John A. Taylor

An Efficient Near-Duplicate Video Shot Detection Method Using Shot-Based Interest Points

We propose a shot-based interest point selection approach for effective and efficient near-duplicate search over a large collection of video shots. The basic idea is to eliminate the local descriptors with lower frequencies among the selected video frames from a shot to ensure that the shot representation is compact and discriminative. Specifically, we propose an adaptive frame selection strategy called furthest point voronoi (FPV) to produce the shot frame set according to the shot content and frame distribution. We describe a novel strategy named reference extraction (RE) to extract the shot interest descriptors from a keyframe with the support of the selected frame set. We demonstrate the effectiveness and efficiency of the proposed approaches with extensive experiments.

Toolbox for advanced x-ray image processing

A software system has been developed for high-performance Computed Tomography (CT) reconstruction, simulation and other X-ray image processing tasks utilizing remote computer clusters optionally equipped with multiple Graphics Processing Units (GPUs). The system has a streamlined Graphical User Interface for interaction with the cluster. Apart from extensive functionality related to X-ray CT in plane-wave and cone-beam forms, the software includes multiple functions for X-ray phase retrieval and simulation of phase-contrast imaging (propagation-based, analyzer crystal based and Talbot interferometry). Other features include several methods for image deconvolution, simulation of various phase-contrast microscopy modes (Zernike, Schlieren, Nomarski, dark-field, interferometry, etc.) and a large number of conventional image processing operations (such as FFT, algebraic and geometrical transformations, pixel value manipulations, simulated image noise, various filters, etc.). The architectural design of the system is described, as well as the two-level parallelization of the most computationally-intensive modules utilizing both the multiple CPU cores and multiple GPUs available in a local PC or a remote computer cluster. Finally, some results about the current system performance are presented. This system can potentially serve as a basis for a flexible toolbox for X-ray image analysis and simulation, that can efficiently utilize modern multi-processor hardware for advanced scientific computations.

Multiscale cardiac modelling reveals the origins of notched T waves in long QT syndrome type 2

The heart rhythm disorder long QT syndrome (LQTS) can result in sudden death in the young or remain asymptomatic into adulthood. The features of the surface electrocardiogram (ECG), a measure of the electrical activity of the heart, can be equally variable in LQTS patients, posing well-described diagnostic dilemmas. Here we report a correlation between QT interval prolongation and T-wave notching in LQTS2 patients and use a novel computational framework to investigate how individual ionic currents, as well as cellular and tissue level factors, contribute to notched T waves. Furthermore, we show that variable expressivity of ECG features observed in LQTS2 patients can be explained by as little as 20% variation in the levels of ionic conductances that contribute to repolarization reserve. This has significant implications for interpretation of whole-genome sequencing data and underlies the importance of interpreting the entire molecular signature of disease in any given individual.

Adaptive Subspace Symbolization for Content-Based Video Detection

Efficiently and effectively identifying similar videos is an important and nontrivial problem in content-based video retrieval. This paper proposes a subspace symbolization approach, namely SUDS, for content-based retrieval on very large video databases. The novelty of SUDS is that it explores the data distribution in subspaces to build a visual dictionary with which the videos are processed by deriving the string matching techniques with two-step data simplification. Specifically, we first propose an adaptive approach, called VLP, to extract a series of dominant subspaces of variable lengths from the whole visual feature space without the constraint of dimension consecutiveness. A stable visual dictionary is built by clustering the video keyframes over each dominant subspace. A compact video representation model is developed by transforming each keyframe into a word that is a series of symbols in the dominant subspaces, and further each video into a series of words. Then, we present an innovative similarity measure called CVE, which adopts a complementary information compensation scheme based on the visual features and sequence context of videos. Finally, an efficient two-layered index strategy with a number of query optimizations is proposed to facilitate video retrieval. The experimental results demonstrate the high effectiveness and efficiency of SUDS.

Galaxy + Hadoop: Toward a Collaborative and Scalable Image Processing Toolbox in Cloud

With emergence and adoption of cloud computing, cloud has become an effective collaboration platform for integrating various software tools to deliver as services. In this paper, we present a cloud-based image processing toolbox by integrating Galaxy, Hadoop and our proprietary image processing tools. This toolbox allows users to easily design and execute complex image processing tasks by sharing various advanced image processing tools and scalable cloud computation capacity. The paper provides the integration architecture and technical details about the whole system. In particular, we present our investigations to use Hadoop to handle massive image processing jobs in the system. A number of real image processing examples are used to demonstrate the usefulness and scalability of this class of data-intensive applications.

Applications of heterogeneous computing in computational and simulation science

As the size and complexity of scientific problems and datasets grow, scientists from a broad range of discipline areas are relying more and more on computational methods and simulations to help solve their problems. This paper presents a summary of heterogeneous algorithms and applications that have been developed by a large research organization (CSIRO) for solving practical and challenging science problems faster than is possible with conventional multi-core CPUs alone. The problem domains discussed include biological image analysis, computed tomography reconstruction, marine biogeochemical models, fluid dynamics, and bioinformatics. The algorithms utilize GPUs and multi-core CPUs on a scale ranging from single workstation installations through to large GPU clusters. Results demonstrate that large GPU clusters can be used to accelerate a variety of practical science applications, and justify the significant financial investment and interest being placed into such systems.

X-ray imaging software tools for HPC clusters and the Cloud

Computed Tomography (CT) is a non-destructive imaging technique widely used across many scientific, industrial and medical fields. It is both computationally and data intensive, and therefore can benefit from infrastructure in the “supercomputing” domain for research purposes, such as Synchrotron science. Our group within CSIRO has been actively developing X-ray tomography and image processing software and systems for HPC clusters. We have also leveraged the use of GPUs (Graphical Processing Units) for several codes enabling speedups by an order of magnitude or more over CPU-only implementations. A key goal of our systems is to enable our targeted “end users”, researchers, easy access to the tools, computational resources and data via familiar interfaces and client applications such that specialized HPC expertise and support is generally not required in order to initiate and control data processing, analysis and visualzation workflows. We have strived to enable the use of HPC facilities in an interactive fashion, similar to the familiar Windows desktop environment, in contrast to the traditional batch-job oriented environment that is still the norm at most HPC installations. Several collaborations have been formed, and we currently have our systems deployed on two clusters within CSIRO, Australia. A major installation at the Australian Synchrotron (MASSIVE GPU cluster) where the system has been integrated with the Imaging and Medical Beamline (IMBL) detector to provide rapid on-demand CT-reconstruction and visualization capabilities to researchers whilst on-site and remotely. A smaller-scale installation has also been deployed on a mini-cluster at the Shanghai Synchrotron Radiation Facility (SSRF) in China. All clusters run the Windows HPC Server 2008 R2 operating system. The two large clusters running our software, MASSIVE and CSIRO Bragg are currently configured as “hybrid clusters” in which individual nodes can be dual-booted between Linux and Windows as demand requires. We have also recently explored the adaptation of our CT-reconstruction code to Cloud infrastructure, and have constructed a working “proof-of-concept” system for the Microsoft Azure Cloud. However, at this stage several challenges remain to be met in order to make it a truly viable alternative to our HPC cluster solution. Recently, CSIRO was successful in its proposal to develop eResearch tools for the Australian Government funded NeCTAR Research Cloud. As part of this project our group will be contributing CT and imaging processing components.

A Subspace Symbolization Approach to Content-Based Video Search

We propose a subspace symbolization approach, namely SUDS, for content-based search on very large video databases. The novelty of SUDS is that it explores the data distribution in subspaces to build a visual dictionary. With this dictionary, the video data are processed using string matching techniques with two-step data simplification. A compact video representation model is developed by transforming each keyframe into a word that is a series of symbols in the dominant subspaces. Then, we present an innovative similarity measure called ED, which draws from the concept of the edit distance on strings to conduct video matching. The experimental results demonstrate the higheffectiveness of SUDS with optimal parameters.

Advanced investigation on the change in the streamflow into the water source of the middle route of China's water diversion project

To alleviate water shortage in northern China, the middle route of the South to North Water Diversion Project (MRP) was constructed by the Chinese government. A dramatic reduction in the annual streamflow into Danjiangkou Reservoir (ASDR), the water source of MRP, during 1990 has raised some concerns on the MRPs operation. This paper employed an advanced segmented regression model with more recent data to have a clear picture and understand the changing pattern of the ASDR. Our study firstly revealed a zigzag changing pattern (decreasing-increasing-decreasing-increasing) of ASDR during 1960-2013, which was supported by statistical criteria compared with a monotonic or single abrupt change. Particularly, the significantly decreasing trend from 1990s was reversed after 2000, and such change may relieve the concern about the water availability in the future. Sensitivity analysis showed that changes in streamflow were largely influenced by the combined effects of precipitation (P) and potential evapotranspiration (ET0), and were more sensitive to P than ET0. As ET0 is estimated from other primary variables, further analysis was conducted to understand the sensitivities of ET0 to its primary driving variables (wind speed, actual vapor pressure, temperature and sunshine duration), and indicated that ET0 is mostly sensitive to actual vapor pressure during 1960-2013. The findings will assist the MRPs operation and management. Moreover, the results in this study also indicates that an adaptive water diversion plan, rather than the current plan with a constant annual amount of diversion water, might be a better option in the MRPs operation.

Biomedical image analysis and processing in clouds

Cloud-based Image Analysis and Processing Toolbox project runs on the Australian National eResearch Collaboration Tools and Resources (NeCTAR) cloud infrastructure and allows access to biomedical image processing and analysis services to researchers via remotely accessible user interfaces. By providing user-friendly access to cloud computing resources and new workflow-based interfaces, our solution enables researchers to carry out various challenging image analysis and reconstruction tasks. Several case studies will be presented during the conference.