视频中台体系架构研究与设计

doi:10.3778/j.issn.1002-8331.2402-0114

摘要/Abstract

摘要： 针对城市公共安全系统中存在视图资源利用率低、复杂监控图像的识别与检索难、大量视频数据处理难、功能变更难，体系架构层次间紧密耦合等问题与挑战，提出一种基于微服务的视频中台体系架构，包括系统的体系架构、任务与资源调度算法和视频图像检索算法。研究“大中台、小前台”的中台链接模式，设计由服务层、能力层和资源层构成的多级视频中台架构。将常用共性应用功能分解，设计为标准化细粒度的微服务组件，部署在服务层。将支撑软件运行和处理监控视频数据的功能，设计为基于容器化技术的能力引擎与中间件，部署在能力层。设计基于AI的警务大数据分析链路和算法库，部署在资源层。由此形成一个逻辑独立的嵌入式视频中台系统，所设计的视频中台成功应用于福州等市的城市安全系统，取得良好的应用成效。

关键词: 视频中台, 微服务, 调度引擎, 图像解析, AI算法库

Abstract: In response to the challenges and issues in urban public safety systems such as low utilization of visual resources, identifying and retrieving complex surveillance images, processing large volumes of video data, functional changes, and tight coupling between hierarchical levels of the system architecture, a video middle platform architecture based on microservices is proposed, including the system’s architecture, task and resource scheduling algorithms, and video image retrieval algorithms. The research explores the “large middle platform, small front platform” linking mode and designs a multi-level video middle platform architecture composed of service layer, capability layer, and resource layer. Common shared application functionalities are decomposed and designed as standardized, fine-grained microservice components deployed in the service layer. The functionalities supporting software operation and processing of surveillance video data are designed as capability engines and middleware based on containerization technology, deployed in the capability layer. AI-based police big data analysis chains and algorithm libraries are designed and deployed in the resource layer. This forms a logically independent embedded video middle platform system. The video middle platform designed in this paper has been successfully applied in urban security systems in cities such as Fuzhou, achieving significant application results.

Key words: video middleware, microservices, scheduling engine, image analysis, AI algorithm library

吴志雄, 崔勇. 视频中台体系架构研究与设计[J]. 计算机工程与应用, 2024, 60(23): 286-302.

WU Zhixiong, CUI Yong. Research and Design of Video Middle Platform Architecture[J]. Computer Engineering and Applications, 2024, 60(23): 286-302.

参考文献

[1] 钟华. 企业IT架构转型之道[M]. 北京: 机械工业出版社, 2020.
ZHONG H. The way of enterprise IT architecture trans-formation[M]. Beijing: China Machine Press, 2020.
[2] 刘颖慧, 刘楠. 数字化转型中不同企业的中台战略及架构设计[J]. 电信科学, 2020(7): 126-135.
LIU Y H, LIU N. Design of the strategy and architecture of different enterprises middle platform[J]. Telecommunication Science, 2020(7): 126-135.
[3] 元瑞捧. 基于算子共享的视频中台系统设计与实现[D]. 西安: 西安电子科技大学, 2021: 33-45.
YUAN R P. Design and implementation of video-middle station system based on operator sharing[D], Xi’an: Xidian University, 2021: 33-45.
[4] 英之道. 视频综合平台在安防监控中的应用[J]. 中国公共安全, 2019(1): 134-135.
YING Z D. Application of video integrated platform in security monitoring[J]. China Public Security, 2019(1): 134-135.
[5] WANG R, TIAN G, YING S. MicroCM: a cloud monitoring architecture for microservice invocation[J]. Computer Networks, 2024, 238: 110121.
[6] SILDATKE M, KARWANNI H, KRAFT B, et al. A distributed microservice architecture pattern for the automated generation of information extraction pipelines[J]. SN Computer Science, 2023, 4(6): 833.
[7] 杨舒, 苏放. 基于微服务的分布式数据安全整合应用系统[J]. 计算机工程与应用, 2021, 57(18): 238-247.
YANG S, SU F. Distributed data security integrated application system based on microservices[J]. Computer Engineering and Applications, 2021, 57(18): 238-247.
[8] 刘志杰, 陈佐, 颜思宇. 基于双中台架构的营销 2. 0 移动作业平台的设计与应用[J]. 电工电气, 2022(11): 66-71.
LIU Z J, CHEN Z, YAN S Y. Design and application of marketing 2. 0 mobile operation platform based on dual-middle platform architecture[J]. China Electrical Engineering, 2022(11): 66-71.
[9] 陆钊, 温禧茜. 基于数据中台架构MLSQL的大数据实验平台研发[J]. 计算机应用, 2022(7): 118-121.
LU Z, WEN X Q. Research and development of big data experimental platform based on MLSQL[J]. Journal of Computer Applications, 2022(7): 118-121.
[10] 肖逸, 李俊业. 基于微服务的大电网在线安全稳定分析系统数据中台与架构设计[J]. 中国电机工程学报, 2022, 42(1): 126-135.
XIAO Y, LI J Y. Microservice based data integration platform and architecture design of online security and stability analysis system for a bulk power system[J]. Proceedings of the CSEE, 2022, 42(1): 126-135.
[11] TIAN Y, BROWN L, HAMPAPUR A, et al. IBM smart surveillance system (S3): event based video surveillance system with an open and extensible framework[J]. Machine Vision and Applications, 2008, 19: 315-327.
[12] ANJUM A, ABDULLAH T, TARIQ M F, et al. Video stream analysis in clouds: an object detection and classification framework for high performance video analytics[J]. IEEE Transactions on Cloud Computing, 2016, 7(4): 1152-1167.
[13] 李小将, 王巍, 熊帆. 中台架构在智慧机场建设中的应用初探[J]. 民航学报, 2022, 6(5): 34-37.
LI X J, WANG W, XIONG F. Application of middle plat-form architecture in smart airport construction[J]. Journal of Civil Aviation, 2022, 6(5): 34-37.
[14] 杨茜泽, 朱里越, 汤新坤. 基于智慧广电的视频中台建设[J]. 广播电视网络, 2022, 29(4): 23-25.
YANG Q Z, ZHU L Y, TANG X K. Construction of video medium station based on smart radio and television[J]. Radio and Television Network, 2022, 29(4): 23-25.
[15] 孙苏勇. 视频监控识别预警平台的设计与实现[J]. 福建电脑, 2023, 39(5): 92-97.
SUN S Y. Design and implementation of an early warning platform for video surveillance identification[J]. Fujian Computer , 2023, 39(5): 92-97.
[16] 何瑶. 深圳安博会产品综述新安防迈进 5G 时代[J]. 中国公共安全, 2019(12): 99-105.
HE Y. Shenzhen security expo product overview new security into the 5G era[J]. China Public Security, 2019(12): 99-105.
[17] 王英杰. 海康威视: 多维感知和人工智能打造智慧流域[J]. 中国水利, 2021(12): 112.
WANG Y J. Hikvision: multi-dimensional perception and artificial intelligence to create smart watershed[J]. China Water Resources, 2021(12): 112.
[18] 超高精度人体测温大华股份用科技筑牢防疫堡垒[J]. 中国公共安全, 2020(3): 34-35.
Ultra high precision human body temperature measurement Dahua Shares build a strong fortress of epidemic prevention with science and technology[J]. China Public Security, 2020(3): 34-35.
[19] 李夏风, 傅小龙. 大模型在安防领域的实践应用——以云从科技从容大模型安防领域实践应用为例[J]. 中国安防, 2023(9): 49-53.
LI X F, FU X L. The practical application of large model in the field of security—a case study of the practical application of cloud technology in the field of large model security [J]. China Security, 2023(9): 49-53.
[20] 罗超. 软件定义智能协同智能摄像机新标杆——评测紫光华智400万软件定义枪型网络摄像机[J]. 中国公共安全, 2019(12): 104-105.
LUO C. A new benchmark for software-defined intelligent collaborative intelligent camera: evaluation of 4 million software-defined gun network cameras of Ziguang Huazhi [J]. China Public Security, 2019(12): 104-105.
[21] 邓钰涵. 人脸识别信息保护研究[D]. 上海: 上海师范大学, 2023.
DENG Y H. Research on face recognition information protection[D]. Shanghai: Shanghai Normal University, 2023.
[22] 王荣琤. 基于霍尼韦尔WEBs的智能照明监控实训系统设计与实现[J]. 数字通信世界, 2022(1): 138-140.
WANG R J. Design and implementation of intelligent lighting monitoring training system based on honeywell WEBs[J]. Digital Communication World, 2022(1): 138-140.
[23] ZHENG Y, YANG H, ZHANG T, et al. General facial representation learning in a visual-linguistic manner[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 18697-18709.
[24] ZHAO Y, LV W, XU S, et al. Detrs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024: 16965-16974.
[25] CHENG H K, OH S W, PRICE B, et al. Tracking anything with decoupled video segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023: 1316-1326.
[26] WANG C Y, LIAO H Y M, WU Y H, et al. CSPNet: a new backbone that can enhance learning capability of CNN[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020: 390-391.
[27] LIN T Y, DOLLáR P, GIRSHICK R B, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
[28] DAI X, CHEN Y, XIAO B, et al. Dynamic head: unifying object detection heads with attentions[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 7373-7382.
[29] RADENOVI? F, TOLIAS G, CHUM O. Fine-tuning CNN image retrieval with no human annotation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 41(7): 1655-1668.
[30] 赵月松. 设计院构建“大中台小前台”模式势在必行[J]. 中国勘察设计, 2019(11): 72-74.
ZHAO Y S. It is imperative for design institutes to build a model of “big, middle and small front office”[J]. China Survey and Design, 2019(11): 72-74.
[31] 宁明超, 张俊勃, 陈戈. 基于面向服务架构的工业软件的任务调度算法[J]. 计算机应用, 2023, 43(3): 885-893.
NING M C, ZHANG J B, CHEN G. Task scheduling algorithm for service-oriented architecture-based industrial software[J]. Journal of Computer Applications, 2023, 43(3): 885-893.
[32] LUO Q, HU S, LI C, et al. Resource scheduling in edge computing: a survey[J]. IEEE Communications Surveys & Tutorials, 2021, 23(4): 2131-2165.
[33] XIANG Y, ZHOU Y, YANG X, et al. A many-objective evolutionary algorithm with Pareto-adaptive reference points[J]. IEEE Transactions on Evolutionary Computation, 2019, 24(1): 99-113.
[34] GU Q, CHEN S, JIANG S, et al. Improved strength Pareto evolutionary algorithm based on reference direction and coordinated selection strategy[J]. International Journal of Intelligent Systems, 2021, 36(9): 4693-4722.
[35] LIN X, ZHEN H L, LI Z, et al. Pareto multi-task learning[J]. arXiv:1912.12854, 2019.
[36] ZOU J, HAO T, YU C, et al. A3C-DO: a regional resource scheduling framework based on deep reinforcement learning in edge scenario[J]. IEEE Transactions on Computers, 2020, 70(2): 228-239.
[37] 王迪聪, 白晨帅, 邬开俊. 基于深度学习的视频目标检测综述[J]. 计算机科学与探索, 2021, 15(9): 1563-1577.
WNAG D C. BAI C S. WU K J. Survey of video object detection based on deep learning[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(9): 1563-1577.
[38] 周康, 孔慧慧. 面向多场景的大数据多维分析系统的设计与实现[EB/OL].[2024-01-15]. http://www. paper. edu. cn.
ZHOU K, KONG H H. For many scenes of multidimen-sional data analysis system design and implementation of [EB/OL].[2024-01-15]. http://www. paper.edu.cn.
[39] DERYABIN S, TEMKIN I, RZAZADE U, et al. Models and methods of designing data-centric microservice architectures of digital enterprises[J]. Informatics, 2023, 10(1): 4.
[40] LI P, SONG J, XU H, et al. Resource scheduling optimisation algorithm for containerised microservice architecture in cloud computing[J]. International Journal of High Performance Systems Architecture, 2018, 8(1/2): 51-58.
[41] FREIRE G, PAUCAR H, ESTRELLA J. A distributed software architecture based on reactive microservices for the smart building context[C]//Proceedings of the 12th Latin-American Symposium on Dependable and Secure Computing, 2023: 168-169.
[42] CHAUDHRY N R, ANAGNOSTOU A, TAYLOR S J E. A workflow architecture for cloud-based distributed simulation[J]. ACM Transactions on Modeling and Computer Simulation, 2022, 32(2): 1-26.
[43] SCATTONE F F, BRAGHETTO K R. A microservices architecture for distributed complex event processing in smart cities[C]//Proceedings of the 2018 IEEE 37th International Symposium on Reliable Distributed Systems Workshops, 2018: 6-9.
[44] 鲁逸丁. 基于虚拟机聚合的云平台高能效资源调度框架[J]. 电子技术与软件工程, 2022(16): 258-262.
LU Y D. Cloud platform energy efficient resource scheduling framework based on virtual Machine aggregation[J]. Electronic Technology and Software Engineering, 2022(16): 258-262.
[45] JIAO Q, XU B, FAN Y. Design of cloud native application architecture based on kubernetes[C]//Proceedings of the 2021 IEEE International Conference on Dependable, Autonomic and Secure Computing, International Conference on Pervasive Intelligence and Computing, International Conference on Cloud and Big Data Computing, International Conference on Cyber Science and Technology Congress, 2021: 494-499.
[46] FIORI S, ABENI L, CUCINOTTA T. RT-kubernetes: containerized real-time cloud computing[C]//Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing, 2022: 36-39.
[47] ZENG R, ZHAO Y, SU H, et al. A novel construction technology of enterprise business deployment architecture based on containerized microservice[C]//Proceedings of the 2020 5th International Conference on Communication, Image and Signal Processing, 2020: 256-263.
[48] HAN J, HONG Y, KIM J. Refining microservices placement employing workload profiling over multiple kubernetes clusters[J]. IEEE Access, 2020, 8: 192543-192556.
[49] RAMASAMY B, NA Y, KIM W, et al. HACM: high availability control method in container-based microservice applications over multiple clusters[J]. IEEE Access, 2022, 11: 3461-3471.