Showing papers in "Neurocomputing in 2013"

TL;DR: A comprehensive survey of different systems for fall detection and their underlying algorithms is given, divided into three main categories: wearable device based, ambience device based and vision based.

TL;DR: A weighted ELM which is able to deal with data with imbalanced class distribution while maintain the good performance on well balanced data as unweighted ELM and generalized to cost sensitive learning.

TL;DR: The effectiveness of a detection approach based on machine learning is explored, using the Discriminative Restricted Boltzmann Machine to combine the expressive power of generative models with good classification accuracy capabilities to infer part of its knowledge from incomplete training data.

TL;DR: Several new operations/improvements such as the particle update method based on random sampling and uniform mutation, the infeasible archive, the constrained domination relationship based on collision times with obstacles, are incorporated into the proposed algorithm to improve its effectiveness.

TL;DR: A new kernel is derived by establishing a connection with the Riemannian geometry of symmetric positive definite matrices, effectively replacing the traditional spatial filtering approach for motor imagery EEG-based classification in brain-computer interface applications.

TL;DR: The result demonstrates that the proposed recurrent neural classifier using the energy features extracted from characteristic waves of EEG signals can classify sleep stages more efficiently and accurately using only a single EEG channel.

TL;DR: This paper summarises the experiments performed and the possibilities of the extraction techniques, in particular the newly introduced algorithm of markerless detection in stereo recordings, which is developing software for the time-efficient automatic extraction of accurate pedestrian trajectories.

TL;DR: The proposed hybrid intelligent fault detection and classification method can reliably identify different fault patterns of rolling element bearings based on the vibration signals and can achieve a greater accuracy than the commonly used SVM.

TL;DR: An adaptive filtering approach based on discrete wavelet transform and artificial neural network is proposed for ECG signal noise reduction that can successfully remove a wide range of noise with significant improvement on SNR (signal-to-noise ratio).

TL;DR: This paper is the first review analyzing this new trend in automated surveillance of human activities, proposing a structured snapshot of the state of the art and envisaging novel challenges in the surveillance domain where the cross-pollination of Computer Science technologies and Sociology theories may offer valid investigation strategies.

TL;DR: A hybrid learning algorithm is proposed to overcome the drawbacks of ELM, which uses an improved particle swarm optimization (PSO) algorithm to select the input weights and hidden biases and Moore–Penrose (MP) generalized inverse to analytically determine the output weights.

TL;DR: This improved GA presents an effective and accurate fitness function, improves genetic operators of conventional genetic algorithms and proposes a new genetic modification operator.

TL;DR: An efficient memetic algorithm based on particle swarm optimization algorithm (PSO) and pattern search and a novel probabilistic selection strategy to select the appropriate individuals among the current population to undergo local refinement is proposed, keeping a well balance between exploration and exploitation.

TL;DR: The existence, uniqueness and stability of the equilibrium point of a class of fractional-order neural networks with delay are proved and a numerical example is presented to demonstrate the validity and feasibility of the proposed results.

TL;DR: The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP, LTP, local tetra patterns (LTrP) and LDP with and without Gabor transform.

TL;DR: The experimental results have successfully validated that the integration of the PNN classifier with the proposed feature reduction method can achieve satisfactory classification accuracy.

TL;DR: The proposed ACOSampling that is a novel undersampling method based on the idea of ant colony optimization (ACO) to address class imbalance problem in DNA microarray data outperforms many other sampling approaches, which indicates its superiority.

TL;DR: It is concluded that hLBPI, hLBPH and eLBPH respectively are suitable for face representation under what conditions, and expect providing practitioners with guidance in selecting appropriate approaches for real tasks.

TL;DR: The experimental results demonstrate that the proposed parallel ELM for regression can efficiently handle very large datasets on commodity hardware with a good performance on different evaluation criterions, including speedup, scaleup and sizeup.

TL;DR: This paper investigates various combinations of mid-level features to build an effective image retrieval system based on the bag-of-features (BoF) model and shows that the integrations of these features yield complementary and substantial improvement on image retrieval even with noisy background and ambiguous objects.

TL;DR: This paper offers the researchers a link to public image database for the algorithm assessment of text extraction from natural scene images and draws attention to studies on the first two steps in the extraction process, since OCR is a well-studied area where powerful algorithms already exist.

TL;DR: A greedy heuristic dynamic programming iteration algorithm is developed to solve the zero-sum game problems, which can be used to solves the Hamilton-Jacobi-Isaacs equation associated with H"~ optimal regulation control problems.

TL;DR: Several delay-dependent criteria for checking the global stability of the addressed complex-valued neural networks are established in linear matrix inequality (LMI), which can be checked numerically using the effective LMI toolbox in MATLAB.

TL;DR: A robust framework of LBP is proposed, named Completed Robust Local Binary Pattern (CRLBP), in which the value of each center pixel in a 3x3 local area is replaced by its average local gray level, which is more robust to noise and illumination variants.

TL;DR: Experiments on five sentiment classification datasets show that ADN and IADN outperform classical semi-supervised learning algorithms, and deep learning techniques applied for sentiment classification.

TL;DR: This paper proposes the Extended Complete Orthogonal Decomposition (ECOD) method to solve the computational problem in ELM weights computing via ECODLS algorithm and shows that ECOD can effectively replace SVD.

TL;DR: A composite HAFC (CHAFC), which combines the HAFC with composite adaptation technique, is proposed, and it is proved that the closed-loop system obtains H^~ tracking performance in the sense that both tracking and modeling errors converge to small neighborhoods of zero.

TL;DR: A tracking method that deals with appearance variations separately based on sparse representation in a particle filter framework and shows excellent performance in comparison with two latest state-of-the-art trackers.

TL;DR: This study investigates the accuracy of the hybrid SVM-QPSO model (support vector machine-quantum behaved particle swarm optimization) in predicting monthly streamflows and indicates that SVM is a far better technique for predicting Monthly streamflows as it provides a high degree of accuracy and reliability.

TL;DR: Experimental results show that, compared with SVMs, the ordinary ELM and its modifications can all dramatically speed up the training process while still achieving similar or better vigilance estimation accuracy.