基于Hadoop的气象数据分布式存储技术研究

针对目前气象数据存储所面临的海量扩张、高并发读写、结构化和非结构化数据并存以及长时间序列和大数据集检索效率低下等问题,提出了以Hadoop开源框架为基础的气象数据分布式存储方案。通过对气象数据自身属性和特点进行分析,得出了气象数据在经过充分优化的基础上,在分布式存储框架中具有很强的适应性和规模化应用的潜力;并在HBase数据库中的Row Key设计和小文件合并策略方面做了创新。最后针对气象数据中广泛存在的结构化和非结构化这两种主要数据类型,以自动气象站数据和雷达产品数据为具体实例,给出了详细的设计思路和实现方法。     

Translation speed measurements of hydrogen,helium, and argon field-reversed configurations in the central cell of a KMAX mirror device

A series of experimental results of field-reversed configurations (FRCs) on a KMAX (Keda Mirror with AXisymmetricity) tandem mirror machine are reported. Single-side FRC translation processes with three different gas species were measured by avalanche photodiodes. Consistent with the theoretical prediction, the measured FRC speeds were inversely proportional to the square root of the ion mass. However, the speeds of the hydrogen FRC increased even in a uniform magnetic field region while the speeds of the helium and argon FRCs decreased. Possible mechanisms are discussed. The speed of the second pass due to the reflection of the mirror fields was found to be ∼1/3 of the first pass speed. The internal magnetic fields were measured for a colliding-merging argon FRC, and the results show that, even for very slowmoving FRCs, merging can occur.     

结合位图切割和区域合并的彩色图像分割

就经典分水岭图像分割算法中存在的过分割问题，提出一种结合位图切割和区域合并的彩色图像分割算法。对原始彩色图像通过空域梯度算子求其梯度图像，并利用位图切割重建梯度图像；对新梯度图像进行分水岭预分割；对预分割图像基于异质性最小原则进行区域合并，并获得最终分割结果。相比于现有的同类方法，该算法引入位图切割，抑制噪声对分割结果的影响，在边缘模糊处分割准确，得到符合人类视觉的较小分割区域数目，同时在运行效率上提高。     

Automated coronal hole segmentation from Solar EUV Images using the watershed transform

Region of interest segmentation in solar images is the subject of frequent research in solar physics. This study outlines watershed by immersion segmentation to identify coronal hole areas in solar images acquired using the Extreme UV Imaging Telescope (EIT). Solutions presented here produce highly accurate segmentation results of coronal holes of irregular shape, and what is more, they do so for images representing varied solar activity, recorded in different years and months. In addition, the solutions presented here make all the methods used operate very quickly. These methods include: the preprocessing step before the watershed segmentation, the watershed segmentation itself, and also the postprocessing of solar images after the watershed segmentation. The mean duration of the entire segmentation process of solar images amounts to 342 ms for a single coronal hole, without the parallel implementation of the methods used. The experiments were carried out on a computer with an Intel Core i7 CPU @ 2 GHz and 4 GB RAM. After the seed point is identified inside the coronal hole, the segmentation runs automatically.     

基于多线结构光的弱纹理物体形貌测量方法

大多数现有的主流结构光测量装置在被测物体的表面上投射单条激光,通过移动装置实现物体表面的扫描重建。为了实现物体表面的在线快速重建,将多线结构光和双目立体视觉理论相结合,设计了一种新的三维测量方法。提出了一种结合形态滤波和Zhang-Suen细化算法的光条骨架提取方法和基于光条序列关系的光条匹配算法,并通过极线约束实现光条上特征点的精确匹配。通过实验验证了方法的有效性,相对误差在3%以内。     

基于普吕克直线的多线结构光视觉传感器标定方法

针对多线结构光系统的快速标定,提出了一种基于普吕克直线的标定方法。传统的线结构光标定方法大多以单点的形式进行提取和坐标系转换,从而获得平面方程。然而单点数量较多会导致操作复杂,不适用于多线结构光的标定。为了解决这个问题,提出了一种新的标定方法,该方法在标定过程中全部采用普吕克矩阵的形式,而不是直接使用点特征。该方法有利于在摄像机坐标系下快速准确地获得光平面方程。同时设计了一种与该方法相对应的用于标定多线结构光的平面标定靶标。为了将一般直线转换为普吕克直线,将图像平面维度延展,定义了一个新的图像空间。为了对普吕克直线的坐标系进行转换,重新表达了基于普吕克直线的透视投影模型。基于普吕克空间中直线与平面的性质,对多条结构光线的普吕克矩阵进行合并即可高效地构造出线性的矩阵方程,从而进一步拟合出结构光平面方程。实验验证了所提方法并证明了校准精度的显著提高,当测试距离为1.8 m时,测量得到的三维点的RMS误差在0.08 mm之内。     

多流多科目实时数据处理算法的设计与实现

随着飞行试验测试技术的快速发展,单架次需要处理的试飞科目越来越多,而多科目处理参数分布在不同的网络流中,这对机载实时数据处理的效率带来严峻的挑战。利用“合并-处理-分流”的方法,设计一种多流多科目网络数据实时处理算法。该算法将多科目处理参数按照参数名、采样率等信息进行整合形成合并参数列表；在实时处理时根据合并参数信息实时处理多网络流中的数据,并将处理结果存储在数据缓存中；最后将多科目数据进行分流即可得到处理结果。测试结果表明,该算法可实际应用于多流多科目数据处理中,大幅提高实时处理的效率,同时可以得到准确的数据处理结果。     

微覆盖解决高低层混合居民楼深度覆盖的应用

城区高低层混合居民楼区域作为无线网络覆盖难点,传统宏站无法进行有效的覆盖。本文通过采用“微覆盖、微对打”的建设方式有效解决“高低层混合居民楼”无线网络信号的深度覆盖问题。     

以“耦合、融合、结合”为特征的复合应用型人才培养模式研究——以工程造价专业为例

文章结合工程造价专业教学和应用实践,剖析了目前工程造价专业人才培养模式中存在的问题,借鉴国外工程造价专业人才培养成果,从耦合行业需求,整合相关知识、融入职业标准,工学结合、做学结合、课证结合三方面,对以"耦合、融合、结合"为特征的工程造价复合应用型人才培养模式进行系统研究,旨在为复合应用型工程造价专业人才的培养提供借鉴。     

Improved Region Merging Algorithm for Remote Sensing Images

To segment high-resolution remote sensing images (RSIs) accurately on an object level and meet the precise boundary dividing requirement, an improved superpixel segmentation and region merging algorithm is proposed. Simple linear iterative clustering (SLIC) is widely used because of its advantages in performance and effect; however, it causes over-segmentation, which is very disadvantageous to information extraction. In this proposed method, SLIC is firstly adopted for initial superpixel partition. The second stage follows the iterative merging procedure, which uses a hierarchical clustering algorithm and introduces a local binary pattern (LBP) texture feature operator during the process of merging. The experimental results indicate that the proposed method achieved a good segmentation and region merging performance, and worked effectively on cloud detection preprocessing in high-resolution RSIs with cloud and snow overlap situations.