Pub Date : 2023-06-01DOI: 10.1109/IS3C57901.2023.00044
Ya-Ling Kao, Gin-Der Peng, Jai-Tsung Hong, Iching Lin, Yu-Da Lin
While the number of remote-controlled drones in Taiwan continues to grow, it is necessary to set up a proper regulatory mechanism for public safety and interests. Although Taiwan has the first law in Asia to regulate remote-controlled drones, if a proper regulatory mechanism is not designed, not only will the relevant laws and regulations become a formality and fail to achieve the goal of protecting public interests, but it will even slow down the potential development of more remote-controlled drones. In this study, we have conducted a preliminary introduction and comparison of the documents related to network-based remote identification (remote ID) mechanisms proposed by the FAA and EASA and other advanced national aviation authorities and found that Europe, the United States, and other countries have set the goal of developing network-based remote ID. However, considering the current status of infrastructure and telecommunication services, most countries have set broadcast remote ID as the first target and have gradually developed a clear mechanism to follow. Although Taiwan currently has preliminary regulations in the relevant areas, the details have not been fully released. Therefore, this study will start from the legal and regulatory level and use literature analysis and comparative research to make a preliminary judgment on the relevant regulations as a reference for the future planning and development of the Radio Frequency Identification mechanism in Taiwan.
{"title":"A Preliminary Study on the Development of Drones Network Remote Identification Mechanism in Taiwan","authors":"Ya-Ling Kao, Gin-Der Peng, Jai-Tsung Hong, Iching Lin, Yu-Da Lin","doi":"10.1109/IS3C57901.2023.00044","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00044","url":null,"abstract":"While the number of remote-controlled drones in Taiwan continues to grow, it is necessary to set up a proper regulatory mechanism for public safety and interests. Although Taiwan has the first law in Asia to regulate remote-controlled drones, if a proper regulatory mechanism is not designed, not only will the relevant laws and regulations become a formality and fail to achieve the goal of protecting public interests, but it will even slow down the potential development of more remote-controlled drones. In this study, we have conducted a preliminary introduction and comparison of the documents related to network-based remote identification (remote ID) mechanisms proposed by the FAA and EASA and other advanced national aviation authorities and found that Europe, the United States, and other countries have set the goal of developing network-based remote ID. However, considering the current status of infrastructure and telecommunication services, most countries have set broadcast remote ID as the first target and have gradually developed a clear mechanism to follow. Although Taiwan currently has preliminary regulations in the relevant areas, the details have not been fully released. Therefore, this study will start from the legal and regulatory level and use literature analysis and comparative research to make a preliminary judgment on the relevant regulations as a reference for the future planning and development of the Radio Frequency Identification mechanism in Taiwan.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"314 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116531491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/IS3C57901.2023.00096
Li-Yang Ho, Wei-Jong Yang
With the development of computer vision, more and more systems for autonomous driving are adopting deep learning technology. Among them, lane detection aims to avoid accidents caused by the cars that deviate from their driving lanes. The lane detection task is challenging due to complex scenes and few features of distorted lane lines. Therefore, collecting the useful spatial information of the feature map related lane line becomes an important task for lane line detection. There are some spatial enhancements in feature maps, such as the spatial convolutional neural network (SCNN) [1] and the parallel spatial attention network (PSAN) [2]. To avoid computation in computing spatial attentions from top-to-bottom, left-to-right, bottom-to-top and right-to-left processes., in this paper, we design a more efficient detection system based on the PSAN concept, we shortened the spatial attention ranges, where the module only collects spatial local information and passes to the adjacent feature to reduce the computation time and enhance the lane detection performances. Simulation results show that the proposed parallel shortened spatial attention module can achieve effective and precision lane detection.
{"title":"Parallel Shortened Spatial Attention Module for Effective and Precision Lane Detection","authors":"Li-Yang Ho, Wei-Jong Yang","doi":"10.1109/IS3C57901.2023.00096","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00096","url":null,"abstract":"With the development of computer vision, more and more systems for autonomous driving are adopting deep learning technology. Among them, lane detection aims to avoid accidents caused by the cars that deviate from their driving lanes. The lane detection task is challenging due to complex scenes and few features of distorted lane lines. Therefore, collecting the useful spatial information of the feature map related lane line becomes an important task for lane line detection. There are some spatial enhancements in feature maps, such as the spatial convolutional neural network (SCNN) [1] and the parallel spatial attention network (PSAN) [2]. To avoid computation in computing spatial attentions from top-to-bottom, left-to-right, bottom-to-top and right-to-left processes., in this paper, we design a more efficient detection system based on the PSAN concept, we shortened the spatial attention ranges, where the module only collects spatial local information and passes to the adjacent feature to reduce the computation time and enhance the lane detection performances. Simulation results show that the proposed parallel shortened spatial attention module can achieve effective and precision lane detection.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115061170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/IS3C57901.2023.00049
Jun-Bin Zhang, Pei-Hsuan Tsai, Meng-Hsun Tsai
In real-world interactive applications, where videos are generated in real-time and require immediate feedback, online segmentation has practical advantages over offline inference. Many excellent previous models have been developed for offline scenarios, while real-time prediction for temporal action segmentation (TAS) is a difficult task. Some interactive applications can tolerate a certain amount of delay. In this paper, we propose a node delay embedding of a dynamic graph for real-time TAS. We transform the video stream into a dynamic graph stream that evolves over time. We define past, current, and future nodes to construct sub-graphs at each step. Specifically, future nodes are sampled using our proposed node delay method. A graph model is utilized to aggregate past, current, and future node information to update the representation of current nodes and predict their labels. To the best of our knowledge, it is the first real-time TAS graph model with delay embedding. Experiments show that delay embedding enhances node representation and improves performance. Overall, our proposed approach provides a promising solution for real-time TAS.
{"title":"DEDGraph: Delay Embedding of Dynamic Graph for Temporal Action Segmentation","authors":"Jun-Bin Zhang, Pei-Hsuan Tsai, Meng-Hsun Tsai","doi":"10.1109/IS3C57901.2023.00049","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00049","url":null,"abstract":"In real-world interactive applications, where videos are generated in real-time and require immediate feedback, online segmentation has practical advantages over offline inference. Many excellent previous models have been developed for offline scenarios, while real-time prediction for temporal action segmentation (TAS) is a difficult task. Some interactive applications can tolerate a certain amount of delay. In this paper, we propose a node delay embedding of a dynamic graph for real-time TAS. We transform the video stream into a dynamic graph stream that evolves over time. We define past, current, and future nodes to construct sub-graphs at each step. Specifically, future nodes are sampled using our proposed node delay method. A graph model is utilized to aggregate past, current, and future node information to update the representation of current nodes and predict their labels. To the best of our knowledge, it is the first real-time TAS graph model with delay embedding. Experiments show that delay embedding enhances node representation and improves performance. Overall, our proposed approach provides a promising solution for real-time TAS.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130905700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/IS3C57901.2023.00085
C. Tseng, Su-Ling Lee
Graph Fourier transform (GFT) is an important tool for analyzing the irregular graph signals collected from various real-world networks. One of its applications is the graph Fourier transform centrality (GFTC) which has been developed to find the influential nodes in the graphical representations of networks. In the traditional GFTC computation method, the eigen-decomposition of Laplacian matrix needs to be calculated for obtaining the GFT basis to compute GFTC. To reduce the computational complexity, a rational graph filter (RGF) method is presented in this paper. The main technique is that the spectral-domain computational task is converted to the vertex-domain one by using Parseval’s theorem of GFT. The Pade method and Maclaurin series expansion are applied to obtain the filter coefficients of RGF when the weight function of GFTC is specified. Finally, the Taipei metro network is used to demonstrate the effectiveness of GFTC index for identifying the important stations in the metro network.
{"title":"A Rational Graph Filter Method for GFT Centrality Computation","authors":"C. Tseng, Su-Ling Lee","doi":"10.1109/IS3C57901.2023.00085","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00085","url":null,"abstract":"Graph Fourier transform (GFT) is an important tool for analyzing the irregular graph signals collected from various real-world networks. One of its applications is the graph Fourier transform centrality (GFTC) which has been developed to find the influential nodes in the graphical representations of networks. In the traditional GFTC computation method, the eigen-decomposition of Laplacian matrix needs to be calculated for obtaining the GFT basis to compute GFTC. To reduce the computational complexity, a rational graph filter (RGF) method is presented in this paper. The main technique is that the spectral-domain computational task is converted to the vertex-domain one by using Parseval’s theorem of GFT. The Pade method and Maclaurin series expansion are applied to obtain the filter coefficients of RGF when the weight function of GFTC is specified. Finally, the Taipei metro network is used to demonstrate the effectiveness of GFTC index for identifying the important stations in the metro network.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131104953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/is3c57901.2023.00008
{"title":"Sponsors","authors":"","doi":"10.1109/is3c57901.2023.00008","DOIUrl":"https://doi.org/10.1109/is3c57901.2023.00008","url":null,"abstract":"","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"121 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133540068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With a large number of non-linear loads are integrated into the electric grids, they cause the significant power quality problems; meanwhile, harmonic is one of the problems that be paid more concerns, and it is necessary to seek useful ways for suppressions. Nowadays, solar PV power generations become one of the main streams among various renewable energy. Under the applicable control design, PV power not only can provide general power generation but also can used for power quality improvement. This paper thus mainly focuses on the study of harmonic suppression on low-voltage system by three-phase three-wire PV-APF system. For the proposed PV-APF system, it consists of: dual second order generalized integrator-phase-locked loop design for exact harmonic detection, self-tuning synchronous reference frame method for the calculation of harmonic compensation currents, a multi-mode strategy for the PV-APF to operate in different modes, and a bat-optimization PI controller for the better PV-APF control. MATLAB/Simulink is used for modeling and testing in this paper. Performance of proposed PV-APF design is finally be well validated by different load scenarios.
{"title":"Study on Harmonic Suppressions by optimization-based Three-Phase Three-Wire PV-APF","authors":"Yu-Jen Liu, Po-Yu Hou, Tsung-Han Kuo, Y. Lee, Chin-Chan Cheng, Yen-Fu Chen","doi":"10.1109/IS3C57901.2023.00034","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00034","url":null,"abstract":"With a large number of non-linear loads are integrated into the electric grids, they cause the significant power quality problems; meanwhile, harmonic is one of the problems that be paid more concerns, and it is necessary to seek useful ways for suppressions. Nowadays, solar PV power generations become one of the main streams among various renewable energy. Under the applicable control design, PV power not only can provide general power generation but also can used for power quality improvement. This paper thus mainly focuses on the study of harmonic suppression on low-voltage system by three-phase three-wire PV-APF system. For the proposed PV-APF system, it consists of: dual second order generalized integrator-phase-locked loop design for exact harmonic detection, self-tuning synchronous reference frame method for the calculation of harmonic compensation currents, a multi-mode strategy for the PV-APF to operate in different modes, and a bat-optimization PI controller for the better PV-APF control. MATLAB/Simulink is used for modeling and testing in this paper. Performance of proposed PV-APF design is finally be well validated by different load scenarios.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129432788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
BCI is a technology that enables individuals to interact with computers or other devices using only their brain signals. The mu rhythm is a type of EEG signal that is observed over the sensorimotor cortex during rest or motor tasks [1]. This paper investigates the presence of mu wave in Motor Imagery (MI) based Brain-Computer Interface (BCI) experiments using the Berlin BCI competition IV dataset 1. In this study, an epoch of 4 seconds each was extracted using Event codes and labels. Butterworth Bandpass of 8-12Hz, 8-14Hz, and 8-16Hz were used for preprocessing the data with three different frequency ranges, known to encompass the frequency range of mu waves. Common Spatial Patterns were used for feature extraction. We used the 80/20 method to split the data for training and testing the algorithms. Linear Discriminant Analysis (LDA), Support Vector Machines (SVM) were trained by using these extracted features, and Convolutional Neural Networks (CNN) were trained using the preprocessed data. Results show that the 8-16Hz frequency range is the most suitable for investigating the presence of mu waves in MI BCI experiments, as the classification accuracy of all three algorithms increased significantly in this range compared to the other two ranges. The study highlights the importance of selecting the appropriate frequency range for investigating the presence of mu waves in MI BCI experiments, and the results presented in this paper can aid in designing and optimizing BCI experiments and developing more accurate and reliable BCI systems in the future.
{"title":"Investigating the presence of mu signal during motor movements using SVM, LDA, and CNN","authors":"Maheswar Reddy Yelugoti, Cheng-Yi Lin, Shih-Chung Chen","doi":"10.1109/IS3C57901.2023.00011","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00011","url":null,"abstract":"BCI is a technology that enables individuals to interact with computers or other devices using only their brain signals. The mu rhythm is a type of EEG signal that is observed over the sensorimotor cortex during rest or motor tasks [1]. This paper investigates the presence of mu wave in Motor Imagery (MI) based Brain-Computer Interface (BCI) experiments using the Berlin BCI competition IV dataset 1. In this study, an epoch of 4 seconds each was extracted using Event codes and labels. Butterworth Bandpass of 8-12Hz, 8-14Hz, and 8-16Hz were used for preprocessing the data with three different frequency ranges, known to encompass the frequency range of mu waves. Common Spatial Patterns were used for feature extraction. We used the 80/20 method to split the data for training and testing the algorithms. Linear Discriminant Analysis (LDA), Support Vector Machines (SVM) were trained by using these extracted features, and Convolutional Neural Networks (CNN) were trained using the preprocessed data. Results show that the 8-16Hz frequency range is the most suitable for investigating the presence of mu waves in MI BCI experiments, as the classification accuracy of all three algorithms increased significantly in this range compared to the other two ranges. The study highlights the importance of selecting the appropriate frequency range for investigating the presence of mu waves in MI BCI experiments, and the results presented in this paper can aid in designing and optimizing BCI experiments and developing more accurate and reliable BCI systems in the future.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116255301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/IS3C57901.2023.00093
Chien-Chang Chen, Chia Hung Wei, Cheng-Shian Lin
Anomaly detection is an important research topic in artificial intelligent studies. Among anomaly detection applications, fabric defect detections obtain lots of research interests due to its industrial potential. This study presents an efficient method to detect fabric defect regions by the Siamese network for greatly reducing the training time by only using limited training data. The model identifies texture features by using some normal and defect images. Defect regions can be detected through overlapped blocks identification and the block size determines the precisions of detection correctness and locality. At last, the proposed structure is compared with the conventional Bergmann’s autoencoder, the Alexnet-based autoencoder, and the VGG16-based autoencoder models. Experimental results show that the proposed structure requires limited training time comparing with autoencoder models and exhibits good recognition rate.
{"title":"Fast Detection of Fabric Defects based on Neural Networks","authors":"Chien-Chang Chen, Chia Hung Wei, Cheng-Shian Lin","doi":"10.1109/IS3C57901.2023.00093","DOIUrl":"https://doi.org/10.1109/IS3C57901.2023.00093","url":null,"abstract":"Anomaly detection is an important research topic in artificial intelligent studies. Among anomaly detection applications, fabric defect detections obtain lots of research interests due to its industrial potential. This study presents an efficient method to detect fabric defect regions by the Siamese network for greatly reducing the training time by only using limited training data. The model identifies texture features by using some normal and defect images. Defect regions can be detected through overlapped blocks identification and the block size determines the precisions of detection correctness and locality. At last, the proposed structure is compared with the conventional Bergmann’s autoencoder, the Alexnet-based autoencoder, and the VGG16-based autoencoder models. Experimental results show that the proposed structure requires limited training time comparing with autoencoder models and exhibits good recognition rate.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"199 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115040406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/is3c57901.2023.00089
Jun-Yan Huang, Hsiao-Chuan Liu, Jian-Xing Wu
In endoscopic ultrasound (EUS) imaging, high-resolution and deep imaging depth are often desired. However, achieving high-resolution images requires the use of a higher frequency transducer, which in turn results in reduced imaging depth. This trade-off between high-resolution and imaging depth cannot be easily resolved. To address this issue, a cylindrical probe consisting of three 30 MHz transducers with different imaging depths was developed. This probe enables imaging with both high resolution and deeper depth, and the imaging results from the three transducers combined were found to be superior to those of a single transducer. The depth of field (DOF) of the fused images was also three times that of a single transducer. This technology will provide a clearer, more convenient, and efficient diagnosis of gastrointestinal diseases in the future. In this experiment, a triple-frequency cylindrical probe is designed, in which the frequencies of the three transducer elements are all 30 MHz, and the different detection depths of 10 MHz, 20 MHz and 30 MHz are simulated by the embedded depth. The sound field module uses PWM-PT as the piezoelectric material simulation, and a single frequency transducer is used to generate the sound field to display the imaging status and energy concentration at different imaging depths of 10 MHz, 20 MHz, and 30 MHz to detect whether the size of the piezoelectric material setup is consistent. In addition, in order to test the performance of the transducer and evaluate the imaging situation, the imaging situation of 10 MHz, 20 MHz and 30 MHz was observed at imaging depth of 1 mm for the designed transducer, and the results showed that 30 MHz had the best imaging effect. In addition, the photoacoustic and ultrasound images obtained by scanning pig intestines were fused using MATLAB. The fusion results show the multi-layered tissue echo signals and the distribution of hemoglobin in the tissue.
{"title":"Development of an Multi-frequency Photoacoustic Endoscopy Probe Diagnosis System for Biomedical Applications","authors":"Jun-Yan Huang, Hsiao-Chuan Liu, Jian-Xing Wu","doi":"10.1109/is3c57901.2023.00089","DOIUrl":"https://doi.org/10.1109/is3c57901.2023.00089","url":null,"abstract":"In endoscopic ultrasound (EUS) imaging, high-resolution and deep imaging depth are often desired. However, achieving high-resolution images requires the use of a higher frequency transducer, which in turn results in reduced imaging depth. This trade-off between high-resolution and imaging depth cannot be easily resolved. To address this issue, a cylindrical probe consisting of three 30 MHz transducers with different imaging depths was developed. This probe enables imaging with both high resolution and deeper depth, and the imaging results from the three transducers combined were found to be superior to those of a single transducer. The depth of field (DOF) of the fused images was also three times that of a single transducer. This technology will provide a clearer, more convenient, and efficient diagnosis of gastrointestinal diseases in the future. In this experiment, a triple-frequency cylindrical probe is designed, in which the frequencies of the three transducer elements are all 30 MHz, and the different detection depths of 10 MHz, 20 MHz and 30 MHz are simulated by the embedded depth. The sound field module uses PWM-PT as the piezoelectric material simulation, and a single frequency transducer is used to generate the sound field to display the imaging status and energy concentration at different imaging depths of 10 MHz, 20 MHz, and 30 MHz to detect whether the size of the piezoelectric material setup is consistent. In addition, in order to test the performance of the transducer and evaluate the imaging situation, the imaging situation of 10 MHz, 20 MHz and 30 MHz was observed at imaging depth of 1 mm for the designed transducer, and the results showed that 30 MHz had the best imaging effect. In addition, the photoacoustic and ultrasound images obtained by scanning pig intestines were fused using MATLAB. The fusion results show the multi-layered tissue echo signals and the distribution of hemoglobin in the tissue.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132048925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-06-01DOI: 10.1109/is3c57901.2023.00105
Ke-Wei Lin, Wei-Ling Lin, Y. Tsai, F. Hsiao
We achieved a fault diagnosis for a certain air pump using an artificial neural network. The operating sound of the pump is recorded by a single microphone, after processing by an unsupervised autoencoder, 108 groups of samples containing only 1-second audio data are inputted to the neural network classifier. The training rounds and the neurons of the autoencoder are tested. After training, the provided detection network can finally give the classifying accuracy of up to 99% according to 1-sec sound data.
{"title":"Sound based fault classify diagnosis method using artificial neural network and autoencoder processing","authors":"Ke-Wei Lin, Wei-Ling Lin, Y. Tsai, F. Hsiao","doi":"10.1109/is3c57901.2023.00105","DOIUrl":"https://doi.org/10.1109/is3c57901.2023.00105","url":null,"abstract":"We achieved a fault diagnosis for a certain air pump using an artificial neural network. The operating sound of the pump is recorded by a single microphone, after processing by an unsupervised autoencoder, 108 groups of samples containing only 1-second audio data are inputted to the neural network classifier. The training rounds and the neurons of the autoencoder are tested. After training, the provided detection network can finally give the classifying accuracy of up to 99% according to 1-sec sound data.","PeriodicalId":142483,"journal":{"name":"2023 Sixth International Symposium on Computer, Consumer and Control (IS3C)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125045611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：