CN112464757A - High-definition video-based target real-time positioning and track reconstruction method - Google Patents
High-definition video-based target real-time positioning and track reconstruction method Download PDFInfo
- Publication number
- CN112464757A CN112464757A CN202011278432.0A CN202011278432A CN112464757A CN 112464757 A CN112464757 A CN 112464757A CN 202011278432 A CN202011278432 A CN 202011278432A CN 112464757 A CN112464757 A CN 112464757A
- Authority
- CN
- China
- Prior art keywords
- camera
- video
- target person
- target
- specific target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000012544 monitoring process Methods 0.000 claims abstract description 21
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Library & Information Science (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a high-definition video-based target real-time positioning and track reconstruction method, which is used for positioning a target through cameras arranged on each road in a monitoring place and is characterized by comprising the following steps of: step S1: storing the relation between the camera and the road in a database in a graph structure; step S2: acquiring a video shot by a camera, carrying out target detection analysis on the video, and storing and recording the analyzed personnel information of the target personnel and the serial number of the camera corresponding to the video; step S3: displaying the position of the target person in real time by using the position coordinate corresponding to the camera; step S4: aiming at the selected specific target person, all videos of adjacent cameras are obtained according to a graph structure; step S5: the number of a camera of a specific target person is shot through target detection and analysis; step S6: repeating steps S4 through S5 until no specific target person is analyzed; step S7: the location of a particular target person is displayed.
Description
Technical Field
The invention belongs to the field of target monitoring, and relates to a high-definition video-based target real-time positioning and track reconstruction method.
Background
The current coordinate positioning generally relies on satellite positioning systems, such as GPS (global positioning system) and beidou satellite navigation system, through which users can accurately position the position anywhere and anytime outdoors through portable positioning devices. However, in many cases, satellite positioning systems cannot be used, for example, in the fields of pedestrian identification, target monitoring, etc., because these fields generally identify the target to be monitored through a camera, it is difficult to determine the identity of the monitored target, and it is even impossible to make the target carry a positioning device, so it is difficult to position the monitored target by satellite positioning.
The target detection technology is mainly used for detecting targets in images or videos, is developed relatively mature at present, and can detect people in cameras by embedding a target detection algorithm into a plurality of network cameras. However, the problem of positioning a specific monitoring target remains to be solved, and especially the number of cameras is often large, and a lot of computing resources are consumed for detecting the target of the specific monitoring target in a lot of shot videos.
Disclosure of Invention
In order to solve the problems, the invention provides a high-definition video-based target real-time positioning and track reconstruction method, which is used for carrying out real-time positioning and track reconstruction on characteristic pedestrians in a specific scene, and adopts the following technical scheme:
the invention provides a high-definition video-based target real-time positioning and track reconstruction method, which is used for positioning a target through cameras arranged on all roads in a monitoring place and is characterized by comprising the following steps of: step S1: storing the relationship between the camera and the road in a database in a graph structure based on the number of the camera and the position coordinates corresponding to each road; step S2: the method comprises the steps that a video shot by a camera is obtained, target detection analysis is conducted on the video, once a target person is found in the video, communication is conducted on a background management system, and the background management system stores and records person information of the target person and the serial number of the camera corresponding to the video; step S3: the background management system displays the position of the target person in real time by using the position coordinate corresponding to the camera and combining the map JS api; step S4: aiming at the selected specific target person, acquiring all videos of adjacent cameras in a database according to the relation and taking the videos as possible videos based on the number of the camera corresponding to the video of the specific target person; step S5: analyzing the possibility video of the existing specific target person from the possibility video through target detection analysis and acquiring the number of the corresponding camera; step S6: repeating the steps S4 to S5 until the possibility video does not analyze the specific target person; step S7: and displaying the position of the specific target person by combining the map JS api according to the position coordinates of all cameras which appear on the specific target person so as to show the moving track of the specific target person.
The method for real-time target positioning and track reconstruction based on high definition video provided by the invention can also have the technical characteristics that the step S1 comprises the following sub-steps: s1-1, selecting a position of a heavy point in a monitoring place as a camera arrangement point, wherein the position of the heavy point is a road intersection; step S1-2, acquiring the serial number of the camera and the corresponding road intersection positions, and acquiring the road names among the road intersection positions; and step S1-3, storing the camera numbers in a graph structure by taking the positions of the road intersections as nodes and the road names as relations.
The method for real-time target positioning and track reconstruction based on high-definition video provided by the invention can also have the technical characteristics that the video is rtsp video stream, and the step S2 comprises the following substeps: step 2-1, the camera sends the rtsp video stream to a background management system to process the rtsp video stream, and the background management system identifies a human face by using a human face detection algorithm; and 2-2, determining the identity of the target person corresponding to the face, integrating the face and the identity as a record, and further correspondingly storing the record and the serial number of the camera corresponding to the corresponding rtsp video stream in a database.
The method for real-time target positioning and track reconstruction based on high definition video provided by the invention can also have the technical characteristics that in the step 6, the position of each specific target person also shows the occurrence of the video key frame of the specific target person and the time corresponding to the video key frame.
Action and Effect of the invention
According to the high-definition video-based target real-time positioning and track reconstruction method, the position information of the camera is stored in advance, and the video shot by the camera is analyzed in real time through target detection and analysis, so that target personnel can be monitored in real time and positioned according to the position of the camera, a satellite-independent positioning system is realized, the personnel distribution and control cost can be reduced to a certain extent in specific scenes such as a campus and a garden, and the high-definition video-based target real-time positioning and track reconstruction method has high use value. Meanwhile, the serial numbers and the position information of the cameras are stored in a graph structure according to the relationship between the cameras and the roads, so when people searching in an amusement park, dangerous people monitoring in a market and the like aiming at specific target people are monitored, videos shot by adjacent cameras are obtained according to the graph structure, the action tracks of the specific target people are analyzed step by step, and through the mode, the target identification can be carried out only aiming at each camera adjacent to the camera shot with the action track of the specific target people, so that the time and labor consuming calculation aiming at all videos by using a large amount of calculation resources is avoided, and the calculation speed of the action tracks of the specific target people is greatly accelerated.
Drawings
FIG. 1 is a high definition video-based target real-time positioning and trajectory reconstruction method in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a management interface of camera information in an embodiment of the present invention;
FIG. 3 is a schematic diagram of a graph structure in an embodiment of the invention;
FIG. 4 is a diagram illustrating a screen for real-time monitoring of a target person in an embodiment of the present invention; and
fig. 5 is a schematic diagram of a screen for trajectory confirmation for a specific target person in the embodiment of the present invention.
Detailed Description
In order to make the technical means, the creation features, the achievement purposes and the effects of the present invention easy to understand, the following describes the real-time target positioning and track reconstruction method based on high definition video in detail with reference to the embodiments and the accompanying drawings.
< example >
The embodiment relates to a high-definition video-based target real-time positioning and track reconstruction method which is realized based on a target monitoring platform, wherein an operating system of the target monitoring platform is ubuntu16.04 and implemented by Python 3.6 language, a web background uses django1.1, a front end uses vue 2.0.0, a Haokangwei iDS-2PT7T80MX-D4/T3 camera is selected as a camera, and a map JS api uses a Gagde map.
Fig. 1 is a high definition video-based target real-time positioning and trajectory reconstruction method in an embodiment of the present invention.
As shown in fig. 1, the method for real-time target positioning and trajectory reconstruction based on high definition video specifically includes steps S1 to S6.
Step S1: a set of camera system is built, position coordinates and ip for installing the camera are required to be provided, and the data structure of a figure is used for representing in a computer according to the relation between the camera and a road. The step S1 specifically includes sub-steps S1-1 to S1-3:
s1-1, selecting a position of a heavy point in a monitoring place as a camera arrangement point, wherein the position of the heavy point is a road intersection;
step S1-2, acquiring the serial number of the camera and the corresponding road intersection positions, and acquiring the road names among the road intersection positions;
and step S1-3, storing the camera numbers in a graph structure by taking the positions of the road intersections as nodes and the road names as relations.
In this embodiment, as shown in fig. 2, the camera system is built in a campus, and the camera coordinates (i.e., position coordinates), the name (i.e., road name), and the IP of the camera (i.e., camera number) are stored in the database (i.e., camera list) in correspondence. In the interface shown in fig. 2, the record with id 3 in the camera list is in a clicked state, and at this time, the position of the camera is also displayed in the map on the right side by a mark according to the coordinates of the camera. In addition, through the interface, monitoring personnel or other responsible personnel can manage the data of the cameras in the garden, such as deleting, modifying and the like, when the data in the camera list are in the list
Meanwhile, as shown in fig. 3, the cameras are also stored in a graph structure according to the relationship between roads, so as to facilitate the track restoration, and fig. 3 shows a graph structure of 11 cameras numbered SXT1 to SXT11, and the connecting lines between the cameras correspond to the road relationship (i.e., road name) between the cameras. For example, at the intersection corresponding to the camera SXT8, only roads facing the cameras SXT2, SXT6, SXT7 are provided, and thus the person can move only in these three directions (in practical cases, the road relationship may be more complicated, and the diagram structure shown in fig. 3 is only one example).
Step S2: and carrying out target detection and analysis on the video captured by the camera, communicating the found target personnel to a background management system, and recording the position and personnel information of the camera by the background management system. Step S2 includes the following sub-steps:
step 2-1, the camera sends the rtsp video stream to a background management system to process the rtsp video stream, and the background management system identifies a human face by using a human face detection algorithm;
and 2-2, determining the identity of the target person corresponding to the face, integrating the face and the identity as a record, and further correspondingly storing the record and the serial number of the camera corresponding to the corresponding rtsp video stream in a database.
Step S3: and the background management system utilizes the position information of the camera and combines JS api provided by the Gade map to display in real time.
As shown in fig. 4, a map of the campus is displayed on the left side of the view, and all historically monitored target persons are displayed in the "target person record" table on the right side of the map. When the target person is monitored in the camera, the video frame of the monitored target person can be displayed in real time in the rightmost real-time picture, and the picture can be displayed in a map at the same time, so that the position of the target person can be conveniently mastered by the monitoring person.
In this embodiment, the above steps S2 and S3 are performed to analyze the video captured by the camera in real time and show the detected target person in real time, and in addition, the embodiment may also perform trajectory analysis on the specific target person to be monitored through the following steps S4 to S6 for the offline video (i.e., the video captured by the camera and stored and archived).
Step S4: and aiming at the selected specific target person, acquiring all videos of adjacent cameras in the database according to the relation as the possibility videos based on the number of the camera corresponding to the video of the specific target person.
In this embodiment, a specific target person is selected by a monitoring person through the target monitoring platform, and in actual use, the monitoring person calls the offline stored video and the detected target person in the video to specify the target person needing to be monitored, so that at this time, the number of the camera corresponding to the offline video can be obtained.
Next, all neighboring cameras are acquired from the database of the graph structure according to the camera numbers, taking the graph structure shown in fig. 4 as an example, the current camera number is camera SXT8, the acquired numbers of the neighboring cameras are SXT2, SXT6 and SXT7, and all videos captured by the three cameras and stored offline are taken as videos (i.e., possibility videos) in which a specific target person may appear.
Step S5: and analyzing the possibility video of the specific target person from the possibility video through target detection and analysis, and acquiring the number of the corresponding camera.
In step S5 of this embodiment, after each of the possible videos is sequentially analyzed through the target detection analysis, the video frame of the specific target person and the number of the camera recording the video can be obtained through analysis.
Step S6: repeating the steps S4 to S5 until the possibility video does not analyze the specific target person.
Step S7: and displaying the position of the specific target person by combining the map JS api according to the position coordinates of all cameras of the specific target person, thereby showing the moving track of the specific target person.
In this embodiment, by repeating steps S4 and S5, the peripheral cameras can be searched from the initial camera in the database of the graph structure step by step along the action track of the specific target person, and the sequence, time and position of the specific target person passing through the cameras are determined according to the video frames searched to contain the specific target person.
Next, the trajectory of the specific target person can be obtained from the sequence, time and location, and displayed on a map, as shown in fig. 4, in which a map of the campus is shown on the left side and all the historically monitored target persons are shown in the "target person record" on the right side of the map. When the monitoring person clicks the target person with the number 7 as the specific target person, the trajectory processing is performed in the above steps S4 to S7, so that the time and position of the camera of the specific target person and the video frame containing the specific target person are further displayed in the "trajectory information" on the right side of the "target person record", and the action trajectory is marked on the left map based on the position coordinates of the camera.
Examples effects and effects
According to the high-definition video-based target real-time positioning and track reconstruction method provided by the embodiment, the position information of the camera is stored in advance, and the video shot by the camera is analyzed in real time through target detection and analysis, so that target personnel can be monitored in real time and positioned according to the position of the camera, a satellite-independent positioning system is realized, the personnel distribution cost can be reduced to a certain extent in specific scenes such as a campus and a garden, and the method has high use value. Meanwhile, the serial numbers and the position information of the cameras are stored in a graph structure according to the relationship between the cameras and the roads, so when people searching in an amusement park, dangerous people monitoring in a market and the like aiming at specific target people are monitored, videos shot by adjacent cameras are obtained according to the graph structure, the action tracks of the specific target people are analyzed step by step, and through the mode, the target identification can be carried out only aiming at each camera adjacent to the camera shot with the action track of the specific target people, so that the time and labor consuming calculation aiming at all videos by using a large amount of calculation resources is avoided, and the calculation speed of the action tracks of the specific target people is greatly accelerated.
The above-described embodiments are merely illustrative of specific embodiments of the present invention, and the present invention is not limited to the description of the above-described embodiments.
Claims (4)
1. A target real-time positioning and track reconstruction method based on high definition video is used for positioning a target through cameras arranged on all roads in a monitoring place, and is characterized by comprising the following steps:
step S1: storing the relationship between the camera and the road in a database in a graph structure based on the number of the camera and the position coordinates corresponding to each road;
step S2: acquiring a video shot by the camera and carrying out target detection analysis on the video, communicating with a background management system once a target person is found in the video, and storing and recording the person information of the target person and the serial number of the camera corresponding to the video by the background management system;
step S3: the background management system displays the position of the target person in real time by using the position coordinate corresponding to the camera and combining a map JS api;
step S4: aiming at the selected specific target person, acquiring all videos of adjacent cameras in the database according to the relation and taking the videos as possible videos based on the number of the camera corresponding to the video of the specific target person;
step S5: analyzing the possibility video of the specific target person from the possibility video through the target detection analysis, and acquiring the number of the corresponding camera;
step S6: repeating the steps S4 through S5 until the specific target person is not analyzed in the likelihood video;
step S7: according to all the position coordinates of the camera which appear on the specific target person, the position of the specific target person is displayed by combining a map JS api, so that the movement track of the specific target person is shown.
2. The high-definition video-based target real-time positioning and trajectory reconstruction method according to claim 1, characterized in that:
wherein the step S1 includes the following sub-steps:
step S1-1, selecting a heavy point position as a camera arrangement point in the monitoring place, wherein the heavy point position is a road intersection;
step S1-2, acquiring the serial number of the camera and the corresponding road intersection positions, and acquiring the road names among the road intersection positions;
and step S1-3, storing the camera number in a graph structure by taking the intersection position as a node and the road name as a relation.
3. The high-definition video-based target real-time positioning and trajectory reconstruction method according to claim 1, characterized in that:
wherein the video is an rtsp video stream,
the step S2 includes the following sub-steps:
step 2-1, the camera sends the rtsp video stream into the background management system to process the rtsp video stream, and the background management system identifies a human face by using a human face detection algorithm;
and 2-2, determining the identity of the target person corresponding to the face, integrating the face and the identity as a record, and further correspondingly storing the record and the serial number of the camera corresponding to the rtsp video stream in the database.
4. The high-definition video-based target real-time positioning and trajectory reconstruction method according to claim 1, characterized in that:
in step 6, the position of each specific target person also shows the occurrence of the video key frame of the specific target person and the time corresponding to the video key frame.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011278432.0A CN112464757A (en) | 2020-11-16 | 2020-11-16 | High-definition video-based target real-time positioning and track reconstruction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011278432.0A CN112464757A (en) | 2020-11-16 | 2020-11-16 | High-definition video-based target real-time positioning and track reconstruction method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112464757A true CN112464757A (en) | 2021-03-09 |
Family
ID=74837269
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011278432.0A Pending CN112464757A (en) | 2020-11-16 | 2020-11-16 | High-definition video-based target real-time positioning and track reconstruction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112464757A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114422828A (en) * | 2022-01-26 | 2022-04-29 | 复旦大学 | Real-time positioning and track reconstruction method based on high-definition camera |
CN115205341A (en) * | 2022-09-14 | 2022-10-18 | 南京北新智能科技有限公司 | Motion trajectory generation method based on A-Star |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104581000A (en) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | Method for rapidly retrieving motional trajectory of interested video target |
-
2020
- 2020-11-16 CN CN202011278432.0A patent/CN112464757A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104581000A (en) * | 2013-10-12 | 2015-04-29 | 北京航天长峰科技工业集团有限公司 | Method for rapidly retrieving motional trajectory of interested video target |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114422828A (en) * | 2022-01-26 | 2022-04-29 | 复旦大学 | Real-time positioning and track reconstruction method based on high-definition camera |
CN115205341A (en) * | 2022-09-14 | 2022-10-18 | 南京北新智能科技有限公司 | Motion trajectory generation method based on A-Star |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Van Etten et al. | The multi-temporal urban development spacenet dataset | |
CN107370983B (en) | method and device for acquiring track of video monitoring system | |
US9710924B2 (en) | Field of view determiner | |
CN102436738B (en) | Traffic monitoring device based on unmanned aerial vehicle (UAV) | |
US20200374491A1 (en) | Forensic video exploitation and analysis tools | |
CN108446015B (en) | Exhibition method and exhibition system based on mixed reality | |
CN106446002A (en) | Moving target-based video retrieval method for track in map | |
CN106295598A (en) | A kind of across photographic head method for tracking target and device | |
CN102819578A (en) | Suspected target analyzing system and method by video investigation | |
CN111160243A (en) | Passenger flow volume statistical method and related product | |
Bouma et al. | Real-time tracking and fast retrieval of persons in multiple surveillance cameras of a shopping mall | |
KR101645959B1 (en) | The Apparatus and Method for Tracking Objects Based on Multiple Overhead Cameras and a Site Map | |
JP6013923B2 (en) | System and method for browsing and searching for video episodes | |
CN107241572A (en) | Student's real training video frequency tracking evaluation system | |
Höferlin et al. | Uncertainty-aware video visual analytics of tracked moving objects | |
CN112464757A (en) | High-definition video-based target real-time positioning and track reconstruction method | |
CN113255477A (en) | Comprehensive management system and method for pedestrian video images | |
Lin et al. | Moving camera analytics: Emerging scenarios, challenges, and applications | |
CN113469080A (en) | Individual, group and scene interactive collaborative perception method, system and equipment | |
CN113312951A (en) | Dynamic video target tracking system, related method, device and equipment | |
CN115830076A (en) | Intelligent personnel trajectory video analysis system | |
CN113887384B (en) | Pedestrian track analysis method, device, equipment and medium based on multi-track fusion | |
CN115424320A (en) | Method and system for displaying movement track of person in three-dimensional scene | |
CN106777078A (en) | A kind of video retrieval method and system based on information database | |
CN110781797B (en) | Labeling method and device and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210309 |