Vibration frequency measurement using a local multithreshold technique

In this paper, we demonstrate the use of a video camera for measuring the frequency of small-amplitude vibration movements. The method is based on image acquisition and multilevel thresholding and it only requires a video camera with high enough acquisition rate, not being necessary the use of targets or auxiliary laser beams. Our proposal is accurate and robust. We demonstrate the technique with a pocket camera recording low-resolution videos with AVI-JPEG compression and measuring different objects that vibrate in parallel or perpendicular direction to the optical sensor. Despite the low resolution and the noise, we are able to measure the main vibration modes of a tuning fork, a loudspeaker and a bridge. Results are successfully compared with design parameters and measurements with alternative devices.

GPGPU implementation of growing neural gas: application to 3D scene reconstruction

Self-organising neural models have the ability to provide a good representation of the input space. In particular the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time-consuming, especially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This paper proposes a Graphics Processing Unit (GPU) parallel implementation of the GNG with Compute Unified Device Architecture (CUDA). In contrast to existing algorithms, the proposed GPU implementation allows the acceleration of the learning process keeping a good quality of representation. Comparative experiments using iterative, parallel and hybrid implementations are carried out to demonstrate the effectiveness of CUDA implementation. The results show that GNG learning with the proposed implementation achieves a speed-up of 6× compared with the single-threaded CPU implementation. GPU implementation has also been applied to a real application with time constraints: acceleration of 3D scene reconstruction for egomotion, in order to validate the proposal.

Spin dynamics of current-driven single magnetic adatoms and molecules

A scanning tunneling microscope can probe the inelastic spin excitations of a single magnetic atom in a surface via spin-flip assisted tunneling in which transport electrons exchange spin and energy with the atomic spin. If the inelastic transport time, defined as the average time elapsed between two inelastic spin flip events, is shorter than the atom spin-relaxation time, the scanning tunnel microscope (STM) current can drive the spin out of equilibrium. Here we model this process using rate equations and a model Hamiltonian that describes successfully spin-flip-assisted tunneling experiments, including a single Mn atom, a Mn dimer, and Fe Phthalocyanine molecules. When the STM current is not spin polarized, the nonequilibrium spin dynamics of the magnetic atom results in nonmonotonic dI/dV curves. In the case of spin-polarized STM current, the spin orientation of the magnetic atom can be controlled parallel or antiparallel to the magnetic moment of the tip. Thus, spin-polarized STM tips can be used both to probe and to control the magnetic moment of a single atom.

Probing a single nuclear spin in a silicon single electron transistor

We study single electron transport across a single Bi dopant in a silicon nanotransistor to assess how the strong hyperfine coupling with the Bi nuclear spin I = 9/2 affects the transport characteristics of the device. In the sequential tunneling regime we find that at, temperatures in the range of 100 mK, dI/dV curves reflect the zero field hyperfine splitting as well as its evolution under an applied magnetic field. Our non-equilibrium quantum simulations show that nuclear spins can be partially polarized parallel or antiparallel to the electronic spin just tuning the applied bias.

Parallel Computational Intelligence-Based Multi-Camera Surveillance System

In this work, we present a multi-camera surveillance system based on the use of self-organizing neural networks to represent events on video. The system processes several tasks in parallel using GPUs (graphic processor units). It addresses multiple vision tasks at various levels, such as segmentation, representation or characterization, analysis and monitoring of the movement. These features allow the construction of a robust representation of the environment and interpret the behavior of mobile agents in the scene. It is also necessary to integrate the vision module into a global system that operates in a complex environment by receiving images from multiple acquisition devices at video frequency. Offering relevant information to higher level systems, monitoring and making decisions in real time, it must accomplish a set of requirements, such as: time constraints, high availability, robustness, high processing speed and re-configurability. We have built a system able to represent and analyze the motion in video acquired by a multi-camera network and to process multi-source data in parallel on a multi-GPU architecture.