Search

Accurate diagnosis of Lithium-ion batteries (Li-ion batteries) degradation plays a critical role in improving the maintenance of energy storage technology. This paper presents a method based on a novel deep network model combined with a data transformation technique to diagnose Li-ion battery degradation modes. Different from conventional studies based on specific experimental and numerical methods to estimate and predict the degradation, the proposed method is based on data-driven approach, by leveraging datasets consisting of voltage/capacity curves, these were converted into incremental capacity (IC) curves and then transformed into images using the gramian angular summation field (GASF) technique. The study adopted two models: Inception-v3 and the proposed model, both underwent fine-tuning and a subsequent transfer learning process. Degradation modes, namely loss of lithium inventory (LLI) and the loss of active materials in both the positive (LAMPE) and negative electrodes (LAMNE), were diagnosed in relation to IC curves. Finally, the model was tested using two different datasets, and the results showed that the proposed method achieved high performance, especially across three Li-ion batteries, three degradation modes, three cells, and various cycles (totaling 378 cases) the proposed method outperformed in 233 cases, thereby outperforming other methods in comparison. Our method provides a flexible data-driven approach that accurately predicts various degradation modes across different cell chemistries throughout their lifespan. [ABSTRACT FROM AUTHOR]

Distributed generation sources (e.g. photovoltaic, wind power) are raised to fit the energy demands recently in proficient way. Further, adoption of renewable energy sources to the smart grid causes severe issues in quality of power signals. These problems can be overcome with correct identification and classification of power quality disturbances (PQDs) efficiently by the both producer and consumer. In this work, a new method for classification of PQDs based on signal processing technique and deep convolutional neural network (DCNN) is proposed. The proposed method compress the raw data of PQDs by using discrete wavelet transform (DWT) and multi-resolution analysis (MRA). Moreover, CNN has the ability to automatic feature selection and classification in a single stage instead of conventional methods i.e. artificial neural networks (ANNs). In this paper, DWT 1D-CNN is designed to extract and optimize multi-scale characteristics of disturbances. Further, 1-D convolutional, pooling and batch-normalization layers are added to facilitate normalize data, better generalization ability and, effective classification process against noisy data. A typical 11Kv distribution network is simulated in MATLAB to prove the validity of the proposed method in an essential part of smart grid. [ABSTRACT FROM AUTHOR]

This paper presents an effective and accurate method to analyze the network traffic risk of enterprise public cloud financial system by using deep learning algorithm. To build the analytical model, the collected data is preprocessed to extract relevant features that contribute to risk analysis. The pre-processed data is used to train the deep learning model. The model is designed to learn patterns and relationships in network traffic data. The network traffic risk analysis module is constructed by combining the deep learning model with the system architecture of the enterprise public cloud financial system. Through a trained deep learning model, the module is able to accurately identify abnormal behavior in network traffic and provide corresponding risk assessments. The results show that the proposed system module can accurately identify abnormal behavior in network traffic and provide risk assessment. [ABSTRACT FROM AUTHOR]

The rapid growth of online music education has led to increased security risks from cyber intrusions. This paper proposes public cloud networks oriented deep neural networks for effective intrusion detection in online music education environments. Specifically, a novel intrusion detection framework is developed, comprising fuzzy logic based feature selection, chronological salp swarm algorithm optimized deep belief networks, and gated recurrent unit integrated convolutional neural networks. Detailed methodologies are presented for the fuzzy logic system, chronological salp optimization, deep belief network architecture, and convolutional neural networks. Comprehensive experiments are conducted on the NSL-KDD and CICIDS2017 datasets. Various deep neural networks are evaluated and compared, including multi-layer perceptrons, convolutional neural networks, deep belief networks, recurrent neural networks, and the proposed models. Experimental results demonstrate that the proposed fuzzy feature selection and chronological salp swarm algorithm optimized deep belief network achieves a test accuracy of 97.33 %, outperforming other peer models. The gated recurrent unit integrated convolutional neural network obtains a test accuracy of 98.46 %, superior to state-of-the-art methods. While the experiments on the newly created dataset for intrusion detection in cloud-based online music education demonstrate that the proposed models outperform the benchmarks. The experiments verify the effectiveness of the proposed deep learning frameworks for intrusion detection in online music education cloud networks. [ABSTRACT FROM AUTHOR]

• Interpretable DL Model: Our unique deep learning model interprets modulation mechanisms, efficiently extracting time and phase features with minimal resources. • Efficient Performance: Our approach achieves top performance on RML2016 and RML2018 datasets with fewer model parameters. • Fastest Embedded Speed: Outperforming 8 DL-based methods, our approach demonstrates the fastest speed on Jetson Nano and Raspberry Pi systems. Automatic Modulation Classification (AMC) constitutes a fundamental technology for enabling automatic demodulation in Cognitive Communication Systems (CCS). Due to the size, weight, and power (SWaP) constraints of embedded computers employed in CCS, there are limited computational and memory resources. While deep neural networks possess strong feature representation and high accuracy recognition capabilities, they usually come with a high number of network parameters and high computational complexity, thereby reducing the real-time processing ability of CCS. Therefore, neural network structures intended for CCS must be lightweight and computationally efficient. In this paper, we propose a high-performance and resource-friendly network model based on an analysis of the modulation mechanism of communication signals. The network extracts phase features and short-time features sequentially using directional convolutional filters. Long short-term memory (LSTM) units are then used to extract long-term features, and only one fully connected layer is used for classification. Experiments with a standard dataset consisting of 11 communication modulation types demonstrate that our proposed model achieves an accuracy greater than 84.5%, even when the signal-to-noise ratio (SNR) is 0 dB, and the model has only 29187 parameters. On a Jetson Nano embedded platform, the model achieves a processing speed of up to 375366 in-phase and quadrature samples/s. Overall, the results suggest that our proposed approach is both lightweight and highly efficient, making it more suitable for CCS applications. [Display omitted] [ABSTRACT FROM AUTHOR]

• We proposed a 9-layer deep convolutional neural network for leaf disease identification. • We used six different types of image augmentation methods, and compared their validation performances using data augmentation and not using data augmentation. • Fine-tuning the batch size, dropout and epoch improves the performance of our model. • We validated that the proposed model performs better than transfer learning techniques. • Our model produced an average testing accuracy of 96.46%, which was better than the state-of-the-art machine learning approaches. In this paper, we proposed a novel plant leaf disease identification model based on a deep convolutional neural network (Deep CNN). The Deep CNN model is trained using an open dataset with 39 different classes of plant leaves and background images. Six types of data augmentation methods were used: image flipping, gamma correction, noise injection, principal component analysis (PCA) colour augmentation, rotation, and scaling. We observed that using data augmentation can increase the performance of the model. The proposed model was trained using different training epochs, batch sizes and dropouts. Compared with popular transfer learning approaches, the proposed model achieves better performance when using the validation data. After an extensive simulation, the proposed model achieves 96.46% classification accuracy. This accuracy of the proposed work is greater than the accuracy of traditional machine learning approaches. The proposed model is also tested with respect to its consistency and reliability. [ABSTRACT FROM AUTHOR]

Relocation is one of the most common problems in Simultaneous Localization and Mapping (SLAM). This paper presents a novel relocation method, using unsupervised deep learning algorithm to extract the feature of Light Detection and Ranging (LiDAR) data, and narrows the scope of relocation by classifying these features to reduce the randomness of the relocation. Compared with the other methods which is based on matching the manual feature points, this method avoids some limitations of manual features. We modify the Particle Filter SLAM (PF-SLAM), and use our relocation method to replace the original method for experimentation. The experimental results demonstrate that this method can be relocation whit high success rate only use a small amount of computational resource in a short time. Training neural network will consume a lot of computing resources, we also propose a cloud computing framework to the implementation of this method for the mobile robot which computational resources are limited. [ABSTRACT FROM AUTHOR]

• One dimensional emotion recognition method is proposed to realize the dimensional emotion recognition by combing the attention mechanism and global second-order deep representations. • The residual attention network is used to extract emotionally relevant features to solve the negative influence of emotionally irrelevant features on the recognition performance. • The global second-order pooling network is integrated to learn the interdependence relationship between the long-distance features, increasing the nonlinear emotion representation learning ability of the trained model. • The experiments on the Affectnet dataset demonstrate the validity of the proposed model which is superior to some existing methods. For the emotion recognition tasks, the emotion-related features and the emotionally irrelevant ones are fully extracted by the deep neural network is particularly desirable. In this paper, a dimensional emotion recognition algorithm is proposed by combing the attention mechanism and global second-order feature representations. More specially, the residual attention network (RAN) is utilized to select the features related to the task, and then the global second-order pooling network is employed to model the correlation between these features to complete the modeling of the feature interdependence relationship, improving the deficiency of the RAN in this point so as to enhance the capacity of extracting the emotion-related features and the emotionally irrelevant ones. The experiments on the AffectNet dataset demonstrate the validity of the proposed model, which is comparable to and even superior to those from the support vector regression method, the convolutional neural network (CNN)-based method, and the recently-developed CNN-based variants. [Display omitted] [ABSTRACT FROM AUTHOR]

All witnessed the terrible effects of the COVID-19 pandemic on the health and work lives of the population across the world. It is hard to diagnose all infected people in real time since the conventional medical diagnosis of COVID-19 patients takes a couple of days for accurate diagnosis results. In this paper, a novel learning framework is proposed for the early diagnosis of COVID-19 patients using hybrid deep fusion learning models. The proposed framework performs early classification of patients based on collected samples of chest X-ray images and Coswara cough (sound) samples of possibly infected people. The captured cough samples are pre-processed using speech signal processing techniques and Mel frequency cepstral coefficient features are extracted using deep convolutional neural networks. Finally, the proposed system fuses extracted features to provide 98.70% and 82.7% based on Chest-X ray images and cough (audio) samples for early diagnosis using the weighted sum-rule fusion method. [ABSTRACT FROM AUTHOR]

Object detection algorithms based on deep learning have been widely used in the intelligent detection of prohibited items in X-ray images during the security inspection process. However, security inspection is a process that emphasizes timeliness, and the low inference efficiency brought by overly complex deep neural networks is not allowed. In this paper, we propose a lightweight object detection framework based on the YOLOv4 algorithm, named LightRay. Where MobileNetV3 is applied as the feature extraction backbone. To effectively deal with the detection problem of prohibited items with small sizes in complex backgrounds, a feature enhancement network in shallow layers is proposed. Where, the Lightweight Feature Pyramid Network (LFPN) and the Convolutional Block Attention Module (CBAM) effectively achieve feature fusion of different scales, while strengthening the features of small objects. The experimental results show that the mAP of the lightRay model is 87.28% on the SIXray data set, while the FLOPs of the model are reduced to the original 1/5, and Params is reduced to 1/3 of the original. In addition, some ablation experiments confirm the ability of the LightRay model the detection of prohibited items with small sizes. The codes of LightRay are released at https://github.com/zhg-SZPT/LightRay. [ABSTRACT FROM AUTHOR]

• A faster texture based feature vector and a deep learning based method for skin cancer classification. • Use of a bigger neighbourhood (7 × 7) yields a more powerful feature vector with less noise. Thereby the results are better as compared to recent related texture based algorithms. • Hybrid architecture with use of feature vectors along with a deep neural network reduce the time drastically as compared to the existing deep learning based models. • An average increase in accuracy for two publicly available datasets is more than 10%. Skin Cancer is one of the most widespread forms of cancer in the world which can be detected using dermatoscopic images. In this paper, a texture based feature extraction algorithm is presented for the classification of dermatoscopic images. A median based Local Ternary Pattern is extracted followed by the computation of local quantized ternary patterns. The feature set extracted is then classified using a modified convolutional neural network. The images used for the detection of multiple types of skin cancer are obtained from two publicly available datasets, HAM10000 and ISICUDA11. For the proposed technique, the average recall value, average precision and average accuracy is found to be 75.20%, 95.44% and 96% respectively. An average increase in accuracy for the proposed algorithm is up-to 50.6%, 24.1% and 4.7% over LTP, DLTerQEP and a DE ANN based algorithm respectively. [Display omitted] [ABSTRACT FROM AUTHOR]

The paper investigates a noise-robust and bandwidth-efficient audio event detection technique. In this work, audio features are extracted from audio data using a gammatone filter bank. The proposed denoising autoencoder-based fully connected deep neural network scheme is realized in a distributed architecture. Audio features are extracted using a Raspberry Pi edge computing device and those are fed to a local data server to detect audio events. Experiments have shown that feature extraction and event classification tasks using the proposed technique take 840 ms for a 2-second audio clip that promises real-time implementation. The prototype model shows improved classification probability of noisy audio events with an accuracy of 98.12% and an F1-score of 90.6% at a 10 dB signal-to-noise-ratio. The performance of the proposed technique is compared with state-of-the-art techniques using the Freiburg-106 dataset and improved noise-robust audio event detection performance is found. [ABSTRACT FROM AUTHOR]

In this paper, we show a deep learning model for bot detection, named as DeBot, for industrial network traffic. DeBot uses a novel Cascade Forward Back Propagation Neural Network (CFBPNN) model with a subset of features using the Correlation-based Feature Selection (CFS) technique. A time series-based Nonlinear Auto-regressive Network with eXogenous inputs (NARX) technique analyzes the factors having a higher impact on the target variable and predict the behavioral pattern. To the best of our knowledge, we pioneer the use of optimal feature selection and integration with the cascading model of deep learning in bot detection of IIoTs. We conduct a thorough set of experiments on five popular bot datasets: NF-UNSW-NB15, NF-ToN-IoT, NF-BoT-IoT, NF-CSE-CIC-IDS2018, and ToN-IoT-Windows. We compare CFBPNN with other existing neural network models. We observe that CFBPNN in DeBot shows 100% accuracy in all the datasets with subset evaluation and obtains optimum F1-score and zero precision. [Display omitted] • We address the multi-layered network model with IIoT features to enable bot attack detection. • Our DeBot model evaluates the subset with optimal features with high impact on attack class and predict the risk factor. • We pioneer the implementation of time series-based Nonlinear Auto-Regressive eXogenous (NARX) model to predict the risk in the specified time gap. • We introduce a new cascading model with recurrent feature to preserve the relationship between input and output variables. [ABSTRACT FROM AUTHOR]

Effective and timely fault detection and status monitoring of the industrial production process is essential to fully guarantee the operational safety. However, massive multi-source heterogeneous data analysis is facing many challenges. This paper proposes a process monitoring model combined with a parallel deep learning algorithm and Principal Component Analysis (PCA) method. Firstly, PCA is applied to realize fault diagnosis and extract fault characteristic variables, thus abnormal conditions can be revealed. In order to reduce the complexity of processing massive data, a parallel deep learning framework consisting of multi-models of convolutional neural network (CNN) and Long-Short Term Memory (LSTM) neural network is proposed to effectively predict the target variable status. Comprehensive experiments are taken under two real-system scenarios, and comparisons are made against four traditional neural network models to demonstrate its practicability and effectiveness. The generated results clearly show that the CNN-LSTM parallel model combined with PCA outperform other popular models due to its merged advantages of accurate time series prediction and effective fault feature extraction. [Display omitted] • A parallel deep learning approach with feature extraction and long time series prediction. • Multi-source unsupervised faulty variable extraction and detection can be realized. • Successfully solve the distribution power grid fault diagnosis in multiple locations. [ABSTRACT FROM AUTHOR]

In this paper, a new hybrid method of change detection and classification is proposed for precise detection and classification of series arc faults (SAFs) in photovoltaic systems. An artificial neural network (ANN) structure is applied for change detection at the first stage, which is then incorporated together with four different convolutional neural network (CNN) models with various dimensions as classifiers for the discrimination of SAFs at the second stage. The models used in the proposed method are 1D CNN, 2D CNN, 3D CNN, and 2D-based images. A comparison of the proposed approach and the state-of-the-art methods has been carried out in terms of accuracy and computational complexity. For a thorough evaluation of the proposed method's performance, studies have been conducted in both simulation and practice, considering various possible scenarios which may emerge. To such an aim, alongside the records from actual measurements in practice, nine models of SAF are also employed for simulation. The results show that the proposed method satisfies principle criteria such as reliability, fault classification error, overfitting, and vanishing solutions. [ABSTRACT FROM AUTHOR]

As an essential subproblem in the face recognition field, occluded face recognition has received considerable attention in recent years. However, satisfactory recognition accuracy for occluded faces has not been achieved. Therefore, in order to solve this problem, this paper proposes a convolutional neural network that has multidimensional serial feature extraction modules for occluded faces and uses the deep learning method to improve the recognition rate. In order to improve the expression ability of the network, the method extracts features from two dimensions — space and channel — and built a multidimensional serial feature extraction module. By "multiscale" and "dependence" processing features, the generalization ability was significantly improved. Through the experimental test, the recognition accuracy of the multidimensional feature network (MFNet) reached 90.35%. This accuracy is 17.89% (72.36%) better than that of the traditional algorithm. Meanwhile, compared with other convolutional neural networks, this network has different degrees of improvement, which are 0.68% (ArcFace), 2.14% (Visual Geometry Group Face), and 4.83% (DeepID3). [ABSTRACT FROM AUTHOR]

• We present the face recognition framework with computing and neural models. • The novel neural netwoork model is designed to enhance the robustness. • Social robot scenario is considered to validate the performance. • Face detection network is preoposed to improve the efficiency. In recent years, with the continuous breakthrough of computer vision technology, the accuracy of object detection and target recognition has been improved by leaps and bounds. Face recognition is one of the important research directions in the field of computer vision, which is widely used in mobile payment, safe city, criminal investigation and other fields. Traditional face recognition methods need to extract face image features manually. The extracted features are greatly affected by subjective factors, and time-consuming and laborious. Deep learning is the most important technology in the field of computer vision at present. Compared with traditional face recognition methods, it can extract more essential features of face image without manual participation. In this paper, we build a face recognition system based on neural computing model and the principle of neural network. The experimental results show that the proposed method has high detection rate and short processing time. [Display omitted] [ABSTRACT FROM AUTHOR]

• Improve speed of learning process. • Decreases the computational time. • To reduce the complexity of a model and makes it simple and easier to interpret. • To improve the accuracy of a model and reduce over fitting. The Neural Network has a significant impact in developing predictive models in a wide range of applications. In this paper, a neuro-fuzzy prediction model is developed depending on improving the performance of the traditional artificial neural networks using Adaptive Momentum Optimizer. This optimizer simulates the behavior of the International Trade Analysis in the agriculture industry, and this method is used to determine the optimal parameters of artificial neural networks. The proposed model is compared with the existing models such as Support Vector Machine, Random Forest, Decision Tree and traditional Artificial Neural Network models. To examine the forecasting performance of the proposed approach, agriculture datasets is used. The performance of the models was assessed using different performance evaluation criteria and the empirical results show that the back propagation neural network with Adam optimizer attains favorable prediction accuracy of 96.78%, and a better convergence rate. Compared to other benchmark algorithms, the proposed algorithm performs better, and the result validates the effectiveness of the back propagation with Adam optimizer for Natural Language Processing. Image, graphical abstract [ABSTRACT FROM AUTHOR]

• A new deep learning method is proposed and applied to detect arm fractures in X-rays. • The proposed method achieves the state-of-the-art average precision (AP) in arm fracture detection. • Experienced radiologists are invited to annotate nearly 4000 arm fracture X-rays to evaluate our method. In this paper, a novel deep learning method is proposed and applied to fracture detection in arm bone X-rays. The main improvements include three aspects. First, a new backbone network is established based on feature pyramid architecture to gain more fractural information. Second, an image preprocessing procedure including opening operation and pixel value transformation is developed to enhance the contrast of original images. Third, the receptive field adjustment containing anchor scale reduction and tiny RoIs expansion is exploited to find more fractures. In the experiments, nearly 4000 arm fracture X-ray radiographs collected from MURA dataset are annotated by experienced radiologists. The experiment results show that the proposed deep learning method achieves the state-of-the-art AP in arm fracture detection and it has strong potential application in real clinical environments. [ABSTRACT FROM AUTHOR]

It is of vast significance to assess the bone age of hand radiographs automatically in pediatric radiology and legal medicine. In the literature, many papers focus on improving the assessment accuracy but neglecting the existence of poor-quality X-ray images. However, in real medical scenarios, the existence of poor-quality X-ray images is unavoidable. To tackle this problem, we propose a bone age assessment system for real-world X-ray images. Specifically, we first establish a regression model 'BoNet+' based on densely connected convolutional networks. Then, to handle poor-quality X-ray images, we introduce three model architectures that are different in the way of improving image quality. Experiment results show that the proposed models can estimate the bone age of poor-quality images accurately. We also tentatively put forward that if the expressivity of CNN model is enough high, multiple tasks can be handled together just by a single model. [ABSTRACT FROM AUTHOR]