Imaging the failure of a rock-fill dam following liquefaction

Rock-fill dams are popular in developing countries due to their ease of construction and use of local materials. They are used to store water and to provide flood defences. The presence of such dams in earthquake-prone regions poses risks, particularly from ground liquefaction. In this paper, results from physical model tests on dams with different configurations are presented. Model dams with impermeable cores including sheet pile walls and clay cores were tested and the effect of reservoir water was investigated. High-speed photography was used to capture the response of the model dams allowing the movement of foundation soil below the dam to be established. It is concluded that the stiffness of the impermeable core has a significant influence on the ultimate deformation of the dam. The presence of reservoir water led to increased downstream movements of the dam and differential settlements between the upstream and downstream sides.

Amorphous oxide semiconductor TFTs for displays and imaging

© 2013 IEEE. This paper reviews the mechanisms underlying visible light detection based on phototransistors fabricated using amorphous oxide semiconductor technology. Although this family of materials is perceived to be optically transparent, the presence of oxygen deficiency defects, such as vacancies, located at subgap states, and their ionization under illumination, gives rise to absorption of blue and green photons. At higher energies, we have the usual band-to-band absorption. In particular, the oxygen defects remain ionized even after illumination ceases, leading to persistent photoconductivity, which can limit the frame-rate of active matrix imaging arrays. However, the persistence in photoconductivity can be overcome through deployment of a gate pulsing scheme enabling realistic frame rates for advanced applications such as sensor-embedded display for touch-free interaction.

Single camera pose estimation using Bayesian filtering and Kinect motion priors

Traditional approaches to upper body pose estimation using monocular vision rely on complex body models and a large variety of geometric constraints. We argue that this is not ideal and somewhat inelegant as it results in large processing burdens, and instead attempt to incorporate these constraints through priors obtained directly from training data. A prior distribution covering the probability of a human pose occurring is used to incorporate likely human poses. This distribution is obtained offline, by fitting a Gaussian mixture model to a large dataset of recorded human body poses, tracked using a Kinect sensor. We combine this prior information with a random walk transition model to obtain an upper body model, suitable for use within a recursive Bayesian filtering framework. Our model can be viewed as a mixture of discrete Ornstein-Uhlenbeck processes, in that states behave as random walks, but drift towards a set of typically observed poses. This model is combined with measurements of the human head and hand positions, using recursive Bayesian estimation to incorporate temporal information. Measurements are obtained using face detection and a simple skin colour hand detector, trained using the detected face. The suggested model is designed with analytical tractability in mind and we show that the pose tracking can be Rao-Blackwellised using the mixture Kalman filter, allowing for computational efficiency while still incorporating bio-mechanical properties of the upper body. In addition, the use of the proposed upper body model allows reliable three-dimensional pose estimates to be obtained indirectly for a number of joints that are often difficult to detect using traditional object recognition strategies. Comparisons with Kinect sensor results and the state of the art in 2D pose estimation highlight the efficacy of the proposed approach.