WO2020114528A2 - 疫情期间公共场所潜在被感染者追踪方法、装置及系统 - Google Patents

疫情期间公共场所潜在被感染者追踪方法、装置及系统 Download PDF

Info

Publication number
WO2020114528A2
WO2020114528A2 PCT/CN2020/080855 CN2020080855W WO2020114528A2 WO 2020114528 A2 WO2020114528 A2 WO 2020114528A2 CN 2020080855 W CN2020080855 W CN 2020080855W WO 2020114528 A2 WO2020114528 A2 WO 2020114528A2
Authority
WO
WIPO (PCT)
Prior art keywords
person
diagnosed
potential infected
persons
tracking
Prior art date
Application number
PCT/CN2020/080855
Other languages
English (en)
French (fr)
Other versions
WO2020114528A3 (zh
Inventor
罗谦
党婉丽
罗晓
邓睿
耿龙
潘野
陈肇欣
郁二改
Original Assignee
中国民用航空总局第二研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国民用航空总局第二研究所 filed Critical 中国民用航空总局第二研究所
Priority to JP2020542373A priority Critical patent/JP6980928B2/ja
Publication of WO2020114528A2 publication Critical patent/WO2020114528A2/zh
Publication of WO2020114528A3 publication Critical patent/WO2020114528A3/zh
Priority to US17/979,782 priority patent/US20230121997A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20Movements or behaviour, e.g. gesture recognition
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/80ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/783Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/7837Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using objects detected or recognised in the video content
    • G06F16/784Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using objects detected or recognised in the video content the detected or recognised objects being people
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/7867Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/787Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/26Government or public services
    • G06Q50/265Personal security, identity or safety
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/70Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/46Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/50Context or environment of the image
    • G06V20/52Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00ICT specially adapted for the handling or processing of medical images
    • G16H30/40ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Definitions

  • the invention relates to the field of target tracking technology in dense scenes in public places, in particular to a method, device and system for tracking potential infected persons in public places during an epidemic.
  • Class B personnel How to find out whether a potential infected person (called "Class B personnel") has been in public transportation and places in contact with people who have been diagnosed or even the source of the epidemic (called Class A personnel) is important in epidemic prevention. Link.
  • Solution 1 According to the vehicle, flight information and seat information, check whether it is the same flight as the diagnosed person and the seat number is closer;
  • Solution 2 Based on the travel data of people, analyze the probability of virus infection
  • Option Three Based on the trajectory of the diagnosed person, he will actively search for potentially infected persons.
  • the first solution is to find potential infected persons with a relatively small range and can only locate the flight. It is difficult to locate and confirm potential infected persons in the terminal building where the infected person has been confirmed to stay for a longer time.
  • the second scheme is mainly based on the travel data of the personnel, and correlates time to find potential infected people. It has the advantage of a wide range of applications, but the analysis is coarse-grained and belongs to active query confirmation, which cannot be deleted without further action. Select and find and confirm its identity.
  • the third scheme is mainly for humans to actively seek out. By publishing the trajectory of the diagnosed person, the contact person can confirm whether he is a potential infected person.
  • the embodiments of the present invention provide a method, device and system for tracking potential infected persons in public places during epidemic prevention and control. Based on the transmission mechanism of infectious diseases, the location and trajectory of potential infected persons are determined based on multi-target tracking. The characteristics of time and space dimensions enable the identification of potential infected persons.
  • the present invention discloses a method for tracking potential infected persons in public places during an epidemic, including:
  • the associated information of the diagnosed person determine the time node of the confirmed person's activity in the public place, and the location information matching the time node;
  • the potential infected person is tracked, and the identity is confirmed in conjunction with the time node where the potential infected person is active in the public place.
  • the associated information of the diagnosed person includes: identity information of the diagnosed person and operation information of the vehicle used.
  • determining the location information of the confirmed person in a public place includes the following steps:
  • the diagnosed person is detected, extracted and confirmed.
  • detecting, extracting and confirming the diagnosed person in the video frame includes:
  • searching for potential infected persons by using video frame data provided by the video surveillance equipment in the public place includes:
  • mapping a given infectious disease transmission range to the distance in the video frame and constructing a circular area with the distance as the radius;
  • the time node corresponding to the video frame is tracked forward and backward until the position of the diagnosed person satisfies:
  • O i represents the current location of the diagnosed person
  • O i-1 is another location of the diagnosed person on its movement track
  • tracking the potential infected person based on multi-target tracking and confirming the identity in conjunction with the time node where the potential infected person is active in the public place includes the following steps:
  • the above tracking method after extracting the position and features of the diagnosed person in different video frames at different time nodes, it also includes constructing a feature pyramid, and determining the feature space according to the feature pyramid;
  • the diagnosed or potential infected person in different video frames at adjacent time points is judged
  • the similarity is:
  • T i [T i1 , T i2 , T i3 ]
  • T j [T j1 , T j2 , T j3 ];
  • Q is the similarity of diagnosed persons in different video frames or the similarity of potential infected persons in different video frames.
  • i and j are natural numbers.
  • the present invention also provides a tracking device for potential infected persons in public places during an outbreak, including:
  • the diagnosed person search module is used to determine the time node of the confirmed person's activity in the public place and the location information matching the time node according to the diagnosed person's associated information;
  • the potential infected person search module is used to search for the potential infected person based on the transmission time of the diagnosed person in the public place and the location information, based on the transmission mechanism of infectious diseases, with the help of video frame data provided by the video surveillance equipment in the public place ;
  • the identity confirmation module of the potential infected person is used to track the potential infected person based on multi-target tracking, and to confirm the identity in combination with the time node where the potential infected person is active in the public place.
  • the associated information of the diagnosed person includes: identity information of the diagnosed person and operation information of the vehicle used.
  • the confirmed person search module includes:
  • a video monitoring device determining unit configured to determine the corresponding video monitoring device in the public place according to the time node corresponding to the confirmed personnel performing activities in the public place;
  • a receiving unit configured to receive video frames of each time node extracted in the video monitoring device
  • the confirming person confirmation unit is used for detecting, extracting and confirming the confirmed person in the video frame.
  • the confirming person confirmation unit further includes:
  • the first feature space construction sub-unit is used to extract the position and features of the diagnosed person in different video frames at different time nodes to construct a feature space;
  • the first spatial probability feature construction subunit is used to construct a spatial probability feature based on moving distance and trajectory probability
  • the first confirmation unit determines the similarity of the diagnosed persons in different video frames at adjacent time points, and when the similarity is greater than a given threshold, continues on the basis of the diagnosed persons in the video frame The tracking is performed until the node of the confirmed person who is registered is tracked, and the confirmed person is confirmed.
  • the potential infected person search module is used to implement:
  • mapping a given infectious disease transmission range to the distance in the video frame and constructing a circular area with the distance as the radius;
  • the time node corresponding to the video frame is tracked forward and backward until the position of the diagnosed person satisfies:
  • O i represents the current location of the diagnosed person
  • O i-1 is another location of the diagnosed person on its movement track
  • the identity confirmation module of the potential infected person includes:
  • the second feature space construction subunit is used to extract the position and features of the searched potential infected person in different video frames at different time nodes to construct a feature space;
  • the second spatial probability feature construction subunit is used to construct a spatial probability feature based on the moving distance and the trajectory probability
  • the second confirmation subunit determines the similarity of potential infected persons in different video frames at adjacent time points. When the similarity is greater than a given threshold, the potential The infected person continues to follow up until the node of the potential infected person is registered, the identity of the potential infected person is confirmed and reported.
  • the above tracking device after extracting the position and features of the diagnosed person in different video frames at different time nodes, it also includes constructing a feature pyramid and determining the feature space according to the feature pyramid; and, the extraction search After the location and characteristics of the potential infected person in different video frames at different time nodes, it also includes constructing a feature pyramid, and determining the constructed feature space according to the feature pyramid.
  • T i [T i1 , T i2 , T i3 ]
  • T j [T j1 , T j2 , T j3 ];
  • Q is the similarity of diagnosed persons in different video frames or the similarity of potential infected persons in different video frames.
  • i and j are natural numbers.
  • the present invention also discloses a tracking method for potential infected persons in public places during an epidemic.
  • the public place is an airport terminal; the terminal is divided into a check-in area, a security inspection area, and a waiting area; Based on the check-in area, the security check area, and the waiting area, perform any of the above methods for tracking potential infected persons in public places during the epidemic; for the confirmed personnel, obtain the potential in the check-in area
  • the infected person is the first set, and the potential infected person obtained in the security area is the second set; the potential infected person obtained in the waiting area is the third set; then the first set, the The union of the second set and the third set serves as a potential infected person.
  • the present invention is based on the existing monitoring environment in public places (such as airport terminals), and on the basis of the confirmation of the identity of the diagnosed person, actively searches for the potential infected person associated with the confirmed person in the public place and determines its identity. After data testing, its accuracy can reach more than 80%, which largely solves the search and confirmation of potential infected persons in public transportation places, and plays an important role in the prevention and control of epidemics.
  • FIG. 1 is a flowchart of steps of a method for tracking a potential infected person in a public place during an epidemic situation according to an embodiment of the present invention
  • FIG. 2 shows a business flowchart of a method for tracking potential infected persons in public places during epidemic prevention and control according to an embodiment of the present invention
  • FIG. 3 shows a flow chart of how to determine the time node of the confirmed personnel’s activity in the terminal building and the location information matching the time node during the method of tracking potential infected persons in public places during epidemic prevention and control according to an embodiment of the present invention
  • FIG. 4 shows a flowchart of how to search for a potential infected person in a method for tracking a potential infected person in a public place during epidemic prevention and control according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of pedestrian detection and numbering of a video frame where a diagnosed person is located in a tracking method of a potential infected person in a public place during an epidemic situation according to an embodiment of the present invention
  • FIG. 6 is a schematic diagram of determining a first batch of potentially infected persons in a method of tracking a potentially infected person in a public place during an epidemic according to an embodiment of the present invention, using a confirmed person as the center of the circle;
  • FIG. 7 is a schematic diagram of a model for calculating the radius of circle O 1 in the method for tracking potential infected persons in public places during an epidemic situation according to an embodiment of the invention
  • FIG. 8 is a method for tracking a potential infected person in a public place during an epidemic embodiment according to an embodiment of the present invention.
  • FIG. 8 shows a schematic diagram of the second batch of potentially infected persons in circle O 1 circumscribed from circle O;
  • FIG. 9 is a schematic diagram of a feature pyramid in a method for tracking potential infected persons in public places during an epidemic embodiment according to the present invention.
  • FIG. 10 is a structural block diagram of a tracking device for potential infected persons in a public place during an epidemic situation according to an embodiment of the present invention.
  • A is the definitive diagnosis of virus infection.
  • B the other is C.
  • a and B do not know.
  • a and C know.
  • the C-type person is easy to confirm. But it is difficult to find people in category B. This is because, for example, A has encountered B in a public place that he never knew, A does not know the existence of B, and B does not know the existence of A. Therefore, B is the biggest hidden danger and it is not easy to find. No one knows who is B, even if he is B.
  • Embodiments of the present invention provide a method and system for tracking potential infected persons in public places during epidemic prevention and control. Based on the transmission mechanism of infectious diseases, the location and trajectory of potential infected persons are determined based on multi-target tracking, combined with the characteristics of time and space dimensions. To realize the identification of potential infected persons, the process is as follows:
  • FIG. 1 a flowchart of steps of a method for tracking a potential infected person in a public place during an epidemic situation according to an embodiment of the present invention is shown, including the following steps:
  • the present invention is not limited to the airport terminal, and can be extended to bus stations, train stations, docks, etc. where video surveillance equipment is provided and needs to be identified Certified public places.
  • FIG. 2 shows a business flowchart of a method for tracking a potential infected person in a public place during epidemic prevention and control in this embodiment.
  • the mechanism is to locate pedestrian characteristics based on the time axis information of the video frame and search for potentially infected persons.
  • Each area includes three key points: 1) Confirmation of time nodes and location information of the key area of the diagnosed personnel; 2) Calibration of the position and image identity of potential personnel based on the virus propagation mechanism; 3) Potential in key areas Track and confirm the identity information of the infected person.
  • the above process can be implemented by the following steps, refer to FIG. 3.
  • FIG. 3 shows how to determine the time node at which the diagnosed person performs activities in a public place and the position information matching the time node according to the associated information of the diagnosed person. In other words, how to determine the movement position of the diagnosed virus infection in the key area.
  • Step S310 query the identity information and flight information of the diagnosed personnel
  • Step S320 according to the identity information and flight information, determine the three time nodes of the confirmed person in the check-in area, security check area and waiting area of the airport terminal;
  • Time ⁇ C time , S time , B time ⁇
  • Step S340 confirm the corresponding camera ID through the time node, the check-in counter, the security channel, and the waiting ticket check counter;
  • Step S350 Extract the video frame of the Time node according to the camera ID.
  • step S360 the pedestrian in the video frame is detected, and the person confirmed by the rugged virus infection is confirmed.
  • This step can be completed as follows:
  • S120 will be described. That is, based on the time node and location information of the diagnosed personnel in public places, based on the transmission mechanism of infectious diseases, how to search for potential infected persons with the help of video frame data provided by video surveillance equipment in public places. Specifically, referring to FIG. 4, it includes the following steps:
  • Step S410 on the basis of confirming the position of the virus-infected person A, taking the position of A in the three areas as the center of the circle O, taking the range of the spread of the infectious disease as the radius and mapping the distance to the video frame as the radius, construct a circular area, where Pedestrians in the circular area can be identified as category B personnel, that is, potential infected persons.
  • category B personnel that is, potential infected persons.
  • the location of category B personnel is recorded as B i
  • the time of the video frame image where it is located is
  • Step 420 the video frame where the virus infected person A is located is the starting video frame, and the time in the three areas is C time , S time , B time , and A is in front of the C time , S time , and B time time nodes. And then track until its location O i satisfies the following relationship:
  • d i represents the distance in which infectious diseases spread in practice is mapped to the distance in the video frame.
  • step S430 taking O i as the center of the circle, d i continues to construct the circular area, and determine the class B personnel B j
  • Step S440 repeat Step 2 and Step 3 until all the personnel of Class B are confirmed.
  • step S130 will be further described.
  • tracking and identifying the identity based on the potential infected person can be achieved through the following steps:
  • This embodiment is based on the existing monitoring environment of the airport terminal, and on the basis of the confirmation of the identity of the diagnosed person, actively searches for the potential infected person associated with the confirmed person in the public place and determines its identity. After data testing, its accuracy can reach more than 80%, which largely solves the search and confirmation of potentially infected persons in public transportation places, and plays an important role in the prevention and control of epidemics.
  • the following describes a method for tracking potential infected persons in public places during the epidemic prevention and control of the present invention in conjunction with a more specific embodiment.
  • This example selects the terminal security inspection area as an example.
  • Step 1 According to the associated information of the diagnosed person A, determine the time node at which the diagnosed person A is active in the airport terminal, and the location information matching the time node.
  • the identity information and flight information of the diagnosed person A confirm the location and time of the appearance of the security check channel, set the time of the ticket check in the security check area of an airport to 11:23, and the security check channel number to 10.
  • the video frame is retrieved, and one frame of the image is obtained. Then, the frame image is divided into blocks, and the pedestrian is detected using the RCNN model in the prior art, and the pedestrian is numbered.
  • Figure 5 According to the position of the ticket inspection, confirm the diagnosed person A, whose number in the image is 45, and the box marked A in the figure is the confirmed person A.
  • Step 2 According to the time node and location information of the diagnosed person A in the terminal security screening area, based on the transmission mechanism of infectious diseases, with the help of the video frame data provided by the video surveillance equipment in the terminal security screening area, search for potential infected persons, that is, Class B personnel.
  • the distance mapped in the picture is y, and y is defined as the reference distance, which provides a basis for the subsequent radius calculation;
  • the size of m mapped to this picture phase is ma.
  • the historical data is used to track the trajectory before the diagnosed person A appears at this position in reverse order, and the circle O 1 is constructed for the center of the circle at the next position before the diagnosed person A appears.
  • the circle O is tangent to the circle O 1 , and the circle The radius of O 1 is calculated according to the model shown in FIG. 7.
  • y 1 represents the position of the diagnosed person in circle O 1 and the corresponding y-axis value in the established two-dimensional coordinate system
  • x 1 represents the value of X-axis at the position of circle O 1 and the reference line
  • the coordinates are (x, y), y 2 and x 2 respectively represent the position of the diagnosed person in the circle O, then the pixels (area) S 1 and S 2 occupied in the image satisfy the following relationship:
  • FIG. 8 shows a schematic diagram of the second batch of potentially infected persons in circle O 1 circumscribed from circle O.
  • Step 3 Track and confirm the identity based on the potential infected person.
  • the trajectory description and identity confirmation of a potential infected person based on multi-target tracking technology.
  • the location of the infected person in circle O has been confirmed, and its numbers are 19 and 30.
  • the tracking technology for potential infected persons is also applicable to confirmed persons.
  • the video frame of time is input to the mask-RCNN network in the prior art to extract the features of the personnel numbered 19 and 30, and construct their feature pyramid to construct the feature space.
  • the use of feature pyramid can prompt more advanced semantic information to be detected. Let its feature space be:
  • T i [T i1 , T i2 , T i3 ]
  • FIG. 9 is a schematic diagram of a feature pyramid.
  • the semantic information of each layer in Fig. 9 is from low to high from bottom to top, the semantics are stronger and stronger, the detected information is more and more advanced, and the deeper the image information that is mined.
  • T j [T j1 , T j2 , T j3 ]
  • the pedestrian's spatial movement direction is deterministic, and its probability of moving in different directions is different.
  • P i is the statistical experience value, for example: At the security checkpoint, the probability of pedestrians moving towards the security checkpoint is greater than the probability of other directions). Therefore, measuring the similarity (Q) between the two is defined as:
  • the similarity is calculated using the features of the corresponding level, as follows:
  • step 1 the identity of the diagnosed person A in the security inspection area is also continuously verified by using the similarity-based method described above.
  • the algorithm is the same and will not be repeated here.
  • the above embodiment takes the security inspection area as an example to illustrate how the identity of the potential infected person B is confirmed, and the check-in area and the waiting area are not involved, but the identity of the potential infected person B in these two areas is confirmed
  • the method is the same as in this embodiment, and will not be repeated here.
  • the first set of potential infected persons in the check-in area is the first set
  • the second set of potential infected persons in the security check area is the third set; Then, the union of the first set, the second set and the third set is regarded as a potential infected person.
  • a public place such as an airport terminal
  • it on the basis of the confirmation of the identity of the diagnosed person, it actively searches for the potential infected person associated with the confirmed person in the public place and determines its identity. After data testing, its accuracy can reach more than 80%, which largely solves the search and confirmation of potentially infected persons in public transportation places, and plays an important role in the prevention and control of epidemics.
  • the present invention also provides a tracking device for potential infected persons in public places during an outbreak. Referring to Figure 10, including:
  • the diagnosed person search module 1001 is used to determine the time node of the confirmed person's activity in the public place and the location information matching the time node according to the diagnosed person's associated information;
  • the potential infected person search module 1002 is used to search for the potential infection based on the transmission time of the diagnosed person in the public place and the location information, based on the transmission mechanism of infectious diseases, with the help of video frame data provided by the video surveillance device in the public place By;
  • the potential infected person identity confirmation module 1003 is used to track the potential infected person based on multi-target tracking, and to confirm the identity in conjunction with the time node where the potential infected person is active in the public place.
  • the tracking device for potential infected persons in public places during the epidemic has the same principle as the tracking method for potential infected persons in public places during the epidemic.
  • the methods for tracking potential infected persons in public places during the epidemic have been described above. It is sufficient to refer to each other, and the present invention will not repeat them here.
  • embodiments of the present invention also provide a tracking system for potential infected persons in public places during an epidemic.
  • the system includes: one less processor; and a memory communicatively connected to the at least one processor; wherein the memory stores An instruction executable by the at least one processor, the instruction being executed by the at least one processor, so that the at least one processor can execute the method for tracking a potential infected person in a public place during an epidemic in the foregoing method embodiment.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Primary Health Care (AREA)
  • Data Mining & Analysis (AREA)
  • Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Epidemiology (AREA)
  • Library & Information Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Pathology (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Computer Security & Cryptography (AREA)
  • Radiology & Medical Imaging (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Social Psychology (AREA)
  • Human Computer Interaction (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Evolutionary Computation (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Image Analysis (AREA)
  • Medical Treatment And Welfare Office Work (AREA)

Abstract

本发明实施例中提供了一种疫情期间公共场所潜在被感染者追踪方法、装置及系统,该方法包括如下步骤:根据确诊人员关联信息,确定确诊人员在公共场所进行活动的时间节点,以及与时间节点匹配的位置信息;根据确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;基于多目标跟踪技术对潜在感染者进行跟踪,并结合公共场所关键节点时间确认其身份。本发明主动寻找公共场所中与确诊人员关联的潜在被感染者,并确定其身份。经过数据测试,其精度可以达到80%以上,在很大程度上解决了公共交通场所潜在被感染人员的寻找及确认,对疫情的防控发挥重要的作用。

Description

疫情期间公共场所潜在被感染者追踪方法、装置及系统 技术领域
本发明涉及公共场所密集场景下目标追踪技术领域,尤其涉及一种疫情期间公共场所潜在被感染者追踪方法、装置及系统。
背景技术
在疫情传播过程中,假设有三种人,分别为A、B和C。C与A明确认识,则在A诊断为患者后,与之接触的C很容易确定。但假设A在公众场所(例如旅途中的机场、车站或码头等)遇到了平生不认识的B,A不知道B的存在,B也不知道A的存在。但经过若干天的病毒潜伏期,发现A为病毒感染者。那么A在公共场所曾经接触过的B,成为潜在的被病毒感染者,B这种潜在的感染者是非常危险的,因为B被明确诊断为感染者之前,没有人知道他是被感染者,连B自己也无法知道。于是,B在公共场所没有任何防护的继续出现,若A与B的接触果真导致了B的被感染,则B处于潜伏期的这段时间无疑会导致病毒更大范围的扩散与传播。
如何寻找潜在的被感染人(称之为“B类人员”)是否在公共交通工具及场所中与已确诊人员甚至疫源地人员(称之为A类人员)接触过是疫情防疫中的重要环节。
现已有部分方案或者软件可以解决潜在感染者的发现问题,例如在机场或者或者候车室或检票口落采用如方案:
方案一、根据乘坐车辆、航班信息以及座位信息,查询自身是否与确诊人员为同一航班以及座位号较近;
方案二、基于人员旅行数据,分析其感染病毒的概率;
方案三、人为根据确诊人员的运动轨迹,主动寻找潜在感染人员。
在上述三种主流解决方案中,存在以下问题:
(1)第一种方案寻找潜在感染者范围较小,只可以定位到航班,对于已经确认感染者停留且停留时间较长的位置的航站楼潜在感染者难易定位及确认。
(2)第二种方案主要基于人员的旅行数据,进行时间的关联,寻找潜在感染者,其具有应用范围较广的优势,但分析粒度粗,且属于主动查询确认,而不能别动的删选及查找且确认其身份。
(3)第三种方案主要是人为主动寻找,通过公布确诊者的运动轨迹,由接触者自行确认自身是否属于潜在感染者。
上述三种方案在一定程度上均可以删选出部分潜在的感染者,但对于机场航站楼,均不能解决如何确认航站楼内潜在感染人员身份的问题。
综上所述,大型机场航站楼、火车站、公共汽车站等作为公共场所,人群较为聚集,停留时间长,是疫情防御中需要重点关注场所。如何从诸如上述共场合中,根据已确诊的感染人的信息,准确定位潜在的被感染人,并确认被感染人的身份,是本领域技术技术人员亟需解决的技术问题。
发明内容
有鉴于此,本发明实施例提供一种疫情防控期间公共场所潜在感染者追踪方法、装置及系统,以传染病传播机理为途径,基于多目标跟踪确定潜在被感染者的位置及轨迹,结合时间及空间维度特征,实现对潜在被感染者的身份确认。
第一方面,本发明公开了一种疫情期间公共场所潜在被感染者追踪方法,包括:
根据确诊人员关联信息,确定所述确诊人员在公共场所进行活动的时间节点,以及与所 述时间节点匹配的位置信息;
根据所述确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份。
进一步地,上述追踪方法中,所述确诊人员关联信息包括:确诊人员身份信息和所乘交通工具的运营信息。
进一步地,上述追踪方法中,确定所述确诊人员在公共场所的位置信息包括如下步骤:
根据所述确诊人员在公共场所进行活动所对应的时间节点,确定公共场所中对应的视频监控设备;
接收在所述视频监控设备中所提取的各个时间节点的视频帧;
在所述视频帧中,对所述确诊者进行检测、提取并确认。
进一步地,上述追踪方法中,在所述视频帧中,对所述确诊者进行检测及提取并确认,包括:
提取确诊者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
基于移动距离和轨迹概率构建空间概率特征;
基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者的相似度,当相似度大于给定阈值时,以视频帧中的确诊者为基础继续进行追踪,描述完其在公共场所的所有位置,对所述的确诊者在公共区的轨迹进行确认。
进一步地,上述追踪方法中,所述基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者包括:
在所述视频帧中,以所述确诊者的位置为圆心,将给定的传染病传播范围映射到视频帧中的距离后,以该距离为半径构建圆形区域;
将所述圆形区域中的行人确定为潜在被感染者;
将所述确诊者当前的视频帧作为起始视频帧,对该视频帧所对应的时间节点向前、向后进行追踪,直至所述确诊者所在的位置满足:
|O i-O i-1|=d i
其中,O i表示所述确诊者当前所在的位置,O i-1为所述确诊者在其运动轨迹的另一个位置;
以O i为圆心,d i为半径,继续构建圆形区域,再次搜索潜在的被感染者。
重复上述步骤,直至搜索到该公共场所中所有的被感染者。
进一步地,上述追踪方法中,基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份,包括如下步骤:
提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
基于移动距离和轨迹概率构建空间概率特征;
基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中潜在被感染者的相似度,当相似度大于给定阈值时,以视频帧中的潜在被感染者为基础继续进行追踪,直至跟踪至所述潜在被感染者的进行登记的节点时,确认所述潜在被感染者的身份并上报。
进一步地,上述追踪方法中,所述提取确诊者在不同时间节点不同视频帧中的位置及特征后,还包括构建特征金字塔,依据所述特征金字塔确定所述特征空间;以及
所述提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征后,也包括 构建特征金字塔,依据所述特征金字塔确定所述构建特征空间。
进一步地,上述追踪方法中,基于所述确诊者或所述潜在被感染者基于视频帧获得的特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者或潜在被感染者的相似度为:
令第i时刻从视频帧中提取的特征金字塔构建的特征空间为T i=[T i1,T i2,T i3]
令第j=i+1时刻从视频帧中提取的特征金字塔构建的特征空间为T j=[T j1,T j2,T j3];
令所述确诊者或所述潜在被感染者在不同方向的空间概率特征为P i
则不同视频帧中确诊者或潜在被感染者的相似度通过如下方式计算:
Figure PCTCN2020080855-appb-000001
Figure PCTCN2020080855-appb-000002
Figure PCTCN2020080855-appb-000003
Q=P 1·Q 1+P·Q 2+P 3·Q 3(P 1+P 2+P 3=1)
其中,Q为不同视频帧中确诊者的相似度或不同视频帧中,潜在被感染者的相似度,i和j均为自然数,在实际应用中,P 1>P 2>P 3
第二方面,本发明还提供了一种疫情期间公共场所潜在被感染者追踪装置,包括:
确诊人员搜索模块,用于根据确诊人员关联信息,确定所述确诊人员在公共场所进行活动的时间节点,以及与所述时间节点匹配的位置信息;
潜在被感染者搜索模块,用于根据所述确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
潜在被感染者身份确认模块,用于基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份。
进一步地,上述追踪装置中,所述确诊人员关联信息包括:确诊人员身份信息和所乘交通工具的运营信息。
进一步地,上述追踪装置中,所述确诊人员搜索模块包括:
视频监控设备确定单元,用于根据所述确诊人员在公共场所进行活动所对应的时间节点,确定公共场所中对应的视频监控设备;
接收单元,用于接收在所述视频监控设备中所提取的各个时间节点的视频帧;
确诊人员确认单元,用于在所述视频帧中,对所述确诊者进行检测、提取并确认。
进一步地,上述追踪装置中,在所述视频帧中,所述确诊人员确认单元还包括:
第一特征空间构建子单元,用于提取确诊者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
第一空间概率特征构建子单元,用于基于移动距离和轨迹概率构建空间概率特征;
第一确认单元,基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者的相似度,当相似度大于给定阈值时,以视频帧中的确诊者为基础继续进行追踪,直至跟踪至所述确诊者的进行登记的节点时,对所述确诊者进行确认。
进一步地,上述追踪装置中,所述潜在被感染者搜索模块用于实现:
在所述视频帧中,以所述确诊者的位置为圆心,将给定的传染病传播范围映射到视频帧中的距离后,以该距离为半径构建圆形区域;
将所述圆形区域中的行人确定为潜在被感染者;
将所述确诊者当前的视频帧作为起始视频帧,对该视频帧所对应的时间节点向前、向后进行追踪,直至所述确诊者所在的位置满足:
|O i-O i-1|=d i
其中,O i表示所述确诊者当前所在的位置,O i-1为所述确诊者在其运动轨迹的另一个位置;
以O i为圆心,d i为半径,继续构建圆形区域,再次搜索潜在的被感染者。
进一步地,上述追踪装置中,潜在被感染者身份确认模块包括:
第二特征空间构建子单元,用于提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
第二空间概率特征构建子单元,用于基于移动距离和轨迹概率构建空间概率特征;
第二确认子单元,基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中潜在被感染者的相似度,当相似度大于给定阈值时,以视频帧中的潜在被感染者为基础继续进行追踪,直至跟踪至所述潜在被感染者的进行登记的节点时,确认所述潜在被感染者的身份并上报。
进一步地,上述追踪装置中,所述提取确诊者在不同时间节点不同视频帧中的位置及特征后,还包括构建特征金字塔,依据所述特征金字塔确定所述特征空间;以及,所述提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征后,也包括构建特征金字塔,依据所述特征金字塔确定所述构建特征空间。
进一步地,上述追踪装置中,所述第一确认子单元和所述第二确认子单元中:
令第i时刻从视频帧中提取的特征金字塔构建的特征空间为T i=[T i1,T i2,T i3]
令第j=i+1时刻从视频帧中提取的特征金字塔构建的特征空间为T j=[T j1,T j2,T j3];
令所述确诊者或所述潜在被感染者在不同方向的空间概率特征为P i
则不同视频帧中确诊者或潜在被感染者的相似度通过如下方式计算:
Figure PCTCN2020080855-appb-000004
Figure PCTCN2020080855-appb-000005
Figure PCTCN2020080855-appb-000006
Q=P 1·Q 1+P·Q 2+P 3·Q 3(P 1+P 2+P 3=1)
其中,Q为不同视频帧中确诊者的相似度或不同视频帧中,潜在被感染者的相似度,i和j均为自然数,在实际应用中,P 1>P 2>P 3
第三方面,本发明还公开了一种疫情期间公共场所潜在被感染者追踪方法,所述公共场所为航站楼;将所述航站楼划分为值机区域、安检区域和候机区域;基于所述值机区域、所述安检区域和所述候机区域执行上述任一项所述的疫情期间公共场所潜在被感染者追踪方法;针对所述确诊人员,在所述值机区域获得潜在被感染者为第一集合,在所述安检区域获得潜在被感染者为第二集合;在所述候机区域获得的潜在被感染者为第三集合;则将所述第一集合、所述第二集合与所述第三集合的并集作为潜在被感染者。
本发明基于公共场所(例如机场航站楼)现有的监视环境,在确诊者身份确认的基础上,主动寻找公共场所中与确诊人员关联的潜在被感染者,并确定其身份。经过数据测试,其精度可以达到80%以上,在很大程度上解决了公共交通场所潜在被感染人员的寻找及确认,对疫情的防控发挥重要的作用。
附图说明
为了更清楚地说明本发明实施例的技术方案,下面将对实施例中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其它的附图。
图1为本发明实施例疫情期间公共场所潜在被感染者追踪方法的步骤流程图;
图2示出了本发明实施例疫情防控期间公共场所潜在感染者追踪方法的业务流程图;
图3示出了本发明实施例疫情防控期间公共场所潜在感染者追踪方法,如何确定确诊人员在航站楼所进行活动的时间节点,以及与时间节点匹配的位置信息的步骤流程图;
图4示出了本发明实施例疫情防控期间公共场所潜在感染者追踪方法中,如何搜索潜在被感染者的步骤流程图;
图5为本发明实施例疫情期间公共场所潜在感染者追踪方法中,确诊人员所在的视频帧进行行人检测及编号示意图;
图6为本发明实施例疫情期间公共场所潜在感染者追踪方法中,以确诊人员为圆心构建平面圆,确定第一批潜在感染人员的示意图;
图7为本发明实施例疫情期间公共场所潜在感染者追踪方法中,计算圆O 1半径的模型示意图;
图8为本发明实施例疫情期间公共场所潜在感染者追踪方法中,图8示出了与圆O外切的圆O 1中的第二批潜在感染人员的示意图;
图9为本发明实施例疫情期间公共场所潜在感染者追踪方法中,特征金字塔的示意图;
图10为本发明实施例疫情期间公共场所潜在被感染者追踪装置的结构框图。
具体实施方式
下面结合附图对本发明实施例进行详细描述。
以下通过特定的具体实例说明本公开的实施方式,本领域技术人员可由本说明书所揭露的内容轻易地了解本公开的其他优点与功效。显然,所描述的实施例仅仅是本公开一部分实 施例,而不是全部的实施例。本公开还可以通过另外不同的具体实施方式加以实施或应用,本说明书中的各项细节也可以基于不同观点与应用,在没有背离本公开的精神下进行各种修饰或改变。需说明的是,在不冲突的情况下,以下实施例及实施例中的特征可以相互组合。基于本公开中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本公开保护的范围。
需要说明的是,下文描述在所附权利要求书的范围内的实施例的各种方面。应显而易见,本文中所描述的方面可体现于广泛多种形式中,且本文中所描述的任何特定结构及/或功能仅为说明性的。基于本公开,所属领域的技术人员应了解,本文中所描述的一个方面可与任何其它方面独立地实施,且可以各种方式组合这些方面中的两者或两者以上。举例来说,可使用本文中所阐述的任何数目个方面来实施设备及/或实践方法。另外,可使用除了本文中所阐述的方面中的一或多者之外的其它结构及/或功能性实施此设备及/或实践此方法。
还需要说明的是,以下实施例中所提供的图示仅以示意方式说明本公开的基本构想,图式中仅显示与本公开中有关的组件而非按照实际实施时的组件数目、形状及尺寸绘制,其实际实施时各组件的型态、数量及比例可为一种随意的改变,且其组件布局型态也可能更为复杂。
另外,在以下描述中,提供具体细节是为了便于透彻理解实例。然而,所属领域的技术人员将理解,可在没有这些特定细节的情况下实践所述方面。
在进行方案步骤描述之前,对B类人员进行定义。
在疫情传播过程中,假设有三种人A、B、C。A为病毒感染的确诊人员。在A确诊前,可能与两类人相遇,一类是B,一类是C。A与B不认识。A与C认识。在A确诊为病毒感染者后,C类人很容易确认。但B类人很难发现。这是因为,例如,A在公共场所中遇到过平生绝不认识的B,A不知道B的存在,B不知道A的存在。因此,B是最大的隐患也是不易找到的,没有人知道谁是B,哪怕自己是B也难以知晓。
本发明实施例提供一种疫情防控期间公共场所潜在感染者追踪方法及系统,以传染病传播机理为途径,基于多目标跟踪确定潜在被感染者的位置及轨迹,结合时间及空间维度特征,实现对潜在被感染者的身份确认,流程如下:
(1)结合公共场所的业务流程,描述确诊人员在关键区域的运动位置;
(2)根据确诊者的运动位置,基于传染病传播机理的潜在被感染人员的寻找确认;
(3)基于多目标跟踪技术的潜在感染者的轨迹描述及身份确认方法.
参照图1,示出了本发明实施例疫情期间公共场所潜在被感染者追踪方法的步骤流程图,包括如下步骤:
S110,根据确诊人员关联信息,确定确诊人员在公共场所进行活动的时间节点,以及与时间节点匹配的位置信息;
S120,根据确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
S130,基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份。
下面,以机场航站楼为例,对本发明疫情防控期间公共场所潜在感染者追踪方法作进一步地详细说明。需要说明的是,本实施例虽然以机场航站楼为例进行说明,但本发明并非局限于机场航站楼,可扩展至汽车站、火车站、码头等设置有视频监控设备并且需要进行身份认证公共场所。
参照图2,图2示出了本实施例疫情防控期间公共场所潜在感染者追踪方法的业务流程图。
简单地说,就是病毒感染的确诊人员在离港前,根据确诊人员关联信息,(包括确诊人 员个人信息和航班的运营信息),在值机区域、安检区域和候机区域,基于传染病传播机理,依据视频帧时间轴信息,定位行人特征,搜索潜在被感染人员。其中每个区域都包括三个关键点:1)确诊人员关键区域时间节点及位置信息的确认;2)基于病毒传播学机理,对潜在人员位置及图像身份的标定;3)关键区域中进行潜在被感染者身份信息的追踪及确认。在一个实施例中,可以通过如下步骤实现上述过程,参照图3。
图3示出了根据确诊人员关联信息,如何确定确诊人员在公共场所进行活动的时间节点,以及与时间节点匹配的位置信息。也就是说,如何确定病毒感染的确诊者在关键区域的运动位置。
步骤S310,查询确诊人员的身份信息和航班信息;
步骤S320,依据身份信息和航班信息,确定该确诊人员在机场航站楼的值机区域、安检区域和候机区域三个时间节点;
步骤S330,建立确诊者信息集合A={Face,Fli,Time},其中,Face身份信息中的一种表现形式,指的是面部信息,Fli指的是航班信息,Time代表时间序列集合,其中包含A的值机时间C time,安检时间S time,登机检票时间B time,具体表示为:
Time={C time,S time,B time}
步骤S340,通过时间节点及值机柜台、安检通道、候机检票柜台确认对应的摄像头ID;
步骤S350,根据摄像头ID提取Time时间节点的视频帧。
步骤S360,对视频帧中的行人进行检测,并崎岖病毒感染确诊者确认者。该步骤中可以通过如下方式完成:
提取确诊者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
基于移动距离和轨迹概率构建空间概率特征;
基于特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者的相似度,当相似度大于给定阈值时,以视频帧中的确诊者为基础继续进行追踪,直至跟踪至确诊者的进行登记的节点时,对确诊者进行确认。
其上述确认方式在下面涉及潜在被感染者(B类人)时还会做进一步地说明,故在此不再赘述。相关之处,参照下文说明即可。
下面对S120进行说明。也就是,根据确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助公共场所的视频监控设备提供的视频帧数据,如何搜索潜在被感染者。具体来说,参照图4,包括如下步骤:
步骤S410,在确认病毒感染者A的位置的基础上,以A在三个区域的位置为圆心O,以传染病传播的范围为半径映射到视频帧的距离为半径,构建圆形区域,所在圆形区域中的行人可确定为B类人员,也就是潜在的被感染者,此时B类人员的位置记为B i,,其所在的视频帧图像的时间为
Figure PCTCN2020080855-appb-000007
步骤420,以确认病毒感染者A所在的视频帧为起始视频帧,其在三个区域的时间为C time,S time,B time,对A在C time,S time,B time时间节点前、后进行追踪,直至其所在位置O i满足以下关系:
|O i-O i-1|=d i
其中d i代表的是实际中传染病传播的距离映射到视频帧中的距离。
步骤S430,以O i为圆心,d i继续构建圆形区域,确定B类人员B j
步骤S440,重复步骤2、步骤3,直至所有的B类人员确认。
下面,对步骤S130做进一步地说明。在一个实施例中,基于潜在被感染者进行追踪并进行身份确认可以通过如下步骤实现:
i以提取的
Figure PCTCN2020080855-appb-000008
时间节点的视频帧为起始点确定B i的位置及特征;
ii提取并构建此视频帧中B i的特征金字塔,构成特征空间;
iii构建空间概率特征,主要指移动的距离、轨迹概率,对其进行跟踪;
iv跟踪至值机、安检、或者候机检票处,根据节点时间确认B类人员的身份信息。
本实施例基于机场航站楼现有的监视环境,在确诊者身份确认的基础上,主动寻找公共场所中与确诊人员关联的潜在被感染者,并确定其身份。经过数据测试,其精度可以达到80%以上,在很大程度上解决了公共交通场所潜在被感染人员的寻找及确认,对疫情的防控发挥重要的作用。
下面结合一个更加具体的实施例,对本发明疫情防控期间公共场所潜在感染者追踪方法进行说明。本实例选择航站楼安检区域为例进行说明。
步骤1:根据确诊人员A的关联信息,确定确诊人员A在机场航站楼进行活动的时间节点,以及与所述时间节点匹配的位置信息。
具体地,根据确诊者A的身份信息和航班信息,确认其出现在安检通道的位置以及时间,设其出现在某机场安检区域进行检票的时间为11:23,出现的安检通道号为10。
根据时间节点11:23及安检通道10,调取视频帧,获取其中的一帧图像。然后,将该帧图像进行分块,利用现有技术中的RCNN模型对行人进行检测,并将行人进行编号。参照图5所示。根据检票位置,确定确诊人员A,其在图像中的编号为45,图中标记有A的框为确诊人员A。
步骤2,根据确诊人员A在航站楼安检区域的时间节点与位置信息,基于传染病传播机理,借助航站楼安检区域的视频监控设备提供的视频帧数据,搜索潜在被感染者,也就是B类人员。
根据传染病的传播机理,设其在m米范围之内,停留时间为t的条件下均可以认为其为潜在感染者B。
设安检区域检票柜台行人站立的安全距离为d,映射到图片中的距离为y,y定义为基准距离,为后续半径计算提供基础依据;
根据影视透视模型,三维空间中的距离变化与二维空间中的距离变化存在比例关系,设
Figure PCTCN2020080855-appb-000009
此时根据确诊者A在图片中的位置,m映射到此图相中的大小为ma。
在视频帧中构建以确诊者A所在位置为圆心,ma为半径的圆,如图6所示。图6为以确诊人员为圆心构建平面圆,确定第一批潜在感染人员的示意图。其中,m=1米,a=0.05。图6中圆O区域中的人员即为第一批寻找的潜在感染者B 1的位置,其在图6中的编号为19 和30。
接下来,利用历史数据对确诊人员A出现在此位置之前的轨迹进行倒序跟踪,并在确诊人员A出现的之前的下一个位置为圆心构建圆O 1,圆O与圆O 1相切,圆O 1的半径计算按照如图7所示的模型进行计算。
在图7中,y 1代表确诊者在圆O 1中的位置及对应的建立的二维坐标系下y轴值,x 1代表其在圆O 1位置的X轴的值,其参考线的坐标为(x,y),y 2及x 2分别代表确诊者在圆O中的位置,则其在图像中所占的像素点(面积)S 1与S 2满足下述关系:
Figure PCTCN2020080855-appb-000010
设O 1的半径为r 1,圆O的半径为r,r 1的计算公式为:
Figure PCTCN2020080855-appb-000011
r与r 1满足下述关系:
Figure PCTCN2020080855-appb-000012
根据上述方法,确定圆O 1。进而确定圆O 1中的潜在被感染人员。参照图8。图8示出了与圆O外切的圆O 1中的第二批潜在感染人员的示意图。
重复圆O至圆O 1的方式,根据感染者A的运动轨迹,确认所有潜在感染人员也就是B类人员的位置。
步骤3,基于潜在被感染者进行追踪并进行身份确认。在本实施例中,是基于多目标跟踪技术的潜在感染者的轨迹描述及身份确认。
如上所述,圆O中的感染者位置已经确认,其编号为19、30。对潜在感染者的追踪技术对于确诊者同样适用。
Figure PCTCN2020080855-appb-000013
时间的视频帧输入至现有技术中的mask-RCNN网络,提取编号为19、30的人员的特征,并构建其特征金字塔,构建特征空间。特征金字塔的使用可以促使更多高级语义信息被检测到。设其特征空间为:
T i=[T i1,T i2,T i3]
参照图9,图9为特征金字塔的示意图。图9中每一层的语义信息由下至上从低至高,语义越来越强,检测到的信息越来越高级,挖掘到的图片信息越深入。
Figure PCTCN2020080855-appb-000014
帧图像进行行人检测,同样构建其特征金字塔。记为:
T j=[T j1,T j2,T j3]
根据安检区域的地理位置特征,行人的空间运动方向具有确定性,其在运动向不同的方向的概率不同,设其在不同方向的运动概率为P i(P i为统计经验值,例如:对于安检口,行人朝安检口运动的概率大于其它方向的概率值)。因此衡量两者之间的相似度(Q)定义为:
Figure PCTCN2020080855-appb-000015
为了减少其计算量,在实际相似度的计算中,使用对应层次的特征进行相似度的计算,如下:
Figure PCTCN2020080855-appb-000016
Figure PCTCN2020080855-appb-000017
Figure PCTCN2020080855-appb-000018
最终:
Q=P 1·Q 1+P·Q 2+P 3·Q 3(P 1+P 2+P 3=1)
且在实际运用中,P 1>P 2>P 3
由此以来,可以避免整体特征进行计算的冗余性,进一步提高了在实际应用中计算速度。
根据此相似度,确认潜在感染者在
Figure PCTCN2020080855-appb-000019
帧中的位置,以此为跟踪,直至其位置为安检通道的检票柜台,根据检票信息,确认潜在感染者B的身份,并上报。
在一个实施例中,在步骤1中,也采用上述基于相似度的方式确定对确诊人员A在安检区域的身份不断进行核实。算法相同,在此不再赘述。
需要说明的是,上述实施例是以安检区域为例,说明潜在感染者B的身份是如何确认的,没有涉及值机区域和候机区域,但这两个区域其中潜在感染者B的身份确认方法与该实施例相同,在此不再赘述。具体实施时,针对确诊人员,在值机区域获得潜在被感染者为第一集合,在安检区域获得潜在被感染者为第二集合;在候机区域获得的潜在被感染者为第三集合;则将第一集合、第二集合与第三集合的并集作为潜在被感染者。
本实施例基于公共场所(例如机场航站楼)现有的监视环境,在确诊者身份确认的基础上,主动寻找公共场所中与确诊人员关联的潜在被感染者,并确定其身份。经过数据测试,其精度可以达到80%以上,在很大程度上解决了公共交通场所潜在被感染人员的寻找及确认,对疫情的防控发挥重要的作用。
第二方面,本发明还提供了一种疫情期间公共场所潜在被感染者追踪装置。参照图10所示,包括:
确诊人员搜索模块1001,用于根据确诊人员关联信息,确定所述确诊人员在公共场所进行活动的时间节点,以及与所述时间节点匹配的位置信息;
潜在被感染者搜索模块1002,用于根据所述确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
潜在被感染者身份确认模块1003,用于基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份。
需要说明的是,疫情期间公共场所潜在被感染者追踪装置与疫情期间公共场所潜在被感染者追踪方法原理相同,上文已经对疫情期间公共场所潜在被感染者追踪方法做了说明,相关之处相互参考即可,本发明在此不做赘述。
此外,本发明实施例还提供了一种疫情期间公共场所潜在被感染者追踪系统,该系统包括:少一个处理器;以及,与该至少一个处理器通信连接的存储器;其中,该存储器存储有可被该至少一个处理器执行的指令,该指令被该至少一个处理器执行,以使该至少一个处理器能够执行前述方法实施例中疫情期间公共场所潜在被感染者追踪方法。
上文已经对疫情期间公共场所潜在被感染者追踪方法做了说明,相关之处相互参考即可,本发明在此不做赘述。
以上所述,仅为本发明的具体实施方式,但本发明的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本发明揭露的技术范围内,可轻易想到的变化或替换,都应涵盖在本发明的保护范围之内。因此,本发明的保护范围应以权利要求的保护范围为准。

Claims (10)

  1. 一种疫情期间公共场所潜在被感染者追踪方法,其特征在于,包括如下步骤:
    根据确诊人员关联信息,确定所述确诊人员在公共场所进行活动的时间节点,以及与所述时间节点匹配的位置信息;
    根据所述确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
    基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份。
  2. 根据权利要求1所述的追踪方法,其特征在于,
    所述确诊人员关联信息包括:确诊人员身份信息和所乘交通工具的运营信息;并且,确定所述确诊人员在公共场所的位置信息包括如下步骤:
    根据所述确诊人员在公共场所进行活动所对应的时间节点,确定公共场所中对应的视频监控设备;
    接收在所述视频监控设备中所提取的各个时间节点的视频帧;
    在所述视频帧中,对所述确诊者进行检测、提取并持续进行跟踪确认。
  3. 根据权利要求2所述的追踪方法,其特征在于,在所述视频帧中,对所述确诊者进行检测及提取并持续进行跟踪确认,包括:
    提取确诊者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
    基于移动距离和轨迹概率构建空间概率特征;
    基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者的相似度,当相似度大于给定阈值时,以视频帧中的确诊者为基础继续进行追踪,直至描述完所述确诊者在公共场所的所有位置,对所述的确诊者在公共场所的轨迹进行确认。
  4. 根据权利要求3所述的追踪方法,其特征在于,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者包括:
    在所述视频帧中,以所述确诊者的位置为圆心,将给定的传染病传播范围映射到视频帧中的距离后,以该距离为半径构建圆形区域;
    将所述圆形区域中的行人确定为潜在被感染者;
    将所述确诊者当前的视频帧作为起始视频帧,对该视频帧所对应的时间节点向前、向后进行追踪,直至所述确诊者所在的位置满足:
    |O i-O i-1|=d i
    其中,O i表示所述确诊者当前所在的位置,O i-1为所述确诊者在其运动轨迹的另一个位置;
    以O i为圆心,d i为半径,继续构建圆形区域,再次搜索潜在的被感染者。
    重复上述步骤,直至搜索到该公共场所中所有的被感染者。
  5. 根据权利要求4所述的追踪方法,其特征在于,所述基于多目标跟踪,对所述潜在感染者进行跟踪,并结合所述潜在感染者在所述公共场所活动的时间节点确认身份,包括如下步骤:
    提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征,构建特征空间;
    基于移动距离和轨迹概率构建空间概率特征;
    基于所述特征空间以及空间概率特征,判断相邻时间点的不同视频帧中潜在被感染者的相似度,当相似度大于给定阈值时,以视频帧中的潜在被感染者为基础继续进行追踪,直至跟踪至所述潜在被感染者的进行登记的节点时,确认所述潜在被感染者的身份并上报。
  6. 根据权利要求5所述的追踪方法,其特征在于,
    所述提取确诊者在不同时间节点不同视频帧中的位置及特征后,还包括构建特征金字塔,依据所述特征金字塔确定所述特征空间;以及
    所述提取搜索到的潜在被感染者在不同时间节点不同视频帧中的位置及特征后,也包括构建特征金字塔,依据所述特征金字塔确定所述构建特征空间。
  7. 根据权利要求6所述的追踪方法,其特征在于,
    基于所述确诊者或所述潜在被感染者基于视频帧获得的特征空间以及空间概率特征,判断相邻时间点的不同视频帧中确诊者或潜在被感染者的相似度为:
    令第i时刻从视频帧中提取的特征金字塔构建的特征空间为T i=[T i1,T i2,T i3]
    令第j=i+1时刻从视频帧中提取的特征金字塔构建的特征空间为T j=[T j1,T j2,T j3];
    令所述确诊者或所述潜在被感染者在不同方向的空间概率特征为P i
    则不同视频帧中确诊者或潜在被感染者的相似度通过如下方式计算:
    Figure PCTCN2020080855-appb-100001
    Figure PCTCN2020080855-appb-100002
    Figure PCTCN2020080855-appb-100003
    Q=P 1·Q 1+P·Q 2+P 3·Q 3(P 1+P 2+P 3=1)
    其中,Q为不同视频帧中确诊者的相似度或不同视频帧中,潜在被感染者的相似度,i和j均为自然数,在实际应用中,P 1>P 2>P 3
  8. 一种疫情期间公共场所潜在被感染者追踪方法,其特征在于,
    所述公共场所为航站楼;
    将所述航站楼划分为值机区域、安检区域和候机区域;
    基于所述值机区域、所述安检区域和所述候机区域执行如权利要求1至7中任一项所述的疫情期间公共场所潜在被感染者追踪方法;
    针对所述确诊人员,在所述值机区域获得潜在被感染者为第一集合,在所述安检区域获得潜在被感染者为第二集合;在所述候机区域获得的潜在被感染者为第三集合;
    则将所述第一集合、所述第二集合与所述第三集合的并集作为潜在被感染者。
  9. 一种疫情期间公共场所潜在被感染者追踪装置,其特征在于,包括:
    确诊人员搜索模块,用于根据确诊人员关联信息,确定所述确诊人员在公共场所进行活动的时间节点,以及与所述时间节点匹配的位置信息;
    潜在被感染者搜索模块,用于根据所述确诊人员在公共场所的时间节点与位置信息,基于传染病传播机理,借助所述公共场所的视频监控设备提供的视频帧数据,搜索潜在被感染者;
    潜在被感染者身份确认模块,用于基于多目标跟踪,对所述潜在感染者进行跟踪,并结 合所述潜在感染者在所述公共场所活动的时间节点确认身份。
  10. 一种疫情期间公共场所潜在被感染者追踪系统,其特征在于,包括:
    至少一个处理器;以及,
    与所述至少一个处理器通信连接的存储器;其中,
    所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行前述任一权利要求1至9中任一项所述的疫情期间公共场所潜在被感染者追踪方法。
PCT/CN2020/080855 2020-03-09 2020-03-24 疫情期间公共场所潜在被感染者追踪方法、装置及系统 WO2020114528A2 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2020542373A JP6980928B2 (ja) 2020-03-09 2020-03-24 疫病期間中の公共場所における潜在感染者の追跡方法、装置及びシステム
US17/979,782 US20230121997A1 (en) 2020-03-09 2022-11-03 Method, device and system for tracking potential infected persons in public place during epidemic period

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010154976.X 2020-03-09
CN202010154976.XA CN111027525B (zh) 2020-03-09 2020-03-09 疫情期间公共场所潜在被感染者追踪方法、装置及系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/979,782 Continuation US20230121997A1 (en) 2020-03-09 2022-11-03 Method, device and system for tracking potential infected persons in public place during epidemic period

Publications (2)

Publication Number Publication Date
WO2020114528A2 true WO2020114528A2 (zh) 2020-06-11
WO2020114528A3 WO2020114528A3 (zh) 2021-01-14

Family

ID=70199437

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/080855 WO2020114528A2 (zh) 2020-03-09 2020-03-24 疫情期间公共场所潜在被感染者追踪方法、装置及系统

Country Status (4)

Country Link
US (1) US20230121997A1 (zh)
JP (1) JP6980928B2 (zh)
CN (1) CN111027525B (zh)
WO (1) WO2020114528A2 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112153575A (zh) * 2020-09-27 2020-12-29 深圳前海微众银行股份有限公司 一种接触者追踪方法、装置、设备及存储介质
CN112367622A (zh) * 2020-11-04 2021-02-12 南京科技职业学院 一种用于紧急公共卫生事件预警的方法
CN113132912A (zh) * 2021-04-12 2021-07-16 东南大学 基于人地数字画像的传染病密接者的双重追溯方法及系统
CN113365211A (zh) * 2021-04-19 2021-09-07 中科劲点(北京)科技有限公司 基于活动轨迹的疫情预报方法及其装置、介质及电子设备

Families Citing this family (39)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111816320A (zh) * 2020-04-22 2020-10-23 杭州溪塔科技有限公司 基于区块链的病患接触者匹配方法及系统
CN111461677A (zh) * 2020-04-27 2020-07-28 深圳市城市公共安全技术研究院有限公司 疫情期间人员管理方法、装置、终端及存储介质
CN111695048B (zh) * 2020-05-09 2023-06-02 珠海中科先进技术研究院有限公司 疫情溯源方法及介质
CN111863270A (zh) * 2020-05-20 2020-10-30 京东城市(北京)数字科技有限公司 疾病传染概率确定方法、装置、系统及存储介质
CN111627564B (zh) * 2020-05-25 2024-02-27 日立楼宇技术(广州)有限公司 用户检测模型的训练及检测方法、装置、设备和存储介质
CN111369721B (zh) * 2020-05-26 2021-02-19 浙江宇视科技有限公司 一种检疫方法、闸机单元和闸机系统
CN112530600A (zh) * 2020-05-26 2021-03-19 第四范式(北京)技术有限公司 传染病的溯源方法、装置、电子设备及存储介质
CN111695472A (zh) * 2020-06-02 2020-09-22 广州市泰睿科技有限公司 一种辅助进行信息快速录入的方法、电子设备及存储介质
CN111711676B (zh) * 2020-06-04 2021-12-10 中国联合网络通信集团有限公司 传染病患者的室内密切接触者的确认方法及服务器
CN111711925A (zh) * 2020-06-04 2020-09-25 中国联合网络通信集团有限公司 一种密切接触者判断方法及装置
CN111901753A (zh) * 2020-06-04 2020-11-06 中国联合网络通信集团有限公司 一种基于区块链的传染病密切接触者确定方法及区块链节点
CN111812688B (zh) * 2020-06-05 2021-09-14 中南大学湘雅二医院 一种基于gps的呼吸道传染病潜在携带者追踪方法
US20210391089A1 (en) * 2020-06-15 2021-12-16 Honeywell International Inc. Methods and systems for reducing a risk of spread of an illness in a building
CN111798353A (zh) * 2020-06-17 2020-10-20 广州通达汽车电气股份有限公司 一种公共交通的信息通知方法及装置
CN111798354A (zh) * 2020-06-17 2020-10-20 广州通达汽车电气股份有限公司 一种公共交通的乘车人员信息管理方法及装置
CN111710424A (zh) * 2020-06-19 2020-09-25 浙江新芮信息科技有限公司 餐饮人员健康监测方法、设备和计算机可读存储介质
JP7396214B2 (ja) * 2020-06-26 2023-12-12 トヨタ自動車株式会社 情報処理システム、情報処理方法、及び、プログラム
CN111885502B (zh) * 2020-06-28 2021-07-27 华东师范大学 一种保护隐私的疫情防控预警与追溯系统及方法
CN113873441B (zh) * 2020-06-30 2024-02-09 香港科技大学 接触者追踪的方法及相应的处理方法、系统、设备和介质
CN111818450A (zh) * 2020-07-01 2020-10-23 中国医学科学院生物医学工程研究所 人员定位方法、装置、设备及存储介质
CN112382400B (zh) * 2020-07-02 2024-02-13 浙江宇视科技有限公司 一种疫情防控方法、装置、电子设备和存储介质
CN111798342A (zh) * 2020-07-07 2020-10-20 德能森智能科技(无锡)有限公司 一种基于云平台的智慧社区系统
CN111885498B (zh) * 2020-07-13 2023-03-24 中国联合网络通信集团有限公司 一种应急响应等级的预测方法及预测装置
CN113938835B (zh) * 2020-07-13 2023-07-07 华为技术有限公司 一种确定密切接触终端设备的方法及装置
US10991185B1 (en) 2020-07-20 2021-04-27 Abbott Laboratories Digital pass verification systems and methods
CN111737585B (zh) * 2020-08-07 2020-11-17 中航信移动科技有限公司 位置的选择方法、装置、计算机设备及存储介质
CN112037927A (zh) * 2020-08-24 2020-12-04 北京金山云网络技术有限公司 与被追踪人关联的同行人确定方法、装置及电子设备
CN112135261A (zh) * 2020-09-23 2020-12-25 维沃移动通信有限公司 潜在感染者确定方法、装置、电子设备和存储介质
CN112635073B (zh) * 2020-12-18 2022-02-08 四川省疾病预防控制中心 一种密切接触者排查方法、装置、计算机设备及存储介质
CN112735605A (zh) * 2021-01-22 2021-04-30 中国银行股份有限公司 人员密切接触识别追踪分析方法及装置
CN113643824B (zh) * 2021-07-19 2024-03-26 南京理工大学 基于γ-疑似感染社区模型的疑似流行病感染人员搜索方法
CN113539512B (zh) * 2021-07-21 2023-06-16 医渡云(北京)技术有限公司 传染病传播趋势的预测方法及装置、存储介质、电子设备
CN113538678A (zh) * 2021-07-28 2021-10-22 视伴科技(北京)有限公司 一种确定潜在目标人的方法及装置
CN113539513A (zh) * 2021-08-03 2021-10-22 卫宁健康科技集团股份有限公司 患者行踪追溯方法、系统、计算机设备及存储介质
CN113327103B (zh) * 2021-08-03 2021-10-26 深圳市知酷信息技术有限公司 智慧校园疫情在线监测预警方法、系统及存储介质
CN113793690B (zh) * 2021-08-31 2024-03-12 医渡云(北京)技术有限公司 传染病的区域风险评估方法、装置、存储介质及设备
CN114067935B (zh) * 2021-11-03 2022-05-20 广西壮族自治区通信产业服务有限公司技术服务分公司 一种流行病调查方法、系统、电子设备及存储介质
CN115187916B (zh) * 2022-09-13 2023-02-17 太极计算机股份有限公司 基于时空关联的建筑内疫情防控方法、装置、设备和介质
CN117476250B (zh) * 2023-12-21 2024-03-12 华侨大学 基于多目标跟踪的流行病调查预测方法、装置及可读介质

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7993266B2 (en) * 2002-04-26 2011-08-09 Lawrence Livermore National Security, Llc Early detection of contagious diseases
CN1567329A (zh) * 2003-06-25 2005-01-19 北京行者华视网络系统集成技术有限公司 无污染健康身份证明和健康申报的方法和装置
JP2011142558A (ja) * 2010-01-08 2011-07-21 Fujitsu Ltd イメージセンサおよび撮像システム
CN101999888B (zh) * 2010-12-01 2012-07-25 北京航空航天大学 一种对体温异常者进行检测与搜寻的疫情防控系统
JP2012118775A (ja) * 2010-12-01 2012-06-21 Nec Casio Mobile Communications Ltd 食中毒管理システム、食中毒管理方法および食中毒管理プログラム
CN103970977A (zh) * 2013-02-06 2014-08-06 翦宜军 疫情监控的方法
CN103259962B (zh) * 2013-04-17 2016-02-17 深圳市捷顺科技实业股份有限公司 一种目标追踪方法和相关装置
CN103310013A (zh) * 2013-07-02 2013-09-18 北京航空航天大学 一种面向主题的网页采集系统
US9916485B1 (en) * 2015-09-09 2018-03-13 Cpg Technologies, Llc Method of managing objects using an electromagnetic guided surface waves over a terrestrial medium
CN105740615B (zh) * 2016-01-28 2018-10-16 中山大学 利用手机轨迹追踪传染源和预测传染病流行趋势的方法
KR101806521B1 (ko) * 2016-06-08 2017-12-07 (주)에코비즈 열화상 카메라를 이용하는 가축의 방역 시스템
KR101885031B1 (ko) * 2017-05-24 2018-08-03 서울시립대학교 산학협력단 교통 네트워크에 기반하여 질병 위험 지역을 감시하고 정보를 제공하는 장치 및 방법
CN107239655A (zh) * 2017-05-26 2017-10-10 陈明福 一种在线健康证明文件的识别方法
CN207067411U (zh) * 2017-08-02 2018-03-02 长沙九兴动物防检用品有限公司 一种用于畜禽防疫的智能跟踪脚标装置
CN108700468A (zh) * 2017-09-29 2018-10-23 深圳市大疆创新科技有限公司 对象检测方法、对象检测终端及计算机可读介质
CN108615214A (zh) * 2018-05-17 2018-10-02 邵春芝 出入境口岸旅客卫生检疫系统及操作方法
CN108812407B (zh) * 2018-05-23 2020-11-27 平安科技(深圳)有限公司 动物健康状态监测方法、设备及存储介质
CN109036579B (zh) * 2018-08-22 2021-11-26 泰康保险集团股份有限公司 基于区块链的信息预测方法、装置、介质及电子设备
CN110771064A (zh) * 2018-09-28 2020-02-07 深圳市大疆创新科技有限公司 一种估计方法、监听设备及计算机可读存储介质
CN109360660A (zh) * 2018-10-31 2019-02-19 河南省疾病预防控制中心 一种疾控与出行信息互联的防控方法及防控系统
CN110197502B (zh) * 2019-06-06 2021-01-22 山东工商学院 一种基于身份再识别的多目标跟踪方法及系统
CN110717414B (zh) * 2019-09-24 2023-01-03 青岛海信网络科技股份有限公司 一种目标检测追踪方法、装置及设备

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112153575A (zh) * 2020-09-27 2020-12-29 深圳前海微众银行股份有限公司 一种接触者追踪方法、装置、设备及存储介质
CN112153575B (zh) * 2020-09-27 2024-01-19 深圳前海微众银行股份有限公司 一种接触者追踪方法、装置、设备及存储介质
CN112367622A (zh) * 2020-11-04 2021-02-12 南京科技职业学院 一种用于紧急公共卫生事件预警的方法
CN113132912A (zh) * 2021-04-12 2021-07-16 东南大学 基于人地数字画像的传染病密接者的双重追溯方法及系统
CN113365211A (zh) * 2021-04-19 2021-09-07 中科劲点(北京)科技有限公司 基于活动轨迹的疫情预报方法及其装置、介质及电子设备

Also Published As

Publication number Publication date
WO2020114528A3 (zh) 2021-01-14
JP6980928B2 (ja) 2021-12-15
JP2021508896A (ja) 2021-03-11
CN111027525A (zh) 2020-04-17
US20230121997A1 (en) 2023-04-20
CN111027525B (zh) 2020-06-30

Similar Documents

Publication Publication Date Title
WO2020114528A2 (zh) 疫情期间公共场所潜在被感染者追踪方法、装置及系统
JP4970195B2 (ja) 人物追跡システム、人物追跡装置および人物追跡プログラム
US8891820B2 (en) Multi-modal sensor fusion
Cui et al. Laser-based detection and tracking of multiple people in crowds
US20220180534A1 (en) Pedestrian tracking method, computing device, pedestrian tracking system and storage medium
CN106447698B (zh) 一种基于距离传感器的多行人跟踪方法和系统
Cheng et al. Pedestrian group detection in shared space
CN112634369A (zh) 空间与或图模型生成方法、装置、电子设备和存储介质
Mandeljc et al. Tracking by identification using computer vision and radio
Himeur et al. Deep visual social distancing monitoring to combat COVID-19: A comprehensive survey
CN112613668A (zh) 基于人工智能的景区危险区域管控方法
WO2018227491A1 (zh) 视频多目标模糊数据关联方法及装置
Jiao et al. Anatomy of a multicamera video surveillance system
Prokaj et al. Using 3d scene structure to improve tracking
Wang et al. Effective multiple pedestrian tracking system in video surveillance with monocular stationary camera
Gupta et al. SSDT: distance tracking model based on deep learning
Zhao et al. Scalable building height estimation from street scene images
CN112513870A (zh) 用于利用改进的高度计算对感兴趣的人类对象进行检测、跟踪和计数的系统和方法
Levchuk et al. Activity and function recognition for moving and static objects in urban environments from wide-area persistent surveillance inputs
Liu et al. Vision meets wireless positioning: Effective person re-identification with recurrent context propagation
Wu et al. Indoor surveillance video based feature recognition for pedestrian dead reckoning
CN115690157A (zh) 一种基于Transformer的毫米波雷达行人轨迹预测方法
CN112541403B (zh) 一种利用红外摄像头的室内人员跌倒检测方法
Wang et al. DelvMap: Completing Residential Roads in Maps Based on Couriers’ Trajectories and Satellite Imagery
Lee et al. BIPS: Building Information Positioning System

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2020542373

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20730551

Country of ref document: EP

Kind code of ref document: A2