Pub Date : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00059
Fengrong Lv, Zeyu Dong, T. Wan, Kaili Jiang, Xueli Fang, Lei Zhang
The ambiguity function describes the joint characteristics of the signal in the time domain and the frequency domain, and is usually used for signal recognition. The time domain and frequency domain distribution of noise is more spurious than the signal. Its ambiguity function can also reflect its characteristics. However, the direct use of the ambiguity function as the signal feature will increase the amount of calculation of the detection system, so slice it, which reduces the burden on the system for detecting signal information while preserving the information. At the same time, the radar signal detection based on neural network can avoid the influence of artificially setting the threshold on the detection performance of the system. So, we propose an algorithm for inputting the extracted fuzzy function slice sequence into a neural network to realize signal detection. Simulation proves that when the signal-to-noise ratio of the received signal is very low, our proposed method still has a high detection probability.
{"title":"Radar Signal Detection Based on Ambiguity Function and Deep Learning","authors":"Fengrong Lv, Zeyu Dong, T. Wan, Kaili Jiang, Xueli Fang, Lei Zhang","doi":"10.1109/ICCEA53728.2021.00059","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00059","url":null,"abstract":"The ambiguity function describes the joint characteristics of the signal in the time domain and the frequency domain, and is usually used for signal recognition. The time domain and frequency domain distribution of noise is more spurious than the signal. Its ambiguity function can also reflect its characteristics. However, the direct use of the ambiguity function as the signal feature will increase the amount of calculation of the detection system, so slice it, which reduces the burden on the system for detecting signal information while preserving the information. At the same time, the radar signal detection based on neural network can avoid the influence of artificially setting the threshold on the detection performance of the system. So, we propose an algorithm for inputting the extracted fuzzy function slice sequence into a neural network to realize signal detection. Simulation proves that when the signal-to-noise ratio of the received signal is very low, our proposed method still has a high detection probability.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129978406","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00032
Yin Yue, Zhou Jun
Traditional restoration algorithms are not effective enough to fully repair rock paintings. The barley lithology is used as an example in this work to demonstrate a novel repair algorithm based on saliency detection and total variation (TV). The damage detection area of the rock painting image is extracted by saliency detection, then the improved TV model is applied to complete the repair. Experiments show that the proposed algorithm can effectively improve the poor robustness and virtual boundary of the traditional TV model image restoration algorithm, and make the image smoother and achieve better repair results. At the same time, a set of solutions for the extraction and restoration of structural rock paintings has been formed.
{"title":"Rock painting restoration method based on saliency detection and TV model","authors":"Yin Yue, Zhou Jun","doi":"10.1109/ICCEA53728.2021.00032","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00032","url":null,"abstract":"Traditional restoration algorithms are not effective enough to fully repair rock paintings. The barley lithology is used as an example in this work to demonstrate a novel repair algorithm based on saliency detection and total variation (TV). The damage detection area of the rock painting image is extracted by saliency detection, then the improved TV model is applied to complete the repair. Experiments show that the proposed algorithm can effectively improve the poor robustness and virtual boundary of the traditional TV model image restoration algorithm, and make the image smoother and achieve better repair results. At the same time, a set of solutions for the extraction and restoration of structural rock paintings has been formed.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128774797","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00056
Z. Jiawen, Cao Xiaomei, Wang Shaohui
Feature extraction is a critical step when using machine learning. High-quality features help improve model performance. Feature extraction methods are still not perfect in current webshell detection schemes based on machine learning, so there is still room for improvement in webshell detection accuracy. Aimed at solving the imperfect feature extraction problem, a webshell attack detection method based on multi-dimensional dynamic features is proposed. In addition to source code of a PHP file, its opcode sequence is analyzed and processed to reduce the impact of obfuscation and encryption technology on attack detection. To improve webshell detection accuracy, TextRank algorithm of natural language processing is introduced to extract relevant features, providing optimized data for subsequent training machine learning algorithm. Simulation experiment results show that the proposed method can accurately distinguish between normal pages and webshell attacks, and the accuracy rate is as high as 99.574%.
{"title":"Detection approach of webshell attacks based on multi-dimensional dynamic features","authors":"Z. Jiawen, Cao Xiaomei, Wang Shaohui","doi":"10.1109/ICCEA53728.2021.00056","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00056","url":null,"abstract":"Feature extraction is a critical step when using machine learning. High-quality features help improve model performance. Feature extraction methods are still not perfect in current webshell detection schemes based on machine learning, so there is still room for improvement in webshell detection accuracy. Aimed at solving the imperfect feature extraction problem, a webshell attack detection method based on multi-dimensional dynamic features is proposed. In addition to source code of a PHP file, its opcode sequence is analyzed and processed to reduce the impact of obfuscation and encryption technology on attack detection. To improve webshell detection accuracy, TextRank algorithm of natural language processing is introduced to extract relevant features, providing optimized data for subsequent training machine learning algorithm. Simulation experiment results show that the proposed method can accurately distinguish between normal pages and webshell attacks, and the accuracy rate is as high as 99.574%.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122134935","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00022
Wenjuan Cheng, Siyi Chen
More and more studies are using sophisticated textual sentiment models to help us better understand behavioral finance patterns across financial market participants. Most of the existing sentiment analysis methods are oriented to general fields. Most text representation extraction methods use fixed token encoders. Generally, sentiment analysis models are invalid for financial applications. To overcome these challenges, a text sentiment analysis model (BBiLSTM-Attention) for financial fields is proposed. The model uses the pre-training language model FinBERT as a feature extractor to dynamically obtain the context information of comments, and combinate BiLSTM and multiple attention mechanisms to extract the sentiment of financial comments. Experiments is performed using financial field commentary dataset. The results show improved accuracy and generalization ability, accuracy reached 79.33%, and F1-score reached 0.8068.
{"title":"Sentiment Analysis of Financial Texts Based on Attention Mechanism of FinBERT and BiLSTM","authors":"Wenjuan Cheng, Siyi Chen","doi":"10.1109/ICCEA53728.2021.00022","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00022","url":null,"abstract":"More and more studies are using sophisticated textual sentiment models to help us better understand behavioral finance patterns across financial market participants. Most of the existing sentiment analysis methods are oriented to general fields. Most text representation extraction methods use fixed token encoders. Generally, sentiment analysis models are invalid for financial applications. To overcome these challenges, a text sentiment analysis model (BBiLSTM-Attention) for financial fields is proposed. The model uses the pre-training language model FinBERT as a feature extractor to dynamically obtain the context information of comments, and combinate BiLSTM and multiple attention mechanisms to extract the sentiment of financial comments. Experiments is performed using financial field commentary dataset. The results show improved accuracy and generalization ability, accuracy reached 79.33%, and F1-score reached 0.8068.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"433 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122802013","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00015
Guiping Sun
Image hashing is an essential technology for the community of multimedia security, and has been extensively applied in image authentication and content protection. This paper designs a new image hashing scheme with CVA and texture structure descriptor. A valuable contribution is the color DWT (Discrete Wavelet Transform) representation by incorporating CVA (color vector angle) matrix into LL sub-band. Since CVA can well retain the color information, desirable discrimination of our hashing is achieved. In addition, as LBP (local binary pattern) is texture structure descriptor, feature extraction with LBP from the color DWT representation ensures good classification performance between robustness and discrimination. The effectiveness of our hashing scheme is validated by various experiments with open databases. Experimental results demonstrate that the proposed image hashing scheme is superior to the state-of-the-art schemes in terms of classification performance between perceptual robustness and discriminative capability.
{"title":"Robustness and Discrimination Oriented Image Hashing with CVA and Texture Structure Descriptor","authors":"Guiping Sun","doi":"10.1109/ICCEA53728.2021.00015","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00015","url":null,"abstract":"Image hashing is an essential technology for the community of multimedia security, and has been extensively applied in image authentication and content protection. This paper designs a new image hashing scheme with CVA and texture structure descriptor. A valuable contribution is the color DWT (Discrete Wavelet Transform) representation by incorporating CVA (color vector angle) matrix into LL sub-band. Since CVA can well retain the color information, desirable discrimination of our hashing is achieved. In addition, as LBP (local binary pattern) is texture structure descriptor, feature extraction with LBP from the color DWT representation ensures good classification performance between robustness and discrimination. The effectiveness of our hashing scheme is validated by various experiments with open databases. Experimental results demonstrate that the proposed image hashing scheme is superior to the state-of-the-art schemes in terms of classification performance between perceptual robustness and discriminative capability.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122845855","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}
On the premise of introducing the knowledge graph and its application in the daily scenarios of power grids, and building and displaying the knowledge graph for specific scenarios, this study intuitively shows the correlation of each link in each scenario and discovers the corresponding problems. Reasonable suggestions are accordingly made to improve the daily work patterns, enhance the service quality of each link, and boost the refinement of the management of daily affairs. Through the application of knowledge graph in the typical daily scenarios in power marketing, this paper aims to form a panoramic knowledge graph of the power grids on the basis of smart brain among all businesses and organizations in the near future to promote the intelligent operation of the power grids [1].
{"title":"Research on the Typical Application of Knowledge Graph in Power Marketing","authors":"W. Meng, Dongning Zhang, Tengxuan Guo, Zhenguo Zong, Yijuan Liu, Yanmei Wang, Jing Li, Weiyi Zhu","doi":"10.1109/ICCEA53728.2021.00069","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00069","url":null,"abstract":"On the premise of introducing the knowledge graph and its application in the daily scenarios of power grids, and building and displaying the knowledge graph for specific scenarios, this study intuitively shows the correlation of each link in each scenario and discovers the corresponding problems. Reasonable suggestions are accordingly made to improve the daily work patterns, enhance the service quality of each link, and boost the refinement of the management of daily affairs. Through the application of knowledge graph in the typical daily scenarios in power marketing, this paper aims to form a panoramic knowledge graph of the power grids on the basis of smart brain among all businesses and organizations in the near future to promote the intelligent operation of the power grids [1].","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124548401","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00031
Gao Xirong, Li Dan
The research believes that there may be some mutual conversion and interaction mechanism between entity and virtual mirroring. The mathematical structure of virtual space and real space were described by matrix algebra tools. Then, the physical and cyberspace matrix corresponding to virtual-real space were set. After the feature information of entity was classified from the perspective of technical realization, a path of deconstruction and reconstruction of entity to mirror conversion was designed to prove the process of virtual-real fusion. Influence factor were considered, sensitivity and conversion noise, then the dual mapping model for virtual-real fusion is constructed. The case of Internet of vehicle for the mapping was demonstrated. The results above had enriched the theory of the virtual-real fusion of physical space and cyberspace.
{"title":"A Dual Mapping Model for Virtual-real Fusion in 5G","authors":"Gao Xirong, Li Dan","doi":"10.1109/ICCEA53728.2021.00031","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00031","url":null,"abstract":"The research believes that there may be some mutual conversion and interaction mechanism between entity and virtual mirroring. The mathematical structure of virtual space and real space were described by matrix algebra tools. Then, the physical and cyberspace matrix corresponding to virtual-real space were set. After the feature information of entity was classified from the perspective of technical realization, a path of deconstruction and reconstruction of entity to mirror conversion was designed to prove the process of virtual-real fusion. Influence factor were considered, sensitivity and conversion noise, then the dual mapping model for virtual-real fusion is constructed. The case of Internet of vehicle for the mapping was demonstrated. The results above had enriched the theory of the virtual-real fusion of physical space and cyberspace.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130456045","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}
In the complex electromagnetic environment of modern electronic warfare, active deception jamming poses a serious threat to radar. This paper proposes the identification of active deception jamming based on residual neural network (RESNET), and studies the bispectral characteristics of the signal and the method of identifying the active deception jamming of RESNET. This algorithm firstly extracts the bispectral features of the jamming signal, and uses the bispectral diagonal slice as the input of RSNET to realize the recognition of active deception jamming. Simulation results show that the algorithm has high recognition accuracy, not only has good anti-noise performance, but also has excellent robustness under low JSR.
{"title":"Research on Radar Active Deception Jamming Identification Method Based on RESNET and Bispectrum Features","authors":"Kunteng Wang, Zeyu Dong, T. Wan, Kaili Jiang, Wanan Xiong, Xueli Fang","doi":"10.1109/ICCEA53728.2021.00102","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00102","url":null,"abstract":"In the complex electromagnetic environment of modern electronic warfare, active deception jamming poses a serious threat to radar. This paper proposes the identification of active deception jamming based on residual neural network (RESNET), and studies the bispectral characteristics of the signal and the method of identifying the active deception jamming of RESNET. This algorithm firstly extracts the bispectral features of the jamming signal, and uses the bispectral diagonal slice as the input of RSNET to realize the recognition of active deception jamming. Simulation results show that the algorithm has high recognition accuracy, not only has good anti-noise performance, but also has excellent robustness under low JSR.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134382628","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00013
Jun He, Jun Yang
In recent years, the development of deep reinforcement learning has become a new research frontier in the field of artificial intelligence. In the complex and changeable battlefield environment of military confrontation, how to use reinforcement learning methods to assist in formulating a set of efficient military game strategies has become a new research direction. The military game confrontation environment usually uses static reward settings ignoring the importance of captured target in the tactical arrangement. This paper starts from the use of episodic memory Q-network model to train drone swarm confrontation and uses expert knowledge to design a local state target threat assessment method. By analyzing the threat indicator of killed target, it provides drones with dynamic rewards to help drones comb local situation. Aiming at the problem that the static weight of episodic memory is very inefficient by manual adjustment in complex military confrontation environment, we use dynamic weight adjustment strategy inspired by multitask learning. Through the improvement of above method, we establish a dynamic gain drones confrontation model based on episodic memory with methods of dynamic reward mechanism and dynamic weight adjustment. The effectiveness of model is verified by machine-to-machine confrontation which provides thoughts for the analysis of military game decision-making in the complex battlefield environment.
{"title":"Dynamic Gain Military Game Algorithm Based on Episodic Memory","authors":"Jun He, Jun Yang","doi":"10.1109/ICCEA53728.2021.00013","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00013","url":null,"abstract":"In recent years, the development of deep reinforcement learning has become a new research frontier in the field of artificial intelligence. In the complex and changeable battlefield environment of military confrontation, how to use reinforcement learning methods to assist in formulating a set of efficient military game strategies has become a new research direction. The military game confrontation environment usually uses static reward settings ignoring the importance of captured target in the tactical arrangement. This paper starts from the use of episodic memory Q-network model to train drone swarm confrontation and uses expert knowledge to design a local state target threat assessment method. By analyzing the threat indicator of killed target, it provides drones with dynamic rewards to help drones comb local situation. Aiming at the problem that the static weight of episodic memory is very inefficient by manual adjustment in complex military confrontation environment, we use dynamic weight adjustment strategy inspired by multitask learning. Through the improvement of above method, we establish a dynamic gain drones confrontation model based on episodic memory with methods of dynamic reward mechanism and dynamic weight adjustment. The effectiveness of model is verified by machine-to-machine confrontation which provides thoughts for the analysis of military game decision-making in the complex battlefield environment.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"162 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133288289","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 : 2021-06-01DOI: 10.1109/ICCEA53728.2021.00088
Wei Zhang, Dahai Jin, Zhiqing Shi, Xiangxuan Tian
With the rapid development of information technology, the status and role of software are more and more prominent. At the same time, due to the increasing scale and complexity, software problems are also increasing. Therefore, the importance of software support is also increasing. This paper briefly introduces the concept and classification of software support, puts forward the process of pre-deployment software support and post-deployment software support, and then discusses the principles and agency of software support.
{"title":"Research on Software Support Process","authors":"Wei Zhang, Dahai Jin, Zhiqing Shi, Xiangxuan Tian","doi":"10.1109/ICCEA53728.2021.00088","DOIUrl":"https://doi.org/10.1109/ICCEA53728.2021.00088","url":null,"abstract":"With the rapid development of information technology, the status and role of software are more and more prominent. At the same time, due to the increasing scale and complexity, software problems are also increasing. Therefore, the importance of software support is also increasing. This paper briefly introduces the concept and classification of software support, puts forward the process of pre-deployment software support and post-deployment software support, and then discusses the principles and agency of software support.","PeriodicalId":325790,"journal":{"name":"2021 International Conference on Computer Engineering and Application (ICCEA)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131294527","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：