MULTI-INSTRUMENT DETECTION IN POLYPHONIC MUSIC USING GAUSSIAN MIXTURE BASED FACTORIAL HMM

We formulate the problem of detecting the constituent instruments in a polyphonic music piece as a joint decoding problem. From monophonic data, parametric Gaussian Mixture Hidden Markov Models (GM-HMM) are obtained for each instrument. We propose a method to use the above models in a factorial framework, termed as Factorial GM-HMM (F-GM-HMM). The states are jointly inferred to explain the evolution of each instrument in the mixture observation sequence. The dependencies are decoupled using variational inference technique. We show that the joint time evolution of all instruments' states can be captured using F-GM-HMM. We compare performance of proposed method with that of Student's-t mixture model (tMM) and GM-HMM in an existing latent variable framework. Experiments on two to five polyphony with 8 instrument models trained on the RWC dataset, tested on RWC and TRIOS datasets show that F-GM-HMM gives an advantage over the other considered models in segments containing co-occurring instruments.

A fluorescent molecularly-imprinted polymer gate with temperature and pH as inputs for detection of alpha-fetoprotein

In this work, we have reported a new approach on the use of stimuli-responsive molecularly imprinted polymer (MIP) for trace level sensing of alpha-fetoprotein (AFP), which is a well know cancer biomarker. The stimuli-responsive MIP is composed of three components, a thermo-responsive monomer, a pH responsive component (tyrosine derivative) and a highly fluorescent vinyl silane modified carbon dot. The synthesized AFP-imprinted polymer possesses excellent selectivity towards their template molecule and dual-stimuli responsive behavior. Along with this, the imprinted polymer was also explored as `OR' logic gate with two stimuli (pH and temperature) as inputs. However, the non-imprinted polymers did not have such `OR' gate property, which confirms the role of template binding. The imprinted polymer was also used for estimation of AFP in the concentration range of 3.96-80.0 ng mL(-1), with limit of detection (LOD) 0.42 ng mL(-1). The role of proposed sensor was successfully exploited for analysis of AFP in real human blood plasma, serum and urine sample. (C) 2015 Elsevier B.V. All rights reserved.

Detection of quantum well induced single degenerate-transition-dipoles in ZnO nanorods

Quantifying and characterising atomic defects in nanocrystals is difficult and low-throughput using the existing methods such as high resolution transmission electron microscopy (HRTEM). In this article, using a defocused wide-field optical imaging technique, we demonstrate that a single ultrahigh-piezoelectric ZnO nanorod contains a single defect site. We model the observed dipole-emission patterns from optical imaging with a multi-dimensional dipole and find that the experimentally observed dipole pattern and model-calculated patterns are in excellent agreement. This agreement suggests the presence of vertically oriented degenerate-transition-dipoles in vertically aligned ZnO nanorods. The HRTEM of the ZnO nanorod shows the presence of a stacking fault, which generates a localised quantum well induced degenerate-transition-dipole. Finally, we elucidate that defocused wide-field imaging can be widely used to characterise defects in nanomaterials to answer many difficult questions concerning the performance of low-dimensional devices, such as in energy harvesting, advanced metal-oxide-semiconductor storage, and nanoelectromechanical and nanophotonic devices.

Sparse representation-based human detection: a scale-embedded dictionary approach

Human detection is a complex problem owing to the variable pose that they can adopt. Here, we address this problem in sparse representation framework with an overcomplete scale-embedded dictionary. Histogram of oriented gradient features extracted from the candidate image patches are sparsely represented by the dictionary that contain positive bases along with negative and trivial bases. The object is detected based on the proposed likelihood measure obtained from the distribution of these sparse coefficients. The likelihood is obtained as the ratio of contribution of positive bases to negative and trivial bases. The positive bases of the dictionary represent the object (human) at various scales. This enables us to detect the object at any scale in one shot and avoids multiple scanning at different scales. This significantly reduces the computational complexity of detection task. In addition to human detection, it also finds the scale at which the human is detected due to the scale-embedded structure of the dictionary.

Investigation Into Harmful Patterns Over Multitrack Shingled Magnetic Detection Using the Voronoi Model

Two-dimensional magnetic recording 2-D (TDMR) is a promising technology for next generation magnetic storage systems based on a systems-level framework involving sophisticated signal processing at the core. The TDMR channel suffers from severe jitter noise along with electronic noise that needs to be mitigated during signal detection and recovery. Recently, we developed noise prediction-based techniques coupled with advanced signal detectors to work with these systems. However, it is important to understand the role of harmful patterns that can be avoided during the encoding process. In this paper, we investigate the Voronoi-based media model to study the harmful patterns over multitrack shingled recording systems. Through realistic quasi-micromagnetic simulation studies, we identify 2-D data patterns that contribute to high media noise. We look into the generic Voronoi model and present our analysis on multitrack detection with constrained coded data. We show that the 2-D constraints imposed on input patterns result in an order of magnitude improvement in the bit-error rate for the TDMR systems. The use of constrained codes can reduce the complexity of 2-D intersymbol interference (ISI) signal detection, since the lesser 2-D ISI span can be accommodated at the cost of a nominal code rate loss. However, a system must be designed carefully so that the rate loss incurred by a 2-D constraint does not offset the detector performance gain due to more distinguishable readback signals.

Tetraphenylethene-Based Conjugated Fluoranthene: A Potential Fluorescent Probe for Detection of Nitroaromatic Compounds

This study reports the synthesis and photophysical properties of a star-shaped, novel, fluoranthene-tetraphenylethene (TFPE) conjugated luminogen, which exhibits aggregation-induced blue-shifted emission (AIBSE). The bulky fluoranthene units at the periphery prevent intramolecular rotation (IMR) of phenyl rings and induces a blueshift with enhanced emission. The AIBSE phenomenon was investigated by solvatochromic and temperature-dependent emission studies. Nanoaggregates of TFPE, formed by varying the water/THF ratio, were investigated by SEM and TEM and correlated with optical properties. The TFPE conjugate was found to be a promising fluorescent probe towards the detection of nitroaromatic compounds (NACs), especially for 2,4,6-trinitrophenol (PA) with high sensitivity and a high Stern-Volmer quenching constant. The study reveals that nanoaggregates of TFPE formed at 30 and 70% water in THF showed unprecedented sensitivity with detection limits of 0.8 and 0.5ppb, respectively. The nanoaggregates formed at water fractions of 30 and 70% exhibit high Stern-Volmer constants (K-sv=79998 and 51120m(-1), respectively) towards PA. Fluorescence quenching is ascribed to photoinduced electron transfer between TFPE and NACs with a static quenching mechanism. Test strips coated with TFPE luminogen demonstrate fast and ultra-low-level detection of PA for real-time field analysis.

A mode-shape classification technique for robust speech rate estimation and syllable nuclei detection

Acoustic feature based speech (syllable) rate estimation and syllable nuclei detection are important problems in automatic speech recognition (ASR), computer assisted language learning (CALL) and fluency analysis. A typical solution for both the problems consists of two stages. The first stage involves computing a short-time feature contour such that most of the peaks of the contour correspond to the syllabic nuclei. In the second stage, the peaks corresponding to the syllable nuclei are detected. In this work, instead of the peak detection, we perform a mode-shape classification, which is formulated as a supervised binary classification problem - mode-shapes representing the syllabic nuclei as one class and remaining as the other. We use the temporal correlation and selected sub-band correlation (TCSSBC) feature contour and the mode-shapes in the TCSSBC feature contour are converted into a set of feature vectors using an interpolation technique. A support vector machine classifier is used for the classification. Experiments are performed separately using Switchboard, TIMIT and CTIMIT corpora in a five-fold cross validation setup. The average correlation coefficients for the syllable rate estimation turn out to be 0.6761, 0.6928 and 0.3604 for three corpora respectively, which outperform those obtained by the best of the existing peak detection techniques. Similarly, the average F-scores (syllable level) for the syllable nuclei detection are 0.8917, 0.8200 and 0.7637 for three corpora respectively. (C) 2016 Elsevier B.V. All rights reserved.

SALIENT OBJECT DETECTION VIA OBJECTNESS MEASURE

Salient object detection has become an important task in many image processing applications. The existing approaches exploit background prior and contrast prior to attain state of the art results. In this paper, instead of using background cues, we estimate the foreground regions in an image using objectness proposals and utilize it to obtain smooth and accurate saliency maps. We propose a novel saliency measure called `foreground connectivity' which determines how tightly a pixel or a region is connected to the estimated foreground. We use the values assigned by this measure as foreground weights and integrate these in an optimization framework to obtain the final saliency maps. We extensively evaluate the proposed approach on two benchmark databases and demonstrate that the results obtained are better than the existing state of the art approaches.