Pub Date : 2019-03-15DOI: 10.5220/0008098200840091
Siqi Huang, Bo Yan, Dongmei Zhang
DGA (domain generation algorithms) domain names are a class of domain names generated by specific algorithms which are used to communicate with malicious C&C servers. DGA based on the PCFG model has been proposed Lately. Under the test of existing DGA detection techniques, its anti-detection effect is very obvious. One of the reasons is that these domains are generated by legal domain names and have the same statistical characteristics of legitimate domain names. This paper proposes a detection model which combines deep learning and Multi-head attention mechanism. It employs these two techniques to extract the features of the domain names. Experiment results show that the model has a good effect on detecting domain names based on PCFG model.
{"title":"A Detection Algorithm of Malicious Domain Based on Deep Learning and Multi-Head Attention Mechanism","authors":"Siqi Huang, Bo Yan, Dongmei Zhang","doi":"10.5220/0008098200840091","DOIUrl":"https://doi.org/10.5220/0008098200840091","url":null,"abstract":"DGA (domain generation algorithms) domain names are a class of domain names generated by specific algorithms which are used to communicate with malicious C&C servers. DGA based on the PCFG model has been proposed Lately. Under the test of existing DGA detection techniques, its anti-detection effect is very obvious. One of the reasons is that these domains are generated by legal domain names and have the same statistical characteristics of legitimate domain names. This paper proposes a detection model which combines deep learning and Multi-head attention mechanism. It employs these two techniques to extract the features of the domain names. Experiment results show that the model has a good effect on detecting domain names based on PCFG model.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132054653","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 : 2019-03-15DOI: 10.5220/0008098602130221
Lin Guo, Bin Wu
Evaluation of Attack Effect in Ad Hoc networks is one of the key technologies for Ad Hoc network security applications. In order to solve the traditional attack effect evaluation with the constant weighted summation can’t adjust the relevant weights in time to the change of the attack situation, which causes the limitation and one-sidedness of the judgment. This paper proposes an attack effect evaluation model based on variable weight theory. First, comprehensively considering the relevance of the attack's own complexity and the attack effect, establish an attack effect evaluation indicator system. Then, construct a state variable weight vector expression, so that the weights are adjusted accordingly with the change of the situation. Finally, combined with TOPSIS method, the attack effect evaluation model based on variable weight TOPSIS is constructed. The experimental simulations show that the evaluation results obtained by the model are scientific and reasonable, which makes up for the deviation caused by the decision of the constant weight, and provides a theory and method for the evaluation of the attack effect in Ad Hoc network.
Ad Hoc网络攻击效果评估是Ad Hoc网络安全应用的关键技术之一。为了解决传统的常加权求和攻击效果评价不能根据攻击情况的变化及时调整相关权重，造成判断的局限性和片面性的问题。提出了一种基于变权理论的攻击效果评价模型。首先，综合考虑攻击本身的复杂性与攻击效果的相关性，建立攻击效果评价指标体系。然后，构造状态变量权向量表达式，使权值随情况的变化而相应调整。最后，结合TOPSIS方法，构建了基于变权TOPSIS的攻击效果评价模型。实验仿真结果表明，该模型得到的评价结果科学合理，弥补了定权决策带来的偏差，为Ad Hoc网络攻击效果的评价提供了理论和方法。
{"title":"Evaluation of Attack Effect in Ad Hoc Networks Based on Variable Weight TOPSIS Method","authors":"Lin Guo, Bin Wu","doi":"10.5220/0008098602130221","DOIUrl":"https://doi.org/10.5220/0008098602130221","url":null,"abstract":"Evaluation of Attack Effect in Ad Hoc networks is one of the key technologies for Ad Hoc network security applications. In order to solve the traditional attack effect evaluation with the constant weighted summation can’t adjust the relevant weights in time to the change of the attack situation, which causes the limitation and one-sidedness of the judgment. This paper proposes an attack effect evaluation model based on variable weight theory. First, comprehensively considering the relevance of the attack's own complexity and the attack effect, establish an attack effect evaluation indicator system. Then, construct a state variable weight vector expression, so that the weights are adjusted accordingly with the change of the situation. Finally, combined with TOPSIS method, the attack effect evaluation model based on variable weight TOPSIS is constructed. The experimental simulations show that the evaluation results obtained by the model are scientific and reasonable, which makes up for the deviation caused by the decision of the constant weight, and provides a theory and method for the evaluation of the attack effect in Ad Hoc network.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126808720","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 : 2019-03-15DOI: 10.5220/0008098000440048
Fuping Yang, Zhichun Yuan, Xi Cheng
The text-to-scene conversion is such a process that converts the input text to 3D scenes automatically based on the natural language processing. The scene division is one of the basic research contents of the text-to-scene conversion, and it mainly identifies and divides the number and structure of scenes in multi-scene text. In this paper, the general expression of the scene is obtained through the analysis of scene features, and then the method for divide multi-scene text which to use scene themes has been proposed. In the experiment, the LDA model is used to divide the multi-scene text, and this method is applied to the standard data set. The experiments show that the expected results are achieved.
{"title":"A Scene Division Method Based on Theme","authors":"Fuping Yang, Zhichun Yuan, Xi Cheng","doi":"10.5220/0008098000440048","DOIUrl":"https://doi.org/10.5220/0008098000440048","url":null,"abstract":"The text-to-scene conversion is such a process that converts the input text to 3D scenes automatically based on the natural language processing. The scene division is one of the basic research contents of the text-to-scene conversion, and it mainly identifies and divides the number and structure of scenes in multi-scene text. In this paper, the general expression of the scene is obtained through the analysis of scene features, and then the method for divide multi-scene text which to use scene themes has been proposed. In the experiment, the LDA model is used to divide the multi-scene text, and this method is applied to the standard data set. The experiments show that the expected results are achieved.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125178719","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 : 2019-03-15DOI: 10.5220/0008095900590064
Junchang Zhang, Chenyang Xia, Jinjin Wan
For most algorithms, the problem of Tracking target performance degradation in the case of fast moving, illumination changes, target deformation, occlusion, out-of-plane rotation, low-resolution images, etc. This paper proposes a tracking verification algorithm based on channel reliability. The tracker part of the algorithm is tracked by the method of correlation filter based on channel reliability. By calculating the reliability weight of each feature channel of the input correlation filter, and multiplying the weight by the response of the corresponding channel to obtain the final response, so that the target positioning will be more accurate. The validator part uses the Siamese dual input network in the deep learning convolutional neural network. Every few frames, the verifier will verify the results of the tracker part of the algorithm. If the reliability is verified, the tracking result will not be modified. Otherwise, the validator will re-detect the target location and verify the reliability through the Siamese dual-input network. The tracker will regard this location as the new location of our target continues to be tracked, making target tracking more durable and robust. The experimental evaluation of the OTB13 video sequence proves that the proposed algorithm has good adaptability to target fast motion, illumination change, target deformation, occlusion, and out-of-plane rotation, and has good robustness.
{"title":"Tracking Verification Algorithm Based on Channel Reliability","authors":"Junchang Zhang, Chenyang Xia, Jinjin Wan","doi":"10.5220/0008095900590064","DOIUrl":"https://doi.org/10.5220/0008095900590064","url":null,"abstract":"For most algorithms, the problem of Tracking target performance degradation in the case of fast moving, illumination changes, target deformation, occlusion, out-of-plane rotation, low-resolution images, etc. This paper proposes a tracking verification algorithm based on channel reliability. The tracker part of the algorithm is tracked by the method of correlation filter based on channel reliability. By calculating the reliability weight of each feature channel of the input correlation filter, and multiplying the weight by the response of the corresponding channel to obtain the final response, so that the target positioning will be more accurate. The validator part uses the Siamese dual input network in the deep learning convolutional neural network. Every few frames, the verifier will verify the results of the tracker part of the algorithm. If the reliability is verified, the tracking result will not be modified. Otherwise, the validator will re-detect the target location and verify the reliability through the Siamese dual-input network. The tracker will regard this location as the new location of our target continues to be tracked, making target tracking more durable and robust. The experimental evaluation of the OTB13 video sequence proves that the proposed algorithm has good adaptability to target fast motion, illumination change, target deformation, occlusion, and out-of-plane rotation, and has good robustness.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132610598","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 : 2019-03-15DOI: 10.5220/0008098701480152
Runqi Li, Cen Chen, Mengyao Chen
In order to improve the efficiency of urban traffic operation, this paper combines deep learning technology and UAV photography of drones. With the algorithm in this paper, we can classify traffic area and static area in high quality and speed. We make a test in Fengtai District of Beijing to conduct regional traffic identification research on the traffic content of the area. The primary identification area includes the vehicle travel area and coordinates the pedestrian area in a coordinated manner. The main research result of this algorithm is to propose a key frame extraction scheme for UAV image and then combine it with the application of Mask R-CNN in high-altitude image to identify the ground area. The experimental results are similar to the same algorithm (refer to FCNs for this article). Comparative benchmarks) have obvious advantages of high speed and high accuracy, which are of great help to traffic safety and urban planning.
{"title":"Research on UAV Image Classification Based on Deep Learning","authors":"Runqi Li, Cen Chen, Mengyao Chen","doi":"10.5220/0008098701480152","DOIUrl":"https://doi.org/10.5220/0008098701480152","url":null,"abstract":"In order to improve the efficiency of urban traffic operation, this paper combines deep learning technology and UAV photography of drones. With the algorithm in this paper, we can classify traffic area and static area in high quality and speed. We make a test in Fengtai District of Beijing to conduct regional traffic identification research on the traffic content of the area. The primary identification area includes the vehicle travel area and coordinates the pedestrian area in a coordinated manner. The main research result of this algorithm is to propose a key frame extraction scheme for UAV image and then combine it with the application of Mask R-CNN in high-altitude image to identify the ground area. The experimental results are similar to the same algorithm (refer to FCNs for this article). Comparative benchmarks) have obvious advantages of high speed and high accuracy, which are of great help to traffic safety and urban planning.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131865240","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 : 2019-03-15DOI: 10.5220/0008095800050009
Li Leyang, Guo Shuaichen, Hao Peifeng
Environmental protection has become an important issue in modern enterprises, usually involving harmful liquid and harmful gas to be detected. In order to address this issue, we propose a new intelligent algorithm to identify and estimate the emission based on the color and smoke. To be specific, the three-frame difference method is used to achieve the background segregation, and the area of the smoke is separated after the images are proceeded by a mask. The color dictionary is employed to determine what color it is. Then, the background separation algorithm and the physics knowledge are used to establish an estimation model. In the experiment, we use the proposed approach to estimate harmful gas emissions in the real-world production process of enterprises. Results validate the effectiveness and efficiency of the proposed algorithm, in terms of the prediction accuracy.
{"title":"Intelligent Identification and Emissions Estimation of Harmful Gas Based on Background Segregation Algorithm","authors":"Li Leyang, Guo Shuaichen, Hao Peifeng","doi":"10.5220/0008095800050009","DOIUrl":"https://doi.org/10.5220/0008095800050009","url":null,"abstract":"Environmental protection has become an important issue in modern enterprises, usually involving harmful liquid and harmful gas to be detected. In order to address this issue, we propose a new intelligent algorithm to identify and estimate the emission based on the color and smoke. To be specific, the three-frame difference method is used to achieve the background segregation, and the area of the smoke is separated after the images are proceeded by a mask. The color dictionary is employed to determine what color it is. Then, the background separation algorithm and the physics knowledge are used to establish an estimation model. In the experiment, we use the proposed approach to estimate harmful gas emissions in the real-world production process of enterprises. Results validate the effectiveness and efficiency of the proposed algorithm, in terms of the prediction accuracy.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121661601","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 : 2019-03-15DOI: 10.5220/0008098102030212
Yan Li, Bin Wu
Traditional honeynets cannot dynamically migrate traffic. The flexibility of SDN can solve this problem. At the same time, the traditional honeynets have the disadvantages of complicated alarm logs and inability to carry out targeted analysis, and lacks protection for the honeypot. It is easy to completely destroy the honeypot and make it a jumper for the attacker to launch the next attack on the intranet. This paper proposes a modular honeynet system based on SDN, which can respond to the scanning probe-exploit-worm injected attack chain, reducing the complexity of the alarm log and improving the efficiency of the researchers in analyzing attacks. Also, a honeypot switching strategy based on the detection of the attack tree phase is proposed in the module of vulnerability response, which can delay the attacker's attack progress and reduces the risk of the honeypot. The experiment also verified the feasibility of the modular system.
{"title":"Design and Implementation of Modular Honeynet System Based on SDN","authors":"Yan Li, Bin Wu","doi":"10.5220/0008098102030212","DOIUrl":"https://doi.org/10.5220/0008098102030212","url":null,"abstract":"Traditional honeynets cannot dynamically migrate traffic. The flexibility of SDN can solve this problem. At the same time, the traditional honeynets have the disadvantages of complicated alarm logs and inability to carry out targeted analysis, and lacks protection for the honeypot. It is easy to completely destroy the honeypot and make it a jumper for the attacker to launch the next attack on the intranet. This paper proposes a modular honeynet system based on SDN, which can respond to the scanning probe-exploit-worm injected attack chain, reducing the complexity of the alarm log and improving the efficiency of the researchers in analyzing attacks. Also, a honeypot switching strategy based on the detection of the attack tree phase is proposed in the module of vulnerability response, which can delay the attacker's attack progress and reduces the risk of the honeypot. The experiment also verified the feasibility of the modular system.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"312 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123678357","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 : 2019-03-15DOI: 10.5220/0008098401420147
Kun Mo, Jian Li
With the development of the network in recent years, cyber security has become one of the most challenging aspects of modern society. Machine learning is one of extensively used techniques in Intrusion Detection System, which has achieved comparable performance. To extract more important features, this paper proposes an efficient model based Auto-Encoder and LightGBM to classify network traffic. KDD99 dataset [1], as the benchmark dataset, is used for computing the performance and analyse the metrics of the method. Based on Auto-Encoder, we extract more important features, and then mix them with existing features to improve the effectiveness of the LightGBM [2] model. The experimental results show that the proposed algorithm produces the best performance in terms of overall accuracy.
{"title":"A Deep Auto-Encoder based LightGBM Approach for Network Intrusion Detection System","authors":"Kun Mo, Jian Li","doi":"10.5220/0008098401420147","DOIUrl":"https://doi.org/10.5220/0008098401420147","url":null,"abstract":"With the development of the network in recent years, cyber security has become one of the most challenging aspects of modern society. Machine learning is one of extensively used techniques in Intrusion Detection System, which has achieved comparable performance. To extract more important features, this paper proposes an efficient model based Auto-Encoder and LightGBM to classify network traffic. KDD99 dataset [1], as the benchmark dataset, is used for computing the performance and analyse the metrics of the method. Based on Auto-Encoder, we extract more important features, and then mix them with existing features to improve the effectiveness of the LightGBM [2] model. The experimental results show that the proposed algorithm produces the best performance in terms of overall accuracy.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"215 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116141902","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 : 1900-01-01DOI: 10.5220/0008096101030110
Runchu Xu
{"title":"Total Variation and Tight Frame Iteration Regularization Based Image Reconstruction for Short-scanning Few-view Computer Tomography","authors":"Runchu Xu","doi":"10.5220/0008096101030110","DOIUrl":"https://doi.org/10.5220/0008096101030110","url":null,"abstract":"","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124608598","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}
This paper proposes a new three-point method to locate the spatial sphere center from a single image. In monocular vision system with known intrinsic parameters, the traditional methods of locating the center of a spatial sphere with known radius require fitting its image points to an ellipse from which the sphere center is extracted. The ellipse fitting procedure requires at least five image points whereas the projection ellipse of a sphere essentially is a three-degree-of-freedom problem, which implies that over-parametrization is introduced in ellipse fitting. In this paper, the ellipse is represented with the three coordinates of the sphere center, and then at least three image points on the ellipse are used to construct a set of quadratic equations of the coordinates from which the Gröbner basis method is used to solve for the coordinates. The experimental results show that the three-point method can solve the problem with less than five image points, and when the number of image points increases to five or more, the new method can also improve sphere localization accuracy and have improved robustness.
{"title":"Sphere Localization from a Minimal Number of Points in a Single Image","authors":"Kunfeng Shi, Xuebin Li, Huikun Xu, Hongmei Zhao, Huanlong Zhang","doi":"10.5220/0008096300650070","DOIUrl":"https://doi.org/10.5220/0008096300650070","url":null,"abstract":"This paper proposes a new three-point method to locate the spatial sphere center from a single image. In monocular vision system with known intrinsic parameters, the traditional methods of locating the center of a spatial sphere with known radius require fitting its image points to an ellipse from which the sphere center is extracted. The ellipse fitting procedure requires at least five image points whereas the projection ellipse of a sphere essentially is a three-degree-of-freedom problem, which implies that over-parametrization is introduced in ellipse fitting. In this paper, the ellipse is represented with the three coordinates of the sphere center, and then at least three image points on the ellipse are used to construct a set of quadratic equations of the coordinates from which the Gröbner basis method is used to solve for the coordinates. The experimental results show that the three-point method can solve the problem with less than five image points, and when the number of image points increases to five or more, the new method can also improve sphere localization accuracy and have improved robustness.","PeriodicalId":415288,"journal":{"name":"Proceedings of the International Conference on Advances in Computer Technology, Information Science and Communications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116915343","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：