A Multi‐Grid Method of High Accuracy Surface Modeling and Its Validation

A method of high accuracy surface modeling (HASM) has been constructed to find a solution for error problems that had long troubled surface modeling in geographical information systems (GIS). It is found that when a preconditioned conjugate gradient (PCG) algorithm is used to solve the large sparse linear system, which HASM can be transferred into, HASM performs best in terms of simulation compared with all other algorithms. But its computing speed is not fast enough for all applications. A multi‐grid method is introduced into HASM to try to shorten its computing time. Both numerical and real‐world tests demonstrate that there is a range of stop error (SE). The multi‐grid method of HASM (HASM‐MG) greatly increases computing speed when SEs are within this range, compared with the PCG algorithm of HASM (HASM‐PCG). HASM‐MG is suitable for applications with a need for less accuracy and a shorter computing time. HASM‐PCG is appropriate for issues needing higher accuracy. HASM‐MG performs better than HASM‐PCG in flat areas, while HASM‐PCG does better in complex terrainm in terms of accuracy and computing time.     

HASM解算的2维双连续投影方法

基于曲面论原理的高精度曲面建模方法（HASM）解决了长期困扰地理信息系统界的误差问题,但使用该方法需要求解大规模线性方程系统导致其计算效率较低并限制了其广泛应用。为提高HASM线性系统的解算速度和精度,本文基于二维双连续投影提出了一种解算HASM的新方法（HASM-DSPM）。为了提高收敛速度和解算精度,该方法采用选取优化的投影空间和两次校正策略。理论分析表明,使用该方法比使用高斯赛尔方法（Gauss-Seidel, GS）和改进的高斯赛德尔方法(Modified Gauss-Seidel, MGS)效率明显提高。分别用本文引入的方法、高斯赛德尔方法和效率较高的MGS方法分别进行了HASM方程组的解算,并对高斯合成曲面和实际项目区模拟进行了模拟试验。结果表明,和GS、MGS方法相比,本文提出的HASM-DSPM解算方法无论收敛速度和计算时间都有明显改善。     

An Adaptive Method of High Accuracy Surface Modeling and Its Application to Simulating Elevation Surfaces

An adaptive method is employed to speed up computation of high accuracy surface modeling (HASM), for which an error indicator and an error estimator are developed. Root mean‐square error (RMSE) is used as the error estimator that is formulated as a function of gully density and grid cell size. The error indicator is developed on the basis of error surfaces for different spatial resolutions, which are interpolated in terms of the absolute errors calculated at sampled points while paying attention to the landform characteristics. The error surfaces indicate the magnitude and distribution of errors in each step of adaptive refinement and make spatial changes to the errors in the simulation process visualized. The adaptive method of high accuracy surface modeling (HASM‐AM) is applied to simulating elevation surface of the Dong‐Zhi tableland with 27.24 million pixels at a spatial resolution of 10 m × 10 m. Test results show that HASM‐AM has greatly speeded up computation by avoiding unnecessary calculations and saving memory. In addition, HASM‐AM improves simulation accuracy.     

地球表层建模研究进展

本文在讨论地球表层概念和地球表层建模定义的基础上,总结分析了地球表面形态表达、地球气候系统模拟、生态系统空间模拟分析和地球表层模拟系统等主要研究进展。分析结果表明,误差问题、多尺度问题、地球表层模拟速度问题和三维实时可视化问题是地球表层建模面临的主要挑战。为了解决这些问题,必须发展和采用诸如以遥感数据为初始场、以地面实测数据为优化控制条件的高精度高速度曲面建模新方法和新思路。     

高精度曲面模型的解算

为了提高Gass-Seidel(GS)算法的收敛速度,提出了改进的GS算法(MGS),用于解算高精度曲面模型(HASM)(HASM-MGS)。以高斯合成曲面为研究对象,将HASM-MGS与HASM-GS和Matlab提供的函数进行对比,结果表明,达到相同的模拟中误差,HASM-MGS计算时间远小于HASM-GS和Matlab提供的函数;HASM-MGS计算时间与模拟区域的网格数呈非常好的线性关系,时间复杂度比传统的方法降低两个数量级。     

Method of reducing HASM boundary error

High accuracy surface modeling (HASM) constructed based on the fundamental theorem of surface is more accurate than the classical methods. However, because of boundary error, location error, etc, HASM has a big accuracy loss in real-world examples. In former researches we solved the location error with Taylor expansion. In order to reduce the HASM boundary error and improve its accuracy further, this paper presents a new method of Laplace interpolation to compute the region boundary value. Gaussian synthetic surface and the real world test region are employed to validate the efficiency of this method. Results show that the boundary value computed with Laplace interpolation is more accurate than the classical methods, which can be regarded as an alternative for boundary value computation.     

HASM-AD algorithm based on the sequential least squares

The HASM(high accuracy surface modeling) technique is based on the fundamental theory of surfaces,which has been proved to improve the interpolation accuracy in surface fitting.However,the integral iterative solution in previous studies resulted in high temporal complexity in computation and huge memory usage so that it became difficult to put the technique into application,especially for large-scale datasets.In the study,an innovative model(HASM-AD) is developed according to the sequential least squares on...     

逐次最小二乘在高精度曲面建模方法（HASM）中的应用

从高精度曲面建模方法（HASM）曲面方程出发,结合数据平差理论,建立了基于独立单元的计算模型,并且采用逐次最小二乘法对计算方程进行分组求解。对HASM-AD和几种经典方法的精度进行了交叉验证,结果表明,该算法模拟误差的均方根差远小于其他空间插值方法。     

一种由等高线构建DEM的新方法

基于HASM,提出了一种由等高线建立DEM的新方法HASM-OC。HASM—OC方法保证地形曲面的整体光滑性,保证DEM最大程度上忠实于原始等高线数据。实际等高线案例结果表明,HASM—OC方法与基于薄板样条原理的Hutchinson方法的DEM模拟结果及其回放的等高线差相比,前者比后者保留更多的地形特征信息,前者的回放等高线比后者回放等高线更忠实于原始数据。     

Geocomputation over the Emerging Heterogeneous Computing Infrastructure

Emerging computer architectures and systems that combine multi‐core CPUs and accelerator technologies, like many‐core Graphic Processing Units (GPUs) and Intel's Many Integrated Core (MIC) coprocessors, would provide substantial computing power for many time‐consuming spatial‐temporal computation and applications. Although a distributed computing environment is suitable for large‐scale geospatial computation, emerging advanced computing infrastructure remains unexplored in GIScience applications. This article introduces three categories of geospatial applications by effectively exploiting clusters of CPUs, GPUs and MICs for comparative analysis. Within these three benchmark tests, the GPU clusters exemplify advantages in the use case of embarrassingly parallelism. For spatial computation that has light communication between the computing nodes, GPU clusters present a similar performance to that of the MIC clusters when large data is applied. For applications that have intensive data communication between the computing nodes, MIC clusters could display better performance than GPU clusters. This conclusion will be beneficial to the future endeavors of the GIScience community to deploy the emerging heterogeneous computing infrastructure efficiently to achieve high or better performance spatial computation over big data.     

基于GPU并行计算的巨幅遥感影像坐标转换研究与实现

随着航空航天遥感技术的不断发展,以遥感影像为代表的栅格数据分辨率越来越高,遥感影像处理呈现出数据量大、复杂度高的特点。近年来,通用GPU的运算性能不断提高为加速密集运算提供了新的途径,目前,采用GPU并行技术进行遥感影像处理成为新的研究热点。本文提出了基于GPU并行计算的巨幅遥感影像坐标转换方法,实践证明,相比于传统的转换方法基于GPU的算法有较为明显的提速。     

GPU并行区域网平差

为了进一步解决大数据量带来的平差效率低下的问题,引入GPU并行计算技术,同时使用预条件共轭梯度法以及不精确牛顿解法求解区域网平差过程中的法方程,构建了适用于GPU并行计算的全新的区域网平差技术流程。本文方法避免了存储法方程系数矩阵,而是在需要的时候实时的计算该矩阵,使得本文算法相较于传统的算法所需的计算机内存空间大幅减少(仅需要存储平差原始数据即可),平差计算速度明显提升,同时计算精度与传统方法相当。初步试验证明,本文的方法在普通电脑上仅需要约1.5min即可完成对4500张影像、近900万像点数据的平差计算,且计算精度达到子像素级。     

集群技术下的实景三维建模

计算机集群技术的出现为倾斜摄影测量中实景三维自动化建模时遇到的计算资源匮乏问题提供了解决途径。本文论述了集群在实景三维建模中的两种应用模式：基于物理并行计算的集群和基于逻辑并行计算的集群;阐述了在逻辑并行计算的集群模式下ContextCapture软件进行实景三维建模的流程;提出了该集群模式下子测区划分的5个原则：起降点原则、飞行高度原则、建筑物密度/植被密度原则、最大计算量原则、航线角度原则,并有效控制了划子区后的精度问题。最后,本文利用实验室多台普通PC机实现了某测区实景三维集群建模,结果表明：该方法自动化建立模型精度满足测区需求,可有效加快自动化建模速度,降低生产成本。     

高速铁路轨道中长波平顺性参数计算方法

高速铁路轨道中长波平顺性参数的计算方法是高速铁路轨道精测与精调的核心技术,现有的中长波平顺性参数计算多采用以轨道点法向偏移量代替矢距差(即设计矢距与实测矢距之差)的近似计算方法。由于该算法忽略实测弦线端点偏差的影响,导致轨道实测轨向(高低)计算结果与严密算法计算结果存在一定的偏差。鉴于现有近似算法存在准确度低的缺点,文中在现有近似算法的基础上提出一种改进算法,并通过对轨道实测坐标数据的计算和对比,验证改进算法的可行性和准确性,并提高高铁轨道中长波平顺性参数计算结果的准确度,可为高速铁路轨道精测与精调提供参考。     

基于GPGPU的全波形并行分解算法

针对EM(Expectation Maximization)波形分解算法具有多次迭代和大量乘、除、累加等高密集运算的特点,提出一套将EM算法在通用计算图形处理器GPGPU上并行化的方案。针对通用并行计算架构CUDA的存储层次特点,设计总体的并行方案,充分挖掘共享存储器、纹理存储器的高速访存的潜能;根据波形采样值采用字节存储的特征,利用波形采样值的直方图求取中位数,从而降低求噪音阈值的计算复杂度;最后,采用求和规约的并行策略提高EM算法迭代过程中大量累加的计算效率。实验结果表明,当设置合理的并行参数、EM迭代次数大于16次、数据量大于64 M时,与单核CPU处理相比,GPU的加速比达到了8,能够显著地提高全波形分解的效率。     

一种利用GPU加速的轨迹线热力图生成显示方法

为解决大数据量带来的热力图生成效率低的问题,引入基于图形处理器(graphic processing unit,GPU)的并行计算方法,并结合轨迹线模型,提出了一种利用GPU加速的轨迹线热力图生成显示方法。首先,针对轨迹点分布不均、邻域半径设置不合理等条件下产生的热力值不连续、不均等问题,采用轨迹线模型提升了热力图的效果。其次,针对大规模数据计算产生的热力图生成效率低的问题,通过GPU并行计算并配合内核函数参数调优、循环展开、像素缓冲对象显示等策略大幅提升算法计算效率。实验结果表明,所提方法较传统的基于中央处理器（central processing unit, CPU）的方法计算效率提升了5~30倍,且随着图像分辨率和轨迹数据的增加,算法加速比有逐步上升的趋势。     

基于观测量校正的星载GNSS-R海面风速快速反演方法

快速反演(FDI)算法是一种典型的星载全球卫星导航反射技术(GNSS-R)海面风速反演方法,具有计算复杂度低、快速实时反演的特点,但是FDI算法中的反演观测量提取精度不高,导致风速反演精度低. 针对于此,提出基于观测量校正的改进FDI算法,用于实现海面风速的快速高精度反演. 该方法首先利用辅助测量信息对观测量进行校正以降低干扰因素的影响,然后基于统计分析方法对ASCAT卫星风速数据进行海面风速值提取,最后建立了海面风速与校正观测量的地球物理模型函数(GMF)关系式,实现对海面风速的反演. 与传统FDI算法相比,该方法的反演偏差值更小,均方根误差(RMSE)降低了29%.      

基于高精度曲面模型的DEM构建与误差分析

引入地形表达误差(terrain representation error,Etr),选择标准曲面和甘肃省董志塬地区作为研究对象,利用窗口分析法实现Etr的提取;用统计分析法得出Etr随网格分辨率变化的回归方程;根据误差传播定律计算DEM中误差。数值结果表明,该方法能更准确的计算HASM生成的DEM精度;相同的采样数下,HASM较传统方法(IDW,Spline和Kriging)能生成更高精度和分辨率的DEM。在难以获取已知数据的地区,HASM提供了生成相对准确DEM的高效工具。     

Parallel computing solutions for Markov chain spatial sequential simulation of categorical fields

The Markov chain random field (MCRF) model is a spatial statistical approach for modeling categorical spatial variables in multiple dimensions. However, this approach tends to be computationally costly when dealing with large data sets because of its sequential simulation processes. Therefore, improving its computational efficiency is necessary in order to run this model on larger sizes of spatial data. In this study, we suggested four parallel computing solutions by using both central processing unit (CPU) and graphics processing unit (GPU) for executing the sequential simulation algorithm of the MCRF model, and compared them with the nonparallel computing solution on computation time spent for a land cover post-classification. The four parallel computing solutions are: (1) multicore processor parallel computing (MP), (2) parallel computing by GPU-accelerated nearest neighbor searching (GNNS), (3) MP with GPU-accelerated nearest neighbor searching (MP-GNNS), and (4) parallel computing by GPU-accelerated approximation and GPU-accelerated nearest neighbor searching (GA-GNNS). Experimental results indicated that all of the four parallel computing solutions are at least 1.8× faster than the nonparallel solution. Particularly, the GA-GNNS solution with 512 threads per block is around 83× faster than the nonparallel solution when conducting a land cover post-classification with a remotely sensed image of 1000?×?1000 pixels.     

GPS信号FFT捕获的GPU实现

缩短GPS接收机冷启动时间一直是GPS领域的热点问题,而决定冷启动时间的关键是捕获速度。针对快速付里叶变换（FFT）捕获算法的并行运算特点和图形处理单元（GPU）适合于进行并行的优势,简单介绍了FFT捕获算法原理和对比了GPU与FPGA的特点,重点设计了各通道和各频点均进行并行计算的FFT捕获算法的GPU实现方案。利用实测的GPS中频数据初步验证了本文捕获方案的正确性和运行时间。试验结果表明：与基于CPU的捕获方案相比,本文的捕获方案对卫星PRN和CA码相位的捕获结果完全正确,而捕获时间大幅度缩短了。