An online AUC formulation for binary classification

The area under the ROC curve (AUC) provides a good scalar measure of ranking performance without requiring a specific threshold for performance comparison among classifiers. AUC is useful for imprecise environments since it operates independently with respect to class distributions and misclassification costs. A direct optimization of this AUC criterion thus becomes a natural choice for binary classifier design. However, a direct formulation based on the AUC criterion would require a high computational cost due to the drastically increasing input pair features. In this paper, we propose an online learning algorithm to circumvent this computational problem for binary classification. Different from those conventional recursive formulations, the proposed formulation involves a pairwise cost function which pairs up a newly arrived data point with those of opposite class in stored data. Moreover, with incorporation of a sparse learning into the online formulation, the computational effort can be significantly reduced. Our empirical results on three different scales of public databases show promising potential in terms of classification AUC, accuracy, and computational efficiency.     

Optimization of Surface Registrations Using Beltrami Holomorphic Flow

In shape analysis, finding an optimal 1-1 correspondence between 3D surfaces within a large class of admissible bijective mappings is of great importance. Such a process is called surface registration. The difficulty lies in the fact that the space of all surface diffeomorphisms is a complicated functional space, making it challenging to exhaustively search for the best mapping. To tackle this problem, we propose a simple representation of bijective surface maps using Beltrami coefficients (BCs)—complex-valued functions defined on surfaces with supremum norm less than 1. Fixing any 3 points on a pair of surfaces, there is a 1-1 correspondence between the set of surface diffeomorphisms between them and the set of BCs. Hence, every bijective surface map may be represented by a unique BC. Conversely, given a BC, we can reconstruct the unique surface map associated with it using the Beltrami Holomorphic flow (BHF) method. Using BCs to represent surface maps is advantageous because it is a much simpler functional space, which captures many essential features of a surface map. By adjusting BCs, we equivalently adjust surface diffeomorphisms to obtain the optimal map with desired properties. More specifically, BHF gives us the variation of the associated map under the variation of BC. Using this, a variational problem over the space of surface diffeomorphisms can be easily reformulated into a variational problem over the space of BCs. This makes the minimization procedure much easier. More importantly, the diffeomorphic property is always preserved. We test our method on synthetic examples and real medical applications. Experimental results demonstrate the effectiveness of our proposed algorithm for surface registration.     

Bifurcation and chaos analysis of atomic force microscope system

Atomic force microscope (AFM) is a very high-resolution type of scanning probe microscope, which is an essential characterization and actuation tool in modern nanoscience or engineering. This paper investigates the bifurcation and chaos behavior of the probe tip from AFM system by the differential transformation method (DTM). The dynamic behavior of the probe tip is characterized by reference to bifurcation diagrams, phase portraits, power spectra, Poincaré maps and maximum Lyapunov exponent plots produced using the time-series data obtained from DTM. The results indicate that the probe tip behavior is significantly dependent on the magnitude of the vibrational amplitude. Specifically, the probe tip motion changes from T-periodic to 3T-periodic, then from 2T-periodic to multi-periodic, and finally to chaotic motion with windows of periodic motion as the vibrational amplitude is increased from 0 to 2.0. Furthermore, it is demonstrated that the DTM is in good agreement for the considered system.     

The effects of platform motion and target orientation on the performance of trackball manipulation

The trackball has been widely employed as a control/command input device on moving vehicles, but few studies have explored the effects of platform motion on its manipulation. Fewer still have considered this issue in designing the user interface and the arrangement of console location and orientation simultaneously. This work describes an experiment carried out to investigate the performance of trackball users on a simple point-and-click task in a motion simulator. By varying the orientation of onscreen targets, the effect of cursor movement direction on performance is investigated. The results indicate that the platform motion and target orientation both significantly affect the time required to point and click, but not the accuracy of target selection. The movement times were considerably longer under rolling and pitching motions and for targets located along the diagonal axes of the interface. Subjective evaluations carried out by the participants agree with these objective results. These findings could be used to optimise console and graphical menu design for use on maritime vessels. STATEMENT OF RELEVANCE: In military situations, matters of life or death may be decided in milliseconds. Any delay or error in classification and identification will thus affect the safety of the ship and its crew. This study demonstrates that performance of manipulating a trackball is affected by the platform motion and target orientation. The results of the present study can guide the arrangement of consoles and the design of trackball-based graphical user interfaces on maritime vessels.     

Residual orientation modeling for fingerprint enhancement and singular point detection

This paper presents a novel method for fingerprint orientation modeling, which executes in two phases. Firstly, the orientation field is reconstructed using a lower order Legendre polynomial to capture the global orientation pattern in the fingerprint structure. Then the preliminary model around the region with presence of fingerprint singularities is dynamically refined using a higher order Legendre polynomial. The singular region is automatically detected through the analysis on the orientation residual field between the original orientation field and the orientation model. The method does not require any prior knowledge on the fingerprint structure. To validate the performance, the method has been applied to fingerprint image enhancement, fingerprint singularity detection and fingerprint recognition using the FVC 2004 data sets. Compared with the recently published Legendre polynomial model, the proposed method attains higher accuracy in fingerprint singularity detection, lower error rates in fingerprint matching.     

Application of the differential transformation method to bifurcation and chaotic analysis of an AFM probe tip

The AFM (atomic force microscope) has become a popular and useful instrument for measuring intermolecular forces with atomic resolution, that can be applied in electronics, biological analysis, and studying materials, semiconductors etc. This paper conducts a systematic investigation into the bifurcation and chaotic behavior of the probe tip of an AFM using the differential transformation (DT) method. The validity of the analytical method is confirmed by comparing the DT solutions for the displacement and velocity of the probe tip at various values of the vibrational amplitude with those obtained using the Runge–Kutta (RK) method. The behavior of the probe tip is then characterized utilizing bifurcation diagrams, phase portraits, power spectra, Poincaré maps, and maximum Lyapunov exponent plots. The results indicate that the probe tip behavior is significantly dependent on the magnitude of the vibrational amplitude. Specifically, the tip motion changes first from subharmonic to chaotic motion, then from chaotic to multi-periodic motion, and finally from multi-periodic motion to subharmonic motion with windows of chaotic behavior as the non-dimensional vibrational amplitude is increased from 1.0 to 5.0.     

Accelerated catadioptric omnidirectional view image unwrapping processing using GPU parallelisation

Catadioptric omnidirectional view sensors have found increasing adoption in various robotic and surveillance applications due to their 360° field of view. However, the inherent distortion caused by the sensors prevents their direct utilisations using existing image processing techniques developed for perspective images. Therefore, a correction processing known as “unwrapping” is commonly performed. However, the unwrapping process incurs additional computational loads on central processing units. In this paper, a method to reduce this burden in the computation is investigated by exploiting the parallelism of graphical processing units (GPUs) based on the Compute Unified Device Architecture (CUDA). More specifically, we first introduce a general approach of parallelisation to the said process. Then, a series of adaptations to the CUDA platform is proposed to enable an optimised usage of the hardware platform. Finally, the performances of the unwrapping function were evaluated on a high-end and low-end GPU to demonstrate the effectiveness of the parallelisation approach.     

Absolute phase-assisted three-dimensional data registration for a dual-camera structured light system

For a three-dimensional shape measurement system with a single projector and multiple cameras, registering patches from different cameras is crucial. Registration usually involves a complicated and time-consuming procedure. We propose a new method that can robustly match different patches via absolute phase without significantly increasing its cost. For y and z coordinates, the transformations from one camera to the other are approximated as third-order polynomial functions of the absolute phase. The x coordinates involve only translations and scalings. These functions are calibrated and only need to be determined once. Experiments demonstrated that the alignment error is within RMS 0.7 mm.     

Stress-corrosion cracking and its propagation in aligned short-fibre composites

Stress-corrosion tests are conducted on aligned short-fibre composites to assess their performance in both air and 1 N H₂SO₄ environments. The materials used are 3 mm E-glass, AS4 carbon, and their hybrid fibre reinforced epoxy composites. Several crack propagation models are developed to describe different cracking behaviour; the observed composite fracture modes verify these proposed models. Pre-immersion has significant effects onK _l-t _f and crack arrest of hybrid composites. A comparison of theoretical and experimentalK _l-t _f curves indicates good agreement of time to failure. The present research has quantified the susceptibility of these composites to environmental attack.