Database Independent Human Emotion Recognition with Meta-Cognitive Neuro-Fuzzy Inference System

Facial emotions are the most expressive way to display emotions. Many algorithms have been proposed which employ a particular set of people (usually a database) to both train and test their model. This paper focuses on the challenging task of database independent emotion recognition, which is a generalized case of subject-independent emotion recognition. The emotion recognition system employed in this work is a Meta-Cognitive Neuro-Fuzzy Inference System (McFIS). McFIS has two components, a neuro-fuzzy inference system, which is the cognitive component and a self-regulatory learning mechanism, which is the meta-cognitive component. The meta-cognitive component, monitors the knowledge in the neuro-fuzzy inference system and decides on what-to-learn, when-to-learn and how-to-learn the training samples, efficiently. For each sample, the McFIS decides whether to delete the sample without being learnt, use it to add/prune or update the network parameter or reserve it for future use. This helps the network avoid over-training and as a result improve its generalization performance over untrained databases. In this study, we extract pixel based emotion features from well-known (Japanese Female Facial Expression) JAFFE and (Taiwanese Female Expression Image) TFEID database. Two sets of experiment are conducted. First, we study the individual performance of both databases on McFIS based on 5-fold cross validation study. Next, in order to study the generalization performance, McFIS trained on JAFFE database is tested on TFEID and vice-versa. The performance The performance comparison in both experiments against SVNI classifier gives promising results.

Human action recognition in H.264/AVC compressed domain using meta-cognitive radial basis function network

In this paper, we propose a H.264/AVC compressed domain human action recognition system with projection based metacognitive learning classifier (PBL-McRBFN). The features are extracted from the quantization parameters and the motion vectors of the compressed video stream for a time window and used as input to the classifier. Since compressed domain analysis is done with noisy, sparse compression parameters, it is a huge challenge to achieve performance comparable to pixel domain analysis. On the positive side, compressed domain allows rapid analysis of videos compared to pixel level analysis. The classification results are analyzed for different values of Group of Pictures (GOP) parameter, time window including full videos. The functional relationship between the features and action labels are established using PBL-McRBFN with a cognitive and meta-cognitive component. The cognitive component is a radial basis function, while the meta-cognitive component employs self-regulation to achieve better performance in subject independent action recognition task. The proposed approach is faster and shows comparable performance with respect to the state-of-the-art pixel domain counterparts. It employs partial decoding, which rules out the complexity of full decoding, and minimizes computational load and memory usage. This results in reduced hardware utilization and increased speed of classification. The results are compared with two benchmark datasets and show more than 90% accuracy using the PBL-McRBFN. The performance for various GOP parameters and group of frames are obtained with twenty random trials and compared with other well-known classifiers in machine learning literature. (C) 2015 Elsevier B.V. All rights reserved.

Territorial Integrity and Self-Determination : The Approach of the International Court of Justice in the Advisory Opinion on Kosovo

[EN] On 17 February 2008 Kosovo approved its declaration of independence from Serbia. The declaration was raised as a unilateral secession, a category which to date is widely debated by the international community, but supported in that case by a respectable number of the United Nation member states. A great many legal issues have been raised by the International Court of Justice's Advisory Opinion on Kosovo. This opinion was eagerly awaited by legal scholars due to both its possible effects and the scope of its principles outside the context of decolonization in what it could constitute of new approach to the international scenario for the twenty-first century. The ICJ stated that the declaration of independence was in accordance with international law if it was not prohibited. The answer turned on whether or not international law prohibited the declaration of independence, without ever examining whether an entity seeking secession is entitled with a positive right to secede and if so, under which circumstances. The basic issue can be summarised as whether or not we are facing a new course in the interpretation of certain classical categories of international law: the principle of territorial integrity, statehood, sovereignty, recognition, the right to external self-determination, etc. In this study we shall analyse some of the aspects arising from the Advisory Opinion of the International Court of Justice on the Accordance with international law of the unilateral declaration of independence in respect of Kosovo focusing on the territorial issue. Firstly we shall analyse the scope of the principle of territorial integrity of States and how it operates ; secondly, we shall focus on the scope of that principle in relation to the interior of the State, and ask ourselves how international law operates in relation to declarations of independence. Lastly, we shall deal with the principle of respect for territorial integrity in the specific case of Serbia with respect to Kosovo, and then end with a series of general conclusions. This study aims, definitely, to contribute to the theoretical debate on the challenges to the traditional certainties of international law in this area.

The body social: an enactive approach to the self

This paper takes a new look at an old question: what is the human self? It offers a proposal for theorizing the self from an enactive perspective as an autonomous system that is constituted through interpersonal relations. It addresses a prevalent issue in the philosophy of cognitive science: the body-social problem. Embodied and social approaches to cognitive identity are in mutual tension. On the one hand, embodied cognitive science risks a new form of methodological individualism, implying a dichotomy not between the outside world of objects and the brain-bound individual but rather between body-bound individuals and the outside social world. On the other hand, approaches that emphasize the constitutive relevance of social interaction processes for cognitive identity run the risk of losing the individual in the interaction dynamics and of downplaying the role of embodiment. This paper adopts a middle way and outlines an enactive approach to individuation that is neither individualistic nor disembodied but integrates both approaches. Elaborating on Jonas' notion of needful freedom it outlines an enactive proposal to understanding the self as co-generated in interactions and relations with others. I argue that the human self is a social existence that is organized in terms of a back and forth between social distinction and participation processes. On this view, the body, rather than being identical with the social self, becomes its mediator

Dealing with the Effects of Sensor Displacement in Wearable Activity Recognition

Most wearable activity recognition systems assume a predefined sensor deployment that remains unchanged during runtime. However, this assumption does not reflect real-life conditions. During the normal use of such systems, users may place the sensors in a position different from the predefined sensor placement. Also, sensors may move from their original location to a different one, due to a loose attachment. Activity recognition systems trained on activity patterns characteristic of a given sensor deployment may likely fail due to sensor displacements. In this work, we innovatively explore the effects of sensor displacement induced by both the intentional misplacement of sensors and self-placement by the user. The effects of sensor displacement are analyzed for standard activity recognition techniques, as well as for an alternate robust sensor fusion method proposed in a previous work. While classical recognition models show little tolerance to sensor displacement, the proposed method is proven to have notable capabilities to assimilate the changes introduced in the sensor position due to self-placement and provides considerable improvements for large misplacements.

The modified Kanerva model for automatic speech recognition

A parallel processing network derived from Kanerva's associative memory theory Kanerva 1984 is shown to be able to train rapidly on connected speech data and recognize further speech data with a label error rate of 0·68%. This modified Kanerva model can be trained substantially faster than other networks with comparable pattern discrimination properties. Kanerva presented his theory of a self-propagating search in 1984, and showed theoretically that large-scale versions of his model would have powerful pattern matching properties. This paper describes how the design for the modified Kanerva model is derived from Kanerva's original theory. Several designs are tested to discover which form may be implemented fastest while still maintaining versatile recognition performance. A method is developed to deal with the time varying nature of the speech signal by recognizing static patterns together with a fixed quantity of contextual information. In order to recognize speech features in different contexts it is necessary for a network to be able to model disjoint pattern classes. This type of modelling cannot be performed by a single layer of links. Network research was once held back by the inability of single-layer networks to solve this sort of problem, and the lack of a training algorithm for multi-layer networks. Rumelhart, Hinton & Williams 1985 provided one solution by demonstrating the "back propagation" training algorithm for multi-layer networks. A second alternative is used in the modified Kanerva model. A non-linear fixed transformation maps the pattern space into a space of higher dimensionality in which the speech features are linearly separable. A single-layer network may then be used to perform the recognition. The advantage of this solution over the other using multi-layer networks lies in the greater power and speed of the single-layer network training algorithm. © 1989.

A simple bioconjugate attachment protocol for use in single molecule force spectroscopy experiments based on mixed self-assembled monolayers.

Single molecule force spectroscopy is a technique that can be used to probe the interaction force between individual biomolecular species. We focus our attention on the tip and sample coupling chemistry, which is crucial to these experiments. We utilised a novel approach of mixed self-assembled monolayers of alkanethiols in conjunction with a heterobifunctional crosslinker. The effectiveness of the protocol is demonstrated by probing the biotin-avidin interaction. We measured unbinding forces comparable to previously reported values measured at similar loading rates. Specificity tests also demonstrated a significant decrease in recognition after blocking with free avidin.

Self-starting passively mode-locking all-solid-state laser with GaAs absorber grown at low temperature

We realize a stable self-starting passively mode-locking all-solid-state laser by using novel GaAs mirrors as the absorber and output coupler. The GaAs mirror is grown by the technology of metal organic chemical vapour deposition at low temperature. With such an absorber as the output coupler in the laser resonator, laser pulses with duration of 42ps were generated at a repetition rate of 400MHz, corresponding to the average power of 590mW.

2-ps passively mode-locked Nd : YVO4 laser using an output-coupling-type semiconductor saturable absorber mirror

We have demonstrated stable self-starting passive mode-locking in a diode-end-pumped Nd: YVO4 laser using a semiconductor saturable absorber mirror (SESAM). An ln(0.25)Ga(0.75)As single quantum-well SESAM, which was grown by the metalorganic chemical-vapor deposition technique at low temperature, acts as a passive mode-locking device and an output coupler at the same time. Continuous-wave mode-locked transform-limited pulses were obtained at 1064 nm with a pulse duration of 2.1 ps and an average output power of 1.28 W at a repetition rate of 96.5 MHz. (c) 2005 American Institute of Physics.

Novel folding 8x8 silicon-based optical matrix switch with tapered waveguides and self-aligned corner mirrors

Novel folding 8 x 8 matrix switches based on silicon on insulator were demonstrated. In the design, single-mode rib waveguides and multimode interferences are connected by optimized tapered waveguides to reduce the mode coupling loss between the two types of waveguides. The self-aligned method was applied to the key integrated turning mirrors for perfect positions and low loss of them. A mixed etching process including inductively coupled plasma and chemical etching was employed to etch waveguides and mirrors, respectively. The compact size of the device is only 20 x 3.2 mm(2). The switch element with high switching speed and low power consumption is presented in the matrix. The average insertion loss of the matrix is about -21 dB, and the excess loss of one mirror is measured of -1.4 dB. The worst crosstalk is larger than 21 dB. Experimental results illuminate that some of the main characteristics of optical matrix switches are. developed in the modified design, which is in accord with theoretic analyses.

Size evolution and optical properties of self-assembled InAs quantum dots on different matrix

InAs quantum dots have been grown by solid source molecular beam epitaxy on different matrix to investigate the effect on the structure and optical properties. High density of 1.02 x 10(11) cm(-2) of InAs islands on In0.15Ga0.85As and In0.15Al0.85As underlying layer has been achieved. Atomic force microscopy and photoluminescence spectra show the size evolution of InAs islands on In0.15Ga0.85As underlying layer. A strong 1.3 mum photoluminescence from InAs islands on In0.15Ga0.85As underlying layer and with InGaAs strain-reduced layer has been obtained. Single-mirror light emitting diode structures with InAs quantum dots capped by InGaAs grown on InGaAs layer as active layer were fabricated and the corresponding radiative efficiency was deduced to be as high as 20.5%. Our results provide important information for optimizing the epitaxial structures of 1.3 mum wavelength quantum dots devices. (C) 2003 Elsevier B.V. All rights reserved.

Self-assembly of single-stranded RNA on carbon nanotube: Polyadenylic acid to form a duplex structure

All messenger-RNA (mRNA) molecules in eukaryotic cells have a polyadenylic acid [poly (rA)] tail at the 3'-end and human poly (rA) polymerase (PAP) has been considered as a tumor-specific target. A ligand that is capable of recognizing and binding to the poly(M) tail of mRNA might interfere with the full processing of mRNA by PAP and can be a potential therapeutic agent. We report here for the first time that single-walled carbon nanotubes (SWNTs) can cause single-stranded poly (M) to self-structure and form a duplex structure, which is studied by UV melting, atomic force microscopy, circular dichroism spectroscopy, and NMR spectrometry.

Colorimetric recognition of the coralyne-poly(dA) interaction using unmodified gold nanoparticle probes, and further detection of coralyne based upon this recognition system

Among the functional nucleic acids studied, adenine-rich nucleic acids have attracted attention due to their critical roles in many biological processes and self-assembly-based nanomaterials, especially deoxyribonucleic acids (abbreviated as poly(dA)). Therefore the ligands binding to poly(dA) might serve as potential therapeutic agents. Coralyne, a kind of planar alkaloid, has been firstly found that it could bind strongly to poly(dA). This work herein reports an approach for visual sensing of the coralyne-poly(dA) interaction. This method was based on the coralyne inducing poly(dA) into the homo-adenine DNA duplex and the difference in electrostatic affinity between single-stranded DNA and double-stranded DNA with gold nanoparticles (GNPs). Furthermore, we applied the recognition process of the interaction between coralyne and poly(dA) into specific coralyne detection with the assistance of certain software (such as Photoshop). A linear response from 0 to 728 nM was obtained for coralyne, and a detection limit of 91 nM was achieved.