Sensors最新文献

{"title":"Research on Network Intrusion Detection Model Based on Hybrid Sampling and Deep Learning.","authors":"Derui Guo, Yufei Xie","doi":"10.3390/s25051578","DOIUrl":"10.3390/s25051578","url":null,"abstract":"<p><p>This study proposes an enhanced network intrusion detection model, 1D-TCN-ResNet-BiGRU-Multi-Head Attention (TRBMA), aimed at addressing the issues of incomplete learning of temporal features and low accuracy in the classification of malicious traffic found in existing models. The TRBMA model utilizes Temporal Convolutional Networks (TCNs) to improve the ResNet18 architecture and incorporates Bidirectional Gated Recurrent Units (BiGRUs) and Multi-Head Self-Attention mechanisms to enhance the comprehensive learning of temporal features. Additionally, the ResNet network is adapted into a one-dimensional version that is more suitable for processing time-series data, while the AdamW optimizer is employed to improve the convergence speed and generalization ability during model training. Experimental results on the CIC-IDS-2017 dataset indicate that the TRBMA model achieves an accuracy of 98.66% in predicting malicious traffic types, with improvements in precision, recall, and F1-score compared to the baseline model. Furthermore, to address the challenge of low identification rates for malicious traffic types with small sample sizes in unbalanced datasets, this paper introduces TRBMA (BS-OSS), a variant of the TRBMA model that integrates Borderline SMOTE-OSS hybrid sampling. Experimental results demonstrate that this model effectively identifies malicious traffic types with small sample sizes, achieving an overall prediction accuracy of 99.88%, thereby significantly enhancing the performance of the network intrusion detection model.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902645/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Skeleton Reconstruction Using Generative Adversarial Networks for Human Activity Recognition Under Occlusion.","authors":"Ioannis Vernikos, Evaggelos Spyrou","doi":"10.3390/s25051567","DOIUrl":"10.3390/s25051567","url":null,"abstract":"<p><p>Recognizing human activities from motion data is a complex task in computer vision, involving the recognition of human behaviors from sequences of 3D motion data. These activities encompass successive body part movements, interactions with objects, or group dynamics. Camera-based recognition methods are cost-effective and perform well under controlled conditions but face challenges in real-world scenarios due to factors such as viewpoint changes, illumination variations, and occlusion. The latter is the most significant challenge in real-world recognition; partial occlusion impacts recognition accuracy to varying degrees depending on the activity and the occluded body parts while complete occlusion can render activity recognition impossible. In this paper, we propose a novel approach for human activity recognition in the presence of partial occlusion, which may be applied in cases wherein up to two body parts are occluded. The proposed approach works under the assumptions that (a) human motion is modeled using a set of 3D skeletal joints, and (b) the same body parts remain occluded throughout the whole activity. Contrary to previous research, in this work, we address this problem using a Generative Adversarial Network (GAN). Specifically, we train a Convolutional Recurrent Neural Network (CRNN), whose goal is to serve as the generator of the GAN. Its aim is to complete the missing parts of the skeleton due to occlusion. Specifically, the input to this CRNN consists of raw 3D skeleton joint positions, upon the removal of joints corresponding to occluded parts. The output of the CRNN is a reconstructed skeleton. For the discriminator of the GAN, we use a simple long short-term memory (LSTM) network. We evaluate the proposed approach using publicly available datasets in a series of occlusion scenarios. We demonstrate that in all scenarios, the occlusion of certain body parts causes a significant decline in performance, although in some cases, the reconstruction process leads to almost perfect recognition. Nonetheless, in almost every circumstance, the herein proposed approach exhibits superior performance compared to previous works, which varies between 2.2% and 37.5%, depending on the dataset used and the occlusion case.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902716/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CNN-Based Classification of Optically Critical Cutting Tools with Complex Geometry: New Insights for CNN-Based Classification Tasks.","authors":"Mühenad Bilal, Ranadheer Podishetti, Tangirala Sri Girish, Daniel Grossmann, Markus Bregulla","doi":"10.3390/s25051575","DOIUrl":"10.3390/s25051575","url":null,"abstract":"<p><p>Sustainability has increasingly emphasized the importance of recycling and repairing materials. Cutting tools, such as milling cutters and drills, play a crucial role due to the high demands placed on products used in CNC machining. As a result, the repair and regrinding of these tools have become more essential. The geometric differences among machining tools determine their specific applications: twist drills have spiral flutes and pointed cutting edges designed for drilling, while end mills feature multiple sharp edges around the shank, making them suitable for milling. Taps and form cutters exhibit unique geometries and cutting-edge shapes, enabling the creation of complex profiles. However, measuring and classifying these tools for repair or regrinding is challenging due to their optical properties and coatings. This research investigates how lighting conditions affect the classification of tools for regrinding, addressing the shortage of skilled workers and the increasing need for automation. This paper compares different training strategies on two unique tool-specific datasets, each containing 36 distinct tools recorded under two lighting conditions-direct diffuse ring lighting and normal daylight. Furthermore, Grad-CAM heatmap analysis provides new insights into relevant classification features.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902434/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Texture Reconstructive Downsampling for Multi-Scale Object Detection in UAV Remote-Sensing Images.","authors":"Wenhao Zheng, Bangshu Xiong, Jiujiu Chen, Qiaofeng Ou, Lei Yu","doi":"10.3390/s25051569","DOIUrl":"10.3390/s25051569","url":null,"abstract":"<p><p>Unmanned aerial vehicle (UAV) remote-sensing images present unique challenges to the object-detection task due to uneven object densities, low resolution, and drastic scale variations. Downsampling is an important component of deep networks that expands the receptive field, reduces computational overhead, and aggregates features. However, object detectors using multi-layer downsampling result in varying degrees of texture feature loss for various scales in remote-sensing images, degrading the performance of multi-scale object detection. To alleviate this problem, we propose a lightweight texture reconstructive downsampling module called TRD. TRD models part of the texture features lost as residual information during downsampling. After modeling, cascading downsampling and upsampling operators provide residual feedback to guide the reconstruction of the desired feature map for each downsampling stage. TRD structurally optimizes the feature-extraction capability of downsampling to provide sufficiently discriminative features for subsequent vision tasks. We replace the downsampling module of the existing backbone network with the TRD module and conduct a large number of experiments and ablation studies on a variety of remote-sensing image datasets. Specifically, the proposed TRD module improves 3.1% AP over the baseline on the NWPU VHR-10 dataset. On the VisDrone-DET dataset, the TRD improves 3.2% AP over the baseline with little additional cost, especially the APS, APM, and APL by 3.1%, 8.8%, and 13.9%, respectively. The results show that TRD enriches the feature information after downsampling and effectively improves the multi-scale object-detection accuracy of UAV remote-sensing images.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902378/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Special Issue: Digital Healthcare Leveraging Edge Computing and the Internet of Things.","authors":"Antonio Celesti, Ivanoe De Falco, Giovanna Sannino, Lorenzo Carnevale","doi":"10.3390/s25051571","DOIUrl":"10.3390/s25051571","url":null,"abstract":"<p><p>The global Internet of Things (IoT) medical device market is always growing [...].</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902605/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Innovative Regression Model for Frequency-Dependent Acoustic Source Strength in the Aquatic Environment: Bridging Scientific Insight and Practical Applications.","authors":"Moshe Greenberg, Uri Kushnir, Vladimir Frid","doi":"10.3390/s25051560","DOIUrl":"10.3390/s25051560","url":null,"abstract":"<p><p>This study addresses the challenge of predicting acoustic source strength in freshwater environments, focusing on frequencies between 100-400 kHz. Acoustic signal attenuation is inherently frequency-dependent and influenced by water properties as well as the total propagation path of the acoustic wave, complicating the accurate determination of source strength. To address this challenge, we developed a non-linear regression model for solving the inverse problem of attenuation correction in reflected signals from typical aquatic reflectors, addressing the current absence of robust correction tools in this frequency range. The novelty of our approach lies in designing a non-linear regression framework that incorporates key physical parameters-signal energy, propagation distance, and frequency-enabling accurate source strength prediction. Using an experimental setup comprising ultrasonic transducers and a signal generator under controlled conditions, we collected a comprehensive dataset of 366 samples. The results demonstrate that our proposed model achieves reliable source strength prediction by simplifying Thorpe's equation for freshwater environments. This research represents a significant advancement in underwater acoustics, providing a practical and reliable tool for source strength estimation in freshwater systems. The developed methodology may have broad applications across sonar technology, environmental monitoring, and aquatic research domains.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902304/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650525","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of Smart Watch-Based Functional Evaluation for Upper Extremity Impairment: A Preliminary Study on Older Emirati Stroke Population.","authors":"Yeo Hyung Kim, Sarah Kim, Hyung Seok Nam","doi":"10.3390/s25051554","DOIUrl":"10.3390/s25051554","url":null,"abstract":"<p><p>Smartwatch-based functional assessments for upper extremity movement are a promising tool for a detailed and serial assessment during stroke rehabilitation, but their clinical application remains challenging. In this study, nine patients with hemiparesis due to a stroke participated in occupational therapy sessions using virtual reality-based rehabilitation devices. An Action Research Arm Test (ARAT) was performed at baseline and after intervention, with wrist smartwatch sensors recording motion data. We extracted acceleration and gyro sensor data from smartwatches and calculated the average motion segment size (MSS) as a measure of motion smoothness. Among the included patients, four participants completed all 10 therapy sessions and the follow-up evaluation. The average MSSs of acceleration for all <i>x</i>, <i>y</i>, and <i>z</i> directions were significantly correlated with the ARAT scores across all task domains. For angular motion, the average MSS in the gross movement task (domain 4) showed strong correlations with the ARAT scores: roll (r_s = 0.735, <i>p</i> = 0.004), pitch (r_s = 0.715, <i>p</i> = 0.009), and yaw (r_s = 0.704, <i>p</i> = 0.007). At the serial follow-ups, most participants showed a considerable increase in the average MSSs of the roll, pitch, and yaw angles measured during domain 4, alongside improvements in their clinical ARAT scores. Our findings support the feasibility of using commercial smartwatch-based parameters for upper extremity functional evaluations during stroke rehabilitation and highlight their potential for serial follow-up assessments.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902849/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application Research on High-Precision Tiltmeter with Rapid Deployment Capability.","authors":"Fuxi Yang, Dongxiao Guan, Xiaodong Li, Chen Dou","doi":"10.3390/s25051559","DOIUrl":"10.3390/s25051559","url":null,"abstract":"<p><p>This article introduces a high-precision vertical pendulum tiltmeter with rapid deployment capability to improve the observation efficiency, practicality, and reliability of geophysical site tilt observation instruments. The system consists of a pendulum body, a triangular platform, a locking pendulum motor, a sealed cover, a ratio measurement bridge, a high-precision ADC, and an embedded data acquisition unit. The sensing unit adopts a vertical pendulum system suspended by a cross spring and a differential capacitance bridge measurement circuit, which can simultaneously measure two orthogonal directions of ground tilt. The pendulum is installed on a short baseline triangular platform, sealed as a whole with the platform, and equipped with a locking pendulum motor. When the pendulum is locked and packaged, it can withstand a 2 m free fall impact, with high reliability and easy use. It can be quickly deployed without the need for professional technicians. This article analyzes its various performance and technical indicators based on its application in the rapid deployment of the Zeketai seismic station in Xinjiang. It is of great significance for emergency response, mobile observation, base detection, anomaly verification, and other applications of ground tilt.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902578/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"IRFNet: Cognitive-Inspired Iterative Refinement Fusion Network for Camouflaged Object Detection.","authors":"Guohan Li, Jingxin Wang, Jianming Wei, Zhengyi Xu","doi":"10.3390/s25051555","DOIUrl":"10.3390/s25051555","url":null,"abstract":"<p><p>Camouflaged Object Detection (COD) aims to identify objects that are intentionally concealed within their surroundings through appearance, texture, or pattern adaptations. Despite recent advances, extreme object-background similarity causes existing methods struggle with accurately capturing discriminative features and effectively modeling multiscale patterns while preserving fine details. To address these challenges, we propose Iterative Refinement Fusion Network (IRFNet), a novel framework that mimics human visual cognition through progressive feature enhancement and iterative optimization. Our approach incorporates the following: (1) a Hierarchical Feature Enhancement Module (HFEM) coupled with a dynamic channel-spatial attention mechanism, which enriches multiscale feature representations through bilateral and trilateral fusion pathways; and (2) a Context-guided Iterative Optimization Framework (CIOF) that combines transformer-based global context modeling with iterative refinement through dual-branch supervision. Extensive experiments on three challenging benchmark datasets (CAMO, COD10K, and NC4K) demonstrate that IRFNet consistently outperforms fourteen state-of-the-art methods, achieving improvements of 0.9-13.7% across key metrics. Comprehensive ablation studies validate the effectiveness of each proposed component and demonstrate how our iterative refinement strategy enables progressive improvement in detection accuracy.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902440/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650235","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of Distributed Acoustic Sensing for Active Near-Surface Seismic Monitoring.","authors":"Eslam Roshdy, Mariusz Majdański, Szymon Długosz, Artur Marciniak, Paweł Popielski","doi":"10.3390/s25051558","DOIUrl":"10.3390/s25051558","url":null,"abstract":"<p><p>High-resolution imaging of the near-surface structures of critical objects is necessary in various applications including geohazard studies, the structural health of artificial structures, and generally in environmental seismology. This study explores the use of fiber optic sensor technology in active seismic surveys to monitor the embankment structure of the Rybnik Reservoir in Poland. We discuss the technical aspects, including sensor types and energy sources, and provide a comparison of the data collected with a standard geophone-based survey conducted simultaneously. A thorough data processing methodology is presented to directly compare both datasets. The results show a comparable data quality, with DAS offering significant advantages in terms of both the spatial and temporal resolution, facilitating more accurate interpretations. DAS demonstrates its ability to operate effectively in complex geological environments, such as areas with high seismic noise, rough terrain, and variable surface conditions, making it highly adaptable for monitoring critical infrastructure. Additionally, DAS provides long-term monitoring capabilities, essential for ongoing structural health assessments and geohazard detection. For example, the multichannel analysis of surface waves (MASW) using DAS data clearly identifies S-wave velocities down to 13 m with an RMS error of 3.26%, compared to an RMS error of 6.2% for geophone data. Moreover, the DAS-based data were easier to process and interpret. The integration of DAS with traditional seismic data can provide a more comprehensive understanding of subsurface properties, facilitating more accurate and reliable geophysical assessments over time. This innovative approach is particularly valuable in challenging environments, underscoring its importance in monitoring critical infrastructure.</p>","PeriodicalId":21698,"journal":{"name":"Sensors","volume":"25 5","pages":""},"PeriodicalIF":3.4,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11902483/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143650272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}