useful methods for observing the ocean parameters. The Doppler frequency or interferometric phase of the radar echoes can be used for an ocean surface current speed retrieval,which is widely used in spaceborne and airborne radars. While the effect of the ocean currents and waves is interactional. It is impossible to retrieve the ocean surface current speed from Doppler frequency shift directly. In order to study the relationship between the ocean surface current speed and the Doppler frequency shift, a numerical ocean surface Doppler spectrum model is established and validated with a reference. The input parameters of ocean Doppler spectrum include an ocean wave elevation model, a directional distribution function, and wind speed and direction. The suitable ocean wave elevation spectrum and the directional distribution function are selected by comparing the ocean Doppler spectrum in C band with an empirical geophysical model function(CDOP). What is more, the error sensitivities of ocean surface current speed to the wind speed and direction are analyzed. All these simulations are in Ku band. The simulation results show that the ocean surface current speed error is sensitive to the wind speed and direction errors. With VV polarization, the ocean surface current speed error is about 0.15 m/s when the wind speed error is 2 m/s, and the ocean surface current speed error is smaller than 0.3 m/s when the wind direction error is within 20° in the cross wind direction. 相似文献
16.
Bang-Fuh CHEN 《中国海洋工程》2007,21(4):659-675
In multiresolution analysis(MRA)by wavelet function Daubechies(db),we decompose the signal to two parts,the low and high frequency content.The high-frequency content of the data is removed first and a new "de-noise" signal is reconstructed by using inverse wavelet transform.The wavelet spectrum and harmonic analysis were used to analyze the characteristics of tidal data before constructing the input and output structure of ANN model.That is,the concept of tidal constituent phase-lags was introduced and the new "de-noise" signal was used as the input data set of ANN and the forecasting accuracy of ANN model is significantly improved. 相似文献
17.
本文选取142幅RADARSAT-2全极化合成孔径雷达（SAR）影像，在没有入射角输入的情况下，首先利用C-2PO模型进行海面风速反演。随后，将同一时空下的ASCAT散射计风向作为初始风向，提取相应雷达入射角，利用地球物理模式函数（GMF） CMOD5.N对142幅SAR影像进行风速计算。反演结果与美国国家资料浮标中心海洋浮标风速数据对比，结果显示:CMOD5.N GMF和C-2PO模型均可反演出较高精确度的海面风速，其均方根误差分别为1.68 m/s和1.74 m/s。此外，研究发现，在低风速段，CMOD5.N GMF的风速反演精度要明显优于C-2PO模型。针对这一现象，本文以SAR系统成像机理为基础，以低风速SAR图像为具体案例，给出了3种造成这一现象的原因。 相似文献
18.
The geophysical model function (GMF) describes the relationship between a backscattering and a sea surface wind, and enables a wind vector retrieval from backscattering measurements. It is clear that t... 相似文献
19.
为了对海道测量中测线定位精度进行评估,以及为补偿测线系统误差的需要,提出了实测定位折线模型,同时在利用小波降噪算法求得拟合航迹线的基础上,提出了拟合定位测线的概念,针对测线的不同形态给出了拟合测线的另外两种模型,即拟合直线和拟合折线。仿真实验表明:提出的三种测线模型均满足1:2 000以下比例尺测图要求;在满足定位精度和测线正横水深梯度不超限的约束条件下,可以用拟合测线代替原始测线进行数据的后处理;三种测线模型对系统误差不存在"淹没效应";拟合折线可以方便地减弱测线内受方向、船速等系统误差的影响。 相似文献
20.
使用参数化初猜测谱算法提取VV极化Sentinel-1 SAR图像的海浪要素并与ECMWF再分析数据以及浮标数据进行验证对比 总被引：2，自引：1，他引：1
The purpose is to study the accuracy of ocean wave parameters retrieved from C-band VV-polarization Sentinel-1Synthetic Aperture Radar(SAR) images, including both significant wave height(SWH) and mean wave period(MWP), which are both calculated from a SAR-derived wave spectrum. The wind direction from in situ buoys is used and then the wind speed is retrieved by using a new C-band geophysical model function(GMF) model,denoted as C-SARMOD. Continuously, an algorithm parameterized first-guess spectra method(PFSM) is employed to retrieve the SWH and the MWP by using the SAR-derived wind speed. Forty–five VV-polarization Sentinel-1 SAR images are collected, which cover the in situ buoys around US coastal waters. A total of 52 subscenes are selected from those images. The retrieval results are compared with the measurements from in situ buoys. The comparison performs good for a wind retrieval, showing a 1.6 m/s standard deviation(STD) of the wind speed, while a 0.54 m STD of the SWH and a 2.14 s STD of the MWP are exhibited with an acceptable error.Additional 50 images taken in China's seas were also implemented by using the algorithm PFSM, showing a 0.67 m STD of the SWH and a 2.21 s STD of the MWP compared with European Centre for Medium-range Weather Forecasts(ECMWF) reanalysis grids wave data. The results indicate that the algorithm PFSM works for the wave retrieval from VV-polarization Sentinel-1 SAR image through SAR-derived wind speed by using the new GMF C-SARMOD. 相似文献