Electronics

Hua, Yan;Yang, Lin;Yang, Yingyun

doi: 10.3390/electronics11121803pmid: N/A

Deep learning-based methods have made remarkable progress for stereo matching in terms of accuracy. However, two issues still hinder producing a perfect disparity map: (1) blurred boundaries and the discontinuous disparity of a continuous region on disparity estimation maps, and (2) a lack of effective means to restore resolution precisely. In this paper, we propose to utilize multiple frequency inputs and an attention mechanism to construct the deep stereo matching model. Specifically, high-frequency and low-frequency information of the input image together with the RGB image are fed into a feature extraction network with 2D convolutions. It is conducive to produce a distinct boundary and continuous disparity of the smooth region on disparity maps. To regularize the 4D cost volume for disparity regression, we propose a 3D context-guided attention module for stacked hourglass networks, where high-level cost volumes as context guide low-level features to obtain high-resolution yet precise feature maps. The proposed approach achieves competitive performance on SceneFlow and KITTI 2015 datasets.

Kamalzadeh, Saeed;Soleimani, Mohammad

doi: 10.3390/electronics11121804pmid: N/A

A novel, periodic, leaky-wave array antenna using substrate-integrated waveguide (SIW) technology is proposed for continuous beam scanning applications. For this purpose, a periodic structure with the ability to radiate from backward to forward is proposed. The unit cell of this periodic structure includes a longitudinal slot and an H-plane discontinuity. The H-plane step discontinuity is suggested to suppress the open stopband (OSB) and enable continuous beam scanning from backward to forward through the broadside. The impedance matching technique is used to suppress the open stopband. In contrast to phased array antennas, this form of antenna is distinguished by its ability to scan without requiring a complex feeding network. These antennas are used for different factors such as scanning the beam, determining the direction of arrival, avoiding collisions, indoor communications, etc. A prototype of the proposed antenna was fabricated for experimental characterization. The overall physical dimensions of the fabricated antenna are 7.9 mm × 128 mm. The results demonstrate that an adequate level of agreement between measurement and simulation is satisfactory. The results indicate that the suggested antenna can scan continuously in the frequency range of 14.5 to 22.5 GHz between −60 and +57.5 degrees through broadside with a maximum gain of 16 dBi and radiation efficiency of 71%.

Woltran, Flora;Lindner, Katharina-Theresa;Dzojic, Tanja;Schwab, Susanne

doi: 10.3390/electronics11121805pmid: N/A

Although comprehensive digitalization (e.g., the provision of skills and resources) had already been placed on Austria’s education policy agenda prior to the emergence of COVID-19, there is evidence that educators had some difficulty ensuring digital learning opportunities for their students when schools closed in early 2020. Against this backdrop, the present study, which drew on qualitative data from the large-scale INCL-LEA (Inclusive Home Learning) study, aimed to determine whether secondary school teachers (n = 17) from Viennese schools met the prerequisites for successfully implementing digital instruction, formulated in the Will–Skill–Tool model developed by Christensen and Kzenek (2008). Findings reveal that teachers primarily associated their sufficient digital skills with three factors: (1) basic interest and competence, (2) recently attended training, and/or (3) a positive attitude toward changing teaching practices. Interestingly, some educators recognized that digitization offers great potential for implementing individualized teaching approaches. However, the findings point to the didactic necessity of digital socialization in terms of social communication and inclusion when establishing emergency digital education.

Zhang, Linchao;Kim, Dohyeun

doi: 10.3390/electronics11121806pmid: N/A

The era of mobile information has arrived, and people’s lifestyles have undergone tremendous changes. Ordering takeaways through takeout apps on smartphones is one of them. However, most existing takeaway platforms charge high commissions in the middle. There are many fake reviews in restaurants, the authenticity of restaurant ratings is low, and the recommended dishes have low customer satisfaction. This paper aims to solve this problem by introducing a peer-to-peer architecture based on blockchain smart contracts. The proposed architecture leverages the automation of smart contracts to provide autonomous, commission-free food ordering and delivery services. In addition, the smart contract reward mechanism is used to collect order information and rating information, and a deep learning recommendation model is introduced to analyze the data to recommend restaurants and menus to the client accurately. To demonstrate the usability and efficiency of the proposed method, we conducted a case study using public chain-based technologies. At the same time, comprehensive evaluation experiments are carried out, and the results show the importance of the proposed food delivery system.

Fan, Lei;Wang, Yongjun;Zhang, Hongxin;Li, Chao;Xin, Xiangjun

doi: 10.3390/electronics11121807pmid: N/A

In this paper, we propose an image denoising algorithm based on the quaternion wavelet transform (QWT) to address sinusoidal fringe images under strong noise in structured light 3D profilometry. The analysis of a quaternion wavelet shows that the amplitude image of the quaternion wavelet is easily affected by noise. However, the three phase images, which mainly reflect edge and texture information, are randomly and disorderly distributed with respect to noise. The QWT denoising algorithm is suitable for processing sinusoidal fringe images of complex structures in a high-accuracy 3D measurement system. Sinusoidal fringe images are collected and denoised by using the QWT algorithm and classical Gaussian smoothing (GS) denoising algorithm, and GS is used as a reference for the QWT algorithm. The results indicate that the standard deviation is reduced from 0.1448 for raw sinusoidal fringe images to 0.0192, and the signal-to-noise ratio is improved from 4.6213 dB to 13.3463 dB by using the QWT algorithm. The two algorithms have the same denoising effect for a surface with less information. For a surface with rich information, the details of the 3D contour are lost because of the image “blurring” caused by using the GS algorithm, while all edge details of the 3D contour are reconstructed by using the QWT denoising algorithm because of its characteristic of information and noise being separated from the source. For the measured face mask, the error is less than ±0.02 mm. In addition, it takes less than 20 s to run the QWT algorithm to process eight sinusoidal fringe images, which meets the requirements of high-precision measurements.

Qais, Mohammed H.;Hasanien, Hany M.;Turky, Rania A.;Alghuwainem, Saad;Loo, Ka Hong;Elgendy, Mohmmed

doi: 10.3390/electronics11121808pmid: N/A

The aim of this article is to introduce a novel Circle Search Algorithm (CSA) with the purpose of obtaining a precise electrical model of a proton exchange membrane fuel cell (PEMFC). Current-voltage and current-power curves are used to characterize the performance of PEMFCs. A nonlinear model with seven unknown parameters is used to describe these polarization curves. Estimating these unknown parameters is a critical issue because they influence the dynamic analysis of fuel cells in a variety of applications such as transportation and smart grids. The suggested method is based on minimizing the fitness function (the sum of the squared errors (SSE)) between estimated and measured voltage values. The CSA is compared to the neural network algorithm (NNA), grey wolf optimization (GWO), and the sine cosine algorithm (SCA). The optimization results reveal that the simulation times of the CSA, NNA, GWO, and SCA are 5.2, 6, 5.8, and 5.75 s, respectively. Moreover, the CSA converges to the best minimum within the first 100 iterations, which is faster than the other algorithms. The robustness of the CSA is verified using 20 independent runs, where the CSA achieves the smallest average and standard deviation. In addition, the t-test proves the superiority of the CSA compared to the other algorithms, where all p-values are less than 5%. The simulated I-V and I-P curves of the CSA-PEMFC model match the measured curves very closely. Moreover, the efficacy of the CSA-PEMFC model is evaluated under a variety of temperature and pressure conditions. Therefore, the suggested CSA-PEMFC model has the potential to be an accurate and efficient model.

Jin, Woobeen;Kim, Hyeonjin;Lee, Hyukjoon

doi: 10.3390/electronics11121809pmid: N/A

Accurate and efficient path loss prediction in mmWave communication plays an important role in large-scale deployment of the mmWave-based 5G mobile communication systems. Existing methods often present limitations in accuracy and efficiency and fail to fulfill the requirements of cell planning, especially in dense urban environments. In this paper, we propose a novel training method called multi-way local attentive learning, which allows for learning from multiple perspectives on the same set of training samples with local attention paid to each subset of the entire dataset. The sample data set can be partitioned in various ways with respect to different attributes, such that a larger amount of knowledge can be extracted from the same data set. The proposed scheme outperforms the existing schemes in terms of prediction accuracy at the average RMSE of 6.01 dBm.

Ding, Hengbing;Huang, Shan;Jin, Weiqiang;Shan, Yuan;Yu, Hang

doi: 10.3390/electronics11121810pmid: N/A

The end-to-end aspect-based social comment sentiment analysis (E2E-ABSA) task aims to discover human’s fine-grained sentimental polarity, which can be refined to determine the attitude in response to an object revealed in a social user’s textual description. The E2E-ABSA problem includes two sub-tasks, i.e., opinion target extraction and target sentiment identification. However, most previous methods always tend to model these two tasks independently, which inevitably hinders the overall practical performance. This paper investigates the critical collaborative signals between these two sub-tasks and thus proposes a novel cascade social comment sentiment analysis model for jointly tackling the E2E-ABSA problem, namely CasNSA. Instead of treating the opinion target extraction and target sentiment identification as discrete procedures in previous works, our new framework takes the contextualized target semantic encoding into consideration to yield better sentimental polarity judgment. Additionally, extensive empirical results show that the proposed approach effectively achieves a 68.13% F1-score on SemEval-2014, 62.34% F1-Score on SemEval-2015, 56.40% F1-Score on SemEval-2016, and 50.05% F1-score on a Twitter dataset, which is higher than the existing approaches. Ablated experiments demonstrate that the CasNSA model substantially outperforms state-of-the-art methods, even when using fixed words embedding rather than pre-trained BERT fine tuning. Moreover, in-depth performance analysis on the social comment datasets further validates that our work gains superior performance and reliability effectively and efficiently in realistic scenarios.

Mohino-Herranz, Inma;García-Gómez, Joaquín;Aguilar-Ortega, Miguel;Utrilla-Manso, Manuel;Gil-Pita, Roberto;Rosa-Zurera, Manuel

doi: 10.3390/electronics11121811pmid: N/A

This paper presents the Real-Life Indoor Sound Event Dataset (ReaLISED), a new database which has been developed to contribute to the scientific advance by providing a large amount of real labeled indoor audio event recordings. They offer the scientific community the possibility of testing Sound Event Classification (SEC) algorithms. The full set is made up of 2479 sound clips of 18 different events, which were recorded following a precise recording process described along the proposal. This, together with a described way of testing the similarity of new audio, makes the dataset scalable and opens up the door to its future growth, if desired by the researchers. The full set presents a good balance in terms of the number of recordings of each type of event, which is a desirable characteristic of any dataset. Conversely, the main limitation of the provided data is that all the audio is recorded in indoor environments, which was the aim behind this development. To test the quality of the dataset, both the intraclass and the interclass similarities were evaluated. The first has been studied through the calculation of the intraclass Pearson correlation coefficient and further discard of redundant audio, while the second one has been evaluated with the creation, training and testing of different classifiers: linear and quadratic discriminants, k-Nearest Neighbors (kNN), Support Vector Machines (SVM), Multilayer Perceptron (MLP), and Deep Neural Networks (DNN). Firstly, experiments were carried out over the entire dataset, and later over three different groups (impulsive sounds, non-impulsive sounds, and appliances) composed of six classes according to the results from the entire dataset. This clustering shows the usefulness of following a two-step classification process.

Huang, Zedong;Gu, Jinan;Li, Jing;Li, Shuwei;Hu, Junjie

doi: 10.3390/electronics11121812pmid: N/A

To improve the accuracy of using deep neural networks to predict the depth information of a single image, we proposed an unsupervised convolutional neural network for single-image depth estimation. Firstly, the network is improved by introducing a dense residual module into the encoding and decoding structure. Secondly, the optimized hybrid attention module is introduced into the network. Finally, stereo image is used as the training data of the network to realize the end-to-end single-image depth estimation. The experimental results on KITTI and Cityscapes data sets show that compared with some classical algorithms, our proposed method can obtain better accuracy and lower error. In addition, we train our models on PCB data sets in industrial environments. Experiments in several scenarios verify the generalization ability of the proposed method and the excellent performance of the model.