CN109977806A - Realize the system and method for IDC computer room disengaging personnel's fine-grained management - Google Patents
Realize the system and method for IDC computer room disengaging personnel's fine-grained management Download PDFInfo
- Publication number
- CN109977806A CN109977806A CN201910180700.6A CN201910180700A CN109977806A CN 109977806 A CN109977806 A CN 109977806A CN 201910180700 A CN201910180700 A CN 201910180700A CN 109977806 A CN109977806 A CN 109977806A
- Authority
- CN
- China
- Prior art keywords
- computer room
- personnel
- subsystem
- room
- information
- 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 OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/758—Involving statistics of pixels or of feature values, e.g. histogram matching
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Alarm Systems (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of system for realizing the disengaging personnel's fine-grained management of IDC computer room, it is used to register including Login Register subsystem and carrys out the personnel of computer room for the first time and provide login service as computer room operator；Data base management subsystem is used to save the information and computer room information for logging in typing；Gate inhibition's room entry/exit management subsystem；Computer room analog map visual subsystem is used to show the location information in computer room by analog map；Personnel tracking subsystem in computer room；Alarm subsystem is used for the personnel that cross the border by combined aural and visual alarm warning.The invention further relates to a kind of methods for realizing the disengaging personnel's fine-grained management of IDC computer room based on the system.Computer vision technique is combined with depth learning technology using the system and method, can be crossed the border security risk with real-time early warning IDC computer room personnel's time-out delay computer room and position, it crosses the border to identification, time and position is crossed the border and accurately identified and auto-alarming, be able to satisfy the management requirement to IDC computer room disengaging personnel automation, standardization and fining.
Description
Technical field
The present invention relates to personal management fields, more particularly to according to the personal management of deep learning and computer vision technique
Field, in particular to a kind of system and method for realizing the disengaging personnel's fine-grained management of IDC computer room.
Background technique
With the development of Internet technology and big data technology, the number of servers of the computer room quantity and configuration of data center
It is sharply increased with data storage and the increase of data processing needs.Since the area of mono- computer room of IDC is huger, computer room
A large amount of server is inside distributed with, and these servers would generally rent different units and use, the server of some unit
Operator can only allow access into the server that designated room specifies region operation our unit to rent within a certain period of time.
How the multiple operators of large area zone of action are tracked automatically by the multi-cam in IDC computer room, such as
Automatically the overtime operator being detained or position is crossed the border of discovery in time, these difficulties all propose the management of IDC computer room personnel for what
Very big challenge is gone out.
The management of IDC computer room disengaging personnel mainly or by manually registration and certificate is verified at present, manually dig-ins machine
Video monitoring in room carries out tracing management.Such way to manage causes administrative staff's labor intensity high, due to automation
Management degree is low, therefore not only the efficiency of management is low, but also is easy to miss the supervision to security risk in IDC computer lab management.
Summary of the invention
The purpose of the present invention is overcoming the above-mentioned prior art, a kind of automation, standardization and fine are provided
The system and method for realizing the disengaging personnel's fine-grained management of IDC computer room of change.
To achieve the goals above, the system and method for realizing the disengaging personnel's fine-grained management of IDC computer room of the invention
It is as follows:
The system that realization IDC computer room passes in and out personnel's fine-grained management, is mainly characterized by, the system includes:
Login Register subsystem carrys out the personnel of computer room for the first time and provides login service as computer room operator for registering；
Data base management subsystem is connected with the Login Register subsystem, for saving the information for logging in typing
And computer room information；
Gate inhibition's room entry/exit management subsystem, is connected with the data base management subsystem, for according to recognition of face journey
Sequence opens computer room gate inhibition and captures and save human body image；
Computer room analog map visual subsystem, is connected with the data base management subsystem, for passing through simulation
Location information in map denotation computer room, and real-time video is retrieved for examination by each camera；
Personnel tracking subsystem in computer room is connected with the data base management subsystem, for passing through camera reality
When detection human body and extract information carry out similarity comparison；
Subsystem is alerted, is connected with personnel tracking subsystem in the computer room, for being alerted by combined aural and visual alarm
Cross the border personnel, and computer room analog map visual subsystem described in linkage shows the real-time video for position of crossing the border.
Preferably, the Login Register subsystem typing first log into the ID card No. of personnel, identity card picture and
Living body faces photo.
Preferably, the computer room information of the data base management subsystem typing includes computer room number, computer room entrance people
Face recognition access control system number, server in server room number and camera position number and machine in position and rental Unit code, computer room
Whole Body photo treated the provincial characteristics image that room entrance is captured.
Preferably, personnel tracking subsystem includes: in the computer room
Human body target detection module is connected with the data base management subsystem, for being passed through based on deep learning
Camera real-time monitoring human body；
Area image similarity comparison module, is connected with the data base management subsystem and human body target detection module
It connects, carries out histogram equalization processing for extracting block image based on computer vision technique, carry out cross-correlation calculation determination and take the photograph
As head institute tracking individuals identity, and determine whether this person is in out-of-range conditions.
Preferably, cross-correlation calculation is carried out in the area image similarity comparison module, specifically:
Cross-correlation calculation is carried out according to the following formula:
Wherein, A and B is respectively the YUV value or rgb value of A image and B image, and E (A) and E (B) are respectively the mean value of A and B,
A length of M of block image size, the width of block image size are that the range of N, m and n are respectively M and N.
The method for realizing the disengaging personnel's fine-grained management of IDC computer room based on above system, is mainly characterized by, described
Method the following steps are included:
(1) system typing personal information and computer room information described in simultaneously check identity information automatically, and system passes through simulation ground
Figure carries out visualization processing；
(2) gate inhibition's room entry/exit management subsystem described in captures personnel's photo and carries out recognition of face；
(3) system described in tracks personnel present position in computer room by camera, and according to data base management subsystem
Interior information to overtime delay personnel and position cross the border personnel carry out real-time early warning.
Preferably, the step (1) specifically includes the following steps:
(1.1) Login Register subsystem typing personal information described in simultaneously checks identity information automatically；
(1.2) the data base management subsystem typing computer room information described in；
(1.3) information in data base management subsystem is passed through simulation ground by the computer room analog map visual subsystem described in
Figure carries out visualization processing.
Preferably, the step (2) specifically includes the following steps:
(2.1) gate inhibition's room entry/exit management subsystem described in captures personnel's photo；
(2.2) gate inhibition's room entry/exit management subsystem described in carries out face knowledge according to the identity information in data base management subsystem
Not, judge whether it is the personnel for allowing access into the computer room, capture human figure if it is, opening computer room gate inhibition and triggering camera
Picture；Otherwise, continue step (2.1).
Preferably, the step (3) specifically includes the following steps:
(3.1) personnel tracking subsystem tracks personnel present position in computer room by camera in the computer room described in；
(3.2) in the computer room described in personnel tracking subsystem judge personnel whether the period of computer room and/or position with
The information recorded in data base management subsystem is not consistent, if it is, time of origin crosses the border and/or position is crossed the border；Otherwise, after
Continuous step (3.1)；
(3.2) the alarm subsystem described in alerts the personnel that cross the border, the simulation of computer room described in linkage by combined aural and visual alarm
Map visualization subsystem shows the real-time video for position of crossing the border.
Preferably, personnel tracking subsystem includes human body target detection module and area image similarity in the computer room
Comparison module, the step (3.1) specifically includes the following steps:
Human body target detection module described in (3.1.1) is based on deep learning and carries out human testing, and judges whether to detect
To human body, if it is, continuing step (3.1.2)；Otherwise, continue step (3.1.1)；
Area image similarity comparison module described in (3.1.2) extracts area image and carries out scaling processing；
Area image similarity comparison module described in (3.1.3) carries out histogram equalization processing to area image；
Area image similarity comparison module described in (3.1.4) is by area image and the data base management subsystem
Personnel's image that the entrance of middle preservation is captured carries out similarity comparison calculating, and determines the identity of tracking individuals.
Using the system and method for realizing the disengaging personnel's fine-grained management of IDC computer room of the invention, computer is regarded
Feel technology is combined with depth learning technology, even if the camera in IDC computer room captures the facial image less than complete display,
The whole-process automatic tracking IDC computer room of energy passes in and out personnel, is real-time and accurately tracked in computer room by the camera of dense deployment in computer room
More people present positions can be detained computer room according to database corresponding information with real-time early warning IDC computer room personnel's time-out and position is crossed the border
Security risk.IDC computer room provided by the invention disengaging personnel's fine-grained management method and system cross the border to identification, time and
Position, which is crossed the border, to be accurately identified and auto-alarming, is able to satisfy to IDC computer room disengaging personnel automation, standardization and fining
Management requires.
Detailed description of the invention
Fig. 1 is that IDC computer room passes in and out the composition schematic diagram of the system of personnel's fine-grained management for of the invention realizing.
Fig. 2 is that IDC computer room passes in and out the flow diagram of the method for personnel's fine-grained management for of the invention realizing.
Fig. 3 is that IDC computer room passes in and out the process signal of the personnel tracking of the method for personnel's fine-grained management for of the invention realizing
Figure.
Specific embodiment
It is further to carry out combined with specific embodiments below in order to more clearly describe technology contents of the invention
Description.
The system of realization IDC computer room disengaging personnel's fine-grained management of the invention, including:
Login Register subsystem carrys out the personnel of computer room for the first time and provides login service as computer room operator for registering；
Data base management subsystem is connected with the Login Register subsystem, for saving the information for logging in typing
And computer room information；
Gate inhibition's room entry/exit management subsystem, is connected with the data base management subsystem, for according to recognition of face journey
Sequence opens computer room gate inhibition and captures and save human body image；
Computer room analog map visual subsystem, is connected with the data base management subsystem, for passing through simulation
Location information in map denotation computer room, and real-time video is retrieved for examination by each camera；
Personnel tracking subsystem in computer room is connected with the data base management subsystem, for passing through camera reality
When detection human body and extract information carry out similarity comparison；
Subsystem is alerted, is connected with personnel tracking subsystem in the computer room, for being alerted by combined aural and visual alarm
Cross the border personnel, and computer room analog map visual subsystem described in linkage shows the real-time video for position of crossing the border.
As the preferred embodiment of the present invention, the Login Register subsystem typing first logs into the identity card of personnel
Number, identity card picture and living body faces photo.
As the preferred embodiment of the present invention, the computer room information of the data base management subsystem typing includes computer room
It is taken the photograph in number, computer room entrance recognition of face gate inhibition number, server in server room number and position and rental Unit code, computer room
The Whole Body photo captured as head Position Number and computer room entrance treated provincial characteristics image.
As the preferred embodiment of the present invention, personnel tracking subsystem includes: in the computer room
Human body target detection module is connected with the data base management subsystem, for being passed through based on deep learning
Camera real-time monitoring human body；
Area image similarity comparison module, is connected with the data base management subsystem and human body target detection module
It connects, carries out histogram equalization processing for extracting block image based on computer vision technique, carry out cross-correlation calculation determination and take the photograph
As head institute tracking individuals identity, and determine whether this person is in out-of-range conditions.
As the preferred embodiment of the present invention, cross-correlation meter is carried out in the area image similarity comparison module
It calculates, specifically:
Cross-correlation calculation is carried out according to the following formula:
Wherein, A and B is respectively the YUV value or rgb value of A image and B image, and E (A) and E (B) are respectively the mean value of A and B,
A length of M of block image size, the width of block image size are that the range of N, m and n are respectively M and N.
Of the invention method that the disengaging personnel's fine-grained management of IDC computer room should be realized based on above system, including with
Lower step:
(1) system typing personal information and computer room information described in simultaneously check identity information automatically, and system passes through simulation ground
Figure carries out visualization processing；
(1.1) Login Register subsystem typing personal information described in simultaneously checks identity information automatically；
(1.2) the data base management subsystem typing computer room information described in；
(1.3) information in data base management subsystem is passed through simulation ground by the computer room analog map visual subsystem described in
Figure carries out visualization processing；
(2) gate inhibition's room entry/exit management subsystem described in captures personnel's photo and carries out recognition of face；
(2.1) gate inhibition's room entry/exit management subsystem described in captures personnel's photo；
(2.2) gate inhibition's room entry/exit management subsystem described in carries out face knowledge according to the identity information in data base management subsystem
Not, judge whether it is the personnel for allowing access into the computer room, capture human figure if it is, opening computer room gate inhibition and triggering camera
Picture；
Otherwise, continue step (2.1)；
(3) system described in tracks personnel present position in computer room by camera, and according to data base management subsystem
Interior information to overtime delay personnel and position cross the border personnel carry out real-time early warning；
(3.1) personnel tracking subsystem tracks personnel present position in computer room by camera in the computer room described in；
Human body target detection module described in (3.1.1) is based on deep learning and carries out human testing, and judges whether to detect
To human body, if it is, continuing step (3.1.2)；Otherwise, continue step (3.1.1)；
Area image similarity comparison module described in (3.1.2) extracts area image and carries out scaling processing；
Area image similarity comparison module described in (3.1.3) carries out histogram equalization processing to area image；
Area image similarity comparison module described in (3.1.4) is by area image and the data base management subsystem
Personnel's image that the entrance of middle preservation is captured carries out similarity comparison calculating, and determines the identity of tracking individuals；
(3.2) in the computer room described in personnel tracking subsystem judge personnel whether the period of computer room and/or position with
The information recorded in data base management subsystem is not consistent, if it is, time of origin crosses the border and/or position is crossed the border；Otherwise, after
Continuous step (3.1)；
(3.2) the alarm subsystem described in alerts the personnel that cross the border, the simulation of computer room described in linkage by combined aural and visual alarm
Map visualization subsystem shows the real-time video for position of crossing the border.
In a specific embodiment of the invention, the present invention is based on artificial intelligence technologys, provide a kind of IDC computer room disengaging personnel
The method and system of fine-grained management.System is by personnel tracking in Login Register subsystem, gate inhibition's room entry/exit management subsystem, computer room
Subsystem, computer room analog map visual subsystem, alarm subsystem and data base management subsystem composition.
System numbers computer room and entrance recognition of face gate inhibition number, server in server room are numbered and position and hire order
Camera position number in position code, computer room entrance and computer room, in Login Register subsystem typing identity card and face characteristic
The computer room registered in information and its affiliated unit's code, Login Register subsystem passes in and out the time, computer room entrance camera is captured
Whole Body photo treated data inputs such as provincial characteristics image are to data base management subsystem.
IDC computer room provided by the invention passes in and out personnel's fine-grained management method, first with based on deep neural network
The human face photo pair that face recognition technology carries out living body faces and identity card face to Login Register personnel and provides with unit
Than verifying, entering in the computer room period in database registration after login authentication passes through, pass through the automatic gate inhibition of recognition of face
The camera for entering corresponding computer room, while triggering computer room entrance captures lower Whole Body photo, handles by computer vision
Afterwards, characteristics of human body's area image is saved to database.After someone enters computer room, examined using the target based on deep learning
Survey technology triggers the camera real-time perfoming human testing in computer room, once detect human body image, then using based on efficient
Computer vision processing technique, the characteristics of human body's area image captured with the camera by computer room entrance carry out similarity pair
Than to differentiate identity, then being numbered and position, disengaging time with camera position number, server in the computer room in database
And rent unit code information and compare, if information is not inconsistent in the position of this person's appearance, the period of appearance and database,
It then triggers alarm subsystem and pops up this road video in computer room analog map visual subsystem and checked for computer lab management personnel
Confirmation.
IDC computer room passes in and out the composition schematic diagram of personnel's fine-grained management system as shown in Figure 1, it is made of six parts.
First part's Login Register subsystem, 101 modules as shown in figure 1.When certain unit personnel needs IDC computer room to operate
When its server, it is necessary first to be logged in Login Register subsystem, if it is first logging into regard to first being registered, in identity
Automatic input ID card No. and identity card picture in card reader, and typing living body faces photo are demonstrate,proved, starting is based on deep learning
The face alignment algorithm routine of technology retains photo in the database to living body faces photo, identity card picture and unit
Identification is compared, once compare failure, then the artificial examination of starting.Typing simultaneously allows access into the period of computer room, Suo Youlu
Enter information preservation to data base management subsystem.
Second part data base management subsystem, 102 modules as shown in figure 1.Data base management subsystem is stepped in addition to saving
Record the information of registration subsystem typing, it is also necessary in typing computer room number and computer room entrance recognition of face gate inhibition number, computer room
Server number is shone with the Whole Body that camera position number, computer room entrance are captured in position and rental Unit code, computer room
Piece treated information such as provincial characteristics image.
Part III gate inhibition's room entry/exit management subsystem, 103 modules as shown in figure 1.It is embedded in this subsystem based on depth
The face recognition algorithms program of study starts recognition of face when living body faces are directed at the Face datection window of this system automatically
Program is opened this computer room gate inhibition, is opened simultaneously when face alignment verification is consistent by the period that rear and this person enters this computer room
The Whole Body video camera of motivation door mouth captures Whole Body photo and after treatment saves provincial characteristics image to data
Library management subsystem, the characteristics of human body's area image into computer room entrance captured here will be used for personnel tracking subsystem in computer room
System.
Part IV computer room analog map visual subsystem, 104 modules as shown in figure 1.It can will be in each computer room
The information such as server location, camera position, passageway position are shown by way of analog map, visual in analog map
In sub-systems, the camera for clicking some position can directly retrieve for examination real-time video.Such as there is alarm signal in alarm subsystem,
The pop-up that can link in computer room analog map visual subsystem responds the real-time video of position.
Personnel tracking subsystem in Part V computer room, 105 modules as shown in figure 1, this is core of the invention module.This
It include the human body target detection module based on depth learning technology and the administrative division map based on computer vision technique in subsystem
As similarity comparison module.After there are personnel to enter some computer room, the human body mesh of all cameras in this computer room will be triggered
Mark real-time detection module extracts the block image in Bounding Box after detecting human body, will be in Bounding Box
The size scaling of block image be that M pixel is long and N pixel is wide (i.e. M × N size), in order to reduce the differences of illumination intensities of shooting image
Influence, to M × block S image carry out histogram equalization processing, treated in this way, and block image is set as A image；Equally
Method is handled the Whole Body image that computer room doorway is captured is entered, and is detected human body first, is extracted Bounding
Block image in Box, then it is scaled M × N size, then carry out histogram equalization processing, treated in this way, and block image is set
For B image.Then computing cross-correlation, the formula of computing cross-correlation are carried out to Y, U of A and B image, V component (or R, G, B component)
It is as follows:
Wherein A, B respectively represent the YUV value (or rgb value) of A, B image, and E (A), E (B) respectively refer to the mean value of A, B, m, n's
Range is respectively M, N.The product of the cross-correlation coefficient of 3 components (YUV or RGB) is denoted as R, will be protected in A image and database
The characteristic area B image into all personnel in this computer room deposited carries out computing cross-correlation, maximum former according to similarity (R value)
Reason can determine tracked personnel identity.Pass through camera position information in inquiry database and personnel's entry time section letter again
Breath judges whether this person is in that position is crossed the border or the time crosses the border, and triggers alarm subsystem alarm if in out-of-range conditions.
Part VI alerts subsystem, 106 modules as shown in figure 1.When personnel tracking subsystem sends out assigned personnel in computer room
When position is crossed the border or the time crosses the border, the combined aural and visual alarm of triggering alarm subsystem causes the concern of computer lab management personnel, accuses simultaneously
Alert signal linkage triggering computer room analog map visual subsystem pop-up display is crossed the border the real-time video of position.
It is of the invention based on above system realize the disengaging personnel's fine-grained management of IDC computer room method as shown in Fig. 2, its
In, comprising the following steps:
Be personnel's Login Register process such as 201 modules in Fig. 2, by personnel identity demonstrate,prove number and identity card picture information,
Living body faces image rents Unit code, allows access into the data inputs such as computer room period to database, and checking body automatically
Part information.
Such as 202 modules in Fig. 2, to establish database process, other than the data input to database of 201 modules,
Typing computer room number, server location and corresponding rental Unit code, camera position, recognition of face gate inhibition is also needed to compile
Number, the letter such as computer room entrance camera number and its whole body human body photo provincial characteristics image after treatment captured
Breath.
Such as 203 modules in Fig. 2, it is computer room analog map visible process, is taken orally computer room according to information in database
The information such as device position, camera position, passageway position of being engaged in carry out analog map visualization processing, the wherein camera shooting in analog map
Head, which can be put to open, sees real-time video.
It is that entrance guard management and whole body photo capture process such as 204 modules in Fig. 2, when there is face alignment computer room doorway
When the Face datection frame of face recognition instrument, start face alignment program, numbered according to the face information library in database, computer room,
Personnel allow access into the information such as period starting access control and then open computer room if it is determined that allowing access into computer room personnel
Gate inhibition, while triggering doorway camera and capturing lower this person's whole body photograph image, this image is passed through into the human body based on deep learning
Detection Programmable detection goes out human region, is M × N size by image scaling in Bounding Box, in order to reduce different illumination
It influences, then histogram equalization processing is carried out to M × n-quadrant image, then this area image is saved to database as in computer room
Personnel tracking subsystem uses.
Such as 205 modules in Fig. 2, be in computer room whether the judgement process of someone, according to some computer room gate inhibition in database
Entrance personnel amount, which subtracts this machine room door nonexportation personnel amount and whether is more than or equal to 1, to be judged, if it is larger than or equal to 1, just
Judge someone in this computer room, enters 206 modules and carry out computer room personnel tracking process；Otherwise judge nobody in this computer room.
It is computer room personnel tracking process such as 206 modules in Fig. 2, includes the human body target inspection based on depth learning technology
Survey module and the area image similarity comparison module based on computer vision technique.Pass through the camera of dense deployment in computer room
More people present positions in computer room are real-time and accurately tracked, computer room and position can be detained to personnel's time-out according to database corresponding information
It sets security risk of crossing the border and carries out real-time early warning.It is illustrated in figure 3 personnel tracking flow chart in computer room.301 modules are based on depth
The human detection module of study；302 modules judge whether to detect human body, if detecting human body, enter 303 modules；
303 modules extract the block image in Bounding Box, and carry out scaling processing, by characteristics of human body's administrative division map
As zooming to M × N size；
304 modules carry out histogram equalization processing to the area image of M × N size；
The personnel's block image for entering this computer room that 305 modules capture the entrance saved in block image and database
Carry out similarity comparison operation；
306 modules determine the identity of tracking individuals according to similarity maximum principle.
Such as 207 modules in Fig. 2, for whether someone crosses the border the judgement process crossed the border with position the time in computer room, if machine
The record that the personnel of room personnel tracking subsystem tracking occurred in the period and database of this computer room is not inconsistent, then judges herein
Time of origin crosses the border；If there is the note in the position and database of this computer room in the personnel of computer room personnel tracking subsystem tracking
Record is not inconsistent, then judges that position occurs herein crosses the border.It crosses the border in case of the time or position cross-border phenomenon, then 208 alarm of starting is flowed
Otherwise journey is back to 206 modules and continues computer room personnel tracking process.
Such as 208 modules in Fig. 2, to touch alarm processing, when judgement is crossed the border or position there are personnel time in 207 modules
When crossing the border, starts acousto-optic warning instrument, while popping up the camera real-time video of the corresponding position in 203 modules, remind computer room pipe
Reason personnel manually verify security risk.
Using the system and method for realizing the disengaging personnel's fine-grained management of IDC computer room of the invention, computer is regarded
Feel technology is combined with depth learning technology, even if the camera in IDC computer room captures the facial image less than complete display,
The whole-process automatic tracking IDC computer room of energy passes in and out personnel, is real-time and accurately tracked in computer room by the camera of dense deployment in computer room
More people present positions can be detained computer room according to database corresponding information with real-time early warning IDC computer room personnel's time-out and position is crossed the border
Security risk.IDC computer room provided by the invention disengaging personnel's fine-grained management method and system cross the border to identification, time and
Position, which is crossed the border, to be accurately identified and auto-alarming, is able to satisfy to IDC computer room disengaging personnel automation, standardization and fining
Management requires.
In this description, the present invention is described with reference to its specific embodiment.But it is clear that can still make
Various modifications and alterations are without departing from the spirit and scope of the invention.Therefore, the description and the appended drawings should be considered as illustrative
And not restrictive.
Claims (10)
1. a kind of system for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that the system includes:
Login Register subsystem carrys out the personnel of computer room for the first time and provides login service as computer room operator for registering；
Data base management subsystem is connected with the Login Register subsystem, for saving the information and machine that log in typing
Room information；
Gate inhibition's room entry/exit management subsystem, is connected with the data base management subsystem, for being opened according to recognition of face program
It opens computer room gate inhibition and captures and save human body image；
Computer room analog map visual subsystem, is connected, for passing through analog map with the data base management subsystem
It shows the location information in computer room, and real-time video is retrieved for examination by each camera；
Personnel tracking subsystem in computer room is connected with the data base management subsystem, for being examined in real time by camera
It surveys human body and extracts information and carry out similarity comparison；
Subsystem is alerted, is connected with personnel tracking subsystem in the computer room, for being crossed the border by combined aural and visual alarm warning
Personnel, computer room analog map visual subsystem described in linkage show the real-time video for position of crossing the border.
2. the system according to claim 1 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
Login Register subsystem typing first log into ID card No., identity card picture and the living body faces photo of personnel.
3. the system according to claim 1 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
Data base management subsystem typing computer room information include computer room number, computer room entrance recognition of face gate inhibition number, computer room
The Whole Body that server number is captured with camera position number in position and rental Unit code, computer room and computer room entrance
Photo treated provincial characteristics image.
4. the system according to claim 1 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
Computer room in personnel tracking subsystem include:
Human body target detection module is connected with the data base management subsystem, for passing through camera shooting based on deep learning
Head real-time monitoring human body；
Area image similarity comparison module, is connected with the data base management subsystem and human body target detection module,
Histogram equalization processing is carried out for extracting block image based on computer vision technique, cross-correlation calculation is carried out and determines camera
Institute's tracking individuals identity, and determine whether this person is in out-of-range conditions.
5. the system according to claim 4 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
Area image similarity comparison module in carry out cross-correlation calculation, specifically:
Cross-correlation calculation is carried out according to the following formula:
Wherein, A and B is respectively the YUV value or rgb value of A image and B image, and E (A) and E (B) are respectively the mean value of A and B, block
A length of M of picture size, the width of block image size are that the range of N, m and n are respectively M and N.
6. a kind of method for realizing the disengaging personnel's fine-grained management of IDC computer room based on system described in claim 1, feature exist
In, the method the following steps are included:
(1) system typing personal information and computer room information described in simultaneously check identity information automatically, system by analog map into
Row visualization processing；
(2) gate inhibition's room entry/exit management subsystem described in captures personnel's photo and carries out recognition of face；
(3) system described in tracks personnel present position in computer room by camera, and according to believing in data base management subsystem
Breath to overtime delay personnel and position cross the border personnel carry out real-time early warning.
7. the method according to claim 6 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
The step of (1) specifically includes the following steps:
(1.1) Login Register subsystem typing personal information described in simultaneously checks identity information automatically；
(1.2) the data base management subsystem typing computer room information described in；
(1.3) computer room analog map visual subsystem described in by information in data base management subsystem by analog map into
Row visualization processing.
8. the method according to claim 6 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
The step of (2) specifically includes the following steps:
(2.1) gate inhibition's room entry/exit management subsystem described in captures personnel's photo；
(2.2) gate inhibition's room entry/exit management subsystem described in carries out recognition of face according to the identity information in data base management subsystem,
Judge whether it is the personnel for allowing access into the computer room, captures human body image if it is, opening computer room gate inhibition and triggering camera；
Otherwise, continue step (2.1).
9. the method according to claim 6 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
The step of (3) specifically includes the following steps:
(3.1) personnel tracking subsystem tracks personnel present position in computer room by camera in the computer room described in；
(3.2) whether personnel tracking subsystem judges personnel in the period of computer room and/or position and data in the computer room described in
The information recorded in the management subsystem of library is not consistent, if it is, time of origin crosses the border and/or position is crossed the border；Otherwise, continue to walk
Suddenly (3.1)；
(3.2) the alarm subsystem described in alerts the personnel that cross the border, computer room analog map described in linkage by combined aural and visual alarm
Visual subsystem shows the real-time video for position of crossing the border.
10. the method according to claim 9 for realizing the disengaging personnel's fine-grained management of IDC computer room, which is characterized in that described
Computer room in personnel tracking subsystem include human body target detection module and area image similarity comparison module, the step
(3.1) specifically includes the following steps:
Human body target detection module described in (3.1.1) is based on deep learning and carries out human testing, and judges whether to detect people
Body, if it is, continuing step (3.1.2)；Otherwise, continue step (3.1.1)；
Area image similarity comparison module described in (3.1.2) extracts area image and carries out scaling processing；
Area image similarity comparison module described in (3.1.3) carries out histogram equalization processing to area image；
Area image similarity comparison module described in (3.1.4) will protect in area image and the data base management subsystem
Personnel's image that the entrance deposited is captured carries out similarity comparison calculating, and determines the identity of tracking individuals.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910180700.6A CN109977806A (en) | 2019-03-11 | 2019-03-11 | Realize the system and method for IDC computer room disengaging personnel's fine-grained management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910180700.6A CN109977806A (en) | 2019-03-11 | 2019-03-11 | Realize the system and method for IDC computer room disengaging personnel's fine-grained management |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109977806A true CN109977806A (en) | 2019-07-05 |
Family
ID=67078451
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910180700.6A Pending CN109977806A (en) | 2019-03-11 | 2019-03-11 | Realize the system and method for IDC computer room disengaging personnel's fine-grained management |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977806A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126219A (en) * | 2019-12-16 | 2020-05-08 | 国网浙江省电力有限公司电力科学研究院 | Transformer substation personnel identity recognition system and method based on artificial intelligence |
| CN111811658A (en) * | 2020-07-13 | 2020-10-23 | 深圳市多威尔科技有限公司 | A method and system for measuring body temperature through face recognition and infrared thermal imaging |
| CN116027725A (en) * | 2023-03-27 | 2023-04-28 | 深圳市森辉智能自控技术有限公司 | A group control optimization analysis system based on high-efficiency computer room |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101794382A (en) * | 2010-03-12 | 2010-08-04 | 华中科技大学 | Method for counting passenger flow of buses in real time |
| CN106096566A (en) * | 2016-06-21 | 2016-11-09 | 成都科曦科技有限公司 | A kind of security protection identification system and method supporting image typing retrieval, identification and track to describe |
| CN106682603A (en) * | 2016-12-19 | 2017-05-17 | 陕西科技大学 | Real time driver fatigue warning system based on multi-source information fusion |
| CN108171842A (en) * | 2017-12-28 | 2018-06-15 | 深圳市泛海三江科技发展有限公司 | A kind of personnel management system |
| CN108650485A (en) * | 2018-03-21 | 2018-10-12 | 江苏同袍信息科技有限公司 | Personnel monitoring's method and system based on image detection and wireless probe |
-
2019
- 2019-03-11 CN CN201910180700.6A patent/CN109977806A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101794382A (en) * | 2010-03-12 | 2010-08-04 | 华中科技大学 | Method for counting passenger flow of buses in real time |
| CN106096566A (en) * | 2016-06-21 | 2016-11-09 | 成都科曦科技有限公司 | A kind of security protection identification system and method supporting image typing retrieval, identification and track to describe |
| CN106682603A (en) * | 2016-12-19 | 2017-05-17 | 陕西科技大学 | Real time driver fatigue warning system based on multi-source information fusion |
| CN108171842A (en) * | 2017-12-28 | 2018-06-15 | 深圳市泛海三江科技发展有限公司 | A kind of personnel management system |
| CN108650485A (en) * | 2018-03-21 | 2018-10-12 | 江苏同袍信息科技有限公司 | Personnel monitoring's method and system based on image detection and wireless probe |
Non-Patent Citations (1)
| Title |
|---|
| 熊有伦 等: "《机器人学 建模、控制与视觉》", 华中科技大学出版社 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126219A (en) * | 2019-12-16 | 2020-05-08 | 国网浙江省电力有限公司电力科学研究院 | Transformer substation personnel identity recognition system and method based on artificial intelligence |
| CN111811658A (en) * | 2020-07-13 | 2020-10-23 | 深圳市多威尔科技有限公司 | A method and system for measuring body temperature through face recognition and infrared thermal imaging |
| CN116027725A (en) * | 2023-03-27 | 2023-04-28 | 深圳市森辉智能自控技术有限公司 | A group control optimization analysis system based on high-efficiency computer room |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9977971B2 (en) | Role-based tracking and surveillance | |
| CN106373240B (en) | A kind of intelligent monitor system and its monitoring method of self-service bank | |
| CN112233300A (en) | Community passing epidemic prevention monitoring joint defense system and method based on artificial intelligence | |
| CN109145789A (en) | Power supply system safety work support method and system | |
| CN106384285A (en) | Intelligent self-service bank system | |
| CN101556717A (en) | ATM intelligent security system and monitoring method | |
| CN110674761B (en) | A regional behavior early warning method and system | |
| CN108549854A (en) | A kind of human face in-vivo detection method | |
| CN108537922A (en) | Visitor's method for early warning based on recognition of face and system | |
| CN109558785A (en) | Safety defense monitoring system and the unmanned convenience store for applying it | |
| CN112132048A (en) | Community patrol analysis method and system based on computer vision | |
| RU2713876C1 (en) | Method and system for detecting alarm events when interacting with self-service device | |
| CN109977806A (en) | Realize the system and method for IDC computer room disengaging personnel's fine-grained management | |
| CN114821934A (en) | A kind of park perimeter security control system and method | |
| CN111064931A (en) | System and method for accurately identifying identity of dangerous person | |
| CN118917980A (en) | Community management system combining image analysis and big data deep learning | |
| Zin et al. | A Markov random walk model for loitering people detection | |
| CN109831634A (en) | The density information of target object determines method and device | |
| EP3109837A1 (en) | System and method of smart incident analysis in control system using floor maps | |
| CN109961588A (en) | A kind of monitoring system | |
| CN209044724U (en) | A kind of gate inhibition's net connection safety management system | |
| CN107368728A (en) | Visualize alarm management system and method | |
| CN118138726A (en) | Intelligent building security system and method | |
| CN119399662A (en) | Human-vehicle association method and retrieval matching method based on video surveillance | |
| Ho et al. | Public space behavior modeling with video and sensor analytics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for 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: 20190705 |