CN112258026B - Dynamic positioning scheduling method and system based on video identity recognition - Google Patents
Dynamic positioning scheduling method and system based on video identity recognition Download PDFInfo
- Publication number
- CN112258026B CN112258026B CN202011129592.9A CN202011129592A CN112258026B CN 112258026 B CN112258026 B CN 112258026B CN 202011129592 A CN202011129592 A CN 202011129592A CN 112258026 B CN112258026 B CN 112258026B
- Authority
- CN
- China
- Prior art keywords
- staff
- video
- identity recognition
- site
- dispatching desk
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000012544 monitoring process Methods 0.000 claims abstract description 31
- 238000013507 mapping Methods 0.000 claims abstract description 8
- 238000003062 neural network model Methods 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 6
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Security & Cryptography (AREA)
- Molecular Biology (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Tourism & Hospitality (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Signal Processing (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a dynamic positioning scheduling method and a system based on video identity recognition, wherein the method comprises the following steps: (1) acquiring an on-site video stream through video monitoring equipment; (2) When a worker enters the site for the first time, an appearance image is collected and used as a standard characteristic template of the worker; (3) Acquiring video stream information slices to form a plurality of image frames, and respectively inputting the image frames and a standard characteristic template into two input ends of an identity recognition model to obtain an identity recognition result of a worker; (4) Mapping the corresponding staff to a global map according to the identified corresponding staff, acquiring a dispatching desk closest to the staff, and establishing a staff-dispatching desk corresponding relation table to realize the positioning of all the staff; (5) When the staff needs to be called, the staff-dispatching desk corresponding relation table is searched, the dispatching desk corresponding to the staff is obtained, and the call is automatically forwarded to the dispatching desk. The invention can realize dynamic positioning and efficient and accurate dispatch calling.
Description
Technical Field
The invention relates to a scheduling technology, in particular to a dynamic positioning scheduling method and system based on video identity recognition.
Background
All functions in the traditional program control dispatching desk system depend on a dispatching exchanger, and the program control exchanger is limited by a technical system, has obvious defects of single service function, difficult system expansion, difficult quick deployment and the like, and is difficult to support a novel dispatching system with active safety, intelligent cooperative control, comprehensive balance and overall excellent performance. The IP intelligent dispatching desk system based on the IP technology realizes the automation and the intellectualization of dispatching voice stream, video stream and data stream fusion and dispatching command connection. And the digital relay is interconnected with the existing dispatching exchange network and administrative exchange network to realize bidirectional intercommunication with the dispatching exchange network and single intercommunication with the administrative exchange network (namely, the administrative exchange network user cannot actively call the dispatching desk system user), and the integrated butt joint is successfully realized with a dispatching operation ticket system, a regulation and control operation assistant, a voice recognition engine and other systems.
Although the whole function construction of the dispatching desk is finished at present, the corresponding application service is still relatively lacking, and how to reasonably fuse high-definition audio and video with on-site data streams is still to be developed for realizing more complex service functions. The existing scheduling system still stays on basic services such as video call and the like for the utilization of audio and video signals, and a large number of mature schemes in the related fields need to be referred to. On the other hand, the existing IP dispatcher station still makes a call to related personnel through a fixed dispatcher equipment, and the call to a specific dispatcher station is still poor in flexibility and low in execution efficiency although automatic forwarding is realized for the call to the specific personnel.
Disclosure of Invention
The invention aims to: aiming at the problems existing in the prior art, the invention provides a dynamic positioning scheduling method and a system based on video identity recognition, which can dynamically position staff and realize automatic calling.
The technical scheme is as follows: the dynamic positioning scheduling method based on video identity recognition comprises the following steps:
(1) The method comprises the steps that video stream information of a dispatching system site is collected in real time through video monitoring equipment and uploaded to a cloud storage server, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of obtaining monitoring video of each position in the site;
(2) When a worker in a dispatching system enters the site for the first time in a working period, acquiring an appearance image of the worker as a standard characteristic template of the current working period of the worker;
(3) Acquiring video stream information from a cloud storage server, slicing to form a plurality of image frames, respectively inputting the image frames and standard feature templates of all staff in the current working period of the staff into two input ends of an identity recognition model to obtain an identity recognition result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
(4) Mapping the corresponding staff to a global map according to the video monitoring equipment for acquiring the image frames and the identified corresponding staff, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff to realize the positioning of all the staff;
(5) When the staff needs to be called, the staff-dispatching desk corresponding relation table is searched, the dispatching desk corresponding to the staff is obtained, and the call is automatically forwarded to the dispatching desk.
Further, the method further comprises:
(6) And judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
Further, the training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
The dynamic positioning scheduling system based on video identity recognition comprises:
the cloud storage server is used for storing video stream information of a dispatching system site acquired in real time through video monitoring equipment, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of acquiring monitoring video of each position in the site;
the image collector is used for collecting an appearance image of a worker when the worker in the dispatching system enters the site for the first time in a working period and taking the appearance image as a standard characteristic template of the current working period of the worker;
the identification module is used for acquiring video stream information from the cloud storage server, slicing the video stream information to form a plurality of image frames, respectively inputting the image frames and standard characteristic templates of all staff in the current working period into two input ends of the identification model to obtain the identification result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
the positioning module is used for mapping the corresponding staff to the global map according to the image frames and the identified corresponding staff and the video monitoring equipment for acquiring the image frames, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff so as to realize the positioning of all the staff;
and the calling module is used for searching a personnel-dispatching desk correspondence table when the personnel need to be called, obtaining a dispatching desk corresponding to the personnel, and automatically forwarding the call to the dispatching desk.
Further, the method further comprises the following steps:
and the positioning legal judgment module is used for judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
Further, the training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
The beneficial effects are that: compared with the prior art, the invention has the remarkable advantages that: the invention carries out identity recognition through the image characteristics, carries out position mapping through the relation of the dispatching equipment and distributes the positioning result, thereby realizing the dynamic positioning of the staff and realizing higher efficiency of calling the staff.
Drawings
FIG. 1 is a block diagram of a dynamic positioning scheduling system based on video identification provided by the invention.
Detailed Description
The embodiment provides a dynamic positioning scheduling method based on video identity recognition, as shown in fig. 1, which comprises the following steps:
(1) The method comprises the steps that video stream information of a dispatching system site is collected in real time through video monitoring equipment and uploaded to a cloud storage server, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of obtaining monitoring video of each position in the site;
(2) When a worker in a dispatching system enters the site for the first time in a working period, acquiring an appearance image of the worker as a standard characteristic template of the current working period of the worker;
(3) Acquiring video stream information from a cloud storage server, slicing to form a plurality of image frames, respectively inputting the image frames and standard feature templates of all staff in the current working period of the staff into two input ends of an identity recognition model to obtain an identity recognition result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
(4) Mapping the corresponding staff to a global map according to the video monitoring equipment for acquiring the image frames and the identified corresponding staff, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff to realize the positioning of all the staff;
(5) When a worker needs to be called, searching a worker-dispatching desk correspondence table, obtaining a dispatching desk corresponding to the worker, and automatically forwarding the call to the dispatching desk;
(6) And judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
The training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
The embodiment also provides a dynamic positioning scheduling system based on video identity recognition, which comprises:
the cloud storage server is used for storing video stream information of a dispatching system site acquired in real time through video monitoring equipment, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of acquiring monitoring video of each position in the site;
the image collector is used for collecting an appearance image of a worker when the worker in the dispatching system enters the site for the first time in a working period and taking the appearance image as a standard characteristic template of the current working period of the worker;
the identification module is used for acquiring video stream information from the cloud storage server, slicing the video stream information to form a plurality of image frames, respectively inputting the image frames and standard characteristic templates of all staff in the current working period into two input ends of the identification model to obtain the identification result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
the positioning module is used for mapping the corresponding staff to the global map according to the image frames and the identified corresponding staff and the video monitoring equipment for acquiring the image frames, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff so as to realize the positioning of all the staff;
the calling module is used for searching a personnel-dispatching desk correspondence table when the personnel need to be called, obtaining a dispatching desk corresponding to the personnel, and automatically forwarding the call to the dispatching desk;
and the positioning legal judgment module is used for judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
The training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
The above disclosure is only a preferred embodiment of the present invention and should not be construed as limiting the scope of the invention, which is defined by the appended claims.
Claims (6)
1. A dynamic positioning scheduling method based on video identity recognition is characterized by comprising the following steps:
(1) The method comprises the steps that video stream information of a dispatching system site is collected in real time through video monitoring equipment and uploaded to a cloud storage server, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of obtaining monitoring video of each position in the site;
(2) When a worker in a dispatching system enters the site for the first time in a working period, acquiring an appearance image of the worker as a standard characteristic template of the current working period of the worker;
(3) Acquiring video stream information from a cloud storage server, slicing to form a plurality of image frames, respectively inputting the image frames and standard feature templates of all staff in the current working period of the staff into two input ends of an identity recognition model to obtain an identity recognition result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
(4) Mapping the corresponding staff to a global map according to the video monitoring equipment for acquiring the image frames and the identified corresponding staff, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff to realize the positioning of all the staff;
(5) When the staff needs to be called, the staff-dispatching desk corresponding relation table is searched, the dispatching desk corresponding to the staff is obtained, and the call is automatically forwarded to the dispatching desk.
2. The video identity recognition-based dynamic positioning scheduling method as claimed in claim 1, wherein the method comprises the following steps: further comprises:
(6) And judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
3. The video identity recognition-based dynamic positioning scheduling method as claimed in claim 1, wherein the method comprises the following steps: the training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
4. A dynamic positioning scheduling system based on video identity recognition is characterized by comprising:
the cloud storage server is used for storing video stream information of a dispatching system site acquired in real time through video monitoring equipment, wherein the video monitoring equipment is arranged at each position of the dispatching system site so as to be capable of acquiring monitoring video of each position in the site;
the image collector is used for collecting an appearance image of a worker when the worker in the dispatching system enters the site for the first time in a working period and taking the appearance image as a standard characteristic template of the current working period of the worker;
the identification module is used for acquiring video stream information from the cloud storage server, slicing the video stream information to form a plurality of image frames, respectively inputting the image frames and standard characteristic templates of all staff in the current working period into two input ends of the identification model to obtain the identification result of the staff in each image frame; the identity recognition model is specifically a trained twin convolution network with two input ends;
the positioning module is used for mapping the corresponding staff to the global map according to the image frames and the identified corresponding staff and the video monitoring equipment for acquiring the image frames, acquiring a dispatching desk closest to the staff according to the global map, and establishing a staff-dispatching desk corresponding relation table with the corresponding staff so as to realize the positioning of all the staff;
and the calling module is used for searching a personnel-dispatching desk correspondence table when the personnel need to be called, obtaining a dispatching desk corresponding to the personnel, and automatically forwarding the call to the dispatching desk.
5. The video identity-based dynamic positioning scheduling system of claim 4, wherein: further comprises:
and the positioning legal judgment module is used for judging whether the positioning of the staff in the staff-dispatching desk corresponding relation table belongs to a specified legal range according to the authority and the service range of each staff, and if not, sending an alarm signal to a manager terminal of the staff.
6. The video identity-based dynamic positioning scheduling system of claim 4, wherein: the training method of the identity recognition model comprises the following steps:
establishing a twin convolutional network, wherein the twin convolutional network comprises two ResNet convolutional networks with the same structure, each ResNet convolutional network comprises a convolutional layer, a normalization layer and a downsampling layer which are sequentially connected, a global pooling layer is connected with the output of the two downsampling layers, all the layers are connected by a ReLU activation function, and the parameters of the two ResNet convolutional networks are shared;
the network training loss function is established as follows:
wherein v is supervision information, x c ,y c The neural network model uses f (x) c ,y c ) Representing a neural network model function, wherein T represents the number of training samples in the batch;
the method comprises the steps of obtaining video monitoring flow information of the existing staff on site, slicing to form a plurality of image frames, inputting the image frames into one ResNet convolutional network, inputting the appearance picture of the existing staff entering the site to be collected into another ResNet convolutional network, performing network training by adopting a counter-propagation and gradient descent method, and finally obtaining a trained twin convolutional network as an identity recognition model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011129592.9A CN112258026B (en) | 2020-10-21 | 2020-10-21 | Dynamic positioning scheduling method and system based on video identity recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011129592.9A CN112258026B (en) | 2020-10-21 | 2020-10-21 | Dynamic positioning scheduling method and system based on video identity recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112258026A CN112258026A (en) | 2021-01-22 |
| CN112258026B true CN112258026B (en) | 2023-12-15 |
Family
ID=74263244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011129592.9A Active CN112258026B (en) | 2020-10-21 | 2020-10-21 | Dynamic positioning scheduling method and system based on video identity recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112258026B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109034535A (en) * | 2018-06-21 | 2018-12-18 | 中国化学工程第六建设有限公司 | Construction site wisdom monitoring method, system and computer readable storage medium |
| CN109145789A (en) * | 2018-08-09 | 2019-01-04 | 炜呈智能电力科技（杭州）有限公司 | Power supply system safety work support method and system |
| CN109615158A (en) * | 2018-10-15 | 2019-04-12 | 平安科技（深圳）有限公司 | Asset management monitoring method, device, computer equipment and storage medium |
| CN109871804A (en) * | 2019-02-19 | 2019-06-11 | 上海宝尊电子商务有限公司 | A kind of method and system of shop stream of people discriminance analysis |
| WO2019184124A1 (en) * | 2018-03-30 | 2019-10-03 | 平安科技（深圳）有限公司 | Risk-control model training method, risk identification method and apparatus, and device and medium |
| CN111597999A (en) * | 2020-05-18 | 2020-08-28 | 常州工业职业技术学院 | 4S shop sales service management method and system based on video detection |
| WO2020186801A1 (en) * | 2019-03-18 | 2020-09-24 | 北京市商汤科技开发有限公司 | Driver attentiveness monitoring method and apparatus, and electronic device |
| CN111754368A (en) * | 2020-01-17 | 2020-10-09 | 天津师范大学 | College teaching evaluation method and college teaching evaluation system based on edge intelligence |
-
2020
- 2020-10-21 CN CN202011129592.9A patent/CN112258026B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019184124A1 (en) * | 2018-03-30 | 2019-10-03 | 平安科技（深圳）有限公司 | Risk-control model training method, risk identification method and apparatus, and device and medium |
| CN109034535A (en) * | 2018-06-21 | 2018-12-18 | 中国化学工程第六建设有限公司 | Construction site wisdom monitoring method, system and computer readable storage medium |
| CN109145789A (en) * | 2018-08-09 | 2019-01-04 | 炜呈智能电力科技（杭州）有限公司 | Power supply system safety work support method and system |
| CN109615158A (en) * | 2018-10-15 | 2019-04-12 | 平安科技（深圳）有限公司 | Asset management monitoring method, device, computer equipment and storage medium |
| CN109871804A (en) * | 2019-02-19 | 2019-06-11 | 上海宝尊电子商务有限公司 | A kind of method and system of shop stream of people discriminance analysis |
| WO2020186801A1 (en) * | 2019-03-18 | 2020-09-24 | 北京市商汤科技开发有限公司 | Driver attentiveness monitoring method and apparatus, and electronic device |
| CN111754368A (en) * | 2020-01-17 | 2020-10-09 | 天津师范大学 | College teaching evaluation method and college teaching evaluation system based on edge intelligence |
| CN111597999A (en) * | 2020-05-18 | 2020-08-28 | 常州工业职业技术学院 | 4S shop sales service management method and system based on video detection |
Non-Patent Citations (1)
| Title |
|---|
| 基于神经网络人脸识别的钢企应急防控系统;朱俊杰;;冶金自动化(S1);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112258026A (en) | 2021-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106888266B (en) | Ship supervision system based on multi-network fusion and data exchange method thereof | |
| CN104125349B (en) | A kind of interactive voice management method based on traffic forecast and system thereof | |
| CN105049812A (en) | Unmanned aerial vehicle portable ground station processing method and system | |
| CN101888452A (en) | Multi-access customer service system and method thereof | |
| CN104125354A (en) | Method and system for regulating and controlling voice service based on call requirements | |
| CN104125087A (en) | Alarm information processing method and device | |
| CN104021513A (en) | Digital urban management control method based on cloud computing and Internet of Things | |
| CN101471803B (en) | Conference service system and method | |
| CN110135609A (en) | A kind of intelligent tourism system and method based on big data Joint construction and sharing | |
| CN111586373A (en) | Visual comprehensive supervision big data platform for local-level traffic bureau | |
| CN113724415A (en) | Interactive automatic inspection method and equipment for unmanned aerial vehicle of power distribution line | |
| CN112258026B (en) | Dynamic positioning scheduling method and system based on video identity recognition | |
| CN105657328A (en) | Method, device and system for scheduling conference resources | |
| CN111740867B (en) | Method for automatically optimizing, scheduling and routing of wide area network service traffic | |
| CN202310005U (en) | Multi-business communication device possessing voice program control exchange function | |
| CN108632236B (en) | Data processing method and device for video network | |
| CN112233406B (en) | Quick data response control method based on 5G communication | |
| CN107124583A (en) | A kind of monitoring system of quick obtaining video monitoring information | |
| CN110213344B (en) | Multi-center remote sign language online translation system and method | |
| CN113297237A (en) | Intelligent integrated communication command system | |
| CN112911212A (en) | Scenic spot command system and method based on big data processing | |
| CN216930054U (en) | Integrated management system | |
| CN209625782U (en) | A kind of audio conference system | |
| CN111262338A (en) | Main dispatching system graph unified browsing integration method based on wide area service bus | |
| CN116154959B (en) | Multi-fusion terminal edge collaboration system and method for power distribution area |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |