Pub Date : 2024-11-12DOI: 10.1109/TIM.2024.3488147
Tun Wang;Hao Sheng;Rongshan Chen;Ruixuan Cong;Mingyuan Zhao;Zhenglong Cui
Light field (LF) technology captures information from multiple directions and angles, enabling precise disparity estimation. Recently, matching cost-based approaches have advanced rapidly and shown satisfactory results. However, these methods typically depend on fixed disparity candidates, leading to inadequate utilization of candidates and making them unsuitable for LF scenes with varying baselines. Multidirection line structures of epipolar-plane images (EPIs) associate multiple viewpoints, adaptively perceiving disparity ranges and accurately matching features in real scenes. In this article, we propose an adaptive EPI-matching cost (AEMC) for LF disparity estimation, which is proven to enhance the adaptability across datasets with varying baselines. Our approach calculates pixel-level disparity candidates to keep the predicted distribution near the ground truth (GT) and matches line structures to improve accuracy. Then, to enhance robustness during the adaptive process, we introduce an intra-EPI extraction module that dynamically establishes correlations in the local EPI while supplementing spatial information. Finally, we present a network named adaptive EPI-matching cost network (AEMCNet) for LF disparity estimation. Experimental results demonstrate that AEMCNet achieves state-of-the-art (SOTA) performance and robustness on various LF datasets with different baselines. Specifically, on the sparse LF dataset, our method reduces the mean square error (mse) by 49.6%.
{"title":"Adaptive EPI-Matching Cost for Light Field Disparity Estimation","authors":"Tun Wang;Hao Sheng;Rongshan Chen;Ruixuan Cong;Mingyuan Zhao;Zhenglong Cui","doi":"10.1109/TIM.2024.3488147","DOIUrl":"https://doi.org/10.1109/TIM.2024.3488147","url":null,"abstract":"Light field (LF) technology captures information from multiple directions and angles, enabling precise disparity estimation. Recently, matching cost-based approaches have advanced rapidly and shown satisfactory results. However, these methods typically depend on fixed disparity candidates, leading to inadequate utilization of candidates and making them unsuitable for LF scenes with varying baselines. Multidirection line structures of epipolar-plane images (EPIs) associate multiple viewpoints, adaptively perceiving disparity ranges and accurately matching features in real scenes. In this article, we propose an adaptive EPI-matching cost (AEMC) for LF disparity estimation, which is proven to enhance the adaptability across datasets with varying baselines. Our approach calculates pixel-level disparity candidates to keep the predicted distribution near the ground truth (GT) and matches line structures to improve accuracy. Then, to enhance robustness during the adaptive process, we introduce an intra-EPI extraction module that dynamically establishes correlations in the local EPI while supplementing spatial information. Finally, we present a network named adaptive EPI-matching cost network (AEMCNet) for LF disparity estimation. Experimental results demonstrate that AEMCNet achieves state-of-the-art (SOTA) performance and robustness on various LF datasets with different baselines. Specifically, on the sparse LF dataset, our method reduces the mean square error (mse) by 49.6%.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-13"},"PeriodicalIF":5.6,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the above article [1], there is an error in (9). The correct equation is as below: begin{equation*} hat {sigma }^{2}=frac {sum _{j=1}^{16} Delta V_{j}^{2}}{16-6} =frac {sum _{j=1}^{16}left ({{V_{mathrm {cal}, j}-V_{mathrm {meas}, j}}}right )^{2}}{10}.end{equation*} The above error resulted in the calculated values for uncertainty displayed in Fig. 5 and Table III being smaller by 18%. The figure and table reflecting the correct values are as follows:TABLE IIISensor Parameters With Uncertainties channelx(mm)y(mm)z(mm) $n_x$ $n_y$ $n_z$ g(nT/V)CH240.39±0.5820.72±0.56163.84±0.410.1028±0.00840.1757±0.00810.9791±0.00171.176±0.011CH28116.76±0.22-14.51±0.293.61±0.230.9929±0.0008-0.1178±0.00690.0157±0.00691.135±0.008CH55-10.63±0.32116.38±0.2122.02±0.440.0262±0.00780.9993±0.0003-0.0254±0.00031.065±0.008CH115-135.52±0.577.07±0.6278.59±0.60-0.2907±0.00280.3784±0.00960.8788±0.00271.137±0.016CH140-19.74±0.42-116.89±0.3226.20±0.56-0.0926±0.0094-0.9951±0.00110.0352±0.00141.038±0.009The values following the plus-minus signs (±) correspond to uncertainties with a coverage factor $k=1$ . Fig. 5.Fig. 5.
{"title":"Errata to “A Spherical Coil Array for the Calibration of Whole-Head Magnetoencephalograph Systems”","authors":"Yoshiaki Adachi;Daisuke Oyama;Masanori Higuchi;Gen Uehara","doi":"10.1109/TIM.2024.3475788","DOIUrl":"https://doi.org/10.1109/TIM.2024.3475788","url":null,"abstract":"In the above article [1], there is an error in (9). The correct equation is as below: begin{equation*} hat {sigma }^{2}=frac {sum _{j=1}^{16} Delta V_{j}^{2}}{16-6} =frac {sum _{j=1}^{16}left ({{V_{mathrm {cal}, j}-V_{mathrm {meas}, j}}}right )^{2}}{10}.end{equation*} The above error resulted in the calculated values for uncertainty displayed in Fig. 5 and Table III being smaller by 18%. The figure and table reflecting the correct values are as follows:TABLE IIISensor Parameters With Uncertainties channelx(mm)y(mm)z(mm) $n_x$ $n_y$ $n_z$ g(nT/V)CH240.39±0.5820.72±0.56163.84±0.410.1028±0.00840.1757±0.00810.9791±0.00171.176±0.011CH28116.76±0.22-14.51±0.293.61±0.230.9929±0.0008-0.1178±0.00690.0157±0.00691.135±0.008CH55-10.63±0.32116.38±0.2122.02±0.440.0262±0.00780.9993±0.0003-0.0254±0.00031.065±0.008CH115-135.52±0.577.07±0.6278.59±0.60-0.2907±0.00280.3784±0.00960.8788±0.00271.137±0.016CH140-19.74±0.42-116.89±0.3226.20±0.56-0.0926±0.0094-0.9951±0.00110.0352±0.00141.038±0.009The values following the plus-minus signs (±) correspond to uncertainties with a coverage factor $k=1$ . Fig. 5.Fig. 5.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-1"},"PeriodicalIF":5.6,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10750891","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142600445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-08DOI: 10.1109/TIM.2024.3493880
Bin Ju;Chao An;Yuzhi Gao;Ke Zhang;Siliang Lu;Yongbin Liu
This study aims to enhance the energy output of a cantilever beam (CB) by employing an analytical approach centered on stiffness matching. First, a single spring with the simplest structure is used as the driving load, with the root, middle, and end of the CB as driving sources, establishing a multiposition driving model. Theoretical analysis and finite element simulations are then conducted to elucidate the correlation between the energy output at each driving position of the CB and the stiffness of the spring. Subsequently, in order to test and evaluate the external excitation performance of the CB, the load structure of which the output is easy to observe, test and quantify must be selected. A diaphragm volume pump (DVP) is, hence, chosen as the driving load instead of the spring. A CB-driven DVP structure is established, and dynamic model analysis and fluid-solid coupling simulation are conducted. Findings suggest that the optimal stiffness for the diaphragm to match with the CB decreases as the CB’s external output stiffness diminishes, irrespective of the CB’s operational mode. An experimental setup featuring the CB-driven DVP is constructed for empirical validation, and the experimental outcomes corroborate the simulation results.
{"title":"Stiffness Matching of Cantilever Beam at Multipositions for Diaphragm Volume Pump Driving","authors":"Bin Ju;Chao An;Yuzhi Gao;Ke Zhang;Siliang Lu;Yongbin Liu","doi":"10.1109/TIM.2024.3493880","DOIUrl":"https://doi.org/10.1109/TIM.2024.3493880","url":null,"abstract":"This study aims to enhance the energy output of a cantilever beam (CB) by employing an analytical approach centered on stiffness matching. First, a single spring with the simplest structure is used as the driving load, with the root, middle, and end of the CB as driving sources, establishing a multiposition driving model. Theoretical analysis and finite element simulations are then conducted to elucidate the correlation between the energy output at each driving position of the CB and the stiffness of the spring. Subsequently, in order to test and evaluate the external excitation performance of the CB, the load structure of which the output is easy to observe, test and quantify must be selected. A diaphragm volume pump (DVP) is, hence, chosen as the driving load instead of the spring. A CB-driven DVP structure is established, and dynamic model analysis and fluid-solid coupling simulation are conducted. Findings suggest that the optimal stiffness for the diaphragm to match with the CB decreases as the CB’s external output stiffness diminishes, irrespective of the CB’s operational mode. An experimental setup featuring the CB-driven DVP is constructed for empirical validation, and the experimental outcomes corroborate the simulation results.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-13"},"PeriodicalIF":5.6,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-08DOI: 10.1109/TIM.2024.3485394
Lin Jiewei;Gou Xin;Zhu Xiaolong;Liu Zhisheng;Dai Huwei;Liu Xiaolei;Zhang Junhong
Due to the operation conditions of variable loads, it is challenging to achieve high-accuracy fault diagnosis of power machinery. The attention mechanism is widely used in this issue because of its ability to capture domain-invariant features of vibration signals. However, when the problem is specific to thermal engine diagnosis, the attention collapse can be caused by the interaction between load patterns and fault patterns. Consequently, the deep features converge to decrease the network generalization. To address this issue, this research employs the ensemble learning of crowd intelligence strategy, which is opposite to the attention mechanism of elite strategy. A multidepth step-training convolutional neural network (MDNN) is proposed. The multidepth architecture enhances feature diversity, and the step-training feature ensemble incorporates features into decision-making, thus overcoming feature convergence. The MDNN is tested using two datasets: a light-duty rotor-bearing test rig (electromechanical system) and a heavy-duty diesel engine test rig (thermodynamic machinery). According to the results, for the load-varying diesel engine, the attention mechanism exacerbates feature convergence, whereas MDNN effectively mitigates it. Meanwhile, with the mixture of four engine loads, the diagnosis accuracy of the attention mechanism-based network falls sharply to 54.27% from 59.20%, while the MDNN rises to 95.46%. The results offer a promising method for load-varying fault diagnosis of thermodynamic machinery and give a comprehensive understanding of the importance of avoiding feature convergence in the prognostic diagnosis of diesel engines.
{"title":"A Multidepth Step-Training Convolutional Neural Network for Power Machinery Fault Diagnosis Under Variable Loads","authors":"Lin Jiewei;Gou Xin;Zhu Xiaolong;Liu Zhisheng;Dai Huwei;Liu Xiaolei;Zhang Junhong","doi":"10.1109/TIM.2024.3485394","DOIUrl":"https://doi.org/10.1109/TIM.2024.3485394","url":null,"abstract":"Due to the operation conditions of variable loads, it is challenging to achieve high-accuracy fault diagnosis of power machinery. The attention mechanism is widely used in this issue because of its ability to capture domain-invariant features of vibration signals. However, when the problem is specific to thermal engine diagnosis, the attention collapse can be caused by the interaction between load patterns and fault patterns. Consequently, the deep features converge to decrease the network generalization. To address this issue, this research employs the ensemble learning of crowd intelligence strategy, which is opposite to the attention mechanism of elite strategy. A multidepth step-training convolutional neural network (MDNN) is proposed. The multidepth architecture enhances feature diversity, and the step-training feature ensemble incorporates features into decision-making, thus overcoming feature convergence. The MDNN is tested using two datasets: a light-duty rotor-bearing test rig (electromechanical system) and a heavy-duty diesel engine test rig (thermodynamic machinery). According to the results, for the load-varying diesel engine, the attention mechanism exacerbates feature convergence, whereas MDNN effectively mitigates it. Meanwhile, with the mixture of four engine loads, the diagnosis accuracy of the attention mechanism-based network falls sharply to 54.27% from 59.20%, while the MDNN rises to 95.46%. The results offer a promising method for load-varying fault diagnosis of thermodynamic machinery and give a comprehensive understanding of the importance of avoiding feature convergence in the prognostic diagnosis of diesel engines.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-14"},"PeriodicalIF":5.6,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142645488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Entropy has long been a subject that has attracted researchers from a diverse range of fields, including healthcare, finance, and fault detection. Slope entropy (SE) has recently been proposed as a new approach to address the shortcomings of permutation entropy (PE), which ignores magnitude information; however, SE is sensitive to parameters $boldsymbol {gamma }$ and $boldsymbol {delta }$ , and some information may be lost when segmenting symbols. The $boldsymbol {delta }$ , moreover, has only a limited gain on the time series classification performance of SE and increases the algorithm complexity. Considering the aforementioned limitations, this study introduces the concept of fuzzification to the SE and eliminates the $boldsymbol {delta }$ to simplify the parameters, resulting in the proposal of fuzzy SE (FuSE); furthermore, we incorporate the artificial rabbit optimization (ARO) algorithm to optimize the parameter $boldsymbol {gamma }$ to enhance the effectiveness of FuSE for time series classification and finally proposed an optimized FuSE (OFuSE). OFuSE can greatly reduce the information loss in the mapping process and adaptively search for the optimal parameter. The study evaluated FuSE and OFuSE on several synthetic datasets and concluded that FuSE is more sensitive to changes in signal amplitude and frequency while confirming the advantage of OFuSE in classification. The application of OFuSE on three different real datasets verifies that its classification performance and generalization ability are better than other entropy methods.
{"title":"Optimized Fuzzy Slope Entropy: A Complexity Measure for Nonlinear Time Series","authors":"Yuxing Li;Ge Tian;Yuan Cao;Yingmin Yi;Dingsong Zhou","doi":"10.1109/TIM.2024.3493878","DOIUrl":"https://doi.org/10.1109/TIM.2024.3493878","url":null,"abstract":"Entropy has long been a subject that has attracted researchers from a diverse range of fields, including healthcare, finance, and fault detection. Slope entropy (SE) has recently been proposed as a new approach to address the shortcomings of permutation entropy (PE), which ignores magnitude information; however, SE is sensitive to parameters \u0000<inline-formula> <tex-math>$boldsymbol {gamma }$ </tex-math></inline-formula>\u0000 and \u0000<inline-formula> <tex-math>$boldsymbol {delta }$ </tex-math></inline-formula>\u0000, and some information may be lost when segmenting symbols. The \u0000<inline-formula> <tex-math>$boldsymbol {delta }$ </tex-math></inline-formula>\u0000, moreover, has only a limited gain on the time series classification performance of SE and increases the algorithm complexity. Considering the aforementioned limitations, this study introduces the concept of fuzzification to the SE and eliminates the \u0000<inline-formula> <tex-math>$boldsymbol {delta }$ </tex-math></inline-formula>\u0000 to simplify the parameters, resulting in the proposal of fuzzy SE (FuSE); furthermore, we incorporate the artificial rabbit optimization (ARO) algorithm to optimize the parameter \u0000<inline-formula> <tex-math>$boldsymbol {gamma }$ </tex-math></inline-formula>\u0000 to enhance the effectiveness of FuSE for time series classification and finally proposed an optimized FuSE (OFuSE). OFuSE can greatly reduce the information loss in the mapping process and adaptively search for the optimal parameter. The study evaluated FuSE and OFuSE on several synthetic datasets and concluded that FuSE is more sensitive to changes in signal amplitude and frequency while confirming the advantage of OFuSE in classification. The application of OFuSE on three different real datasets verifies that its classification performance and generalization ability are better than other entropy methods.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-14"},"PeriodicalIF":5.6,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-08DOI: 10.1109/TIM.2024.3485393
Pengju Tian;Xianghong Hua
Most of the existing surface area measurement methods suffer from poor efficiency, low precision, and high computational cost, especially for inaccessible, large-scale, rough, and curved surfaces. In this article, we propose a method to directly measure the surface areas of unorganized point clouds applicable to various scenes. First, an adaptive supervoxel segmentation algorithm is adopted to divide the input point cloud into a collection of facets with multiple scales. For each facet, all points belonging to it are projected onto its corresponding accurately fit plane. Second, for each projected facet, rigid transform is performed so that its normal vector is parallel to the Z-axis. For each 2-D facet point cloud, the x-coordinates and-coordinates are utilized to abstract its boundary points. Third, the boundary points are sorted in clockwise order so that every two adjacent points and the center point determine a triangle. Next, an improved interpolation method is adopted to interpolate the sparse edge points. The surface area calculation results of different scales can be obtained by counting the sum of the triangular area inside each facet. Finally, the optimum value is determined from these results. The proposed method is tested on various types of point clouds acquired in different ways. Comprehensive experiments demonstrate that the proposed method is efficient and effective and is capable of obtaining good performances in both simple regular planes and complex surfaces. In particular, compared with traditional reconstruction-based methods, the proposed method significantly outperforms when dealing with large-scale and complex scenes.
现有的表面积测量方法大多存在效率低、精度低和计算成本高等问题，尤其是对于无法接近的、大尺度的、粗糙的和弯曲的表面。本文提出了一种直接测量无组织点云表面积的方法，适用于各种场景。首先，采用自适应上像素分割算法将输入点云划分为多个尺度的面集合。对于每个面，属于它的所有点都会被投影到相应的精确拟合平面上。其次，对每个投影面进行刚性变换，使其法线向量平行于 Z 轴。对于每个二维面点云，利用 x 坐标和坐标来抽象其边界点。第三，按顺时针顺序对边界点进行排序，使每两个相邻点和中心点组成一个三角形。然后，采用改进的插值方法对稀疏的边缘点进行插值。通过计算每个面内的三角形面积之和，可以得到不同尺度的表面积计算结果。最后，根据这些结果确定最佳值。所提出的方法在以不同方式获取的各类点云上进行了测试。综合实验证明，所提出的方法高效、有效，无论是在简单规则平面还是复杂曲面上，都能获得良好的性能。特别是，与传统的基于重建的方法相比，所提出的方法在处理大尺度和复杂场景时有明显的优势。
{"title":"Robust Surface Area Measurement of Unorganized Point Clouds Based on Multiscale Supervoxel Segmentation","authors":"Pengju Tian;Xianghong Hua","doi":"10.1109/TIM.2024.3485393","DOIUrl":"https://doi.org/10.1109/TIM.2024.3485393","url":null,"abstract":"Most of the existing surface area measurement methods suffer from poor efficiency, low precision, and high computational cost, especially for inaccessible, large-scale, rough, and curved surfaces. In this article, we propose a method to directly measure the surface areas of unorganized point clouds applicable to various scenes. First, an adaptive supervoxel segmentation algorithm is adopted to divide the input point cloud into a collection of facets with multiple scales. For each facet, all points belonging to it are projected onto its corresponding accurately fit plane. Second, for each projected facet, rigid transform is performed so that its normal vector is parallel to the Z-axis. For each 2-D facet point cloud, the x-coordinates and-coordinates are utilized to abstract its boundary points. Third, the boundary points are sorted in clockwise order so that every two adjacent points and the center point determine a triangle. Next, an improved interpolation method is adopted to interpolate the sparse edge points. The surface area calculation results of different scales can be obtained by counting the sum of the triangular area inside each facet. Finally, the optimum value is determined from these results. The proposed method is tested on various types of point clouds acquired in different ways. Comprehensive experiments demonstrate that the proposed method is efficient and effective and is capable of obtaining good performances in both simple regular planes and complex surfaces. In particular, compared with traditional reconstruction-based methods, the proposed method significantly outperforms when dealing with large-scale and complex scenes.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-17"},"PeriodicalIF":5.6,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recently, long-term gas detection has attracted much attention due to its being a key factor for electronic nose (E-Nose) applications. However, the sensor drift effect can significantly reduce the performance of the sensor. Therefore, in this work, we proposed a new drift compensation method by optimizing feature selection, model construction, and training methods to study drift-resistant gas detection based on convolutional neural network (CNN) methods. First, the attention mechanism is used to screen the specific features of the gas data and remove the low-weight features. Moreover, a multiscale feature extraction network is designed so that the features fused by the three-layer convolution are used as the final classification feature input to extract the depth features keeping the drift unchanged. Simultaneously, the segmented training method and the targeted cyclic training model are adopted to reduce the required experimental data. Importantly, based on the largest gas drift dataset currently, the proposed method maintains the average gas detection accuracy beyond 80% in three years, and the long-term stability of gas detection is effectively improved. Therefore, our findings provide an effective way to solve the sensor drift effect.
{"title":"Anti-Drift Gas Detection Algorithm Based on Neural Network","authors":"Jiayi Guo;Xu Li;Xiulei Li;Zheng Liang;Juexian Cao;Xiaolin Wei","doi":"10.1109/TIM.2024.3488159","DOIUrl":"https://doi.org/10.1109/TIM.2024.3488159","url":null,"abstract":"Recently, long-term gas detection has attracted much attention due to its being a key factor for electronic nose (E-Nose) applications. However, the sensor drift effect can significantly reduce the performance of the sensor. Therefore, in this work, we proposed a new drift compensation method by optimizing feature selection, model construction, and training methods to study drift-resistant gas detection based on convolutional neural network (CNN) methods. First, the attention mechanism is used to screen the specific features of the gas data and remove the low-weight features. Moreover, a multiscale feature extraction network is designed so that the features fused by the three-layer convolution are used as the final classification feature input to extract the depth features keeping the drift unchanged. Simultaneously, the segmented training method and the targeted cyclic training model are adopted to reduce the required experimental data. Importantly, based on the largest gas drift dataset currently, the proposed method maintains the average gas detection accuracy beyond 80% in three years, and the long-term stability of gas detection is effectively improved. Therefore, our findings provide an effective way to solve the sensor drift effect.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-8"},"PeriodicalIF":5.6,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-07DOI: 10.1109/TIM.2024.3488158
Xiang Liu
Defect detection is a critical component of quality control in the manufacturing of printed circuit boards (PCBs). However, accurately detecting PCB defects is challenging because they are very small and inconspicuous. In this article, an adaptive defect-aware attention network (ADANet) is proposed for PCB defect detection, and it contains two main modules: small defect preserving and location (SDPL) and defect segmentation prediction (DSP), where the SDPL module is designed to extract the high-resolution and multiscale defect feature representations to avoid the loss of small defects caused by model depth and then locate their positions with a deformable Transformer, and the DSP module is developed to predict their categories and masks. Experimental results conducted on two PCB datasets show that the proposed ADANet can surpass state-of-the-art approaches and achieve high performance in multiscale defect classification and detection results.
{"title":"An Adaptive Defect-Aware Attention Network for Accurate PCB-Defect Detection","authors":"Xiang Liu","doi":"10.1109/TIM.2024.3488158","DOIUrl":"https://doi.org/10.1109/TIM.2024.3488158","url":null,"abstract":"Defect detection is a critical component of quality control in the manufacturing of printed circuit boards (PCBs). However, accurately detecting PCB defects is challenging because they are very small and inconspicuous. In this article, an adaptive defect-aware attention network (ADANet) is proposed for PCB defect detection, and it contains two main modules: small defect preserving and location (SDPL) and defect segmentation prediction (DSP), where the SDPL module is designed to extract the high-resolution and multiscale defect feature representations to avoid the loss of small defects caused by model depth and then locate their positions with a deformable Transformer, and the DSP module is developed to predict their categories and masks. Experimental results conducted on two PCB datasets show that the proposed ADANet can surpass state-of-the-art approaches and achieve high performance in multiscale defect classification and detection results.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-11"},"PeriodicalIF":5.6,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142672062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The detection of defect severity in cable terminations plays a critical role in ensuring the safe and stable operation of high-speed trains (HSTs). However, the partial discharge (PD) characteristics of the same type of defect can appear similar across different severities, posing challenges for accurate insulation defect degree identification. Consequently, this article proposes an image transformation method, named the equivalent bandwidth matrix of relative locations (EBMRLs), coupled with the self-guided transformer (SG-Former) algorithm, which is more effective for fine-grained image recognition, to accurately identify different degrees of defects with similar PD characteristics. In the proposed approach, the original PD signals are first converted into images using EBMRL. This transformation embeds the characteristic and bandwidth information from the original PD data into the images, thereby reducing the similarity of information between classes in the transformed images and enhancing their distinguishability. Subsequently, the local and global features of the transformed EBMRL images are extracted to train the SG-Former model. The model is finally utilized to identify the severity of defects in cable terminations. The results demonstrate that the method proposed in this article achieves better performance compared with some of the state-of-the-art methods.
{"title":"Equivalent Bandwidth Matrix of Relative Locations: Image Modeling Method for Defect Degree Identification of In-Vehicle Cable Termination","authors":"Kai Liu;Shibo Jiao;Guangbo Nie;Hui Ma;Bo Gao;Chuanming Sun;Dongli Xin;Tapan Kumar Saha;Guangning Wu","doi":"10.1109/TIM.2024.3481567","DOIUrl":"https://doi.org/10.1109/TIM.2024.3481567","url":null,"abstract":"The detection of defect severity in cable terminations plays a critical role in ensuring the safe and stable operation of high-speed trains (HSTs). However, the partial discharge (PD) characteristics of the same type of defect can appear similar across different severities, posing challenges for accurate insulation defect degree identification. Consequently, this article proposes an image transformation method, named the equivalent bandwidth matrix of relative locations (EBMRLs), coupled with the self-guided transformer (SG-Former) algorithm, which is more effective for fine-grained image recognition, to accurately identify different degrees of defects with similar PD characteristics. In the proposed approach, the original PD signals are first converted into images using EBMRL. This transformation embeds the characteristic and bandwidth information from the original PD data into the images, thereby reducing the similarity of information between classes in the transformed images and enhancing their distinguishability. Subsequently, the local and global features of the transformed EBMRL images are extracted to train the SG-Former model. The model is finally utilized to identify the severity of defects in cable terminations. The results demonstrate that the method proposed in this article achieves better performance compared with some of the state-of-the-art methods.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-10"},"PeriodicalIF":5.6,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142598615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-05DOI: 10.1109/TIM.2024.3485406
Min Zhang;Jiamin Li;Jiliang Mo;Mingxue Shen;Zaiyu Xiang;Zhongrong Zhou
The braking system of high-speed trains is directly related to the operation safety of the train. The brake pads, which play a crucial role, will inevitably undergo uneven wear in long-term use, posing safety hazards to train braking. As the trains are in normal operating condition for long periods, it is difficult to collect usable uneven wear data, and there is a situation of data imbalance. This article proposes a trade-off contrastive learning network (TCLN), utilizing the differences between data and balancing the weights of different classes, which can realize the condition monitoring under the data imbalance of brake pads. First, data augmentation is employed to provide sufficient and diverse data for contrastive learning, and nonlinear features are extracted by a quadratic convolutional neural network (QCNN). Then, the designed class-weighted method is utilized to improve the characterization ability of the minority class data and realize the equidistant representation of features for each class, which in turn achieves the purpose of paying equal attention to all classes. Finally, the effectiveness of the proposed method is verified using the dataset collected from the scaling experiments, and the results show that the proposed method has higher accuracy and efficiency compared to other methods, which can still accurately identify the brake pad condition when the data are highly imbalanced.
{"title":"High-Speed Train Brake Pads Condition Monitoring Based on Trade-Off Contrastive Learning Network","authors":"Min Zhang;Jiamin Li;Jiliang Mo;Mingxue Shen;Zaiyu Xiang;Zhongrong Zhou","doi":"10.1109/TIM.2024.3485406","DOIUrl":"https://doi.org/10.1109/TIM.2024.3485406","url":null,"abstract":"The braking system of high-speed trains is directly related to the operation safety of the train. The brake pads, which play a crucial role, will inevitably undergo uneven wear in long-term use, posing safety hazards to train braking. As the trains are in normal operating condition for long periods, it is difficult to collect usable uneven wear data, and there is a situation of data imbalance. This article proposes a trade-off contrastive learning network (TCLN), utilizing the differences between data and balancing the weights of different classes, which can realize the condition monitoring under the data imbalance of brake pads. First, data augmentation is employed to provide sufficient and diverse data for contrastive learning, and nonlinear features are extracted by a quadratic convolutional neural network (QCNN). Then, the designed class-weighted method is utilized to improve the characterization ability of the minority class data and realize the equidistant representation of features for each class, which in turn achieves the purpose of paying equal attention to all classes. Finally, the effectiveness of the proposed method is verified using the dataset collected from the scaling experiments, and the results show that the proposed method has higher accuracy and efficiency compared to other methods, which can still accurately identify the brake pad condition when the data are highly imbalanced.","PeriodicalId":13341,"journal":{"name":"IEEE Transactions on Instrumentation and Measurement","volume":"73 ","pages":"1-10"},"PeriodicalIF":5.6,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142636293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","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：