Pub Date : 2024-08-26DOI: 10.1109/JSEN.2024.3445125
Yirong Wang;Jiacheng Xie;Ziying Zheng;Yichen Wang;Xuewen Wang
To address the issues of low accuracy and insufficient information in the pose measurement of hydraulic support caused by complex working conditions, this article proposes a method for pose optimization and collision resolution. First, based on the structure and significance, the pose is divided into “three layers of base pose-one layer of internal pose.” A mapping model of each layer’s pose and sensing error is established to accurately resolve errors. Then, the Bayesian algorithm is employed to infer the single pose under a normal distribution. A multimodel system is constructed through five steps: updating measurement, updating error, prior calculation, Bayesian estimation, and pose correction, to optimize the complete pose of support. Subsequently, three highly credible virtual support scenes are constructed. Sensor information is input into the pose optimization scene to generate the pose distribution, which is synchronized to the collision-solving scene to visualize and solve collision probabilities. This probability is then fed back to the monitoring scene for warning judgment. Finally, various conditions are set for pose optimization and collision-solving experiments. The results show that the accuracy of the optimized support pose is enhanced by an average of 11.57% and the collision between supports is accurately simulated. This verifies the accuracy of the theoretical research and the Bayesian algorithm. The approach is beneficial to warn the timely abnormal situation warnings, enhancing the reliability of underground support, and providing an important reference to support following control and adjustment.
{"title":"Method for Collision Relationship of Hydraulic Supports Considering Multilayer Sensing Errors","authors":"Yirong Wang;Jiacheng Xie;Ziying Zheng;Yichen Wang;Xuewen Wang","doi":"10.1109/JSEN.2024.3445125","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3445125","url":null,"abstract":"To address the issues of low accuracy and insufficient information in the pose measurement of hydraulic support caused by complex working conditions, this article proposes a method for pose optimization and collision resolution. First, based on the structure and significance, the pose is divided into “three layers of base pose-one layer of internal pose.” A mapping model of each layer’s pose and sensing error is established to accurately resolve errors. Then, the Bayesian algorithm is employed to infer the single pose under a normal distribution. A multimodel system is constructed through five steps: updating measurement, updating error, prior calculation, Bayesian estimation, and pose correction, to optimize the complete pose of support. Subsequently, three highly credible virtual support scenes are constructed. Sensor information is input into the pose optimization scene to generate the pose distribution, which is synchronized to the collision-solving scene to visualize and solve collision probabilities. This probability is then fed back to the monitoring scene for warning judgment. Finally, various conditions are set for pose optimization and collision-solving experiments. The results show that the accuracy of the optimized support pose is enhanced by an average of 11.57% and the collision between supports is accurately simulated. This verifies the accuracy of the theoretical research and the Bayesian algorithm. The approach is beneficial to warn the timely abnormal situation warnings, enhancing the reliability of underground support, and providing an important reference to support following control and adjustment.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368565","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Although numerous efforts have been dedicated toward developing optical communication system with high performances, challenges still remain in achieving communication in special scenarios, such as mines where flammable and explosive gases are present. Aiming at the problem, a passive bidirectional audio-over-fiber (PB-AOF) system that integrates sensing, power supply, and communication has been proposed, enabling passive bidirectional audio transmission across a 10-km single-mode fiber (SMF). In the uplink system, distributed acoustic sensing (DAS) technology is used to achieve distributed audio sensing across the entire span of the optical fiber. In the downlink system, power-over-fiber (PWoF) technology is used, using a self-designed InGaAs/InP photovoltaic power converter (PPC), to achieve simultaneous power and signal transmission to the downlink terminal. The system not only achieves distributed audio transmission with frequency response range up to 5 kHz and signal-to-noise ratio (SNR) of more than 50 dB over a distance of 10 km in the uplink but also provides an SNR of more than 50 dB of the audio signal after 10 km of fiber-optic transmission in the downlink. The downlink has a power transfer efficiency of up to 24%.
尽管人们一直致力于开发高性能的光通信系统，但在特殊场景（如存在易燃易爆气体的矿井）中实现通信仍面临挑战。针对这一问题，有人提出了一种集传感、供电和通信于一体的无源双向光纤音频（PB-AOF）系统，可在 10 千米长的单模光纤（SMF）上实现无源双向音频传输。在上行系统中，分布式声学传感（DAS）技术用于实现整个光纤跨度上的分布式音频传感。在下行链路系统中，利用自行设计的 InGaAs/InP 光电功率转换器（PPC），采用光纤功率（PWoF）技术实现向下行链路终端同时传输功率和信号。该系统不仅在上行链路中实现了频率响应范围高达 5 kHz、信噪比（SNR）超过 50 dB 的分布式音频传输，而且在下行链路中经过 10 km 的光纤传输后，音频信号的信噪比（SNR）也超过了 50 dB。下行链路的功率传输效率高达 24%。
{"title":"Passive Bidirectional Audio-Over-Fiber System Integrating Sensing, Power Supply, and Communication","authors":"Cong Liu;Haixin Qin;Chenggang Guan;Xuan Chen;Jingqi Li;Linfeng Zhan;Weiqi Wang;Yifan Xiao;Sheng Hu;Junchang Huang;Xueyou Zhang","doi":"10.1109/JSEN.2024.3443153","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3443153","url":null,"abstract":"Although numerous efforts have been dedicated toward developing optical communication system with high performances, challenges still remain in achieving communication in special scenarios, such as mines where flammable and explosive gases are present. Aiming at the problem, a passive bidirectional audio-over-fiber (PB-AOF) system that integrates sensing, power supply, and communication has been proposed, enabling passive bidirectional audio transmission across a 10-km single-mode fiber (SMF). In the uplink system, distributed acoustic sensing (DAS) technology is used to achieve distributed audio sensing across the entire span of the optical fiber. In the downlink system, power-over-fiber (PWoF) technology is used, using a self-designed InGaAs/InP photovoltaic power converter (PPC), to achieve simultaneous power and signal transmission to the downlink terminal. The system not only achieves distributed audio transmission with frequency response range up to 5 kHz and signal-to-noise ratio (SNR) of more than 50 dB over a distance of 10 km in the uplink but also provides an SNR of more than 50 dB of the audio signal after 10 km of fiber-optic transmission in the downlink. The downlink has a power transfer efficiency of up to 24%.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-26DOI: 10.1109/MEI.2024.10646177
The 2024 Electrical Insulation Conference (EIC 2024) was held at the Radisson Blu Mall of America Hotel in Minneapolis, Minnesota, USA, between June 2 and 6, 2024.
{"title":"Bulletin Board: IEEE DEIS Electrical Insulation Conference 2024","authors":"","doi":"10.1109/MEI.2024.10646177","DOIUrl":"https://doi.org/10.1109/MEI.2024.10646177","url":null,"abstract":"The 2024 Electrical Insulation Conference (EIC 2024) was held at the Radisson Blu Mall of America Hotel in Minneapolis, Minnesota, USA, between June 2 and 6, 2024.","PeriodicalId":444,"journal":{"name":"IEEE Electrical Insulation Magazine","volume":null,"pages":null},"PeriodicalIF":2.6,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10646177","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142077605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vortex pumps play a crucial role in industrial and municipal settings by transferring high-viscosity and particle-laden fluids. However, their performance and reliability are significantly compromised by cavitation. Identifying and diagnosing cavitation promptly is essential for maintaining the proper operation of vortex pumps. The use of current signals as a noninvasive monitoring method has shown great promise in detecting multicavitation states. The proposed method integrates dual-tree complex wavelet transform (DT-CWT) with variational mode decomposition (VMD) to decompose current signals into multiple modes. Subsequently, the Bayesian optimized locally weighted k-nearest neighbor (LW-KNN) algorithm is employed to accurately identify multicavitation states. High-speed photography is also utilized to observe the incipient, developing, and collapsing phases of cavitation. The results indicate that the proposed method achieves a detection accuracy of 96.67% at a flow rate of 40 m3/h, outperforming other flow conditions. The recognition accuracy reaches 98.33% under stable flow conditions, while accuracies of 92.33% and 93.67% are observed for flow rates of 35 and 45 m3/h, respectively. The overall average recognition rate across all tested flow conditions is 94.22%. This methodology not only demonstrates high effectiveness in identifying cavitation states but also offers a reliable and practical solution for fault diagnosis in fluid mechanical systems. It significantly contributes to the improvement of operational efficiency, reliability, and maintenance strategies in industrial and municipal pumping systems.
{"title":"Multicavitation States Diagnosis of the Vortex Pump Using a Combined DT-CWT-VMD and BO-LW-KNN Based on Motor Current Signals","authors":"Weitao Zeng;Peijian Zhou;Yanzhao Wu;Denghao Wu;Maosen Xu","doi":"10.1109/JSEN.2024.3446170","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3446170","url":null,"abstract":"Vortex pumps play a crucial role in industrial and municipal settings by transferring high-viscosity and particle-laden fluids. However, their performance and reliability are significantly compromised by cavitation. Identifying and diagnosing cavitation promptly is essential for maintaining the proper operation of vortex pumps. The use of current signals as a noninvasive monitoring method has shown great promise in detecting multicavitation states. The proposed method integrates dual-tree complex wavelet transform (DT-CWT) with variational mode decomposition (VMD) to decompose current signals into multiple modes. Subsequently, the Bayesian optimized locally weighted k-nearest neighbor (LW-KNN) algorithm is employed to accurately identify multicavitation states. High-speed photography is also utilized to observe the incipient, developing, and collapsing phases of cavitation. The results indicate that the proposed method achieves a detection accuracy of 96.67% at a flow rate of 40 m3/h, outperforming other flow conditions. The recognition accuracy reaches 98.33% under stable flow conditions, while accuracies of 92.33% and 93.67% are observed for flow rates of 35 and 45 m3/h, respectively. The overall average recognition rate across all tested flow conditions is 94.22%. This methodology not only demonstrates high effectiveness in identifying cavitation states but also offers a reliable and practical solution for fault diagnosis in fluid mechanical systems. It significantly contributes to the improvement of operational efficiency, reliability, and maintenance strategies in industrial and municipal pumping systems.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Osteoarthritis (OA) is a degenerative joint disease characterized by cartilage degradation and changes in bone morphology, typically assessed through magnetic resonance imaging (MRI). This study introduces a method using a posterior shape model (PSM) to estimate cartilage thickness based solely on bone geometry. Utilizing the SKI10 public MRI dataset, we developed bone shape and combined bone-cartilage shape models through a leave-one-out (LOO) experiment involving 99 folds. Cartilage estimation in the tibiofemoral contact and surgical areas relied solely on bone geometry, using a PSM. This novel method, compared against current state-of-the-art techniques, demonstrated a predictable correlation in cartilage thickness in regions where bone relationship information is available. The validation of the model was conducted using a cross-validation technique on the dataset, comparing the predicted cartilage thickness with actual measurements obtained through manual segmentation. Employing bone gap data at the tibiofemoral contact point, our cartilage thickness prediction achieved a root mean square error (RMSE) compared to the manual segmentation of 0.64 mm for the femur and 0.58 mm. Preliminary results indicate that the proposed method can successfully estimate cartilage information in scenarios where direct cartilage imaging is unavailable. This approach holds promise for enhancing diagnostic capabilities in knee joint conditions where cartilage assessment is critical.
{"title":"Knee Cartilage Estimation Based on Knee Bone Geometry Using Posterior Shape Model","authors":"Hao Chen;Tao Tan;Yan Kang;Yue Sun;Hui Xie;XinYe Wang;Nico Verdonschot","doi":"10.1109/JSEN.2024.3443994","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3443994","url":null,"abstract":"Osteoarthritis (OA) is a degenerative joint disease characterized by cartilage degradation and changes in bone morphology, typically assessed through magnetic resonance imaging (MRI). This study introduces a method using a posterior shape model (PSM) to estimate cartilage thickness based solely on bone geometry. Utilizing the SKI10 public MRI dataset, we developed bone shape and combined bone-cartilage shape models through a leave-one-out (LOO) experiment involving 99 folds. Cartilage estimation in the tibiofemoral contact and surgical areas relied solely on bone geometry, using a PSM. This novel method, compared against current state-of-the-art techniques, demonstrated a predictable correlation in cartilage thickness in regions where bone relationship information is available. The validation of the model was conducted using a cross-validation technique on the dataset, comparing the predicted cartilage thickness with actual measurements obtained through manual segmentation. Employing bone gap data at the tibiofemoral contact point, our cartilage thickness prediction achieved a root mean square error (RMSE) compared to the manual segmentation of 0.64 mm for the femur and 0.58 mm. Preliminary results indicate that the proposed method can successfully estimate cartilage information in scenarios where direct cartilage imaging is unavailable. This approach holds promise for enhancing diagnostic capabilities in knee joint conditions where cartilage assessment is critical.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-26DOI: 10.1109/JSEN.2024.3441839
Dandan Meng;Xin Li;Wei Wang
Parameter estimation in electromagnetic vector sensor (EMVS) multiple-input multiple-output (MIMO) radar faces challenges due to spatial noise and small array apertures. Therefore, a robust tensor decomposition approach, which is tailored for a monostatic EMVS-MIMO radar with sparse L-shaped, is proposed for high-resolution 2-D direction-of-arrival (DOA) estimation by constructing a covariance tensor to suppress spatial noise and utilizing higher order singular value decomposition (HOSVD) to obtain an exact subspace. Subsequently, the vector-cross product (VCP) technique achievable by the EMVS is utilized to obtain low-resolution but unique DOA estimates, and the estimation of signal parameters with rotational invariance technique (ESPRIT) technique based on sparse uniform array geometry is exploited to obtain high-resolution but ambiguous DOA estimates. By combining the characteristics of both, a unique high-resolution DOA estimate is derived. The results indicate that the framework exhibits better estimation accuracy under low signal-to-noise ratios compared with the existing methods. Furthermore, it is more adaptable than current sparse array methods. Experimental simulation results validate the correctness of the theoretical derivations.
由于空间噪声和阵列孔径较小，电磁矢量传感器（EMVS）多输入多输出（MIMO）雷达的参数估计面临挑战。因此，针对具有稀疏 L 形的单静态 EMVS-MIMO 雷达，提出了一种鲁棒张量分解方法，通过构建协方差张量来抑制空间噪声，并利用高阶奇异值分解（HOSVD）来获得精确子空间，从而实现高分辨率二维到达方向（DOA）估计。随后，利用 EMVS 可实现的矢量交叉积（VCP）技术获得低分辨率但唯一的 DOA 估计值，并利用基于稀疏均匀阵列几何的旋转不变性信号参数估计技术（ESPRIT）获得高分辨率但模糊的 DOA 估计值。通过结合两者的特点，得出了独特的高分辨率 DOA 估计值。结果表明，与现有方法相比，该框架在低信噪比条件下表现出更高的估计精度。此外，它比现有的稀疏阵列方法更具适应性。实验模拟结果验证了理论推导的正确性。
{"title":"Robust Tensor Decomposition Approach for DOA Estimation With EMVS-MIMO Radar","authors":"Dandan Meng;Xin Li;Wei Wang","doi":"10.1109/JSEN.2024.3441839","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3441839","url":null,"abstract":"Parameter estimation in electromagnetic vector sensor (EMVS) multiple-input multiple-output (MIMO) radar faces challenges due to spatial noise and small array apertures. Therefore, a robust tensor decomposition approach, which is tailored for a monostatic EMVS-MIMO radar with sparse L-shaped, is proposed for high-resolution 2-D direction-of-arrival (DOA) estimation by constructing a covariance tensor to suppress spatial noise and utilizing higher order singular value decomposition (HOSVD) to obtain an exact subspace. Subsequently, the vector-cross product (VCP) technique achievable by the EMVS is utilized to obtain low-resolution but unique DOA estimates, and the estimation of signal parameters with rotational invariance technique (ESPRIT) technique based on sparse uniform array geometry is exploited to obtain high-resolution but ambiguous DOA estimates. By combining the characteristics of both, a unique high-resolution DOA estimate is derived. The results indicate that the framework exhibits better estimation accuracy under low signal-to-noise ratios compared with the existing methods. Furthermore, it is more adaptable than current sparse array methods. Experimental simulation results validate the correctness of the theoretical derivations.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Reconstruction-based methods, as one of the mainstream and advanced methods for anomaly detection, have attracted significant attention in the academic community. Although these methods may achieve good performance on some ideal industrial datasets, background factors have considerable influence on detecting anomalies due to a complex and ever-changing environment, resulting in overkill and false detection. In this work, we extend our previous implicit foreground-guided network (IFgNet) with a comprehensive consideration of the interference from complex backgrounds, and an incorporation of foreground constraints throughout the entire process. Thus, we propose a foreground collaboration and augmentation (ForeCA) network for anomaly detection, consisting of foreground homology augmentation (FHA) and foreground collaboration reconstruction (FCR). To be specific, FHA adopts a shuffled homology augmentation (SHA) strategy to synthesize pseudo-anomalous samples, as inputs of FCR, disrupting the original spatial structure of normal samples while preserving some structural relevance. Furthermore, FCR flexibly injects two sets of task-specific attention blocks into each convolutional block as task attention, integrating foreground detection with image reconstruction. We discriminate anomalies by the difference between the reconstructed images and the inputs and utilize the obtained foreground predictions to refine the coarse anomaly map. Extensive experiments on two challenging, widely used industrial anomaly detection datasets, including visual anomaly (VisA) and metal parts defect detection (MPDD), demonstrate our proposed method can achieve competitive results in both anomaly detection and localization. Our code is available at https://github.com/gloriacxl/ForeCA .
{"title":"Leveraging Foreground Collaboration and Augmentation for Industrial Anomaly Detection","authors":"Xiaolu Chen;Haote Xu;Jiaxiang Wang;Xiaotong Tu;Xinghao Ding;Yue Huang","doi":"10.1109/JSEN.2024.3446249","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3446249","url":null,"abstract":"Reconstruction-based methods, as one of the mainstream and advanced methods for anomaly detection, have attracted significant attention in the academic community. Although these methods may achieve good performance on some ideal industrial datasets, background factors have considerable influence on detecting anomalies due to a complex and ever-changing environment, resulting in overkill and false detection. In this work, we extend our previous implicit foreground-guided network (IFgNet) with a comprehensive consideration of the interference from complex backgrounds, and an incorporation of foreground constraints throughout the entire process. Thus, we propose a foreground collaboration and augmentation (ForeCA) network for anomaly detection, consisting of foreground homology augmentation (FHA) and foreground collaboration reconstruction (FCR). To be specific, FHA adopts a shuffled homology augmentation (SHA) strategy to synthesize pseudo-anomalous samples, as inputs of FCR, disrupting the original spatial structure of normal samples while preserving some structural relevance. Furthermore, FCR flexibly injects two sets of task-specific attention blocks into each convolutional block as task attention, integrating foreground detection with image reconstruction. We discriminate anomalies by the difference between the reconstructed images and the inputs and utilize the obtained foreground predictions to refine the coarse anomaly map. Extensive experiments on two challenging, widely used industrial anomaly detection datasets, including visual anomaly (VisA) and metal parts defect detection (MPDD), demonstrate our proposed method can achieve competitive results in both anomaly detection and localization. Our code is available at \u0000<uri>https://github.com/gloriacxl/ForeCA</uri>\u0000.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-23DOI: 10.1109/JSEN.2024.3444053
Tianchi Li;Can Li;Ning Li
The rainbow scattering technique holds significant promise for simultaneously detecting micrometer-sized droplet size and refractive index. Reported here is a fast and lightweight deep-learning-based rainbow inversion algorithm containing a unique signal preprocessor and a novel signal inversion network. Rainbow signal is preprocessed based on scattering intensity, scattering angle, and its intensity-angle area to effectively preserve the physical features. A multilayer perceptron (MLP)-based rainbow inversion network is constructed for more effective inversion of droplet size and refractive index at the global sensory field level. The algorithm achieves a 50fold speedup compared with conventional methods without compromising accuracy. Extensive simulations demonstrate an average relative error of ${0}.{89}%$ in droplet size estimation and an average absolute error of ${3}.{81}times {10} ^{-{5}}$ in refractive index determination. Experimental validation further confirms the algorithm’s reliability. This work not only offers significant improvements in computational speed and accuracy but also opens new avenues for advanced droplet parameter inversion algorithms.
{"title":"A Precise and Fast Droplet Parameter Inversion Algorithm for Rainbow Scattering Detection","authors":"Tianchi Li;Can Li;Ning Li","doi":"10.1109/JSEN.2024.3444053","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3444053","url":null,"abstract":"The rainbow scattering technique holds significant promise for simultaneously detecting micrometer-sized droplet size and refractive index. Reported here is a fast and lightweight deep-learning-based rainbow inversion algorithm containing a unique signal preprocessor and a novel signal inversion network. Rainbow signal is preprocessed based on scattering intensity, scattering angle, and its intensity-angle area to effectively preserve the physical features. A multilayer perceptron (MLP)-based rainbow inversion network is constructed for more effective inversion of droplet size and refractive index at the global sensory field level. The algorithm achieves a 50fold speedup compared with conventional methods without compromising accuracy. Extensive simulations demonstrate an average relative error of \u0000<inline-formula> <tex-math>${0}.{89}%$ </tex-math></inline-formula>\u0000 in droplet size estimation and an average absolute error of \u0000<inline-formula> <tex-math>${3}.{81}times {10} ^{-{5}}$ </tex-math></inline-formula>\u0000 in refractive index determination. Experimental validation further confirms the algorithm’s reliability. This work not only offers significant improvements in computational speed and accuracy but also opens new avenues for advanced droplet parameter inversion algorithms.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-23DOI: 10.1109/JSEN.2024.3444496
Yongxing Yu;Dan Huang;Yueming Hu;Rongjun Chen;Xu Lu
As the fabrication of high-density integrated circuit (IC) substrates advances in precision, the task of sensing and visually inspecting defects becomes increasingly challenging. These challenges arise from the need for precise segmentation of metallographic substrate images, complicated by variations in gray levels, noise interference, and rich textures. To address these issues, an adaptive fractional differentiation (AFD)-based active contour model (ACM) is proposed, which integrates global and local terms for a more accurate depiction of sensed image information. Using local image statistics, we construct an adaptive fractional-order model to find the optimal order for the fractional gradient. This gradient is then integrated into the Chan-Vese model as the global fitting term. The fusion of the original image with the fractional gradient image forms the local fitting object, refined by a Gaussian kernel function. Adaptive weight parameters are designed to enhance segmentation performance and accelerate evolution, adjusting the balance between global and local terms. Additionally, a distance penalty term is introduced to prevent reinitialization and improve segmentation efficiency. Experimental results show that the AFD model achieves accurate segmentation of high-density IC substrate images, with average improvements in DSC and JS scores of ${16}.{95}%$ and ${28}.{52}%$ , respectively. Moreover, it reduces average processing times by ${59}.{47}%$ , demonstrating better efficiency and robustness than other models.
{"title":"Sensing High-Density IC Substrates: Adaptive Fractional Differentiation for Accurate Image Segmentation","authors":"Yongxing Yu;Dan Huang;Yueming Hu;Rongjun Chen;Xu Lu","doi":"10.1109/JSEN.2024.3444496","DOIUrl":"https://doi.org/10.1109/JSEN.2024.3444496","url":null,"abstract":"As the fabrication of high-density integrated circuit (IC) substrates advances in precision, the task of sensing and visually inspecting defects becomes increasingly challenging. These challenges arise from the need for precise segmentation of metallographic substrate images, complicated by variations in gray levels, noise interference, and rich textures. To address these issues, an adaptive fractional differentiation (AFD)-based active contour model (ACM) is proposed, which integrates global and local terms for a more accurate depiction of sensed image information. Using local image statistics, we construct an adaptive fractional-order model to find the optimal order for the fractional gradient. This gradient is then integrated into the Chan-Vese model as the global fitting term. The fusion of the original image with the fractional gradient image forms the local fitting object, refined by a Gaussian kernel function. Adaptive weight parameters are designed to enhance segmentation performance and accelerate evolution, adjusting the balance between global and local terms. Additionally, a distance penalty term is introduced to prevent reinitialization and improve segmentation efficiency. Experimental results show that the AFD model achieves accurate segmentation of high-density IC substrate images, with average improvements in DSC and JS scores of \u0000<inline-formula> <tex-math>${16}.{95}%$ </tex-math></inline-formula>\u0000 and \u0000<inline-formula> <tex-math>${28}.{52}%$ </tex-math></inline-formula>\u0000, respectively. Moreover, it reduces average processing times by \u0000<inline-formula> <tex-math>${59}.{47}%$ </tex-math></inline-formula>\u0000, demonstrating better efficiency and robustness than other models.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":null,"pages":null},"PeriodicalIF":4.3,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-23DOI: 10.1109/TPS.2024.3421897
S. Trieu;F. Hoffmann;M. de Haas;G. Tchilinguirian;B. P. LeBlanc
The multipoint Thomson scattering (MPTS) diagnostic system at the National Spherical Torus Experiment Upgrade (NSTX-U) facility is undergoing an upgrade to operate in real-time and interface with the plasma control system (PCS) for NSTX-U. Previous prototyping efforts have shown that spectral analysis and rapid calculations of electron temperature and density are possible on a real-time Linux machine when using up to a 100-Hz laser pulse repetition rate. A remaining challenge was transferring the real-time data to NSTX-U’s PCS, which utilizes the front panel data port (FPDP) protocol. The original proposed method was to convert the real-time data into analog values, but a new solution was developed to keep the output format digital by using an Ethernet controller with a field-programmable gate array (FPGA). This article focuses on a new input module that has been developed to accept incoming user datagram protocol (UDP) packets sent over Ethernet, convert into FPDP format, and integrate into the existing data stream under NSTX-U’s real-time framework.
{"title":"Real-Time Ethernet Interface for NSTX-U’s Thomson Scattering Diagnostic (2023)","authors":"S. Trieu;F. Hoffmann;M. de Haas;G. Tchilinguirian;B. P. LeBlanc","doi":"10.1109/TPS.2024.3421897","DOIUrl":"10.1109/TPS.2024.3421897","url":null,"abstract":"The multipoint Thomson scattering (MPTS) diagnostic system at the National Spherical Torus Experiment Upgrade (NSTX-U) facility is undergoing an upgrade to operate in real-time and interface with the plasma control system (PCS) for NSTX-U. Previous prototyping efforts have shown that spectral analysis and rapid calculations of electron temperature and density are possible on a real-time Linux machine when using up to a 100-Hz laser pulse repetition rate. A remaining challenge was transferring the real-time data to NSTX-U’s PCS, which utilizes the front panel data port (FPDP) protocol. The original proposed method was to convert the real-time data into analog values, but a new solution was developed to keep the output format digital by using an Ethernet controller with a field-programmable gate array (FPGA). This article focuses on a new input module that has been developed to accept incoming user datagram protocol (UDP) packets sent over Ethernet, convert into FPDP format, and integrate into the existing data stream under NSTX-U’s real-time framework.","PeriodicalId":450,"journal":{"name":"IEEE Transactions on Plasma Science","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142175280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：