Search

Facial emotion recognition (FER) is a computer vision process aimed at detecting and classifying human emotional expressions. FER systems are currently used in a vast range of applications from areas such as education, healthcare, or public safety; therefore, detection and recognition accuracies are very important. Similar to any computer vision task based on image analyses, FER solutions are also suitable for integration with artificial intelligence solutions represented by different neural network varieties, especially deep neural networks that have shown great potential in the last years due to their feature extraction capabilities and computational efficiency over large datasets. In this context, this paper reviews the latest developments in the FER area, with a focus on recent neural network models that implement specific facial image analysis algorithms to detect and recognize facial emotions. This paper's scope is to present from historical and conceptual perspectives the evolution of the neural network architectures that proved significant results in the FER area. This paper endorses convolutional neural network (CNN)-based architectures against other neural network architectures, such as recurrent neural networks or generative adversarial networks, highlighting the key elements and performance of each architecture, and the advantages and limitations of the proposed models in the analyzed papers. Additionally, this paper presents the available datasets that are currently used for emotion recognition from facial expressions and micro-expressions. The usage of FER systems is also highlighted in various domains such as healthcare, education, security, or social IoT. Finally, open issues and future possible developments in the FER area are identified. [ABSTRACT FROM AUTHOR]

Catastrophic forgetting, which means a rapid forgetting of learned representations while learning new data/samples, is one of the main problems of deep neural networks. In this paper, we propose a novel incremental learning framework that can address the forgetting problem by learning new incoming data in an online manner. We develop a new incremental learning framework that can learn extra data or new classes with less catastrophic forgetting. We adopt the hippocampal memory process to the deep neural networks by defining the effective maximum of neural activation and its boundary to represent a feature distribution. In addition, we incorporate incremental QR factorization into the deep neural networks to learn new data with both existing labels and new labels with less forgetting. The QR factorization can provide the accurate subspace prior, and incremental QR factorization can reasonably express the collaboration between new data with both existing classes and new class with less forgetting. In our framework, a set of appropriate features (i.e., nodes) provides improved representation for each class. We apply our method to the convolutional neural network (CNN) for learning Cifar-100 and Cifar-10 datasets. The experimental results show that the proposed method efficiently alleviates the stability and plasticity dilemma in the deep neural networks by providing the performance stability of a trained network while effectively learning unseen data and additional new classes. [ABSTRACT FROM AUTHOR]

In this study, we propose a new model for optical character recognition (OCR) based on both CNNs (convolutional neural networks) and RNNs (recurrent neural networks). The distortions affecting the document image can take different forms, such as blur (focus blur, motion blur, etc.), shadow, bad contrast, etc. Document-image distortions significantly decrease the performance of OCR systems, to the extent that they reach a performance close to zero. Therefore, a robust OCR model that performs robustly even under hard (distortion) conditions is still sorely needed. However, our comprehensive study in this paper shows that various related works can somewhat improve their respective OCR recognition performance of degraded document images (e.g., captured by smartphone cameras under different conditions and, thus, distorted by shadows, contrast, blur, etc.), but it is worth underscoring, that improved recognition is neither sufficient nor always satisfactory—especially in very harsh conditions. Therefore, in this paper, we suggest and develop a much better and fully different approach and model architecture, which significantly outperforms the aforementioned previous related works. Furthermore, a new dataset was gathered to show a series of different and well-representative real-world scenarios of hard distortion conditions. The new OCR model suggested performs in such a way that even document images (even from the hardest conditions) that were previously not recognizable by other OCR systems can be fully recognized with up to 97.5% accuracy/precision by our new deep-learning-based OCR model. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) is an important research problem in computer vision. This problem is widely applied to building applications in human–machine interactions, monitoring, etc. Especially, HAR based on the human skeleton creates intuitive applications. Therefore, determining the current results of these studies is very important in selecting solutions and developing commercial products. In this paper, we perform a full survey on using deep learning to recognize human activity based on three-dimensional (3D) human skeleton data as input. Our research is based on four types of deep learning networks for activity recognition based on extracted feature vectors: Recurrent Neural Network (RNN) using extracted activity sequence features; Convolutional Neural Network (CNN) uses feature vectors extracted based on the projection of the skeleton into the image space; Graph Convolution Network (GCN) uses features extracted from the skeleton graph and the temporal–spatial function of the skeleton; Hybrid Deep Neural Network (Hybrid–DNN) uses many other types of features in combination. Our survey research is fully implemented from models, databases, metrics, and results from 2019 to March 2023, and they are presented in ascending order of time. In particular, we also carried out a comparative study on HAR based on a 3D human skeleton on the KLHA3D 102 and KLYOGA3D datasets. At the same time, we performed analysis and discussed the obtained results when applying CNN-based, GCN-based, and Hybrid–DNN-based deep learning networks. [ABSTRACT FROM AUTHOR]

Smart grids are able to forecast customers' consumption patterns, i.e., their energy demand, and consequently electricity can be transmitted after taking into account the expected demand. To face today's demand forecasting challenges, where the data generated by smart grids is huge, modern data-driven techniques need to be used. In this scenario, Deep Learning models are a good alternative to learn patterns from customer data and then forecast demand for different forecasting horizons. Among the commonly used Artificial Neural Networks, Long Short-Term Memory networks—based on Recurrent Neural Networks—are playing a prominent role. This paper provides an insight into the importance of the demand forecasting issue, and other related factors, in the context of smart grids, and collects some experiences of the use of Deep Learning techniques, for demand forecasting purposes. To have an efficient power system, a balance between supply and demand is necessary. Therefore, industry stakeholders and researchers should make a special effort in load forecasting, especially in the short term, which is critical for demand response. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel technique for the inspection of high-density polyethylene (HDPE) pipes using ultrasonic sensors, signal processing, and deep neural networks (DNNs). Specifically, we propose a technique that detects whether there is a diversion on a pipe or not. The proposed model transmits ultrasound signals through a pipe using a custom-designed array of piezoelectric transmitters and receivers. We propose to use the Zadoff–Chu sequence to modulate the input signals, then utilize its correlation properties to estimate the pipe channel response. The processed signal is then fed to a DNN that extracts the features and decides whether there is a diversion or not. The proposed technique demonstrates an average classification accuracy of 90.3 % (when one sensor is used) and 99.6 % (when two sensors are used) on 3 4 inch pipes. The technique can be readily generalized for pipes of different diameters and materials. [ABSTRACT FROM AUTHOR]

This paper focuses on the classification of seven locomotion modes (sitting, standing, level ground walking, ramp ascent and descent, stair ascent and descent), the transitions among these modes, and the gait phases within each mode, by only using data in the frequency domain from one or two inertial measurement units. Different deep neural network configurations are investigated and compared by combining convolutional and recurrent layers. The results show that a system composed of a convolutional neural network followed by a long short-term memory network is able to classify with a mean F 1 -score of 0.89 and 0.91 for ten healthy subjects, and of 0.92 and 0.95 for one osseointegrated transfemoral amputee subject (excluding the gait phases because they are not labeled in the data-set), using one and two inertial measurement units, respectively, with a 5-fold cross-validation. The promising results obtained in this study pave the way for using deep learning for the control of transfemoral prostheses with a minimum number of inertial measurement units. [ABSTRACT FROM AUTHOR]

The complexity of polyphonic sounds imposes numerous challenges on their classification. Especially in real life, polyphonic sound events have discontinuity and unstable time-frequency variations. Traditional single acoustic features cannot characterize the key feature information of the polyphonic sound event, and this deficiency results in poor model classification performance. In this paper, we propose a convolutional recurrent neural network model based on the temporal-frequency (TF) attention mechanism and feature space (FS) attention mechanism (TFFS-CRNN). The TFFS-CRNN model aggregates Log-Mel spectrograms and MFCCs feature as inputs, which contains the TF-attention module, the convolutional recurrent neural network (CRNN) module, the FS-attention module and the bidirectional gated recurrent unit (BGRU) module. In polyphonic sound events detection (SED), the TF-attention module can capture the critical temporal–frequency features more capably. The FS-attention module assigns different dynamically learnable weights to different dimensions of features. The TFFS-CRNN model improves the characterization of features for key feature information in polyphonic SED. By using two attention modules, the model can focus on semantically relevant time frames, key frequency bands, and important feature spaces. Finally, the BGRU module learns contextual information. The experiments were conducted on the DCASE 2016 Task3 dataset and the DCASE 2017 Task3 dataset. Experimental results show that the F1-score of the TFFS-CRNN model improved 12.4% and 25.2% compared with winning system models in DCASE challenge; the ER is reduced by 0.41 and 0.37 as well. The proposed TFFS-CRNN model algorithm has better classification performance and lower ER in polyphonic SED. [ABSTRACT FROM AUTHOR]

As artificial neural network architectures grow increasingly more efficient in time-series prediction tasks, their use for day-ahead electricity price and demand prediction, a task with very specific rules and highly volatile dataset values, grows more attractive. Without a standardized way to compare the efficiency of algorithms and methods for forecasting electricity metrics, it is hard to have a good sense of the strengths and weaknesses of each approach. In this paper, we create models in several neural network architectures for predicting the electricity price on the HUPX market and electricity load in Montenegro and compare them to multiple neural network models on the same basis (using the same dataset and metrics). The results show the promising efficiency of neural networks in general for the task of short-term prediction in the field, with methods combining fully connected layers and recurrent neural or temporal convolutional layers performing the best. The feature extraction power of convolutional layers shows very promising results and recommends the further exploration of temporal convolutional networks in the field. [ABSTRACT FROM AUTHOR]

Traditional time series forecasting techniques can not extract good enough sequence data features, and their accuracies are limited. The deep learning structure SeriesNet is an advanced method, which adopts hybrid neural networks, including dilated causal convolutional neural network (DC-CNN) and Long-short term memory recurrent neural network (LSTM-RNN), to learn multi-range and multi-level features from multi-conditional time series with higher accuracy. However, they didn't consider the attention mechanisms to learn temporal features. Besides, the conditioning method for CNN and RNN is not specific, and the number of parameters in each layer is tremendous. This paper proposes the conditioning method for two types of neural networks, and respectively uses the gated recurrent unit network (GRU) and the dilated depthwise separable temporal convolutional networks (DDSTCNs) instead of LSTM and DC-CNN for reducing the parameters. Furthermore, this paper presents the lightweight RNN-based hidden state attention module (HSAM) combined with the proposed CNN-based convolutional block attention module (CBAM) for time series forecasting. Experimental results show our model is superior to other models from the viewpoint of forecasting accuracy and computation efficiency. [ABSTRACT FROM AUTHOR]

Hospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work using a Gaussian blur and smoothing–convolutional neural network combined with a recurrent neural network (GBS-CR) method. Gaussian blur and smoothing (GBS) is a novel preprocessing method that can fix the breakpoint font in medical invoices. The combination of convolutional neural network (CNN) and recurrent neural network (RNN) was used to raise the recognition rate of the breakpoint font in medical invoices. RNN was designed to be the semantic revision module. In the aspect of image preprocessing, Gaussian blur and smoothing were used to fix the breakpoint font. In the period of making the self-built dataset, a certain proportion of the breakpoint font (the font of breakpoint is 3, the original font is 7) was added, in this paper, so as to optimize the Alexnet–Adam–CNN (AA-CNN) model, which is more suitable for the recognition of the breakpoint font than the traditional CNN model. In terms of the identification methods, we not only adopted the optimized AA-CNN for identification, but also combined RNN to carry out the semantic revisions of the identified results of CNN, meanwhile further improving the recognition rate of the medical invoices. The experimental results show that compared with the state-of-art invoice recognition method, the method presented in this paper has an average increase of 10 to 15 percentage points in recognition rate. [ABSTRACT FROM AUTHOR]

Landslides pose a constant threat to the lives and property of mountain people and may also cause geomorphological destruction such as soil and water loss, vegetation destruction, and land cover change. Landslide susceptibility assessment (LSA) is a key component of landslide risk evaluation. There are many related studies, but few analyses and comparisons of models for optimization. This paper aims to introduce the Tree-structured Parzen Estimator (TPE) algorithm for hyperparameter optimization of three typical neural network models for LSA in Shuicheng County, China, as an example, and to compare the differences of predictive ability among the models in order to achieve higher application performance. First, 17 influencing factors of landslide multiple data sources were selected for spatial prediction, hybrid ensemble oversampling and undersampling techniques were used to address the imbalanced sample and small sample size problem, and the samples were randomly divided into a training set and validation set. Second, deep neural network (DNN), recurrent neural network (RNN), and convolutional neural network (CNN) models were adopted to predict the regional landslides susceptibility, and the TPE algorithm was used to optimize the hyperparameters respectively to improve the assessment capacity. Finally, to compare the differences and optimization effects of these models, several objective measures were applied for validation. The results show that the high-susceptibility regions mostly distributed in bands along fault zones, where the lithology is mostly claystone, sandstone, and basalt. The DNN, RNN, and CNN models all perform well in LSA, especially the RNN model. The TPE optimization significantly improves the accuracy of the DNN and CNN (3.92% and 1.52%, respectively), but does not improve the performance of the RNN. In summary, our proposed RNN model and TPE-optimized DNN and CNN model have robust predictive capability for landslide susceptibility in the study area and can also be applied to other areas containing similar geological conditions. [ABSTRACT FROM AUTHOR]

Sound event detection (SED) recognizes the corresponding sound event of an incoming signal and estimates its temporal boundary. Although SED has been recently developed and used in various fields, achieving noise-robust SED in a real environment is typically challenging owing to the performance degradation due to ambient noise. In this paper, we propose combining a pretrained time-domain speech-separation-based noise suppression network (NS) and a pretrained classification network to improve the SED performance in real noisy environments. We use group communication with a context codec method (GC3)-equipped temporal convolutional network (TCN) for the noise suppression model and a convolutional recurrent neural network for the SED model. The former significantly reduce the model complexity while maintaining the same TCN module and performance as a fully convolutional time-domain audio separation network (Conv-TasNet). We also do not update the weights of some layers (i.e., freeze) in the joint fine-tuning process and add an attention module in the SED model to further improve the performance and prevent overfitting. We evaluate our proposed method using both simulation and real recorded datasets. The experimental results show that our method improves the classification performance in a noisy environment under various signal-to-noise-ratio conditions. [ABSTRACT FROM AUTHOR]

Multiple types of sleep arousal account for a large proportion of the causes of sleep disorders. The detection of sleep arousals is very important for diagnosing sleep disorders and reducing the risk of further complications including heart disease and cognitive impairment. Sleep arousal scoring is manually completed by sleep experts by checking the recordings of several periods of sleep polysomnography (PSG), which is a time-consuming and tedious work. Therefore, the development of efficient, fast, and reliable automatic sleep arousal detection system from PSG may provide powerful help for clinicians. This paper reviews the automatic arousal detection methods in recent years, which are based on statistical rules and deep learning methods. For statistical detection methods, three important processes are typically involved, including preprocessing, feature extraction and classifier selection. For deep learning methods, different models are discussed by now, including convolution neural network (CNN), recurrent neural network (RNN), long-term and short-term memory neural network (LSTM), residual neural network (ResNet), and the combinations of these neural networks. The prediction results of these neural network models are close to the judgments of human experts, and these methods have shown robust generalization capabilities on different data sets. Therefore, we conclude that the deep neural network will be the main research method of automatic arousal detection in the future. [ABSTRACT FROM AUTHOR]

Photovoltaic (PV) power forecasting urges in economic and secure operations of power systems. To avoid an inaccurate individual forecasting model, we propose an approach for a one-day to three-day ahead PV power hourly forecasting based on the stacking ensemble model with a recurrent neural network (RNN) as a meta-learner. The proposed approach is built by using real weather data and forecasted weather data in the training and testing stages, respectively. To accommodate uncertain weather, a daily clustering method based on statistical features, e.g., daily average, maximum, and standard deviation of PV power is applied in the data sets. Historical PV power output and weather data are used to train and test the model. The single learner considered in this research are artificial neural network, deep neural network, support vector regressions, long short-term memory, and convolutional neural network. Then, RNN is used to combine the forecasting results of each single learner. It is also important to observe the best combination of the single learners in this paper. Furthermore, to compare the performance of the proposed method, a random forest ensemble instead of RNN is used as a benchmark for comparison. Mean relative error (MRE) and mean absolute error (MAE) are used as criteria to validate the accuracy of different forecasting models. The MRE of the proposed RNN ensemble learner model is 4.29%, which has significant improvements by about 7–40%, 7–30%, and 8% compared to the single models, the combinations of fewer single learners, and the benchmark method, respectively. The results show that the proposed method is promising for use in real PV power forecasting systems. [ABSTRACT FROM AUTHOR]

In this paper, we address various challenges in multi-pedestrian and vehicle tracking in high-resolution aerial imagery by intensive evaluation of a number of traditional and Deep Learning based Single- and Multi-Object Tracking methods. We also describe our proposed Deep Learning based Multi-Object Tracking method AerialMPTNet that fuses appearance, temporal, and graphical information using a Siamese Neural Network, a Long Short-Term Memory, and a Graph Convolutional Neural Network module for more accurate and stable tracking. Moreover, we investigate the influence of the Squeeze-and-Excitation layers and Online Hard Example Mining on the performance of AerialMPTNet. To the best of our knowledge, we are the first to use these two for regression-based Multi-Object Tracking. Additionally, we studied and compared the L 1 and Huber loss functions. In our experiments, we extensively evaluate AerialMPTNet on three aerial Multi-Object Tracking datasets, namely AerialMPT and KIT AIS pedestrian and vehicle datasets. Qualitative and quantitative results show that AerialMPTNet outperforms all previous methods for the pedestrian datasets and achieves competitive results for the vehicle dataset. In addition, Long Short-Term Memory and Graph Convolutional Neural Network modules enhance the tracking performance. Moreover, using Squeeze-and-Excitation and Online Hard Example Mining significantly helps for some cases while degrades the results for other cases. In addition, according to the results, L 1 yields better results with respect to Huber loss for most of the scenarios. The presented results provide a deep insight into challenges and opportunities of the aerial Multi-Object Tracking domain, paving the way for future research. [ABSTRACT FROM AUTHOR]

El Niño is an important quasi-cyclical climate phenomenon that can have a significant impact on ecosystems and societies. Due to the chaotic nature of the atmosphere and ocean systems, traditional methods (such as statistical methods) are difficult to provide accurate El Niño index predictions. The latest research shows that Ensemble Empirical Mode Decomposition (EEMD) is suitable for analyzing non-linear and non-stationary signal sequences, Convolutional Neural Network (CNN) is good at local feature extraction, and Recurrent Neural Network (RNN) can capture the overall information of the sequence. As a special RNN, Long Short-Term Memory (LSTM) has significant advantages in processing and predicting long, complex time series. In this paper, to predict the El Niño index more accurately, we propose a new hybrid neural network model, EEMD-CNN-LSTM, which combines EEMD, CNN, and LSTM. In this hybrid model, the original El Niño index sequence is first decomposed into several Intrinsic Mode Functions (IMFs) using the EEMD method. Next, we filter the IMFs by setting a threshold, and we use the filtered IMFs to reconstruct the new El Niño data. The reconstructed time series then serves as input data for CNN and LSTM. The above data preprocessing method, which first decomposes the time series and then reconstructs the time series, uses the idea of symmetry. With this symmetric operation, we extract valid information about the time series and then make predictions based on the reconstructed time series. To evaluate the performance of the EEMD-CNN-LSTM model, the proposed model is compared with four methods including the traditional statistical model, machine learning model, and other deep neural network models. The experimental results show that the prediction results of EEMD-CNN-LSTM are not only more accurate but also more stable and reliable than the general neural network model. [ABSTRACT FROM AUTHOR]

Personnel performance is important for the high-technology industry to ensure its core competitive advantages are present. Therefore, predicting personnel performance is an important research area in human resource management (HRM). In this paper, to improve prediction performance, we propose a novel framework for personnel performance prediction to help decision-makers to forecast future personnel performance and recruit the best suitable talents. Firstly, a hybrid convolutional recurrent neural network (CRNN) model based on self-attention mechanism is presented, which can automatically learn discriminative features and capture global contextual information from personnel performance data. Moreover, we treat the prediction problem as a classification task. Then, the k-nearest neighbor (KNN) classifier was used to predict personnel performance. The proposed framework is applied to a real case of personnel performance prediction. The experimental results demonstrate that the presented approach achieves significant performance improvement for personnel performance compared to existing methods. [ABSTRACT FROM AUTHOR]

Domain generation algorithms (DGAs) represent a class of malware used to generate large numbers of new domain names to achieve command-and-control (C2) communication between the malware program and its C2 server to avoid detection by cybersecurity measures. Deep learning has proven successful in serving as a mechanism to implement real-time DGA detection, specifically through the use of recurrent neural networks (RNNs) and convolutional neural networks (CNNs). This paper compares several state-of-the-art deep-learning implementations of DGA detection found in the literature with two novel models: a deeper CNN model and a one-dimensional (1D) Capsule Networks (CapsNet) model. The comparison shows that the 1D CapsNet model performs as well as the best-performing model from the literature. [ABSTRACT FROM AUTHOR]

This survey paper describes a literature review of deep learning (DL) methods for cyber security applications. A short tutorial-style description of each DL method is provided, including deep autoencoders, restricted Boltzmann machines, recurrent neural networks, generative adversarial networks, and several others. Then we discuss how each of the DL methods is used for security applications. We cover a broad array of attack types including malware, spam, insider threats, network intrusions, false data injection, and malicious domain names used by botnets. [ABSTRACT FROM AUTHOR]