CN112036306A - System and method for realizing target tracking based on monitoring video analysis - Google Patents
System and method for realizing target tracking based on monitoring video analysis Download PDFInfo
- Publication number
- CN112036306A CN112036306A CN202010897506.2A CN202010897506A CN112036306A CN 112036306 A CN112036306 A CN 112036306A CN 202010897506 A CN202010897506 A CN 202010897506A CN 112036306 A CN112036306 A CN 112036306A
- Authority
- CN
- China
- Prior art keywords
- target
- tracking
- video
- analysis
- task
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012544 monitoring process Methods 0.000 title abstract description 14
- 238000012216 screening Methods 0.000 claims abstract description 35
- 238000010187 selection method Methods 0.000 claims description 7
- 238000012217 deletion Methods 0.000 claims description 6
- 230000037430 deletion Effects 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 4
- 238000012106 screening analysis Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 3
- 238000012732 spatial analysis Methods 0.000 abstract description 3
- 238000012731 temporal analysis Methods 0.000 abstract description 3
- 230000002123 temporal effect Effects 0.000 abstract description 3
- 239000000523 sample Substances 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/53—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Library & Information Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a target tracking system based on monitoring video analysis, which comprises a video tracking task management module, a video tracking task processing module and a video tracking task processing module, wherein the video tracking task management module is used for supporting the establishment and management of a tracking task and establishing the video tracking task; the video analysis module is used for analyzing the tracking target in the point position video within the selected time range; the target retrieval module is used for inquiring and screening the analysis result; and the clue module is used for managing clues and supporting the establishment of a tracking subtask by taking the clue picture location as the center. The invention also relates to a target tracking method based on the surveillance video analysis. By adopting the system and the method for realizing target tracking based on monitoring video analysis, disclosed by the invention, through establishing a video tracking task and a subtask, studying and judging results, establishing clues for analysis, assisting managers to perform important attention target temporal and spatial analysis to the greatest extent, and determining important characteristic information such as a motion track, an outline, an easily-identified characteristic and a suspected foot-falling range of a target.
Description
Technical Field
The invention relates to the field of surveillance videos, in particular to the field of target analysis and application of surveillance videos, and particularly relates to a system and a method for achieving target tracking based on surveillance video analysis.
Background
In recent years, with the rapid development of information technology and network technology, especially the development of city work with strong scientific and technological caution scope, video monitoring systems, electronic police systems and public security checkpoint systems all over the country develop very rapidly, and play a good role in the aspects of attacking and preventing illegal crimes by government and law system organs.
At present, target attribute analysis, image retrieval, target retrieval and control of human faces and vehicle checkpoints are mature and applied, and play a great role and value. Compared with human face and vehicle bayonet probes, the number of the monitoring video probes is larger and the monitoring video probes are widely distributed. In the aspect of the utilization of the monitoring video, the manual video investigation after the case event occurs is mainly used, the manpower investment is large, the case solving period is long, the effect is not obvious, and the value of the monitoring video cannot be fully exerted.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a system and a method for realizing target tracking based on monitoring video analysis, which have the advantages of fast processing, simple and convenient operation and wider application range.
In order to achieve the above object, the system and method for tracking a target based on surveillance video analysis according to the present invention are as follows:
the target tracking system based on the analysis of the surveillance video is mainly characterized by comprising the following components:
the video tracking task management module is used for supporting the creation and management of a tracking task, creating the video tracking task, and selecting a task type, a tracking target type, a task name, an analyzed point location video and a time range;
the video analysis module is connected with the video tracking task management module and is used for analyzing the tracking target in the point location video within the selected time range, and the analysis comprises text screening and image screening;
the target retrieval module is connected with the video analysis module and used for inquiring and screening analysis results;
the clue module is connected with the target retrieval module and used for managing clues, inquiring, deleting and opening the clues, and establishing a tracking subtask by using a plurality of algorithms with a certain clue picture place as a center to select a point range and a time range on a map in the clues;
the video tracking task management module creates a video tracking task and a subtask, calls the video analysis module to analyze, queries and screens analysis results through the target retrieval module, adds clues corresponding to the tracking task, and the clue module reproduces and analyzes a target action track.
Preferably, the management of the video tracking task management module includes search query, deletion, query of a corresponding target analysis result of the tracking task, and display of detailed information of the tracking task.
Preferably, the detailed information includes a tracking target type, a task name, and a task start-stop time, and further includes video tracking subtask information including an analysis progress and an analysis time period.
Preferably, the task types include real-time and historical, and the tracking target types include pedestrians, non-target vehicles and motor vehicles.
Preferably, the analyzing of the tracking target in the point location video by the video analyzing module includes detecting and tracking the target in the video, screening a target optimal frame, and performing structural description and image feature description on the target in the optimal frame to generate structural description text information and a high-dimensional image feature vector of each target.
Preferably, the text screening is performed according to the structured text information of the target.
Preferably, the image screening includes text retrieval and image retrieval, the text retrieval is performed according to the structural information of the uploaded image target identification, the image retrieval searches the target image by using an image, and the query results are sorted from high to low according to the feature similarity.
Preferably, the query of the analysis result is performed by the target retrieval module, and the latest retrieval result is pushed constantly as time goes on.
Preferably, the multiple algorithms select the point range on the map, including the squared figure, the four-quadrant and the annular selection method, and the corresponding algorithms are selected according to requirements.
The target tracking method based on the surveillance video analysis by using the system is mainly characterized by comprising the following steps:
(1) appointing a case place as a central point and a case time range;
(2) performing video analysis according to the specified range and time range;
(3) carrying out image screening or text screening on the analysis result;
(4) judging whether a suspected target is found, if so, manually confirming the suspected target, adding the clue image into a clue corresponding to the tracking task, otherwise, still taking the current point location as the center, dividing an analysis range by adopting different technical and tactical methods, and continuing the step (2);
(5) judging whether the tracking is finished or not, if so, researching and judging a target space-time trajectory by clues, and finishing the step; otherwise, starting the video tracking subtask by taking the clue image location in the clue as the center, and continuing to the step (2).
By adopting the system and the method for realizing target tracking based on monitoring video analysis, disclosed by the invention, through establishing a video tracking task and a subtask, studying and judging results, establishing clues for analysis, assisting managers to perform important attention target temporal and spatial analysis to the greatest extent, and determining important characteristic information such as a motion track, an outline, an easily-identified characteristic and a suspected foot-falling range of a target. The historical target tracking task restores the action track of the target in a historical period of time and is used for analyzing the historical track and the action of the target; the real-time target tracking task can grasp the action track of the target in real time and is used for real-time tracking and capturing of the target.
Drawings
Fig. 1 is a video target tracking flow chart of a target tracking method based on surveillance video analysis according to the present invention.
Fig. 2 is a range selection algorithm diagram of the target tracking method based on surveillance video analysis according to the present invention.
Detailed Description
In order to more clearly describe the technical contents of the present invention, the following further description is given in conjunction with specific embodiments.
The target tracking system based on the surveillance video analysis comprises:
the video tracking task management module is used for supporting the creation and management of a tracking task, creating the video tracking task, and selecting a task type, a tracking target type, a task name, an analyzed point location video and a time range;
the video analysis module is connected with the video tracking task management module and is used for analyzing the tracking target in the point location video within the selected time range, and the analysis comprises text screening and image screening;
the target retrieval module is connected with the video analysis module and used for inquiring and screening analysis results;
the clue module is connected with the target retrieval module and used for managing clues, inquiring, deleting and opening the clues, and establishing a tracking subtask by using a plurality of algorithms with a certain clue picture place as a center to select a point range and a time range on a map in the clues;
the video tracking task management module creates a video tracking task and a subtask, calls the video analysis module to analyze, queries and screens analysis results through the target retrieval module, adds clues corresponding to the tracking task, and the clue module reproduces and analyzes a target action track.
As a preferred embodiment of the present invention, the management of the video tracking task management module includes search query, deletion, query corresponding to the target analysis result of the tracking task, and display of detailed information of the tracking task.
As a preferred embodiment of the present invention, the detailed information includes a tracking target type, a task name, and a task start/stop time, and further includes video tracking subtask information including an analysis progress and an analysis time period.
As a preferred embodiment of the invention, the task types comprise real time and history, and the tracking target types comprise pedestrians, non-standard vehicles and motor vehicles.
As a preferred embodiment of the present invention, the analyzing of the tracking target in the point location video by the video analyzing module includes detecting and tracking the target in the video, screening a target optimal frame, and performing structural description and image feature description on the target in the optimal frame to generate structural description text information and a high-dimensional image feature vector of each target.
As a preferred embodiment of the present invention, the text filtering is performed according to the structured text information of the target.
As a preferred embodiment of the present invention, the image screening includes text retrieval and image retrieval, the text retrieval is performed according to the structured information of the uploaded image target identification, the image retrieval searches the target image in a graph, and the query results are sorted from high to low according to the feature similarity.
In a preferred embodiment of the present invention, the query of the analysis result by the target search module is performed, and the latest search result is pushed constantly as time goes by.
As a preferred embodiment of the invention, the various algorithms select the point range on the map by frames, including the Sudoku, the four-quadrant and the annular selection method, and the corresponding algorithms are selected according to requirements.
The invention discloses a target tracking method based on monitoring video analysis by using the system, which comprises the following steps:
(1) appointing a case place as a central point and a case time range;
(2) performing video analysis according to the specified range and time range;
(3) carrying out image screening or text screening on the analysis result;
(4) judging whether a suspected target is found, if so, manually confirming the suspected target, adding the clue image into a clue corresponding to the tracking task, otherwise, still taking the current point location as the center, dividing an analysis range by adopting different technical and tactical methods, and continuing the step (2);
(5) judging whether the tracking is finished or not, if so, researching and judging a target space-time trajectory by clues, and finishing the step; otherwise, starting the video tracking subtask by taking the clue image location in the clue as the center, and continuing to the step (2).
In the specific implementation manner of the invention, the video tracking system firstly establishes the video depth description task by taking the scheduled place and the scheduled time as the center, and the case handling personnel manually judge the suspected target in the preliminary analysis result and add the suspected target into the clue corresponding to the tracking task to provide new clues and conditions for the subsequent video tracking. The next video tracking subtask is established by taking the video tracking clue of the previous round as the center, and the tracking of identifying target objects such as pedestrians, non-motor vehicles and motor vehicles is realized through a plurality of video tracking tasks which are continuously related by clues, so that a necessary clue chain is provided for finding a target track.
A target tracking system based on a surveillance video comprises a video tracking task management module, a video analysis module, a target retrieval module and a clue module.
And creating a video tracking task through a video tracking task management module, and selecting a task type, a tracking target type, a task name, an analyzed point location video and a time range.
After the video tracking task is created, a tracking target in the point location video within the selected time range is analyzed through a video analysis module.
And after the video analysis is finished or in the process of video analysis, the analysis result is inquired and screened by a target retrieval module, wherein the inquiry and screening comprises character screening and image screening.
And adding a clue corresponding to the tracking task after the suspected target image is searched and manually judged through the target. In clues, a plurality of technical and tactical methods take a certain clue picture place as a center to select a point range on a map and select a time range to establish a tracking subtask.
And calling a video analysis module after the video tracking subtask is created, inquiring an analysis result through a target retrieval module after the analysis is completed or in the process of the analysis, and adding a clue corresponding to the tracking task after the suspected target is retrieved and manually judged.
And creating and collecting pictures of the suspected target through multiple rounds of video tracking subtasks, and adding the pictures into the clue. After the tracking is finished, the target action track is reproduced and analyzed in the clues.
The video tracking task management module supports the creation and management of tracking tasks, and manages the retrieval query and deletion of the tracking tasks, the query of corresponding target analysis results and the display of detailed information of the tracking tasks.
The detailed information of the video tracking task comprises the type of a tracking target (motor vehicle/non-motor vehicle/pedestrian), the name of the task and the starting and ending time of the task; including video tracking subtask information including analysis progress and analysis time period.
The task types comprise real time and history, and the tracking target types comprise pedestrians, non-standard vehicles and motor vehicles.
The method for analyzing the tracking target in the positioning video comprises the steps of detecting and tracking the target in the video, screening the optimal frame of the target, and performing structural description and image feature description on the target in the optimal frame to generate structural description text information and high-dimensional image feature vectors of each target.
The text screening is carried out according to the structured text information of the target, the image screening comprises text retrieval and image retrieval, the text retrieval is carried out according to the structured information identified by the uploaded image target, the image retrieval refers to the image searching and screening of the target image, and the screening results are sorted from high to low according to the feature similarity.
The image retrieval comprises searching images of vehicles, pedestrians and non-vehicles. The map searching of the non-motor vehicles supports person-by-person searching, vehicle-by-vehicle searching and searching of the whole person-by-person and vehicle.
The analysis result query is carried out through the target retrieval module, the latest retrieval result is constantly pushed along with the time lapse in the real-time video tracking task, and the time range T and the retrieval number N of comparison can be set in the image retrieval mode.
The multiple technical and combat methods select the point range on the map, including the Sudoku, the four-quadrant and the annular selection method, and the corresponding technical and combat methods are selected according to requirements.
The clues comprise clue pictures of the target and the time information of corresponding points, and support the starting and stopping of a certain clue picture, the deletion of the clue picture and the display of the position of the clue picture on the map; the motion trail of the target is generated according to the time and the point information, and the trail animation is automatically played on the electronic map.
The thread module supports management of all threads, query, deletion and opening of the threads.
The annular selection method is characterized in that the inner radius and the outer radius of the annular are adjustable, the four-quadrant selection method takes the map position of the selected clue picture as a reference center, the nearby area is quickly divided into four coordinate quadrants, the four quadrants are adjustable in size, and all cameras covered by one or more quadrants can be selected. The nine-grid selection method takes the map position of a selected clue as a reference center, quickly divides the nearby area into nine-grid areas, the size of the nine-grid areas can be adjusted, and all cameras covered by one or more grids can be selected according to the tracking direction of the clue track. The scope selection technique is shown in figure 2. The range selection technical and tactical method aims to flexibly select the range of the camera and provide tracking efficiency under the conditions that video analysis consumes large computing resources and the computing resources are limited.
A target tracking system based on monitoring video analysis is divided into history and real-time according to time types, and motor vehicles, non-motor vehicles and pedestrians are classified according to tracking targets.
A flow chart of the video target tracking method of the present invention is shown in fig. 1.
The invention takes the history tracking task of the non-motor vehicle as a specific embodiment, and comprises the following specific steps:
1) selecting cameras on a map according to the event occurrence time and the event occurrence place, wherein the number of the cameras is less than the maximum supported analysis road number m determined by system computing resources; selecting a historical time period, selecting a tracking target type as a non-motor vehicle, selecting a time type as history, filling in a task name, and starting a video tracking task.
2) After video analysis is finished, result query is carried out, text screening is carried out according to the provided suspected target information (the male riding the white electric motorcycle and wearing the black short sleeves), and if the type of the non-motor car is selected: electric motorcycle, select non-motor car colour: white, selecting a rider characteristic jacket type: short sleeves, jacket color: black, sex: male, screening for results. And checking in the screening result record, after manually carrying out secondary judgment to determine the suspected target, selecting the suspected target image and automatically adding the suspected target image into a clue corresponding to the video tracking task. And downloading a suspected target image.
3) Selecting a certain cable image from the cable, selecting a certain range to select a technical and tactical method, such as a ring shape, selecting a camera on a map by taking the snapping place of the cable image as a center frame, adjusting the size of the inner radius and the outer radius of the ring shape, selecting a time range, and starting a video tracking subtask 1.
4) After video analysis is completed, result query is carried out, image screening is selected, downloaded suspected target images are uploaded, the system automatically detects targets in the images, retrieval according to the whole images of people and vehicles is selected after the targets are selected, the system displays retrieval results of topN, and after manual secondary research and judgment, the suspected images are selected and added into clues.
5) And 3) repeating the steps and 4), selecting different ranges to select technical and tactical methods and retrieval modes according to needs until the tracking task is finished. The purpose of the different-range selection technical and tactical method is to flexibly select the range of the camera and provide the tracking efficiency under the conditions that the video analysis consumes large computing resources and the computing resources are limited.
6) And after the tracking task is finished, entering a clue, checking the candid record of the suspected target, selecting track generation, and playing the action track of the target on the map according to time and place.
The invention takes a real-time tracking task of a pedestrian as a specific embodiment, and comprises the following specific steps:
1) selecting cameras on a map according to the event occurrence time and the event occurrence place, wherein the number of the cameras is less than the maximum supported analysis road number m determined by system computing resources; and selecting the tracking target type as a pedestrian vehicle, selecting the time type as real time, filling in a task name, and starting a video tracking task.
2) In the analysis process, result query is carried out, and text screening is carried out according to the provided suspicion target information (white-clothing-back double-shoulder men wearing hats), such as jacket color: white, sex: male, cap: is, backpack: then, result screening is performed. And continuously refreshing the latest result in the result query, manually judging and determining the suspected target, adding the suspected target into the clue, and downloading the suspected target image.
3) Selecting a certain cable image from the cable, selecting a certain range to select a technical and tactical method, selecting a camera by taking the snapping place of the cable image as a center frame on a map, and starting a video tracking subtask 1.
4) In video analysis, result query is carried out, image screening is selected, downloaded suspected target images are uploaded, a system automatically detects targets in the images, image retrieval is selected after the targets are selected, the number of default retrieval results is 10, the time interval is 120s, and the system pushes the image retrieval results of a target library closest to 120s at intervals of 30s and ranks 10 records. After the artificial secondary judgment, a suspected image is selected and added into the clue.
5) And 3) repeating the steps and 4), selecting different ranges to select technical and tactical methods and retrieval modes according to needs until the tracking task is finished.
By adopting the system and the method for realizing target tracking based on monitoring video analysis, disclosed by the invention, through establishing a video tracking task and a subtask, studying and judging results, establishing clues for analysis, assisting managers to perform important attention target temporal and spatial analysis to the greatest extent, and determining important characteristic information such as a motion track, an outline, an easily-identified characteristic and a suspected foot-falling range of a target. The historical target tracking task restores the action track of the target in a historical period of time and is used for analyzing the historical track and the action of the target; the real-time target tracking task can grasp the action track of the target in real time and is used for real-time tracking and capturing of the target.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (10)
1. A system for realizing target tracking based on surveillance video analysis is characterized by comprising:
the video tracking task management module is used for supporting the creation and management of a tracking task, creating the video tracking task, and selecting a task type, a tracking target type, a task name, an analyzed point location video and a time range;
the video analysis module is connected with the video tracking task management module and is used for analyzing the tracking target in the point location video within the selected time range, and the analysis comprises text screening and image screening;
the target retrieval module is connected with the video analysis module and used for inquiring and screening analysis results;
the clue module is connected with the target retrieval module and used for managing clues, inquiring, deleting and opening the clues, and establishing a tracking subtask by using a plurality of algorithms with a certain clue picture place as a center to select a point range and a time range on a map in the clues;
the video tracking task management module creates a video tracking task and a subtask, calls the video analysis module to analyze, queries and screens analysis results through the target retrieval module, adds clues corresponding to the tracking task, and the clue module reproduces and analyzes a target action track.
2. The system for achieving target tracking based on surveillance video analysis as claimed in claim 1, wherein the management of the video tracking task management module includes search query, deletion, query corresponding to target analysis result of tracking task and display detailed information of tracking task.
3. The system for achieving target tracking based on surveillance video analysis as claimed in claim 2, wherein the detailed information includes tracking target type, task name, task start and stop time, and video tracking subtask information including analysis progress and analysis time period.
4. The system for target tracking based on surveillance video analytics as claimed in claim 1, wherein the task types include real-time and historical, and the tracked target types include pedestrians, non-target vehicles and motor vehicles.
5. The system for achieving target tracking based on surveillance video analysis as claimed in claim 1, wherein the video analysis module analyzes the tracking target in the point location video, including detecting and tracking the target in the video, screening the optimal frame of the target, and performing structural description and image feature description on the target in the optimal frame to generate structural description text information and high-dimensional image feature vectors of each target.
6. The system for target tracking based on surveillance video analytics as claimed in claim 1, wherein the text filtering is based on structured text information of the target.
7. The system for achieving target tracking based on surveillance video analysis as claimed in claim 1, wherein the image screening includes text retrieval and image retrieval, the text retrieval is performed according to the structured information of uploaded image target identification, the image retrieval searches the target image by image, and the query results are ranked from high to low according to feature similarity.
8. The system for achieving target tracking based on surveillance video analysis as claimed in claim 1, wherein the query of analysis results by the target retrieval module pushes the latest retrieval results continuously over time.
9. The system for achieving target tracking based on surveillance video analysis according to claim 1, wherein the plurality of algorithms select the point ranges on the map in a box mode, including a nine-square grid, a four-quadrant and a ring selection method, and the corresponding algorithms are selected according to requirements.
10. A method for implementing target tracking based on surveillance video analysis by using the system of claim 1, wherein the method comprises the following steps:
(1) appointing a case place as a central point and a case time range;
(2) performing video analysis according to the specified range and time range;
(3) carrying out image screening or text screening on the analysis result;
(4) judging whether a suspected target is found, if so, manually confirming the suspected target, adding the clue image into a clue corresponding to the tracking task, otherwise, still taking the current point location as the center, dividing an analysis range by adopting different technical and tactical methods, and continuing the step (2);
(5) judging whether the tracking is finished or not, if so, researching and judging a target space-time trajectory by clues, and finishing the step; otherwise, starting the video tracking subtask by taking the clue image location in the clue as the center, and continuing to the step (2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010897506.2A CN112036306A (en) | 2020-08-31 | 2020-08-31 | System and method for realizing target tracking based on monitoring video analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010897506.2A CN112036306A (en) | 2020-08-31 | 2020-08-31 | System and method for realizing target tracking based on monitoring video analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112036306A true CN112036306A (en) | 2020-12-04 |
Family
ID=73586461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010897506.2A Pending CN112036306A (en) | 2020-08-31 | 2020-08-31 | System and method for realizing target tracking based on monitoring video analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112036306A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824045A (en) * | 2012-11-16 | 2014-05-28 | 中兴通讯股份有限公司 | Face recognition and tracking method and face recognition and tracking system |
| CN106096577A (en) * | 2016-06-24 | 2016-11-09 | 安徽工业大学 | Target tracking system in a kind of photographic head distribution map and method for tracing |
| CN106354816A (en) * | 2016-08-30 | 2017-01-25 | 东软集团股份有限公司 | Video image processing method and video image processing device |
| CN109344267A (en) * | 2018-09-06 | 2019-02-15 | 苏州千视通视觉科技股份有限公司 | Relay method for tracing and system based on PGIS map |
-
2020
- 2020-08-31 CN CN202010897506.2A patent/CN112036306A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824045A (en) * | 2012-11-16 | 2014-05-28 | 中兴通讯股份有限公司 | Face recognition and tracking method and face recognition and tracking system |
| CN106096577A (en) * | 2016-06-24 | 2016-11-09 | 安徽工业大学 | Target tracking system in a kind of photographic head distribution map and method for tracing |
| CN106354816A (en) * | 2016-08-30 | 2017-01-25 | 东软集团股份有限公司 | Video image processing method and video image processing device |
| CN109344267A (en) * | 2018-09-06 | 2019-02-15 | 苏州千视通视觉科技股份有限公司 | Relay method for tracing and system based on PGIS map |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8300924B2 (en) | Tracker component for behavioral recognition system | |
| US20110096149A1 (en) | Video surveillance system with object tracking and retrieval | |
| CN108564052A (en) | Multi-cam dynamic human face recognition system based on MTCNN and method | |
| CN103294775B (en) | Police service cloud image recognition vehicle administrating system based on geographic space-time constraint | |
| US20100287161A1 (en) | System and related techniques for detecting and classifying features within data | |
| CN108234961B (en) | Multi-path camera coding and video stream guiding method and system | |
| CN111639620B (en) | Fire analysis method and system based on visible light image recognition | |
| US20090192990A1 (en) | Method and apparatus for realtime or near realtime video image retrieval | |
| CN110659391A (en) | Video detection method and device | |
| KR20160109761A (en) | Method and System for Recognition/Tracking Construction Equipment and Workers Using Construction-Site-Customized Image Processing | |
| CN103020275A (en) | Video analysis method based on video abstraction and video retrieval | |
| CN104239309A (en) | Video analysis retrieval service side, system and method | |
| CN112102367B (en) | Video analysis computing power real-time distribution scheduling method based on motion model | |
| Tung et al. | Large-scale object detection of images from network cameras in variable ambient lighting conditions | |
| CN112329691A (en) | Monitoring video analysis method and device, electronic equipment and storage medium | |
| Gochoo et al. | FishEye8K: a benchmark and dataset for fisheye camera object detection | |
| Hampapur | Smart video surveillance for proactive security [in the spotlight] | |
| CN112036306A (en) | System and method for realizing target tracking based on monitoring video analysis | |
| WO2023165219A1 (en) | Target retrieval method and device, and storage medium | |
| US9275140B2 (en) | Method of optimizing the search for a scene on the basis of a stream of images archived in a video database | |
| KR20170095599A (en) | System and method for video searching | |
| Zhu et al. | Person re-identification in the real scene based on the deep learning | |
| CN112347996A (en) | Scene state judgment method, device, equipment and storage medium | |
| CN112449155A (en) | Video monitoring method and system for protecting privacy of personnel | |
| CN110996043A (en) | Intelligent police service inspection management system |
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 |