CN104966006A - Intelligent face identification system based on cloud variation platform - Google Patents
Intelligent face identification system based on cloud variation platform Download PDFInfo
- Publication number
- CN104966006A CN104966006A CN201510243929.1A CN201510243929A CN104966006A CN 104966006 A CN104966006 A CN 104966006A CN 201510243929 A CN201510243929 A CN 201510243929A CN 104966006 A CN104966006 A CN 104966006A
- Authority
- CN
- China
- Prior art keywords
- cloud
- platform
- service
- data
- management
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2471—Distributed queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2221/00—Indexing scheme relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F2221/21—Indexing scheme relating to G06F21/00 and subgroups addressing additional information or applications relating to security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F2221/2141—Access rights, e.g. capability lists, access control lists, access tables, access matrices
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Bioethics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an intelligent face identification system based on a cloud variation platform. The system comprises a cloud variation management platform, a front end data acquisition subsystem, a cloud data platform and a cloud variation intelligent face identification platform. The system has high-performance processing efficiency, high reliable security strategy, completely ensured concurrent processing capacity, flexible and convenient system configuration, simple and easy-to-use data extraction strategy, and is easy to install and maintain, update, popularize, and use. The system is comprehensively optimized on hardware equipment, software system architecture, a server cluster, a database platform, and an application server platform and other aspects in advancement, stability, security, openness, expandability, and other aspects.
Description
Technical field
the present invention relates to a kind of intelligent face recognition system, relate to a kind of intelligent face recognition system of the platform that flattens based on cloud particularly.
Background technology
under global network environment, understanding three angles of the integrated framework of enterprise, element and relation each other thereof are set out, condensed go out five critical support technology: cloud agent skill group, SOA, Modeling Theory, cloud computing and semantic technology.Take SOA as primary structure framework, rely on modeling technique to realize the integrated data center of public security, cloud agency with the basis of cloud computing integrate public security software and hardware resources, relies on semantic technology to realize new Police Information interactive mode.
the cloud platform that flattens is the dynamic calculation platform technology that grows up in cloud agency's (Cloud Agent) technology and general fit calculation (Ubiquitous Computing) basis.Cloud becomes technical research for magnanimity information collection, transmission, storage and the efficiency involved by analytical calculation, the Distributed Computing Technology of availability and reliability method.It has employed cloud agency (Cloud Agent) technology and carries out the virtual real-time promoting magnanimity node system of dynamic depth to various information equipment, controllability, calculate and transfer efficiency, thus the system with magnanimity Internet of things node can more efficiently and more reliably be run.
cloud becomes the problem that technology not only solves concurrency parallel computation, and more solve the problem of software dynamic structure, resource is more easily shared, and resource more easily optimizes use, ensure that use the unified management of existing resource and Future from structure.The availability of prior art and WeiLai Technology is ensure that from design.Both can meet current needs, can the future development of easy expansion coupling need again, more can meet large-scale application requirement.
cloud becomes technology easily can carry out universal remote autonomous management and maintenance by units at different levels computer resource, and resource can be allocated mutually, and resource utilization is maximized.Both solve each school ability of maintenance weak, solve again the waste that constituent parts Informationization Equipment use level inequality causes.
cloud flatten platform provide one imitate the mankind obtain, accumulation and working knowledge platform.For the necessary self study of intelligent system and adaptation function provide the foundation.
one of key problem in technology of recognition of face allows authentication information not suffer a loss to low n-dimensional subspace n variable drop at higher dimensional space by feature extracting method.Thereby produce linear (principal component analysis (PCA), linear discriminatory analysis) and non-linear (kernel method, manifold learning) two class methods.The recognition of face of cloud connection science and technology have employed nonlinear method, and its principal feature is:
the cloud agency becoming technology based on cloud can make face characteristic extract, aspect ratio pair, and the speed of face characteristic storehouse training significantly improves.
intelligent learning method based on artificial neural network (ANN) can improve Face datection precision fast by the training of a large amount of machines and study.
adaboost Face datection algorithmic method based on Haar-like feature more can by training and learning the Face datection precision that adaptively selected optimal combination of characters reaches 95%, flatten on the basis of platform at cloud, the detection speed of piece image is also less than part second.
principal component analysis (PCA) (or claiming Karhunen – Loeve conversion) method makes Face datection precision can reach more than 90% on the standard picture storehouse based on front face.
Summary of the invention
the present invention seeks to: the intelligent face recognition system that a kind of platform that flattens based on cloud is provided, the concurrent processing ability there is high performance treatment effeciency, highly reliably security strategy, ensureing completely, flexible system configuration, the data pick-up strategy be simple and easy to, is easy to installation and maintenance, upgrading, popularization, use.
technical scheme of the present invention is:
flatten based on cloud the intelligent face recognition system of platform, comprises cloud and become management platform, front end data acquisition subsystem, cloud data platform and cloud and become Intelligent human-face Identification platform;
described cloud change management platform comprises management server, monitoring module and cloud proxy server for managing; Described management server comprises cloud proxy management module, resource management module, performance monitoring module, safety management module, exception management module, alarm management module, user management module, Role Management module and device management module; Described monitoring module is used for registering to centring system, download the business of disposing with implementation center's system, implementation center's system is to its traffic scheduling, and the order of load platform, resource and equipment control, uploads the control task of service executing result data and the business of execution; Described cloud proxy server, for managing cloud agency, realizes cloud agency of trademark registration, audit, storage, request response and distribution;
described front end data acquisition subsystem is for gathering video data;
described cloud data platform comprises message server, Streaming Media processing server, data cloud service and data access servers; Described message server is used for communication between each sensing layer hardware device and the many covers application layer operation system being deployed in zones of different; Described Streaming Media processing server is used for forwarding and distribution flow media data and Video processing in parallel agency to Video processing; Described data cloud service comprises the cloud service of video image analysis class, and the cloud service of described video image analysis class is for monitoring video camera apparatus and providing the Automatic face recognition service to real-time video and video segment; Described data access servers provides the data-interface service of standard, for application provides data;
described cloud becomes Intelligent human-face Identification platform and comprises the portrait Compare System operated in based in the Distributed Computing Platform of cloud change technology, and described portrait Compare System is for extracting portrait feature and carrying out matching identification.
preferably, described cloud becomes management platform and also comprises for providing clock synchronous interface to all networked devices and the system time that can correct management and for managing system cloud gray model daily record and Operation Log and providing the log management module of query function.
preferably, described cloud proxy management module comprises cloud agency of trademark registration, cloud agency audit, cloud agency distribution use; Described cloud agency of trademark registration provides the registration interface of standard for Connection Service pond, and registration process comprises the type selecting service, the information of service and the resource of service; Management server provides the unified access interface of service and authentication information to use for cloud agency audit and cloud agency distribution.
preferably, described resource management module comprises resource performance monitoring, scheduling of resource, resource capacity expansion and resource optimization.
preferably, described safety management module comprises user identity management and access control, described user identity management is used for user identity identification and certification and user authority management, and described access control comprises can the client ip address Range Access in connection data storehouse to control and to sensitive data access control.
preferably, the cloud service of described video image analysis class comprises video quality detection module and face recognition module, described video quality detection module is used for video camera apparatus and carries out poll detection to its Working Status Monitoring, Timeliness coverage connectivity problem, picture signal problem, the video quality detection module of image quality issues; Described face recognition module is for providing the Automatic face recognition service of real-time video and video segment.
advantage of the present invention is:
1. this system concurrent processing ability of having high performance treatment effeciency, highly reliably security strategy, ensureing completely, flexible system configuration, the data pick-up strategy be simple and easy to, is easy to installation and maintenance, upgrading, popularization, use.This system is at hardware device, software system architecture, server cluster, and the aspect such as database platform, applied server platform is comprehensively optimized from aspects such as advance, stability, security, opening, extensibilities.
2. system has extensibility, realizes the modularization of overall system architecture, and ensure that software and hardware and relevant environment are with good expansibility, under meeting Future Services Development, system stability is upgraded fast; Possess the security mechanism of height, ensure the safety of business and data, safe and reliable recovery and back mechanism must be possessed simultaneously; The technological means such as the automatic distribution of destructuring NoSQL and multi-duplicate technology, technology of data copy, disaster tolerance are adopted to ensure the fast quick-recovery of the reliability of data, integrality and data; The reliability of single equipment refers to select the equipment with high reliability technology, ensures less trouble.Particularly disk array, must adopt Redundancy Design, without Single Point of Faliure; Utilize distributed computing technology, by a large amount of server ratio of components, parallel computation is carried out to cluster, build magnanimity static person as the process of image data and analysis platform with this, provide magnanimity static person as the high performance ratio of data to analytical calculation ability to user.
3. cloud becomes each physical machine in system architecture is a node, service module on each machine is the Agent in cloud administration and supervision authorities, can carry out adjusting and change it dynamically flexibly and run formula, thus the computing ability of raising cloud service, fault-tolerant ability and reduction single-point collapse equivalent risk.
4. cloud change system has cloud computing dynamic self-adapting load balance ability; Dynamic assignment calculation task, raising system availability, handling capacity, Strengthens network processing power, shortening task processing time.
5. all node serve that cloud becomes in system architecture are all monitored by cloud administration and supervision authorities, cloud administration and supervision authorities can according to the loading condition of service node and node tasks processing power dynamic migration task in operational process, each service agent baseload is balanced, thus effectively improves system performance.
6. cloud act on behalf of the scheduling feature of Agent node can according to the request of certain tactful Resources allocation.Utilize the data structures such as many queues by the request buffer memory of user to infrastructure service, then carry out allocation process according to certain policing algorithm.
7. data access has to pass through the access rights flow process that cloud becomes administration and supervision authorities, and be may have access to called by application, approval, configuration rear, cloud administration and supervision authorities can be supervised the process of data access simultaneously.
Accompanying drawing explanation
below in conjunction with drawings and Examples, the invention will be further described:
fig. 1 the present invention is based on cloud to flatten the logical organization block diagram of intelligent face recognition system of platform;
fig. 2 the present invention is based on cloud to flatten the network structure block diagram of intelligent face recognition system of platform;
fig. 3 the present invention is based on cloud to flatten the network topological diagram of intelligent face recognition system of platform;
fig. 4 the present invention is based on cloud to flatten the software flow pattern of intelligent face recognition system of platform.
Embodiment
for making the object, technical solutions and advantages of the present invention clearly understand, below in conjunction with embodiment also with reference to accompanying drawing, the present invention is described in more detail.Should be appreciated that, these describe just exemplary, and do not really want to limit the scope of the invention.In addition, in the following description, the description to known features and technology is eliminated, to avoid unnecessarily obscuring concept of the present invention.
embodiment:
as shown in Figure 1, three levels can be divided into based on the flatten intelligent face recognition system framework of platform of cloud, be respectively and gather Access Layer, cloud service platform and service application layer (application centring system).Gather Access Layer mainly to the docking of portrait photo and the access of portrait feature modeling multi-source, system support sets up portrait feature database from relevant database with based on importing portrait photo file system architecture unstructured data storehouse.And cloud becomes the core that public security cloud service platform is service layer, it assume responsibility for the access of portrait picture, portrait modeling, distributed Identification of Images comparison, the calculation tasks such as portrait signature analysis.
as Figure 1-4, the composition of system is divided into following components: cloud becomes management platform, front end data acquisition subsystem, cloud data platform, cloud change Intelligent human-face Identification platform.
system support hardware platform comprises: computer network system (the above trunk of gigabit); Server cluster (according to video camera how many easily extensible) and database.
this system composition is except meeting system to counting yield, outside the demand in calculating and transmission time, more consider the efficiency when system extension in the future, query time and the problem associated with other data sources, flatten on the basis of platform at cloud, build minimum system thus save primary construction expense, but structurally still there is cloud become expansion structure, in the data platform construction in future, associated support platform is organically integrated, form a complete cloud data platform.
front end data acquisition subsystem, can for video camera at the scene be respectively for gathering video data or other collecting device.
one, cloud change intelligent management platform comprises management server, monitoring module and the cloud proxy server for managing.
1 management server (Admin Server) comprises cloud proxy management module, resource management module, performance monitoring module, safety management module, exception management module, alarm management module, user management module, Role Management module, device management module, system time manages, log management module.
equipment involved by cloud computing, server and Internet resources huger, the Appropriate application of system resource is the key problem of cloud computing.Magnanimity calculates, concurrency requirement of real-time is higher, and this system adopts famous cloud changes technology, system resource clustering is managed achieve the high concurrent requirement with real-time.On this basis, system is also optimized the utilization factor of equipment and computational resource, thus greatly saves hardware cost, improves the efficiency of hardware device.The another one advantage adopting cloud to become technology makes complicated system maintenance work become robotization more.
1.1 cloud proxy management modules
1.1.1 cloud agency of trademark registration
admin Server should provide the registration interface of standard, so that the service of our exploitation and the respective services of User Exploitation can use unified interface to be registered in the service pool (service drawer) of Admin Server.Registration process comprises the type selecting service, is application service or data, services or infrastructure services; Also to comprise the essential information such as reference address, agreement, unique identification, title of service.In corresponding type, every service also needs to tell that this service of Admin Server needs to possess which resource, so Admin Server can judge whether oneself has these resources, but does not need to be concerned about to the current behaviour in service of resource.
for application service, if resource can be used, then complete registration, be designated available; If resource is unavailable, then registration failure.
1.1.2 cloud agency audit
for data, services and infrastructure services, application examination & verification is by rear, and Admin Server can provide unified access interface and the authentication information of service, and user can use service by unified service access interface and access mode.
1.1.3 cloud agency distribution uses
for data, services and infrastructure services, application examination & verification is by rear, and Admin Server can provide unified access interface and the authentication information of service, and user can use service by unified service access interface and access mode.
1.2 resource management module
1.2.1 resource performance monitoring
resource performance monitors that main monitoring host computer performance, disk size, network traffics etc. comprise CPU, internal memory, JVM, I/O, database index of correlation, capacity of memory device, network interface card go out inbound traffics etc.; By patterned mode, represent the variation tendency of all kinds of monitor control index intuitively.
arranged by threshold values, the index that logarithm value transfinites is reported to the police simultaneously, facilitates user's Timeliness coverage performance bottleneck or Network Abnormal situation.
statistical study by the cycle is realized to monitored results, the rush hour of positioning system operation, low ebb time, reasonably serves distribution thus, namely meet the requirements for access of daily front end user, take into account again the video analysis on backstage, the services such as image retrieval.
1.2.2 scheduling of resource
cloud is become management platform and is managed all services by the mode of registration center.Supplier's (as: intelligent video analysis service, intrusion detection service etc.) of service and consumer's (as: front end types of applications, third-party access request etc.) of service are before being initiated, all need to carry out authentication registration to platform, not only increase level of security and authority definition, also can ensure the realization of service dispatch of overall importance simultaneously.
service dispatch of overall importance and Service Statistics (as: allocating object, service duration etc.) will be aggregated into Surveillance center, for service dilatation provides data foundation.
cloud becomes system provides perfect server hardware monitoring and the scheduling of service, to ensure that the performance of hardware device obtains maximum using, serves the recovery of dilatation or resource simultaneously dynamically, reach demand assigned object according to the result of monitoring.
1.2.3 resource capacity expansion
cloud becomes management platform provides level and smooth service dilatation strategy.Log-in protocol is passed through between all services and platform, realize real distributed and loose coupling mechanism, platform monitoring Central Radical factually time monitored results, reasonably arrange the information channel of service consumer and ISP and the selection mechanism based on load balancing.Simultaneously according to result and monitoring threshold values, in good time startup agent node, the dilatation as required that level and smooth carrying out is served or the recovery of Service Source.
1.2.4 resource optimization
cloud becomes the resource management of management platform and mainly includes storage resources, computational resource, the management of Internet resources and Optimized Operation.Support average and pooling of resources scheduling mode, and by the mode that idling-resource reclaims, carry out the optimization of resource.
1.3 performance monitoring module
platform carries out performance monitoring to equipment, server and network and shows.
1.4 safety management
1.4.1 user identity management
user identity identification: platform identity certification adopts double factor authentication, i.e. password and ID certification, supports client ip and MAC Address binding, identifies database user identity.; Or digital certificate authentication logs in, take client and MAC Address binding simultaneously.
user authority management: configure its access rights to sensitive data in database according to database user identity, priority assignation granularity can arrive table level, record level and field level.
support the authentication (CA) based on PKI art simultaneously.
1.4.2 access control
unified data resource sharing access interface is provided, realizes safe access control;
in database server side to can the client ip address scope in connection data storehouse to conduct interviews control;
classify to the sensitive data in whole data resource, mark and manage, management granularity can be divided into table level, record level and field level;
realize sensitive data access control: should allow according to the sensitivity label of data or refuse data to pass through, and database of record user is to the daily record of sensitive data access; When the identity of database user and authority undesirable time, adopt and to report to the police or blocking way realizes sensitive assets access security.
1.5 exception management
system was supported based on the time, the alarm inquiry of event and equipment, alarm statistics, the functions such as alarm elimination.
1.6 log management
platform daily record comprises system cloud gray model daily record and Operation Log.Status information and the time of origin such as device start, self-inspection, exception, fault, recovery, closedown in running log essential record system; Operation Log essential record operating personnel log in, time of logging off and main operational circumstances, and Operation Log content comprises the module of operation, the user of operation, running time, content of operation.
support log query function, user can carry out retrieving according to running time, information such as operation user, operational module etc. and check the detailed content of Operation Log; Also according to information query system running logs such as device type, equipment failure, time of origins, machine operation can be understood, support the functions such as inquiry, statistics, derivation, printing.
1.7 alarm management
system was supported based on the time, the alarm inquiry of event and equipment, alarm statistics, the functions such as alarm elimination.
1.8 user management
be entered in system by all user profile, form personnel base library, personnel manage according to department, and associate with system user, mate.
1.9 Role Management
dynamically create, delete and revise role, form System Privileges set, to distribute to user, reach the object of leading subscriber authority.
1.10 equipment control
login system administration interface, in equipment control, can find Back ground Information and the status information of all devices.
or the control point icon clicked in the GIS map of application interface, can check Back ground Information and the status information of equipment of this control point.
facility information comprises a position title, numbering, installation site, monitoring objective, camera type, brand, affiliated unit, IP address, channel number etc. and current device running status.
1.11 system time manages
platform has unified time management system, provides clock synchronous interface to all networked devices, comprises front end codec, DVR, each application server, memory device, client etc.Software can setup times calibration cycle flexibly, as 24 hours, 1 hour, 10 minutes.
2 monitoring modules (OSERVER)
oserver is that cloud becomes system equipment and serves the generic adaptation layer accessed, and it bears two roles---manage and communicate.To register to centring system on the one hand, download the business of disposing with implementation center's system, want implementation center's system to its traffic scheduling on the other hand, load platform, the order of resource and equipment control, also carries the data upload of service executing result and the control task of the business of execution.
3 cloud proxy servers (RESERVER)
cloud agency service is that cloud becomes system and pays the base unit that user uses, as long as make user when using service just as buying the application bought water power and be concerned about that oneself needs, and need not be concerned about service operation flow below and architecture.
service is dynamic concept, produces along with the request of user, completes and wither away along with task.The management to cloud agency is responsible for by cloud proxy server.Realize cloud agency of trademark registration, audit, store, request response and distribution.
two, cloud data platform
1 message server (XServer)
in cloud change system, it is the ingredient of transmission cloud to XServer.XServer is that cloud becomes the bridge connecting sensing layer (DC end) and application layer (operation system) in system platform, the network layer namely in cloud change system.
the application at all levels of the Internet of Things of present stage is scattered, and XServer will realize accessible communication between each sensing layer hardware device and the many covers application layer operation system being deployed in zones of different.
mXServer, as the special XServer of streaming media, in QOS management, in the application of the special scenes such as the allocated bandwidth of offline storage data and real-time Data Transmission process, provides high-quality service ability.
2 Streaming Media processing servers (MXServer)
mXServer is as the special XServer of streaming media, and it can also Video processing agency in parallel process video except streaming media data carry out forwarding and distribute, and is an efficient streaming media server.In QOS management, in the application of the special scenes such as the allocated bandwidth of offline storage data and real-time Data Transmission process, provide high-quality service ability.
3 data cloud services (CSERVICE)
3.1 video image analysis class cloud services
3.1.1 video quality detects
this service is carried out poll to video camera apparatus and is detected its Working Status Monitoring, Timeliness coverage connectivity problem, picture signal problem, and as flower screen and blank screen, image quality issues is as frame per second and image quality: fuzzy, mosaic, water mark.Also cradle head control and presetting bit are detected.
3.1.2 recognition of face
automatic face recognition service to real-time video and video segment is provided.
3.1.2.1 the basic technical indicator of face recognition system software
4 data access servers (DATAMART)
data access servers provides the data-interface service of standard, for application provides data.The data, services of this server supporting structure and non-structural data.
it can realize Information issued, upload, networking interface, realizes video monitoring system, face recognition system is synchronous with the portrait view data of public security bureau of Suzhou City.
the data-interface such as system image call, playing back videos, information inquiry is also provided, meets the overall application requirement of information sharing, mutual, interlock.
to other related application in city, as smart city also can provide relevant interface, meet the overall application requirement of its information sharing, mutual, interlock.
there is provided external interface, realize video, recording order, the control of upper wall and camera control etc. in native system, other infosystems of public security realize real-time video and history video recording operation like a cork by interface.
three, Intelligent human-face Identification platform
cloud management platform is deployed in central machine room, and system is managed the registration of algorithm, examination & verification by the resource management of cloud management platform; By the task management functions of cloud management platform, the scheduling of task and task run situation are supervised in overall process.Managed by the request process of Service Management to service.
multiple portrait alignment algorithm is deployed in distributed type assemblies environment, the mode that different human Facial Image Recognition Algorithm is acted on behalf of by cloud under the unified platform is run jointly, and the issue of uniform service interface encapsulation is carried out to algorithms of different, being published to internal services bus by portrait service externally provides unified interface to access.
it is the carrier that different people is run as program that cloud acts on behalf of OServer program, and cloud acting head is as the docking of control data of algorithm and cloud management platform and the docking of communication management data.It is separate operation that portrait algorithm routine and cloud act on behalf of OServer, and cloud Agent can not change to portrait algorithm routine, and the operation of portrait algorithm routine itself depends on first floor system environment and cloud Agent has nothing to do.But portrait algorithm routine itself cannot independent operating at this platform for the side of application provides service, the interface of portrait algorithm routine must meet native system platform and could be acted on behalf of OServer procedure identification by cloud for the interface specification of portrait service and take up operation.
system is based upon on multiple servers cluster, adopts distributed structure/architecture, and support that distributed storage is in calculating, dynamic adds server, improves system computing capacity and memory capacity.
portrait Compare System is by the Distributed Computing Platform that operates in based on cloud change technology, Distributed Computing Platform construction is the high performance computing service ability for large data, utilize the X86 server that multiple stage cost performance is higher, utilize distributed structure/architecture, build storage and the portrait contrast conting platform of magnanimity portrait data, provide high performance portrait analytical calculation ability to user.
portrait Compare System becomes resource management and the Real-time Task Dispatch ability of system cloud management platform by cloud, realizes horizontal extension and the deployment of simple and easy physics.When system cloud gray model task amount increase severely or current machine can not meet existing user than to request amount or alignment algorithm program requirements more multiple resource time, by increasing the mode of service, expansion physical computing nodes can be carried out in non-stop-machine situation.
when after increase physical server, first registration service type in cloud management platform, dissimilar service cloud management platform can act on behalf of working procedure by the different cloud of configuration automatically.System can specify this server to be used as contrast conting purposes to expand computational resource; Or distributed portrait data store increases storage space; Or internal data bus increases the network bandwidth, improve network data transmission amount.
mass data fast throughput is provided based on distributed type assemblies, solve conventional architectures poor performance, inefficient problem when carrying out mass data and analyzing, realize hundred million DBMS fuzzy queries can complete at short notice, enable the application of the types such as the large data search of portrait, compare of analysis really possess application in actual combat and be worth.
the construction of Distributed Computing Platform utilizes distributed computing technology, use a large amount of PC Server server to form cluster and carry out parallel computation, build process and the analysis platform of large data with this, utilize data mining and intellectual analysis to solve effective retrieval use problem of large data.
system's composition comprises comparison server, Web server, storage server etc., and adopt different configuration for different server, reasonable distribution calculates, and is stored in bandwidth resources.
dissimilar service cloud management platform can act on behalf of working procedure by the different cloud of configuration automatically.System can specify this server to be that computational resource is used as contrast conting; Or storage server user distribution formula portrait data store increases storage space; Or internal data bus is to increase the network bandwidth, improve network data transmission amount.
cloud becomes system and has cloud computing dynamic self-adapting load balance ability.By the serve resources management of cloud management platform, task management and internal data bus, that serves comparison for user or Third party system calls, and native system generates adaptive load balancing by acting on behalf of the performance monitoring data Real-time Feedback of OServer program to physical node according to the assessment of the task treatment capacity of each node and cloud.
system, to comparison task dynamic assignment computational resource, makes full use of idle machine resource, improves system availability, shortens the task processing time.By design Strengthens network processing power and the handling capacity of internal data bus.
support cluster server controls, and can control the start and stop of system service, and the interpolation of server node and deletion;
in the cloud management of native system, the application of service, approval and closedown are comprised for Service Management.Here service, without comprising comparison service, also comprises stores service, internal bus service, or even a certain concrete alignment algorithm in distributed type assemblies runs node.
native system is acted on behalf of OServer program by cloud and is responsible for monitoring the real-time running state of each physical node, to accepting number of request, CPU use amount, request process used time, internal memory use amount etc., carry out detailed record, when system exception, recording exceptional information is to daily record; Can guiding system key log fast, system exception is made mistakes investigation.
under the support of cloud agent skill group, when a machine goes wrong, another machine can replace, thus the normal operation of the system of guarantee.
become the management of intelligent management platform at cloud under, by the load factor of monitoring server, uniform distribution server load, maintenance system stability.
should be understood that, above-mentioned embodiment of the present invention only for exemplary illustration or explain principle of the present invention, and is not construed as limiting the invention.Therefore, any amendment made when without departing from the spirit and scope of the present invention, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.In addition, claims of the present invention be intended to contain fall into claims scope and border or this scope and border equivalents in whole change and modification.
Claims (6)
1. flatten based on cloud the intelligent face recognition system of platform, it is characterized in that, comprises cloud and become management platform, front end data acquisition subsystem, cloud data platform and cloud and become Intelligent human-face Identification platform;
Described cloud change management platform comprises management server, monitoring module and cloud proxy server for managing; Described management server comprises cloud proxy management module, resource management module, performance monitoring module, safety management module, exception management module, alarm management module, user management module, Role Management module and device management module; Described monitoring module is used for registering to centring system, download the business of disposing with implementation center's system, implementation center's system is to its traffic scheduling, and the order of load platform, resource and equipment control, uploads the control task of service executing result data and the business of execution; Described cloud proxy server, for managing cloud agency, realizes cloud agency of trademark registration, audit, storage, request response and distribution;
Described front end data acquisition subsystem is for gathering vedio data;
Described cloud data platform comprises message server, Streaming Media processing server, data cloud service and data access servers; Described message server is used for communication between each sensing layer hardware device and the many covers application layer operation system being deployed in zones of different; Described Streaming Media processing server is used for forwarding and distribution flow media data and Video processing in parallel agency to Video processing; Described data cloud service comprises the cloud service of video image analysis class, and the cloud service of described video image analysis class is for monitoring video camera apparatus and providing the Automatic face recognition service to real-time video and video segment; Described data access servers provides the data-interface service of standard, for application provides data;
Described cloud becomes Intelligent human-face Identification platform and comprises the portrait Compare System operated in based in the Distributed Computing Platform of cloud change technology, and described portrait Compare System is for extracting portrait feature and carrying out matching identification.
2. the intelligent face recognition system of the platform that flattens based on cloud according to claim 1, it is characterized in that, described cloud becomes management platform and also comprises for providing clock synchronous interface to all networked devices and the system time that can correct management and for managing system cloud gray model daily record and Operation Log and providing the log management module of query function.
3. the intelligent face recognition system of the platform that flattens based on cloud according to claim 1, is characterized in that, described cloud proxy management module comprises cloud agency of trademark registration, cloud agency audit, cloud agency distribution use; Described cloud agency of trademark registration provides the registration interface of standard for Connection Service pond, and registration process comprises the type selecting service, the information of service and the resource of service; Management server provides the unified access interface of service and authentication information to use for cloud agency audit and cloud agency distribution.
4. the intelligent face recognition system of the platform that flattens based on cloud according to claim 1, it is characterized in that, described resource management module comprises resource performance monitoring, scheduling of resource, resource capacity expansion and resource optimization.
5. the intelligent face recognition system of the platform that flattens based on cloud according to claim 1, it is characterized in that, described safety management module comprises user identity management and access control, described user identity management is used for user identity identification and certification and user authority management, and described access control comprises can the client ip address Range Access in connection data storehouse to control and to sensitive data access control.
6. the intelligent face recognition system of the platform that flattens based on cloud according to claim 1, it is characterized in that, the cloud service of described video image analysis class comprises video quality detection module and face recognition module, described video quality detection module is used for video camera apparatus and carries out poll detection to its Working Status Monitoring, Timeliness coverage connectivity problem, picture signal problem, the video quality detection module of image quality issues; Described face recognition module is for providing the Automatic face recognition service of real-time video and video segment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510243929.1A CN104966006A (en) | 2015-05-14 | 2015-05-14 | Intelligent face identification system based on cloud variation platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510243929.1A CN104966006A (en) | 2015-05-14 | 2015-05-14 | Intelligent face identification system based on cloud variation platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104966006A true CN104966006A (en) | 2015-10-07 |
Family
ID=54220044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510243929.1A Pending CN104966006A (en) | 2015-05-14 | 2015-05-14 | Intelligent face identification system based on cloud variation platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104966006A (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975948A (en) * | 2016-05-23 | 2016-09-28 | 南京甄视智能科技有限公司 | Cloud service platform architecture for face identification |
| CN106529494A (en) * | 2016-11-24 | 2017-03-22 | 深圳市永达电子信息股份有限公司 | Human face recognition method based on multi-camera model |
| CN106570087A (en) * | 2016-10-20 | 2017-04-19 | 中国电子科技集团公司第二十八研究所 | Distributed face recognition system |
| CN107343182A (en) * | 2017-08-20 | 2017-11-10 | 成都才智圣有科技有限责任公司 | Video analytic system based on cloud computing |
| CN108090419A (en) * | 2017-11-30 | 2018-05-29 | 安徽四创电子股份有限公司 | A kind of dynamic human face recognition methods based on WebSocket agreements |
| CN108229260A (en) * | 2016-12-21 | 2018-06-29 | 杭州海康威视系统技术有限公司 | A kind of identity information checking method and system |
| CN108810099A (en) * | 2018-05-22 | 2018-11-13 | 四川南亚丝路科技有限公司 | A kind of cell management system based on physical layering formula |
| CN108920924A (en) * | 2018-06-20 | 2018-11-30 | 甘肃万维信息技术有限责任公司 | A kind of data sharing method based on recognition of face |
| CN109165624A (en) * | 2018-09-10 | 2019-01-08 | 湖北众锐智能技术股份有限公司 | A kind of recognition of face analysis method based on public cloud |
| CN109523431A (en) * | 2018-09-29 | 2019-03-26 | 四川南亚丝路科技有限公司 | A kind of Property Management of residence security system and method |
| CN109614951A (en) * | 2018-12-27 | 2019-04-12 | 广东金杭科技有限公司 | Portrait compares task distribution processor algorithm |
| CN110097007A (en) * | 2019-04-30 | 2019-08-06 | 福建威盾科技集团有限公司 | A kind of method and device for integrating a variety of face recognition engines |
| CN111144384A (en) * | 2020-01-16 | 2020-05-12 | 四川天翼网络服务有限公司 | Police affair data processing system based on video structured application |
| CN111143594A (en) * | 2019-12-26 | 2020-05-12 | 北京橘拍科技有限公司 | Portrait searching method, server, storage medium, video processing method and system |
| CN111353057A (en) * | 2019-11-12 | 2020-06-30 | 北京中盾安全技术开发公司 | Face recognition service platform |
| CN111611973A (en) * | 2020-06-01 | 2020-09-01 | 广州市百果园信息技术有限公司 | Method, device and storage medium for identifying target user |
| CN111637335A (en) * | 2020-07-03 | 2020-09-08 | 国网安徽省电力有限公司培训中心 | Face recognition equipment and recognition system thereof |
| CN112783575A (en) * | 2021-02-03 | 2021-05-11 | 杭州海宴科技有限公司 | Plug-in type elastic cloud video image analysis system and method |
| CN112954034A (en) * | 2021-02-02 | 2021-06-11 | 苏州信颐系统集成有限公司 | Smart city public safety information management system based on cloud service |
| CN113098951A (en) * | 2021-03-30 | 2021-07-09 | 中电科航空电子有限公司 | Civil aircraft passenger cabin wireless network system and server software architecture thereof |
| CN114419717A (en) * | 2022-01-27 | 2022-04-29 | 睿云联(厦门)网络通讯技术有限公司 | Face detection and recognition acceleration method and system for terminal equipment |
| CN115495231A (en) * | 2022-08-09 | 2022-12-20 | 徐州医科大学 | Dynamic resource scheduling method and system under complex scene of high concurrent tasks |
| CN117408395A (en) * | 2023-12-14 | 2024-01-16 | 成都乐超人科技有限公司 | Method and device for optimizing running stability of wind control platform based on digital supply chain |
| CN117499163A (en) * | 2024-01-03 | 2024-02-02 | 成都数之联科技股份有限公司 | WebRTC-based server remote maintenance method, system and equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102055730A (en) * | 2009-11-02 | 2011-05-11 | 华为终端有限公司 | Cloud processing system, cloud processing method and cloud computing agent device |
| CN102222213A (en) * | 2010-07-29 | 2011-10-19 | 郑文明 | Distributed vision computing method based on open type Web Service framework |
| CN202870852U (en) * | 2012-08-01 | 2013-04-10 | 成都亿友科技有限公司 | Face identification system based on cloud computing |
| CN103839034A (en) * | 2012-11-22 | 2014-06-04 | 成都江法科技有限公司 | Security protection system based on facial identification cloud server |
| CN104284141A (en) * | 2013-07-08 | 2015-01-14 | 株式会社日立制作所 | Video monitoring system |
-
2015
- 2015-05-14 CN CN201510243929.1A patent/CN104966006A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102055730A (en) * | 2009-11-02 | 2011-05-11 | 华为终端有限公司 | Cloud processing system, cloud processing method and cloud computing agent device |
| CN102222213A (en) * | 2010-07-29 | 2011-10-19 | 郑文明 | Distributed vision computing method based on open type Web Service framework |
| CN202870852U (en) * | 2012-08-01 | 2013-04-10 | 成都亿友科技有限公司 | Face identification system based on cloud computing |
| CN103839034A (en) * | 2012-11-22 | 2014-06-04 | 成都江法科技有限公司 | Security protection system based on facial identification cloud server |
| CN104284141A (en) * | 2013-07-08 | 2015-01-14 | 株式会社日立制作所 | Video monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 相海瀛: "公安网吧数字视频监控系统的设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105975948B (en) * | 2016-05-23 | 2019-03-29 | 南京甄视智能科技有限公司 | Cloud service platform framework for recognition of face |
| CN105975948A (en) * | 2016-05-23 | 2016-09-28 | 南京甄视智能科技有限公司 | Cloud service platform architecture for face identification |
| CN106570087A (en) * | 2016-10-20 | 2017-04-19 | 中国电子科技集团公司第二十八研究所 | Distributed face recognition system |
| CN106570087B (en) * | 2016-10-20 | 2019-05-21 | 中国电子科技集团公司第二十八研究所 | One kind being based on distributed face identification system |
| CN106529494A (en) * | 2016-11-24 | 2017-03-22 | 深圳市永达电子信息股份有限公司 | Human face recognition method based on multi-camera model |
| CN108229260A (en) * | 2016-12-21 | 2018-06-29 | 杭州海康威视系统技术有限公司 | A kind of identity information checking method and system |
| CN108229260B (en) * | 2016-12-21 | 2020-12-29 | 杭州海康威视系统技术有限公司 | Identity information verification method and system |
| CN107343182A (en) * | 2017-08-20 | 2017-11-10 | 成都才智圣有科技有限责任公司 | Video analytic system based on cloud computing |
| CN108090419A (en) * | 2017-11-30 | 2018-05-29 | 安徽四创电子股份有限公司 | A kind of dynamic human face recognition methods based on WebSocket agreements |
| CN108090419B (en) * | 2017-11-30 | 2020-10-30 | 安徽四创电子股份有限公司 | Dynamic face recognition method based on WebSocket protocol |
| CN108810099A (en) * | 2018-05-22 | 2018-11-13 | 四川南亚丝路科技有限公司 | A kind of cell management system based on physical layering formula |
| CN108920924A (en) * | 2018-06-20 | 2018-11-30 | 甘肃万维信息技术有限责任公司 | A kind of data sharing method based on recognition of face |
| CN108920924B (en) * | 2018-06-20 | 2022-05-03 | 中电万维信息技术有限责任公司 | Data sharing method based on face recognition |
| CN109165624A (en) * | 2018-09-10 | 2019-01-08 | 湖北众锐智能技术股份有限公司 | A kind of recognition of face analysis method based on public cloud |
| CN109523431A (en) * | 2018-09-29 | 2019-03-26 | 四川南亚丝路科技有限公司 | A kind of Property Management of residence security system and method |
| CN109614951A (en) * | 2018-12-27 | 2019-04-12 | 广东金杭科技有限公司 | Portrait compares task distribution processor algorithm |
| CN110097007A (en) * | 2019-04-30 | 2019-08-06 | 福建威盾科技集团有限公司 | A kind of method and device for integrating a variety of face recognition engines |
| CN111353057A (en) * | 2019-11-12 | 2020-06-30 | 北京中盾安全技术开发公司 | Face recognition service platform |
| CN111143594A (en) * | 2019-12-26 | 2020-05-12 | 北京橘拍科技有限公司 | Portrait searching method, server, storage medium, video processing method and system |
| CN111144384A (en) * | 2020-01-16 | 2020-05-12 | 四川天翼网络服务有限公司 | Police affair data processing system based on video structured application |
| CN111611973A (en) * | 2020-06-01 | 2020-09-01 | 广州市百果园信息技术有限公司 | Method, device and storage medium for identifying target user |
| CN111637335A (en) * | 2020-07-03 | 2020-09-08 | 国网安徽省电力有限公司培训中心 | Face recognition equipment and recognition system thereof |
| CN112954034A (en) * | 2021-02-02 | 2021-06-11 | 苏州信颐系统集成有限公司 | Smart city public safety information management system based on cloud service |
| CN112783575A (en) * | 2021-02-03 | 2021-05-11 | 杭州海宴科技有限公司 | Plug-in type elastic cloud video image analysis system and method |
| CN113098951A (en) * | 2021-03-30 | 2021-07-09 | 中电科航空电子有限公司 | Civil aircraft passenger cabin wireless network system and server software architecture thereof |
| CN114419717A (en) * | 2022-01-27 | 2022-04-29 | 睿云联(厦门)网络通讯技术有限公司 | Face detection and recognition acceleration method and system for terminal equipment |
| CN115495231A (en) * | 2022-08-09 | 2022-12-20 | 徐州医科大学 | Dynamic resource scheduling method and system under complex scene of high concurrent tasks |
| CN115495231B (en) * | 2022-08-09 | 2023-09-19 | 徐州医科大学 | Dynamic resource scheduling method and system under high concurrency task complex scene |
| CN117408395A (en) * | 2023-12-14 | 2024-01-16 | 成都乐超人科技有限公司 | Method and device for optimizing running stability of wind control platform based on digital supply chain |
| CN117408395B (en) * | 2023-12-14 | 2024-04-02 | 成都乐超人科技有限公司 | Method and device for optimizing running stability of wind control platform based on digital supply chain |
| CN117499163A (en) * | 2024-01-03 | 2024-02-02 | 成都数之联科技股份有限公司 | WebRTC-based server remote maintenance method, system and equipment |
| CN117499163B (en) * | 2024-01-03 | 2024-03-01 | 成都数之联科技股份有限公司 | WebRTC-based server remote maintenance method, system and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104966006A (en) | Intelligent face identification system based on cloud variation platform | |
| US11637762B2 (en) | MDL-based clustering for dependency mapping | |
| US11924240B2 (en) | Mechanism for identifying differences between network snapshots | |
| US11470159B2 (en) | API key security posture scoring for microservices to determine microservice security risks | |
| CN104573611B (en) | A kind of distributed recognition of face group system | |
| US11044170B2 (en) | Network migration assistant | |
| CN106709003A (en) | Hadoop-based mass log data processing method | |
| Zhang et al. | A video cloud platform combing online and offline cloud computing technologies | |
| US10826803B2 (en) | Mechanism for facilitating efficient policy updates | |
| US10917438B2 (en) | Secure publishing for policy updates | |
| US11308114B1 (en) | Platform for provisioning a data analytics environment | |
| US9792334B2 (en) | Large-scale processing and querying for real-time surveillance | |
| CN103258027A (en) | Context awareness service platform based on intelligent terminal | |
| CN104834911A (en) | Portrait dynamic comparison system based on distributed cloud transformation platform | |
| US20230012641A1 (en) | Securing network resources from known threats | |
| US20210392135A1 (en) | Securing workload and application access from unauthorized entities | |
| CN113505260A (en) | Face recognition method and device, computer readable medium and electronic equipment | |
| Prakash et al. | Smart city video surveillance using fog computing | |
| EP4176568A1 (en) | Systems and methods for determining effectiveness of network segmentation policies | |
| Zhang et al. | Efficient online surveillance video processing based on spark framework | |
| US11765046B1 (en) | Endpoint cluster assignment and query generation | |
| CN111818310A (en) | Public safety management platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151007 |