Search

In order to accurately estimate the period of X-ray pulsars to improve navigation performance, this paper proposes a pulsar period estimation method based on photon energy using reverse Hilbert coding and double Convolutional Neural Networks (CNNs) discrimination. This method uses photon energy information to perform epoch folding. It converts one-dimensional contour information into two-dimensional image signals through reverse Hilbert curve and recognizes contour image features through CNN models to determine the optimal period. According to the simulation results, the period estimation accuracy of this method is 0.5350 ns, which is 65.71% higher than the χ 2 epoch folding method. The contour phase accuracy is 0.0004767, which is 41.93% higher than the χ 2 epoch folding method. At the same time, the observation time and noise interference that may affect the accuracy of period estimation are also quantitatively analyzed. In addition, in order to simulate real navigation scenarios, this paper also conducts dynamic period estimation experiments on PSR B0531+21 pulsar in a specific period of time. The method proposed in this paper has the advantages of high precision, fast calculation speed, strong anti-interference ability and accurate dynamic period estimation, which can significantly improve the navigation performance of X-ray pulsars. • The proposed period estimation method uses photon energy information for contour weighted folding. • The proposed method uses reverse Hilbert curve to convert one-dimensional contour information into two-dimensional image information. • Convolutional Neural Network models are used to optimize pulsar period estimation. [Display omitted] [ABSTRACT FROM AUTHOR]

Deep neural networks have achieved outstanding performance in computer vision tasks. Convolutional Neural Networks (CNNs) typically operate in the spatial domain with raw images, but in practice, images are usually stored and transmitted in their compressed representation where JPEG is one of the most widely used encoder. Also, these networks are computationally intensive and slow. This paper proposes performing the learning and inference processes in the compressed domain in order to reduce the computational complexity and improve the speed of popular CNNs. For this purpose, a novel graph-based frequency channel selection method is proposed to identify and select the most important frequency channels. The computational complexity is reduced by retaining the important frequency components and discarding the insignificant ones as well as eliminating the unnecessary layers of the network. Experimental results show that the modified ResNet-50 operating in the compressed domain is up to 70% faster than the spatial-based traditional ResNet-50 while resulting in similar classification accuracy. Moreover, this paper proposes a preprocessing step with partial encoding to improve the resilience to distortions caused by low-quality encoded images. Finally, we show that training a network with highly compressed data can achieve a good classification accuracy with up to 93% reduction in the storage requirements of the training data. [ABSTRACT FROM AUTHOR]

• We propose a rotation invariant Gabor convolutional neural network (RIGCN). • We learn Gabor-guided convolutional features in the Siamese network architecture. • We explore a convolutional fusion operator to obtain rotation invariant features. • We show the effectiveness of RIGCN for rotation invariant image classification. Gabor filters have been recently integrated with deep convolutional neural networks to learn better features with fewer model parameters. However, during feature learning the rotation invariance is not well addressed. In this paper, we propose a rotation invariant Gabor convolutional neural network (RIGCN) for image classification. First, we transform each input image to generate multiple rotated image instances and feed them into a weight-sharing Siamese network architecture to learn Gabor-guided deep convolutional features. Then, we compute the maximum and average feature responses from all the rotated instances of the same input image and send them into a convolutional fusion module to obtain a rotation invariant image representation. Finally, we use the cross-entropy loss for image classification. In our method, the use of Siamese architecture enables us to obtain rotation invariant features from rotated image instances. The use of convolutional fusion operator enables us to obtain richer statistical features with high efficiency. Experimental results on several benchmark datasets demonstrate the effectiveness of RIGCN for rotation invariant image classification. [ABSTRACT FROM AUTHOR]

With the significant increase in photovoltaic (PV) electricity generation, more attention has been given to PV power forecasting. Indeed, accurate forecasting allows power grid operators to better schedule and dispatch their assets, such as energy storage systems and reserve. In this paper, a hybrid deep learning model and a convolutional neural network with memory is proposed, to provide intraday (2 h) solar irradiance forecasts using sequentially-collected images from public webcams. The performance of the proposed model is compared to those of a standard time-series forecast models, a linear regression as well as state-of-the-art neural networks. All models are trained and tested using images collected from two webcams on EPFL's campus for just over a year. The results show that the proposed method outperforms all other models and matches the state-of-the-art methodology while providing simplicity of implementation and efficient computation. • A novel method for estimating global horizontal irradiance is proposed. • The method uses images public webcams instead of all-sky cameras. • The method combines images from different cameras to improve forecasts' accuracy. • The method can be extended to include other types of data (meteorological data). • This simple method matches state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In this paper, a Raman spectral detection and identification method combined with a portable Raman device is proposed to realize the real-time detection of trace substances. By employing a composite Surface-Enhanced Raman Scattering (SERS) substrate coupled with surface plasmon polaritons (SPP) and localized surface plasmons (LSP), the Raman detection performance for trace can be improved significantly. The one-dimensional dilated convolutional neural network (1D CNN) model proposed in this paper combined with a spatial attention mechanism can accurately identify Raman spectra. Together, a combination of the two provides great advantages, especially for the identification of weak signals with indistinguishable characteristic peaks. At the same time, for the public Raman dataset Rruff, the model has higher accuracy and generalization ability than traditional Raman spectral recognition algorithms. The evidence presented in this paper reveals that the LSP–SPP​ coupled SERS substrate combined with a 1D CNN-based spectral identification algorithm can pave the way for the development of an accurate and portable Raman detection of trace. • A convenient LSP–SPP coupled SERS substrate is proposed. • A SERS spectrum recognition model based on a one-dimensional (1D) dilated CNN combined with spatial attention mechanisms is designed. [ABSTRACT FROM AUTHOR]

In this paper, we tackle the task of similar question retrieval (QR) which is essential for Community Question Answering (cQA) and aims to retrieve historical questions that are semantically equivalent to the new queries. Over time, with the sharp increase of community archives and the accumulation of duplicated questions, the QR problem has become increasingly challenging due to the shortness of the community questions as well as the word mismatch problem as users can formulate the same query using different wording. Although many efforts have been devoted to address this problem, existing methods mostly relied on supervised models which significantly depend on massive training data sets and manual feature engineering. Such methods are chiefly constrained by their specificities that ignore the word order and do not capture enough syntactic and semantic information in questions. In this paper, we rely on Neural Networks (NNs) which use a deep analysis of words and questions to take into consideration the semantics as well as the structure of questions to predict the semantic text similarity. We propose a deep learning approach based on a Siamese architecture with Long Short-Term Memory (LSTM) networks, augmented with an attention mechanism to let the model give different words different attention while modeling questions. We also explore the use of Convolutional Neural Networks (CNN) nested within the Siamese architecture to retrieve relevant questions. Different similarity measures were tested to predict the semantic similarity between the pairs of questions. To evaluate the proposed approach, we conducted experiments on large-scale datasets in English and Arabic. [ABSTRACT FROM AUTHOR]

• A novel hand gesture identification is proposed using CNN, ORB and Gabor filter. • It can be used as a pre-trained structure for any type of the hand gesture dataset. • Remarkable results are obtained on the famous databases. • Less computational intricacy in the hardware implementation and online application. In this paper, a new structure of deep learning neural network is introduced to identify the static hand gesture in the sign language. The proposed structure includes the convolutional neural network (CNN) and the classical non-intelligent feature extraction method. In the proposed structure, the hand gesture image, after preprocessing and removing its background, passes through three different streams of feature extraction, to well extract of effective features and determine the hand gesture class. These three streams, that independently extract their own specific features, consist of three widely used methods in the hand gesture classification named CNN, Gabor filter and ORB feature descriptor. Then these features are merged and formed the final feature vector. By combining these efficient methods, in addition to achieving a very high accuracy in hand gestures classifying, the proposed structure becomes more resistant to uncertainties such as rotation and ambiguity in the hand gestures. Another prominent feature of the proposed structure is its comprehensiveness on different image databases, compared to the similar methods. The transfer learning technique demonstrates that the proposed structure has the ability to be used as a pre-trained structure for any type of image database. Finally, the proposed structure is applied to the three different databases of Massey, ASL Alphabet and ASL, which have 2520, 87,000 and 23,400 of hand gesture images, respectively. The results show the mean accuracy of the proposed structure for the Massey test set of 758 images, ASL with 7020 test images, and ASL Alphabet with 26,100 test images, at 99.92%, 99.8%, and 99.80% respectively. [ABSTRACT FROM AUTHOR]

Deep neural networks (dnn) techniques for aspect-based sentiment classification have been widely studied. The success of these methods depends largely on training data which are often inadequate because of the rigor involved in manually tagging large collection of opinionated texts. Attempts have been made to transfer knowledge from document-level to aspect-level sentiment task. However, the success of this approach is also dependent on the model because aspect sentiment data like other type of texts contain complex semantic features. In this paper, we present an attention-based deep learning technique which jointly learns on document and aspect-level sentiment data and which also transfers learning from the document-level data to aspect-level sentiment classification. It basically consists of a convolutional layer and a bidirectional long short-term memory (Bi lstm) layer. The first variant of our technique uses convolutional neural network (cnn) to extract high-level semantic features. The output of the feature extraction is then fed into the Bi lstm layer which captures the contextual feature representation of the texts. The second variant applies the Bi lstm layer directly on the input data. In both variants, the output hidden representation is passed to an output layer using softmax activation function for sentiment polarity classification. We evaluate our model on four standard benchmark datasets which shows the effectiveness of our approach with improvements over baselines. We also conduct ablation studies to show the effect of the different document-level weights on the learning techniques. [ABSTRACT FROM AUTHOR]

• The double convolutional neural network model for fault data identification is proposed in this paper. The identification rate of distribution network fault types is improved. • A novel convolutional auto-encoding network is used to realize the auto-extraction of the features. This method improves the feature extraction ability and efficiency of the network. • The F 1 score and robustness of the DCNN model were analysed. The optimal structural parameters of DCNN model were determined by parameter sensitivity analysis in discussion. The rapid and accurate identification of different types of faults in the power grid is of great significance to the stable operation of the power grid. An identification model of transient fault recording data for the distribution network based on a double convolutional neural network is proposed in this study. The 1-dimension convolutional auto-encoder (1-D CAE) is used to learn features from the power grid transient data. The obtained low-dimensional fault features are imported into the 1-dimension convolutional neural network (1-D CNN) identification model. The identification accuracy of the proposed model is higher than that of the traditional methods by the verification of the measured transient fault data of the power distribution network. The robustness study implies that the DCNN model can be applied in practical situations prone to using contaminated samples. [ABSTRACT FROM AUTHOR]

• Expanded the insufficient citrus greening diseased leaves to improve the generalization of the classification network. The diseased leaves with insufficient are expanded to strengthen the generalization of the classification network. Meanwhile, the generated citrus yellow dragon disease leaf fake data has the phenotypic characteristics of the real data, and the yellow dragon disease researchers are unable to identify the generated data real and fake, testing out TRL-GAN to achieve certain generation effect. • Effective background culling of direct collection of diseased leaf data in the field to achieve shooting effect under indoor conditions. • To make the classification network achieve a certain improvement in the accuracy rate of citrus greening diseased leaves. The monitoring and prevention and control of citrus yellow dragon disease is a significant measure to ensure citrus production. If yellow dragon disease appears in citrus orchards, it will cause root rot, fruit deformation and wilting of fruit trees, which will eventually spread to every fruit tree in the whole orchard and cause the death of fruit trees, so it is very meaningful to detect the symptoms of citrus yellow dragon disease early and take appropriate treatment and prevention measures. Pratically, the orchard owner will remove the corresponding fruit trees as soon as they are found to be infected with Huanglong disease, so that it is extremely problematic to obtain a large number of Huanglong disease leaf data. Meanwhile, due to the uncertainty of the pathological trait distribution of citrus yellow dragon disease leaves and the extreme shortage of data, the convolutional neural network model learned in a small number of samples is not capable of generalization. In order to improve the accuracy and generalization of Citrus Greening Disease recognition algorithm, this paper introduces Texture Reconstruction Loss CycleGAN(TRL-GAN) to generate citrus diseased leaf data in realistic scene to increase the richness of samples, and thus proposes the Recognizing Citrus Greening Based on TRL-GAN(RCG TRL-GAN). This algorithm firstly performs background culling by using the instance segmentation network Mask RCNN for realistic scenes citrus yellow dragon disease mottled, zinc deficiency, magnesium deficiency, leaf veins yellowing and other corresponding symptomatic leaves, then introduces texture reconstruction loss improvement CycleGAN as training and migrates the diseased leaf style to ordinary green leaves for data expansion, and finally uses the expanded dataset to train the convolutional neural network. Experimental results on the constructed dataset of 4516 images (762 mottled, 749 Zn deficient, 737 Mg deficient, 721 Vein yellowing, 783 Diachyma yellowing, 764 green leaves) reveal that TRL-GAN has 13.49% and 1.1% improvement in FID and KID, respectively, relative to the original structure CycleGAN, and has been identified by six citrus yellow dragon disease experts and three vision professionals identify that the fake data generated by TRL-GAN have similarity with the leaf pathological characteristics and real data, and also by using T-SNE technique it is observed that the real data have similar distribution with the generated fake data in two-dimensional plane. Meanwhile, the more outstanding accuracy performance in the classification network is ResNeXt101 with 97.45% accuracy, and the average accuracy of RCG TRL-GAN technique in the recognition of classification network is improved 2.76%. The study proves that the RCG TRL-GAN effectively improves the citrus greening disease phenotype data generation and recognition, and can provide method reference for the expansion and recognition of complex plant disease phenotype images. [ABSTRACT FROM AUTHOR]

• A deep convolutional neural network model based on dynamic time warping was proposed for the fault diagnosis. • Supersaturation control based on temperature and flow was applied. • Dynamic time warping makes the data from semi-batch process steady. • Compared to the traditional model, the new model has an outstanding performance in the fault diagnosis. • An average fault diagnosis rate of 88.6% is achieved. The abnormal conditions of the crystallization process seriously affect the crystal quality and the smooth operation of the process. Compared to the continuous steady process, it is a big challenge to realize the fault detection and diagnosis (FDD) in a batch or semi-batch crystallization process which is unsteady and nonlinear. In this paper, a coupled method combining convolutional neural network (CNN) with dynamic time warping (DTW) is proposed for FDD in semi-batch crystallization process based on temperature and flow supersaturation control (TF-SSC). DTW solves the problem that the data is unsteady in a semi-batch process. Different fault data produced by introducing disturbances are calculated through DTW to obtain the similarity which is steady. Then, the similarity of different operating states is preprocessed and classified by the CNN. Compared to the traditional CNN, Resnet18 and Inception10, DTW-CNN method has an outstanding performance in FDD, especially under a small number of samples. [ABSTRACT FROM AUTHOR]

To combat the recent coronavirus disease 2019 (COVID-19), academician and clinician are in search of new approaches to predict the COVID-19 outbreak dynamic trends that may slow down or stop the pandemic. Epidemiological models like Susceptible–Infected–Recovered (SIR) and its variants are helpful to understand the dynamics trend of pandemic that may be used in decision making to optimize possible controls from the infectious disease. But these epidemiological models based on mathematical assumptions may not predict the real pandemic situation. Recently the new machine learning approaches are being used to understand the dynamic trend of COVID-19 spread. In this paper, we designed the recurrent and convolutional neural network models: vanilla LSTM, stacked LSTM, ED_LSTM, BiLSTM, CNN, and hybrid CNN+LSTM model to capture the complex trend of COVID-19 outbreak and perform the forecasting of COVID-19 daily confirmed cases of 7, 14, 21 days for India and its four most affected states (Maharashtra, Kerala, Karnataka, and Tamil Nadu). The root mean square error (RMSE) and mean absolute percentage error (MAPE) evaluation metric are computed on the testing data to demonstrate the relative performance of these models. The results show that the stacked LSTM and hybrid CNN+LSTM models perform best relative to other models. • Various variants of LSTM and CNN deep learning architectures have been designed. • Predicted COVID-19 confirmed cases for India and its four most affected states. • RMSE and MAPE are computed to compare the performance of the designed models. • Experimental results show that predicted cases are very close to actual cases. [ABSTRACT FROM AUTHOR]

Protein amino acid sequences can be used to determine the functions of the protein. However, determining the function of a single protein requires many resources and a tremendous amount of time. Computational Intelligence methods such as Deep learning have been shown to predict the proteins' functions. This paper proposes a hybrid deep neural network model to predict an unknown protein's functions from sequences. The proposed model is named Deep_CNN_LSTM_GO. Deep_CNN_LSTM_GO is an Integration between Convolutional Neural network (CNN) and Long Short-Term Memory (LSTM) Neural Network to learn features from amino acid sequences and outputs the three different Gene Ontology (GO). The gene ontology represents the protein functions in the three sub-ontologies: Molecular Functions (MF), Biological Process (BP), and Cellular Component (CC). The proposed model has been trained and tested using UniProt-SwissProt's dataset. Another test has been done using Computational Assessment of Function Annotation (CAFA) on the three sub-ontologies. The proposed model outperforms different methods proposed in the field with better performance using three different evaluation metrics (Fmax, Smin, and AUPR) in the three sub-ontologies (MF, BP, CC). [Display omitted] [ABSTRACT FROM AUTHOR]

• Construct deep learning models for multi energy on multi simulated building type. • Long short-term memory performs the best in terms of absolute and relative errors. • Long short-term memory model predicts 50.7% of the task with CVRMSE lower than 20% • Convolutional neural network model is sensitive to the network layer number. • Two-layer structure with Huber loss is suitable for long short-term memory model. Accurate prediction of building energy consumption plays a critical role in energy sustainability. Most research focuses on single task prediction on one type of building, which makes model less robust when applied to multi energy vectors prediction on multiple building types. Meanwhile, multi vectors prediction is beneficial for detailed and precise management of energy supply in different buildings. This paper leveraged deep learning models to predict hourly-based multi energy vectors on multiple building types. The deep learning model is trained and tested on a large scale of datasets simulated by EnergyPlus over 16 DOE building archetypes in 936 cities of the US. Convolutional neural network and long short-term memory models were constructed and compared in terms of accuracy and computation efficiency under different hyperparameter designs, including layers, input steps, and loss functions. Artificial neural network is the best none-deep learning model and chosen as the baseline for deep learning models. The results also showed that long short-term memory performs the best on this multi vectors prediction for both absolute and relative errors. It predicts 50.7% of the tasks from different building types with CVRMSE lower than 20%. A two-layer structure is good enough for the long short-term memory model, and it is more stable under conditions of different input steps and layer structures. The features reduction through principal component analysis shows a more comprehensive and ample embedding method should be explored. Further research should be carried out to deal with the inconsistent performance of gas-related energy prediction. [ABSTRACT FROM AUTHOR]

• Using a CNN model to classify investor sentiments from a major Chinese stock forum. • Using a LTSM model to build a stock price prediction model. • We showed that the hybrid model with sentiments outperforms the baseline model. • The results support that investor sentiment is a driver of stock prices. Whether stock prices are predictable has been the center of debate in academia. In this paper, we propose a hybrid model that combines a deep learning approach with a sentiment analysis model for stock price prediction. We employ a Convolutional Neural Network model for classifying the investors' hidden sentiments, which are extracted from a major stock forum. We then propose a hybrid research model by applying the Long Short-Term Memory (LSTM) Neural Network approach for analyzing the technical indicators from the stock market and the sentiment analysis results from the first step. Furthermore, this work has conducted real-life experiments from six key industries of three time intervals on the Shanghai Stock Exchange (SSE) to validate the effectiveness and applicability of the proposed model. The experiment results indicate that the proposed model has achieved better performance in classifying investor sentiments than the baseline classifiers, and this hybrid approach performs better in predicting stock prices compared to the single model and the models without sentiment analysis. [ABSTRACT FROM AUTHOR]