CN110532923A - A kind of personage's trajectory retrieval method and its system - Google Patents
A kind of personage's trajectory retrieval method and its system Download PDFInfo
- Publication number
- CN110532923A CN110532923A CN201910774887.2A CN201910774887A CN110532923A CN 110532923 A CN110532923 A CN 110532923A CN 201910774887 A CN201910774887 A CN 201910774887A CN 110532923 A CN110532923 A CN 110532923A
- Authority
- CN
- China
- Prior art keywords
- face
- camera
- target person
- characteristic value
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; 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
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- 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
-
- 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
-
- 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
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
- G06V40/168—Feature extraction; Face representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Library & Information Science (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present invention provides a kind of personage's trajectory retrieval method and its system, and this method is applied to the system, which comprises according to target camera number and period acquisition of information video data;When there is target person in the video data, the facial image about target person in the video data is intercepted;Recognition of face is carried out to the facial image, calculates the face characteristic value of target person;Face retrieval index is generated according to the face characteristic value;The multiple image data about target person are obtained according to the face characteristic value and the face retrieval search index face database;It obtains and data, the shooting time of the multiple image data and shooting location generates personage's motion profile according to the map.Implement the present invention, enabling to can be grid-connected to non-grid-connected camera progress, makes it possible to image recognition personage, and can support quickly to retrieve image recognition personage.
Description
Technical field
Route tracer technique of the present invention field, in particular to a kind of personage's trajectory retrieval method and its system.
Background technique
The one kind of image recognition technology as information discriminating technology, can effectively identify, face, dress, vehicle, license plate etc.
Characteristic information.Image recognition technology has been applied to multiple fields in life, such as personage's trajectory retrieval.But it is now traditional
Personage's trajectory retrieval still rests on artificial carry out video play-reverse, and time-consuming and laborious, timeliness is not high.If being intended only as criminal investigation
Or searching person and object, application demand can also be met, but if being user's purchasing habits follow-up investigation for multiple people markets,
Then it is unable to satisfy application demand.
Also, in field of video monitoring, by the development of several years, also each community, place are assembled with very much
Camera, wherein having the camera of security protection, camera of traffic etc..For these cameras, there is independent part camera,
Then there are no grid-connected for the system being made of multi-cam, the camera for having small part grid-connected, but most of camera.
Summary of the invention
The present invention is directed to propose a kind of personage's trajectory retrieval method and its system so that can to non-grid-connected camera into
Row is grid-connected, makes it possible to image recognition personage, and can support quickly to retrieve image recognition personage.
In a first aspect, the embodiment of the present invention proposes a kind of personage's trajectory retrieval method, comprising:
According to target camera number and period acquisition of information video data;
When there is target person in the video data, the face about target person in the video data is intercepted
Image;
Recognition of face is carried out to the facial image, calculates the face characteristic value of target person;
Face retrieval index is generated according to the face characteristic value;
It is obtained according to the face characteristic value and the face retrieval search index face database about target person
Multiple image data;
It obtains and data, the shooting time of the multiple image data and shooting location generates personage's movement rail according to the map
Mark.
Wherein, the face database real-time reception and storage and distribution view captured by multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index frequency accordingly and in video data, each video data packet
Multiple image data are included, every frame image data includes image, shooting time and the camera number of camera shooting, wherein often
One camera has unique number.
Wherein, it is obtained according to the face characteristic value and the face retrieval search index face database about target person
The multiple image data of object include:
Step S1, centered on the camera site of center, pre-determined distance is that radius generates border circular areas;Wherein, institute
State the center camera that the corresponding camera site of target camera number is query context when inquiry starts;
Step S2, the circle according to the face characteristic value and the face retrieval search index face database
Video data captured by camera in domain obtains the multiple image data about target person;
If step S3, inquiring includes image about target person in the video data of a certain camera in border circular areas
Data, then using the camera as center camera return step S201.
Wherein, described to be obtained according to the face characteristic value and the face retrieval search index face database about mesh
Mark personage multiple image data include:
The face characteristic value and face retrieval index are sent to each child servers;
According to the face characteristic value and the face retrieval search index, respectively face database obtains each child servers
Multiple image data about target person;
Receive the multiple image data about target person that each child servers return;
Wherein, the face database real-time reception of each child servers and storage and distribution are in multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index, described each in captured video data and video data
Video data includes multiple image data, and every frame image data includes image, shooting time and the camera of camera shooting
Number, wherein each camera has unique number.
Wherein, it obtains and data, the shooting time of the multiple image data and shooting location generates personage according to the map
Motion profile includes:
The multiple image data are ranked up by shooting time;
The shooting location that target person sequentially occurs is determined according to ranking results;
The shooting location sequentially occurred according to target person generates personage's motion profile of target person on map.
Second aspect, the embodiment of the present invention propose a kind of personage's trajectory retrieval system, including primary server, the main service
Device includes:
Video acquisition unit, for according to target camera number and period acquisition of information video data;
Video processing unit, for intercepting in the video data when there is target person in the video data
Facial image about target person;
Image identification unit calculates the face characteristic value of target person for carrying out recognition of face to the facial image;
Generation unit is indexed, for generating face retrieval index according to the face characteristic value;
Data query unit is closed according to the face characteristic value and the face retrieval search index face database
In the multiple image data of target person;
Track generation unit, for obtaining and data, the shooting time of the multiple image data and shooting according to the map
Place generates personage's motion profile.
Wherein, the face database real-time reception and storage and distribution view captured by multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index frequency accordingly and in video data, each video data packet
Multiple image data are included, every frame image data includes image, shooting time and the camera number of camera shooting, wherein often
One camera has unique number.
Wherein, the data query unit includes:
Area generation unit, for centered on the camera site of center, pre-determined distance to be that radius generates circle
Domain;Wherein, the corresponding camera site of the target camera number is that the center of query context when inquiry starts is taken the photograph
As head;
Subelement is inquired, for according to institute in the face characteristic value and the face retrieval search index face database
Video data captured by the camera in border circular areas is stated, the multiple image data about target person are obtained;
Judging unit, for judging whether the data query unit inquires the video of a certain camera in border circular areas
It include the image data about target person in data, if so, in determining that the camera is new center camera and generates
Heart camera more new information is sent to the Area generation unit.
It wherein, further include each child servers;
The data query unit includes:
Data transmission unit, for the face characteristic value and face retrieval index to be sent to each child servers;
Data transmission unit receives the multiple image data about target person that each child servers return;
Each child servers are respectively used to according to the respective face number of the face characteristic value and the face retrieval search index
The multiple image data about target person are obtained according to library;
Wherein, the face database real-time reception of each child servers and storage and distribution are in multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index, described each in captured video data and video data
Video data includes multiple image data, and every frame image data includes image, shooting time and the camera of camera shooting
Number, wherein each camera has unique number.
Wherein, the track generation unit includes:
Sorting subunit, for being ranked up by shooting time to the multiple image data;
Place subelement, for determining shooting location that target person sequentially occurs according to ranking results;
Track subelement, the shooting location for sequentially being occurred according to target person generate the people of target person on map
Object motion profile.
The embodiment of the present invention proposes a kind of personage's trajectory retrieval method and its system, by by camera each in region and clothes
Business device connection, establishes region cameras monitoring network, and server set passes through target person at image identification unit and database
Face characteristic value and face retrieval index can rapidly inquire the facial image in database about target person, according to face
The shooting time of image and place can determine the motion track of target person, and combine map datum, finally raw on map
At the motion profile of target person.Therefore, implement the embodiment of the present invention, enable to carry out simultaneously non-grid-connected camera
Net makes it possible to image recognition personage, and can support quickly to retrieve image recognition personage.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that being emerged from by implementing the present invention.The objectives and other advantages of the invention can by specification,
Specifically noted structure is achieved and obtained in claims and attached drawing.Certainly, implement any of the products of the present invention or
Method does not necessarily require achieving all the advantages described above at the same time.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of personage's trajectory retrieval method flow schematic diagram described in the embodiment of the present invention one.
Fig. 2 is a kind of personage's trajectory retrieval system schematic described in the embodiment of the present invention two.
Appended drawing reference:
Video acquisition unit -1, video processing unit -2, image identification unit -3 index generation unit -4, data query
Unit -5, track generation unit -6.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear and complete
Ground description, it is clear that described embodiment is only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art without making creative work it is obtained it is all its
Its embodiment, shall fall within the protection scope of the present invention.
Here, it should also be noted that, in order to avoid having obscured the present invention because of unnecessary details, in the accompanying drawings only
Show with closely related structure according to the solution of the present invention, and be omitted and the little other details of relationship of the present invention.
Embodiment one
The embodiment of the present invention one proposes a kind of personage's trajectory retrieval method, can be applied to personage described in embodiment two
Trajectory retrieval system, the system include a primary server, the primary server include video acquisition unit, video processing unit,
Image identification unit, index generation unit, data query unit and track generation unit.Fig. 1 is process described in embodiment one
Schematic diagram.
Refering to fig. 1, which comprises
Step S101, according to target camera number and period acquisition of information video data.
Specifically, user inputs target camera number and period by client in step when needing to inquire
Information, is sent to the primary server in cloud, and the video acquisition unit of primary server transfers corresponding period target from database
Video data captured by camera is simultaneously sent to client.
Step S102, when there is target person in the video data, intercept in the video data about target
The facial image of personage.
Specifically, the video data that user returns to primary server at client in step is checked, judgement should
Whether there is target person in video data, if so, user sends a confirmation message, main clothes to primary server by client
The video processing unit of business device intercepts the facial image about target person in the video data according to the confirmation message.
Step S103, recognition of face is carried out to the facial image, calculates the face characteristic value of target person.
Specifically, primary server identify using the facial image that image identification unit intercepts step S102
Face characteristic value calculating is carried out to face corresponding with target person, and to the face.It should be noted that the meter of face characteristic value
Calculation mode is not present subject matter, and the present embodiment is intended to carry out the retrieval of personage's motion profile, this field using face characteristic value
The calculation method of the known related face characteristic value of technical staff, can choose any face characteristic value according to actual needs and calculate
Method is applied in the present invention, therefore, does not carry out particular technique means to the calculation method of face characteristic value in the present embodiment
It limits.
Step S104, face retrieval index is generated according to the face characteristic value.
Specifically, face retrieval index is generated by index generation unit in step, the wherein life of face retrieval index
At being for the ease of quick-searching, face retrieval index is a kind of face character, and the face character of target person includes but unlimited
In gender, nose types, eyes type, mouth type etc., for example, target person is male, then it is available fast according to search index
Speed inquires the relevant face image data of male.
Step S105, it is obtained according to the face characteristic value and the face retrieval search index face database about mesh
Mark the multiple image data of personage.
View captured by multiple cameras of different location in region is distributed in specifically, being stored in face database
Frequency evidence, each video data include multiple image data, and every frame includes that the image data of face can all extract corresponding people
Face characteristic value and face retrieval index, are then stored in face database, in case inquiring at any time.Therefore, when getting target
When face characteristic value and the face retrieval index of personage, data query unit is examined according to the face characteristic value and face of target person
Rustling sound draws inquiry face database, can find and index matched multiframe figure with the face characteristic value of target person and face retrieval
As data.
Step S106, it obtains and data, the shooting time of the multiple image data and shooting location generates according to the map
Personage's motion profile.
Specifically, can determine target person according to the shooting time of the multiple image data and shooting location in step
Which place object reached according to time sequencing, and track generation unit generates personage's motion profile using map datum on map.
Wherein, the face database real-time reception and storage and distribution view captured by multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index frequency accordingly and in video data, each video data packet
Multiple image data are included, every frame image data includes image, shooting time and the camera number of camera shooting, wherein often
One camera has unique number.
Based on one content of above embodiments, as one of improvement, in one embodiment, the step S105 includes:
Step S201, centered on the camera site of center, pre-determined distance is that radius generates border circular areas;Wherein,
The corresponding camera site of the target camera number is the center camera of query context when inquiry starts.
Specifically, the pre-determined distance can be any value in 2000-3000 meters of ranges.If using all take the photograph is traversed
As head video data, then calculation amount is too big, and retrieval time is long, and working efficiency is low.And target person is in moving process, it is most absolutely
Number situation moves on the ground, and maximum probability will not walk out border circular areas within the section time, and therefore, the present embodiment proposes logical
It crosses and retrieves whether camera in border circular areas takes target person first, rather than use and traverse all camera video numbers
According to improve recall precision, reduction calculation amount realizes quick-searching.
Step S202, the circle according to the face characteristic value and the face retrieval search index face database
Video data captured by camera in region obtains the multiple image data about target person.
Specifically, as shown in the above, if an image data includes facial image, image data can exist and it
Corresponding face characteristic value and face retrieval index can index accordingly according to the face characteristic value and the face retrieval
Video data captured by camera in border circular areas described in quick search face database, obtains about target person
Multiple image data.
If step S203, inquiring includes figure about target person in the video data of a certain camera in border circular areas
As data, then using the camera as center camera return step S201, then again according to center camera in step S203
New border circular areas is generated, and continues to inquire.It is understood that in query process, the change procedure of border circular areas is also
It is corresponding with the motion profile of target person.
As one of improvement, in another embodiment, the step S105 includes:
Step S301, the face characteristic value and face retrieval index are sent to each child servers.
Specifically, each community, place are assembled with many cameras, wherein there is the camera of security protection, traffic is taken the photograph
As first-class etc..If all camera data in one region are handled by a server, possible server will appear
Therefore the case where overload, proposes distributed server in present embodiment, multiple child servers are arranged, will be one big
The multiple zonules of region division, such as an administrative area is divided into five zonules, each cell by all directions and middle part
A child servers are arranged in domain, and all cameras in child servers and zonule communicate to connect, and establish monitoring network.Work as needs
When inquiry, primary server sends inquiry request information to each child servers, and the inquiry request includes face characteristic value and face
Search index.
Step S302, each child servers are according to the respective face number of the face characteristic value and the face retrieval search index
The multiple image data about target person are obtained according to library;
Specifically, each child servers are after receiving inquiry request, according to the face characteristic value and the face
Whether the video data that search index is inquired in respective face database includes target person.
Step S303, the multiple image data about target person that each child servers return are received.
Wherein, the face database real-time reception of each child servers and storage and distribution are in multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index, described each in captured video data and video data
Video data includes multiple image data, and every frame image data includes image, shooting time and the camera of camera shooting
Number, wherein each camera has unique number.
Wherein, each child servers query process is specific as follows:
Step S401, centered on the camera site of center, pre-determined distance is that radius generates border circular areas;Wherein,
The corresponding camera site of the target camera number is the center camera of query context when inquiry starts.
Specifically, the pre-determined distance can be any value in 2000-3000 meters of ranges.If using all take the photograph is traversed
As head video data, then calculation amount is too big, and retrieval time is long, and working efficiency is low.And target person is in moving process, it is most absolutely
Number situation moves on the ground, and maximum probability will not walk out border circular areas within the section time, and therefore, the present embodiment proposes logical
It crosses and retrieves whether camera in border circular areas takes target person first, rather than use and traverse all camera video numbers
According to improve recall precision, reduction calculation amount realizes quick-searching.
Step S402, the circle according to the face characteristic value and the face retrieval search index face database
Video data captured by camera in region obtains the multiple image data about target person.
Specifically, as shown in the above, if an image data includes facial image, image data can exist and it
Corresponding face characteristic value and face retrieval index can index accordingly according to the face characteristic value and the face retrieval
Video data captured by camera in border circular areas described in quick search face database, obtains about target person
Multiple image data.
If step S403, inquiring includes figure about target person in the video data of a certain camera in border circular areas
As data, then using the camera as center camera return step S401, then again according to center camera in step S403
New border circular areas is generated, and continues to inquire.It is understood that in query process, the change procedure of border circular areas is also
It is corresponding with the motion profile of target person.
Based on one content of above embodiments, as one of improvement, the step S106 includes:
Step S501, the multiple image data are ranked up by shooting time.
Specifically, each frame image data all has the shooting time of image.
Step S502, shooting location that target person sequentially occurs is determined according to ranking results.
Specifically, by multiple image data according to time sequence after, can learn that target person temporally elapses successively
Which shooting location appeared in.
Step S503, the shooting location sequentially occurred according to target person generates personage's movement of target person on map
Track.
Specifically, can include the map datum of one's respective area by means of existing map software, map software, pass through
Map software can generate the motion track of target person on map.
The embodiment of the present invention increases the preservation of face recognition features' value and the guarantor of search index on the basis of tradition monitoring
It deposits.When needed, it is only necessary to confirm someone camera shooting inside system, it can plant the rail task quickly through feature
Mark quick-searching and the personage's motion profile for generating target person.
Embodiment two
The embodiment of the present invention two proposes a kind of personage's trajectory retrieval system, a kind of personage's rail as described in Fig. 2 is embodiment two
The schematic diagram of mark searching system.
Referring to Fig.2, the system comprises primary server, the primary server includes:
Video acquisition unit 1, for according to target camera number and period acquisition of information video data;
Video processing unit 2, for intercepting in the video data when there is target person in the video data
Facial image about target person;
Image identification unit 3 calculates the face characteristic of target person for carrying out recognition of face to the facial image
Value;
Generation unit 4 is indexed, for generating face retrieval index according to the face characteristic value;
Data query unit 5 is obtained according to the face characteristic value and the face retrieval search index face database
Multiple image data about target person;
Track generation unit 6, for obtaining and data, the shooting time of the multiple image data and shooting according to the map
Place generates personage's motion profile.
Based on two or more content of embodiment, as one of improvement, in one embodiment, the face database connects in real time
Receive every frame facial image in simultaneously storage and distribution video data captured by multiple cameras of different location and video data
Face characteristic value and face retrieval index, each video data includes multiple image data, and every frame image data includes
Image, shooting time and the camera number of camera shooting, wherein each camera has unique number.
Wherein, the data query unit includes:
Area generation unit, for centered on the camera site of center, pre-determined distance to be that radius generates circle
Domain;Wherein, the corresponding camera site of the target camera number is that the center of query context when inquiry starts is taken the photograph
As head;
Subelement is inquired, for according to institute in the face characteristic value and the face retrieval search index face database
Video data captured by the camera in border circular areas is stated, the multiple image data about target person are obtained;
Judging unit, for judging whether the data query unit inquires the video of a certain camera in border circular areas
It include the image data about target person in data, if so, in determining that the camera is new center camera and generates
Heart camera more new information is sent to the Area generation unit.
It further include each child servers in another embodiment as one of improvement;
The data query unit includes:
Data transmission unit, for the face characteristic value and face retrieval index to be sent to each child servers;
Data transmission unit receives the multiple image data about target person that each child servers return;
Each child servers are respectively used to according to the respective face number of the face characteristic value and the face retrieval search index
The multiple image data about target person are obtained according to library;
Wherein, the face database real-time reception of each child servers and storage and distribution are in multiple cameras of different location
The face characteristic value of every frame facial image and face retrieval index, described each in captured video data and video data
Video data includes multiple image data, and every frame image data includes image, shooting time and the camera of camera shooting
Number, wherein each camera has unique number.
Based on two or more content of embodiment, as one of improvement, the track generation unit includes:
Sorting subunit, for being ranked up by shooting time to the multiple image data;
Place subelement, for determining shooting location that target person sequentially occurs according to ranking results;
Track subelement, the shooting location for sequentially being occurred according to target person generate the people of target person on map
Object motion profile.
It should be noted that system described in embodiment two is corresponding with one the method for embodiment, therefore, it is described in embodiment two
Non- detailed portion of uniting can be obtained refering to the content of one the method for embodiment, and details are not described herein again.
By the description above with respect to embodiment it is found that the embodiment of the present invention propose a kind of personage's trajectory retrieval method and its
System establishes region cameras monitoring network, server integrated image is known by the way that camera each in region to be connect with server
Other unit and database can be inquired rapidly in database by face characteristic value and the face retrieval index of target person and be closed
The motion track of target person can be determined according to the shooting time of facial image and place in the facial image of target person,
And map datum is combined, the motion profile of target person is finally generated on map.Therefore, implement the embodiment of the present invention, it can
Allow to carry out non-grid-connected camera grid-connected, makes it possible to image recognition personage, and can support quickly to image knowledge
Others retrieves object.
It should be noted that it is based on present disclosure, it is apparent to those skilled in the art that, the present invention can borrow
Software and required common hardware is helped to realize, naturally it is also possible to which the former is more preferably by hardware realization, but in many cases
Embodiment.Based on this understanding, technical solution of the present invention substantially the part that contributes to existing technology in other words
It can be embodied in the form of software products, which may be stored in a computer readable storage medium,
Floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (RandomAccess such as computer
Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) realize method/system described in each embodiment of the present invention.
The above is only a specific embodiment of the invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of personage's trajectory retrieval method characterized by comprising
According to target camera number and period acquisition of information video data;
When there is target person in the video data, the face figure about target person in the video data is intercepted
Picture;
Recognition of face is carried out to the facial image, calculates the face characteristic value of target person;
Face retrieval index is generated according to the face characteristic value;
The multiframe about target person is obtained according to the face characteristic value and the face retrieval search index face database
Image data;
It obtains and data, the shooting time of the multiple image data and shooting location generates personage's motion profile according to the map.
2. personage's trajectory retrieval method as described in claim 1, which is characterized in that the face database real-time reception is simultaneously deposited
Storage is distributed in the face of every frame facial image in video data captured by multiple cameras of different location and video data
Characteristic value and face retrieval index, each video data includes multiple image data, and every frame image data includes camera
Image, shooting time and the camera number of shooting, wherein each camera has unique number.
3. personage's trajectory retrieval method as claimed in claim 2, which is characterized in that according to the face characteristic value and the people
Face search index inquiry face database obtains
Step S1, centered on the camera site of center, pre-determined distance is that radius generates border circular areas;Wherein, the mesh
The corresponding camera site of mark camera number is the center camera of query context when inquiry starts;
Step S2, in the border circular areas according to the face characteristic value and the face retrieval search index face database
Camera captured by video data, obtain the multiple image data about target person;
If step S3, inquiring includes picture number about target person in the video data of a certain camera in border circular areas
According to then using the camera as center camera return step S201.
4. personage's trajectory retrieval method as described in claim 1, which is characterized in that described according to the face characteristic value and institute
It states face retrieval search index face database and obtains and include: about the multiple image data of target person
The face characteristic value and face retrieval index are sent to each child servers;
Each child servers according to the face characteristic value and the face retrieval search index respectively face database obtain about
The multiple image data of target person;
Receive the multiple image data about target person that each child servers return;
Wherein, the face database real-time reception of each child servers simultaneously clapped by multiple cameras of the storage and distribution in different location
The face characteristic value of every frame facial image and face retrieval index, each video in the video data and video data taken the photograph
Data include multiple image data, and every frame image data includes image, shooting time and the camera number of camera shooting,
Wherein each camera has unique number.
5. personage's trajectory retrieval method according to any one of claims 1-4, which is characterized in that obtain and count according to the map
Generating personage's motion profile according to, the shooting time of the multiple image data and shooting location includes:
The multiple image data are ranked up by shooting time;
The shooting location that target person sequentially occurs is determined according to ranking results;
The shooting location sequentially occurred according to target person generates personage's motion profile of target person on map.
6. a kind of personage's trajectory retrieval system, which is characterized in that including primary server, the primary server includes:
Video acquisition unit, for according to target camera number and period acquisition of information video data;
Video processing unit, for when there is target person in the video data, intercept in the video data about
The facial image of target person;
Image identification unit calculates the face characteristic value of target person for carrying out recognition of face to the facial image;
Generation unit is indexed, for generating face retrieval index according to the face characteristic value;
Data query unit is obtained according to the face characteristic value and the face retrieval search index face database about mesh
Mark the multiple image data of personage;
Track generation unit, for obtaining and data, the shooting time of the multiple image data and shooting location according to the map
Generate personage's motion profile.
7. personage's trajectory retrieval system as claimed in claim 6, which is characterized in that the face database real-time reception is simultaneously deposited
Storage is distributed in the face of every frame facial image in video data captured by multiple cameras of different location and video data
Characteristic value and face retrieval index, each video data includes multiple image data, and every frame image data includes camera
Image, shooting time and the camera number of shooting, wherein each camera has unique number.
8. personage's trajectory retrieval method as claimed in claim 7, which is characterized in that the data query unit includes:
Area generation unit, for centered on the camera site of center, pre-determined distance to be that radius generates border circular areas;Its
In, the corresponding camera site of the target camera number is the center camera of query context when inquiry starts;
Subelement is inquired, for the circle according to the face characteristic value and the face retrieval search index face database
Video data captured by camera in shape region obtains the multiple image data about target person;
Judging unit, for judging whether the data query unit inquires the video data of a certain camera in border circular areas
In include image data about target person, if so, determining that the camera is new center camera and generates center and take the photograph
As head more new information is sent to the Area generation unit.
9. personage's trajectory retrieval system as claimed in claim 6, which is characterized in that further include each child servers;
The data query unit includes:
Data transmission unit, for the face characteristic value and face retrieval index to be sent to each child servers;
Data transmission unit receives the multiple image data about target person that each child servers return;
Each child servers are respectively used to according to the respective face database of the face characteristic value and the face retrieval search index
Obtain the multiple image data about target person;
Wherein, the face database real-time reception of each child servers simultaneously clapped by multiple cameras of the storage and distribution in different location
The face characteristic value of every frame facial image and face retrieval index, each video in the video data and video data taken the photograph
Data include multiple image data, and every frame image data includes image, shooting time and the camera number of camera shooting,
Wherein each camera has unique number.
10. personage's trajectory retrieval system as claim in any one of claims 6-9, which is characterized in that the track generation unit
Include:
Sorting subunit, for being ranked up by shooting time to the multiple image data;
Place subelement, for determining shooting location that target person sequentially occurs according to ranking results;
Track subelement, the shooting location for sequentially being occurred according to target person generate personage's fortune of target person on map
Dynamic rail mark.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910774887.2A CN110532923A (en) | 2019-08-21 | 2019-08-21 | A kind of personage's trajectory retrieval method and its system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910774887.2A CN110532923A (en) | 2019-08-21 | 2019-08-21 | A kind of personage's trajectory retrieval method and its system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110532923A true CN110532923A (en) | 2019-12-03 |
Family
ID=68662445
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910774887.2A Pending CN110532923A (en) | 2019-08-21 | 2019-08-21 | A kind of personage's trajectory retrieval method and its system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110532923A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111222492A (en) * | 2020-01-19 | 2020-06-02 | 四川联众竞达科技有限公司 | Tracking method of moving track |
CN111242077A (en) * | 2020-01-20 | 2020-06-05 | 西安奥卡云数据科技有限公司 | Figure tracking method, system and server |
CN112232153A (en) * | 2020-09-30 | 2021-01-15 | 广东职业技术学院 | Method and system for acquiring track of target person |
CN112633063A (en) * | 2020-11-18 | 2021-04-09 | 深圳供电局有限公司 | Person action tracking system and method thereof |
CN113010545A (en) * | 2019-12-20 | 2021-06-22 | 浙江宇视科技有限公司 | Data searching method, device, server, storage medium and system |
WO2021135138A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳市商汤科技有限公司 | Target motion trajectory construction method and device, and computer storage medium |
CN113840078A (en) * | 2021-06-10 | 2021-12-24 | 阿波罗智联(北京)科技有限公司 | Target detection system |
CN114579780A (en) * | 2022-03-04 | 2022-06-03 | 浪潮软件集团有限公司 | Method and system for inquiring human activity track under domestic CPU and operating system |
CN116069976A (en) * | 2023-03-06 | 2023-05-05 | 南京和电科技有限公司 | Regional video analysis method and system |
WO2024108539A1 (en) * | 2022-11-25 | 2024-05-30 | 京东方科技集团股份有限公司 | Target people tracking method and apparatus |
Citations (9)
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 |
CN106650652A (en) * | 2016-12-14 | 2017-05-10 | 黄先开 | Trajectory tracking system and method based on face recognition technology |
CN107016374A (en) * | 2017-04-12 | 2017-08-04 | 电子科技大学 | Intelligent Measurement tracking and the generation method of space-time track towards specific objective |
CN108200382A (en) * | 2017-12-15 | 2018-06-22 | 北京奇虎科技有限公司 | A kind of method and apparatus of video monitoring |
CN108197565A (en) * | 2017-12-29 | 2018-06-22 | 深圳英飞拓科技股份有限公司 | Target based on recognition of face seeks track method and system |
CN108985218A (en) * | 2018-07-10 | 2018-12-11 | 上海小蚁科技有限公司 | People flow rate statistical method and device, calculates equipment at storage medium |
CN109344285A (en) * | 2018-09-11 | 2019-02-15 | 武汉魅瞳科技有限公司 | A kind of video map construction and method for digging, equipment towards monitoring |
CN109871815A (en) * | 2019-02-26 | 2019-06-11 | 新华三技术有限公司 | A kind of method and device of query monitor information |
CN109934176A (en) * | 2019-03-15 | 2019-06-25 | 艾特城信息科技有限公司 | Pedestrian's identifying system, recognition methods and computer readable storage medium |
-
2019
- 2019-08-21 CN CN201910774887.2A patent/CN110532923A/en active Pending
Patent Citations (9)
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 |
CN106650652A (en) * | 2016-12-14 | 2017-05-10 | 黄先开 | Trajectory tracking system and method based on face recognition technology |
CN107016374A (en) * | 2017-04-12 | 2017-08-04 | 电子科技大学 | Intelligent Measurement tracking and the generation method of space-time track towards specific objective |
CN108200382A (en) * | 2017-12-15 | 2018-06-22 | 北京奇虎科技有限公司 | A kind of method and apparatus of video monitoring |
CN108197565A (en) * | 2017-12-29 | 2018-06-22 | 深圳英飞拓科技股份有限公司 | Target based on recognition of face seeks track method and system |
CN108985218A (en) * | 2018-07-10 | 2018-12-11 | 上海小蚁科技有限公司 | People flow rate statistical method and device, calculates equipment at storage medium |
CN109344285A (en) * | 2018-09-11 | 2019-02-15 | 武汉魅瞳科技有限公司 | A kind of video map construction and method for digging, equipment towards monitoring |
CN109871815A (en) * | 2019-02-26 | 2019-06-11 | 新华三技术有限公司 | A kind of method and device of query monitor information |
CN109934176A (en) * | 2019-03-15 | 2019-06-25 | 艾特城信息科技有限公司 | Pedestrian's identifying system, recognition methods and computer readable storage medium |
Non-Patent Citations (1)
Title |
---|
都伊林 著: "《智能安防新发展与应用》", 31 July 2018 * |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113010545B (en) * | 2019-12-20 | 2023-02-28 | 浙江宇视科技有限公司 | Data searching method, device, server, storage medium and system |
CN113010545A (en) * | 2019-12-20 | 2021-06-22 | 浙江宇视科技有限公司 | Data searching method, device, server, storage medium and system |
WO2021135138A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳市商汤科技有限公司 | Target motion trajectory construction method and device, and computer storage medium |
CN111222492A (en) * | 2020-01-19 | 2020-06-02 | 四川联众竞达科技有限公司 | Tracking method of moving track |
CN111242077A (en) * | 2020-01-20 | 2020-06-05 | 西安奥卡云数据科技有限公司 | Figure tracking method, system and server |
CN112232153A (en) * | 2020-09-30 | 2021-01-15 | 广东职业技术学院 | Method and system for acquiring track of target person |
CN112633063A (en) * | 2020-11-18 | 2021-04-09 | 深圳供电局有限公司 | Person action tracking system and method thereof |
CN112633063B (en) * | 2020-11-18 | 2023-06-30 | 深圳供电局有限公司 | Figure action tracking system and method thereof |
CN113840078A (en) * | 2021-06-10 | 2021-12-24 | 阿波罗智联(北京)科技有限公司 | Target detection system |
CN114579780A (en) * | 2022-03-04 | 2022-06-03 | 浪潮软件集团有限公司 | Method and system for inquiring human activity track under domestic CPU and operating system |
WO2024108539A1 (en) * | 2022-11-25 | 2024-05-30 | 京东方科技集团股份有限公司 | Target people tracking method and apparatus |
CN116069976A (en) * | 2023-03-06 | 2023-05-05 | 南京和电科技有限公司 | Regional video analysis method and system |
CN116069976B (en) * | 2023-03-06 | 2023-09-12 | 南京和电科技有限公司 | Regional video analysis method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110532923A (en) | A kind of personage's trajectory retrieval method and its system | |
CN110532432A (en) | A kind of personage's trajectory retrieval method and its system, computer readable storage medium | |
Gummidi et al. | A survey of spatial crowdsourcing | |
Zhao et al. | Spatial crowdsourcing: current state and future directions | |
EP3241366B1 (en) | Method and system for estimating a number of persons in a crowd | |
EP3241368B1 (en) | Method and system for a real-time counting of a number of participants at a public happening | |
US8760519B2 (en) | Threat-detection in a distributed multi-camera surveillance system | |
CN111756848B (en) | QoS optimization method based on federal learning and mobile perception under mobile edge environment | |
CN107370983A (en) | Acquisition methods and device for the whereabouts track of video monitoring system | |
US10467311B2 (en) | Communication system and method of generating geographic social networks in virtual space | |
US20050255842A1 (en) | Communication system and method for comprehensive collection, aggregation and dissemination of geospatial information | |
CN107318087A (en) | A kind of method and system being automatically positioned to arrival specified location | |
Cho et al. | Collaborative occupancy reasoning in visual sensor network for scalable smart video surveillance | |
CN106101224A (en) | Identify the method and device of the classification of user geographic location | |
CN105629196B (en) | Alignment system and correlation method based on computer vision and dynamic fingerprint | |
EP3241367B1 (en) | Method and system for real-time computing of origin-destination matrices relating to attendees at a public happening through analysis of mobile communication network data | |
CN109074772B (en) | Dynamic ad hoc network for content sharing based on line of sight | |
CN109902681B (en) | User group relation determining method, device, equipment and storage medium | |
WO2016155769A1 (en) | Method and system for a real-time counting of a number of persons in a crowd by means of aggregated data of a telecommunication network | |
CN102638657A (en) | Information processing apparatus and imaging region sharing determination method | |
Rigby et al. | An opportunistic client user interface to support centralized ride share planning | |
Shaikh et al. | An AHP/TOPSIS-based approach for an optimal site selection of a commercial opening utilizing geospatial data | |
Wang et al. | User recruitment system for efficient photo collection in mobile crowdsensing | |
CN111461460B (en) | Police service inspection route planning method and personnel inspection method | |
EP3241369B1 (en) | Method and system for a posteriori computation of origin-destination matrices relating to gathering of people through analysis of mobile communication network data |
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 |
Application publication date: 20191203 |
|
RJ01 | Rejection of invention patent application after publication |