Application of a feature-based approach to debris flow detection by numerical simulation     

Chih-Wei Lin  Cheng-Wu Chen  Wen-Ko Hsu  Chia-Yen Chen  Chung-Hung Tsai  Yi-Ping Hung  Wei-Ling Chiang 《Natural Hazards》2013,67(2):783-796

A debris flow is a serious natural disaster which can occur anywhere whether in a valley or on a mountain slope, destroying everything it passes through. Debris flows can occur suddenly and cause residents in the path to suffer casualties and property loss. An early warning system is necessary to reduce the damage in order to protect human life and personal property. However, most debris flow detection systems, like wireless sensors, satellite images and radar, are not suitable for general public use. Vision surveillance systems are generally erected in Taiwan as public devices for security. Therefore, we propose a novel debris early warning system that uses a computer vision technique and build a simulation environment to prove the feasibility.  相似文献   

Identification of the controlling factors for hydrological responses by artificial neural networks     

Ruo-Nan Hao  Yue-Ping Xu  Yen-Ming Chiang 《水文研究》2021,35(11):e14420

Identifying the controlling factors for hydrological responses is of great importance for artificial neural network-based flood forecasting models, which are often hindered by the lack of physical mechanisms. To explore the first-order controlling factors of hydrograph patterns, a hybrid neural network was designed to analyse the impacts of potential driving variables with different temporal and spatial resolutions on hydrograph patterns. The Jinhua River Basin in Southeast China was used as an example in this study. Flood events with different hydrograph patterns and six external factors denoting potential controlling factors were individually classified into specific clusters using self-organizing maps (SOMs). Based on the back-propagation neural network (BPNN) and leave-one-out cross-validation methods, the controlling factors of different flood patterns were identified by comparing the performances of flood simulation models trained with datasets before and after the potential controlling factor classification. The results showed that (i) the classification of controlling factors indicating various runoff regimes significantly improved the performance of data-driven models in flood simulation in terms of correlation coefficient, Nash-Sutcliffe coefficient, and normalized root mean square error; (ii) the spatial distribution of antecedent soil moisture and vegetation conditions as well as the temporal distribution of rainfall dominated different hydrograph patterns; and (iii) the transition of dominant rainfall-runoff processes could be identified in an individual flood event using the hybrid SOM–BPNN model, indicating the varying influence of potential controlling factors on streamflow. Overall, the hybrid neural network models trained with datasets classified by controlling factors provide a general analytical framework to identify the governing dynamics for different flood patterns and improve the accuracy of flood simulations. Additionally, more attention should be devoted to improving the time to peak error of hydrological models, which cannot be settled by data-driven models trained with different data-splitting strategies.  相似文献   

Automatic alignment of contemporary vector data and georeferenced historical maps using reinforcement learning     

Weiwei Duan  Yao-Yi Chiang  Stefan Leyk  Johannes H. Uhl  Craig A. Knoblock 《International journal of geographical information science》2020,34(4):824-849

ABSTRACT

With large amounts of digital map archives becoming available, automatically extracting information from scanned historical maps is needed for many domains that require long-term historical geographic data. Convolutional Neural Networks (CNN) are powerful techniques that can be used for extracting locations of geographic features from scanned maps if sufficient representative training data are available. Existing spatial data can provide the approximate locations of corresponding geographic features in historical maps and thus be useful to annotate training data automatically. However, the feature representations, publication date, production scales, and spatial reference systems of contemporary vector data are typically very different from those of historical maps. Hence, such auxiliary data cannot be directly used for annotation of the precise locations of the features of interest in the scanned historical maps. This research introduces an automatic vector-to-raster alignment algorithm based on reinforcement learning to annotate precise locations of geographic features on scanned maps. This paper models the alignment problem using the reinforcement learning framework, which enables informed, efficient searches for matching features without pre-processing steps, such as extracting specific feature signatures (e.g. road intersections). The experimental results show that our algorithm can be applied to various features (roads, water lines, and railroads) and achieve high accuracy.  相似文献   

A LiDAR study of the effective size of cirrus ice crystals over Chung-Li,Taiwan     

Subrata Kumar Das  Jan-Bai Nee  Chih-Wei Chiang 《Journal of Atmospheric and Solar》2010,72(9-10):781-788

In this paper, we estimated the effective size of ice crystals in cirrus clouds using fall velocity derived from LiDAR (light detection and ranging) measurements at Chung-Li (24.58°N, 121.1°E), Taiwan. Nine shapes of the ice crystals, viz. hexagonal plates, hexagonal columns, rimed long columns, crystals with sector-like branches, broad-branched crystals, stellar crystal with broad arms, side planes, bullet rosettes and assemblages of planar poly-crystals of specific dimensions have been analyzed. The results show that the lidar derived most probable mean effective size of ice crystals is 340±180 μm with a dominant size range of 200–300 μm. The lidar derived mean effective size of cirrus crystals are parameterized in terms of cloud mid-height temperature as well as optical depth. The discussed method will be useful to study the most probable effective size distribution of ice crystals in cirrus cloud.  相似文献   

Heterotrophic bacterial and <Emphasis Type="Italic">Synechococcus</Emphasis> spp. Growth and mortality along the inshore-offshore in the East China Sea in summer     

An-Yi Tsai  Gwo-Ching Gong  Robert W. Sanders  Kuo-Ping Chiang  Chien-Fu Chao 《Journal of Oceanography》2012,68(1):151-162

This study used the dilution method to examine growth and grazing rates of heterotrophic bacteria and an autotrophic picoplankton, Synechococcus spp., from 1 to 11 July 2007 in the East China Sea. The main influence of oceanographic conditions in this aquatic system was the introduction of fresh, high-nutrient water from Changjiang River and the extremely nutrient-poor, high-salinity waters of Kuroshio Water. In these experiments, deviation from linearity in the relationship between dilution factor and net growth rate was significant in a large number of cases. Growth rates for heterotrophic bacteria ranged from 0.024 to 0.24, and for Synechococcus spp. from 0.03 to 0.21 h⁻¹. Grazing rates ranged from 0.02 to 0.19 and 0.01 to 0.13 h⁻¹, respectively. The spatial variations of Synechococcus spp. production to the primary production ratio (SP/PP) were low (<5%) in high Chl a environments and increased exponentially in low Chl a environments, indicating that Synechococcus spp. contributes to a large extent to the photosynthetic biomass in the open sea, especially in the more oligotrophic Kuroshio Water. Furthermore, the results of our dilution experiments suggest that nanoflagellates largely depend on heterotrophic bacteria as an important energy source. On average, heterotrophic bacteria contributes to 76 and 59% of carbon consumed by nanoflagellates within the plume (salinity <31) and outside of it (salinity >31).  相似文献   

Implications of coseismic groundwater level changes observed at multiple-well monitoring stations     

Yeeping Chia  Jessie J. Chiu  Yi-Hsuan Chiang  Tsai-Ping Lee  Yih-Min Wu  Ming-Jame Horng 《Geophysical Journal International》2008,172(1):293-301

  相似文献   

Preliminary results of the Japan-Taiwan joint survey on combining ocean radar data in the Kuroshio upstream region     

Kaoru Ichikawa  Wen Chang Yang  Akihiko Morimoto  Yutaka Yoshikawa  Shigeo Sugitani  Wen-Shan Chiang  Jian-Wu Lai  En Yu Liang  Cho-Teng Liu  Chang-Wei Lee  Kei Yufu  Moeto Kyushima  Satoshi Fujii  Tomoharu Senju  Yoshihiko Ide 《Ocean Science Journal》2013,48(1):141-148

  相似文献   

Return mapping of phases and the analysis of the gravitational clustering hierarchy     

Lung-Yih Chiang  Peter Coles  Pavel Naselsky 《Monthly notices of the Royal Astronomical Society》2002,337(2):488-494

We present spectroscopic and high-speed photometric data of the eclipsing polar V895 Cen. We find that the eclipsed component is consistent with it being the accretion regions on the white dwarf. This is in contrast to Stobie et al. who concluded that the eclipsed component was not the white dwarf. Further, we find no evidence for an accretion disc in our data. From our Doppler tomography results, we find that the white dwarf has   M ≳0.7 M_⊙  . Our indirect imaging of the accretion stream suggests that the stream is brightest close to the white dwarf. When we observed V895 Cen in its highest accretion state, emission was concentrated along field lines leading to the upper pole. There is no evidence for enhanced emission at the magnetic coupling region.  相似文献   

Multisensor integration using neuron computing for land-vehicle navigation     

Kai-Wei Chiang  Aboelmagd Noureldin  Naser El-Sheimy 《GPS Solutions》2003,6(4):209-218

Most of the present navigation sensor integration techniques are based on Kalman-filtering estimation procedures. Although Kalman filtering represents one of the best solutions for multisensor integration, it still has some drawbacks in terms of stability, computation load, immunity to noise effects and observability. Furthermore, Kalman filters perform adequately only under certain predefined dynamic models. Neuron computing, a technology of artificial neural network (ANN), is a powerful tool for solving nonlinear problems that involve mapping input data to output data without having any prior knowledge about the mathematical process involved. This article suggests a multisensor integration approach for fusing data from an inertial navigation system (INS) and differential global positioning system (DGPS) hardware utilizing multilayer feed-forward neural networks with a back propagation learning algorithm. In addition, it addresses the impact of neural network (NN) parameters and random noise on positioning accuracy. Electronic Publication  相似文献   

The Relationship Between Monitoring Well and Aquifer Solute Concentrations     

Chen Chiang  Gary Raven  Clint Dawson 《Ground water》1995,33(5):718-726

  相似文献