Tung-Ju Hsieh

A GPU-Based Simulation of Tsunami Propagation and Inundation

Tsunami simulation consists of fluid dynamics, numerical computations, and visualization techniques. Nonlinear shallow water equations are often used to model the tsunami propagation. By adding the friction slope to the conservation of momentum, it also can model the tsunami inundation. To solve these equations, we use the second order finite difference MacCormack method. Since it is a finite difference method, it brings the possibility to be parallelized. We use the parallelism provided by GPU to speed up the computations. By loading data as textures in GPU memory, the computation processes can be written as shader programs and the operations will be done by GPU in parallel. The results show that with the help of GPU, the simulation can get a significant improvement in the execution time for each of the computation steps.

GPU Acceleration of Tsunami Propagation Model

Tsunami propagation in shallow water zone is often modeled by the shallow water equations (also called Saint-Venant equations) that are derived from conservation of mass and conservation of momentum equations. Adding friction slope to the conservation of momentum equations enables the system to simulate the propagation over the coastal area. This means the system is also able to estimate inundation zone caused by the tsunami. Applying Neumann boundary condition and Hansen numerical filter bring more interesting complexities into the system. We solve the system using the two-step finite-difference MacCormack scheme which is potentially parallelizable. In this paper, we discuss the parallel implementation of the MacCormack scheme for the shallow water equations in modern graphics processing unit (GPU) architecture using NVIDIA CUDA technology. On a single Fermi-generation NVIDIA GPU C2050, we achieved 223x speedup with the result output at each time step over the original C code compiled with -O3 optimization flag. If the experiment only outputs the final time step result to the host, our CUDA implementation achieved around 818x speedup over its single-threaded CPU counterpart.

Vision-Based Finger Detection, Tracking, and Event Identification Techniques for Multi-Touch Sensing and Display Systems

This study presents efficient vision-based finger detection, tracking, and event identification techniques and a low-cost hardware framework for multi-touch sensing and display applications. The proposed approach uses a fast bright-blob segmentation process based on automatic multilevel histogram thresholding to extract the pixels of touch blobs obtained from scattered infrared lights captured by a video camera. The advantage of this automatic multilevel thresholding approach is its robustness and adaptability when dealing with various ambient lighting conditions and spurious infrared noises. To extract the connected components of these touch blobs, a connected-component analysis procedure is applied to the bright pixels acquired by the previous stage. After extracting the touch blobs from each of the captured image frames, a blob tracking and event recognition process analyzes the spatial and temporal information of these touch blobs from consecutive frames to determine the possible touch events and actions performed by users. This process also refines the detection results and corrects for errors and occlusions caused by noise and errors during the blob extraction process. The proposed blob tracking and touch event recognition process includes two phases. First, the phase of blob tracking associates the motion correspondence of blobs in succeeding frames by analyzing their spatial and temporal features. The touch event recognition process can identify meaningful touch events based on the motion information of touch blobs, such as finger moving, rotating, pressing, hovering, and clicking actions. Experimental results demonstrate that the proposed vision-based finger detection, tracking, and event identification system is feasible and effective for multi-touch sensing applications in various operational environments and conditions.

Hyperspectral band selection based on parallel particle swarm optimization and impurity function band prioritization schemes

Abstract In recent years, satellite imaging technologies have resulted in an increased number of bands acquired by hyperspectral sensors, greatly advancing the field of remote sensing. Accordingly, owing to the increasing number of bands, band selection in hyperspectral imagery for dimension reduction is important. This paper presents a framework for band selection in hyperspectral imagery that uses two techniques, referred to as particle swarm optimization (PSO) band selection and the impurity function band prioritization (IFBP) method. With the PSO band selection algorithm, highly correlated bands of hyperspectral imagery can first be grouped into modules to coarsely reduce high-dimensional datasets. Then, these highly correlated band modules are analyzed with the IFBP method to finely select the most important feature bands from the hyperspectral imagery dataset. However, PSO band selection is a time-consuming procedure when the number of hyperspectral bands is very large. Hence, this paper proposes a parallel computing version of PSO, namely parallel PSO (PPSO), using a modern graphics processing unit (GPU) architecture with NVIDIA’s compute unified device architecture technology to improve the computational speed of PSO processes. The natural parallelism of the proposed PPSO lies in the fact that each particle can be regarded as an independent agent. Parallel computation benefits the algorithm by providing each agent with a parallel processor. The intrinsic parallel characteristics embedded in PPSO are, therefore, suitable for parallel computation. The effectiveness of the proposed PPSO is evaluated through the use of airborne visible/infrared imaging spectrometer hyperspectral images. The performance of PPSO is validated using the supervised K-nearest neighbor classifier. The experimental results demonstrate that the proposed PPSO/IFBP band selection method can not only improve computational speed, but also offer a satisfactory classification performance.

Band selection for hyperspectral images based on impurity function

Band selection for hyperspectral images is an effective technique to mitigate the curse of dimensionality. A variety of band selection methods have been suggested in the past. This paper presents a novel band prioritization based on impurity function (IF) for the band selection of hyperspectral images. The proposed IF band selection (IFBS) is incorporated with particle swarm optimization (PSO) band selection which has been developed to effectively group highly correlated bands of hyperspectral images into high corrected modules. It uses a particle swarm optimization scheme, which is a well-known method to solve the optimization problems, to develop an effective feature extraction algorithm for hyperspectral imagery. After PSO method is applied to the band reduction of hyperspectral images, the proposed IFBS is applied to enhance the efficiency of band selection. The propose method is evaluated by MODIS/ASTER airborne simulator (MASTER) for land cover classification during the Pacrim II campaign. The performance of IFBS is validated by the supervised k-nearest neighbor (KNN) classifier. Experimental results demonstrate that the proposed IFBS approach is an effective method for dimensionality reduction and feature extraction. Compared to other band selection methods, IFBS can effectively select the most significant bands for the image classification of hyperspectral images.

GPU Parallel Computing of Spherical Panorama Video Stitching

This paper presents a GPU-based spherical coordinate conversion system for panorama video image stitching. Modern programmable GPU makes it possible to process multiple images in an interactive frame rates. To perform image stitching to form a panorama view, we use OpenCL to stitch multiple images and then texture map it to a spherical object. This allows us to compose an immersive environment. In the case study presented in this paper, we achieve a speedup factor of 76x.

Visual analytics of terrestrial lidar data for cliff erosion assessmenton large displays

Heavy development on cliffs place a heavy emphasis on maintaining a healthy natural environment. The ability to explore, conceptualize and correlate spatial and temporal changes of topographical records, is required for the development of new analytical models that capture the mechanisms contributing towards cliff erosion. This paper presents a visualization based approach using large displays in a digital immersive environment. Visual analytics are performed for cliff erosion assessment from a terrestrial LIDAR (LIght Detection And Ranging) data, including visualization techniques for the delineation, segmentation, and classification of features, change detection and annotation. Research findings are described in the context of a cliff failure observed in Solana Beach in California. The visualization system presented in this paper demonstrates the insights that can be gained by observing the temporal change of a failure mass using frequent site monitoring.

Parallel positive Boolean function approach to classification of remote sensing images

We present a parallel image classification approach, referred to as the parallel positive Boolean function (PPBF), to multisource remote sensing images. PPBF is originally from the positive Boolean function (PBF) classifier scheme. The PBF multiclassifier is developed from a stack filter to classify specific classes of land covers. In order to enhance the efficiency of PBF, we propose PPBF to reduce the execution time using parallel computing techniques. PPBF fully utilizes the significant parallelism embedded in PBF to create a set of PBF stack filters on each parallel node based on different classes of land uses. It is implemented by combining the message-passing interface library and the open multiprocessing (OpenMP) application programing interface in a hybrid mode. The experimental results demonstrate that PPBF significantly reduces the computational loads of PBF classification.

A Parallel Simulated Annealing Approach for Floorplanning in VLSI

One of the critical issues in floorplanning is to minimize area and/or wire length of a given design with millions of transistors while considering other factors which may influence the success of design flow or even manufacturing. To deal with the floorplan design with enormous amount of interconnections and design blocks, we adopt a parallel computing environment to increase the throughput of solution space searching. Also, we include the fractional factorial analysis to further reduce the time needed to search the acceptable solution. The experimental results indicate that our approach can obtain better space utility rate and it takes less time than the traditional method and parallel method do.

High-performance computing and visualization of earthquake simulations and ground-motion sensor network data

Comparing numerical simulation results with accelerograph readings is essential in earthquake investigations and discoveries. We provide a case study on the magnitude 7.6 Taiwan Chi-Chi earthquake in 1999. More than 400 seismic sensor stations recorded this event, and the readings from this event increased global strong-motion records fivefold so that the accuracy of the earthquake simulation was enhanced significantly. Direct volume rendering is used to depict the space-time relationships of numerical results and seismic readings. When earthquake simulation data are volume rendered, it reveals the sequence of seismic wave initiation, propagation, attenuation, and energy releasing events of fault ruptures so that the direction of seismic wave propagation can be observed. Both accelerograph readings and earthquake simulation data are used to generate a sequence of ground-motion maps. Stacking these maps up in sequence forms a volume data. Visual analysis of the time-varying component reveals hidden features for better comparison and evaluation. Earthquake scientists are able to obtain insights and evaluate their simulation criteria from volume rendering.