2021 International Conference on Communications, Information System and Computer Engineering (CISCE)最新文献

筛选
英文 中文
An adversarial sample defense model based on computer attention mechanism 基于计算机注意机制的对抗性样本防御模型
Yuwei Han, Qing Wang
{"title":"An adversarial sample defense model based on computer attention mechanism","authors":"Yuwei Han, Qing Wang","doi":"10.1109/CISCE52179.2021.9446015","DOIUrl":"https://doi.org/10.1109/CISCE52179.2021.9446015","url":null,"abstract":"The traditional neural network can not realize the invariance of image rotation and distortion well, so the attacker can cheat the neural network by adding tiny disturbance on the image. If signposts are attacked, automatic driving will be misguided, leading to disastrous consequences. This paper proposes a new adversarial sample defense model based on attention mechanism, which used LSTM network to get the key features. Firstly, we describe our model which recognizes an object through enough main characteristics. It can discard large parts of the noise distributed throughout the image area. Secondly, we show recognition process of the model, receiving adversarial samples and then outputting the salient mask image which can be classified correctly by neural networks in deep learning. At last, we designed experiments to test model effect on the traffic sign and the general picture. The experimental results show that our defense model can strongly defend adversarial examples.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115267616","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}
引用次数: 0
The role of activation function in image classification 激活函数在图像分类中的作用
Yuxian Sun
{"title":"The role of activation function in image classification","authors":"Yuxian Sun","doi":"10.1109/CISCE52179.2021.9445868","DOIUrl":"https://doi.org/10.1109/CISCE52179.2021.9445868","url":null,"abstract":"Image classification is an important branch in the field of computer vision. Improving the accuracy of image classification has always been the goal pursued by many researchers. From the perspective of activation function, this paper studies the role of activation function in image classification. In our experiments, we found that under the LeNet network structure, the classification accuracy rate of 98.55% was achieved when using Swish, which exceeded the 98.18% of ReLU and 98.23% of LeakyReLU. In the AlexNet network structure, the highest classification accuracy rate is 90.23% when using ReLU, 90.20% for PReLU, and 89.71% for Swish, which are 0.03% and 0.49% lower than Mish respectively. In the VGGNet network structure, the use of ReLU achieves the highest classification accuracy of 90.23%, while Swish is 89.71% and Mish is 89.64%, which are 0.52% and 0.59% lower than ReLU respectively. Through our experiments, we can get that different activation functions have different performances in different network structures.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"151 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123266539","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}
引用次数: 5
The Current Status and progress of Adversarial Examples Attacks 对抗性样例攻击的现状与进展
Chaoran Yuan, Xiaobin Liu, Zhengyuan Zhang
{"title":"The Current Status and progress of Adversarial Examples Attacks","authors":"Chaoran Yuan, Xiaobin Liu, Zhengyuan Zhang","doi":"10.1109/cisce52179.2021.9445917","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9445917","url":null,"abstract":"Applications based on deep learning are prevailing in many real-world scenarios. However, due to the interpretability of deep learning and the emergence of adversarial examples attacks, their reliability has been put into doubt. As for Natural Language Processing, adversarial samples that can be semantically similar to the original ones can also be generated so that NLP classification can be fooled without human observers' attention. In this paper, we give an overview of the existing method in implementing adversarial examples attacks. Firstly, we introduce the implementation of these methods and the damage they impose on the security, integrity, and robustness of NLP systems. Secondly, we discuss problems like the limited attention on the logic of defense, the curse of dimensionality, and the difficulty of implementing white-box adversaries. Finally, to present our opinion on ML security's structure, our discussion upon the standardization of both attack and defense is also included.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123592308","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}
引用次数: 6
Metasurface Broadband Circularly Polarized Antenna with Characteristic Mode Theory 基于特征模理论的超表面宽带圆极化天线
Liu Miao, Lu Zhong Liang, P. Qiang
{"title":"Metasurface Broadband Circularly Polarized Antenna with Characteristic Mode Theory","authors":"Liu Miao, Lu Zhong Liang, P. Qiang","doi":"10.1109/cisce52179.2021.9446009","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9446009","url":null,"abstract":"A metasurface broadband circularly polarized antenna is proposed based on the theory of characteristic mode. The antenna is composed of a feeder layer, an air layer, and a metasurface. The theory of characteristic mode is used to analyze and predict the mode features of the metasurface. Besides, mode 1 and mode 2 working at 7GHz and 8.7GHz as the expected modes, a metasurface broadband circularly polarized antenna is designed. It is indicated by the simulation result that $vertmathrm{S}11vert < -10$ dB bandwidth of 5.32GHz–9.18GHz, with the relative bandwidth reaching 53.25%; 3 dB Axial Ratio bandwidth of 6.87GHz–8.93GHz, with the relative bandwidth of up to 26.07%. It can be confirmed that the antenna possesses performance of broadband circular polarization.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126998755","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}
引用次数: 0
Classifying Bone Metastasized SPECT Images in RGB Format with VGGNets 用VGGNets对RGB格式骨转移SPECT图像进行分类
Xiaoyan Chen, Ziwen Zheng, Liangxia Liu, Weili Wang, Qiang Lin
{"title":"Classifying Bone Metastasized SPECT Images in RGB Format with VGGNets","authors":"Xiaoyan Chen, Ziwen Zheng, Liangxia Liu, Weili Wang, Qiang Lin","doi":"10.1109/cisce52179.2021.9446035","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9446035","url":null,"abstract":"SPECT imaging is widely used for displaying lesions of bone metastasis in nuclear medicine. With SPECT images in RGB format, in this work, we propose to develop automated classification models based on VGG framework, which can answer that whether or not an image contain lesions. Experimental evaluation conducted on a set of real-world data of SPECT imaging shows that the VGG13 model can effectively classify bone metastasis SPECT images, obtaining a value of 0.992, 0.99, 0.99 and 0.95 for accuracy Acc, precision Pre, recall Rec and F-1 scores, respectively.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"216 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115093381","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}
引用次数: 0
Anti-noise motor fault diagnosis method based on decision tree and the feature extraction methods in the time domain and frequency domain 基于决策树的抗噪声电机故障诊断方法以及时域和频域的特征提取方法
Liu-Zhuang Zhi, Ju Yuanliang, Xie Zhongheng
{"title":"Anti-noise motor fault diagnosis method based on decision tree and the feature extraction methods in the time domain and frequency domain","authors":"Liu-Zhuang Zhi, Ju Yuanliang, Xie Zhongheng","doi":"10.1109/cisce52179.2021.9446018","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9446018","url":null,"abstract":"Motor bearing state directly affects the reliability of motor operation. The decision tree algorithm has strong anti-noise and adaptability, which is widely used in motor bearing fault diagnosis. This paper proposes constructing many decision trees with similar structures by using the classification and regression tree (CART) algorithm to diagnose the fault of motor bearing. To verify its generality, effectiveness, and anti-interference under the actual operation of the motor, we add noises on the bearing dataset obtained from the Case Western Reserve University and use it to verify the effect of our method. Firstly, we construct feature vectors by the time domain and frequency domain feature extraction method. Then we use the new feature vector dataset to train and test the CART-based diagnosis model. The experimental results show that the proposed algorithm is suitable for motor bearing diagnosis under a noisy environment with high accuracy.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115358789","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}
引用次数: 2
CTG:A Controllable Text Generation Method based on the Joint Work of Language Model and Text Classifier CTG:一种基于语言模型和文本分类器协同工作的可控文本生成方法
Xuyuan Liang, Lihua Tian, Chen Li, Zhang Mandi
{"title":"CTG:A Controllable Text Generation Method based on the Joint Work of Language Model and Text Classifier","authors":"Xuyuan Liang, Lihua Tian, Chen Li, Zhang Mandi","doi":"10.1109/cisce52179.2021.9445873","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9445873","url":null,"abstract":"Text generation is an important research field in natural language processing. Today pre-trained models can generate text with high readability and fluency. However, the generated text content is usually not restricted, and the generated content cannot be controlled. Therefore, there is an urgent need for a controllable text generation method. Traditional controllable text generation methods usually have the disadvantages of high training cost and difficulty in adjustment. We propose a lightweight and easy-to-adjust method to update the language model through the discrimination results of the text classification model on the language model, thereby completing controllable text generation. And we have solved the problem of imbalance and quality degradation in text generation. Experiments show that the quality of the text generated by this method can also reach a higher level with strong subjectivity, can achieves the purpose of controlling text generation. And this method is easy to adjust, suitable for various scenarios.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115592557","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}
引用次数: 1
Explicit Construction of ($k+r, k$) MDS Code with Small Sub-packetization level and Optimal Access Property for All Nodes ($k+r, k$)具有小子分组级别和所有节点最优访问属性的MDS代码的显式构造
Yi Liu, Xiaohu Tang, Xianfu Lei
{"title":"Explicit Construction of ($k+r, k$) MDS Code with Small Sub-packetization level and Optimal Access Property for All Nodes","authors":"Yi Liu, Xiaohu Tang, Xianfu Lei","doi":"10.1109/CISCE52179.2021.9445943","DOIUrl":"https://doi.org/10.1109/CISCE52179.2021.9445943","url":null,"abstract":"In practical distributed storage systems, the sub-packetization level of MDS codes is required as small as possible. In the literature, by sacrificing the optimality of the repair bandwidth, some constructions of MDS codes with small sub-packetization level were presented in the work of Rawat et al. (IEEE Trans. Inform. Theory, 64(10), 6506–6525, Oct. 2018) and Li et al. (IEEE Trans. Inform. Theory, 67(4), 2162–2180, Apr. 2021). However, the sub-packetization level of those constructions given in the two works are the power of the number of the parity nodes, which is still relatively large for the practical systems. In this paper, we construct an MDS code over a small finite field, whose sub-packetization level can be the power of a positive integer $s$, where $s$ is less than the number of the parity nodes.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122044128","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}
引用次数: 0
Two Universal Learnig Frameworks Based on TL-GAN 基于TL-GAN的两种通用学习框架
Sun Jianfeng, C. Hanyu, Lou Chengkang
{"title":"Two Universal Learnig Frameworks Based on TL-GAN","authors":"Sun Jianfeng, C. Hanyu, Lou Chengkang","doi":"10.1109/cisce52179.2021.9446008","DOIUrl":"https://doi.org/10.1109/cisce52179.2021.9446008","url":null,"abstract":"When generative adversarial networks (GAN) have just become popular, many GAN models have room for expansion. Firstly, this paper introduced document text localization generation adversarial network (TL-GAN), a deep neural network used to perform text localization from digital images. Furthermore, this paper conducts corresponding research on the TL-GAN algorithm, innovating the application of TL-GAN from two aspects. By analyzing the TL-GAN framework, it can generate any image. At the same time, based on its fuzzy characteristic axis, the internal logic algorithm of TL-GAN is studied through the analysis of the results.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124665418","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}
引用次数: 0
A New Method for Characteristic Analysis of Gravity and Aeromagnetic Anomaly Data Based on Intelligent Clustering-FASAGA 基于智能聚类的重磁异常数据特征分析新方法——fasaga
Mingli Yang, Linfu Xue, Xiangjin Ran, Haijun Wang
{"title":"A New Method for Characteristic Analysis of Gravity and Aeromagnetic Anomaly Data Based on Intelligent Clustering-FASAGA","authors":"Mingli Yang, Linfu Xue, Xiangjin Ran, Haijun Wang","doi":"10.1109/CISCE52179.2021.9445942","DOIUrl":"https://doi.org/10.1109/CISCE52179.2021.9445942","url":null,"abstract":"In geophysical field, Gravity and Aeromagnetic characteristics can represent the deep geological characteristics of a region. It is difficult to find a general mathematical formula for joint analysis of Gravity and Aeromagnetic characteristics while analyzing the Gravity and Aeromagnetic anomaly data. In this paper, a new method, fuzzy analysis simulated annealing genetic algorithm (FASAGA), is proposed for pattern classification of combined Gravity and Aeromagnetic anomaly data. Apply the model to the test area. The experimental results show that FASAGA can get a good recognition effect.","PeriodicalId":308337,"journal":{"name":"2021 International Conference on Communications, Information System and Computer Engineering (CISCE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129916202","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}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书