Depthwise Separable Convolutional Neural Network Research Articles

A framework for electricity load forecasting based on attention mechanism time series depthwise separable convolutional neural network

Electricity load exhibits daily and weekly cyclical patterns as well as random characteristics. At present, prevailing deep learning models cannot learn electricity load cyclical and stochastic features adequately. This results in insufficient prediction accuracy and the scalability of current methods. To tackle these difficulties, this paper proposes a framework for electrical load prediction based on an Attention Mechanism Time Series Depthwise Separable Convolutional Neural Network (ELPF-ATDSCN). The framework starts by using the Maximum Information Coefficient for exogenous variable selection. It then incorporates a seasonal decomposition algorithm with manual feature engineering to extract the cyclical and stochastic features of the electrical load. Subsequently, the framework employs the ATDSCN to learn the cyclical and stochastic features of the electrical load. In addition, the Bayesian algorithm optimizes model hyperparameters for optimal model performance. Experimental results of point and interval load prediction on datasets from the US and Nordic power markets reveal that the ATDSCN model proposed in this paper enhances load prediction accuracy compared with other models. It can provide more reliable predictions for power system operation and dispatch.

Enhanced multimodal emotion recognition in healthcare analytics: A deep learning based model-level fusion approach

Deep learning techniques have drawn considerable interest in emotion recognition due to recent technological developments in healthcare analytics. Automatic patient emotion recognition can assist healthcare analytics by providing feedback to the stakeholders of competent healthcare about the conditions of the patients and their satisfaction levels. In this paper, we propose a novel model-level fusion technique based on deep learning for enhanced emotion recognition from multimodal signals to monitor patients in connected healthcare. The representative visual features from the video signals are extracted through the Depthwise Separable Convolution Neural Network, and the optimized temporal attributes are derived from the multiple physiological data utilizing Bi-directional Long Short-Term Memory. A soft attention method fused the high multimodal features obtained from the two data modalities to retrieve the most significant features by focusing on emotionally salient parts of the features. We exploited two face detection methods, Histogram of Oriented Gradients and Convolutional Neural Network-based face detector (ResNet-34), to observe the effects of facial features on emotion recognition. Lastly, extensive experimental evaluations have been conducted using the widely used Bio Vid Emo DB multimodal dataset to verify the performance of the proposed architecture. Experimental results show that the developed fusion architecture improved the accuracy of emotion recognition from multimodal signals and outperformed the performance of both state-of-the-art techniques and baseline methods.

Web service reliability and scalability determination using optimized depth wise separable convolutional neural network

AbstractWeb service composition (WSC), a distributed architecture, creates new services atop existing ones. Ensuring trust and assessing performance and dependability in online services coordination is essential. In this paper, “Web Service Reliability and Scalability Determination Using Depth Wise Separable Convolutional Neural Network” (WSRS‐DWSCNN) is proposed to assess the trustworthiness of online service compositions, particularly focusing on performance and dependability. This work addresses the need to predict the reliability and scalability of Business Process Execution Language (BPEL) composite web services. The proposed approach transforms the BPEL specification into a Depth Wise Separable Convolutional Neural Network (DWSCNN) and annotates it with probabilistic properties for prediction. The DWSCNN model classifies the outcomes as correct or incorrect, and to enhances the prediction of web service composition scalability and reliability, we optimize the DWSCNN's weight parameters using the Adolescent Identity Search Algorithm (AISA). The proposed technique is activated in Python and its efficacy is analyzed under some metrics, such as reliability, scalability, accuracy, sensitivity, specificity, precision, F‐measure. The proposed method provides 12.36%, 45.39%, and 25.97% better reliability, 41.39%, 11.39%, 34.16% better accuracy compared with existing methods like, Web service reliability prediction depending on machine learning (WSRS‐K‐means), reliability prediction method for multiple state cloud/edge‐basis network utilizing deep neural network (WSRS‐DNN‐BO), and improving reliability of mobile social cloud computing utilizing machine learning in content addressable network (WSRS‐CAN), respectively.

An Efficient FPGA-based Depthwise Separable Convolutional Neural Network Accelerator with Hardware Pruning

Convolutional neural networks (CNNs) have been widely deployed in computer vision tasks. However, the computation and resource intensive characteristics of CNN bring obstacles to its application on embedded systems. This article proposes an efficient inference accelerator on Field Programmable Gate Array (FPGA) for CNNs with depthwise separable convolutions. To improve the accelerator efficiency, we make four contributions: (1) an efficient convolution engine with multiple strategies for exploiting parallelism and a configurable adder tree are designed to support three types of convolution operations; (2) a dedicated architecture combined with input buffers is designed for the bottleneck network structure to reduce data transmission time; (3) a hardware padding scheme to eliminate invalid padding operations is proposed; and (4) a hardware-assisted pruning method is developed to support online tradeoff between model accuracy and power consumption. Experimental results show that for MobileNetV2 the accelerator achieves 10× and 6× energy efficiency improvement over the CPU and GPU implementation, and 302.3 frames per second and 181.8 GOPS performance that is the best among several existing single-engine accelerators on FPGAs. The proposed hardware-assisted pruning method can effectively reduce 59.7% power consumption at the accuracy loss within 5%.

Efficient skeleton-based action recognition via multi-stream depthwise separable convolutional neural network

Efficient skeleton-based action recognition via multi-stream depthwise separable convolutional neural network

An effective stacked autoencoder based depth separable convolutional neural network model for face mask detection.

The COVID-19 pandemic has been infecting the entire world over the past years. To prevent the spread of COVID-19, people have acclimatised to the new normal, which includes working from home, communicating online, and maintaining personal cleanliness. There are numerous tools required to prepare to compact transmissions in the future. One of these elements for protecting individuals from fatal virus transmission is the mask. Studies have indicated that wearing a mask may help to reduce the risk of viral transmission of all kinds. It causes many public places to take efforts to ensure that its guests wear adequate face masks and keep a safe distance from one another. Screening systems need to be installed at the doors of businesses, schools, government buildings, private offices, and/or other important areas. A variety of face detection models have been designed using various algorithms and techniques. Most of the articles in the previously published research have not worked on dimensionality reduction in conjunction with depth-wise separable neural networks. The necessity of determining the identities of people who do not cover their faces when they are in public is the driving factor for the development of this methodology. This research work proposes a deep learning technique to determine if a person is wearing mask or not and identifies whether it is properly worn or not. Stacked Auto Encoder (SAE) technique is implemented by stacking the following components: Principal Component Analysis (PCA) and Depth-wise Separable Convolutional Neural Network (DWSC-NN). PCA is used to reduce the irrelevant features in the images and resulted high true positive rate in the detection of mask. We achieved an accuracy score of 94.16% and an F1 score of 96.009% by the application of the method described in this research.

Minimalist DCT-based Depthwise Separable Convolutional Neural Network Approach for Tangut Script

The Tangut script, a lesser-explored dead script comprising numerous characters, has received limited attention in deep learning research, particularly in the field of optical character recognition (OCR). Existing OCR studies primarily focus on widely-used characters like Chinese characters and employ deep convolutional neural networks (CNNs) or combinations with recurrent neural networks (RNNs) to enhance accuracy in character recognition. In contrast, this study takes a counterintuitive approach to develop an OCR model specifically for the Tangut script. We utilize shorter layers with slimmer filters using a depthwise separable convolutional neural network (DSCNN) architecture. Furthermore, we preprocess the dataset using a frequency-based transformation, namely the Discrete Cosine Transform (DCT). The results demonstrate successful training of the model, showcasing faster convergence and higher accuracy compared to traditional deep neural networks commonly used in OCR applications.

A 34.7 µW Speech Keyword Spotting IC Based on Subband Energy Feature Extraction

In the era of the Internet of Things (IoT), voice control has enhanced human–machine interaction and the accuracy of keyword spotting (KWS) algorithms has reached 97%; however, the high power consumption of KWS algorithms caused by their huge computing and storage requirements has limited their application in Artificial Intelligence of Things (AIoT) devices. In this study, voice features are extracted by utilizing the fast discrete cosine transform (FDCT) for frequency-domain transformation and to shorten the process of calculating the logarithmic spectrum and cepstrum. The designed KWS system is a two-stage wake-up system, with a sound detection (SD) awakening KWS. The inference process of the KWS network is achieved using time-division computation, reducing the KWS clock to an ultra-low frequency of 24 kHz.At the same time, the implementation of a depthwise separable convolution neural network (DSCNN) greatly reduces the parameter quantity and computation. Under the GSMC 0.11 µm technology, post-layout simulation results show that the total synthesized area of the entire system circuit is 0.58 mm2, the power consumption is 34.7 µW, and the F1-score of the KWS is 0.89 with 10 dB noise, which makes it suitable as a KWS system in AIoT devices.

Prediction of brown tide algae using improved Gramian angular field and deep learning based on laser-induced fluorescence spectrum

The frequent occurrence of algal blooms has seriously affected the marine environment and human production activities. Therefore, it is crucial to monitor the phytoplankton concentration in water bodies. In this study, a prediction method for brown tide algae using improved Gramian angular field (IGAF) and deep learning based on the laser-induced fluorescence spectrum was proposed. The method combined one-dimensional (1D) fluorescence spectrum with IGAF for image coding. The internal normalizing approach of the original Gramian angle field algorithm was upgraded from local to global, which can increase the difference between samples with various concentrations. Then, we established a novel technique that fully takes into account the Gramian angular difference field and Gramian angular summation field features, allowing it to control the main and sub-diagonal features and successfully convert 1D sequences into images by adding various weight factors. Using depthwise separable convolutional neural network to extract image features helps reduce model training parameters, paired with long short-term memory network to rapidly predict the concentration of brown tide. To confirm the actual performance of the given approach, ablation and contrast experiments were carried out, and the results showed that the method’s regression accuracy, R 2 was 97.8%, with the lowest mean square error and mean absolute error. This study investigated the transformation of 1D spectra into images using IGAF, which not only explored the application of the fluorescence spectrum image coding method for algal regression but also enabled the introduction of the potent benefits of deep learning image processing into the field of spectral analysis.

Deep dive into hybrid networks: A comparative study and novel architecture for efficient power prediction

The prediction of electric power consumption (PC) and power generation (PG) plays an important role in the management, trading, and storage of energy, and in saving resources. Traditional approaches based on machine learning (ML) are subject to several issues arising from the use of handcrafted feature extraction, the requirement to learn the nonlinear relationships between output and input sequences, and inadequate adjustability to real-world scenarios. Similarly, the performance of deep-plane hybrid networks decreases when their depth increases. There is a need for more robust models in the energy forecasting domain that have high accuracy and strong generalizability to real-world implementations. To tackle the problems described above and to obtain a robust model, we propose a hybrid network based on a dilated depthwise separable convolutional neural network (DDSCNN) and a bidirectional gated recurrent unit (BGRU) in which a skip connection strategy is used to forecast short-term power production and consumption. In our framework, the obtained data are passed through a preprocessing stage for cleaning, and the refined data are input to the residual DDSCNN block to extract spatial features. These features are then fed to the residual BGRU block to learn the sequential and temporal patterns from them, and finally, dense layers are included at the end of the model to forecast the power. A comprehensive ablation study is conducted on various spatiotemporal hybrid models, and the best performer in terms of accuracy is selected using six power datasets. The root mean square error (RMSE) values obtained by the proposed model for the Korea south east solar power (KSESP), Australia Alice Springs solar power (AASSP), and Korea Yeongam solar power (KYSP) datasets are 0.8335, 0.2317, and 0.0954, respectively. Similarly, the RMSE values for the Korea south east wind power (KSEWP), individual household electric power consumption (IHEPC), and advanced institute of convergence technology (AICT) datasets are recorded as 0.5596, 0.1054, and 0.1068, respectively.

Complex-Valued Depthwise Separable Convolutional Neural Network for Automatic Modulation Classification

Complex-Valued Depthwise Separable Convolutional Neural Network for Automatic Modulation Classification

A double-channel multiscale depthwise separable convolutional neural network for abnormal gait recognition.

Abnormal gait recognition is important for detecting body part weakness and diagnosing diseases. The abnormal gait hides a considerable amount of information. In order to extract the fine, spatial feature information in the abnormal gait and reduce the computational cost arising from excessive network parameters, this paper proposes a double-channel multiscale depthwise separable convolutional neural network (DCMSDSCNN) for abnormal gait recognition. The method designs a multiscale depthwise feature extraction block (MDB), uses depthwise separable convolution (DSC) instead of standard convolution in the module and introduces the Bottleneck (BK) structure to optimize the MDB. The module achieves the extraction of effective features of abnormal gaits at different scales, and reduces the computational cost of the network. Experimental results show that the gait recognition accuracy is up to 99.60%, while the memory size of the model is reduced 4.21 times than before optimization.

LWSleepNet: A lightweight attention-based deep learning model for sleep staging with singlechannel EEG.

Sleep is vital to human health, and sleep staging is an essential process in sleep assessment. However, manual classification is an inefficient task. Along with the increased demand for portable sleep quality detection devices, lightweight automatic sleep staging needs to be developed. This study proposes a novel attention-based lightweight deep learning model called LWSleepNet. A depthwise separable multi-resolution convolutional neural network is introduced to analyze the input feature map and captures features at multiple frequencies using two different sized convolutional kernels. The temporal feature extraction module divides the input into patches and feeds them into a multi-head attention block to extract time-dependent information from sleep recordings. The model's convolution operations are replaced with depthwise separable convolutions to minimize its number of parameters and computational cost. The model's performance on two public datasets (Sleep-EDF-20 and Sleep-EDF-78) was evaluated and compared with those of previous studies. Then, an ablation study and sensitivity analysis were performed to evaluate further each module. LWSleepNet achieves an accuracy of 86.6% and Macro-F1 score of 79.2% for the Sleep-EDF-20 dataset and an accuracy of 81.5% and Macro-F1 score of 74.3% for the Sleep-EDF-78 dataset with only 55.3 million floating-point operations per second and 180 K parameters. On two public datasets, LWSleepNet maintains excellent prediction performance while substantially reducing the number of parameters, demonstrating that our proposed Light multiresolution convolutional neural network and temporal feature extraction modules can provide excellent portability and accuracy and can be easily integrated into portable sleep monitoring devices.

A Lightweight Bearing Fault Diagnosis Method Based on Multi-Channel Depthwise Separable Convolutional Neural Network

The existing rolling bearing fault diagnosis method based on the deep convolutional neural network has the issues of insufficient feature extraction ability, poor anti-noise ability, and a large number of model parameters. A lightweight bearing fault diagnosis method based on depthwise separable convolutions is proposed. The proposed method can simultaneously extract different features from vibration signals in different directions to enhance the stability of the diagnosis model. The lightweight unit based on depthwise separable convolutions in the feature extraction layer reduces the size of the model and the number of parameters that need to be learned. The vibration signals of bearings in different directions are converted into time-frequency signals by the short-time Fourier transform (STFT) and then into pictures as the input of the model. In order to verify the effectiveness and generalization of the method, this paper uses the gearbox data set of Southeast University and the CWRU (Case Western Reserve University) bearing data set for experiments. Comparisons of bearing fault diagnosis results using the proposed model with other classical deep learning models are implemented. The results show that the proposed model is superior to other classical deep learning models; thus, it has a smaller model size, higher accuracy, and less computation burden. Compared with using a single-direction vibration signal as input, the proposed model that uses multiple vibration signals in different directions as input has more accuracy.

DSCNN-LSTMs: A Lightweight and Efficient Model for Epilepsy Recognition.

Epilepsy is the second most common disease of the nervous system. Because of its high disability rate and the long course of the disease, it is a worldwide medical problem and social public health problem. Therefore, the timely detection and treatment of epilepsy are very important. Currently, medical professionals use their own diagnostic experience to identify seizures by visual inspection of the electroencephalogram (EEG). Not only does it require a lot of time and effort, but the process is also very cumbersome. Machine learning-based methods have recently been proposed for epilepsy detection, which can help clinicians make rapid and correct diagnoses. However, these methods often require extracting the features of EEG signals before using the data. In addition, the selection of features often requires domain knowledge, and feature types also have a significant impact on the performance of the classifier. In this paper, a one-dimensional depthwise separable convolutional neural network and long short-term memory networks (1D DSCNN-LSTMs) model is proposed to identify epileptic seizures by autonomously extracting the features of raw EEG. On the UCI dataset, the performance of the proposed 1D DSCNN-LSTMs model is verified by cross-validation and time complexity comparison. Compared with other previous models, the experimental results show that the highest recognition rates of binary and quintuple classification are 99.57% and 81.30%, respectively. It can be concluded that the 1D DSCNN-LSTMs model proposed in this paper is an effective method to identify seizures based on EEG signals.

MS-CheXNet: An Explainable and Lightweight Multi-Scale Dilated Network with Depthwise Separable Convolution for Prediction of Pulmonary Abnormalities in Chest Radiographs

Pulmonary diseases are life-threatening diseases commonly observed worldwide, and timely diagnosis of these diseases is essential. Meanwhile, increased use of Convolution Neural Networks has promoted the advancement of computer-assisted clinical recommendation systems for diagnosing diseases using chest radiographs. The texture and shape of the tissues in the diagnostic images are essential aspects of prognosis. Therefore, in the latest studies, the vast set of images with a larger resolution is paired with deep learning techniques to enhance the performance of the disease diagnosis in chest radiographs. Moreover, pulmonary diseases have irregular and different sizes; therefore, several studies sought to add new components to existing deep learning techniques for acquiring multi-scale imaging features from diagnostic chest X-rays. However, most of the attempts do not consider the computation overhead and lose the spatial details in an effort to capture the larger receptive field for obtaining the discriminative features from high-resolution chest X-rays. In this paper, we propose an explainable and lightweight Multi-Scale Chest X-ray Network (MS-CheXNet) to predict abnormal diseases from the diagnostic chest X-rays. The MS-CheXNet consists of four following main subnetworks: (1) Multi-Scale Dilation Layer (MSDL), which includes multiple and stacked dilation convolution channels that consider the larger receptive field and captures the variable sizes of pulmonary diseases by obtaining more discriminative spatial features from the input chest X-rays; (2) Depthwise Separable Convolution Neural Network (DS-CNN) is used to learn imaging features by adjusting lesser parameters compared to the conventional CNN, making the overall network lightweight and computationally inexpensive, making it suitable for mobile vision tasks; (3) a fully connected Deep Neural Network module is used for predicting abnormalities from the chest X-rays; and (4) Gradient-weighted Class Activation Mapping (Grad-CAM) technique is employed to check the decision models’ transparency and understand their ability to arrive at a decision by visualizing the discriminative image regions and localizing the chest diseases. The proposed work is compared with existing disease prediction models on chest X-rays and state-of-the-art deep learning strategies to assess the effectiveness of the proposed model. The proposed model is tested with a publicly available Open-I Dataset and data collected from a private hospital. After the comprehensive assessment, it is observed that the performance of the designed approach showcased a 7% to 18% increase in accuracy compared to the existing method.

Research on Fast Recognition of Vulnerable Traffic Participants in Intelligent Connected Vehicles on Edge Computing

Real-time and fast recognition of all kinds of traffic participants in intelligent driving has always been a major difficulty in the research of internet of vehicles. With the advent of edge computing, we try to deploy an image recognition algorithm directly to the intelligent vehicles. However, the original image recognition algorithm is difficult to be directly deployed on the vehicles due to limited edge device resources. Based on this, a fast recognition model of vulnerable traffic participants based on depthwise separable convolutional neural network (DSCYOLO) is proposed in this paper. The algorithm can significantly reduce the convolutional parameter quantity and computing load, making it suitable for deployment on the vehicle-mounted edge embedded devices. In order to validate the effectiveness of the proposed method, its simulation results are compared with the main target detection models Faster R-CNN, SSD and YOLOv3. The results show that the recognition time of the proposed model is reduced by 80.28%, 66.80% and 86.74%, respectively, on the basis of a relatively high recognition precision. The model can realize real-time detection and fast recognition of vulnerable traffic participants, so as to avoid a large number of traffic accidents. It has significant social and economic benefits.

Uncertain and biased facial expression recognition based on depthwise separable convolutional neural network with embedded attention mechanism

Datasets play a crucial role in the development of facial expression recognition (FER), but most of the datasets suffer from obvious uncertainties and biases caused by different cultures and collection conditions. To look deeper into these issues, this paper first conducts two sets of experiments, face detection, and facial expression (FE) classification. They are based on three datasets (CK+, FER2013, and RAF–DB), which are collected from lab and wild environments. This paper proposes a network, depthwise separable convolutional neural network (CNN) with an embedded attention mechanism (DSA–CNN) for expression recognition. First, at the preprocessing stage, we obtain the maximum expression range clipping, which is calculated from 81 facial landmark points to filter nonface interferences. Then, we use DSA–CNN, which is based on a coordinate squeeze-and-excitation (CSE) attention for feature extraction. Finally, to further deal with imbalanced class biases and uncertainties issues, this paper proposes a class-weighted cross-entropy loss (CCE-loss) to alleviate the imbalance among seven emotional classes. Then, we combine CCE-loss with ranking regularization loss (RR-loss) and self-importance weighting cross-entropy loss (SCE-loss) at label amend stage, to jointly guide the training of the network. Extensive experiments on three FER datasets demonstrate that our proposed method outperforms most of the state-of-the-art methods eventually.

Warship formation extraction and recognition based on density‐based spatial clustering of applications with noise and improved convolutional neural network

Formation recognition is a significant focus of maritime target recognition. Automatic formation extraction and recognition facilitate autonomous decision-making. However, few studies have explored formation extraction prior to recognition. This paper introduces a density-based spatial clustering of applications with noise (DBSCAN) method based on Gaussian kernel to extract formation targets. On this basis, a depthwise separable convolutional neural network (DSCNN) method is proposed for formation recognition. A track simulation system is established to form a track dataset containing three different proportions of clutter, and the formation extraction method is examined using track dataset. Subsequently, the image dataset with eight different types of formation is formulated, on the basis of various detection errors, the DSCNN method for formation recognition is compared with several typical deep learning methods. As exposed in experimental results, the DBSCAN method based on Gaussian kernel can guarantee accurate extraction of formation targets subject to different proportions of clutter. Hence, it is greatly robust and capable of effective formation extraction. Under different radar detection errors, the formation recognition accuracy of DSCNN is 91.5%–99.5%, which achieves performance improvement by up to 12.5% compared with other deep learning methods. The combination of DBSCAN and DSCNN can well realise formation extraction and recognition with different proportions of clutter in tracks and various radar detection errors.

A classification model based on depthwise separable convolutional neural network to identify rice plant diseases

<p><span>Every year a number of rice diseases cause major damage to crop around the world. Early and accurate prediction of various rice plant diseases has been a major challenge for farmers and researchers. Recent developments in the convolutional neural networks (CNNs) have made image processing techniques more convenient and precise. Motivated from that in this research, a depthwise separable convolutional neural network based classification model has been proposed for identifying 12 types of rice plant diseases. Also, 8 different state-of-the-art convolution neural network model has been fine-tuned specifically for identifying the rice plant diseases and their performance has been evaluated. The proposed model performs considerably well in contrast to existing state-of-the-art CNN architectures. The experimental analysis indicates that the proposed model can correctly diagnose rice plant diseases with a validation and testing accuracy of 96.5% and 95.3% respectively while having a substantially smaller model size.</span></p>