摘要
针对视频经过编解码传输后时延测量效率低、误差大、成本高、无法在线无损监测等问题,提出并实现一种能在线无损监测视频时延的精确测量装置.该装置的硬件由现场可编程门阵列(FPGA)与处理器组成,其中FPGA负责计算量大的视频采集与散列编码,处理器负责散列特征匹配并计算出视频时延.在算法上利用视频帧散列特征提取匹配原理进行视频时延的实时测量.测量结果除了在装置上可显示外,还可以通过网页或屏幕信息显示(OSD)形式体现.试验结果表明,该装置运行稳定,测量标准差在56.6 ns之内,满足在线无损视频时延监测的需求.
Aiming at the problem of low efficiency,large error,high cost,and the inability of online non-destructive monitoring after video transmission through encoding and decoding,an accurate measurement device capable of online non-destructive monitoring of video delay and quality is proposed and implemented.The measurement device is composed of an field programmable gate array(FPGA)and a processor.The FPGA is responsible for the computationally intensive video capture and hash coding,and the processor is responsible for hash feature matching and calculating the video delay.In the algorithm,a method for real-time hardware implementation using the principle of video frame hash feature extraction and matching is proposed for measurement.In addition to displaying the measurement results on the device,it can also be reflected in the form of web pages or on screen display(OSD).The actual test results show that the device runs stably,and the measurement standard deviation is within 56.6 ns,which meets the needs of online nondestructive video delay monitoring.
作者
吴一亮
胡彦多
郑佳春
WU Yiliang;HU Yanduo;ZHENG Jiachun(College of Information Engineering,Jimei University,Xiamen 361021,China)
出处
《厦门大学学报（自然科学版）》
CAS
CSCD
北大核心
2021年第6期1032-1038,共7页
Journal of Xiamen University：Natural Science
基金
福建省引才“百人计划”项目
福建省教育厅资助省属高校专项(JK2017027)
集美大学国家基金培育计划(ZP2020041)。
关键词
时延
无损测量
视频
感知散列算法
latency
nondestructive measurement
video
perceptual hash algorithm
