Search

Visual saliency (VS) is an important mechanism for defining which areas of an image will attract more attention of the HVS. Thus, VS can be employed to weight the just noticeable difference (JND) with different attention levels. Some VS-based JND profiles have been proposed in the DCT domain, which used the bottom-up features, such as luminance and texture only. Recently, the research about saliency detection has shown that a better saliency model considering both bottom-up features and top-down features will lead to a significant improvement for the overall saliency detection performance. In this paper, we propose a novel two-layer VS-induced JND profile, which is composed of the bottom-up features and the top-down feature extracted from DCT blocks in the transformed domain. In this model, the luminance and texture features are adapted to calculate the bottom-up features maps, while the top-down feature of focus is used to guide the generation of final salient regions since the camerapersons are used to facilitate the attention regions in focus. The proposed two-layer saliency-induced JND model is further applied to modulate the quantization step in the watermarking framework, which can make full use of its individual merits to achieve a better tradeoff between fidelity and robustness. The experimental results show that the proposed scheme has superior performance than the previous watermarking schemes.

In order to solve the problem of the speech recognition, this paper proposes a novel framework to process speech signals by using the deep fully convolutional neural network (DFCNN) as the acoustic model and the Transformer as the linguistic model. DFCNN can learn more about speech signals through a deep network structure, while Transformer's self-attention mechanism makes it an efficient linguistic model. In order to better extract the characteristic information of speech signal, we transform the speech signal into a two-dimensional language spectrum by frame division and Fourier transform. In this way, the language spectrum can be processed by using convolutional neural network, which has promising ability to process two-dimensional data. The output of the acoustic model is passed into the linguistic model, and a complete sentence is output. At the same time, considering that there are many hyper parameters in the model, in order to adjust these hyper parameters efficiently, we employ the Bayesian optimization algorithm to optimize the hyper parameters of the acoustic model and the linguistic model, so as to further improve the accuracy of speech recognition. Experimental results on speech recognition library THCHS30 show that compared with other models, the optimized model has higher recognition rate for audio and stronger generalization ability.

Time series prediction is an active area and attracts the attention of researchers from lots of fields. With the expansion of the application of time series forecasting in the real world, many traditional methods that predicted time series with a few data are no longer applicable. Motivated by the demand of accurate prediction, we come up with the piecewise time series prediction model combining stacked long short-term memory network with genetic algorithm. Stacked long short-term memory network (LSTM) is an advanced sequence learning technology. We divide the time series into several segments and then use the stacked LSTM trained on the previous segment to predict the next segment to realize online prediction. The parameters of the stacked LSTM for each segment prediction are optimized by improved genetic algorithm. For purpose of demonstrating the feasibility and effectivity of the proposed model, we carry out simulation experiments on basis of a hybrid benchmark chaotic time series and a real-world dataset of the hourly ozone concentration. Experimental results demonstrate that the proposed model can automatically select the proper structure according to the data, which avoids the structure being too redundant or too simple. Furthermore, it can improve the prediction accuracy of time series and has good practicability.

Differential evolution for multi-objective optimization is a useful and straightforward evolutionary optimization algorithm. Since the algorithm randomly searches in problem space, it may get into trouble when we adapt it to solve multi-objective problems. With regard to the limitation of DE, we propose a new algorithm PSDE to solve multi-objective functions based on the PSO operator and DE operator. In this algorithm, we utilize DE operator to promote the uniform spread of solutions and use PSO operator to obtain good local search ability by retaining the best solutions obtained so far. Experimental results on a set of benchmark test functions sustain the effectiveness of the proposed algorithm.

Time series prediction is an important branch of mathematical statistics, that uses stochastic process theory and machine learning method to study the data sequence. It is based on the research of structural models, with the purpose of predicting the future developing trend. The method proposed in this paper is based on ARMA model and Lasso regression, ARMA model is an effective model for time series prediction, and Lasso regression is a kind of compression estimation method to refine the ARMA model. First, we build an ARMA model. Second, we use the extend autocorrelation function to do the model specification. Third, we use Lasso to calculate the model parameters. Finally, the method presented in this paper is compared with some other methods on two real-world datasets. Experimental results demonstrate its effectiveness for time series prediction.

To deal with “curse of dimensionality” caused by high-dimensional data and the underutilization of spatial information in the classification of hyperspectral imagery (HSI), a new spectral-spatial classification method based on probabilistic support vector machine (SVM) and guided filter is proposed. The proposed method utilizes spatial information in two aspects. Firstly, the extended morphological profiles (EMP), differential morphological profiles (DMP) and Gabor textural features of hyperspectral image are extracted, which can preserve the texture, outline and shape of the original image. The spectral and extracted spatial features are joint classified by probabilistic SVM to obtain the initial classification probabilistic maps. Secondly, the guided filter is used to regularize the initial classification probabilistic maps, which can enhance the smoothness of local neighbourhood and preserve the edge information. The proposed algorithm is examined by the Indian Pines and University of Pavia hyperspectral datasets. On the same experimental conditions, the proposed method achieves the highest classification accuracy and the most reasonable classification map, demonstrating the proposed algorithm has obvious advantages in hyperspectral remote image classification.

Recently, the scale of observed time series becomes bigger and bigger, and the learning ability of conventional models is limited to depict the characteristics of large-scale time series. We put forward a deep model that uses the sparse autoencoder to extract the characteristics of the time series layer by layer. Sparse autoencoder can provide a simple explanation of the data by means of a reduced dimensionality of parts and extract the hidden architecture in the data. By this method, on the one hand, the dimension of the matrix that describes the problem is reduced, on the other hand, a large number of data can be compressed and summarized. Then we calculate the output weights by the method of pseudo inverse. Experiment results on Lorenz series and PM2.5 concentration in Shanghai dataset demonstrate the effectiveness of the proposed method.

Kernel recursive least squares (KRLS) has become a new research hotspot, attracting researchers’ attention in the field of time series online prediction. As an online algorithm, its computational form is updating incrementally as a new sample achieved. But as the number of data samples increases, its computational complexity will increase. And there is a lack of ability to track time-varying characteristics in the online prediction process. Therefore, we add a sliding window approach to the sparse KRLS and propose a new improved algorithm, which is sparse kernel recursive least squares based on sliding window (SW-SKRLS). This improved algorithm can limit the computational complexity and perform better online prediction in time-varying environments. To prove the validity of the proposed algorithm, we make simulations on the Lorenz time series and Dalian monthly average temperature and rainfall time series. The simulation results show that this algorithm has good prediction accuracy and efficiency.

Extreme learning machines have attracted increasing attention for setting the input weights and hidden biases randomly. However, excessive hidden neurons in ELMs may cause ill-posed problem when Moore-Penrose generalized inverse method is used to train the neural network. It generates large amplitude of output weights, which leads to bad generalization. To solve this problem, we propose an improved multi-objective particle swarm optimization algorithm to optimize the output weights of ELMs. It consists of two objective functions. One is the sum of the errors between predicted values and target values. The other one is the $L_{2}$ norm of output weights. We improve general MOPSO algorithm by adding chaotic mapping to promote the ergodicity and randomness of new solutions. In addition, a new strategy is proposed to select $pbest$ and $gbest$ to deal with the unbalance of multiple objectives. Experimental results on predicting two benchmark chaotic time series sustain the efficiency and effectiveness of our proposed algorithm. It has better generalization performance and smaller output weights than original ELM and other evolutionary ELMs.

Over the last few years, the combination synchronization of chaotic system receives a special attention. This paper introduces the idea of combination synchronization into complex networks. Based on sliding mode control (SMC) principle, the finite-time combination synchronization (FTCS) of four uncertain complex networks which combining as the form of A+B+C-D is investigated. And there are unknown parameters and uncertain disturbances in all the network models. By using finite time stability theory and linear matrix inequality, designing appropriate network sliding mode surface and control input, then obtain the synchronization condition of the FTCS and corresponding update laws of unknown parameters. The proposed method has a strong robustness which can overcome the influence of uncertain disturbances and identify the unknown parameters of nodes. Furthermore, the method can estimate the high bounds of the synchronous settling time. Simulation experiment proves the feasibility and effectiveness of the proposed method.

For extreme learning machine (ELM), it is difficult to select an appropriate number of hidden layer nodes, which has a great influence on the prediction performance of ELM. Therefore, in order to overcome the influence of a large number of hidden layer nodes to prediction and enhance the prediction performance of ELM, an improved ELM is proposed based on orthogonal projections to latent structures (O-PLS) method. The improved ELM contains two steps. Firstly, the hidden layer of ELM is used to build the nonlinear mapping for input variables and get the output matrix H, where the number of bidden layer nodes can be random assigned. Then O-PLS is used to build the linear regression. Compared with least squares regression (LSR), O-PLS can effectively reduce the high dimensions, overcome the multicollinearity among H, which are caused by an improper number of hidden layer nodes. The simulation experiment results based on Lorenz time series and sunspots-runoff of Yellow river time series demonstrate that the improved ELM can effectively overcome the influence of hidden layer nodes to prediction performance and improve the stability performance and prediction performance of ELM.

Considering the problem that for multivariate time series prediction, the adaptation of a single reservoir may not be sufficient to improve the prediction accuracy, we propose a novel hierarchical neural network herein. In the first hierarchy, several simplified echo state networks - simple cycle reservoirs (SCRs) are used to extract the dynamical features of the multivariate time series. Particle swarm optimization method is conducted in the pre-training stage to optimize the free parameters of SCRs. The reservoir states of SCRs are collected as dynamical features. In the second hierarchy, a feature selection method based on mutual information is used to select a compact feature set as the input for the extreme learning machine (ELM). In order to further improve the prediction accuracy, the optimal number of hidden nodes of the ELM is chosen by a modified recursive algorithm. Simulation results on monthly average temperature and rainfall series in Dalian China sustain that the proposed model is effective for multivariate time series.

The crowdsourcing and internet of things (IoT) have played a significant role in revolutionizing the information age. There is significant literature which has attempted to investigate the role of crowdsourcing and IoT in improving disaster response. However there is hardly any literature which attempted to reflect upon integration of crowdsourcing and IoT to explain better coordination among disaster relief workers to improve disaster response. In response to pressing need, we have attempted to develop a theoretical framework which can help disaster relief workers to improve their coordination using valuable information derived using comprehensive crowdsourcing framework. In this study we have used two-prong research strategies. First we have conducted extensive review of articles published in reputable journals, magazines and blogs by eminent practitioners and policy makers followed by case studies: stampede in Godavari River at Rajahmundry (2015), earthquake in Nepal (2015), flood in Uttarakhand (2013). Finally we have concluded our research findings with further research directions.

This paper presents a method based on partial least squares regression (PLSR) and multiple kernel extreme learning machine (MKLEM) for multivariate chaotic time series prediction. At first, singular spectrum analysis (SSA) is applied for the time series extraction of complex trends and eliminating the influence of noise. Then, partial least squares regression is used to capture the essential structure of the data and extract the compositions, in order to overcome the multicollinearity problem among time series and reduce the input dimension of neural networks. Finally, multiple kernel extreme learning machine is used to predict the time series. Multiple kernel extreme learning machine overcomes the problem that single extreme learning machine with kernels (KELM) doesn't present an effective generalization performance. Root mean square error (RMSE) is used to measure the performance of the proposed prediction model. The simulation experiment results based on Lorenz chaotic time series and Dalian monthly average temperature-rainfall time series demonstrate that the proposed model is effective for time series prediction, and the prediction accuracy is higher than other models.

This paper presents a multivariate time series modeling and prediction method based on reservoir independent components. As a new type of recurrent neural networks (RNNs), reservoir computing methods have become a new hot topic and attracted wide attention from researchers in the field of time series prediction. It has overcome the problems that traditional gradient descent training algorithms present, for example, the process is computationally expensive, and easy to end in a local minimum. However, there are ill-posed solutions when least square estimation methods are used to calculate the output weights because of the collinear columns or rows in the state matrix. Therefore, we use independent component analysis (ICA) to extract the independent components of the state matrix. In addition, this paper proposes an iterative prediction model based on local error compensation to solve the problem of accumulated errors in multiple-step prediction, in order to realize medium-term prediction. The models have been simulated on benchmark dataset of Lorenz time series and a real-world application of Dalian monthly average temperature-rainfall time series. Simulation results substantiate the proposed methods' effectiveness and characteristics.

Fan blade loss is the most serious condition of typical extreme working conditions in turbofan engines. The transient huge unbalance force leads to the complicated and destructive vibration response of the rotor system. The paper proposed a set of safety design method for the rotor system, which ensured the engine to stop safely after the fan blade lost. The variable stiffness supporting structure at the rear of the fan was presented and its parameters were optimized through dynamic characteristics calculation. The results revealed that the vibration response of the rotor system could be suppressed during the speed reduction process. The spherical matching surface between the bearing ring and the rotation shaft could reduce the impact of the large shaft deformation, and enhanced the bearing's safety.

Echo state networks have become increasingly popular for its superior performance in the field of time series prediction. However, it is difficult to implement the complicated ESN topologies in practice. To solve the problem, we propose a diagonal connected reservoir structure with composite functions inside the nodes. The input is first processed by wavelet functions and then passes through sigmoid activation functions. This increases the diversity of the reservoir. A selection method that takes into account the domain of the input data is applied to initialize the wavelet translation and dilation parameters. The output weights are efficiently computed by the least square method after the reservoir state matrix is formed. We exhibit the merits of our model on a benchmark multivariate chaotic dataset and a real-world application. Experimental results substantiate that the proposed model can achieve significantly good performance with a low-dimensional reservoir.

In this paper, we present an echo state network model based on sparse Gaussian process regression, which has been successfully applied to multivariate time series prediction. While combining the Gaussian process with Echo State Network, the computational complexity of the model is very high. We consider using a group of limited basis functions instead of the original covariance function, which reduces the computational complexity and maintains the prediction performance of the model. In the framework of Bayesian inference, the model can combine prior knowledge and observation data perfectly and provide prediction confidence. The model realizes adaptive estimation of the hyper-parameters by using maximum likelihood approach and avoids complex computation process. Two simulation results show the effectiveness and practicality of the proposed method.