共查询到19条相似文献，搜索用时 125 毫秒
1.
根据图像和检测算子的特性,以相关性为准则,使用遗传算法对图像小波变换的尺度进行选择,从而构成一种自适应的高斯小波尺度空间.融合该空间下不同尺度检测的图像边缘,使得整幅图像的边缘细节丰富清晰,具有更好的抗噪性能.对测试图像使用Canny算法、单尺度、二进尺度和自适应尺度小波进行边缘检测,验证了该算法在去除噪声和准确定位方面的有效性. 相似文献
2.
基于SSA-MGF的BP神经网络多步预测模型 总被引：3，自引：2，他引：3
采用奇异谱分析（Singular Spectrum Analysis，SSA）方法对标准化样本序列进行准周期信号分量重建，将重建序列构造均值生成函数（Mean Generating Function．MGF）延拓矩阵作为输入因子，原样本序列作为输出因子，构建BP神经网络多步预测模型。通过实际建模并与逐步回归等方法进行对比预测试验，结果表明，基于SSA-MGF的BP神经网络多步预测模型预测效果优于其他3种模型，说明SSA的去噪及BP神经网络预报模型对于提高预测准确率是相对有效的，是一种具有较高应用价值的多步预测方法。 相似文献
3.
遗传神经网络释用气候模式预测产品的试验研究 总被引：1，自引：0，他引：1
从业务需求出发,提出了面向气候模式产品释用的神经网络。选用主分量作为网络的输入和输出,大大减少了其节点数,重点突出了大尺度影响变化关系,提高了实际预测的稳定性;用全局寻优的遗传算法取代经典BP算法,为高质量的网络学习训练提供了保证;针对实际设计代价函数,保证了网络学习训练能适应气候模式产品释用的基本要求,学习目的更明确,针对性更强。分别以夏季（6～8月）NCEP／NCAR500 hPa高度场、国家气候中心海气耦合模式500hPa高度预测场主分量为外界输入信号,同期中国降水场、华中区域降水场主分量为网络输出信号,进行了拟合预测和独立预测试验。结果表明：用模式500hPa高度预测场主分量为外界输入信号,网络输出（降水场主分量）反演的中国、华中地区降水场预测距平与实况同号率,有可能接近用NCEP／NCAR500 hPa高度场主分量为外界输入信号相当的技巧水平。 相似文献
4.
基于小波域高斯尺度混合模型的天气雷达图像高分辨率插值 总被引：2，自引：0，他引：2
采用可精确刻画雷达回波强度数据统计特征的小波域高斯尺度混合（GSM）模型作为雷达图像先验模型,进行天气雷达图像插值,在提高图像分辨率的同时有效重建降水回波中局部强回波值、小尺度变化细节等一些重要空间分布统计特征。分析和总结雷达回波强度数据小波频率域统计特点,建立小波域GSM模型;匹配天气雷达图像小波系数和GSM模型,利用贝叶斯理论估计更小尺度的小波系数,进行小波逆变换,完成高分辨率天气雷达图像插值。试验表明,该算法能从低分辨率图像中估计出高分辨率高频系数,且所利用的先验模型充分考虑降水数据本身的特点,可有效捕获降水回波结构的非高斯特征和局部相关特性,重建雷达图像中的局部变化细节。 相似文献
5.
针对多元伪线性滑动平均系统,讨论了多元增广随机梯度算法,为减小算法的计算量,将系统分解为一些子系统,给出了子系统增广随机梯度算法,利用耦合辨识概念和多新息辨识理论,推导了部分耦合（子系统）增广随机梯度算法、部分耦合（子系统）多新息增广随机梯度算法.进一步将提出的方法推广到多元伪线性自回归滑动平均系统,给出了部分耦合（子系统）广义增广随机梯度算法、部分耦合（子系统）多新息广义增广随机梯度算法.文中分析了多元增广随机梯度算法、部分耦合增广随机梯度算法、部分耦合多新息增广随机梯度算法的计算量. 相似文献
6.
针对静止轨道遥感卫星上多通道扫描型载荷成像、传输与存储过程中,存在数据质量下降等问题,本文在经典三维块匹配算法（Block Matching 3D,BM3D）基础上,提出一种基于多层级小波分解的并行执行策略。首先,使用小波变换对原始气象遥感图像分解,得到4个图像分量;其次,将所得图像分量进一步进行3级分解,并选择其中的10个图像分量;最后,每个分量并行执行BM3D滤波器去噪,并重构10个分量的输出图像。与传统BM3D去噪算法相比,改进BM3D算法的计算量可有效降低20%以上。通过与中值滤波、均值滤波、NL-Bayes、BM3D四种降噪算法进行实验对比,所提算法的峰值信噪比平均增益在0.39～4.45 dB之间,特别是在高斯白噪声和脉冲噪声的混合噪声去噪方面要显著优于选取的四种对比算法。 相似文献
7.
典型块结构非线性系统包括基本的输入非线性系统、输出非线性系统、输入输出非线性系统、反馈非线性系统等.输入非线性系统包括输入非线性方程误差类系统和输入非线性输出误差类系统.以输入非线性方程误差自回归系统,即输入非线性受控自回归自回归(IN-CARAR)系统为例,分别基于过参数化模型,基于关键项分离原理,基于数据滤波技术以及基于辨识模型分解技术,研究和提出了IN-CARAR系统的随机梯度辨识方法、多新息随机梯度辨识方法、递推最小二乘辨识方法、多新息最小二乘辨识方法.这些方法可以推广到其他输入非线性方程误差系统、输入非线性输出误差类系统、输出非线性方程误差类系统、输出非线性输出类系统、反馈非线性系统等.同时,给出了几个典型辨识算法的计算步骤、流程图和计算量. 相似文献
8.
标志点等距映射（L-ISOMAP）作为一种降维方法,在高光谱图像可视化中极具潜力.针对L-ISOMAP算法标志点代表性不足以及计算代价较高的问题,提出了基于K-medoids聚类算法的改进型L-ISOMAP算法（KL-ISOMAP）,进而形成可视化方法.该可视化方法由以下几部分组成：1）通过改进型K-medoids算法选择标志点;2）根据相似度剔除相似像元;3）实现剩余像元的非线性降维;4）完成降维结果的可视化.实验结果表明,KL-ISOMAP算法一方面有效地提高了标志点的内在结构代表性,进而取得了更好的可视化效果;另一方面可通过对相似阈值的设置,加快算法的运算速度.由此形成的可视化方法有着较好的视觉效果、距离保持特性以及像元可分性. 相似文献
9.
针对输入非线性方程误差系统,即输入非线性受控自回归系统,研究了基于过参数化模型的多新息辨识方法和基于过参数化模型的递阶多新息辨识方法;研究了基于关键项分离原理的多新息辨识方法;使用辨识模型分解技术,研究了基于关键项分离原理的两阶段多新息辨识方法和三阶段多新息辨识方法.这些方法可以推广到其他输入非线性方程误差系统、输入非线性输出误差类系统、输出非线性方程误差类系统、输出非线性输出类系统、反馈非线性系统等.同时,给出了几个典型辨识算法的计算量、计算步骤和流程图. 相似文献
10.
高分辨率的天气雷达数据能揭示探测天气目标的精细结构,对灾害性天气分析和预报预警至关重要。提高天气雷达反射率数据分辨率可以提升现有业务天气雷达对中小尺度强对流灾害性天气的监测和预警能力。本文在不改变雷达硬件的情况下,提出了基于注意力反向投影网络(Attention Back-Projection Network,ABPN)的天气雷达回波超分辨率重建算法用于提高雷达反射率数据分辨率。ABPN通过在深度反向投影网络(Deep Back-Projection Network,DBPN)中加入长短跳跃连接和通道注意力机制,对关键区域精细化重建结构特征。通过对实际天气过程超分辨率重建测试,结果表明,ABPN算法在雷达回波重建质量和主观视觉评估上有明显的优势,特别是在回波细节和天气雷达的边缘结构特征方面。 相似文献
11.
A deep learning objective forecasting solution for severe convective weather(SCW) including short-duration heavy rain(HR), hail, convective gusts(CG), and thunderstorms based on numerical weather prediction(NWP) data was developed. We first established the training datasets as follows. Five years of severe weather observations were utilized to label the NCEP final(FNL) analysis data. A large number of labeled samples for each type of weather were then selected for model training. The local temperature, pressure, humidity, and winds from 1000 to 200 h Pa, as well as dozens of convective physical parameters, were taken as predictors in our model. A six-layer convolutional neural network(CNN) model was then built and trained to obtain optimal model weights. After that, the trained model was used to predict SCW based on the Global Forecast System(GFS) forecast data as input. The performances of the CNN model and other traditional methods were compared. The results show that the deep learning algorithm had a higher classification accuracy on HR and hail than support vector machine, random forests, and other traditional machine learning algorithms. The objective forecasts by use of the deep learning algorithm also showed better forecasting skills than the subjective forecasts by the forecasters. The threat scores(TSs) of thunderstorm, HR, hail, and CG were increased by 16.1%, 33.2%, 178%, and 55.7%, respectively. The deep learning forecast model is currently used in the National Meteorological Center of China to provide guidance for the operational SCW forecasting over China. 相似文献
12.
在去马赛克问题中,为了精确插值倾斜边缘并提高结果图像的整体质量,提出一种基于残余插值的卷积神经网络去马赛克算法.针对Bayer格式的颜色滤波阵列,插值绿色平面时,对于红蓝通道信息不全的问题,采用同通道邻近像素值近似代替,综合考虑3个通道的梯度,运用倾斜方向的边缘检测算子,将倾斜边缘分为不同方向的边缘分别插值.在插值完成后,利用深度卷积神经网络,进一步训练插值结果.在标准的IMAX数据集上,与目前流行的算法相比,本文算法视觉上更接近原图,具有更高的峰值信噪比和更短的运行时间. 相似文献
13.
Comparative analysis of regression and artificial neural network models for wind speed prediction 总被引：1，自引：1，他引：0
In this study, wind speed was modeled by linear regression (LR), nonlinear regression (NLR) and artificial neural network (ANN) methods. A three-layer feedforward artificial neural network structure was constructed and a backpropagation algorithm was used for the training of ANNs. To get a successful simulation, firstly, the correlation coefficients between all of the meteorological variables (wind speed, ambient temperature, atmospheric pressure, relative humidity and rainfall) were calculated taking two variables in turn for each calculation. All independent variables were added to the simple regression model. Then, the method of stepwise multiple regression was applied for the selection of the “best” regression equation (model). Thus, the best independent variables were selected for the LR and NLR models and also used in the input layer of the ANN. The results obtained by all methods were compared to each other. Finally, the ANN method was found to provide better performance than the LR and NLR methods. 相似文献
14.
Convolutional neural networks(CNNs) have been widely studied and found to obtain favorable results in statistical downscaling to derive high-resolution climate variables from large-scale coarse general circulation models(GCMs).However, there is a lack of research exploring the predictor selection for CNN modeling. This paper presents an effective and efficient greedy elimination algorithm to address this problem. The algorithm has three main steps: predictor importance attribution, predictor rem... 相似文献
15.
A dual-resolution(DR) version of a regional ensemble Kalman filter(EnKF)-3D ensemble variational(3DEnVar) coupled hybrid data assimilation system is implemented as a prototype for the operational Rapid Refresh forecasting system. The DR 3DEnVar system combines a high-resolution(HR) deterministic background forecast with lower-resolution(LR) EnKF ensemble perturbations used for flow-dependent background error covariance to produce a HR analysis. The computational cost is substantially reduced by running the ensemble forecasts and EnKF analyses at LR. The DR 3DEnVar system is tested with 3-h cycles over a 9-day period using a 40/13-km grid spacing combination. The HR forecasts from the DR hybrid analyses are compared with forecasts launched from HR Gridpoint Statistical Interpolation(GSI) 3D variational(3DVar)analyses, and single LR hybrid analyses interpolated to the HR grid. With the DR 3DEnVar system, a 90% weight for the ensemble covariance yields the lowest forecast errors and the DR hybrid system clearly outperforms the HR GSI 3DVar.Humidity and wind forecasts are also better than those launched from interpolated LR hybrid analyses, but the temperature forecasts are slightly worse. The humidity forecasts are improved most. For precipitation forecasts, the DR 3DEnVar always outperforms HR GSI 3DVar. It also outperforms the LR 3DEnVar, except for the initial forecast period and lower thresholds. 相似文献
16.
应用2017—2018年5—9月福建省观测资料对华南区域中尺度模式(GTRAMS-3 km-RUC)预报进行站点检验,建立和训练基于卷积神经网络的逐时降水分级订正模型,并与频率匹配法进行2017—2018年测试集的对比试验和2019年数据集的模拟业务检验,探讨了试验过程中遇到的样本不均衡、特征变量选取以及模型过拟合问题。结果表明:模式对于15 mm·h-1以上降水的预报能力弱,各订正方法对原始预报均有不同程度的改进作用。从评估指标来看,基于卷积神经网络的订正方法比频率匹配法表现出优势,其中相关系数判别方案下的网络模型对强降水预报的订正效果显著优于其他方法;在输入特征变量选取方面,应用主成分分析方案的模型训练收敛速度比相关系数判别方案更快,最佳训练期有所提前,但也更早进入严重的过拟合状态,而相关系数判别方案能够使网络模型的训练拥有更长的提升期以达到更具“潜力”的状态;基于卷积神经网络的订正方法对减少分类降水预报的漏报率、晴雨和弱降水预报的空报率具有显著作用,其优化程度明显超过频率匹配法。 相似文献
17.
随着深度学习方法在计算机视觉领域的崛起,如何将其应用于具有全天时、全天候等优点的SAR图像也成为一大研究重点.相较于传统图像,SAR图像由于其难判读、应用人群较少等原因难以获得大量标注数据.本文提出一种基于场景合成和锚点约束的SAR图标检测方法.通过区域生长算法和阈值法对SAR车辆目标及其阴影进行分割,然后随机嵌入SAR复杂场景中的合理区域来合成目标检测数据集.针对SAR车辆目标的几何特性、图像分辨率参数,对Faster-RCNN中的锚点大小进行约束,减少不符合SAR车辆目标检测框尺寸的候选框,大量约简冗余计算,提升训练、测试效率及精度. 相似文献
18.
19.
Prediction of soil temperature using regression and artificial neural network models 总被引：3，自引：0，他引：3
Mehmet Bilgili 《Meteorology and Atmospheric Physics》2010,110(1-2):59-70
In this study, monthly soil temperature was modeled by linear regression (LR), nonlinear regression (NLR) and artificial neural network (ANN) methods. The soil temperature and other meteorological parameters, which have been taken from Adana meteorological station, were observed between the years of 2000 and 2007 by the Turkish State Meteorological Service (TSMS). The soil temperatures were measured at depths of 5, 10, 20, 50 and 100 cm below the ground level. A three-layer feed-forward ANN structure was constructed and a back-propagation algorithm was used for the training of ANNs. In order to get a successful simulation, the correlation coefficients between all of the meteorological variables (soil temperature, atmospheric temperature, atmospheric pressure, relative humidity, wind speed, rainfall, global solar radiation and sunshine duration) were calculated taking them two by two. First, all independent variables were split into two time periods such as cold and warm seasons. They were added to the enter regression model. Then, the method of stepwise multiple regression was applied for the selection of the “best” regression equation (model). Thus, the best independent variables were selected for the LR and NLR models and they were also used in the input layer of the ANN method. Results of these methods were compared to each other. Finally, the ANN method was found to provide better performance than the LR and NLR methods. 相似文献