Use of Artificial Neural Networks for Selective Omission in Updating Road Networks

Abstract

An important problem faced by national mapping agencies is frequent map updates. An ideal solution is only updating the large-scale map with other smaller scale maps undergoing automatic updates. This process may involve a series of operators, among which selective omission has received much attention. This study focuses on selective omission in a road network, and the use of an artificial neural network (i.e. a back propagation neural network, BPNN). The use of another type of artificial neural network (i.e. a self-organizing map, SOM) is investigated as a comparison. The use of both neural networks for selective omission is tested on a real-life road network. The use of a BPNN for practical application road updating is also tested. The results of selective omission are evaluated by overall accuracy. It is found that (1) the use of a BPNN can adaptively determine which and how many roads are to be retained at a specific scale, with an overall accuracy above 80%; (2) it may be hard to determine which and how many roads should be retained at a specific scale using an SOM. Therefore, the BPNN is more effective for selective omission in road updating.     

利用时空神经网络模型的长江经济带气温反演

传统基于遥感的气温反演方法往往使用全局模型,从而忽略了气温分布及其时空影响异质性,特别是在较大区域尺度的研究中存在不足。针对长江经济带区域,引入时空地理加权神经网络模型,建立一种高精度的气温估计方法。通过在广义回归网络模型中建立局部模型来顾及时空异质性的影响,融合遥感数据、同化数据、站点数据,获取面域分布的近地表气温信息。采用基于站点的十折交叉验证方法对模型性能进行评估,结果表明,时空地理加权神经网络有效提高了气温估计的精度（均方根误差为1.899℃,平均绝对误差（mean absolute error,MAE）为1.310℃,相关系数为0.976）,与多元线性回归和传统的全局神经网络方法相比,MAE值分别降低了1.112℃和0.378℃。气温空间分布制图结果显示,该方法结果能很好地反映长江经济带气温空间上的差异和不同季节的特征信息,具有实际应用价值。     

基于神经网络和模糊评判的滑坡敏感性分析

滑坡的敏感性涉及到很多因素,如滑坡体的坡度、坡的朝向、坡度的类型、岩石特性、海拔高度、植被覆盖等特征.神经网络具有非线性映射能力,利用这些与滑坡发生紧密相关的因素作为网络的输入,构造一个具有反映滑坡敏感性的评价网络,输出端为敏感性分析的结果.本文针对某具体地区,提取相关因素,构造评价指标体系并量化,利用该地区样本集数据对滑坡敏感性评价神经网络进行训练,用训练后的网络对实例并结合模糊评判进行了相互验证,结果说明利用神经网络和模糊评判进行滑坡敏感性分析是可行的.     

利用GA-NN模型反演声速剖面的众源水深数据声速改正

针对当前众源水深数据后处理过程中缺少高精度的实测声速剖面,导致测深数据质量偏低的现状,提出了一种基于遗传算法优化反向传播神经网络（genetic algorithm-back propagation neural network,GA-NN）模型反演声速剖面的声速改正方法。首先,利用历史声速剖面群进行正交经验函数分析,提取特征向量与重构系数范围;然后,结合海区的历史声速场数据训练GA-NN模型;最后,将海表声速数据输入模型反演声速剖面,并分析不同方法下的声速剖面分别进行声速改正后的水深和位置误差。实验结果表明,在复杂的海底地形下,与现有方法相比,所提方法反演的声速剖面更适用于众源水深数据的声速改正,削弱了声速误差的影响,提高了众源水深数据的处理精度。     

利用神经网络方法提取水稻种植面积—以湖北省双季早稻为例

在利用NOAA数据提取水稻种植面积的过程中，由于其地面分辨率较低，存在大量混合像元问题，使得提取精度不够理想，该文基于神经网络方法即可以提供多源数据的输入，又不受数据分布假设限制的特点，从NOAA图像演算最能反应朋稻分布信息的绿度指数（NDVI）和日夜温差值，将其重采样，然后加入对水稻生产区域有重要影响的土壤类型，土地利用类型及高程分布等信息，以TM图像作为准直值进行分类，获得较为理想的湖北省双季早稻种植面积。     

回归分析与人工神经网络在建筑物沉降预计中的应用

随着高层建筑物的逐渐增多,对其做好沉降预警工作具有重要意义。本文中分别采用回归分析法和人工神经网络建立沉降预计模型,对工程实例进行沉降预计,通过图表表现出的预测值与真实值的差异,说明两种方法的优缺点以及两者的实际应用价值,最终两种方法均取得了令人满意的预计结果。     

Multi-source Classification Using Artificial Neural Network in a Rugged Terrain

Abstract

This study advocates the use of GIS and remote sensing technologies to establish urban evolution maps and assess the impact of urbanization on agricultural areas over the last three decades. The target area is the city of Béni‐Mellal, located in central Morocco. The methodology adopted makes use of panchromatic SPOT images to survey the urban areas during the 1980s and 1990s. Available topographic maps provided the information for the 1970s. Maps and statistics of land use and urban growth for Béni Mellal were established after manually classifying images on a per-polygon basis and digitizing topographic maps using GIS capabilities. The results show an increase in dense urban area by 980.7 ha from the 1970s to the 1990s. This increase occurred at the expense of forests (24.7 ha), plantations (752.3 ha), rangeland (113.4 ha), non‐irrigated land (69.7 ha), and irrigated land (20.6 ha). During this period, scattered urban areas, predominantly suburbs, increased by 755.9 ha to the detriment of forests (14.9 ha), plantations (109.8 ha), rangeland (138.9 ha), non‐irrigated land(400.5 ha), and irrigated land (91.9 ha). These cartographic and statistic results are efficient decision‐making tools for protecting agricultural land and planning urban and suburban areas.     

Artificial Neural Networks as a Method of Spatial Interpolation for Digital Elevation Models

This paper examines the performance of artificial neural networks (ANNs) as a method of spatial interpolation, when presented with irregular and regular samples of elevation data. The results of the ANN interpolation are compared with results obtained by kriging. Tests of spatial bias in the systematic errors contained in each of the neural network-derived DEMs were conducted using four attributes: slope, aspect, average direction and average distance from the nearest sampled value. Based on RMS and other evaluation measures, the accuracy of estimated DEMs from regular and irregular sample distributions using neural networks is lower than the accuracy level derived from kriging. The accuracy level of the ANN interpolators also decreases as the range of elevation values in DEMs increases. As reported in the literature, ANNs are approximate interpolators, and the pattern of under-prediction and over-prediction of elevation values in this study revealed that all estimated values fell within the range of sample elevations. Neural networks cannot predict values outside the range of elevation values contained in the sample, a property shared by other interpolators such as inverse weighted distance.     

用地球位模型和BP神经网络转换GPS高程

研究了转换GPS高程的地球位模型和BP神经网络的拟合方法.用已知GPS水准点的高程异常移去地球位模型高程异常,然后对剩余高程异常通过BP神经网络拟合和内插,在内插点上恢复地球位模型高程异常,从而得到该点的高程异常.通过实测GPS水准数据将该方法与基于地球位模型和二次曲面的拟合方法进行了比较.试验结果表明,该方法转换GPS高程的精度优于基于地球位模型和二次曲面的拟合方法,能够满足一定的工程应用需求.     

Multicomponent Image Segmentation Using a Genetic Algorithm and Artificial Neural Network

Image segmentation is an essential process for image analysis. Several methods were developed to segment multicomponent images, and the success of these methods depends on several factors including (1) the characteristics of the acquired image and (2) the percentage of imperfections in the process of image acquisition. The majority of these methods require a priori knowledge, which is difficult to obtain. Furthermore, they assume the existence of models that can estimate its parameters and fit to the given data. However, such a parametric approach is not robust, and its performance is severely affected by the correctness of the utilized parametric model. In this letter, a new multicomponent image segmentation method is developed using a nonparametric unsupervised artificial neural network called Kohonen's self-organizing map (SOM) and hybrid genetic algorithm (HGA). SOM is used to detect the main features that are present in the image; then, HGA is used to cluster the image into homogeneous regions without any a priori knowledge. Experiments that are performed on different satellite images confirm the efficiency and robustness of the SOM-HGA method compared to the Iterative Self-Organizing DATA analysis technique (ISODATA).     

基于神经网络的水稻双向反射模型研究

水稻的双向反射特性与其冠层结构，各组分光谱性质以及入射光方向和观测方向之间存在着密切的，非线性的相关关系。运用人工神经网络技术，采用水稻田间实测数据，对这种关系进行拟合，所建立的水稻双向反射BP前向和反演模型，都达到了较高的拟合精度。研究表明：采用人工神经网络技术计算水稻双向反射率和成批反演冠层结构参数是可行的。对所建模型做进一步的改进，可模拟水稻双向反射的实际过程，进而监测作物长势。     

Heterogeneous Space–Time Artificial Neural Networks for Space–Time Series Prediction

Space–time series prediction plays a key role in the domain of geographic data mining and knowledge discovery. In general, the existing methods of space–time series prediction can be divided into two main categories: statistical machine learning methods. Comparatively, machine leaning methods have obvious advantages with respect to handling nonlinear problems. However, space–time dependence and the heterogeneity of space–time data are not well addressed by the existing machine learning methods. Because of this limitation, an accurate prediction of a space–time series is still a challenging problem. Therefore, in this study, both space–time dependence and heterogeneity are incorporated into the feedback artificial neural network, and heterogeneous space–time artificial neural networks (HSTANNs) are developed for space–time series prediction. First, to handle spatial heterogeneity, space–time series clustering is used to divide the study area into a set of homogeneous sub‐areas. Then, a space–time autocorrelation analysis is employed to explore the space–time dependence structure of the dataset. Finally, a HSTANN is established for each sub‐area. Further, HSTANNs are applied to predict the concentrations of fine particulate matter (PM_2.5) in Beijing–Tianjin–Hebei. The experimental results show that when compared with other methods, the accuracy of the forecasting results is considerably improved by using HSTANNs.     

基于稀疏浮动车数据的城市路网交通流速度估计

浮动车数据在时空维度呈现较强的稀疏性,是其应用于城市路网交通流估计所面临的主要难题之一。本文通过分析路网交通流速度的时空特征,构建了一种基于朴素贝叶斯法的估计模型,实现对路网中未被样本覆盖路段交通流速度的估计。时间特征主要考虑目标路段相邻时段的交通流速度,空间特征根据路段间交通流相似关系进行分析,突破了传统基于欧氏空间或拓扑关系的度量方式。结果显示,模型能有效地估计出样本缺失路段的交通流速度,且在精度方面相对传统基于拓扑关系的算法优势显著,较好地解决了数据时空稀疏性问题,对基于浮动车数据的交通应用具有较强的实践意义。     

基于ImageNet预训练卷积神经网络的遥感图像检索

高分辨率遥感图像内容复杂,细节信息丰富,传统的浅层特征在描述这类图像上存在一定难度,容易导致检索中存在较大的语义鸿沟。本文将大规模数据集ImageNet上预训练的4种不同卷积神经网络用于遥感图像检索,首先分别提取4种网络中不同层次的输出值作为高层特征,再对高层特征进行高斯归一化,然后采用欧氏距离作为相似性度量进行检索。在UC-Merced和WHU-RS数据集上的一系列实验结果表明,4种卷积神经网络的高层特征中,以CNN-M特征的检索性能最好;与视觉词袋和全局形态纹理描述子这两种浅层特征相比,高层特征的检索平均准确率提高了15.7%~25.6%,平均归一化修改检索等级减少了17%~22.1%。因此将ImageNet上预训练的卷积神经网络用于遥感图像检索是一种有效的方法。     

多波束海底地形畸变校正与声速剖面反演

提出了一种多波束测量声速剖面反演方法。即采用EOF(empirical orthogonal function)算法,对测区实测声速剖面数据进行模态向量提取,以多波束测量声速改正不完善引起的地形畸变程度为依据构造适应度函数,通过遗传算法优化声速剖面的重构系数,实现声速剖面反演。实验结果表明,反演的声速剖面能有效改正声速误差引起的海底地形失真,显著提高了多波束水深测量数据精度和处理效率。     

大坝变形的神经网络方法研究

大坝的失事带来的不仅是经济损失也是安全隐患,因此,建立一种大坝变形长期预测模型对它的安全评价将具有重要意义。本文针对华东CC大坝5JHJl04监测点的垂直位移变形进行分析,在传统的回归分析模型和常规神经网络模型的基础上建立了将两种方法结合的融合模型,得到大坝变形分析的最优模型。其精度与一般方法相比有了进一步的提升,可以更好地进行大坝变形预测。     

基于自组织神经网络的声速剖面分类方法研究

利用自组织神经网络技术,结合声速剖面特点,研究了声速剖面的描述方法、网络中神经元个数的确定、获胜神经元的邻域及其邻域内神经元的拓扑关系等对网络结构和声速剖面类别划分的影响,给出了分类声速剖面的网络构造思想和神经网络结构。实验验证了该方法的正确性。     

Neural Networks in GPS Navigation

Neural networks have been proposed as nonlinear filters in a variety of applications that involve nonlinear processing of input signals; examples include blind signal separation, image registration, and blind deconvolution. The Global Positioning System (GPS) navigation equations are nonlinear (quadratic) in nature, and a direct closed form solution of the GPS navigation equations does not exist. This article presents a new approach to solving the GPS pseudorange equations using three-layer neural networks. A three-layer radial basis function (RBF) neural network is designed, which solves the non-linear GPS pseudorange equations directly as opposed to the linear least squares or extended Kalman filter approaches in traditional GPS receivers. For training the neural network, a carefully selected cost function is minimized using a variation of the classical conjugate gradient algorithm such that training time for the neural network is reasonable. Simulations have been performed at SiRF Technology Inc. that show stable behavior even under bad geometry conditions where the traditional recursive least squares and extended Kalman filter approaches show high sensitivity to measurement errors. Under good geometry conditions the neural network solution shows slightly improved noise performance compared to the expected performance of traditional leas squares solution. Simulations have been performed with additive white Gaussian noise and correlated noise models to evaluate the performance of the trained neural network. ? 2000 John Wiley & Sons, Inc.     

多尺度并联卷积神经网络遥感影像油罐目标检测算法

针对传统遥感影像油罐目标检测算法依赖油罐圆形特征,对于背景复杂和存在大量小目标的情况检测效果差的问题,提出一种多尺度并联卷积神经网络油罐目标检测算法.首先根据油罐目标尺寸对各神经网络检测效果的影响规律,采用不同网络架构分别检测不同尺寸的油罐目标;其次利用经过训练的分类网络对上述检测结果进行后处理,剔除可能存在的误检;最...     

常梯度声线跟踪中平均声速的改进算法

在对多波束测深常用声线跟踪算法进行分析的基础上,提出了一种改进的声线跟踪算法,即沿声线传播的圆弧路径进行积分来求取层内的平均声速,进而推导了严密的多波束测深波束脚印位置的计算公式。理论推导表明,常用的声线跟踪算法的平均声速是将声线的圆弧传播路径当作对应弧段弦时的改进算法平均声速的近似解。实验表明,改进的声线跟踪算法比常用的声线跟踪算法更严密。