为了提高有限码率下基于高效率视频编码(high efficiency video coding,HEVC)监控类视频的编码性能,提出了一种基于感兴趣区域的码率控制改进算法。对于帧层,以编码帧的信息熵和参数调整因子构造一种新的帧层位权重,用于指导帧层目标比...为了提高有限码率下基于高效率视频编码(high efficiency video coding,HEVC)监控类视频的编码性能,提出了一种基于感兴趣区域的码率控制改进算法。对于帧层,以编码帧的信息熵和参数调整因子构造一种新的帧层位权重,用于指导帧层目标比特分配;对于最大编码单元(largest coding unit,LCU)层,以LCU的空间复杂度纹理因子和时间复杂度运动因子为感兴趣因子,与特征量化加权系数一起用于指导LCU层目标比特分配,并对模型参数进行自适应更新。性能测试结果表明,提出的算法在低延迟P帧和低延时B帧配置下,与HEVC码率控制算法相比,率失真性能平均分别提高了14.2%和12.2%,结构相似性平均分别提高了0.00483和0.00371,且具有稳定的码率控制精度和较低的编码复杂度。展开更多
In the past years smart homes solutions have become more and more popular with the introduction of a high number of both Internet of Things(IoT)applications and smart devices.Home automation and security systems marke...In the past years smart homes solutions have become more and more popular with the introduction of a high number of both Internet of Things(IoT)applications and smart devices.Home automation and security systems market is in a continuous growth,traditional security systems are evolving fast,and more and more people choose Smart home solutions.In this paper,we propose two IoT based systems in the context of Smart homes:qToggle for multiple home automation,and MotionEyeOS,a video surveillance OS for single-board computers.Most qToggle devices are based on ESP8266/ESP8285 chips or on Raspberry Pi boards and smart sensors,while MotionEye uses Raspberry Pi boards.展开更多
文摘In the past years smart homes solutions have become more and more popular with the introduction of a high number of both Internet of Things(IoT)applications and smart devices.Home automation and security systems market is in a continuous growth,traditional security systems are evolving fast,and more and more people choose Smart home solutions.In this paper,we propose two IoT based systems in the context of Smart homes:qToggle for multiple home automation,and MotionEyeOS,a video surveillance OS for single-board computers.Most qToggle devices are based on ESP8266/ESP8285 chips or on Raspberry Pi boards and smart sensors,while MotionEye uses Raspberry Pi boards.