How to use a few defect samples to complete the defect classification is a key challenge in the production of mobile phone screens.An attention-relation network for the mobile phone screen defect classification is pro...How to use a few defect samples to complete the defect classification is a key challenge in the production of mobile phone screens.An attention-relation network for the mobile phone screen defect classification is proposed in this paper.The architecture of the attention-relation network contains two modules:a feature extract module and a feature metric module.Different from other few-shot models,an attention mechanism is applied to metric learning in our model to measure the distance between features,so as to pay attention to the correlation between features and suppress unwanted information.Besides,we combine dilated convolution and skip connection to extract more feature information for follow-up processing.We validate attention-relation network on the mobile phone screen defect dataset.The experimental results show that the classification accuracy of the attentionrelation network is 0.9486 under the 5-way 1-shot training strategy and 0.9039 under the 5-way 5-shot setting.It achieves the excellent effect of classification for mobile phone screen defects and outperforms with dominant advantages.展开更多
With increasingly rampant telephone fraud activities,the social impact and economic losses caused to China have increased dramatically.Precise,convenient,and efficient fraudulent phone call recognition has become a ch...With increasingly rampant telephone fraud activities,the social impact and economic losses caused to China have increased dramatically.Precise,convenient,and efficient fraudulent phone call recognition has become a challenge since telephone fraud became more varied and covert.To deal with this problem,many researchers have extracted some statistical features of telephone fraud behavior and proposed some machine learning algorithms on the field of fraudulent phone call recognition.In this paper,the calling detail records are utilized to construct a classifier for fraudulent phone call recognition.Meantime,a deep learning approach based on convolutional neural network(CNN)is proposed for better features learning and compared with the existing state-of-the-art machine learning algorithms.It learns phone number and call behavior features of telephone fraud,and improves the accuracy of classification.The evaluation results show that the proposed algorithm outperforms competitive algorithms.展开更多
Usefulness of sensor network applications in human life is increasing day by day and the concept of wireless connection promises new application areas. Sensor network can be very beneficial in saving human life from t...Usefulness of sensor network applications in human life is increasing day by day and the concept of wireless connection promises new application areas. Sensor network can be very beneficial in saving human life from terrorist attacks causing explosion in certain areas leading to casualties. But realization of the sensor network application in explosive detection requires high scalability of the sensor network and fast transmission of the information through real time monitoring and control. In this paper a novel mechanism for explosive trace detection in any populated area by the use of mobile telephony has been described. The aim is to create a system that will assure common men, local population and above all the nation a secured environment, without disturbing their freedom of movement. It would further help the police in detection of explosives more quickly, isolation of suicide bombers, remediation of explosives manufacturing sites, and forensic and criminal investigation. To achieve this, the paper has projected an idea that can combine the strength of the mobile phones, the polymer sensor and existing cellular network. The idea is to design and embed a tiny cog-nitive radio sensor node into the mobile phone that adapts to the changing environment by analyzing the RF surroundings and adjusting the spectrum use appropriately. The system would be capable of detecting explo-sives within a defined territory. It would communicate the location of the detected explosives to the respec-tive service provider, which in turn would inform the law and enforcement agency or Police.展开更多
Due to the characteristics of variability and dispersion in traffic information, to get the reliable real-time traffic information has been a bottleneck in the development of intelligent transportation systems. Howeve...Due to the characteristics of variability and dispersion in traffic information, to get the reliable real-time traffic information has been a bottleneck in the development of intelligent transportation systems. However, with the development of wireless network technology and mobile Internet, the mobile phones are rapidly developed and more popular, so it is possible to get road traffic information by locating the mobile phones in vehicles. The system structure for the road traffic information collection is designed based on wireless network and mobile phones in vehicles, and the vehicle recognition and its information computation methods are given and discussed. Also the simulation is done for vehicle recognition and computation based on fuzzy cluster analysis method and the results are obtained and analyzed.展开更多
目的:评价运动处方、普通心理、认知行为和团体辅导这四种不同干预措施对大学生手机依赖的效果。方法:检索PubMed、Web of Science、Embase、知网等数据库,收集四种干预措施对大学生手机依赖的文献,运用Revman5.3和Stata14.2软件进行数...目的:评价运动处方、普通心理、认知行为和团体辅导这四种不同干预措施对大学生手机依赖的效果。方法:检索PubMed、Web of Science、Embase、知网等数据库,收集四种干预措施对大学生手机依赖的文献,运用Revman5.3和Stata14.2软件进行数据分析。结果:共纳入39篇文献,Meta结果显示:与对照组相比,四种干预措施都能够有效减轻大学生手机依赖,均具有统计学意义。SUCRA值的排序结果:运动处方>认知行为>团体辅导>普通心理>无任何干预的对照组。结论:运动处方干预、普通心理干预、认知行为干预和团体辅导干预都能够有效减轻大学生的手机依赖,而相比于其他三种措施,运动处方成为最优措施的可能性最高,其他依次是认知行为干预、团体辅导干预和普通心理干预。展开更多
人群在城市内部空间中的流动是社会活力和资源分配的直接体现,是城市交通规划管理的重要依据。融合手机信令大数据和POI(point of interests)数据,对福州市主城区内的交通小区进行功能分区并构建了交通小区之间的空间交互网络,采用复杂...人群在城市内部空间中的流动是社会活力和资源分配的直接体现,是城市交通规划管理的重要依据。融合手机信令大数据和POI(point of interests)数据,对福州市主城区内的交通小区进行功能分区并构建了交通小区之间的空间交互网络,采用复杂网络方法对各类型功能区的中心性地位及出行距离衰减效应进行了分析。结果表明:福州市主城区功能区以居住及公共服务相关功能区为主。鼓楼区和台江区的人流量最为活跃,仓山区及晋安区的内部空间交互格局存在显著的不平衡现象。各功能区的中心性地位存在明显的空间和周期差异,科教文化及商服类型功能区的中心性普遍较高。居住用地与其他功能区之间的距离衰减效应受到时段的影响较大。基于以上结果可对城市公共交通优化提供一定支持。展开更多
文摘How to use a few defect samples to complete the defect classification is a key challenge in the production of mobile phone screens.An attention-relation network for the mobile phone screen defect classification is proposed in this paper.The architecture of the attention-relation network contains two modules:a feature extract module and a feature metric module.Different from other few-shot models,an attention mechanism is applied to metric learning in our model to measure the distance between features,so as to pay attention to the correlation between features and suppress unwanted information.Besides,we combine dilated convolution and skip connection to extract more feature information for follow-up processing.We validate attention-relation network on the mobile phone screen defect dataset.The experimental results show that the classification accuracy of the attentionrelation network is 0.9486 under the 5-way 1-shot training strategy and 0.9039 under the 5-way 5-shot setting.It achieves the excellent effect of classification for mobile phone screen defects and outperforms with dominant advantages.
基金the National Natural Science Foundation of China(No.61931019).
文摘With increasingly rampant telephone fraud activities,the social impact and economic losses caused to China have increased dramatically.Precise,convenient,and efficient fraudulent phone call recognition has become a challenge since telephone fraud became more varied and covert.To deal with this problem,many researchers have extracted some statistical features of telephone fraud behavior and proposed some machine learning algorithms on the field of fraudulent phone call recognition.In this paper,the calling detail records are utilized to construct a classifier for fraudulent phone call recognition.Meantime,a deep learning approach based on convolutional neural network(CNN)is proposed for better features learning and compared with the existing state-of-the-art machine learning algorithms.It learns phone number and call behavior features of telephone fraud,and improves the accuracy of classification.The evaluation results show that the proposed algorithm outperforms competitive algorithms.
文摘Usefulness of sensor network applications in human life is increasing day by day and the concept of wireless connection promises new application areas. Sensor network can be very beneficial in saving human life from terrorist attacks causing explosion in certain areas leading to casualties. But realization of the sensor network application in explosive detection requires high scalability of the sensor network and fast transmission of the information through real time monitoring and control. In this paper a novel mechanism for explosive trace detection in any populated area by the use of mobile telephony has been described. The aim is to create a system that will assure common men, local population and above all the nation a secured environment, without disturbing their freedom of movement. It would further help the police in detection of explosives more quickly, isolation of suicide bombers, remediation of explosives manufacturing sites, and forensic and criminal investigation. To achieve this, the paper has projected an idea that can combine the strength of the mobile phones, the polymer sensor and existing cellular network. The idea is to design and embed a tiny cog-nitive radio sensor node into the mobile phone that adapts to the changing environment by analyzing the RF surroundings and adjusting the spectrum use appropriately. The system would be capable of detecting explo-sives within a defined territory. It would communicate the location of the detected explosives to the respec-tive service provider, which in turn would inform the law and enforcement agency or Police.
文摘Due to the characteristics of variability and dispersion in traffic information, to get the reliable real-time traffic information has been a bottleneck in the development of intelligent transportation systems. However, with the development of wireless network technology and mobile Internet, the mobile phones are rapidly developed and more popular, so it is possible to get road traffic information by locating the mobile phones in vehicles. The system structure for the road traffic information collection is designed based on wireless network and mobile phones in vehicles, and the vehicle recognition and its information computation methods are given and discussed. Also the simulation is done for vehicle recognition and computation based on fuzzy cluster analysis method and the results are obtained and analyzed.
文摘目的:评价运动处方、普通心理、认知行为和团体辅导这四种不同干预措施对大学生手机依赖的效果。方法:检索PubMed、Web of Science、Embase、知网等数据库,收集四种干预措施对大学生手机依赖的文献,运用Revman5.3和Stata14.2软件进行数据分析。结果:共纳入39篇文献,Meta结果显示:与对照组相比,四种干预措施都能够有效减轻大学生手机依赖,均具有统计学意义。SUCRA值的排序结果:运动处方>认知行为>团体辅导>普通心理>无任何干预的对照组。结论:运动处方干预、普通心理干预、认知行为干预和团体辅导干预都能够有效减轻大学生的手机依赖,而相比于其他三种措施,运动处方成为最优措施的可能性最高,其他依次是认知行为干预、团体辅导干预和普通心理干预。
文摘人群在城市内部空间中的流动是社会活力和资源分配的直接体现,是城市交通规划管理的重要依据。融合手机信令大数据和POI(point of interests)数据,对福州市主城区内的交通小区进行功能分区并构建了交通小区之间的空间交互网络,采用复杂网络方法对各类型功能区的中心性地位及出行距离衰减效应进行了分析。结果表明:福州市主城区功能区以居住及公共服务相关功能区为主。鼓楼区和台江区的人流量最为活跃,仓山区及晋安区的内部空间交互格局存在显著的不平衡现象。各功能区的中心性地位存在明显的空间和周期差异,科教文化及商服类型功能区的中心性普遍较高。居住用地与其他功能区之间的距离衰减效应受到时段的影响较大。基于以上结果可对城市公共交通优化提供一定支持。
