sensors-logo

Journal Browser

Journal Browser

Special Issue "Signal Processing and Machine Learning for Smart Sensing Applications"

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Intelligent Sensors".

Deadline for manuscript submissions: 20 October 2022 | Viewed by 10068

Special Issue Editors

Prof. Dr. Ying-Ren Chien
E-Mail Website
Guest Editor
Department of Electrical Engineering, National Ilan University, Yilan 26047, Taiwan
Interests: adaptive signal processing; machine learning; IoT; noise cancellation
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Mu Zhou
E-Mail Website
Guest Editor
Chongqing Key Lab of Mobile Communications Technology, School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
Interests: wireless localization and sensing; signal processing and detection; machine learning and information fusion
Special Issues, Collections and Topics in MDPI journals
Dr. Ao Peng
E-Mail Website
Guest Editor
School of Informatics, Xiamen University, Xiamen, China
Interests: GNSS; multi-Source positioning; radio navigation
Dr. Ni Zhu
E-Mail Website
Guest Editor
GEOLOC laboratory, University Gustave Eiffel, Marne la Vallée, France
Interests: multisensory fusion for indoor/outdoor pedestrian positioning; GNSS positioning in urban environments; integrity monitoring

Special Issue Information

Dear Colleagues,

This Special Issue focuses on advanced signal processing and machine learning technologies for smart sensing applications. Successful examples include radio navigation, indoor/outdoor positioning, mm-wave sensing, speech denoising, noise cancellation, etc. One of the objectives of this Special Issue is to present smart sensing applications that leverage state-of-the-art signal processing and machine learning technologies. The other main purpose is to promote interdisciplinary collaborations between researchers in the fields of signal processing and machine learning technologies for smart sensing applications.

The emerging trends for smart sensing include: (1) the integration of sensors with low-power embedded signal processing into one system, (2) the integration of multiple sensors in the same system to extract more useful data, and (3) the use of compressive sensing techniques to extract the useful information from original sensor output. To achieve these goals, sophisticated signal processing and machine learning technologies are required.

Prof. Dr. Ying-Ren Chien
Prof. Dr. Mu Zhou
Dr. Ao Peng
Dr. Ni Zhu
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (12 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Other

Article
Multi-Signal Detection Framework: A Deep Learning Based Carrier Frequency and Bandwidth Estimation
Sensors 2022, 22(10), 3909; https://doi.org/10.3390/s22103909 - 21 May 2022
Abstract
Multi-signal detection is of great significance in civil and military fields, such as cognitive radio (CR), spectrum monitoring, and signal reconnaissance, which refers to jointly detecting the presence of multiple signals in the observed frequency band, as well as estimating their carrier frequencies [...] Read more.
Multi-signal detection is of great significance in civil and military fields, such as cognitive radio (CR), spectrum monitoring, and signal reconnaissance, which refers to jointly detecting the presence of multiple signals in the observed frequency band, as well as estimating their carrier frequencies and bandwidths. In this work, a deep learning-based framework named SigdetNet is proposed, which takes the power spectrum as the network’s input to localize the spectral locations of the signals. In the proposed framework, Welch’s periodogram is applied to reduce the variance in the power spectral density (PSD), followed by logarithmic transformation for signal enhancement. In particular, an encoder-decoder network with the embedding pyramid pooling module is constructed, aiming to extract multi-scale features relevant to signal detection. The influence of the frequency resolution, network architecture, and loss function on the detection performance is investigated. Extensive simulations are carried out to demonstrate that the proposed multi-signal detection method can achieve better performance than the other benchmark schemes. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
2D-DOA Estimation in Switching UCA Using Deep Learning-Based Covariance Matrix Completion
Sensors 2022, 22(10), 3754; https://doi.org/10.3390/s22103754 - 14 May 2022
Viewed by 300
Abstract
In this paper, we study the two-dimensional direction of arrival (2D-DOA) estimation problem in a switching uniform circular array (SUCA), which means performing 2D-DOA estimation with a reduction in the number of radio frequency (RF) chains. We propose a covariance matrix completion algorithm [...] Read more.
In this paper, we study the two-dimensional direction of arrival (2D-DOA) estimation problem in a switching uniform circular array (SUCA), which means performing 2D-DOA estimation with a reduction in the number of radio frequency (RF) chains. We propose a covariance matrix completion algorithm for 2D-DOA estimation in a SUCA. The proposed algorithm estimates the complete covariance matrix of a fully sampled UCA (FUCA) from the sample covariance matrix of the SUCA through a neural network. Afterwards, the MUSIC algorithm is performed for 2D-DOA estimation with the completed covariance matrix. We conduct Monte Carlo simulations to evaluate the performance of the proposed algorithm in various scenarios; the performance of 2D-DOA estimation in the SUCA gradually approaches that in the FUCA as the SNR or the number of snapshots increases, which means that the advantages of a FUCA can be preserved with fewer RF chains. In addition, the proposed algorithm is able to implement underdetermined 2D-DOA estimation. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Entropy-Based Concentration and Instantaneous Frequency of TFDs from Cohen’s, Affine, and Reassigned Classes
Sensors 2022, 22(10), 3727; https://doi.org/10.3390/s22103727 - 13 May 2022
Viewed by 282
Abstract
This paper explores three groups of time–frequency distributions: the Cohen’s, affine, and reassigned classes of time–frequency representations (TFRs). This study provides detailed insight into the theory behind the selected TFRs belonging to these classes. Extensive numerical simulations were performed with examples that illustrate [...] Read more.
This paper explores three groups of time–frequency distributions: the Cohen’s, affine, and reassigned classes of time–frequency representations (TFRs). This study provides detailed insight into the theory behind the selected TFRs belonging to these classes. Extensive numerical simulations were performed with examples that illustrate the behavior of the analyzed TFR classes in the joint time–frequency domain. The methods were applied both on synthetic and real-life non-stationary signals. The obtained results were assessed with respect to time–frequency concentration (measured by the Rényi entropy), instantaneous frequency (IF) estimation accuracy, cross-term presence in the TFRs, and the computational cost of the TFRs. This study gives valuable insight into the advantages and limitations of the analyzed TFRs and assists in selecting the proper distribution when analyzing given non-stationary signals in the time–frequency domain. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Gridless Underdetermined Direction of Arrival Estimation in Sparse Circular Array Using Inverse Beamspace Transformation
Sensors 2022, 22(8), 2864; https://doi.org/10.3390/s22082864 - 08 Apr 2022
Viewed by 348
Abstract
Underdetermined DOA estimation, which means estimating more sources than sensors, is a challenging problem in the array signal processing community. This paper proposes a novel algorithm that extends the underdetermined DOA estimation in a Sparse Circular Array (SCA). We formulate this problem as [...] Read more.
Underdetermined DOA estimation, which means estimating more sources than sensors, is a challenging problem in the array signal processing community. This paper proposes a novel algorithm that extends the underdetermined DOA estimation in a Sparse Circular Array (SCA). We formulate this problem as a matrix completion problem. Meanwhile, we propose an inverse beamspace transformation combined with the Gridless SPICE (GLS) algorithm to complete the covariance matrix sampled by SCA. The DOAs are then obtained by solving a polynomial equation with using the Root-MUSIC algorithm. The proposed algorithm is named GSCA. Monte-Carlo simulations are performed to evaluate the GSCA algorithm, the spatial spectrum plots and RMSE curves demonstrated that the GSCA algorithm can give reasonable results of underdetermined DOA estimation in SCA. Meanwhile, the performance of the algorithm under various configurations of SCA is also evaluated. Numerical results indicated that the GSCA algorithm can provide access to solve the DOA estimation problem in Uniform Circular Array (UCA) when random sensor failures occur. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Effectiveness of Artificial Neural Networks for Solving Inverse Problems in Magnetic Field-Based Localization
Sensors 2022, 22(6), 2240; https://doi.org/10.3390/s22062240 - 14 Mar 2022
Viewed by 495
Abstract
Recently, indoor localization has become an active area of research. Although there are various approaches to indoor localization, methods that utilize artificially generated magnetic fields from a target device are considered to be the best in terms of localization accuracy under non-line-of-sight conditions. [...] Read more.
Recently, indoor localization has become an active area of research. Although there are various approaches to indoor localization, methods that utilize artificially generated magnetic fields from a target device are considered to be the best in terms of localization accuracy under non-line-of-sight conditions. In magnetic field-based localization, the target position must be calculated based on the magnetic field information detected by multiple sensors. The calculation process is equivalent to solving a nonlinear inverse problem. Recently, a machine-learning approach has been proposed to solve the inverse problem. Reportedly, adopting the k-nearest neighbor algorithm (k-NN) enabled the machine-learning approach to achieve fairly good performance in terms of both localization accuracy and computational speed. Moreover, it has been suggested that the localization accuracy can be further improved by adopting artificial neural networks (ANNs) instead of k-NN. However, the effectiveness of ANNs has not yet been demonstrated. In this study, we thoroughly investigated the effectiveness of ANNs for solving the inverse problem of magnetic field-based localization in comparison with k-NN. We demonstrate that despite taking longer to train, ANNs are superior to k-NN in terms of localization accuracy. The k-NN is still valid for predicting fairly accurate target positions within limited training times. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Jamming Strategy Optimization through Dual Q-Learning Model against Adaptive Radar
Sensors 2022, 22(1), 145; https://doi.org/10.3390/s22010145 - 26 Dec 2021
Cited by 1 | Viewed by 955
Abstract
Modern adaptive radars can switch work modes to perform various missions and simultaneously use pulse parameter agility in each mode to improve survivability, which leads to a multiplicative increase in the decision-making complexity and declining performance of the existing jamming methods. In this [...] Read more.
Modern adaptive radars can switch work modes to perform various missions and simultaneously use pulse parameter agility in each mode to improve survivability, which leads to a multiplicative increase in the decision-making complexity and declining performance of the existing jamming methods. In this paper, a two-level jamming decision-making framework is developed, based on which a dual Q-learning (DQL) model is proposed to optimize the jamming strategy and a dynamic method for jamming effectiveness evaluation is designed to update the model. Specifically, the jamming procedure is modeled as a finite Markov decision process. On this basis, the high-dimensional jamming action space is disassembled into two low-dimensional subspaces containing jamming mode and pulse parameters respectively, then two specialized Q-learning models with interaction are built to obtain the optimal solution. Moreover, the jamming effectiveness is evaluated through indicator vector distance measuring to acquire the feedback for the DQL model, where indicators are dynamically weighted to adapt to the environment. The experiments demonstrate the advantage of the proposed method in learning radar joint strategy of mode switching and parameter agility, shown as improving the average jamming-to-signal radio (JSR) by 4.05% while reducing the convergence time by 34.94% compared with the normal Q-learning method. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Automatic Sleep-Arousal Detection with Single-Lead EEG Using Stacking Ensemble Learning
Sensors 2021, 21(18), 6049; https://doi.org/10.3390/s21186049 - 09 Sep 2021
Cited by 3 | Viewed by 752
Abstract
Poor-quality sleep substantially diminishes the overall quality of life. It has been shown that sleep arousal serves as a good indicator for scoring sleep quality. However, patients are conventionally asked to perform overnight polysomnography tests to collect their physiological data, which are used [...] Read more.
Poor-quality sleep substantially diminishes the overall quality of life. It has been shown that sleep arousal serves as a good indicator for scoring sleep quality. However, patients are conventionally asked to perform overnight polysomnography tests to collect their physiological data, which are used for the manual judging of sleep arousals. Even worse, not only is this process time-consuming and cumbersome, the judgment of sleep-arousal events is subjective and differs widely from expert to expert. Therefore, this work focuses on designing an automatic sleep-arousal detector that necessitates only a single-lead electroencephalogram signal. Based on the stacking ensemble learning framework, the automatic sleep-arousal detector adopts a meta-classifier that stacks four sub-models: one-dimensional convolutional neural networks, recurrent neural networks, merged convolutional and recurrent networks, and random forest classifiers. This meta-classifier exploits both advantages from deep learning networks and conventional machine learning algorithms to enhance its performance. The embedded information for discriminating the sleep-arousals is extracted from waveform sequences, spectrum characteristics, and expert-defined statistics in single-lead EEG signals. Its effectiveness is evaluated using an open-accessed database, which comprises polysomnograms of 994 individuals, provided by PhysioNet. The improvement of the stacking ensemble learning over a single sub-model was up to 9.29%, 7.79%, 11.03%, 8.61% and 9.04%, respectively, in terms of specificity, sensitivity, precision, accuracy, and area under the receiver operating characteristic curve. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Exploration of Driver Posture Monitoring Using Pressure Sensors with Lower Resolution
Sensors 2021, 21(10), 3346; https://doi.org/10.3390/s21103346 - 12 May 2021
Cited by 2 | Viewed by 1012
Abstract
Pressure sensors are good candidates for measuring driver postural information, which is indicative for identifying driver’s intention and seating posture. However, monitoring systems based on pressure sensors must overcome the price barriers in order to be practically feasible. This study, therefore, was dedicated [...] Read more.
Pressure sensors are good candidates for measuring driver postural information, which is indicative for identifying driver’s intention and seating posture. However, monitoring systems based on pressure sensors must overcome the price barriers in order to be practically feasible. This study, therefore, was dedicated to explore the possibility of using pressure sensors with lower resolution for driver posture monitoring. We proposed pressure features including center of pressure, contact area proportion, and pressure ratios to recognize five typical trunk postures, two typical left foot postures, and three typical right foot postures. The features from lower-resolution mapping were compared with those from high-resolution Xsensor pressure mats on the backrest and seat pan. We applied five different supervised machine-learning techniques to recognize the postures of each body part and used leave-one-out cross-validation to evaluate their performance. A uniform sampling method was used to reduce number of pressure sensors, and five new layouts were tested by using the best classifier. Results showed that the random forest classifier outperformed the other classifiers with an average classification accuracy of 86% using the original pressure mats and 85% when only 8% of the pressure sensors were available. This study demonstrates the feasibility of using fewer pressure sensors for driver posture monitoring and suggests research directions for better sensor designs. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Intelligent Recognition of Chirp Radar Deceptive Jamming Based on Multi-Pulse Information Fusion
Sensors 2021, 21(8), 2693; https://doi.org/10.3390/s21082693 - 11 Apr 2021
Viewed by 807
Abstract
The perception of jamming types is very important for protecting our radar in complex electromagnetic environments. Radar active deceptive jamming based on digital radio frequency memory (DRFM) has a high coherence with the target echo, which confuses the information of the target echo [...] Read more.
The perception of jamming types is very important for protecting our radar in complex electromagnetic environments. Radar active deceptive jamming based on digital radio frequency memory (DRFM) has a high coherence with the target echo, which confuses the information of the target echo and achieves the effect of hiding the real target. Traditional deceptive jamming recognition methods need to extract complex features and artificially set classification thresholds, which is inefficient. The existing neural network-based jamming identification methods still follow the pattern of signal modulation-type identification, so there are fewer types of jamming that can be identified, and the identification accuracy is low in the case of low jamming-to-noise ratios (JNR). This paper studies the input of jamming recognition networks and proposes an improved intelligent identification method for chirp radar deceptive jamming. This method fuses three short-time Fourier transform time–frequency graphs disturbed by three consecutive pulse periods into a new graph as the input of the convolutional neural network (CNN). Using a CNN to classify the time–frequency image has realized the recognition of a variety of common deceptive jamming techniques. Similarly, by changing the network input, the original signal is used to replace the echo signal, which improves the accuracy of the jamming recognition in the case of a low JNR. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
Low-Rank Matrix Recovery from Noise via an MDL Framework-Based Atomic Norm
Sensors 2020, 20(21), 6111; https://doi.org/10.3390/s20216111 - 27 Oct 2020
Cited by 1 | Viewed by 775
Abstract
The recovery of the underlying low-rank structure of clean data corrupted with sparse noise/outliers is attracting increasing interest. However, in many low-level vision problems, the exact target rank of the underlying structure and the particular locations and values of the sparse outliers are [...] Read more.
The recovery of the underlying low-rank structure of clean data corrupted with sparse noise/outliers is attracting increasing interest. However, in many low-level vision problems, the exact target rank of the underlying structure and the particular locations and values of the sparse outliers are not known. Thus, the conventional methods cannot separate the low-rank and sparse components completely, especially in the case of gross outliers or deficient observations. Therefore, in this study, we employ the minimum description length (MDL) principle and atomic norm for low-rank matrix recovery to overcome these limitations. First, we employ the atomic norm to find all the candidate atoms of low-rank and sparse terms, and then we minimize the description length of the model in order to select the appropriate atoms of low-rank and the sparse matrices, respectively. Our experimental analyses show that the proposed approach can obtain a higher success rate than the state-of-the-art methods, even when the number of observations is limited or the corruption ratio is high. Experimental results utilizing synthetic data and real sensing applications (high dynamic range imaging, background modeling, removing noise and shadows) demonstrate the effectiveness, robustness and efficiency of the proposed method. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Article
A Real-Time Dual-Microphone Speech Enhancement Algorithm Assisted by Bone Conduction Sensor
Sensors 2020, 20(18), 5050; https://doi.org/10.3390/s20185050 - 05 Sep 2020
Cited by 2 | Viewed by 1600
Abstract
The quality and intelligibility of the speech are usually impaired by the interference of background noise when using internet voice calls. To solve this problem in the context of wearable smart devices, this paper introduces a dual-microphone, bone-conduction (BC) sensor assisted beamformer and [...] Read more.
The quality and intelligibility of the speech are usually impaired by the interference of background noise when using internet voice calls. To solve this problem in the context of wearable smart devices, this paper introduces a dual-microphone, bone-conduction (BC) sensor assisted beamformer and a simple recurrent unit (SRU)-based neural network postfilter for real-time speech enhancement. Assisted by the BC sensor, which is insensitive to the environmental noise compared to the regular air-conduction (AC) microphone, the accurate voice activity detection (VAD) can be obtained from the BC signal and incorporated into the adaptive noise canceller (ANC) and adaptive block matrix (ABM). The SRU-based postfilter consists of a recurrent neural network with a small number of parameters, which improves the computational efficiency. The sub-band signal processing is designed to compress the input features of the neural network, and the scale-invariant signal-to-distortion ratio (SI-SDR) is developed as the loss function to minimize the distortion of the desired speech signal. Experimental results demonstrate that the proposed real-time speech enhancement system provides significant speech sound quality and intelligibility improvements for all noise types and levels when compared with the AC-only beamformer with a postfiltering algorithm. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Other

Jump to: Research

Case Report
Sensing of Microvascular Vasomotion Using Consumer Camera
Sensors 2021, 21(18), 6256; https://doi.org/10.3390/s21186256 - 18 Sep 2021
Viewed by 768
Abstract
In this paper, we will introduce a method for observing microvascular waves (MVW) by extracting different images from the available images in the video taken with consumer cameras. Microvascular vasomotion is a dynamic phenomenon that can fluctuate over time for a variety of [...] Read more.
In this paper, we will introduce a method for observing microvascular waves (MVW) by extracting different images from the available images in the video taken with consumer cameras. Microvascular vasomotion is a dynamic phenomenon that can fluctuate over time for a variety of reasons and its sensing is used for variety of purposes. The special device, a side stream dark field camera (SDF camera) was developed in 2015 for the medical purpose to observe blood flow from above the epidermis. However, without using SDF cameras, smart signal processing can be combined with a consumer camera to analyze the global motion of microvascular vasomotion. MVW is a propagation pattern of microvascular vasomotions which reflects biological properties of vascular network. In addition, even without SDF cameras, MVW can be analyzed as a spatial and temporal pattern of microvascular vasomotion using a combination of advanced signal processing with consumer cameras. In this paper, we will demonstrate that such vascular movements and MVW can be observed using a consumer cameras. We also show a classification using it. Full article
(This article belongs to the Special Issue Signal Processing and Machine Learning for Smart Sensing Applications)
Show Figures

Figure 1

Planned Papers

The below list represents only planned manuscripts. Some of these manuscripts have not been received by the Editorial Office yet. Papers submitted to MDPI journals are subject to peer-review.

Title: Automatic Sleep Arousal Detection with Single Lead EEG Using Stacking Ensemble Learning
Authors: Ying-Ren Chien; Cheng-Hsuan Wu; Hen-Wai Tsao
Affiliation: Department of Electrical Engineering, National Ilan University, No. 1, Sec. 1, Shen-Lung Rd., I-Lan City, 26041, Taiwan, R.O.C Graduate Institute of Communication Engineering, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, 10617, Taiwan Graduate Institute of Communication Engineering, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, 10617, Taiwan
Abstract: The quality of sleeping could affect human performance in their work and their emotions. Many studies have shown that sleep arousals can induce various sleep disorders. Thus, arousals are a good marker of sleep disruption representing a harmful feature for sleep. Nowadays, the method of detecting sleep arousals is to collect patient’s physiological data such as electroencephalography, electrocardiography, and electromyography through overnight polysomnography (PSG) test. Even though sleep arousal detection could achieve better performance by exploiting much or complete information of the PSG signals, the procedure of collecting PSG signals is time-consuming and cumbersome. Even worse, many cables with contact sensors have to attach to the user. It could make the user feels uncomfortable. In this work, we focus on using a single-lead EEG signal, which is possibly and easily to be collected by using a headphone-like device, to design an automatic sleep arousal detector based on stacking ensemble learning.

Title: Effectiveness of Neural Networks for Solving Inverse Problems in Magnetic-Field-Based Positioning
Authors: Ai-ichiro Sasaki
Affiliation: Kindai Univ.
Abstract: Magnetic-field based positioning is a feasible technology for establishing accurate indoor positioning systems. For realizing the systems, the position of a target device must be calculated from information of magnetic fields detected by multiple sensors. This calculation is not easy because it is a nonlinear inverse problem. We previously proposed a machine-learning approach for solving the inverse problem and demonstrated that the target position can be estimated fairly accurately by using the nearest neighbor methods. However, it was desirable to enhance the estimation accuracy for realizing accurate positioning systems. In this study, we discuss the machine-learning approach by using neural network algorithms. It is demonstrated that the estimation accuracy is enhanced by replacing the nearest neighbor method with the neural network algorithms. The estimation accuracy can be further improved by using predictor functions obtained by considering characteristics of the magnetic-field spatial distribution.

Title: A New Multimedia System Based on Interactive Tangible Interfacing and Computer Vision Techniques for Traditional Rituals
Authors: Chao-Ming Wang, Shih-Mo Tseng and Yu-Sheng Lin
Affiliation: Department of Digital Media Design, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin 64002, Taiwan, R.O.C.
Abstract: To promote the sustainability of the traditional rituals of the popular Mazu religious culture in Taiwan and Southeast Asia, a new digital system named “i-Ritual,” which combines interactive multimedia technology with the ritual process for worshipping the deity Mazu, is proposed for a worshipper to experience the ritual tradition interactively. Three major steps of the Mazu ritual process, namely, “incense waving,” “moon-block casting,” and “fortune-telling poem drawing,” are carried out by multimedia technology using interactive tangible interfacing and computer vision techniques. Interviews with experts and a questionnaire survey of participants’ opinions were carried out to evaluate the effectiveness of the system, yielding the following findings: (1) the proposed system can bring the participants to experience the ritual process effectively; (2) the proposed three major ritual steps not only attracts more participations but also promotes general people’s learning of the religious ritual process; (3) the proposed system can attract participants by integrating multimedia effects with good usability; (4) the uses of tangible interfacing and computer vision techniques bring the users new experiences of the traditional ritual; and (5) the proposed system was affirmed to be good for religious education and can be extended for use by other religions.

Title: Development of Rim Defect Inspection System Using Deep Learning Network and Robot Arm
Authors: Wei-Lung Mao, Yu-Ying Chiu, Bing-Hong Lin, Chun-Chi Wang, Yi-Ting Wu, Cheng-Yu You
Affiliation: Graduate School of Engineering Science and Technology and Department of Electrical Engineering, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin 64002, Taiwan, R.O.C.
Abstract: In this research, the eye-in-hand vision architecture, deep learning and convolutional neural networks are imported to construct an automatic defect detection system. It is a more efficient defect detection way for forged aluminum rims of electric vehicles. The detection system takes ABB robot arm as the core of the motion path planning, and outputs a 3D drawing of the rim using Robotstudio tool, and it can simulate the environment layout and path trajectory. The generative adversarial networks (GAN) and deep convolution generative adversarial networks (DCGAN) are used to generate a large number of defective images to expand the number of training data sets. The defect detection algorithm developed by YOLO achieves a fast and high-performance defect detection, which is better than current methods. The graphical user interface with C# language can find and mark the defect patterns in the detection images. The experimental results prove the accuracy and efficiency of our proposed automatic inspection system.

Title: An MQTT-based Automatic Fire Alarm System for Psychiatric Ward
Authors: Lan-Ying Chang1,3, Chen-Kuei Li1, Shan-Ju Wang1, Wei-Lung Mao2*, Yu-Chen Chang2,Yao-Teng Yang2, Yun-Yi Li2 , Chun-Chi Wang2
Affiliation: 1Department of Nursing, Wanqiao Branch, Taichung Veterans General Hospital, Taiwan 2Graduate School of Engineering Science and Technology and Department of Electrical Engineering, National Yunlin University of Science and Technology, Taiwan 3Graduate Institute of International Business and Administration and Department of Business Administration, National Yunlin University of Science and Technology, Taiwan
Abstract: Keywords: Smoke Sensors, Message Queuing Telemetry Transport (MQTT) Protocol, Wired and Wireless Communication, Hospital Fire Alarm System, Psychiatric Ward. Corresponding Author: Wei-Lung Mao, Professor, Graduate School of Engineering Science and Technology and Department of Electrical Engineering, National Yunlin University of Science and Technology, 123 University Road, Section 3, Douliou, Yunlin 64002, Taiwan, R.O.C. Tel: +886-5-534-2601 Ext 4222 Fax: +886-5-532-1719 E-mail: [email protected], [email protected] As society gradually relies on technologies to prevent dangers and detect the environment, the processing of data transmission and analysis becomes important. In recent years, firefighters are sacrificed during rescue missions. Even with the high-tech equipment in the current era, the lives of firefighters are not prevented in the fire scene. The psychiatric ward is equipped with access control management. When a fire occurs, the main key is to quickly extinguish the fire and evacuate the patient. In the fire drill research, when the alarm bell rings, it takes 9 minutes for the staff to find the fire source, and 19 minutes for reporting, extinguishing, and evacuating to a safe place.Thus, this research designs and develops an alarm system for hospitals and schools. If the fire source can be found in the early stage of the fire and put out the fire in time, the property loss of the hospital can be reduced. Therefore, the effect of intervening scientific and intelligent equipment to shorten the emergency response time is discussed. In fact, there are many addressable fire alarm systems on the market, whether wireless or wired, in fact, the research has a certain scale, and because the implementation site of this project is located in the Wanqiao Branch of Taichung Veterans General Hospital, in order to maintain the hospital. The internal equipment operates normally and reduces interference, and wired transmission is used as the planned method. However, if the wired products on the market cannot rely on the original transmission line, the wiring of the building must be remodeled, which will increase the time. In order to reduce this situation, network transmission is a very good choice, and this project will also take this as the main axis. Relying on the network line or WIFI, the Arduino microprocessor obtains the alarm data of the sensor and uploads it to the network, and adopts the MQTT communication protocol, and finally relies on the host to collect the sensor information at each location. The Message Queuing Telemetry Transport (MQTT) technology and fire sensors are utilized and integrated to extract the conditions of the environment. In the system, the subscriber server collects the data on a personal computer. Also, the website and database server monitor the wards in a hospital. Integrating and designing the systems help to understand the emergent situation.

Title: Segmentation of Articulated Vehicles based on Region Growing
Authors: Chien-Chou Lin1,*, Yu-Jyun Huang1, Chuan-Yu Chang1, Wei-Lung Mao2, and Teng-Wen Chang3
Affiliation: 1 Dept. of Computer Science and Information Engineering, National Yunlin University of Science & Technology, Taiwan; [email protected] 2 Dept. of Electrical Engineering, National Yunlin University of Science & Technology, Taiwan; [email protected] 3 Dept. of Digital Media Design, National Yunlin University of Science & Technology, Taiwan;[email protected]
Abstract: Since an articulated vehicle has diverse shapes with revolute joints, articulating vehicles’ recognition is challenging. However, if an articulated vehicle is segmented into several parts according to its joints, the recognition of the articulated vehicle can be simplified as rigid objects. Furthermore, other possible poses of the articulated vehicle can be predicted to make the recognition easier. This paper proposes a joint detection approach for the point cloud of an articulated vehicle. The proposed method uses two 3D point clouds of the same target vehicle with different poses as inputs. Two point clouds are converted into 2D bearing images. Then the corresponding pixels between two bearing images obtained by the SURF algorithm are used to derive the optimal transformation matrix. After two point clouds are transformed to the same coordinate system, the region growing method segments two point clouds with the surface norms. Experimental results show that the proposed method has a high detection rate of 85.7%. Furthermore, since the proposed method uses 2D images for alignment, the proposed algorithm is very efficient.

Title: Segmentation of Articulated Vehicles based on Region Growing
Authors: Chin-Yi Cheng*, Ilham Saputra, Cheng-En Shi
Affiliation: Department of Mechanical Engineering, National Yunlin University of Science and Technology, Douliu 64002, 5 Taiwan; [email protected] (I.S); [email protected] (C.E.S) * Correspondence: [email protected]; +886-5-5342601
Abstract: Soft pneumatic robotic grippers have been studied for more than 30 years but are still an area of challenge. Today, most of the soft robotic grippers are created through the injection molding process. It is limited in production consistency and difficult in multiple transformations. To solve this problem, a study was performed with a pneumatic soft robotic gripper which has been introduced by applying the Fused deposition modelling method of 3D printing technology in the fabrication process. The introduced pneumatic soft robotic gripper will utilize an actuation system using neumatic compression as an activator, because of the characteristics of the pneumatic system which has fast speed, easy control and easy maintenance. Making soft robotics easy to apply in areas such as industrial robot arm grippers, objects that need to be protected on surfaces, and more. In an effort to create intelligent robots, computer vision was applied to create gripper responsiveness which was created by utilizing real-time image processing and pressure regulation applied by PID control. The combination of soft robots with sensing capabilities from sensors applied to the gripper is a new approach to enable soft robots to be applied in gripping tasks quickly and precisely.

Title: An Insight into Entropy-Based Concentration and Instantaneous Frequency of Time-Frequency Distributions from Cohen's, Affine and Reassigned Classes
Authors: prof. dr. sc. Jonatan Lerga
Affiliation: Head of the Center for Artificial Intelligence and Cybersecurity, University of Rijeka Head of the Department of Computer Engineering, Faculty of Engineering, University of Rijeka Head of the Information Processing Laboratory, Faculty of Engineering, University of Rijeka Vukovarska 58, HR-51000 Rijeka, Croatia
Abstract: This paper explores three groups of time-frequency distributions: Cohen's, Affine, and Reassigned class of time-frequency representations (TFRs). The study provides detailed insight into the theory behind the selected TFRs belonging to these classes. Next, extensive numerical simulations were performed with examples that illustrate the behavior of analyzed TFR classes in the joint time-frequency domain. The methods were applied both on synthetic and real-life non-stationary signals. The obtained results were assessed with respect to time-frequency concentration (measured by the Rényi entropy), instantaneous frequency (IF) estimation accuracy, and cross-terms presence in the TFRs. The study gives valuable insight into the advantages and limitations of the analyzed TFRs helping select the proper distribution when analyzing given non-stationary signals in the time-frequency domain.

Back to TopTop