Sensors

Open AccessCommunication

A Monitoring System for Carbon Dioxide in Honeybee Hives: An Indicator of Colony Health

by

,

,

,

,

and

Sensors 2023, 23(7), 3588; https://doi.org/10.3390/s23073588 (registering DOI) - 29 Mar 2023

Abstract

Non-dispersive infra-red (NDIR) detectors have become the dominant method for measuring atmospheric CO₂, which is thought to be an important gas for honeybee colony health. In this work we describe a microcontroller-based system used to collect data from Senserion SCD41 NDIR [...] Read more.

Non-dispersive infra-red (NDIR) detectors have become the dominant method for measuring atmospheric CO₂, which is thought to be an important gas for honeybee colony health. In this work we describe a microcontroller-based system used to collect data from Senserion SCD41 NDIR sensors placed in the crown boards and queen excluders of honeybee colonies. The same sensors also provide relative humidity and temperature data. Several months of data have been recorded from four different hives. The mass change measurements, from hive scales, when foragers leave the hive were compared with the data from the gas sensors. Our data suggest that it is possible to estimate the colony size from the change in measured CO₂, however no such link with the humidity is observed. Data are presented showing the CO₂ decreasing over many weeks as a colony dies. Full article

(This article belongs to the Special Issue Selected Papers from the 9th International Electronic Conference on Sensors and Applications)

► Show Figures

Figure 1

Open AccessArticle

Automatic Body Segment and Side Recognition of an Inertial Measurement Unit Sensor during Gait

by

,

,

,

,

and

Sensors 2023, 23(7), 3587; https://doi.org/10.3390/s23073587 (registering DOI) - 29 Mar 2023

Abstract

Inertial measurement unit (IMU) sensors are widely used for motion analysis in sports and rehabilitation. The attachment of IMU sensors to predefined body segments and sides (left/right) is complex, time-consuming, and error-prone. Methods for solving the IMU-2-segment (I2S) pairing work properly only for [...] Read more.

Inertial measurement unit (IMU) sensors are widely used for motion analysis in sports and rehabilitation. The attachment of IMU sensors to predefined body segments and sides (left/right) is complex, time-consuming, and error-prone. Methods for solving the IMU-2-segment (I2S) pairing work properly only for a limited range of gait speeds or require a similar sensor configuration. Our goal was to propose an algorithm that works over a wide range of gait speeds with different sensor configurations while being robust to footwear type and generalizable to pathologic gait patterns. Eight IMU sensors were attached to both feet, shanks, thighs, sacrum, and trunk, and 12 healthy subjects (training dataset) and 22 patients (test dataset) with medial compartment knee osteoarthritis walked at different speeds with/without insole. First, the mean stride time was estimated and IMU signals were scaled. Using a decision tree, the body segment was recognized, followed by the side of the lower limb sensor. The accuracy and precision of the whole algorithm were 99.7% and 99.0%, respectively, for gait speeds ranging from 0.5 to 2.2 m/s. In conclusion, the proposed algorithm was robust to gait speed and footwear type and can be widely used for different sensor configurations. Full article

(This article belongs to the Special Issue Advances in Sensor Technologies for Wearable Applications)

Open AccessArticle

MiCrowd: Vision-Based Deep Crowd Counting on MCU

by

,

,

,

,

and

Sensors 2023, 23(7), 3586; https://doi.org/10.3390/s23073586 (registering DOI) - 29 Mar 2023

Abstract

Microcontrollers (MCUs) have been deployed on numerous IoT devices due to their compact sizes and low costs. MCUs are capable of capturing sensor data and processing them. However, due to their low computational power, applications processing sensor data with deep neural networks (DNNs) [...] Read more.

Microcontrollers (MCUs) have been deployed on numerous IoT devices due to their compact sizes and low costs. MCUs are capable of capturing sensor data and processing them. However, due to their low computational power, applications processing sensor data with deep neural networks (DNNs) have been limited. In this paper, we proposeMiCrowd, a floating population measurement system with a tiny DNNs running on MCUs since the data have essential value in urban planning and business. Moreover, MiCrowd addresses the following important challenges: (1) privacy issues, (2) communication costs, and (3) extreme resource constraints on MCUs. To tackle those challenges, we designed a lightweight crowd-counting deep neural network, named MiCrowdNet, which enables on-MCU inferences. In addition, our dataset is carefully chosen and completely re-labeled to train MiCrowdNet for counting people from an mobility view. Experiments show the effectiveness of MiCrowdNet and our relabeled dataset for accurate on-device crowd counting. Full article

(This article belongs to the Section Industrial Sensors)

Open AccessArticle

Research on Transportation Mode Recognition Based on Multi-Head Attention Temporal Convolutional Network

by

Shuyu Cheng

and

Yingan Liu

Sensors 2023, 23(7), 3585; https://doi.org/10.3390/s23073585 (registering DOI) - 29 Mar 2023

Abstract

Transportation mode recognition is of great importance in analyzing people’s travel patterns and planning urban roads. To make more accurate judgments on the transportation mode of the user, we propose a deep learning fusion model based on multi-head attentional temporal convolution (TCMH). First, [...] Read more.

Transportation mode recognition is of great importance in analyzing people’s travel patterns and planning urban roads. To make more accurate judgments on the transportation mode of the user, we propose a deep learning fusion model based on multi-head attentional temporal convolution (TCMH). First, the time-domain features of a more extensive range of sensor data are mined through a temporal convolutional network. Second, multi-head attention mechanisms are introduced to learn the significance of different features and timesteps, which can improve the identification accuracy. Finally, the deep-learned features are fed into a fully connected layer to output the classification results of the transportation mode. The experimental results demonstrate that the TCMH model achieves an accuracy of 90.25% and 89.55% on the SHL and HTC datasets, respectively, which is 4.45% and 4.70% higher than the optimal value in the baseline algorithm. The model has a better recognition effect on transportation modes. Full article

(This article belongs to the Section Internet of Things)

Open AccessArticle

A Novel Master–Slave Interventional Surgery Robot with Force Feedback and Collaborative Operation

by

,

,

,

and

Sensors 2023, 23(7), 3584; https://doi.org/10.3390/s23073584 (registering DOI) - 29 Mar 2023

Abstract

In recent years, master–slave vascular robots have been developed to address the problem of radiation exposure during vascular interventions for surgeons. However, the single visual feedback reduces surgeon immersion and transparency of the system. In this work, we have developed a haptic interface [...] Read more.

In recent years, master–slave vascular robots have been developed to address the problem of radiation exposure during vascular interventions for surgeons. However, the single visual feedback reduces surgeon immersion and transparency of the system. In this work, we have developed a haptic interface based on the magnetorheological fluid (MRF) on the master side. The haptic interface can provide passive feedback force with high force fidelity and low inertia. Additionally, the manipulation of the master device does not change the operating posture of traditional surgery, which allows the surgeon to better adapt to the robotic system. For the slave robot, the catheter and guidewire can be navigated simultaneously which allows the two degrees of action on the catheter and axial action of a guidewire. The resistance force of the catheter navigation is measured and reflected to the user through the master haptic interface. To verify the proposed master–slave robotic system, the evaluation experiments are carried out in vitro, and the effectiveness of the system was demonstrated experimentally. Full article

(This article belongs to the Special Issue Sensors and Imaging for Medical Robotics)

Open AccessArticle

Statistical Characterization and Modeling of Indoor RF-EMF Down-Link Exposure

by

Biruk Ashenafi Mulugeta

,

,

,

,

and

Sensors 2023, 23(7), 3583; https://doi.org/10.3390/s23073583 (registering DOI) - 29 Mar 2023

Abstract

With the increasing use of wireless communication systems, assessment of exposure to radio-frequency electromagnetic field (RF-EMF) has now become very important due to the rise of public risk perception. Since people spend more than 70% of their daily time in indoor environments, including [...] Read more.

With the increasing use of wireless communication systems, assessment of exposure to radio-frequency electromagnetic field (RF-EMF) has now become very important due to the rise of public risk perception. Since people spend more than 70% of their daily time in indoor environments, including home, office, and car, the efforts devoted to indoor RF-EMF exposure assessment has also increased. However, assessment of indoor exposure to RF-EMF using a deterministic approach is challenging and time consuming task as it is affected by uncertainties due to the complexity of the indoor environment and furniture structure, existence of multiple reflection, refraction, diffraction and scattering, temporal variability of exposure, and existence of many obstructions with unknown dielectric properties. Moreover, it is also affected by the existence of uncontrolled factors that can influence the indoor RF-EMF exposure such as the constant movement of people and random movement of furniture and doors as people are working in the building. In this study, a statistical approach is utilized to characterize and model the total indoor RF-EMF down-link (DL) exposure from all cellular bands on each floor over the length of a wing since the significance of distance is very low between any two points on each floor in a wing and the variation of RF-EMF DL exposure is mainly influenced by the local indoor environment. Measurements were conducted in three buildings that are located within a few hundred meters vicinity of two base station sites supporting several cellular technologies (2G, 3G, 4G, and 5G). We apply the one-sample Kolmogorov–Smirnov test on the measurement data, and we prove that the indoor RF-EMF DL exposure on each floor over the length of a wing is a random process governed by a Gaussian distribution. We validate this proposition using leave-one-out cross validation technique. Consequently, we conclude that the indoor RF-EMF DL exposure on each floor over the length of a wing can be modeled by a Gaussian distribution and, therefore, can be characterized by the mean and the standard deviation parameters. Full article

(This article belongs to the Special Issue Challenges and Solutions in Exposure Assessment for Emerging Wireless Networks)

Open AccessArticle

Data Acquisition Control for UAV-Enabled Wireless Rechargeable Sensor Networks

by

Ikjune Yoon

Sensors 2023, 23(7), 3582; https://doi.org/10.3390/s23073582 (registering DOI) - 29 Mar 2023

Abstract

In the realm of Internet of Things (IoT), wireless sensor networks (WSNs) have been the subject of ongoing research into the use of energy harvesting to capture ambient energy, and wireless power transfer (WPT) via a mobile charger to overcome the energy limitations [...] Read more.

In the realm of Internet of Things (IoT), wireless sensor networks (WSNs) have been the subject of ongoing research into the use of energy harvesting to capture ambient energy, and wireless power transfer (WPT) via a mobile charger to overcome the energy limitations of sensors. Moreover, to mitigate energy imbalance and reduce the number of hops, strategies have been developed to leverage cars or unmanned aerial vehicles (UAVs) as mobile sinks. The primary objective of this work is to increase network lifetime by reducing energy consumption of hotspot nodes and increasing the amount of data acquired from all sensors in an environment that combines the methods mentioned above.To achieve this objective, the proposed method involves developing multiple minimum depth trees (MDTs) for all nodes, considering the energy of the UAV and sensor nodes. Parent nodes prevent their own energy depletion and ensure data transmission without imbalance by adaptively controlling the data sensed at the nodes and their child nodes. Consequently, the energy depletion of nodes in hotspots is prevented, more sensory data is acquired, and balanced data collection from all nodes is achieved. Simulation results demonstrate that the proposed scheme outperforms other state-of-the-art methods in terms of reduced energy depletion, increased network connectivity, and the amount of data collected at the sink node. This scheme will be applied to applications that collect environmental data outdoors, such as climate measurement, to collect data uniformly and increase the lifespan of the network, thereby reducing network maintenance costs while collecting data effectively. Full article

(This article belongs to the Special Issue Wireless Sensing and Networking for the Internet of Things II)

► Show Figures

Figure 1

Open AccessArticle

Practical Implementation of the Indirect Control to the Direct 3 × 5 Matrix Converter Using DSP and Low-Cost FPGA

by

Michal Praženica

,

Patrik Resutík

and

Slavomír Kaščák

Sensors 2023, 23(7), 3581; https://doi.org/10.3390/s23073581 (registering DOI) - 29 Mar 2023

Abstract

The popularity of multiphase drives is increasing due to the growing interest in drives with more than three phases. One promising topology is the multiphase matrix converters, which enable the implementation of a single-stage AC/AC power conversion system with bidirectional power flow capability. [...] Read more.

The popularity of multiphase drives is increasing due to the growing interest in drives with more than three phases. One promising topology is the multiphase matrix converters, which enable the implementation of a single-stage AC/AC power conversion system with bidirectional power flow capability. In this paper, we present the implementation of indirect control for a practical sample of the direct matrix converter. To reduce the overall cost of the control solution for these types of converters, we utilized low-cost FPGA and DSP. The usage of only DSP itself was not possible due to low number of available PWM output needed for 3 × 5 MxC driving. Another reason is commutation, which must be precise and fast to avoid any hazardous states in the converter. Due to these problems, the authors decided to implement an algorithm of a combination of DSP and FPGA, where FPGA is used for time critical operations. The indirect algorithm treats the converter as two separate parts, the rectifier and the inverter, with the DC-LINK being fictitious. The matrix converter is composed of compact modules, and the entire system is verified. The practical verification demonstrates that matrix converters can produce a wide range of output frequencies and achieve input power factor control. Finally, we compare and review the practical model with the simulation model, examining efficiency and other parameters. Full article

(This article belongs to the Special Issue Advanced Sensing and Control Technologies in Power Electronics)

► Show Figures

Figure 1

Open AccessArticle

Facial Motion Capture System Based on Facial Electromyogram and Electrooculogram for Immersive Social Virtual Reality Applications

by

,

,

,

,

and

Sensors 2023, 23(7), 3580; https://doi.org/10.3390/s23073580 (registering DOI) - 29 Mar 2023

Abstract

With the rapid development of virtual reality (VR) technology and the market growth of social network services (SNS), VR-based SNS have been actively developed, in which 3D avatars interact with each other on behalf of the users. To provide the users with more [...] Read more.

With the rapid development of virtual reality (VR) technology and the market growth of social network services (SNS), VR-based SNS have been actively developed, in which 3D avatars interact with each other on behalf of the users. To provide the users with more immersive experiences in a metaverse, facial recognition technologies that can reproduce the user’s facial gestures on their personal avatar are required. However, it is generally difficult to employ traditional camera-based facial tracking technology to recognize the facial expressions of VR users because a large portion of the user’s face is occluded by a VR head-mounted display (HMD). To address this issue, attempts have been made to recognize users’ facial expressions based on facial electromyogram (fEMG) recorded around the eyes. fEMG-based facial expression recognition (FER) technology requires only tiny electrodes that can be readily embedded in the HMD pad that is in contact with the user’s facial skin. Additionally, electrodes recording fEMG signals can simultaneously acquire electrooculogram (EOG) signals, which can be used to track the user’s eyeball movements and detect eye blinks. In this study, we implemented an fEMG- and EOG-based FER system using ten electrodes arranged around the eyes, assuming a commercial VR HMD device. Our FER system could continuously capture various facial motions, including five different lip motions and two different eyebrow motions, from fEMG signals. Unlike previous fEMG-based FER systems that simply classified discrete expressions, with the proposed FER system, natural facial expressions could be continuously projected on the 3D avatar face using machine-learning-based regression with a new concept named the virtual blend shape weight, making it unnecessary to simultaneously record fEMG and camera images for each user. An EOG-based eye tracking system was also implemented for the detection of eye blinks and eye gaze directions using the same electrodes. These two technologies were simultaneously employed to implement a real-time facial motion capture system, which could successfully replicate the user’s facial expressions on a realistic avatar face in real time. To the best of our knowledge, the concurrent use of fEMG and EOG for facial motion capture has not been reported before. Full article

(This article belongs to the Special Issue Sensing Technology in Virtual Reality)

► Show Figures

Figure 1

Open AccessArticle

Sensitivity Analysis of RV Reducer Rotation Error Based on Deep Gaussian Processes

by

,

,

,

and

Sensors 2023, 23(7), 3579; https://doi.org/10.3390/s23073579 (registering DOI) - 29 Mar 2023

Abstract

The rotation error is the most important quality characteristic index of a rotate vector (RV) reducer, and it is difficult to accurately optimize the design of a RV reducer, such as the Taguchi type, due to the many factors affecting the rotation error [...] Read more.

The rotation error is the most important quality characteristic index of a rotate vector (RV) reducer, and it is difficult to accurately optimize the design of a RV reducer, such as the Taguchi type, due to the many factors affecting the rotation error and the serious coupling effect among the factors. This paper analyzes the RV reducer rotation error and each factor based on the deep Gaussian processes (DeepGP) model and Sobol sensitivity analysis(SA) method. Firstly, using the optimal Latin hypercube sampling (OLHS) approach and the DeepGP model, a high-precision regression prediction model of the rotation error and each affecting factor was created. On the basis of the prediction model, the Sobol method was used to conduct a global SA of the factors influencing the rotation error and to compare the coupling relationship between the factors. The results show that the OLHS method and the DeepGP model are suitable for predicting the rotation error in this paper, and the accuracy of the prediction model constructed based on both of them is as high as 95%. The rotation error mainly depends on the influencing factors in the second stage cycloidal pinwheel drive part. The primary involute planetary part and planetary output carrier’s rotation error factors have little effect. The coupling effects between the matching clearance between the pin gear and needle gear hole (

δ J

) and the circular position error of the needle gear hole (

δ t

) is noticeably stronger. Full article

(This article belongs to the Special Issue Applications of Manufacturing and Measurement Sensors)

► Show Figures

Figure 1

Open AccessArticle

Identification, 3D-Reconstruction, and Classification of Dangerous Road Cracks

by

,

,

,

and

Sensors 2023, 23(7), 3578; https://doi.org/10.3390/s23073578 (registering DOI) - 29 Mar 2023

Abstract

Advances in semiconductor technology and wireless sensor networks have permitted the development of automated inspection at diverse scales (machine, human, infrastructure, environment, etc.). However, automated identification of road cracks is still in its early stages. This is largely owing to the difficulty obtaining [...] Read more.

Advances in semiconductor technology and wireless sensor networks have permitted the development of automated inspection at diverse scales (machine, human, infrastructure, environment, etc.). However, automated identification of road cracks is still in its early stages. This is largely owing to the difficulty obtaining pavement photographs and the tiny size of flaws (cracks). The existence of pavement cracks and potholes reduces the value of the infrastructure, thus the severity of the fracture must be estimated. Annually, operators in many nations must audit thousands of kilometers of road to locate this degradation. This procedure is costly, sluggish, and produces fairly subjective results. The goal of this work is to create an efficient automated system for crack identification, extraction, and 3D reconstruction. The creation of crack-free roads is critical to preventing traffic deaths and saving lives. The proposed method consists of five major stages: detection of flaws after processing the input picture with the Gaussian filter, contrast adjustment, and ultimately, threshold-based segmentation. We created a database of road cracks to assess the efficacy of our proposed method. The result obtained are commendable and outperform previous state-of-the-art studies. Full article

(This article belongs to the Special Issue Sensors for Construction Automation and Management)

► Show Figures

Figure 1

Open AccessArticle

Facial Micro-Expression Recognition Using Double-Stream 3D Convolutional Neural Network with Domain Adaptation

by

,

,

and

Sensors 2023, 23(7), 3577; https://doi.org/10.3390/s23073577 (registering DOI) - 29 Mar 2023

Abstract

Humans show micro-expressions (MEs) under some circumstances. MEs are a display of emotions that a human wants to conceal. The recognition of MEs has been applied in various fields. However, automatic ME recognition remains a challenging problem due to two major obstacles. As [...] Read more.

Humans show micro-expressions (MEs) under some circumstances. MEs are a display of emotions that a human wants to conceal. The recognition of MEs has been applied in various fields. However, automatic ME recognition remains a challenging problem due to two major obstacles. As MEs are typically of short duration and low intensity, it is hard to extract discriminative features from ME videos. Moreover, it is tedious to collect ME data. Existing ME datasets usually contain insufficient video samples. In this paper, we propose a deep learning model, double-stream 3D convolutional neural network (DS-3DCNN), for recognizing MEs captured in video. The recognition framework contains two streams of 3D-CNN. The first extracts spatiotemporal features from the raw ME videos. The second extracts variations of the facial motions within the spatiotemporal domain. To facilitate feature extraction, the subtle motion embedded in a ME is amplified. To address the insufficient ME data, a macro-expression dataset is employed to expand the training sample size. Supervised domain adaptation is adopted in model training in order to bridge the difference between ME and macro-expression datasets. The DS-3DCNN model is evaluated on two publicly available ME datasets. The results show that the model outperforms various state-of-the-art models; in particular, the model outperformed the best model presented in MEGC2019 by more than 6%. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

Open AccessArticle

Mushroom Detection and Three Dimensional Pose Estimation from Multi-View Point Clouds

by

George Retsinas

,

Niki Efthymiou

,

Dafni Anagnostopoulou

and

Petros Maragos

Sensors 2023, 23(7), 3576; https://doi.org/10.3390/s23073576 (registering DOI) - 29 Mar 2023

Abstract

Agricultural robotics is an up and coming field which deals with the development of robotic systems able to tackle a multitude of agricultural tasks efficiently. The case of interest, in this work, is mushroom collection in industrial mushroom farms. Developing such a robot, [...] Read more.

Agricultural robotics is an up and coming field which deals with the development of robotic systems able to tackle a multitude of agricultural tasks efficiently. The case of interest, in this work, is mushroom collection in industrial mushroom farms. Developing such a robot, able to select and out-root a mushroom, requires delicate actions that can only be conducted if a well-performing perception module exists. Specifically, one should accurately detect the 3D pose of a mushroom in order to facilitate the smooth operation of the robotic system. In this work, we develop a vision module for 3D pose estimation of mushrooms from multi-view point clouds using multiple RealSense active–stereo cameras. The main challenge is the lack of annotation data, since 3D annotation is practically infeasible on a large scale. To address this, we developed a novel pipeline for mushroom instance segmentation and template matching, where a 3D model of a mushroom is the only data available. We evaluated, quantitatively, our approach over a synthetic dataset of mushroom scenes, and we, further, validated, qualitatively, the effectiveness of our method over a set of real data, collected by different vision settings. Full article

(This article belongs to the Section Smart Agriculture)

► Show Figures

Figure 1

Open AccessArticle

Invertible Privacy-Preserving Adversarial Reconstruction for Image Compressed Sensing

by

Di Xiao

,

Yue Li

and

Min Li

Sensors 2023, 23(7), 3575; https://doi.org/10.3390/s23073575 (registering DOI) - 29 Mar 2023

Abstract

Since the advent of compressed sensing (CS), many reconstruction algorithms have been proposed, most of which are devoted to reconstructing images with better visual quality. However, higher-quality images tend to reveal more sensitive information in machine recognition tasks. In this paper, we propose [...] Read more.

Since the advent of compressed sensing (CS), many reconstruction algorithms have been proposed, most of which are devoted to reconstructing images with better visual quality. However, higher-quality images tend to reveal more sensitive information in machine recognition tasks. In this paper, we propose a novel invertible privacy-preserving adversarial reconstruction method for image CS. While optimizing the quality, the reconstructed images are made to be adversarial samples at the moment of generation. For semi-authorized users, they can only obtain the adversarial reconstructed images, which provide little information for machine recognition or training deep models. For authorized users, they can reverse adversarial reconstructed images to clean samples with an additional restoration network. Experimental results show that while keeping good visual quality for both types of reconstructed images, the proposed scheme can provide semi-authorized users with adversarial reconstructed images with a very low recognizable rate, and allow authorized users to further recover sanitized reconstructed images with recognition performance approximating that of the traditional CS. Full article

(This article belongs to the Special Issue Cyber Security and AI)

► Show Figures

Figure 1

Open AccessArticle

Deep Learning-Based Speech Enhancement of an Extrinsic Fabry–Perot Interferometric Fiber Acoustic Sensor System

by

,

,

and

Sensors 2023, 23(7), 3574; https://doi.org/10.3390/s23073574 (registering DOI) - 29 Mar 2023

Abstract

To achieve high-quality voice communication technology without noise interference in flammable, explosive and strong electromagnetic environments, the speech enhancement technology of a fiber-optic external Fabry–Perot interferometric (EFPI) acoustic sensor based on deep learning is studied in this paper. The combination of a complex-valued [...] Read more.

To achieve high-quality voice communication technology without noise interference in flammable, explosive and strong electromagnetic environments, the speech enhancement technology of a fiber-optic external Fabry–Perot interferometric (EFPI) acoustic sensor based on deep learning is studied in this paper. The combination of a complex-valued convolutional neural network and a long short-term memory (CV-CNN-LSTM) model is proposed for speech enhancement in the EFPI acoustic sensing system. Moreover, the 3 × 3 coupler algorithm is used to demodulate voice signals. Then, the short-time Fourier transform (STFT) spectrogram features of voice signals are divided into a training set and a test set. The training set is input into the established CV-CNN-LSTM model for model training, and the test set is input into the trained model for testing. The experimental findings reveal that the proposed CV-CNN-LSTM model demonstrates exceptional speech enhancement performance, boasting an average Perceptual Evaluation of Speech Quality (PESQ) score of 3.148. In comparison to the CV-CNN and CV-LSTM models, this innovative model achieves a remarkable PESQ score improvement of 9.7% and 11.4%, respectively. Furthermore, the average Short-Time Objective Intelligibility (STOI) score witnesses significant enhancements of 4.04 and 2.83 when contrasted with the CV-CNN and CV-LSTM models, respectively. Full article

(This article belongs to the Special Issue Advanced Optical Sensors Based on Machine Learning)

► Show Figures

Figure 1

Open AccessArticle

SR-FEINR: Continuous Remote Sensing Image Super-Resolution Using Feature-Enhanced Implicit Neural Representation

by

,

,

,

and

Sensors 2023, 23(7), 3573; https://doi.org/10.3390/s23073573 (registering DOI) - 29 Mar 2023

Abstract

Remote sensing images often have limited resolution, which can hinder their effectiveness in various applications. Super-resolution techniques can enhance the resolution of remote sensing images, and arbitrary resolution super-resolution techniques provide additional flexibility in choosing appropriate image resolutions for different tasks. However, for [...] Read more.

Remote sensing images often have limited resolution, which can hinder their effectiveness in various applications. Super-resolution techniques can enhance the resolution of remote sensing images, and arbitrary resolution super-resolution techniques provide additional flexibility in choosing appropriate image resolutions for different tasks. However, for subsequent processing, such as detection and classification, the resolution of the input image may vary greatly for different methods. In this paper, we propose a method for continuous remote sensing image super-resolution using feature-enhanced implicit neural representation (SR-FEINR). Continuous remote sensing image super-resolution means users can scale a low-resolution image into an image with arbitrary resolution. Our algorithm is composed of three main components: a low-resolution image feature extraction module, a positional encoding module, and a feature-enhanced multi-layer perceptron module. We are the first to apply implicit neural representation in a continuous remote sensing image super-resolution task. Through extensive experiments on two popular remote sensing image datasets, we have shown that our SR-FEINR outperforms the state-of-the-art algorithms in terms of accuracy. Our algorithm showed an average improvement of 0.05 dB over the existing method on

\times 30

across three datasets. Full article

(This article belongs to the Special Issue Deep Learning-Based Image and Signal Sensing and Processing)

► Show Figures

Figure 1

Open AccessArticle

Dynamic Obstacle Avoidance for USVs Using Cross-Domain Deep Reinforcement Learning and Neural Network Model Predictive Controller

by

Jianwen Li

,

Jalil Chavez-Galaviz

,

Kamyar Azizzadenesheli

and

Nina Mahmoudian

Sensors 2023, 23(7), 3572; https://doi.org/10.3390/s23073572 (registering DOI) - 29 Mar 2023

Abstract

This work presents a framework that allows Unmanned Surface Vehicles (USVs) to avoid dynamic obstacles through initial training on an Unmanned Ground Vehicle (UGV) and cross-domain retraining on a USV. This is achieved by integrating a Deep Reinforcement Learning (DRL) agent that generates [...] Read more.

This work presents a framework that allows Unmanned Surface Vehicles (USVs) to avoid dynamic obstacles through initial training on an Unmanned Ground Vehicle (UGV) and cross-domain retraining on a USV. This is achieved by integrating a Deep Reinforcement Learning (DRL) agent that generates high-level control commands and leveraging a neural network based model predictive controller (NN-MPC) to reach target waypoints and reject disturbances. A Deep Q Network (DQN) utilized in this framework is trained in a ground environment using a Turtlebot robot and retrained in a water environment using the BREAM USV in the Gazebo simulator to avoid dynamic obstacles. The network is then validated in both simulation and real-world tests. The cross-domain learning largely decreases the training time (

28 %

) and increases the obstacle avoidance performance (70 more reward points) compared to pure water domain training. This methodology shows that it is possible to leverage the data-rich and accessible ground environments to train DRL agent in data-poor and difficult-to-access marine environments. This will allow rapid and iterative agent development without further training due to the change in environment or vehicle dynamics. Full article

(This article belongs to the Special Issue Perception, Planning and Control of Marine Robots)

► Show Figures

Figure 1

Open AccessArticle

A Model for Cognitive Personalization of Microtask Design

by

,

,

,

,

and

Sensors 2023, 23(7), 3571; https://doi.org/10.3390/s23073571 (registering DOI) - 29 Mar 2023

Abstract

The study of data quality in crowdsourcing campaigns is currently a prominent research topic, given the diverse range of participants involved. A potential solution to enhancing data quality processes in crowdsourcing is cognitive personalization, which involves appropriately adapting or assigning tasks based on [...] Read more.

The study of data quality in crowdsourcing campaigns is currently a prominent research topic, given the diverse range of participants involved. A potential solution to enhancing data quality processes in crowdsourcing is cognitive personalization, which involves appropriately adapting or assigning tasks based on a crowd worker’s cognitive profile. There are two common methods for assessing a crowd worker’s cognitive profile: administering online cognitive tests, and inferring behavior from task fingerprinting based on user interaction log events. This article presents the findings of a study that investigated the complementarity of both approaches in a microtask scenario, focusing on personalizing task design. The study involved 134 unique crowd workers recruited from a crowdsourcing marketplace. The main objective was to examine how the administration of cognitive ability tests can be used to allocate crowd workers to microtasks with varying levels of difficulty, including the development of a deep learning model. Another goal was to investigate if task fingerprinting can be used to allocate crowd workers to different microtasks in a personalized manner. The results indicated that both objectives were accomplished, validating the usage of cognitive tests and task fingerprinting as effective mechanisms for microtask personalization, including the development of a deep learning model with 95% accuracy in predicting the accuracy of the microtasks. While we achieved an accuracy of 95%, it is important to note that the small dataset size may have limited the model’s performance. Full article

(This article belongs to the Special Issue Deep Reinforcement Learning and IoT in Intelligent System)

► Show Figures

Figure 1

Open AccessArticle

FedBranch: Leveraging Federated Learning for Anomaly-Aware Load Forecasting in Energy Networks

by

,

,

,

,

,

and

Sensors 2023, 23(7), 3570; https://doi.org/10.3390/s23073570 (registering DOI) - 29 Mar 2023

Abstract

Increased demand for fast edge computation and privacy concerns have shifted researchers’ focus towards a type of distributed learning known as federated learning (FL). Recently, much research has been carried out on FL; however, a major challenge is the need to tackle the [...] Read more.

Increased demand for fast edge computation and privacy concerns have shifted researchers’ focus towards a type of distributed learning known as federated learning (FL). Recently, much research has been carried out on FL; however, a major challenge is the need to tackle the high diversity in different clients. Our research shows that using highly diverse data sets in FL can lead to low accuracy of some local models, which can be categorised as anomalous behaviour. In this paper, we present FedBranch, a clustering-based framework that uses probabilistic methods to create branches of clients and assigns their respective global models. Branching is performed using hidden Markov model clustering (HMM), and a round of branching depends on the diversity of the data. Clustering is performed on Euclidean distances of mean absolute percentage errors (MAPE) obtained from each client at the end of pre-defined communication rounds. The proposed framework was implemented on substation-level energy data with nine clients for short-term load forecasting using an artificial neural network (ANN). FedBranch took two clustering rounds and resulted in two different branches having individual global models. The results show a substantial increase in the average MAPE of all clients; the biggest improvement of 11.36% was observed in one client. Full article

(This article belongs to the Special Issue Machine Learning Techniques for Energy Efficient IoT Networks)

► Show Figures

Figure 1

Journal Description

Sensors

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI