In topographic maps, It is difficult to separate the linear elements, including contour lines, roads, latitude and longitude lines from complicated background due to the pixels with aliasing and false colors, and there exists some background in the result images extracted by the existing methods, especially when the color and energy of linear elements and background are similar in some particular maps, or the maps have low contrast contour lines. To solve these problems, this paper introduces the idea of seed spreading, and puts forward a novel method for separating linear elements. In this method, all the seeds carry the color information of the pixels and the energy information in the negative grayscale images, and they can search other pixels as their brothers to be combined into seed groups according to the color and energy similarity. The seeds have good perception of the environment around them, and the shapes of the seed groups are variable. Furthermore, the seeds are determined as linear elements by analyzing the color and energy differences between the seed groups and the areas around them. The experimental results show that our method can distinguish linear elements from the background more accurately than the previous methods.
{"title":"Linear Elements Separation via Vision System Feature and Seed Spreading from Topographic Maps","authors":"Fei Xie, Yanning Zhang, Xinming Guo, Wei Zhang, Zhaoyong Zhou, Pengfei Xu","doi":"10.1109/CIS52066.2020.00011","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00011","url":null,"abstract":"In topographic maps, It is difficult to separate the linear elements, including contour lines, roads, latitude and longitude lines from complicated background due to the pixels with aliasing and false colors, and there exists some background in the result images extracted by the existing methods, especially when the color and energy of linear elements and background are similar in some particular maps, or the maps have low contrast contour lines. To solve these problems, this paper introduces the idea of seed spreading, and puts forward a novel method for separating linear elements. In this method, all the seeds carry the color information of the pixels and the energy information in the negative grayscale images, and they can search other pixels as their brothers to be combined into seed groups according to the color and energy similarity. The seeds have good perception of the environment around them, and the shapes of the seed groups are variable. Furthermore, the seeds are determined as linear elements by analyzing the color and energy differences between the seed groups and the areas around them. The experimental results show that our method can distinguish linear elements from the background more accurately than the previous methods.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115584697","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00034
Xingsi Xue, Jiawei Lu, Junfeng Chen
A biomedical ontology provides a formal definition on the concepts and their relationships in the biomedical domain, which supports applications such as biomedical data annotation, knowledge integration, search and analysis. Different biomedical ontologies are mostly developed independently, and thus, establishing meaningful links between their entities, so-called ontology matching, is critical to implement their inter-operation. Since biomedical research usually spans multiple domains and topics, which motivates a new type of complex ontology matching, i.e. compound ontology matching, which involves more than two ontologies. Due to the complexity of the ontology matching problem, Evolutionary Algorithm (EA) can present a good methodology for determining ontology alignments. However, there exist different aspects of a solution that are partially or wholly in conflict, and the single-objective EA may lead to unwanted bias to one of them and reduce the solution's quality. To improve the ternary compound alignment's quality when matching three biomedical ontologies, in this work, a compact Multi-Objective Evolutionary Algorithm Based On Adaptive Objective Space Decomposition (cMOEA-AOSD) based matching technique is proposed. In particular, a Ternary Compound Concept Similarity Measure (TCCSM) is proposed to calculate the similarity value of three biomedical concepts, a mathematical model for ternary compound matching problem is constructed, and a cMOEA-AOSD is presented to address it, which is able to adaptively decompose the objective space to ensure the diversity of the solutions in Pareto Front (PF) and the quality of the final solution. The experiment uses six testing cases that consists of nine biomedical ontologies to test our proposal's performance, and the experimental results show that cMOEA-AOSD significantly out performs other MOEA-based matching technique and the state-of-the-art ternary compound matching techniques.
生物医学本体提供了生物医学领域中概念及其关系的形式化定义,支持生物医学数据标注、知识集成、搜索和分析等应用。不同的生物医学本体大多是独立开发的,因此,在它们的实体之间建立有意义的联系,即所谓的本体匹配,是实现它们互操作的关键。由于生物医学研究通常跨越多个领域和主题,这就催生了一种新型的复杂本体匹配,即复合本体匹配,它涉及两个以上的本体。由于本体匹配问题的复杂性,进化算法为确定本体对齐提供了一种很好的方法。然而,解决方案的不同方面存在部分或全部冲突,单一目标EA可能会导致对其中一个方面的不必要的偏见,并降低解决方案的质量。为了提高生物医学本体匹配时三元化合物匹配的质量,提出了一种基于自适应目标空间分解的紧凑多目标进化算法(cMOEA-AOSD)匹配技术。在此基础上,提出了三元化合物概念相似度度量(TCCSM)来计算三个生物医学概念的相似度值,构建了三元化合物匹配问题的数学模型,并提出了cMOEA-AOSD来解决该问题,该模型能够自适应分解目标空间,以保证Pareto Front (PF)解的多样性和最终解的质量。实验使用了包含9个生物医学本体的6个测试用例来测试我们的性能,实验结果表明,cMOEA-AOSD显著优于其他基于moea的匹配技术和最先进的三元化合物匹配技术。
{"title":"Ternary Compound Matching of Biomedical Ontologies with Compact Multi-Objective Evolutionary Algorithm Based on Adaptive Objective Space Decomposition","authors":"Xingsi Xue, Jiawei Lu, Junfeng Chen","doi":"10.1109/CIS52066.2020.00034","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00034","url":null,"abstract":"A biomedical ontology provides a formal definition on the concepts and their relationships in the biomedical domain, which supports applications such as biomedical data annotation, knowledge integration, search and analysis. Different biomedical ontologies are mostly developed independently, and thus, establishing meaningful links between their entities, so-called ontology matching, is critical to implement their inter-operation. Since biomedical research usually spans multiple domains and topics, which motivates a new type of complex ontology matching, i.e. compound ontology matching, which involves more than two ontologies. Due to the complexity of the ontology matching problem, Evolutionary Algorithm (EA) can present a good methodology for determining ontology alignments. However, there exist different aspects of a solution that are partially or wholly in conflict, and the single-objective EA may lead to unwanted bias to one of them and reduce the solution's quality. To improve the ternary compound alignment's quality when matching three biomedical ontologies, in this work, a compact Multi-Objective Evolutionary Algorithm Based On Adaptive Objective Space Decomposition (cMOEA-AOSD) based matching technique is proposed. In particular, a Ternary Compound Concept Similarity Measure (TCCSM) is proposed to calculate the similarity value of three biomedical concepts, a mathematical model for ternary compound matching problem is constructed, and a cMOEA-AOSD is presented to address it, which is able to adaptively decompose the objective space to ensure the diversity of the solutions in Pareto Front (PF) and the quality of the final solution. The experiment uses six testing cases that consists of nine biomedical ontologies to test our proposal's performance, and the experimental results show that cMOEA-AOSD significantly out performs other MOEA-based matching technique and the state-of-the-art ternary compound matching techniques.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127600780","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00068
Wei-bin Deng, XinLun Zhang, Zhe Jiang
With the popularization of smart phones, a large number of mobile applications (Apps) have been developed and become more and more important in people's lives. However, most of them need to verify user identify by their username and password to unlock its full functions, which is an inconvenient operation for users. In order to solve this problem, this paper develops an App which employs face recognition technology to check one's identify. Experiments are conducted and experimental results demonstrate that our app has good performance in terms of recognition rate and time efficiency. Meanwhile, we collect a face dataset for checking one's identify.
{"title":"A Mobile Application of Face Recognition Based on Android Platform","authors":"Wei-bin Deng, XinLun Zhang, Zhe Jiang","doi":"10.1109/CIS52066.2020.00068","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00068","url":null,"abstract":"With the popularization of smart phones, a large number of mobile applications (Apps) have been developed and become more and more important in people's lives. However, most of them need to verify user identify by their username and password to unlock its full functions, which is an inconvenient operation for users. In order to solve this problem, this paper develops an App which employs face recognition technology to check one's identify. Experiments are conducted and experimental results demonstrate that our app has good performance in terms of recognition rate and time efficiency. Meanwhile, we collect a face dataset for checking one's identify.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127285214","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00025
Han Zhang, Zhihua Song, B. Feng, Zhongliang Zhou, Fuxian Liu
Steganography has made great progress over the past few years due to the advancement of deep convolutional neural networks (DCNN), which has caused severe problems in the network security field. Ensuring the accuracy of steganalysis is becoming increasingly difficult. In this paper, we designed a two-channel generative adversarial network (TGAN), inspired by the idea of adversarial training that is based on our previous work. The TGAN consisted of three parts: The first hiding network had two input channels and one output channel. For the second extraction network, the input was a hidden image embedded with the secret image. The third detecting network had two input channels and one output channel. Experimental results on two independent image data sets showed that the proposed TGAN performed well and had better detecting capability compared to other algorithms, thus having important theoretical significance and engineering value.
{"title":"Technology of Image Steganography and Steganalysis Based on Adversarial Training","authors":"Han Zhang, Zhihua Song, B. Feng, Zhongliang Zhou, Fuxian Liu","doi":"10.1109/CIS52066.2020.00025","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00025","url":null,"abstract":"Steganography has made great progress over the past few years due to the advancement of deep convolutional neural networks (DCNN), which has caused severe problems in the network security field. Ensuring the accuracy of steganalysis is becoming increasingly difficult. In this paper, we designed a two-channel generative adversarial network (TGAN), inspired by the idea of adversarial training that is based on our previous work. The TGAN consisted of three parts: The first hiding network had two input channels and one output channel. For the second extraction network, the input was a hidden image embedded with the secret image. The third detecting network had two input channels and one output channel. Experimental results on two independent image data sets showed that the proposed TGAN performed well and had better detecting capability compared to other algorithms, thus having important theoretical significance and engineering value.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122155422","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}
Localization of structural regions in pantograph videos is the core and key to solve the problem of pantograph state detection using image processing technology. In this paper, an unsupervised learning based target localization method for pantograph video is proposed. First, an unsupervised learning approach is used to learn and estimate the pixel nodes in the same frame and neighborhood video image sequences that have undergone superpixel segmentation to achieve initialized prediction of the pantograph region and localization of the target region based on the correlation and semantic information between the pixel nodes. Secondly, we segment the target region by constructing the minimum energy function of the CRF model for the localization results. Finally, we construct pantograph video data sets in various environments. The experimental results show that the method is able to obtain accurate localization and segmentation results for different pantograph video data in different complex scenes, and has obvious superiority and robustness compared with other algorithms.
{"title":"Unsupervised learning based target localization method for pantograph video","authors":"Ruigeng Sun, Liming Li, Xingjie Chen, Ji Wang, X. Chai, Shu-bin Zheng","doi":"10.1109/CIS52066.2020.00074","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00074","url":null,"abstract":"Localization of structural regions in pantograph videos is the core and key to solve the problem of pantograph state detection using image processing technology. In this paper, an unsupervised learning based target localization method for pantograph video is proposed. First, an unsupervised learning approach is used to learn and estimate the pixel nodes in the same frame and neighborhood video image sequences that have undergone superpixel segmentation to achieve initialized prediction of the pantograph region and localization of the target region based on the correlation and semantic information between the pixel nodes. Secondly, we segment the target region by constructing the minimum energy function of the CRF model for the localization results. Finally, we construct pantograph video data sets in various environments. The experimental results show that the method is able to obtain accurate localization and segmentation results for different pantograph video data in different complex scenes, and has obvious superiority and robustness compared with other algorithms.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131813902","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00057
Hui Du, Y. Ni
In order to solve the problem of manual selection of cluster centers in density peaks clustering algorithm, an automatic selection algorithm of cluster centers was proposed in this paper, which can calculate the change rate and difference for each data. Firstly, the local density p and the high density nearest distance δ of each data point were multiplied and sorted to calculate the difference value A between two adjacent data points, where A is a group of finite sequences from big to small, and the ratio of each item in the sequence to its next term is θ. Through the threshold range of θ and A, the cluster centers can be selected adaptively, and the number of clusters can be determined automatically. Experiment results have shown that the algorithm is suitable for non-convex data with good clustering effect.
{"title":"The Improvement on Self-Adaption Select Cluster Centers Based on Fast Search and Find of Density Peaks Clustering","authors":"Hui Du, Y. Ni","doi":"10.1109/CIS52066.2020.00057","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00057","url":null,"abstract":"In order to solve the problem of manual selection of cluster centers in density peaks clustering algorithm, an automatic selection algorithm of cluster centers was proposed in this paper, which can calculate the change rate and difference for each data. Firstly, the local density p and the high density nearest distance δ of each data point were multiplied and sorted to calculate the difference value A between two adjacent data points, where A is a group of finite sequences from big to small, and the ratio of each item in the sequence to its next term is θ. Through the threshold range of θ and A, the cluster centers can be selected adaptively, and the number of clusters can be determined automatically. Experiment results have shown that the algorithm is suitable for non-convex data with good clustering effect.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130594670","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00018
Yiming Zhang
The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm is one of the most popular algorithms for solving unconstrained problems. Recently, it has been widely adopted in large-scale optimization problems. However, its use of the Hessian approximation $B_{k}$ is very likely to become ill-conditioned, resulting in an inaccurate search direction. The contracting BFGS algorithm not only retains the positive definiteness of the Hessian approximation $B_{k}$ and the quadratic termination property but also contracts the distribution of the eigenvalues of $B_{k}$ in some sense. However, the argument is not sufficient concerning the improvement of the search direction's accuracy when $B_{k}$ is ill-conditioned. In this paper, we present a modification of the contracting BFGS algorithm for unconstrained optimization. In our algorithm, instead of using a constant contracting factor $c$, we select a different $c_{k}$ in each step. Our algorithm preserves the positive definiteness of $B_{k}$ and the quadratic termination property. Moreover, by choosing a different contracting factor in each iteration, we prove the existence of the ‘best’ $c_{k}$ that minimizes the spectral condition number of $B_{k+1}$. We present a method to find such $c_{k}$ based on the matrix rank-1 perturbation theory and the eigenvalue optimization. Finally, numerical experiments are presented to verify both the convergence property and the improvement of the sensitiveness of the linear systems used to solve for search directions.
{"title":"A Modified Contracting BFGS Update for Unconstrained Optimization","authors":"Yiming Zhang","doi":"10.1109/CIS52066.2020.00018","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00018","url":null,"abstract":"The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm is one of the most popular algorithms for solving unconstrained problems. Recently, it has been widely adopted in large-scale optimization problems. However, its use of the Hessian approximation $B_{k}$ is very likely to become ill-conditioned, resulting in an inaccurate search direction. The contracting BFGS algorithm not only retains the positive definiteness of the Hessian approximation $B_{k}$ and the quadratic termination property but also contracts the distribution of the eigenvalues of $B_{k}$ in some sense. However, the argument is not sufficient concerning the improvement of the search direction's accuracy when $B_{k}$ is ill-conditioned. In this paper, we present a modification of the contracting BFGS algorithm for unconstrained optimization. In our algorithm, instead of using a constant contracting factor $c$, we select a different $c_{k}$ in each step. Our algorithm preserves the positive definiteness of $B_{k}$ and the quadratic termination property. Moreover, by choosing a different contracting factor in each iteration, we prove the existence of the ‘best’ $c_{k}$ that minimizes the spectral condition number of $B_{k+1}$. We present a method to find such $c_{k}$ based on the matrix rank-1 perturbation theory and the eigenvalue optimization. Finally, numerical experiments are presented to verify both the convergence property and the improvement of the sensitiveness of the linear systems used to solve for search directions.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134574101","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 : 2020-11-01DOI: 10.1109/cis52066.2020.00002
{"title":"[Title page iii]","authors":"","doi":"10.1109/cis52066.2020.00002","DOIUrl":"https://doi.org/10.1109/cis52066.2020.00002","url":null,"abstract":"","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116321125","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00071
Jun Yang, Siyuan Jing, Liping Jia
Faced with various security threats from the Internet, remote voting system usually uses cryptography to identify voters, create ballots, cast ballots and count ballots. It not only makes the voting system more complex, but also increases the computational cost. This paper proposes a remote voting scheme based on Three-Ballot which provides voters to cast their ballots to the voting machine and the Storage and Auditor. The scheme reduces the complexity of voters, while keeping the security of voters when casting as they intend. Furthermore, the scheme ensures individual verifiability for voters to check whether the ballots are counted as they cast, and universal verifiability for the public to check whether the voting system has correctly counted all the ballots.
{"title":"RVBT: A Remote Voting Scheme Based on Three-Ballot","authors":"Jun Yang, Siyuan Jing, Liping Jia","doi":"10.1109/CIS52066.2020.00071","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00071","url":null,"abstract":"Faced with various security threats from the Internet, remote voting system usually uses cryptography to identify voters, create ballots, cast ballots and count ballots. It not only makes the voting system more complex, but also increases the computational cost. This paper proposes a remote voting scheme based on Three-Ballot which provides voters to cast their ballots to the voting machine and the Storage and Auditor. The scheme reduces the complexity of voters, while keeping the security of voters when casting as they intend. Furthermore, the scheme ensures individual verifiability for voters to check whether the ballots are counted as they cast, and universal verifiability for the public to check whether the voting system has correctly counted all the ballots.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133848698","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 : 2020-11-01DOI: 10.1109/CIS52066.2020.00080
Wei Ding, Ji Han, Dijin Wang
Seismic waveform data acquired by various seismic monitoring instruments are the base of understanding the mechanism of seismic research and disaster reduction. How to extract data and eliminate noise from a mass of valuable seismic data has become a hot issue in seismic research. A method based on convolutional neural network is proposed to solve the problem of seismic electromagnetic signal recognition, which employed a set of larger than Ms3.6 seismic event data recorded by electromagnetic instrument in Sichuan-Yunnan region. The electromagnetic signal is first visualized into a two-dimensional picture using short-time Fourier transform (STFT), so the problem of electromagnetic signal recognition is transformed into the object detection problem in the field of image recognition. A convolutional neural network method was used to train and test dataset from 1117 earthquake events. The training and detection accuracy rate of the dataset of 164 stations has reached 90%. The experiments show that this algorithm can deal with the problem of electromagnetic signal recognition and classify small sample size waveform data effectively.
{"title":"The seismic electromagnet signal recognition using convolutional neural network","authors":"Wei Ding, Ji Han, Dijin Wang","doi":"10.1109/CIS52066.2020.00080","DOIUrl":"https://doi.org/10.1109/CIS52066.2020.00080","url":null,"abstract":"Seismic waveform data acquired by various seismic monitoring instruments are the base of understanding the mechanism of seismic research and disaster reduction. How to extract data and eliminate noise from a mass of valuable seismic data has become a hot issue in seismic research. A method based on convolutional neural network is proposed to solve the problem of seismic electromagnetic signal recognition, which employed a set of larger than Ms3.6 seismic event data recorded by electromagnetic instrument in Sichuan-Yunnan region. The electromagnetic signal is first visualized into a two-dimensional picture using short-time Fourier transform (STFT), so the problem of electromagnetic signal recognition is transformed into the object detection problem in the field of image recognition. A convolutional neural network method was used to train and test dataset from 1117 earthquake events. The training and detection accuracy rate of the dataset of 164 stations has reached 90%. The experiments show that this algorithm can deal with the problem of electromagnetic signal recognition and classify small sample size waveform data effectively.","PeriodicalId":106959,"journal":{"name":"2020 16th International Conference on Computational Intelligence and Security (CIS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127194023","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}