Sparse Representations of Image via Overcomplete Dictionary Learned by Adaptive Non-orthogonal Sparsifying Transform

How to learn an over complete dictionary for sparse representations of image is an important topic in machine learning, sparse coding, blind source separation, etc. The so-called K-singular value decomposition (K-SVD) method [3] is powerful for this purpose, however, it is too time-consuming to apply. Recently, an adaptive orthogonal sparsifying transform (AOST) method has been developed to learn the dictionary that is faster. However, the corresponding coefficient matrix may not be as sparse as that of K-SVD. For solving this problem, in this paper, a non-orthogonal iterative match method is proposed to learn the dictionary. By using the approach of sequentially extracting columns of the stacked image blocks, the non-orthogonal atoms of the dictionary are learned adaptively, and the resultant coefficient matrix is sparser. Experiment results show that the proposed method can yield effective dictionaries and the resulting image representation is sparser than AOST.

What helps children to move more at school recess and lunchtime? Mid-intervention results from Transform-Us! cluster-randomised controlled trial

Background 

Image reduction operators as aggregation functions: Fuzzy transform and undersampling

After studying several reduction algorithms that can be found in the literature, we notice that there is not an axiomatic definition of this concept. In this work we propose the definition of weak reduction operators and we propose the properties of the original image that reduced images must keep. From this definition, we study whether two methods of image reduction, undersampling and fuzzy transform, satisfy the conditions of weak reduction operators.

Determination of embedment depth of timber poles and piles using wavelet transform

This paper presents an investigation on the wave propagation in timber poles with Wavelet Transform (WT) analysis for identification of the condition and underground depth of embedded timber poles in service. Most of non-destructive testing (NDT) applications for timber poles using wave-based methods consider only single wave mode and no dispersion. However, for wave propagations in timber poles (damaged/undamaged), such simplification may not be correct, especially for broad band excitation using impulse impact. To investigate the problem, a 5m timber pole was investigated numerically and experimentally. A dispersion curve is generated from the numerical results to provide guidance on the velocity and wave mode selection. Continuous wavelet transform (CWT) is applied on the same signal to verify the presence of modes and to process data from experimental testing. The results are presented in both time domain and time-frequency domain for comparison. The results of the investigation showed that, wavelet transform analysis can be a reliable signal processing tool for NDT in terms of condition and embedment length determination.

Attenuated total reflectance fourier transform infrared spectroscopy: an analytical technique to understand therapeutic responses at the molecular level

Rapid monitoring of the response to treatment in cancer patients is essential to predict the outcome of the therapeutic regimen early in the course of the treatment. The conventional methods are laborious, time-consuming, subjective and lack the ability to study different biomolecules and their interactions, simultaneously. Since; mechanisms of cancer and its response to therapy is dependent on molecular interactions and not on single biomolecules, an assay capable of studying molecular interactions as a whole, is preferred. Fourier Transform Infrared (FTIR) spectroscopy has become a popular technique in the field of cancer therapy with an ability to elucidate molecular interactions. The aim of this study, was to explore the utility of the FTIR technique along with multivariate analysis to understand whether the method has the resolution to identify the differences in the mechanism of therapeutic response. Towards achieving the aim, we utilized the mouse xenograft model of retinoblastoma and nanoparticle mediated targeted therapy. The results indicate that the mechanism underlying the response differed between the treated and untreated group which can be elucidated by unique spectral signatures generated by each group. The study establishes the efficiency of non-invasive, label-free and rapid FTIR method in assessing the interactions of nanoparticles with cellular macromolecules towards monitoring the response to cancer therapeutics.

Determinations of stress wave velocity in a timber pole using wavelet transform

This paper presents an application of Wavelet Transfonn (WT) for determination of stress wave velocity for Non-destructive Testing of timber utility poles in service. For surface Non-destructive Testing (NDT), the hammer impact, which produces generally broadband frequency excitation, is used to generate stress wave. Moreover, due to practicality the impact location for field testing of a utility pole is on the side of the pole and 1.5 m above ground level. And the geometry of utility pole could not guarantee non-dispersive longitudinal wave. All of these issues have resulted in lack of accuracy and reliability of results from surface NDT in field testing. In recognition of such problem, this research explores methods to reliably calculate desired wave velocity by isolating wave mode and studying dispersive nature of utility pole. Fast Fourier Transfonn (FFT) is firstly conducted to determine the suitable frequency from a stress wave data. Then WT is applied on the wave data mentioned to perfonn time-frequency analysis. Velocity can be detennined by time history data of desired frequency from WT results which will be compared with the available analytical solution for longitudinal wave velocity. The results of the investigation showed that wavelet transfonn analysis can be a reliable signal processing tool for non-destructive testing in tenns of velocity detennination, which in tum also helps to detennine the embedded length of the timber pole.

Transition, school, friends and obesity risk models (the TranSFORM study)

The TranSFORM study demonstrated the impact that a change of school environment has on physical activity, sedentary and dietary behaviour in late childhood. Results provide strong evidence of the influence of friendship networks, and the potential for changing and reinforcing positive weight related behaviour in early adolescence.