Quality of experience(QoE) has been one of the most important factors for service providers and operators to develop and deploy services. Previous works have shown that Weber-Fechner law(WFL) can be used to promot...Quality of experience(QoE) has been one of the most important factors for service providers and operators to develop and deploy services. Previous works have shown that Weber-Fechner law(WFL) can be used to promote the accuracy of best-effort(BE) services' QoE assessment. In this paper, based on WFL, we take a deep insight into the visual information of BE services and its role in WFL, especially in web browsing services, and propose an enhanced QoE assessment model for BE services by taking the visual information into account. Both images and texts visual information are extracted and combined to get the visual information for web pages. Finally, a subjective experiment is performed to validate performance of proposed QoE assessment model, which shows that it performs better than traditional WFL based model.展开更多
文摘为了解决智能制造领域中云化控制与视觉分选应用相结合的问题,提出了基于深度学习的云化可编程逻辑控制器(Programmable logic controller,PLC)物料识别与定位系统,并在端到端5G与时间敏感网络(Time sensitive networking,TSN)传输网络环境下,实现了对云化PLC架构和控制功能有效性的验证.首先,将传统PLC系统控制功能容器虚拟化,实现PLC的本地和云端自由部署;其次,在云端设计人工智能学习平台,采用基于You only look once v5(YOLOv5)的目标检测算法实现物料的定位和分类,获取目标的像素坐标和类别信息;然后,利用相机标定方法把像素坐标转换成物理世界坐标,并将目标类别、坐标、时间戳信息传输到云化PLC;最后,在5G和TSN融合网络环境下,实现云化PLC对天车设备的实时控制与复杂计算功能整合.结果表明,该系统能够有效的对多天车进行协同控制,物料定位均值平均精度(Mean average precision,mAP)达到99.65%,分选准确率达到96.67%,平均消耗时间225.99 s,满足工业低时延、高精度的视觉分选需求.
基金supported by the Beijing Municipal Science and Technology Commission Research Fund Project (D151100000115002 )the Ministry of Education. CMCC Research fund (MCM20130132)
文摘Quality of experience(QoE) has been one of the most important factors for service providers and operators to develop and deploy services. Previous works have shown that Weber-Fechner law(WFL) can be used to promote the accuracy of best-effort(BE) services' QoE assessment. In this paper, based on WFL, we take a deep insight into the visual information of BE services and its role in WFL, especially in web browsing services, and propose an enhanced QoE assessment model for BE services by taking the visual information into account. Both images and texts visual information are extracted and combined to get the visual information for web pages. Finally, a subjective experiment is performed to validate performance of proposed QoE assessment model, which shows that it performs better than traditional WFL based model.