检索结果-内蒙古大学图书馆

Material Segmentation Using 1-d convolutional neural network With Transient Histogram

IEEE ACCESS 2025年 13卷 59295-59308页

作者： Pratama, Yohanssen Kitano, Kazuya Kushida, Takahiro Fujimura, Yuki Funatomi, Takuya Mukaigawa, Yasuhiro Nara Inst Sci & Technol Grad Sch Sci & Technol Ikoma Nara 6300192 Japan Ritsumeikan Univ Coll Informat Sci & Engn Osaka Ibaraki 5678570 Japan

This study introduces a method for material classification using transient histograms obtained via a single-photon avalanche diode (SPAd) sensor. Temporal resolution in optical sensing plays a crucial role in material classification and surface segmentation, particularly for distinguishing materials with similar visual properties. In this study, SPAd sensors were utilized to capture transient histograms with temporal resolutions ranging from 13 picoseconds to 208 picoseconds, enabling precise extraction of temporal signatures for various materials. A comparative evaluation of classification techniques, including one-dimensional convolutional neural networks (1-d CNN), random forest (RF), support vector classifier (SVC), and k-nearest neighbors (KNN), was conducted to assess the impact of temporal resolution and exposure time on classification accuracy. 1-d CNN achieved the highest classification accuracy of 99.25% at a temporal resolution of 13 ps and an exposure time of 0.09 s, significantly outperforming other methods. Additionally, the proposed SPAd-based system was evaluated for material segmentation on non-planar surfaces. In a real-world experiment, 1-d CNN achieved an overall accuracy of 87.5% in differentiating visually similar materials, demonstrating the effectiveness of transient histograms for material classification where conventional RGB-based methods fail. These findings highlight the potential of SPAd sensors combined with advanced classification techniques to enhance material classification and segmentation, providing a versatile framework for applications in robotics, computer vision, and optical sensing.

关键词： Photonics Single-photon avalanche diodes Transient analysis Histograms Robot sensing systems Imaging convolutional neural networks Lighting Image resolution Cameras Material classification single photon avalanche diode surface segmentation temporal resolution transient histogram 1-d convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Leak detection and size identification in fluid pipelines using a novel vulnerability index and 1-d convolutional neural network

引用

ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUId MECHANICS 2023年第1期17卷 Article: 2165159页

作者： Ahmad, Zahoor Nguyen, Tuan-Khai Kim, Jong-Myon Univ Ulsan Dept Elect Elect & Comp Engn Ulsan South Korea PD Technol Co Ltd Ulsan South Korea PD Technol Co Ltd Dept Elect Elect & Comp Engn Bldg 793 Daehak Ro Ulsan 44610 South Korea

This paper proposes a leak detection and size identification technique in fluid pipelines based on a new leak-sensitive feature called the vulnerability index (VI) and 1-d convolutional neural network (1d-CNN). The acoustic emission hit (AEH) features can differentiate between normal and leak operating conditions of the pipeline. However, the multiple sources of acoustic emission hits, such as fluid pressure on the joints, interference noises, flange vibrations, and leaks in the pipeline, make the features less sensitive toward leak size identification in the pipeline. To address this issue, acoustic emission hit features are first extracted from the acoustic emission (AE) signal using a sliding window with an adaptive threshold. Since the distribution of the acoustic emission hit features changes according to the pipeline working conditions, a newly developed multiscale Mann-Whitney test (MMU-Test) is applied to the acoustic emission hit features to obtain the new vulnerability index feature, which shows the pipeline's susceptibility to leak and changes according to the pipeline working conditions. Finally, the vulnerability index is provided as input to a 1-d-CNN for leak detection and size identification, whose experimental results show a higher accuracy as compared to the reference state-of-the-art methods under variable fluid pressure conditions.

关键词： Pipelines acoustic emission leak detection 1-d convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Cavitation intensity recognition for high-speed axial piston pumps using 1-d convolutional neural networks with multi-channel inputs of vibration signals

引用

ALEXANdRIA ENGINEERING JOURNAL 2020年第6期59卷 4463-4473页

作者： Chao, Qun Tao, Jianfeng Wei, Xiaoliang Wang, Yuanhang Meng, Linghui Liu, Chengliang Shanghai Jiao Tong Univ State Key Lab Mech Syst & Vibrat Shanghai 200240 Peoples R China China Elect Prod Reliabil & Environm Testing Res Guangzhou 510610 Peoples R China

Raising rotational speed is an effective way to improve power density of axial piston pumps, but high rotational speed tends to cause undesirable cavitation in the pump. Although some machine learning methods have been successfully applied to detect the cavitation with high accuracy, these conventional methods suffer from the drawback of time-consuming and experience-dependent manual feature extraction. In this paper, a new model based on 1-d convolutional neural network (CNN) is proposed to recognize the cavitation intensity of axial piston pumps. To improve the recognition accuracy under noisy environment, the 1-d CNN receives multi-channel vibration data instead of single-channel data. The experimental results show that the proposed anti-noise 1-d CNN model with multi-channel inputs can achieve 15% higher recognition accuracy than its counterpart with single-channel input on a testing set with SNR = 5 dB. (C) 2020 The Authors. Published by Elsevier B.V.

关键词： Axial piston pump Cavitation Vibration 1-d convolutional neural network Multi-channel inputs Anti-noise

来源：评论

学校读者我要写书评

暂无评论

Rub-Impact Fault diagnosis of Rotating Machinery Based on 1-d convolutional neural networks

引用

IEEE SENSORS JOURNAL 2020年第15期20卷 8349-8363页

作者： Wu, Xinya Peng, Zhike Ren, Jishun Cheng, Changming Zhang, Wenming Wang, dong Shanghai Jiao Tong Univ State Key Lab Mech Syst & Vibrat Shanghai 200240 Peoples R China Shanghai Elect Power Generat Equipment Co Ltd Shanghai 200240 Peoples R China Beijing Zhongyuan Risen Technol Co Ltd Beijing 100000 Peoples R China

Rub-impact is a kind of serious malfunction, which often occurs in rotating machinery. The non-stationary rub-impact signals are always submerged in the background and noise signals, which makes it difficult to accurately diagnose the rubbing based on the hand-designed features extracted by the traditional methods. This paper presents a 1-d convolutional neural network (CNN) based approach to automatically learn useful features for rub-impact fault diagnosis from the raw vibration signals of a rotor system. The proposed model is trained on a dataset of vibration signals obtained from an industrial hydro turbine rotor. The results show that timely and accurate rub-impact fault detection can be achieved by a simple 1-d CNN configuration.

关键词： Feature extraction Fault diagnosis Vibrations deep learning Convolution Rotors Hydraulic turbines 1-d convolutional neural network rub-impact fault diagnosis rotating machinery

来源：评论

学校读者我要写书评

暂无评论

A Novel Series Arc Fault detection Method Based on CEEMdAN and IFAW-1dCNN

引用

IEEE TRANSACTIONS ON dIELECTRICS ANd ELECTRICAL INSULATION 2024年第2期31卷 1020-1029页

作者： Wu, Jinming Wang, Wei Shang, Tongtong Cao, Junteng Shandong Univ Technol Coll Elect & Elect Engn Zibo 255049 Peoples R China Shandong Univ Technol Dept Elect & Elect Engn Zibo 255049 Peoples R China

Machine learning-based fault detection technology gains significant attention in recent years. However, the practical implementation of such technologies often encounters challenges in selecting appropriate fault characteristics and network parameters. This study proposes a novel method for series arc fault detection based on the combined approach of complete ensemble empirical mode decomposition with adaptive noise (CEEMdAN) and improved fireworks algorithm-1-d convolutional neural network (IFWA-1dCNN). Following CEEMdAN utilization to decompose the current signal to obtain a set of intrinsic mode functions (IMF), the Spearman correlation coefficient and energy value are utilized to select the characteristic IMF, addressing the issue of some IMF lacking or containing insufficient fault features. To address the problem of manual parameter selection for 1dCNN and the difficulty in obtaining optimal parameters, an IFWA is then proposed for parameter optimization and selection in 1dCNN. The experimental results indicate that the proposed method achieved a fault recognition accuracy of 98.23%, outperforming traditional methods that use the original current signal as input for 1dCNN. Overall, these findings highlight the enhanced recognition capability of the proposed method.

关键词： Feature extraction Circuit faults Fault detection Convolution Explosions convolutional neural networks Fault currents 1-d convolutional neural network characterization intrinsic mode function (IMF) extraction complete ensemble empirical mode decomposition with adaptive noise (CEEMdAN) improved fireworks algorithm (FWA) series arc fault

来源：评论

学校读者我要写书评

暂无评论

deep multi-model fusion network based real object tactile understanding from haptic data

引用

APPLIEd INTELLIGENCE 2022年第14期52卷 16605-16620页

作者： Joolee, Joolekha Bibi Uddin, Md Azher Jeon, Seokhee Kyung Hee Univ Dept Comp Sci & Engn 446-701 Seocheon Dong Yongin 446701 Gyeonggi Do South Korea Heriot Watt Univ Dubai Dept Math & Comp Sci Dubai U Arab Emirates

The tactile information of an object is one of the crucial features which define the impression of that object. This paper presents a novel multi-model fusion network for real object's tactile understanding from haptic data. Furthermore, a low-cost 3d printed artificial finger-based tactile sensing system is designed for capturing haptic information in the form of acceleration profile, angular velocity, and normal force. Our proposed multi-model fusion network includes three different networks. First, we introduce a novel ensemble 2d convolutional neural network, namely SpectroNet, which captures the spatial features from the spectrogram of acceleration profile. Second, we design a 1-d convolutional neural network (CNN) with residual connection for extracting detailed spatial information from each segment of collected data. Third, we design bi-directional gated recurrent unit networks (BiGRU) to capture temporal dynamics. Moreover, the attention mechanism is utilized in all three proposed networks to assign weights to the features according to their contributions, which enhance the performance further. Finally, extensive experimental analysis is conducted on our dataset (i.e., 60 real objects, which cover both planner and non-planner surfaces) as well as the TUM surface material database. Empirical evaluations demonstrate that the proposed method significantly outperformed state-of-the-art methods in terms of accuracy, precision, recall and F1-score. Furthermore, we also found that the proposed multi-model fusion network substantially improves the performance compared to the single network.

关键词： Tactile understanding Multi-model fusion network SpectroNet 1-d convolutional neural network Bi-directional gated recurrent unit networks

来源：评论

学校读者我要写书评

暂无评论

LSTM and 1-d convolutional neural networks for Predictive Monitoring of the Anaerobic digestion Process 28th

LSTM and 1-D Convolutional Neural Networks for Predictive Mo...

引用

28th International Conference on Artificial neural networks (ICANN)

作者： McCormick, Mark Villa, Alessandro E. P. Univ Lausanne NeuroHeurist Res Grp Quartier UNIL Dorigny CH-1015 Lausanne Switzerland

ISBN: (纸本)9783030304935;9783030304928

Anaerobic digestion is a natural process that transforms organic substrates to methane and other products. Under controlled conditions the process has been widely applied to manage organic wastes. Improvements in process control are expected to lead to improvements in the technical and economic efficiency of the process. This paper presents and compares 3 different neural network model architectures for use as anaerobic digestion process predictive models. The models predict the future biogas production trend from measured physical and chemical parameters. The first model features an LSTM layer, the second model features a 1-d convolutional layer and the third model combines 2 separate inputs and parallel treatment using LSTM and 1-d convolutional layers followed by merging to produce a single prediction. The predictions can be used to adaptively adjust the substrate feeding rate in accordance with the transient state of the digestion process as defined by liquid feeding rate, the organic acid and ammonium ion concentrations and the pH of the digester liquid phase. The training and testing data were obtained during 1 year of continuous operation of a pilot-plant treating restaurant wastes. PLS regression and ICA were used to select the most relevant process parameters from the data. The 1-d convolutional based model comprising 272 trainable parameters predicted the future biogas flow rate changes with accuracy as high as 89% and an average accuracy of 58%. The work-flow can be applied to optimize the control of the study digester and to control bioreactors in general.

关键词： LSTM 1-d convolutional neural network Bioprocess

来源：评论

学校读者我要写书评

暂无评论

convolutional neural network Based Heart Sounds Recognition on Edge Computing Platform

Convolutional Neural Network Based Heart Sounds Recognition ...

引用

IEEE International Instrumentation and Measurement Technology Conference (I2MTC)

作者： Vakamullu, Venkatesh Trivedy, Sudipto Mishra, Madhusudhan Mukherjee, Anirban Indian Inst Technol Kharagpur Dept Elect Engn Kharagpur W Bengal India Tata Steel Jamshedpur Bihar India North Eastern Reg Inst Sci & Technol Dept Elect & Commun Engn Nirjuli India

ISBN: (纸本)9781665483605

Cardiovascular diseases (CVds) are often characterized by audio properties of heart sounds. The recognition of first and second heart sounds namely S-1 and S-2 is the first and the foremost step in detecting any cardiac abnormality. This work presents an implementation of optimal 1-dimensional convolutional neural network (1-d CNN) for S-1 and S-2 recognition on an edge computing platform. The optimal 1-d CNN model trained in the general purpose computing platform is deployed in the Raspberry Pi 3B based edge device for real time recognition of S-1 and S-2. In addition, to the best of our knowledge, this study is novice and aims for investigation of S-1 and S-2 recognition considering different portions of the segmented S-1 and S-2 signals obtained from the publicly available database as well as experimentally recorded phonocardiogram(PCG) signals. The achieved recognition accuracies are promising for both the databases with different decimation factors of data. The complete setup emulates the real time environment for heart sound recognition, thereby it paves a way for further analysis of the PCG signals for assisting physicians.

关键词： Auscultation cardiovascular abnormalities phonocardiogram 1-d convolutional neural network embedded platform

来源：评论

学校读者我要写书评

暂无评论

Bayesian-Based Hyperparameter Optimization of 1d-CNN for Structural Anomaly detection

引用

SENSORS 2023年第11期23卷 5058-5058页

作者： Li, Xiaofei Guo, Hainan Xu, Langxing Xing, Zezheng Dalian Maritime Univ Coll Transportat Engn Dalian 116026 Peoples R China Univ Jinan Coll Informat Sci & Engn Jinan 250022 Peoples R China

With the rapid development of sensor technology, structural health monitoring data have tended to become more massive. deep learning has advantages when handling big data, and has therefore been widely researched for diagnosing structural anomalies. However, for the diagnosis of different structural abnormalities, the model hyperparameters need to be adjusted according to different application scenarios, which is a complicated process. In this paper, a new strategy for building and optimizing 1d-CNN models is proposed that is suitable for diagnosing damage to different types of structure. This strategy involves optimizing hyperparameters with the Bayesian algorithm and improving model recognition accuracy using data fusion technology. Under the condition of sparse sensor measurement points, the entire structure is monitored, and the high-precision diagnosis of structural damage is performed. This method improves the applicability of the model to different structure detection scenarios, and avoids the shortcomings of traditional hyperparameter adjustment methods based on experience and subjectivity. In preliminary research on the simply supported beam test case, the efficient and accurate identification of parameter changes in small local elements was achieved. Furthermore, publicly available structural datasets were utilized to verify the robustness of the method, and a high identification accuracy rate of 99.85% was achieved. Compared with other methods described in the literature, this strategy shows significant advantages in terms of sensor occupancy rate, computational cost, and identification accuracy.

关键词： structural anomaly detection 1-d convolutional neural network Bayesian optimization algorithm decision-level fusion vibration signals

来源：评论

学校读者我要写书评

暂无评论

A multi-task learning for cavitation detection and cavitation intensity recognition of valve acoustic signals

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2022年 113卷

作者： Sha, Yu Faber, Johannes Gou, Shuiping Liu, Bo Li, Wei Schramm, Stefan Stoecker, Horst Steckenreiter, Thomas Vnucec, domagoj Wetzstein, Nadine Widl, Andreas Zhou, Kai Xidian Univ Acad Adv Interdisciplinary Res Key Lab Intelligent Percept & Image Understanding Minist Educ Xian 710071 Peoples R China Frankfurt Inst Adv Studies D-60438 Frankfurt Germany Giersch Sci Ctr Xidian FIAS Int Joint Res Ctr D-60438 Frankfurt Germany Goethe Univ Frankfurt Inst Theoret Phys D-60438 Frankfurt Germany GSI Helmholtzzentrum Schwerionenforsch GmbH D-64291 Darmstadt Germany SAMSON AG D-60314 Frankfurt Germany

With the rapid development of smart manufacturing, data-driven machinery health management has received a growing attention. As one of the most popular methods in machinery health management, deep learning (dL) has achieved remarkable successes. However, due to the issues of limited samples and poor separability of different cavitation states of acoustic signals, which greatly hinder the eventual performance of dL modes for cavitation intensity recognition and cavitation detection. Also different tasks were performed separately conventionally. In this work, a novel multi-task learning framework for simultaneous cavitation detection and cavitation intensity recognition framework using 1-d double hierarchical residual networks (1-d dHRN) is proposed for analyzing valves acoustic signals. Firstly, a data augmentation method based on sliding window with fast Fourier transform (Swin-FFT) is developed to alleviate the small-sample issue confronted in this study. Secondly, a 1-d double hierarchical residual block (1-d dHRB) is constructed to capture sensitive features from the frequency domain acoustic signals of valve. Then, a new structure of 1-d dHRN is proposed. Finally, the devised 1-d dHRN is evaluated on two datasets of valve acoustic signals without noise (dataset 1 and dataset 2) and one dataset of valve acoustic signals with realistic surrounding noise (dataset 3) provided by SAMSON AG (Frankfurt). Our method has achieved state-of-the-art results. The prediction accuracies of 1-d dHRN for cavitation intensitys recognition are as high as 93.75%, 94.31% and 100%, which indicates that 1-d dHRN outperforms other dL models and conventional methods. At the same time, the testing accuracies of 1-d dHRN for cavitation detection are as high as 97.02%, 97.64% and 100%. In addition, 1-d dHRN has also been tested for different frequencies of samples and shows excellent results for frequency of samples that mobile phones can accommodate.

关键词： Valves acoustics signal Cavitation intensity recognition Cavitation detection Multi-task learning 1-d convolutional neural network 1-d double hierarchical residual block

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：