WO2022016573A1 - 一种视频监控分析系统和方法 - Google Patents

一种视频监控分析系统和方法 Download PDF

Info

Publication number
WO2022016573A1
WO2022016573A1 PCT/CN2020/105183 CN2020105183W WO2022016573A1 WO 2022016573 A1 WO2022016573 A1 WO 2022016573A1 CN 2020105183 W CN2020105183 W CN 2020105183W WO 2022016573 A1 WO2022016573 A1 WO 2022016573A1
Authority
WO
WIPO (PCT)
Prior art keywords
video
module
image data
training
management server
Prior art date
Application number
PCT/CN2020/105183
Other languages
English (en)
French (fr)
Inventor
陈小兵
赵金玲
高青
Original Assignee
南京智金科技创新服务中心
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南京智金科技创新服务中心 filed Critical 南京智金科技创新服务中心
Publication of WO2022016573A1 publication Critical patent/WO2022016573A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/44Secrecy systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • H04N1/32149Methods relating to embedding, encoding, decoding, detection or retrieval operations
    • H04N1/32267Methods relating to embedding, encoding, decoding, detection or retrieval operations combined with processing of the image
    • H04N1/32272Encryption or ciphering
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources

Definitions

  • the present application relates to the technical field of video processing, and in particular, to a video surveillance analysis system and method.
  • the alarm will be issued.
  • the image is stored and archived for evidence collection, and the detailed information of the alarm is stored in the database for later query.
  • real-time intelligent analysis of video consumes resources, and there are usually dozens or even more channels of video in a video surveillance system. Therefore, it is of great significance to perform real-time video analysis on a large number of surveillance videos at the same time.
  • the purpose of the present application is to provide a video surveillance analysis system and method, which have the advantages of high security, high analysis accuracy and improved accuracy, and various functions.
  • the present application discloses a video surveillance analysis system, comprising: a management server, a training server and several video preprocessing terminals; the management server and the training server are respectively connected in communication with several video preprocessing terminals ; the management server is used to control and manage the training server and several of the video preprocessing terminals; the training server is used to use the image data as training samples to train the danger recognition model; the video preprocessing terminal is used to The image data of the monitoring point is collected, the collected image data is analyzed according to the danger identification model, the analysis result is fed back to the management server, and the image data is uploaded to the training server.
  • the video preprocessing terminal includes: a camera, an analysis module and a video encryption module; the camera is connected in communication with the analysis module; the analysis module is connected in communication with the video encryption module; The camera is used to collect the image data of the monitoring point, and send the image data to the analysis module and the video encryption module; the analysis module is used to analyze the image data according to the risk identification model, and analyze the results of the analysis. feedback to the management server; the video encryption module is used to encrypt the collected image data through the video key, and transmit the encrypted image data to the training server and the management server.
  • the management server includes: a management module and a first video decryption module; the first video decryption module is connected in communication with a video encryption module in the video preprocessing terminal; the management module Used to manage the training server and several of the video preprocessing terminals, and when the feedback analysis result is dangerous, execute a preset alarm processing flow; the first video decryption module is used for The video data of the encryption module is decrypted.
  • the management module is further configured to store and manage image data of the camera.
  • the training server includes: a training module and a second video decryption module; the second video decryption module is connected in communication with the video encryption module in the video preprocessing terminal; the training module is used to The image data whose analysis result is dangerous is used as a training sample, and the training sample is trained through a deep belief network to update and improve the dangerous identification model; the second video decryption module is used to perform video data from the video encryption module. decrypt.
  • the video preprocessing terminal further includes a first authentication module and a first key agreement module;
  • the management server further includes a second authentication module and a second key agreement module; the first authentication module and a second key agreement module;
  • An authentication module and a second authentication module are used to realize the authentication between the video preprocessing terminal and the management server through a public key; the first key agreement module and the second key agreement module are used to pass The key establishes a key agreement between the video preprocessing terminal and the management server.
  • the training server further includes a third authentication module and a third key agreement module; the first authentication module and the third authentication module are used to realize the video preprocessing terminal by using a public key Authentication with the training server; the first key agreement module and the third key agreement module are used to establish key agreement between the video preprocessing terminal and the management server through a key .
  • the video key is updatable.
  • the present application also discloses a video surveillance analysis method, comprising: collecting image data of a monitoring point through a video preprocessing terminal; analyzing, identifying and judging the collected image data according to a danger identification model: when it is judged to be normal, The image data is uploaded to the management server and the training server; when it is judged to be dangerous, the analysis results are fed back to the management server, and the management server executes the preset alarm processing process, and uploads the collected image data to the management server and the training server.
  • the video surveillance analysis method further includes: realizing authentication between the video preprocessing terminal and the management server and/or training server by using the public key; establishing the video preprocessing terminal and the management server by using the key and/or key agreement between training servers.
  • the video surveillance analysis method further includes: encrypting the collected image data with a video key; uploading the encrypted image data to the management server and the training server.
  • High security The video surveillance analysis system of this application ensures that the collected video will not be easily stolen and used by people during the transmission process by encrypting the transmission, and the security is high.
  • the analysis is accurate and the accuracy can be improved:
  • the video surveillance analysis system of the present application can continuously improve the accuracy of the analysis through deep learning, and the analysis is more accurate than the traditional analysis system.
  • FIG. 1 shows a schematic diagram of a system structure of a video surveillance analysis system provided by an embodiment of the present application.
  • the present embodiment discloses a video surveillance analysis system, including but not limited to: a management server, a training server, and several video preprocessing terminals; the management server and the training server communicate with several video preprocessing terminals respectively connection for data exchange.
  • the management server, the training server and several video preprocessing terminals can be connected by wired communication or by wireless communication. Since this is a technology well known to those skilled in the art, it will not be repeated here. in,
  • the video preprocessing terminal is used to collect the image data of the monitoring point, analyze the collected image data according to the danger identification model, and feed back the analysis results to the management server, and upload the image data to the training server.
  • the image data includes video data and/or image data.
  • the so-called monitoring points are set according to actual needs. For example, if the video surveillance analysis system is used for fire alarm monitoring in the factory building, then the monitoring point is set in the fire-prone area in the factory building, and the video preprocessing terminal is used to collect the image data of the monitoring point.
  • the video preprocessing terminal includes but is not limited to: a camera, an analysis module, and a video encryption module; the camera is in communication with the analysis module; the analysis module is in communication with the video encryption module; wherein,
  • the camera is used to collect the image data of the monitoring point, and send the image data to the analysis module and the video encryption module.
  • the analysis module is used for analyzing the image data according to the danger identification model, and feeding back the analysis results to the management server; wherein, the danger identification model is obtained by the training server continuously training the training samples through the neural network.
  • the neural network adopts a deep belief network.
  • the video encryption module is used to encrypt the collected image data through the video key, and transmit the encrypted image data to the training server and the management server. Based on the consideration of the security of the image data, in this embodiment, before the video preprocessing terminal sends the image data to the management server and the training server, the image data will be encrypted to prevent the image data from being stolen.
  • the video preprocessing terminal further includes a first authentication module and a first key agreement module, where the first authentication module is used to implement authentication with the management server and/or the training server through a public key.
  • the first key negotiation module is used for key negotiation with the management server and/or the training server through key establishment.
  • the training server is used to train the hazard identification model by using the image data as training samples.
  • the training server includes but is not limited to: a training module and a second video decryption module; the second video decryption module is connected in communication with the video encryption module in the video preprocessing terminal, wherein,
  • the training module is used to use the image data whose analysis result is dangerous as a training sample, and train the training sample through the deep belief network, and continuously update and improve the danger identification model.
  • the deep belief network is a neural network learning method well known to those skilled in the art, and will not be repeated here.
  • the second video decryption module is used for decrypting the video data from the video encryption module.
  • the collected image data is encrypted and processed by the video key at the video preprocessing terminal.
  • a corresponding second video decryption module is set at the training server to perform decryption processing by the video key.
  • the training server also includes a third authentication module and a third key agreement module; wherein, the third authentication module of the training server and the first authentication module of the video preprocessing terminal are used to realize the training server and video preprocessing through the public key. Authentication between terminals; the third key negotiation module of the training server and the first key negotiation module of the video preprocessing terminal establish key negotiation between the training server and the video preprocessing terminal through keys.
  • the management server is used to control and manage the training server and several video preprocessing terminals.
  • the management server includes but is not limited to: a management module and a first video decryption module; the first video decryption module is connected in communication with the video encryption module in the video preprocessing terminal; wherein
  • the management module is used to manage the training server and several video preprocessing terminals, and when the analysis result fed back from the video preprocessing terminal is dangerous, execute the preset alarm processing flow; the management module is also used to store and manage images data.
  • the first video decryption module is used for decrypting the video data from the video encryption module.
  • the management server is provided with a first video decryption module corresponding to the video encryption module of the video preprocessing terminal, for decrypting the image data by using the video key.
  • the management server also includes a second authentication module and a second key agreement module
  • the second authentication module of the management server and the first authentication module of the video preprocessing terminal are used to realize the authentication between the video preprocessing terminal and the management server through the public key;
  • the second key negotiation module of the management server and the first key negotiation module of the video preprocessing terminal are used to establish key negotiation between the video preprocessing terminal and the management server through the key.
  • the video key is updatable.
  • the two-way authentication between the camera and the management server or the training server is performed when the camera first turns on or refreshes the session communication protocol and registers with the management server or the training server.
  • both parties obtain each other's public key, that is, a digital certificate.
  • the public key is used in the key negotiation process when the subsequent key negotiation is established, and negotiates the message authentication key MAK, which is used to authenticate subsequent signaling except registration messages.
  • MAK message authentication key
  • This embodiment discloses a video surveillance analysis method, including:
  • the collected image data is analyzed and identified according to the hazard identification model:
  • the management server executes the preset alarm processing process, and uploads the collected image data to the management server and the training server.
  • the key agreement between the video preprocessing terminal and the management server and/or the training server is established through the key.
  • the management server or the training server sends image data request information to the camera of the video preprocessing terminal, and the image data request information includes signaling and the hashed key MAK;
  • the video preprocessing terminal After the video preprocessing terminal receives the image data request information, it first verifies the key MAK. If the key MAK is verified, it is also used to send information to the management server/training server in two cases:
  • Case 1 If the video preprocessing terminal does not update the video key VKEK, then use the public key of the management server/training server to encrypt the video key VKEK to generate the video key encryption key ciphertext EVKEK, and then encrypt the video key
  • the key ciphertext EVKEK and the video key encryption key version number VKEVVersion are placed in the SDP channel and sent to the management server/training server;
  • the second case if the video preprocessing terminal updates the video key VKEK, the video key VKEK is encrypted with the public key of the management server/training server to generate the video key encryption key ciphertext EVKEK, and then the video key is encrypted and encrypted.
  • the key ciphertext EVKEK, the updated video key encryption key version number VKEVVersion, and the hashed key MAK are sent to the management server/training server; when the video preprocessing terminal receives the information, it verifies the key MAK After the verification is passed, the video key encryption key ciphertext EVKEK and the video key encryption key version number VKEVVersion are put into the SDP channel and sent to the management server/training server.
  • the video preprocessing terminal After the video preprocessing terminal receives the video key encryption key ciphertext EVKEK and the video key encryption key version number VKEVVersion, it also verifies the key MAK. After the verification is passed, it returns the verification receipt to the camera. Key negotiation with management server/training server succeeded.
  • the video surveillance analysis method further includes: encrypting the collected image data with a video key; uploading the encrypted image data to the management server and the training server.
  • the video surveillance analysis system and method provided by the present application ensures that the collected video will not be easily stolen and utilized by people during the transmission process by performing encrypted transmission, and has high security.
  • the video surveillance analysis system of the present application can continuously improve the accuracy of the analysis through deep learning. Compared with the traditional analysis system, the analysis is more accurate and the accuracy is greatly improved.
  • each block in the flowchart or block diagrams may represent a unit, segment, or portion of code that contains one or more functions for implementing the specified logical function. executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented in dedicated hardware-based systems that perform the specified functions or actions , or can be implemented in a combination of dedicated hardware and computer instructions.
  • each functional unit in each embodiment of the present application may be integrated together to form an independent part, or each unit may exist independently, or two or more units may be integrated to form an independent part.
  • the functions, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium.
  • the technical solution of the present application can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution.
  • the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Onl8 Memor8), random access memory (RAM, Random Access Memor8), magnetic disk or optical disk and other media that can store program codes .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Alarm Systems (AREA)

Abstract

本申请公开了一种视频监控分析系统及方法,涉及视频处理技术领域。包括:管理服务器、训练服务器以及若干个视频预处理终端;管理服务器用于控制和管理训练服务器和若干个视频预处理终端;训练服务器用于以影像数据作为训练样本,训练危险识别模型;视频预处理终端用于采集监控点的影像数据,依据危险识别模型分析采集的影像数据,并将分析结果反馈至管理服务器,将影像数据上传至训练服务器。所述视频监控分析系统具有安全性高、分析准确性高且准确性可以提升,以及功能多样的优点。

Description

一种视频监控分析系统和方法 技术领域
本申请涉及视频处理技术领域,具体而言,涉及一种视频监控分析系统和方法。
背景技术
随着模式识别技术的发展,针对视频中前景目标的检测、识别、跟踪等智能分析技术被广泛应用于智能视频监控系统中。从模拟视频监控系统到数字视频监控系统,再到如今如此普及的网络视频监控系统,不断被应用于城市交通管理、公安监控、厂房监督、小区治安等领域,视频监控系统已经成为当今社会生活中不可或缺的一部分。尤其是近年来随着智慧城市、智慧交通、智慧社区等领域迅速发展,更是促进了智能视频分析技术的发展。
目前绝大多数的视频监控系统都是从前端设备获取视频数据后直接解码并显示,仅在后台将解码后的图片录制成录像文件,以作存档,备事后取证,同时需要不间断的有监控工作人员在监控室盯着屏幕,以防在发生异常事件时可及时挽救或制止,但是工作人员总是会有疲惫期或者疏忽的时候,不能及时发现视频中发生的事件,甚至可能导致无法挽回的后果,这就促使智能视频分析系统的发展。智能视频分析系统直接实时分析摄像机前端拍摄的监控视频画面,用户可根据需求和目的设置警戒区域和触发报警的规则,分析结果满足规则后则报警,可在事情即将发生或未发生之前发现,能够做到有效预防或制止,与此同时将该图像画面进行存储留档以备取证,并将报警的详细信息存入数据库,以便后期进行查询。但是对视频进行实时智能分析耗费资源,且视频监控系统中通常有几十路甚至更多路视频,故而如何针对大量的监控视频同时进行实时的视频分析,有着重大意义。
发明内容
本申请的目的在于提供一种视频监控分析系统和方法,具有安全性高、分析准确性高且准确性可以提升,以及功能多样的优点。
为了实现上述目的,本申请公开了一种视频监控分析系统,包括:管理服务器、训练服务器以及若干个视频预处理终端;所述管理服务器和所述训练服务器分别与若干个视频预处理终端通信连接;所述管理服务器用于控制和管理所述训练服务器和若干个所述视频预处理终端;所述训练服务器用于以影像数据作为训练样本,训练危险识别模型;所述视频预处理终端用于采集监控点的影像数据,依据所述危险识别模型分析采集的影像数据,并将分析结果反馈至所述管理服务器,将所述影像数据上传至所述训练服务器。
在本申请一优选实施例中,所述视频预处理终端包括:摄像头、分析模块和视频加密模块;所述摄像头与所述分析模块通信连接;所述分析模块和所述视频加密模块通信连接;所述摄像头,用于采集监控点的影像数据,并将影像数据发送至所述分析模块和所述视频加密模块;所述分析模块用于依据所述危险识别模型分析影像数据,并将分析结果反馈至所述管理服务器;所述视频加密模块用于对采集的影像数据通过视频密钥进行加密处理,并将加密处理后的影像数据传输给训练服务器和管理服务器。
在本申请一优选实施例中,所述管理服务器包括:管理模块和第一视频解密模块;所述第一视频解密模块与所述视频预处理终端中的视频加密模块通信连接;所述管理模块用于管理所述训练服务器和若干个所述视频预处理终端,并且当接收到反馈的分析结果为危险时,执行预设的报警处理流程;所述第一视频解密模块用于对来自于视频加密模块的视频数据进行解密。
在本申请一优选实施例中,所述管理模块还用于存储并管理所述摄像头的影像数据。
在本申请一优选实施例中所述训练服务器包括:训练模块和第二视频解密模块;所述第二视频解密模块与视频预处理终端中的视频加密模 块通信连接;所述训练模块用于将分析结果为危险的影像数据作为训练样本,并通过深度信念网络对该训练样本进行训练,更新完善所述危险识别模型;所述第二视频解密模块用于对来自于视频加密模块的视频数据进行解密。
在本申请一优选实施例中,所述视频预处理终端还包括第一认证模块和第一密钥协商模块;所述管理服务器还包括第二认证模块和第二密钥协商模块;所述第一认证模块和第二认证模块用于通过公钥实现所述视频预处理终端与所述管理服务器之间的认证;所述第一密钥协商模块和所述第二密钥协商模块用于通过密钥建立所述视频预处理终端与所述管理服务器之间的密钥协商。
在本申请一优选实施例中,所述训练服务器还包括第三认证模块和第三密钥协商模块;所述第一认证模块和第三认证模块用于通过公钥实现所述视频预处理终端与所述训练服务器之间的认证;所述第一密钥协商模块和所述第三密钥协商模块用于通过密钥建立所述视频预处理终端与所述管理服务器之间的密钥协商。
在本申请一优选实施例中,所述视频密钥是可更新的。
本申请还公开了以一种视频监控分析方法,包括:通过视频预处理终端采集监控点的影像数据;对采集的影像数据依据危险识别模型进行分析识别判断:当判断为正常时,将采集的影像数据上传至管理服务器和训练服务器;当判断为危险时,反馈分析结果至管理服务器,管理服务器执行预设的报警处理流程,将采集的影像数据上传至管理服务器和训练服务器。
在本申请一优选实施例中,所述视频监控分析方法还包括:通过公钥实现视频预处理终端与管理服务器和/或训练服务器之间的认证;通过密钥建立视频预处理终端与管理服务器和/或训练服务器之间的密钥协商。
在本申请一优选实施例中,所述视频监控分析方法还包括:对采集的影像数据通过视频密钥进行加密;将加密的影像数据上传至管理服务器和训练服务器。
本申请提供的一种视频监控分析系统和方法,具有以下优点:
1.安全性高:本申请的视频监控分析系统通过进行加密传输,保证了采集到的视频不会在传输过程中被人轻易的窃取和利用,安全性较高。
2.分析准确且精准性可以提升:本申请的视频监控分析系统通过深度学习,可以不断提升分析的准确性,比起传统的分析系统,分析更加准确。
为使本申请的上述目的、特征和优点能更明显易懂,下文特举较佳实施例,并配合所附附图,作详细说明如下。
附图说明
为了更清楚地说明本申请实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,应当理解,以下附图仅示出了本申请的某些实施例,因此不应被看作是对范围的限定,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他相关的附图。
图1示出了本申请实施例提供的视频监控分析系统的系统结构示意图。
具体实施方式
以下通过特定的具体实例说明本申请的实施方式,本领域技术人员可由本说明书所揭露的内容轻易地了解本申请的其他优点与功效。本申请还可以通过另外不同的具体实施方式加以实施或应用,本说明书中的各项细节也可以基于不同观点与应用,在没有背离本申请的精神下进行各种修饰或改变。需说明的是,在不冲突的情况下,以下实施例及实施例中的特征可以相互组合。
需要说明的是,以下实施例中所提供的图示仅以示意方式说明本申请的基本构想,遂图示中仅显示与本申请中有关的组件而非按照实际实施时的组件数目、形状及尺寸绘制,其实际实施时各组件的型态、数量及比例可为一种随意的改变,且其组件布局型态也可能更为复杂。
实施例1:
如图1所示,本实施例公开了一种视频监控分析系统,包括但不限于:管理服务器、训练服务器以及若干个视频预处理终端;管理服务器和训练服务器分别与若干个视频预处理终端通信连接,进行数据交换。管理服务器、训练服务器和若干个视频预处理终端之间既可以采用有线通信连接,也可以采用无线通信连接方式。由于这是本领有技术人员所熟知的技术,在此不再赘述。其中,
视频预处理终端用于采集监控点的影像数据,依据危险识别模型分析采集的影像数据,并将分析结果反馈至管理服务器,将影像数据上传至训练服务器。其中,影像数据包括视频数据和/或图像数据。视频预处理终端设置与监控点。所谓监控点是根据实际需要进行设置的。例如,视频监控分析系统是用于厂房的火警监控,那么将监控点设置在厂房当中容易发生火灾的区域,利用视频预处理终端采集监控点的影像数据。
在本申请的一较佳实施例中,视频预处理终端包括但不限于:摄像头、分析模块和视频加密模块;摄像头与分析模块通信连接;分析模块和视频加密模块通信连接;其中,
摄像头用于采集监控点的影像数据,并将影像数据发送至分析模块和视频加密模块。
分析模块用于依据危险识别模型分析影像数据,并将分析结果反馈至管理服务器;其中,危险识别模型是由训练服务器通过神经网络对训练样本不断训练而获得的。优选地,神经网络采用深度信念网络。
视频加密模块用于对采集的影像数据通过视频密钥进行加密处理,并将加密处理后的影像数据传输给训练服务器和管理服务器。基于对影像数据的安全性的考量,在本实施例中,视频预处理终端将影像数据发送至管理服务器和训练服务器之前,都会对影像数据进行加密处理,以防影像数据被盗。
进一步地,视频预处理终端还包括第一认证模块和第一密钥协商模块,第一认证模块是用于通过公钥实现与管理服务器和/或训练服务器之间的认证。第一密钥协商模块用于通过密钥建立与管理服务器和/或训练服务器之间的密钥协商。
训练服务器用于以影像数据作为训练样本,训练危险识别模型。训 练服务器包括但不限于:训练模块和第二视频解密模块;第二视频解密模块与视频预处理终端中的视频加密模块通信连接,其中,
训练模块用于将分析结果为危险的影像数据作为训练样本,并通过深度信念网络对该训练样本进行训练,不断更新完善危险识别模型。深度信念网络是本领域技术人员所熟知的神经网络学习方法,在此不再赘述。
第二视频解密模块用于对来自于视频加密模块的视频数据进行解密。本实施例在视频预处理终端处对采集的影像数据通过视频密钥加密处理,对应地,在训练服务器处设置了相应的第二视频解密模块,通过视频密钥进行解密处理。
进一步地,训练服务器还包括第三认证模块和第三密钥协商模块;其中,训练服务器的第三认证模块和视频预处理终端的第一认证模块用于通过公钥实现训练服务器和视频预处理终端之间的认证;训练服务器的第三密钥协商模块和视频预处理终端的第一密钥协商模块通过密钥建立训练服务器和视频预处理终端之间的密钥协商。
管理服务器用于控制和管理训练服务器和若干个视频预处理终端。管理服务器包括但不限于:管理模块和第一视频解密模块;第一视频解密模块与视频预处理终端中的视频加密模块通信连接;其中
管理模块用于管理训练服务器和若干个视频预处理终端,并且当接收到来自于视频预处理终端反馈的分析结果为危险时,执行预设的报警处理流程;管理模块还用于存储并管理影像数据。
第一视频解密模块用于对来自于视频加密模块的视频数据进行解密。与训练服务器一样,管理服务器设置了与视频预处理终端的视频加密模块相对应的第一视频解密模块,以用于通过视频密钥对影像数据进行解密。
进一步地,管理服务器还包括第二认证模块和第二密钥协商模块;
管理服务器第二认证模块和视频预处理终端的第一认证模块用于通过公钥实现视频预处理终端与管理服务器之间的认证;
管理服务器第二密钥协商模块和视频预处理终端的第一密钥协商模块用于通过密钥建立视频预处理终端与管理服务器之间的密钥协商。
并且,在本申请一优选实施例中,视频密钥是可更新的。
根据本申请的一些实施例,摄像头与管理服务器或训练服务器之间的双向认证是在摄像头首次开启或刷新会话通信协议注册到管理服务器或训练服务器时进行。通过双向认证,双方获取对方的公钥,即数字证书,公钥用于后续密钥协商建立时的密钥协商过程,并协商消息认证密钥MAK,用于认证后续除了注册消息以外的信令。在摄像头与管理服务器或训练服务器之间进行密钥协商时,首次建立视频加密通信之间的密钥协商、以及定时更换密钥时的自动密钥协商。
实施了2
本实施例公开了一种视频监控分析方法,包括:
通过视频预处理终端采集监控点的影像数据;
对采集的影像数据依据危险识别模型进行分析识别判断:
当判断为正常时,将采集的影像数据上传至管理服务器和训练服务器;
当判断为危险时,反馈分析结果至管理服务器,管理服务器执行预设的报警处理流程,将采集的影像数据上传至管理服务器和训练服务器。
优选地,在实施例1的视频监控系统开始工作或重新建立通信连接时,还需进行如下操作:
通过公钥实现视频预处理终端与管理服务器和/或训练服务器之间的认证;
通过密钥建立视频预处理终端与管理服务器和/或训练服务器之间的密钥协商。
例如,在管理服务器与视频预处理终端,或者训练服务器与视频预处理终端的双方认证通过后,管理服务器或训练服务器向视频预处理终端的摄像头发送影像数据请求信息,影像数据请求信息包括信令和经过哈希计算的密钥MAK;
视频预处理终端接收影像数据请求信息后,先验证密钥MAK,如果密钥MAK验证通过,那么还用于分两种情况向管理服务器/训练服务器发送信息:
第一种情况:若视频预处理终端不更新视频密钥VKEK,则利用管理服务器/训练服务器的公钥将视频密钥VKEK加密生成视频密钥加密密钥密文EVKEK,再将视频密钥加密密钥密文EVKEK、视频密钥加密密钥版本号VKEVVersion放到SDP信道里发送给管理服务器/训练服务器;
第二种情况:若视频预处理终端更新视频密钥VKEK,则利用管理服务器/训练服务器的公钥将视频密钥VKEK加密生成视频密钥加密密钥密文EVKEK,再将视频密钥加密密钥密文EVKEK、更新后的视频密钥加密密钥版本号VKEVVersion、经过哈希计算的密钥MAK发给管理服务器/训练服务器;当视频预处理终端收到信息后,对密钥MAK进行验证,验证通过后,再将视频密钥加密密钥密文EVKEK、视频密钥加密密钥版本号VKEVVersion放到SDP信道里发送给管理服务器/训练服务器。
视频预处理终端收到视频密钥加密密钥密文EVKEK、视频密钥加密密钥版本号VKEVVersion后,还对密钥MAK进行验证,验证通过后,验证通过后将验证回执返回给摄像头,那么与管理服务器/训练服务器的密钥协商成功。
在本申请一优选实施例中,所述视频监控分析方法还包括:对采集的影像数据通过视频密钥进行加密;将加密的影像数据上传至管理服务器和训练服务器。
本申请提供的一种视频监控分析系统和方法,通过进行加密传输,保证了采集到的视频不会在传输过程中被人轻易的窃取和利用,安全性较高。并且,本申请的视频监控分析系统通过深度学习,可以不断提升分析的准确性,比起传统的分析系统,分析更加准确,精准性获得了较大提升。
在本申请所提供的几个实施例中,应该理解到,所揭露的装置和方法,也可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的,例如,附图中的流程图和框图显示了根据本申请的多个实施例的装置、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上,流程图或框图中的每个方框可以代表一个单元、程序段或代码 的一部分,所述单元、程序段或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意,在有些作为替换的实现方式中,方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如,两个连续的方框实际上可以基本并行地执行,它们有时也可以按相反的顺序执行,这依所涉及的功能而定。也要注意的是,框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合,可以用执行规定的功能或动作的专用的基于硬件的系统来实现,或者可以用专用硬件与计算机指令的组合来实现。
另外,在本申请各个实施例中的各功能单元可以集成在一起形成一个独立的部分,也可以是各个单元单独存在,也可以两个或两个以上单元集成形成一个独立的部分。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Onl8 Memor8)、随机存取存储器(RAM,Random Access Memor8)、磁碟或者光盘等各种可以存储程序代码的介质。
需要说明的是,在本文中,诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来,而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上所述仅为本申请的优选实施例而已,并不用于限制本申请,对 于本领域的技术人员来说,本申请可以有各种更改和变化。凡在本申请的精神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本申请的保护范围之内。应注意到:相似的标号和字母在下面的附图中表示类似项,因此,一旦某一项在一个附图中被定义,则在随后的附图中不需要对其进行进一步定义和解释。

Claims (11)

  1. 一种视频监控分析系统,包括:管理服务器、训练服务器以及若干个视频预处理终端;所述管理服务器和所述训练服务器分别与若干个视频预处理终端通信连接;
    所述管理服务器用于控制和管理所述训练服务器和若干个所述视频预处理终端;
    所述训练服务器用于以影像数据作为训练样本,训练危险识别模型;
    所述视频预处理终端用于采集监控点的影像数据,依据所述危险识别模型分析采集的影像数据,并将分析结果反馈至所述管理服务器,将所述影像数据上传至所述训练服务器。
  2. 如权利要求1所述的视频监控分析系统,其中,所述视频预处理终端包括:摄像头、分析模块和视频加密模块;所述摄像头与所述分析模块通信连接;所述分析模块和所述视频加密模块通信连接;
    所述摄像头,用于采集监控点的影像数据,并将影像数据发送至所述分析模块和所述视频加密模块;
    所述分析模块用于依据所述危险识别模型分析影像数据,并将分析结果反馈至所述管理服务器;
    所述视频加密模块用于对采集的影像数据通过视频密钥进行加密处理,并将加密处理后的影像数据传输给训练服务器和管理服务器。
  3. 如权利要求2所述的视频监控分析系统,其中,所述管理服务器包括:管理模块和第一视频解密模块;所述第一视频解密模块与所述视频预处理终端中的视频加密模块通信连接;
    所述管理模块用于管理所述训练服务器和若干个所述视频预处理终端,并且当接收到反馈的分析结果为危险时,执行预设的报警处理流程;所述第一视频解密模块用于对来自于视频加密模块的视频数据进行解密。
  4. 如权利要求3所述的视频监控分析系统,其中,所述管理模块 还用于存储并管理所述摄像头的影像数据。
  5. 如权利要求2所述的视频监控分析系统,其中,所述训练服务器包括:训练模块和第二视频解密模块;所述第二视频解密模块与所述视频预处理终端中的视频加密模块通信连接;
    所述训练模块用于将分析结果为危险的影像数据作为训练样本,并通过深度信念网络对该训练样本进行训练,更新完善所述危险识别模型;
    所述第二视频解密模块用于对来自于视频加密模块的视频数据进行解密。
  6. 如权利要求2所述的视频监控分析系统,其中,所述视频预处理终端还包括第一认证模块和第一密钥协商模块;所述管理服务器还包括第二认证模块和第二密钥协商模块;所述第一认证模块和第二认证模块用于通过公钥实现所述视频预处理终端与所述管理服务器之间的认证;
    所述第一密钥协商模块和所述第二密钥协商模块用于通过密钥建立所述视频预处理终端与所述管理服务器之间的密钥协商;
  7. 如权利要求6所述的视频监控分析系统,其中,所述训练服务器还包括第三认证模块和第三密钥协商模块;
    所述第一认证模块和第三认证模块用于通过公钥实现所述视频预处理终端与所述训练服务器之间的认证;
    所述第一密钥协商模块和所述第三密钥协商模块用于通过密钥建立所述视频预处理终端与所述管理服务器之间的密钥协商。
  8. 一种基于权利要求1至7之中任意一项所述的视频监控分析系统的视频监控分析方法,其中,包括:
    通过视频预处理终端采集监控点的影像数据;
    对采集的影像数据依据危险识别模型进行分析识别判断:
    当判断为正常时,将采集的影像数据上传至管理服务器和训练服务器;
    当判断为危险时,反馈分析结果至管理服务器,管理服务器执行预设的报警处理流程,将采集的影像数据上传至管理服务器和训练服务器。
  9. 如权利要求8所述的视频监控分析方法,其中,所述视频监控分析方法还包括:
    通过公钥实现视频预处理终端与管理服务器和/或训练服务器之间的认证;
    通过密钥建立视频预处理终端与管理服务器和/或训练服务器之间的密钥协商。
  10. 如权利要求8所示的视频监控分析方法,其中,所述视频监控分析方法还包括:
    对采集的影像数据通过视频密钥进行加密;
    将加密的影像数据上传至管理服务器和训练服务器。
  11. 如权利要求10所示的视频监控分析方法,其中,所述视频密钥是可更新的。
PCT/CN2020/105183 2020-07-21 2020-07-28 一种视频监控分析系统和方法 WO2022016573A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010707531.X 2020-07-21
CN202010707531.XA CN111818237A (zh) 2020-07-21 2020-07-21 一种视频监控分析系统和方法

Publications (1)

Publication Number Publication Date
WO2022016573A1 true WO2022016573A1 (zh) 2022-01-27

Family

ID=72861587

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/105183 WO2022016573A1 (zh) 2020-07-21 2020-07-28 一种视频监控分析系统和方法

Country Status (2)

Country Link
CN (1) CN111818237A (zh)
WO (1) WO2022016573A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115797850A (zh) * 2023-02-06 2023-03-14 中国石油大学(华东) 基于视频流的油田生产安全预警分析系统

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6611206B2 (en) * 2001-03-15 2003-08-26 Koninklijke Philips Electronics N.V. Automatic system for monitoring independent person requiring occasional assistance
CN101501564A (zh) * 2006-08-03 2009-08-05 国际商业机器公司 具有组合视频和音频识别的视频监视系统和方法
US20110142233A1 (en) * 2009-12-14 2011-06-16 Electronics And Telecommunications Research Institute Server and camera for video surviellance system and method for processing events in the same system
CN108109385A (zh) * 2018-01-18 2018-06-01 南京杰迈视讯科技有限公司 一种输电线防外破的车辆识别与危险行为判别系统与方法
CN108830143A (zh) * 2018-05-03 2018-11-16 深圳市中电数通智慧安全科技股份有限公司 一种基于深度学习的视频分析系统
CN109544870A (zh) * 2018-12-20 2019-03-29 同方威视科技江苏有限公司 用于智能监控系统的报警判断方法与智能监控系统
EP3511862A1 (en) * 2018-01-12 2019-07-17 Qognify Ltd. System and method for dynamically ordering video channels according to rank of abnormal detection
CN110717433A (zh) * 2019-09-30 2020-01-21 华中科技大学 一种基于深度学习的交通违规分析方法及装置
CN110858288A (zh) * 2018-08-24 2020-03-03 中国移动通信集团浙江有限公司 一种异常行为的识别方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8761401B2 (en) * 2006-08-28 2014-06-24 Motorola Mobility Llc System and method for secure key distribution to manufactured products
CN107277456B (zh) * 2017-07-26 2020-04-17 北京计算机技术及应用研究所 一种基于Android设备的安全视频监控系统
CN109151508B (zh) * 2018-11-09 2020-12-01 北京京航计算通讯研究所 一种视频加密方法
CN109218825B (zh) * 2018-11-09 2020-12-11 北京京航计算通讯研究所 一种视频加密系统

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6611206B2 (en) * 2001-03-15 2003-08-26 Koninklijke Philips Electronics N.V. Automatic system for monitoring independent person requiring occasional assistance
CN101501564A (zh) * 2006-08-03 2009-08-05 国际商业机器公司 具有组合视频和音频识别的视频监视系统和方法
US20110142233A1 (en) * 2009-12-14 2011-06-16 Electronics And Telecommunications Research Institute Server and camera for video surviellance system and method for processing events in the same system
EP3511862A1 (en) * 2018-01-12 2019-07-17 Qognify Ltd. System and method for dynamically ordering video channels according to rank of abnormal detection
CN108109385A (zh) * 2018-01-18 2018-06-01 南京杰迈视讯科技有限公司 一种输电线防外破的车辆识别与危险行为判别系统与方法
CN108830143A (zh) * 2018-05-03 2018-11-16 深圳市中电数通智慧安全科技股份有限公司 一种基于深度学习的视频分析系统
CN110858288A (zh) * 2018-08-24 2020-03-03 中国移动通信集团浙江有限公司 一种异常行为的识别方法及装置
CN109544870A (zh) * 2018-12-20 2019-03-29 同方威视科技江苏有限公司 用于智能监控系统的报警判断方法与智能监控系统
CN110717433A (zh) * 2019-09-30 2020-01-21 华中科技大学 一种基于深度学习的交通违规分析方法及装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115797850A (zh) * 2023-02-06 2023-03-14 中国石油大学(华东) 基于视频流的油田生产安全预警分析系统

Also Published As

Publication number Publication date
CN111818237A (zh) 2020-10-23

Similar Documents

Publication Publication Date Title
DE102015215120B4 (de) Verfahren zur verwendung einer vorrichtung zum entriegeln einer weiteren vorrichtung
US20200372184A1 (en) Method and arrangement for detecting digital content tampering
CN109218825B (zh) 一种视频加密系统
WO2016149943A1 (zh) 图片管理方法以及图片同步方法
CN109151508B (zh) 一种视频加密方法
US11082731B1 (en) Privacy-preserving video analytics
US10311215B2 (en) Secure recording and rendering of encrypted multimedia content
CN101741898A (zh) 一种视频型安防系统中的监视方法
CN113875259B (zh) 用于安全视频帧管理的方法、计算设备和计算机可读介质
WO2017193949A1 (zh) 一种码流篡改监控方法、装置及通信系统
CN107302435A (zh) 身份信息处理方法、系统及其对应服务器
Puvvadi et al. Cost-effective security support in real-time video surveillance
CN106650372B (zh) 管理员权限的开通方法及装置
CN111062043A (zh) 基于边缘计算的医疗影像识别方法及系统
WO2022016573A1 (zh) 一种视频监控分析系统和方法
KR20170131785A (ko) 영상 보안 시스템에서 프라이버시 객체 마스킹 및 사용자 인증을 위한 키 관리 방법 및 이러한 방법을 사용하는 영상 보안 시스템
CN111865761A (zh) 一种基于区块链智能合约的社交聊天信息存证方法
CN114640487B (zh) 一种避免中断视频监控设备运行的gb35114标准实时检测系统和方法
Li et al. Application Model of Video Surveillance System Interworking Based on Blockchain
CN110059630B (zh) 隐私保护的可验证外包监控视频行人再识别方法
CN114218603A (zh) 基于国产cpu和os利用区块链保存数据的方法及系统
CN113949848A (zh) 数据传输方法、装置、设备及介质
CN112714358A (zh) 一种视频流媒体加解密传输方法及系统
JP2022013298A (ja) 情報処理システム
KR20230166346A (ko) 멀티속성기반의 객체 재인식 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20946086

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20946086

Country of ref document: EP

Kind code of ref document: A1