2021 7th International Conference on Computing and Artificial Intelligence最新文献_第7页

Robust Deep Facial Attribute Prediction against Adversarial Attacks 对抗攻击的鲁棒深度面部属性预测

2021 7th International Conference on Computing and Artificial Intelligence Pub Date : 2021-04-23 DOI: 10.1145/3467707.3467737

Kun Fang, Jie Yang

{"title":"Robust Deep Facial Attribute Prediction against Adversarial Attacks","authors":"Kun Fang, Jie Yang","doi":"10.1145/3467707.3467737","DOIUrl":"https://doi.org/10.1145/3467707.3467737","url":null,"abstract":"Face recognition has always been a hot topic in research, and has also widely been applied in industry areas and daily life. Nowadays, face recognition models with excellent performance are mostly based on deep neural networks (DNN). However, recently researchers find that images added invisible perturbations could successfully fool neural networks, which is known as the so-called adversarial attack. The perturbed images, also known as adversarial examples, are almost the same as the original images, but neural network could give different and wrong predictions with high confidence on these adversarial examples. Such a phenomenon indicates the vulnerable robustness of neural network and thus casts a shadow on the security of DNN-based face recognition models. Therefore, in this paper, we focus on the facial attribute prediction task in face recognition, investigate the influence of adversarial attack on facial attribute prediction and give a solution on improving the robustness of facial attribute prediction models. Extensive experiment results illustrate that the solution could indeed produce much more robust results in facial attribute prediction against adversarial attacks.","PeriodicalId":145582,"journal":{"name":"2021 7th International Conference on Computing and Artificial Intelligence","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117067042","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

Research on Intelligent Assignment Method of Radar Jamming Task in Electronic Air Defense Operation 电子防空作战中雷达干扰任务智能分配方法研究

2021 7th International Conference on Computing and Artificial Intelligence Pub Date : 2021-04-23 DOI: 10.1145/3467707.3467780

Jun Li, Lei Hu, W. Dai, Bin Fan

引用次数: 0

Design of On-orbit Software Reconfiguration for a General Signal Processing Platform Based on SOC Chip 基于SOC芯片的通用信号处理平台在轨软件重构设计

2021 7th International Conference on Computing and Artificial Intelligence Pub Date : 2021-04-23 DOI: 10.1145/3467707.3467765

Sheng-Chuan Liang, Kang Wang, Haiyuan Wang, Ruichen Wang, Rui Guo

引用次数: 0

Dynamic Load Balancing Method for Urban Surveillance Video Big Data Storage Based on HDFS 基于HDFS的城市监控视频大数据存储动态负载均衡方法

2021 7th International Conference on Computing and Artificial Intelligence Pub Date : 2021-04-23 DOI: 10.1145/3467707.3467730

Yue Li

{"title":"Dynamic Load Balancing Method for Urban Surveillance Video Big Data Storage Based on HDFS","authors":"Yue Li","doi":"10.1145/3467707.3467730","DOIUrl":"https://doi.org/10.1145/3467707.3467730","url":null,"abstract":"HDFS has been widely used by many video service websites, but its load balancing tool does not consider the bandwidth consumption characteristics of video file online playback and the heterogeneous performance difference of NameNode in metadata allocation problem. The dynamic load imbalance of cluster makes the utilization of bandwidth resources low. In this paper, a HDFS NameNode dynamic load balancing tool (NDLBT) for city monitoring video in urban surveillance video big data storage in cloud storage environment is proposed. method. Firstly, it analyzes the relationship between the bandwidth consumption and the bit rate, data block size and access heat of the video file when the video file is played online, and a new load evaluation model is established. On this basis, it adds consideration to the bandwidth consumption factor in the load scheme generation and load scheduling, and through the dynamic adaptive backup of multi-replica heterogeneous nodes of metadata. The dynamic distribution of metadata is realized under the consideration of node performance and load, and the performance of metadata server cluster is guaranteed. Finally, combined with cache strategy and automatic recovery mechanism, the reading and writing of metadata is improved. The simulation results show that compared with the proposed method, we can effectively avoid the aggregation of high bandwidth consumption data blocks. In the experimental scenario where high bandwidth consumption video files are used as service access hotspots, the proposed method is superior to the original load balancing method in 90% scenarios, and can reduce the bandwidth peak value of bottleneck nodes in data node clusters by 20%.","PeriodicalId":145582,"journal":{"name":"2021 7th International Conference on Computing and Artificial Intelligence","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125745857","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