Sensors

19 pages, 941 KiB

Open AccessArticle

Investigating Transfer Learning in Noisy Environments: A Study of Predecessor and Successor Features in Spatial Learning Using a T-Maze

by Incheol Seo and Hyunsu Lee

Sensors 2024, 24(19), 6419; https://doi.org/10.3390/s24196419 - 3 Oct 2024

Abstract

In this study, we investigate the adaptability of artificial agents within a noisy T-maze that use Markov decision processes (MDPs) and successor feature (SF) and predecessor feature (PF) learning algorithms. Our focus is on quantifying how varying the hyperparameters, specifically the reward learning [...] Read more.

In this study, we investigate the adaptability of artificial agents within a noisy T-maze that use Markov decision processes (MDPs) and successor feature (SF) and predecessor feature (PF) learning algorithms. Our focus is on quantifying how varying the hyperparameters, specifically the reward learning rate (

α_{r}

) and the eligibility trace decay rate (

λ

), can enhance their adaptability. Adaptation is evaluated by analyzing the hyperparameters of cumulative reward, step length, adaptation rate, and adaptation step length and the relationships between them using Spearman’s correlation tests and linear regression. Our findings reveal that an

α_{r}

of 0.9 consistently yields superior adaptation across all metrics at a noise level of 0.05. However, the optimal setting for

λ

varies by metric and context. In discussing these results, we emphasize the critical role of hyperparameter optimization in refining the performance and transfer learning efficacy of learning algorithms. This research advances our understanding of the functionality of PF and SF algorithms, particularly in navigating the inherent uncertainty of transfer learning tasks. By offering insights into the optimal hyperparameter configurations, this study contributes to the development of more adaptive and robust learning algorithms, paving the way for future explorations in artificial intelligence and neuroscience. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

12 pages, 4815 KiB

Open AccessArticle

Approximate Observation Weighted ℓ_2/3 SAR Imaging under Compressed Sensing

by Guangtao Li, Dongjin Xin, Weixin Li, Lei Yang, Dong Wang and Yongkang Zhou

Sensors 2024, 24(19), 6418; https://doi.org/10.3390/s24196418 - 3 Oct 2024

Abstract

Compressed Sensing SAR Imaging is based on an accurate observation matrix. As the observed scene enlarges, the resource consumption of the method increases exponentially. In this paper, we propose a weighted ℓ_2/3-norm regularization SAR imaging method based on approximate observation. Initially, [...] Read more.

Compressed Sensing SAR Imaging is based on an accurate observation matrix. As the observed scene enlarges, the resource consumption of the method increases exponentially. In this paper, we propose a weighted ℓ_2/3-norm regularization SAR imaging method based on approximate observation. Initially, to address the issues brought by the precise observation model, we employ an approximate observation operator based on the Chirp Scaling Algorithm as a substitute. Existing approximate observation models typically utilize ℓ_q(q = 1, 1/2)-norm regularization for sparse constraints in imaging. However, these models are not sufficiently effective in terms of sparsity and imaging detail. Finally, to overcome the aforementioned issues, we apply ℓ_2/3 regularization, which aligns with the natural image gradient distribution, and further constrain it using a weighted matrix. This method enhances the sparsity of the algorithm and balances the detail insufficiency caused by the penalty term. Experimental results demonstrate the excellent performance of the proposed method. Full article

(This article belongs to the Section Sensing and Imaging)

► Show Figures

Figure 1

10 pages, 809 KiB

Open AccessArticle

Comparison of Gait in Women with Degenerative Changes of the Hip Joint and Healthy Women Using the MoKA System—A Pilot Study

by Maciej Kuś, Dagmara Wasiuk-Zowada, Magdalena Henke, Justyna Szefler-Derela and Andrzej Knapik

Sensors 2024, 24(19), 6417; https://doi.org/10.3390/s24196417 - 3 Oct 2024

Abstract

Osteoarthritis (OA) is a global problem. There are few reports in the literature regarding the temporal and spatial parameters of gait in people with OA. The aim of this study was to determine spatiotemporal parameters for the pelvis and lower limbs during walking [...] Read more.

Osteoarthritis (OA) is a global problem. There are few reports in the literature regarding the temporal and spatial parameters of gait in people with OA. The aim of this study was to determine spatiotemporal parameters for the pelvis and lower limbs during walking in women with OA and to compare these parameters with healthy people. For this purpose, a 6 min walking test (6MWT) was carried out. OA subjects had worse outcomes compared to the control group (p < 0.05). Data were collected using IMU sensors integrated into the MoKA system and mounted on indicator points on the body. Limited mobility of the pelvis in the frontal plane was observed in the study group, which influenced walking strategy. For the comparison with the control group at each minute, p < 0.05. IMU sensors attached to the body and integrated in one application provide extensive research and diagnostic capabilities. Full article

(This article belongs to the Section Biomedical Sensors)

► Show Figures

Figure 1

18 pages, 18528 KiB

Open AccessArticle

Data Poisoning Attack against Neural Network-Based On-Device Learning Anomaly Detector by Physical Attacks on Sensors

by Takahito Ino, Kota Yoshida, Hiroki Matsutani and Takeshi Fujino

Sensors 2024, 24(19), 6416; https://doi.org/10.3390/s24196416 - 3 Oct 2024

Abstract

In this paper, we introduce a security approach for on-device learning Edge AIs designed to detect abnormal conditions in factory machines. Since Edge AIs are easily accessible by an attacker physically, there are security risks due to physical attacks. In particular, there is [...] Read more.

In this paper, we introduce a security approach for on-device learning Edge AIs designed to detect abnormal conditions in factory machines. Since Edge AIs are easily accessible by an attacker physically, there are security risks due to physical attacks. In particular, there is a concern that the attacker may tamper with the training data of the on-device learning Edge AIs to degrade the task accuracy. Few risk assessments have been reported. It is important to understand these security risks before considering countermeasures. In this paper, we demonstrate a data poisoning attack against an on-device learning Edge AI. Our attack target is an on-device learning anomaly detection system. The system adopts MEMS accelerometers to measure the vibration of factory machines and detect anomalies. The anomaly detector also adopts a concept drift detection algorithm and multiple models to accommodate multiple normal patterns. For the attack, we used a method in which measurements are tampered with by exposing the MEMS accelerometer to acoustic waves of a specific frequency. The acceleration data falsified by this method were trained on an anomaly detector, and the result was that the abnormal state could not be detected. Full article

(This article belongs to the Special Issue Security of Sensor Network Systems and Circuits from a Hardware Perspective)

► Show Figures

Figure 1

14 pages, 3768 KiB

Open AccessArticle

High-Precision Photonics-Assisted Two-Step Microwave Frequency Measurement Combining Time and Power Mapping Method

by Zhangyi Yang, Zuoheng Liu, Yuqing Jiang, Hanbo Liu, Jiaqi Li and Wei Dong

Sensors 2024, 24(19), 6415; https://doi.org/10.3390/s24196415 - 3 Oct 2024

Abstract

Photonics-assisted methods for microwave frequency measurement (MFM) show great potential for overcoming electronic bottlenecks and offer promising applications in radar and communication due to their wide bandwidth and immunity to electromagnetic interference. In common photonics-assisted MFM methods, the frequency-to-time mapping (FTTM) method has [...] Read more.

Photonics-assisted methods for microwave frequency measurement (MFM) show great potential for overcoming electronic bottlenecks and offer promising applications in radar and communication due to their wide bandwidth and immunity to electromagnetic interference. In common photonics-assisted MFM methods, the frequency-to-time mapping (FTTM) method has the capability to measure various types of signals, but with a trade-off between measurement error, measurement range, and real-time performance, while the frequency-to-power mapping (FTPM) method offers low measurement error but faces great difficulty in measuring signal types other than single-tone signals. In this paper, a two-step high-precision MFM method based on the combination of FTTM and FTPM is proposed, which balances real-time performance with measurement precision and resolution compared with other similar works based on the FTTM method. By utilizing high-speed optical sweeping and an optical filter based on stimulated Brillouin scattering (SBS), FTTM is accomplished, enabling the rough identification of multiple different signals. Next, based on the results from the previous step, more precise measurement results can be calculated from several additional sampling points according to the FTPM principle. The demonstration system can perform optical sweeping at a speed of 20 GHz/μs in the measurement range of 1–18 GHz, with a measurement error of less than 10 MHz and a frequency resolution of 40 MHz. Full article

(This article belongs to the Special Issue Advanced Microwave Sensors and Their Applications in Measurement)

► Show Figures

Figure 1

45 pages, 1088 KiB

Open AccessReview

A Survey of Advanced Border Gateway Protocol Attack Detection Techniques

by Ben A. Scott, Michael N. Johnstone and Patryk Szewczyk

Sensors 2024, 24(19), 6414; https://doi.org/10.3390/s24196414 - 3 Oct 2024

Abstract

The Internet’s default inter-domain routing system, the Border Gateway Protocol (BGP), remains insecure. Detection techniques are dominated by approaches that involve large numbers of features, parameters, domain-specific tuning, and training, often contributing to an unacceptable computational cost. Efforts to detect anomalous activity in [...] Read more.

The Internet’s default inter-domain routing system, the Border Gateway Protocol (BGP), remains insecure. Detection techniques are dominated by approaches that involve large numbers of features, parameters, domain-specific tuning, and training, often contributing to an unacceptable computational cost. Efforts to detect anomalous activity in the BGP have been almost exclusively focused on single observable monitoring points and Autonomous Systems (ASs). BGP attacks can exploit and evade these limitations. In this paper, we review and evaluate categories of BGP attacks based on their complexity. Previously identified next-generation BGP detection techniques remain incapable of detecting advanced attacks that exploit single observable detection approaches and those designed to evade public routing monitor infrastructures. Advanced BGP attack detection requires lightweight, rapid capabilities with the capacity to quantify group-level multi-viewpoint interactions, dynamics, and information. We term this approach advanced BGP anomaly detection. This survey evaluates 178 anomaly detection techniques and identifies which are candidates for advanced attack anomaly detection. Preliminary findings from an exploratory investigation of advanced BGP attack candidates are also reported. Full article

(This article belongs to the Section Sensor Networks)

26 pages, 542 KiB

Open AccessReview

WiFi-Based Human Identification with Machine Learning: A Comprehensive Survey

by Manal Mosharaf, Jae B. Kwak and Wooyeol Choi

Sensors 2024, 24(19), 6413; https://doi.org/10.3390/s24196413 - 3 Oct 2024

Abstract

In the modern world of human–computer interaction, notable advancements in human identification have been achieved across fields like healthcare, academia, security, etc. Despite these advancements, challenges remain, particularly in scenarios with poor lighting, occlusion, or non-line-of-sight. To overcome these limitations, the utilization of [...] Read more.

In the modern world of human–computer interaction, notable advancements in human identification have been achieved across fields like healthcare, academia, security, etc. Despite these advancements, challenges remain, particularly in scenarios with poor lighting, occlusion, or non-line-of-sight. To overcome these limitations, the utilization of radio frequency (RF) wireless signals, particularly wireless fidelity (WiFi), has been considered an innovative solution in recent research studies. By analyzing WiFi signal fluctuations caused by human presence, researchers have developed machine learning (ML) models that significantly improve identification accuracy. This paper conducts a comprehensive survey of recent advances and practical implementations of WiFi-based human identification. Furthermore, it covers the ML models used for human identification, system overviews, and detailed WiFi-based human identification methods. It also includes system evaluation, discussion, and future trends related to human identification. Finally, we conclude by examining the limitations of the research and discussing how researchers can shift their attention toward shaping the future trajectory of human identification through wireless signals. Full article

(This article belongs to the Special Issue Wireless Sensors and Wireless Sensor Networks for Engineering Applications)

► Show Figures

Figure 1

14 pages, 2866 KiB

Open AccessArticle

Cyber–Physical Perception Interface for Co-Simulation Applications

by Teodora Mîndra and Ana Magdalena Anghel

Sensors 2024, 24(19), 6412; https://doi.org/10.3390/s24196412 - 3 Oct 2024

Abstract

Co-simulation can bring improvements to the development of cyber–physical perceptive systems (CPPS) in critical fields, allowing uninterrupted system operation and flexibility to use both real-time sensor data and non-real-time data. This paper proposes a co-simulation approach that integrates physical systems and communication systems, [...] Read more.

Co-simulation can bring improvements to the development of cyber–physical perceptive systems (CPPS) in critical fields, allowing uninterrupted system operation and flexibility to use both real-time sensor data and non-real-time data. This paper proposes a co-simulation approach that integrates physical systems and communication systems, including both hardware and software components. This study demonstrates how systems of different natures with discrete or continuous events can be simulated using three methods: time stepped, global event driven, and variable stepped. Through two case studies from the medical and energy fields, CPPS and co-simulation reveal their importance for the future by improving precision and efficiency, which leads to more accurate diagnoses and personalized treatments in the medical field and increases the stability of energy networks. Full article

(This article belongs to the Section Sensors and Robotics)

► Show Figures

Figure 1

24 pages, 14371 KiB

Open AccessArticle

An Enhanced Transportation System for People of Determination

by Uma Perumal, Fathe Jeribi and Mohammed Hameed Alhameed

Sensors 2024, 24(19), 6411; https://doi.org/10.3390/s24196411 - 3 Oct 2024

Abstract

Visually Impaired Persons (VIPs) have difficulty in recognizing vehicles used for navigation. Additionally, they may not be able to identify the bus to their desired destination. However, the bus bay in which the designated bus stops has not been analyzed in the existing [...] Read more.

Visually Impaired Persons (VIPs) have difficulty in recognizing vehicles used for navigation. Additionally, they may not be able to identify the bus to their desired destination. However, the bus bay in which the designated bus stops has not been analyzed in the existing literature. Thus, a guidance system for VIPs that identifies the correct bus for transportation is presented in this paper. Initially, speech data indicating the VIP’s destination are pre-processed and converted to text. Next, utilizing the Arctan Gradient-activated Recurrent Neural Network (ArcGRNN) model, the number of bays at the location is detected with the help of a Global Positioning System (GPS), input text, and bay location details. Then, the optimal bay is chosen from the detected bays by utilizing the Experienced Perturbed Bacteria Foraging Triangular Optimization Algorithm (EPBFTOA), and an image of the selected bay is captured and pre-processed. Next, the bus is identified utilizing a You Only Look Once (YOLO) series model. Utilizing the Sub-pixel Shuffling Convoluted Encoder–ArcGRNN Decoder (SSCEAD) framework, the text is detected and segmented for the buses identified in the image. From the segmented output, the text is extracted, based on the destination and route of the bus. Finally, regarding the similarity value with respect to the VIP’s destination, a decision is made utilizing the Multi-characteristic Non-linear S-Curve-Fuzzy Rule (MNC-FR). This decision informs the bus conductor about the VIP, such that the bus can be stopped appropriately to pick them up. During testing, the proposed system selected the optimal bay in 247,891 ms, which led to deciding the bus stop for the VIP with a fuzzification time of 34,197 ms. Thus, the proposed model exhibits superior performance over those utilized in prevailing works. Full article

(This article belongs to the Section Intelligent Sensors)

► Show Figures

Figure 1

18 pages, 9588 KiB

Open AccessArticle

Vehicle and Pedestrian Traffic Signal Performance Measures Using LiDAR-Derived Trajectory Data

by Enrique D. Saldivar-Carranza, Jairaj Desai, Andrew Thompson, Mark Taylor, James Sturdevant and Darcy M. Bullock

Sensors 2024, 24(19), 6410; https://doi.org/10.3390/s24196410 - 3 Oct 2024

Abstract

Light Detection and Ranging (LiDAR) sensors at signalized intersections can accurately track the movement of virtually all objects passing through at high sampling rates. This study presents methodologies to estimate vehicle and pedestrian traffic signal performance measures using LiDAR trajectory data. Over 15,000,000 [...] Read more.

Light Detection and Ranging (LiDAR) sensors at signalized intersections can accurately track the movement of virtually all objects passing through at high sampling rates. This study presents methodologies to estimate vehicle and pedestrian traffic signal performance measures using LiDAR trajectory data. Over 15,000,000 vehicle and 170,000 pedestrian waypoints detected during a 24 h period at an intersection in Utah are analyzed to describe the proposed techniques. Sampled trajectories are linear referenced to generate Purdue Probe Diagrams (PPDs). Vehicle-based PPDs are used to estimate movement level turning counts, 85th percentile queue lengths (85QL), arrivals on green (AOG), highway capacity manual (HCM) level of service (LOS), split failures (SF), and downstream blockage (DSB) by time of day (TOD). Pedestrian-based PPDs are used to estimate wait times and the proportion of people that traverse multiple crosswalks. Although vehicle signal performance can be estimated from several days of aggregated connected vehicle (CV) data, LiDAR data provides the ability to measure performance in real time. Furthermore, LiDAR can measure pedestrian speeds. At the studied location, the 15th percentile pedestrian walking speed was estimated to be 3.9 ft/s. The ability to directly measure these pedestrian speeds allows agencies to consider alternative crossing times than those suggested by the Manual on Uniform Traffic Control Devices (MUTCD). Full article

(This article belongs to the Section Radar Sensors)

22 pages, 21022 KiB

Open AccessArticle

Ego-Vehicle Speed Correction for Automotive Radar Systems Using Convolutional Neural Networks

by Sunghoon Moon, Daehyun Kim and Younglok Kim

Sensors 2024, 24(19), 6409; https://doi.org/10.3390/s24196409 - 3 Oct 2024

Abstract

The development of autonomous driving vehicles has increased the global demand for robust and efficient automotive radar systems. This study proposes an automotive radar-based ego-vehicle speed detection network (AVSD Net) model using convolutional neural networks for estimating the speed of the ego vehicle. [...] Read more.

The development of autonomous driving vehicles has increased the global demand for robust and efficient automotive radar systems. This study proposes an automotive radar-based ego-vehicle speed detection network (AVSD Net) model using convolutional neural networks for estimating the speed of the ego vehicle. The preprocessing and postprocessing methods used for vehicle speed correction are presented in detail. The AVSD Net model exhibits characteristics that are independent of the angular performance of the radar system and its mounting angle on the vehicle, thereby reducing the loss of the maximum detection range without requiring a downward or wide beam for the elevation angle. The ego-vehicle speed is effectively estimated when the range–velocity spectrum data are input into the model. Moreover, preprocessing and postprocessing facilitate an accurate correction of the ego-vehicle speed while reducing the complexity of the model, enabling its application to embedded systems. The proposed ego-vehicle speed correction method can improve safety in various applications, such as autonomous emergency braking systems, forward collision avoidance assist, adaptive cruise control, rear cross-traffic alert, and blind spot detection systems. Full article

(This article belongs to the Section Radar Sensors)

► Show Figures

Figure 1

22 pages, 7802 KiB

Open AccessArticle

Study on Bionic Design and Tissue Manipulation of Breast Interventional Robot

by Weixi Zhang, Jiaxing Yu, Xiaoyang Yu, Yongde Zhang and Zhihui Men

Sensors 2024, 24(19), 6408; https://doi.org/10.3390/s24196408 - 3 Oct 2024

Abstract

Minimally invasive interventional surgery is commonly used for diagnosing and treating breast cancer, but the high fluidity and deformability of breast tissue reduce intervention accuracy. This study proposes a bionic breast interventional robot that mimics the scorpion’s predation process, actively manipulating tissue deformation [...] Read more.

Minimally invasive interventional surgery is commonly used for diagnosing and treating breast cancer, but the high fluidity and deformability of breast tissue reduce intervention accuracy. This study proposes a bionic breast interventional robot that mimics the scorpion’s predation process, actively manipulating tissue deformation to control target displacement and enhance accuracy. The robot’s structure is designed using a modular method, and its kinematics and workspace are analyzed and solved. To address the nonlinear breast tissue deformation problem, a hierarchical tissue method is proposed to simplify the three-dimensional problem into a two-dimensional one. A two-dimensional tissue deformation solver is established based on the minimum energy method for quick resolution. The problem is treated as quasi-static, deriving the displacement relationship between external manipulation points and internal tissue targets. The method of active manipulation of tissue deformation is simulated using MATLAB (2019-b) software to verify the feasibility of the method. Results show maximum errors of 1.7 mm for prostheses and 2.5 mm for in vitro tissues in the X and Y directions. This method improves intervention accuracy in breast surgery and offers a new solution for breast cancer diagnosis and treatment. Full article

(This article belongs to the Collection Biomedical Imaging & Instrumentation)

► Show Figures

Figure 1

13 pages, 37705 KiB

Open AccessArticle

Design and Experimental Test of Rope-Driven Force Sensing Flexible Gripper

by Zuhao Zhu, Yufei Liu, Jinyong Ju and En Lu

Sensors 2024, 24(19), 6407; https://doi.org/10.3390/s24196407 - 3 Oct 2024

Abstract

Robotic grasping is a common operation scenario in industry and agriculture, in which the force sensing function is a significant factor to achieve reliable grasping. Existing force sensing methods of flexible grippers require intelligent materials or force sensors embedded in the flexible gripper, [...] Read more.

Robotic grasping is a common operation scenario in industry and agriculture, in which the force sensing function is a significant factor to achieve reliable grasping. Existing force sensing methods of flexible grippers require intelligent materials or force sensors embedded in the flexible gripper, which causes such problems of higher manufacturing requirements and contact surface properties changing. In this paper, a novel rope-driven force sensing flexible gripper is designed based on the fin-shaped gripper structure, which can realize the grasping sensing functions of contact nodes and contact forces without the need for force sensors. Firstly, the rope-driven force sensing flexible gripper is designed, including the driving unit, the transmission part, the gripper unit, and the force sensing unit. The force sensing unit and the gripper unit are connected by rope, and the prototype of the rope-driven force sensing flexible gripper is completed. Secondly, a force sensing algorithm and control system based on finite element method and grasping geometric relationship are designed to realize the rope-driven force sensing flexible gripper grasping control and sensor data acquisition and processing. Finally, the experimental system of the rope-driven force sensing flexible gripper is built, and the grasping experimental tests of objects with different diameters and different contact nodes are carried out to verify the force sensing function of the rope-driven force sensing flexible gripper. The force sensing flexible gripper designed in this paper can provide a new idea for the design and force sensing method of intelligent robotic grasping system in robotic teaching, scientific research, and industrial applications. Full article

(This article belongs to the Special Issue Sensors and Robotic Systems for Agriculture Applications)

► Show Figures

Figure 1

21 pages, 2184 KiB

Open AccessArticle

Design of a High-Power Nanosecond Electromagnetic Pulse Radiation System for Verifying Spaceborne Detectors

by Tianchi Zhang, Zongxiang Li, Changjiao Duan, Lihua Wang, Yongli Wei, Kejie Li, Xin Li and Baofeng Cao

Sensors 2024, 24(19), 6406; https://doi.org/10.3390/s24196406 - 2 Oct 2024

Abstract

The Spaceborne Global Lightning Location Network (SGLLN) serves the purpose of identifying transient lightning events occurring beneath the ionosphere, playing a significant role in detecting and warning of disaster weather events. To ensure the effective functioning of the wideband electromagnetic pulse detector, which [...] Read more.

The Spaceborne Global Lightning Location Network (SGLLN) serves the purpose of identifying transient lightning events occurring beneath the ionosphere, playing a significant role in detecting and warning of disaster weather events. To ensure the effective functioning of the wideband electromagnetic pulse detector, which is a crucial component of the SGLLN, it must be tested and verified with specific signals. However, the inherent randomness and unpredictability of lightning occurrences pose challenges to this requirement. Consequently, a high-power electromagnetic pulse radiation system with a 20 m aperture reflector is designed. This system is capable of emitting nanosecond electromagnetic pulse signals under pre-set spatial and temporal conditions, providing a controlled environment for assessing the detection capabilities of SGLLN. In the design phase, an exponentially TEM feed antenna has been designed firstly based on the principle of high-gain radiation. The feed antenna adopts a pulser-integrated design to mitigate insulation risks, and it is equipped with an asymmetric protective loading to reduce reflected energy by 85.7%. Moreover, an innovative assessment method for gain loss, based on the principle of Love’s equivalence, is proposed to quantify the impact of feed antenna on the radiation field. During the experimental phase, a specialized E-field sensor is used in the far-field experiment at a distance of 400 m. The measurements indicate that at this distance, the signal has a peak field strength of 2.2 kV/m, a rise time of 1.9 ns, and a pulse half-width of 2.5 ns. Additionally, the beamwidth in the time domain is less than 10°. At an altitude of 500 km, the spaceborne detector records a signal with a peak field strength of approximately 10 mV/m. Particularly, this signal transformed into a nonlinear frequency-modulated signal in the microsecond range across its frequency spectrum, which is consistent with the law of radio wave propagation in the ionosphere. This study offers a stable and robust radiation source for verifying spaceborne detectors and establishes an empirical foundation for investigating the impact of the ionosphere on signal propagation characteristics. Full article

(This article belongs to the Section Electronic Sensors)

16 pages, 10550 KiB

Open AccessArticle

Click to Correction: Interactive Bidirectional Dynamic Propagation Video Object Segmentation Network

by Shuting Yang, Xia Yuan and Sihan Luo

Sensors 2024, 24(19), 6405; https://doi.org/10.3390/s24196405 - 2 Oct 2024

Abstract

High-quality video object segmentation is a challenging visual computing task. Interactive segmentation can improve segmentation results. This paper proposes a multi-round interactive dynamic propagation instance-level video object segmentation network based on click interaction. The network consists of two parts: a user interaction segmentation [...] Read more.

High-quality video object segmentation is a challenging visual computing task. Interactive segmentation can improve segmentation results. This paper proposes a multi-round interactive dynamic propagation instance-level video object segmentation network based on click interaction. The network consists of two parts: a user interaction segmentation module and a bidirectional dynamic propagation module. A prior segmentation network was designed in the user interaction segmentation module to better segment objects of different scales that users click on. The dynamic propagation network achieves high-precision video object segmentation through the bidirectional propagation and fusion of segmentation masks obtained from multiple rounds of interaction. Experiments on interactive segmentation datasets and video object segmentation datasets show that our method achieves state-of-the-art segmentation results with fewer click interactions. Full article

(This article belongs to the Section Internet of Things)

19 pages, 3054 KiB

Open AccessArticle

Performance Evaluation of Real-Time Kinematic Global Navigation Satellite System with Survey-Grade Receivers and Short Observation Times in Forested Areas

by Mihnea Cățeanu and Maria Alexandra Moroianu

Sensors 2024, 24(19), 6404; https://doi.org/10.3390/s24196404 - 2 Oct 2024

Abstract

The Real-Time Kinematic (RTK) method is currently the most widely used method for positioning using Global Navigation Satellite Systems (GNSSs) due to its accuracy, efficiency and ease of use. In forestry, position is a critical factor for numerous applications, with GNSS currently being [...] Read more.

The Real-Time Kinematic (RTK) method is currently the most widely used method for positioning using Global Navigation Satellite Systems (GNSSs) due to its accuracy, efficiency and ease of use. In forestry, position is a critical factor for numerous applications, with GNSS currently being the preferred solution for obtaining such data. However, the decreased performance of GNSS observations in challenging environments, such as under the forest canopy, must be considered. This paper analyzes the performance of a survey-grade GNSS receiver under coniferous/deciduous tree cover. Unlike most previous research concerning this topic, the focus here is on employing a methodology that is as close as possible to real working conditions in the field of forestry. To achieve this, short observation times of 30 s were used, with corrections received directly in the field from a Continuously Operating Reference Station (CORS) of the national RTK network in Romania. In total, 84 test points were determined, randomly distributed under the canopy, with reference data collected by topographical surveys using total station equipment. In terms of the overall horizontal accuracy, an RMSE of 2.03 m and MAE of 1.63 m are found. Meanwhile, the overall vertical accuracy is lower, as expected, with an RMSE of 4.85 m and MAE of 4.01 m. The variation in GNSS performance under the different forest compositions was found to be statistically significant, while GNSS-specific factors such as DOP values only influenced the precision and not the accuracy of observations. We established that this methodology offers sufficient accuracy, which is application-dependent, even if the majority of GNSS solutions were code-based, rather than carrier-phase-based, due to strong interference from the vegetation. Full article

(This article belongs to the Special Issue GNSS Signals and Precise Point Positioning)

11 pages, 4060 KiB

Open AccessCommunication

Study of a Crosstalk Suppression Scheme Based on Double-Stage Semiconductor Optical Amplifiers

by Xintong Lu, Xinyu Ma and Baojian Wu

Sensors 2024, 24(19), 6403; https://doi.org/10.3390/s24196403 - 2 Oct 2024

Abstract

An all-optical crosstalk suppression scheme is desirable for wavelength and space division multiplexing optical networks by improving the performance of the corresponding nodes. We put forward a scheme comprising double-stage semiconductor optical amplifiers (SOAs) for wavelength-preserving crosstalk suppression. The wavelength position of the [...] Read more.

An all-optical crosstalk suppression scheme is desirable for wavelength and space division multiplexing optical networks by improving the performance of the corresponding nodes. We put forward a scheme comprising double-stage semiconductor optical amplifiers (SOAs) for wavelength-preserving crosstalk suppression. The wavelength position of the degenerate pump in the optical phase conjugation (OPC) is optimized for signal-to-crosstalk ratio (SXR) improvement. The crosstalk suppression performance of the double-stage SOA scheme for 20 Gb/s quadrature phase shift keying (QPSK) signals is investigated by means of simulations, including the input SXR range and the crosstalk wavelength deviation. For the case with identical-frequency crosstalk, the double-stage SOA scheme can achieve equivalent SXR improvement of 1.5 dB for an input SXR of 10 dB. Thus, the double-stage SOA scheme proposed here is more suitable for few-mode fiber systems and networks. Full article

(This article belongs to the Special Issue Signal Processing for Next-Generation Optical Communications and Networks)

► Show Figures

Figure 1

15 pages, 3636 KiB

Open AccessArticle

Extraction of Features for Time Series Classification Using Noise Injection

by Gyu Il Kim and Kyungyong Chung

Sensors 2024, 24(19), 6402; https://doi.org/10.3390/s24196402 - 2 Oct 2024

Abstract

Time series data often display complex, time-varying patterns, which pose significant challenges for effective classification due to data variability, noise, and imbalance. Traditional time series classification techniques frequently fall short in addressing these issues, leading to reduced generalization performance. Therefore, there is a [...] Read more.

Time series data often display complex, time-varying patterns, which pose significant challenges for effective classification due to data variability, noise, and imbalance. Traditional time series classification techniques frequently fall short in addressing these issues, leading to reduced generalization performance. Therefore, there is a need for innovative methodologies to enhance data diversity and quality. In this paper, we introduce a method for the extraction of features for time series classification using noise injection to address these challenges. By employing noise injection techniques for data augmentation, we enhance the diversity of the training data. Utilizing digital signal processing (DSP), we extract key frequency features from time series data through sampling, quantization, and Fourier transformation. This process enhances the quality of the training data, thereby maximizing the model’s generalization performance. We demonstrate the superiority of our proposed method by comparing it with existing time series classification models. Additionally, we validate the effectiveness of our approach through various experimental results, confirming that data augmentation and DSP techniques are potent tools in time series data classification. Ultimately, this research presents a robust methodology for time series data analysis and classification, with potential applications across a broad spectrum of data analysis problems. Full article

(This article belongs to the Special Issue Augmented Reality-Based Navigation System for Healthcare)

► Show Figures

Figure 1

14 pages, 1186 KiB

Open AccessArticle

Multi-Stage Corn-to-Syrup Process Monitoring and Yield Prediction Using Machine Learning and Statistical Methods

by Sheng-Jen Hsieh and Jeff Hykin

Sensors 2024, 24(19), 6401; https://doi.org/10.3390/s24196401 - 2 Oct 2024

Abstract

Corn syrup is a cost-effective sweetener ingredient for the food industry. In producing syrup from corn, process control to enhance and/or maintain a constant dextrose equivalent value (DE) is a constant challenge, especially in semi-automated/batch production settings, which are common in small to [...] Read more.

Corn syrup is a cost-effective sweetener ingredient for the food industry. In producing syrup from corn, process control to enhance and/or maintain a constant dextrose equivalent value (DE) is a constant challenge, especially in semi-automated/batch production settings, which are common in small to medium-size factories. Existing work has focused on continuous process control to keep parameter values within a setpoint. The machine learning method applied is for time series data. This study focuses on building process control models to enable semi-automation in small to medium-size factories in which the data are not as time dependent. Correlation coefficients were used to identify key process parameters that contribute to feed pH value and DE. Artificial neural network (ANN), support vector machine (SVM), and linear regression (LR) models were built to predict feed pH and DE. The results suggest (1) model accuracy ranges from 91% to 96%; (2) the ANN models yielded about 1% to 3% higher accuracy than the SVM and LR models and the prediction accuracy is robust even with as few as six data sets; (3) both the SVM and ANN models have noise tolerant properties, but ANN has a higher noise tolerance than SVM; (4) SVM performance can be hindered when using high-dimensional data sets; (5) the LR model yields higher variation in accuracy prediction than ANN and SVM; (6) distribution fitting is a good approach for generating data; however, fidelity of fitting can greatly impact accuracy; and (7) multi-stage models yield higher accuracy than single-stage models, but there are pros and cons to each approach. Full article

(This article belongs to the Section Industrial Sensors)

► 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