CN109271859A - It combines related cases method and apparatus, electronic equipment, computer storage medium - Google Patents
It combines related cases method and apparatus, electronic equipment, computer storage medium Download PDFInfo
- Publication number
- CN109271859A CN109271859A CN201810922364.3A CN201810922364A CN109271859A CN 109271859 A CN109271859 A CN 109271859A CN 201810922364 A CN201810922364 A CN 201810922364A CN 109271859 A CN109271859 A CN 109271859A
- Authority
- CN
- China
- Prior art keywords
- suspect
- image
- case
- coincidence
- collections
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000000605 extraction Methods 0.000 claims description 13
- 235000013399 edible fruits Nutrition 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 15
- 230000001815 facial effect Effects 0.000 description 49
- 238000011835 investigation Methods 0.000 description 12
- 230000006854 communication Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 230000004044 response Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000011840 criminal investigation Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000036039 immunity Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000116 mitigating effect Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- 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
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Economics (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Resources & Organizations (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Primary Health Care (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present application discloses one kind and combines related cases method and apparatus, electronic equipment, computer storage medium, wherein method includes: the associated video based on multiple cases, obtains the corresponding suspect's image collection of each case；At least one set of coincidence suspect's image is obtained from multiple suspect's image collections, the corresponding at least two suspect's image collections of coincidence suspect's image described in every group；Corresponding at least two case of coincidence suspect's image described in every group is subjected to processing of combining related cases.Based on the above embodiments of the present application, the image in multiple cases is identified by image recognition technology, realizes conspiring with people for unartificial identification, human resources is saved and improves the accuracy rate combined related cases.
Description
Technical field
This application involves case investigation field, especially one kind combine related cases method and apparatus, electronic equipment, computer storage
Medium.
Background technique
Part of combining related cases analysis refers to the case two or more, from live trace evidence, modus operandi feature, the time,
Point, selection target, attack position etc. are analysed and compared, and are judged whether it is carried out by same people or same partner offender
Process.It is combine related cases part investigation the first step.Combine related cases the suspect in part criminal offence often have it is many altogether
Same, public security organ can use these common ground to combine related cases.The main method of part of combining related cases is had and is conspired with case, with people
It conspires, conspired with object, conspired with case property.By analysis of combining related cases, clue to solve the case and information can be provided, obtained from many aspects
Evidence of crime is taken, grasps whole crimes of offender in time, predicts simultaneously crime prevention.
Summary of the invention
The embodiment of the present application provides one kind and combines related cases technology.
According to the one aspect of the embodiment of the present application, the one kind provided is combined related cases method, comprising:
Based on the associated video of multiple cases, the corresponding suspect's image collection of each case is obtained；
At least one set of coincidence suspect's image is obtained from multiple suspect's image collections, is overlapped suspicion described in every group
People's image corresponds at least two suspect's image collections；
Corresponding at least two case of coincidence suspect's image described in every group is subjected to processing of combining related cases.
Optionally, the associated video based on multiple cases obtains the corresponding suspect's image set of each case
It closes, comprising:
Recognition of face is carried out at least two field pictures in the associated video of each case respectively, obtains at least one
Suspect's image including face；
Suspect's image of the acquisition is based on case classification, obtains the corresponding suspect's image set of each case
It closes.
It is optionally, described that at least one set of coincidence suspect's image is obtained from multiple suspect's image collections, comprising:
The multiple suspect's image collection is clustered, at least one cluster result is obtained, it is each described poly-
It include at least two suspect's images in class result, at least two suspect's images in the cluster result correspond to different institutes
State suspect's image collection；
Using each cluster result as suspect's image is overlapped described in one group, at least one set of coincidence suspicion is obtained
People's image.
Optionally, described that the multiple suspect's image collection is clustered, at least one cluster result is obtained,
Include:
Face characteristic extraction is carried out to suspect's images all in the multiple suspect's image collection, obtains multiple faces
Feature；
All suspect's images in the multiple suspect's image collection are clustered based on the face characteristic, are obtained
At least one cluster result.
It is optionally, described that at least one set of coincidence suspect's image is obtained from multiple suspect's image collections, comprising:
Respectively by each suspect's image and other described suspect's images in each suspect's image collection
Each suspect's image in set matches, and obtains at least one matching result, includes in each matching result
At least two suspect's images, at least two suspect's images in the matching result correspond to different suspect's images
Set；
Using each matching result as suspect's image is overlapped described in one group, at least one set of coincidence suspicion is obtained
People's image.
Optionally, it is described respectively by each suspect's image collection each suspect's image and other described in
Each suspect's image in suspect's image collection matches, and obtains at least one matching result, comprising:
It is obtained in each suspect's image and the second suspect's image collection in first suspect's image collection respectively
Each suspect's image between similarity, the first suspect image collection be multiple suspect's image collections
In any one, the second suspect image collection be multiple suspect's image collections in addition to first suspicion
All suspect's image collections outside people's image collection；
The similarity is greater than or equal at least two suspect's images of preset value as matching result.
Optionally, described that corresponding at least two case of coincidence suspect's image described in every group is subjected to the place that combines related cases
Reason, comprising:
Corresponding at least two institute of coincidence suspect's image is obtained respectively based on coincidence suspect's image described in every group
State suspect's image collection；
Corresponding at least two case is obtained based at least two suspect's image collections；
At least two of the acquisition cases are subjected to processing of combining related cases.
Optionally, further includes:
It combines related cases described in storage and/or display the corresponding suspect's image collection of at least two cases of processing
And its corresponding relevant information.
Optionally, the corresponding relevant information of the case comprises at least one of the following information:
The docket information of corresponding suspect's image, with detecing number information, case type information, case description letter
Breath.
Optionally, further includes:
By each suspicion in the received corresponding facial image to be checked of people to be checked and multiple suspect's image collections
People's image is compared, and determines whether the people to be checked is suspect based on comparison result.
Optionally, described will be in the received corresponding facial image to be checked of people to be checked and multiple suspect's image collections
Each suspect's image is compared, and determines whether the people to be checked is suspect based on comparison result, comprising:
It obtains in the facial image to be checked and multiple suspect's image collections between each suspect's image
Similarity；
It is to obtain at least one target suspect's image in response to the comparison result, determines the artificial suspicion to be checked
People；Similarity between the target suspect image and the facial image to be checked is greater than or equal to default similarity；
It is not obtain the target suspect image in response to the comparison result, determines that the people to be checked is not suspicion
People.
Optionally, described in response to the comparison result is to obtain at least one target suspect's image, determine it is described to
It looks into after artificial suspect, further includes:
New suspect's image collection is established for the facial image to be checked, and will be described in the facial image deposit to be checked
New suspect's image collection.
Optionally, described to obtain each suspect in the facial image to be checked and multiple suspect's image collections
Similarity between image, comprising:
Face characteristic extraction is carried out to the facial image to be checked and each suspect's image respectively, is obtained described to be checked
The corresponding face characteristic to be checked of facial image and the corresponding suspicion face characteristic of suspect's image；
Based on the distance between the face characteristic to be checked and each suspicion face characteristic, the face figure to be checked is obtained
Picture and similarity between each suspect's image.
According to the other side of the embodiment of the present application, the one kind provided is combined related cases device, comprising:
Gather obtaining unit, for the associated video based on multiple cases, obtains the corresponding suspect of each case
Image collection；
It is overlapped suspect's unit, for obtaining at least one set of coincidence suspect figure from multiple suspect's image collections
Picture, the corresponding at least two suspect's image collections of coincidence suspect's image described in every group；
It combines related cases processing unit, for carrying out corresponding at least two case of coincidence suspect's image described in every group
It combines related cases processing.
Optionally, the set obtaining unit, specifically in the associated video respectively to each case at least
Two field pictures carry out recognition of face, obtain suspect's image that at least one includes face；By suspect's image of the acquisition
Based on case classification, the corresponding suspect's image collection of each case is obtained.
Optionally, coincidence suspect's unit, comprising:
Cluster module obtains at least one cluster knot for clustering to the multiple suspect's image collection
Fruit includes at least two suspect's images in each cluster result, at least two suspects figure in the cluster result
As corresponding different suspect's image collection；
First suspect's determining module, for will each cluster result as described in one group coincidence suspect's image,
Obtain at least one set of coincidence suspect's image.
Optionally, the cluster module is specifically used for all suspect's images in the multiple suspect's image collection
Face characteristic extraction is carried out, multiple face characteristics are obtained；Based on the face characteristic in the multiple suspect's image collection
All suspect's images cluster, and obtain at least one cluster result.
Optionally, coincidence suspect's unit, comprising:
Matching module, for respectively by each suspect's image collection each suspect's image and other institutes
Each suspect's image stated in suspect's image collection matches, and obtains at least one matching result, and each described
With including at least two suspect's images in result, at least two suspect's images in the matching result correspond to different institutes
State suspect's image collection；
Second suspect's determining module, for will each matching result as described in one group coincidence suspect's image,
Obtain at least one set of coincidence suspect's image.
Optionally, the matching module, specifically for obtaining each suspicion in first suspect's image collection respectively
The similarity between each suspect's image in people's image and second suspect's image collection, the first suspect image
Collect any one being combined into multiple suspect's image collections, the second suspect image collection is multiple suspicion
All suspect's image collections in people's image collection other than the first suspect image collection；The similarity is greater than
Or equal to preset value at least two suspect's images as matching result.
Optionally, the processing unit of combining related cases, specifically for being obtained respectively based on coincidence suspect's image described in every group
The corresponding at least two suspect's image collection of coincidence suspect's image；Based at least two suspect's images
Set obtains corresponding at least two case；At least two of the acquisition cases are subjected to processing of combining related cases.
Optionally, described device further include:
Storage and display unit, at least two cases for storing and/or showing the processing of combining related cases are corresponding
Suspect's image collection and its corresponding relevant information.
Optionally, the corresponding relevant information of the case comprises at least one of the following information:
The docket information of corresponding suspect's image, with detecing number information, case type information, case description letter
Breath.
Optionally, described device further include:
Determination unit is compared, is used for the received corresponding facial image to be checked of people to be checked and multiple suspect's images
Each suspect's image is compared in set, determines whether the people to be checked is suspect based on comparison result.
Optionally, the comparison determination unit is specifically used for obtaining the facial image to be checked and multiple suspects
Similarity in image collection between each suspect's image；It is to obtain at least one target to dislike in response to the comparison result
People's image is doubted, determines the artificial suspect to be checked；Phase between the target suspect image and the facial image to be checked
It is greater than or equal to default similarity like degree；It is not obtain the target suspect image in response to the comparison result, determines institute
Stating people to be checked is not suspect.
Optionally, the comparison determination unit is also used to establish new suspect's image set for the facial image to be checked
It closes, and the facial image to be checked is stored in new suspect's image collection.
Optionally, the comparison determination unit is obtaining the facial image to be checked and multiple suspect's image collections
In similarity between each suspect's image when, for respectively to the facial image to be checked and each suspect's image
Face characteristic extraction is carried out, the corresponding face characteristic to be checked of facial image to be checked is obtained and suspect's image is corresponding
Suspicion face characteristic；Based on the distance between the face characteristic to be checked and each suspicion face characteristic, obtain described to be checked
Similarity between facial image and each suspect's image.
Other side according to an embodiment of the present invention, a kind of electronic equipment provided, including processor, the processor
Including device of combining related cases as described above.
Other side according to an embodiment of the present invention, a kind of electronic equipment provided, comprising: memory, for storing
Executable instruction；
And processor, string as described above is completed for communicating with the memory to execute the executable instruction
And the operation of case method.
Other side according to an embodiment of the present invention, a kind of computer storage medium provided, for storing computer
The instruction that can be read, described instruction are performed the operation for executing method of combining related cases as described above.
Other side according to an embodiment of the present invention, a kind of computer program provided, including computer-readable code,
When the computer-readable code is run in equipment, the processor in the equipment is executed for realizing string as described above simultaneously
The instruction of case method.
Based on the above embodiments of the present application provide one kind combine related cases method and apparatus, electronic equipment, computer storage be situated between
Matter obtains the corresponding suspect's image collection of each case based on the associated video of multiple cases；From multiple suspect's image sets
At least one set of coincidence suspect's image, the corresponding at least two suspect's image collections of every group of coincidence suspect's image are obtained in conjunction；
Corresponding at least two case of every group of coincidence suspect's image is subjected to processing of combining related cases；By image recognition technology by multiple cases
Image in part is identified, realizes conspiring with people for unartificial identification, is saved human resources and is improved the standard combined related cases
True rate.
Below by drawings and examples, the technical solution of the application is described in further detail.
Detailed description of the invention
The attached drawing for constituting part of specification describes embodiments herein, and together with description for explaining
The principle of the application.
The application can be more clearly understood according to following detailed description referring to attached drawing, in which:
Fig. 1 is that the application combines related cases the flow chart of method one embodiment.
Fig. 2 is that the application combines related cases the structural schematic diagram of device one embodiment.
Fig. 3 is the structural representation suitable for the electronic equipment of the terminal device or server that are used to realize the embodiment of the present application
Figure.
Specific embodiment
The various exemplary embodiments of the application are described in detail now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
The range of application.
Simultaneously, it should be appreciated that for ease of description, the size of various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to the application
And its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as part of specification.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
With the development of depth learning technology, face identification functions are increasingly mature, with the acquisition of its data is convenient, operation is fast,
The advantages such as accuracy rate height go deep into rapidly the various aspects of people's life.Using based on deep learning Face datection and feature extraction calculate
Method, can be in extensive face picture database, and fast search compares out human face data similar with face to be checked, and then assists
Public security carries out cracking of cases.Face recognition technology is applied to analysis of combining related cases, a line policeman can be assisted to investigate series crime,
The workload for mitigating analysis personnel, reduces the input cost of case.
Fig. 1 is that the application combines related cases the flow chart of method one embodiment.As shown in Figure 1, the embodiment method includes:
Step 110, the associated video based on multiple cases obtains the corresponding suspect's image collection of each case.
Optionally, the correlation of case can be obtained by the way that the picture pick-up device acquisition near spot or spot is arranged in
Video, or obtain from the case video being saved the associated video of case；By carrying out people to the image in associated video
Face identification or other processing can get corresponding suspect's image collection, may include at least one in each suspect's set
Suspect's image, each suspect can correspond to suspect's image or multiple suspect's images.
Step 120, at least one set of coincidence suspect's image is obtained from multiple suspect's image collections.
Wherein, the corresponding at least two suspect's image collections of every group of coincidence suspect's image.
Optionally, at least one set of coincidence suspect figure is obtained from multiple suspect's image collections by face recognition technology
Picture is overlapped suspect's image and refers to that the same or similar face appears at least two suspect's image collections, it is believed that should
Suspect has simultaneously participated in the corresponding case of at least two suspects image collection.
Step 130, corresponding at least two case of every group of coincidence suspect's image is subjected to processing of combining related cases.
The one kind provided based on the above embodiments of the present application is combined related cases method, based on the associated video of multiple cases, is obtained
The corresponding suspect's image collection of each case；At least one set of coincidence suspect figure is obtained from multiple suspect's image collections
Picture, the corresponding at least two suspect's image collections of every group of coincidence suspect's image；Every group of coincidence suspect's image is corresponding extremely
Few two cases carry out processing of combining related cases；The image in multiple cases is identified by image recognition technology, is realized inhuman
Being conspired with people for work identification, is saved human resources and improves the accuracy rate combined related cases.
In one or more optional embodiments, step 110 may include:
Recognition of face is carried out at least two field pictures in the associated video of each case respectively, obtaining at least one includes
Suspect's image of face.
Optionally, recognition of face can be realized based on convolutional neural networks, and the image including face is identified from video
Come, the screening of face quality can also be carried out to the image including face of acquisition, to guarantee the people in the suspect's image obtained
Face quality is preferable, one suspect of expression that can be clear and accurate, and face quality can include but is not limited to: face accounts for image ratio
Example, face clarity, deflection angle etc..
Suspect's image of acquisition is based on case classification, obtains the corresponding suspect's image collection of each case.
It in the prior art can only be by the experience and memory of personnel in charge of the case, or by manually investigating for the identification combined related cases
Mode part database on record in find the case with same characteristic features, carry out identification of combining related cases on this basis.The application is logical
Cross and combined related cases using face recognition technology, provide a kind of strong evidence source for identification of combining related cases, calculating process without
Need to artificially it interfere, personnel in charge of the case greatly reduces the consumption to human resources without screening one by one.And face recognition technology is accurate
Rate is high, identifies strong interference immunity, can exclude influence of the variation such as hair style, fat or thin, age, expression to result, can effectively improve and detect
Look into efficiency.
In one or more optional embodiments, step 120 may include:
Multiple suspect's image collections are clustered, at least one cluster result is obtained, includes in each cluster result
At least two suspect's images, at least two suspect's images in cluster result correspond to different suspect's image collections；
Using each cluster result as one group of coincidence suspect's image, at least one set of coincidence suspect's image is obtained.
All suspect's images in multiple suspect's image collections are aggregated to by the present embodiment by clustering according to face
In multiple clusters, if only including suspect's image in these clusters, it is believed that it is not cluster result, and will include extremely
The cluster of few two suspect's images is as cluster result, at this point, the multiple suspect's images for including in cluster result prove
There are the same or similar faces in corresponding suspect's image collection.In case A, A1, A2, A3, the face of A4 are detected；
It is detected in case B, B1, B2, B3, B4, the face of B5；It is detected in case C, C1, C2, C3 face.After clustering method,
Identify that A1, B2, C3 are same people, B1, C1 are same people, then judge that the two people combine related cases suspicion.
Optionally, multiple suspect's image collections are clustered, obtain at least one cluster result, comprising:
Face characteristic extraction is carried out to suspect's images all in multiple suspect's image collections, it is special to obtain multiple faces
Sign；
All suspect's images in multiple suspect's image collections are clustered based on face characteristic, obtain at least one
Cluster result.
In the present embodiment, suspect's image is clustered by face characteristic, the process for obtaining face characteristic can pass through
Convolutional neural networks carry out feature extraction, can also obtain by other means, and the present embodiment, which is not shown, obtains face characteristic
Concrete mode is clustered multiple face characteristics by the distance between multiple face characteristics, specifically, can by feature it
Between distance be less than preset value multiple face characteristics aggregate into one kind.
In one or more optional embodiments, step 120 may include:
Respectively by each suspicion in each suspect's image and other suspect's image collections in each suspect's image collection
It doubts people's image to match, obtains at least one matching result；
It wherein, include at least two suspect's images in each matching result, at least two suspects in matching result
Image corresponds to different suspect's image collections；
Using each matching result as one group of coincidence suspect's image, at least one set of coincidence suspect's image is obtained.
It obtaining and is overlapped except suspect's image except through cluster, the present embodiment is proposed to be compared using batch, such as: with
It on the basis of each face in case A, is compared respectively with the face occurred in case B, case C, exports similarity at certain
Result more than a threshold value.It again on the basis of each face in case B, is compared, exports with the face occurred in case C
Result of the similarity more than some threshold value.For example face similar with A1 has B2, C3, then suspects that A1 combines related cases suspicion.
It optionally, respectively will be in each suspect's image and other suspect's image collections in each suspect's image collection
Each suspect's image match, obtain at least one matching result, comprising:
Obtain respectively each suspect's image in first suspect's image collection with it is each in second suspect's image collection
Similarity between suspect's image；
Similarity is greater than or equal at least two suspect's images of preset value as matching result.
Wherein, first suspect's image collection is any one in multiple suspect's image collections, the second suspect figure
Image set is combined into multiple suspect's image collections all suspect's image collections other than first suspect's image collection；When with
As soon as suspect's image collection is as first suspect's image collection, other suspect's image collections are used as the second suspicion
People's image collection, can be to avoid omission.
In one or more optional embodiments, step 130 may include:
It is obtained respectively based on every group of coincidence suspect image and is overlapped the corresponding at least two suspects image of suspect's image
Set；
Corresponding at least two case is obtained based at least two suspect's image collections；
At least two cases of acquisition are subjected to processing of combining related cases.
The number that occurs in different cases of the statistics same person, if it is greater than or equal to then judging that the people may occur to go here and there simultaneously twice
Case behavior.
The processing that combine related cases of at least two cases there is into following benefit: 1, being conducive to reinforce trans-regional criminal investigation cooperation:
Break the situations such as barrier between different departments in current investigation, people's case separation of jurisdiction of case.Only carry out and case is investigated, breaks region
Boundary carries out combined operation, can just construct the pattern of big criminal investigation.More case unifications, are conducive to concentrate police strength, save manpower, wealth
Power and material resources receive the effect got twice the result with half the effort.2, it shares Crime Information resource: being conducive to make full use of on different scenes of a crime and receive
The various Crime Informations collected.It can make the trace evidence of a lot of cases, the trace evidence of strange land discovery, small case and fragile case
The Crime Informations resource such as trace evidence is comprehensively utilized.By series connection and case, analyzes, compares, confirming various Crime Informations, it is quasi-
Criminal's personal feature really is portrayed, distinguishes investigation direction and range.3, be conducive to obtain various evidence of crime: right in criminal suit
The requirement of evidence is higher and higher, more strictly.Punishment tells method clearly stipulate that when conclusion of investigation, it is necessary to reach " crime fact understands,
Evidence is certain, abundant " requirement.Each evidence being collected into have passed through examination and examine, and be authentic, reversible
's；The evidence of asserting crime has comparable quantity, can make up a complete system of proof, ring ring linking, without omitted.It can
See, requirement of the law to evidence should guarantee matter guaranteed discharge again.Part of combining related cases is conducive to obtain evidence of crime from many aspects,
And whole crimes of offender are grasped in time.4, be conducive to deep-cut remaining crime: a string of band together can be broken by combining related cases, and obtain simultaneously case one
It goes here and there, effect that is a piece of, deep-cutting a gang of of solving a case.5, be conducive to make an initiative sally, crime prevention: by part of combining related cases, can find crime
The crime rule and feature of molecule, understand crime new trend, carry out targeted prevention work in time, adjust investigation scheme,
Strive that crime object is arrested at scene.
In one or more optional embodiments, further includes:
Storage and/or display combine related cases processing the corresponding suspect's image collection of at least two cases and its corresponding phase
Close information.
It, can be on an electronic device to suspect's image after combining related cases after the processing that carries out combining related cases to multiple cases
Set and its corresponding relevant information are shown, to obtain more case information, provide condition for clear up a criminal case；Storage is
In order to it post-process other combine related cases identify when, can be by the case and other as a whole of the case after combining related cases
Case carries out new processing of combining related cases.
Optionally, the corresponding relevant information of case comprises at least one of the following information:
The docket information of corresponding suspect's image, with detecing number information, case type information, case description information.
In one or more optional embodiments, can also include:
By each suspect's image in the corresponding facial image to be checked of received people to be checked and multiple suspect's image collections into
Row compares, and determines whether people to be checked is suspect based on comparison result.
When case collection in worksite is to new facial image, can using facial image as facial image to be checked with it is known
Multiple suspect's image collections in each suspect's image be compared, determine whether the corresponding people of the facial image to be checked has
Criminal record (there are matched suspect's images in suspect's image collection), if any criminal record can using the people to be checked as suspect into
One step is examined, is compared by recognition of face, and the efficiency that suspect determines is improved, and can quickly determine suspicion in spot
People.
Optionally, by each suspect in the received corresponding facial image to be checked of people to be checked and multiple suspect's image collections
Image is compared, and determines whether people to be checked is suspect based on comparison result, comprising:
Obtain the similarity in facial image to be checked and multiple suspect's image collections between each suspect's image；
It is to obtain at least one target suspect's image in response to comparison result, determines artificial suspect to be checked；Target is disliked
It doubts the similarity between people's image and facial image to be checked and is greater than or equal to default similarity；
It is not obtain target suspect's image in response to comparison result, determines that people to be checked is not suspect.
In the present embodiment, face retrieval comparison is carried out to people to be checked: by the default similarity of setting, uploading people's to be checked
The information (optional) such as face picture or input docket, the picture in the face picture of people to be checked and case portrait library is carried out
Retrieval compares；The face characteristic value that people to be checked is extracted by deep learning, by the owner in this feature value and case portrait library
The characteristic value of face is compared one by one, calculates the similarity between two characteristic values；It optionally, is more than the inspection of threshold value by similarity
Hitch fruit presses the Sequential output of similarity from high to low, and the search result of output includes the picture of every face, similarity, case
Relevant information.
Optionally, it is to obtain at least one target suspect's image in response to comparison result, determines artificial suspect to be checked
Later, further includes:
New suspect's image collection is established for facial image to be checked, and facial image to be checked is stored in new suspect and is schemed
Image set closes.
It after determining artificial suspect to be checked, can put on file for investigation and prosecution, establish suspect's image set after putting on record for the case
It closes, and facial image to be checked is stored in suspect's image collection, in case basis, example are done in the processing of subsequent case and processing of combining related cases
Such as, using suspect's image collection as suspect's image collection in step 110, processing of combining related cases is carried out.
Optionally, it obtains similar between facial image to be checked and each suspect's image in multiple suspect's image collections
Degree, comprising:
Face characteristic extraction is carried out to facial image to be checked and each suspect's image respectively, it is corresponding to obtain facial image to be checked
Face characteristic to be checked and the corresponding suspicion face characteristic of suspect's image；
Based on the distance between face characteristic to be checked and each suspicion face characteristic, facial image to be checked and each suspect are obtained
Similarity between image.
In the present embodiment, determined between the corresponding image of two face characteristics by the distance between two face characteristics
Similarity optionally carries out processing to the image for including face based on convolutional neural networks and obtains face characteristic, is carrying out face
Before feature extraction, it can also include that face is divided, facial image is divided from facial image to be checked and/or suspect's image
Out, so that human face ratio reaches preset ratio in the facial image obtained, it is more suitable for doing face feature extraction.
Case string and after, provide foundation for solving criminal cases.It sufficiently to find, study the case where origin and crime line
Road, it is determined whether flee or native, acquaintance carried out by, clearly investigation direction, select breach, carry out investigation.
1. according to the universal law of crime, the characteristics of paying attention to origin case, especially to find the first case be
When, where to occur.Majority crime objects just the origin of case or around, carry out the investigation of subrange to find line
Rope is the effective way that detection is conspired.
2. according to the precedence of incidence of criminal offenses, walking route when analysis case crime, home to return to after distinguishing criminal's crime and
It stops over place, therefrom finds clue to solve the case.
3. the characteristics of according to serial case, found from previous criminal's dossier have the characteristics that similar crime means, pair
As therefrom finding criminal.
4. being searched in fingerprint file library using the fingerprint on site extracted, directly assert and solve a case.
5. taking various investigative measures in time according to the development of merit, quickly track down.Part of combining related cases generally all has certain
Continuity find out the rule of development by going here and there and analyze, judge commit a crime rule trend, scope of activities, crime number, delimitation is touched
Corresponding investigative measures are taken in row's condition and key area in time, are a big advantages of simultaneously case investigation.Can take work at a selected spot wait for,
Patrol ambuscade sets chuck and the methods of looks into, and arrests offender.
6. breaking major case by combining related cases with small case, being the important channel for currently tracking down great violent crime.Criminal is bold
It is mad to make major case, generally all have previous conviction or made case.Simultaneously by the string of major case and small case, some detection lines are obtained from small case
Rope can play a multiplier effect.
7. pair trans-regional, part of combining related cases on a large scale investigation, will accomplish information resources share, carry out combined operation.It is existing
Increasing in the span fled about to commit crimes, the case understood in time elsewhere sends out broken situation, and it is suitable to flee crime for strike
It is important.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can be stored in a computer readable storage medium, the program
When being executed, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: ROM, RAM, magnetic disk or light
The various media that can store program code such as disk.
Fig. 2 is the structural schematic diagram of device one embodiment of the invention of combining related cases.The device of the embodiment can be used for realizing
The above-mentioned each method embodiment of the present invention.As shown in Fig. 2, the device of the embodiment includes:
Gather obtaining unit 21, for the associated video based on multiple cases, obtains the corresponding suspect's figure of each case
Image set closes.
Optionally, the correlation of case can be obtained by the way that the picture pick-up device acquisition near spot or spot is arranged in
Video, or obtain from the case video being saved the associated video of case；By carrying out people to the image in associated video
Face identification or other processing can get corresponding suspect's image collection, may include at least one in each suspect's set
Suspect's image, each suspect can correspond to suspect's image or multiple suspect's images.
It is overlapped suspect's unit 22, for obtaining at least one set of coincidence suspect figure from multiple suspect's image collections
Picture.
Wherein, the corresponding at least two suspect's image collections of every group of coincidence suspect's image.
Optionally, at least one set of coincidence suspect figure is obtained from multiple suspect's image collections by face recognition technology
Picture is overlapped suspect's image and refers to that the same or similar face appears at least two suspect's image collections, it is believed that should
Suspect has simultaneously participated in the corresponding case of at least two suspects image collection.
It combines related cases processing unit 23, for corresponding at least two case of every group of coincidence suspect's image to be combined related cases
Processing.
The one kind provided based on the above embodiments of the present application is combined related cases device, will be in multiple cases by image recognition technology
Image identified, realize conspiring with people for unartificial identification, save human resources and improve the accuracy rate combined related cases.
In one or more optional embodiments, gather obtaining unit 21, specifically for the phase respectively to each case
At least two field pictures closed in video carry out recognition of face, obtain suspect's image that at least one includes face；By acquisition
Suspect's image is based on case classification, obtains the corresponding suspect's image collection of each case.
Optionally, recognition of face can be realized based on convolutional neural networks, and the image including face is identified from video
Come, the screening of face quality can also be carried out to the image including face of acquisition, to guarantee the people in the suspect's image obtained
Face quality is preferable, one suspect of expression that can be clear and accurate, and face quality can include but is not limited to: face accounts for image ratio
Example, face clarity, deflection angle etc..
It in the prior art can only be by the experience and memory of personnel in charge of the case, or by manually investigating for the identification combined related cases
Mode part database on record in find the case with same characteristic features, carry out identification of combining related cases on this basis.The application is logical
Cross and combined related cases using face recognition technology, provide a kind of strong evidence source for identification of combining related cases, calculating process without
Need to artificially it interfere, personnel in charge of the case greatly reduces the consumption to human resources without screening one by one.And face recognition technology is accurate
Rate is high, identifies strong interference immunity, can exclude influence of the variation such as hair style, fat or thin, age, expression to result, can effectively improve and detect
Look into efficiency.
In one or more optional embodiments, being overlapped suspect's unit 22 may include:
Cluster module obtains at least one cluster result, Mei Geju for clustering to multiple suspect's image collections
It include at least two suspect's images in class result, at least two suspect's images in cluster result correspond to different suspects
Image collection；
First suspect's determining module, for obtaining at least using each cluster result as one group of coincidence suspect's image
One group of coincidence suspect's image.
All suspect's images in multiple suspect's image collections are aggregated to by the present embodiment by clustering according to face
In multiple clusters, if only including suspect's image in these clusters, it is believed that it is not cluster result, and will include extremely
The cluster of few two suspect's images is as cluster result, at this point, the multiple suspect's images for including in cluster result prove
There are the same or similar faces in corresponding suspect's image collection.
Optionally, cluster module is specifically used for carrying out face to suspect's images all in multiple suspect's image collections
Feature extraction obtains multiple face characteristics；Based on face characteristic to all suspect's images in multiple suspect's image collections into
Row cluster, obtains at least one cluster result.
In one or more optional embodiments, being overlapped suspect's unit 22 may include: matching module, for distinguishing
Each suspect's image in each suspect's image and other suspect's image collections in each suspect's image collection is carried out
Matching, obtains at least one matching result, includes at least two suspect's images in each matching result, in matching result extremely
Few two suspect's images correspond to different suspect's image collections；
Second suspect's determining module, for obtaining at least using each matching result as one group of coincidence suspect's image
One group of coincidence suspect's image.
It obtaining and is overlapped except suspect's image except through cluster, the present embodiment is proposed to be compared using batch, such as: with
It on the basis of each face in case A, is compared respectively with the face occurred in case B, case C, exports similarity at certain
Result more than a threshold value.It again on the basis of each face in case B, is compared, exports with the face occurred in case C
Result of the similarity more than some threshold value.For example face similar with A1 has B2, C3, then suspects that A1 combines related cases suspicion.
Optionally, matching module, specifically for obtain respectively each suspect's image in first suspect's image collection with
The similarity between each suspect's image in second suspect's image collection, first suspect's image collection are multiple suspects
Any one in image collection, second suspect's image collection are in multiple suspect's image collections in addition to the first suspect schemes
All suspect's image collections outside image set conjunction；Using similarity be greater than or equal to preset value at least two suspect's images as
Matching result.
In one or more optional embodiments, processing unit 23 of combining related cases is specifically used for being based on every group of coincidence suspicion
People's image obtains respectively is overlapped the corresponding at least two suspects image collection of suspect's image；Schemed based at least two suspects
Image set, which closes, obtains corresponding at least two case；At least two cases of acquisition are subjected to processing of combining related cases.
The number that occurs in different cases of the statistics same person, if it is greater than or equal to then judging that the people may occur to go here and there simultaneously twice
Case behavior.
In one or more optional embodiments, the present embodiment device further include:
Storage and display unit, for storing and/or showing at least two cases corresponding suspect figure of processing of combining related cases
Image set closes and its corresponding relevant information.
It, can be on an electronic device to suspect's image after combining related cases after the processing that carries out combining related cases to multiple cases
Set and its corresponding relevant information are shown, to obtain more case information, provide condition for clear up a criminal case；Storage is
In order to it post-process other combine related cases identify when, can be by the case and other as a whole of the case after combining related cases
Case carries out new processing of combining related cases.
Optionally, the corresponding relevant information of case comprises at least one of the following information:
The docket information of corresponding suspect's image, with detecing number information, case type information, case description information.
In one or more optional embodiments, the present embodiment device further include:
Determination unit is compared, is used for the received corresponding facial image to be checked of people to be checked and multiple suspect's image collections
In each suspect's image be compared, determine whether people to be checked is suspect based on comparison result.
When case collection in worksite is to new facial image, can using facial image as facial image to be checked with it is known
Multiple suspect's image collections in each suspect's image be compared, determine whether the corresponding people of the facial image to be checked has
Criminal record (there are matched suspect's images in suspect's image collection), if any criminal record can using the people to be checked as suspect into
One step is examined, is compared by recognition of face, and the efficiency that suspect determines is improved, and can quickly determine suspicion in spot
People.
Optionally, determination unit is compared, is specifically used for obtaining each in facial image to be checked and multiple suspect's image collections
Similarity between suspect's image；It is to obtain at least one target suspect's image in response to comparison result, determines people to be checked
For suspect；Similarity between target suspect image and facial image to be checked is greater than or equal to default similarity；In response to
Comparison result is not obtain target suspect's image, determines that people to be checked is not suspect.
Optionally, determination unit is compared, is also used to establish new suspect's image collection for facial image to be checked, and will be to
It looks into facial image and is stored in new suspect's image collection.
Optionally, determination unit each suspect in obtaining facial image to be checked and multiple suspect's image collections is compared to scheme
When similarity as between, for carrying out face characteristic extraction to facial image to be checked and each suspect's image respectively, obtain to
Look into the corresponding face characteristic to be checked of facial image and the corresponding suspicion face characteristic of suspect's image；Based on face characteristic to be checked with
The distance between each suspicion face characteristic obtains similarity between facial image to be checked and each suspect's image.
According to an aspect of an embodiment of the present invention, a kind of electronic equipment provided, including processor, processor include this
The device of combining related cases of any of the above-described embodiment of invention classification method.
According to an aspect of an embodiment of the present invention, a kind of electronic equipment provided, comprising: memory, it can for storing
It executes instruction；
And processor, it is combined related cases thereby completing the present invention method for being communicated with memory with executing executable instruction
State the operation of any embodiment.
According to an aspect of an embodiment of the present invention, a kind of computer storage medium provided, can for storing computer
The instruction of reading, instruction are performed the operation for executing method any of the above-described embodiment of the invention of combining related cases.
According to an aspect of an embodiment of the present invention, a kind of computer program provided, including computer-readable code, when
When running in equipment, processor in the equipment executes any for realizing present invention method of combining related cases computer-readable code
The instruction of one embodiment.
The embodiment of the present disclosure additionally provides a kind of electronic equipment, such as can be mobile terminal, personal computer (PC), puts down
Plate computer, server etc..Below with reference to Fig. 3, it illustrates the terminal device or the services that are suitable for being used to realize the embodiment of the present application
The structural schematic diagram of the electronic equipment 300 of device: as shown in figure 3, electronic equipment 300 includes one or more processors, communication unit
For example Deng, one or more of processors: one or more central processing unit (CPU) 301, and/or one or more figures
As processor (GPU) 313 etc., processor can according to the executable instruction being stored in read-only memory (ROM) 302 or from
Executable instruction that storage section 308 is loaded into random access storage device (RAM) 303 and execute various movements appropriate and place
Reason.Communication unit 312 may include but be not limited to network interface card, and the network interface card may include but be not limited to IB (Infiniband) network interface card.
Processor can with communicate in read-only memory 302 and/or random access storage device 303 to execute executable instruction,
It is connected by bus 304 with communication unit 312 and is communicated through communication unit 312 with other target devices, to completes the application implementation
The corresponding operation of any one method that example provides, for example, the associated video based on multiple cases, obtains the corresponding suspicion of each case
Doubt people's image collection；At least one set of coincidence suspect's image, every group of coincidence suspect are obtained from multiple suspect's image collections
Image corresponds at least two suspect's image collections；Corresponding at least two case of every group of coincidence suspect's image is gone here and there simultaneously
Case processing.
In addition, in RAM 303, various programs and data needed for being also stored with device operation.CPU301,ROM302
And RAM303 is connected with each other by bus 304.In the case where there is RAM303, ROM302 is optional module.RAM303 storage
Executable instruction, or executable instruction is written into ROM302 at runtime, executable instruction makes central processing unit (CPU)
301 execute the corresponding operation of above-mentioned communication means.Input/output (I/O) interface 305 is also connected to bus 304.Communication unit 312
It can integrate setting, may be set to be with multiple submodule (such as multiple IB network interface cards), and in bus link.
I/O interface 305 is connected to lower component: the importation 306 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 307 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage section 308 including hard disk etc.；
And the communications portion 309 of the network interface card including LAN card, modem etc..Communications portion 309 via such as because
The network of spy's net executes communication process.Driver 310 is also connected to I/O interface 305 as needed.Detachable media 311, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on as needed on driver 310, in order to read from thereon
Computer program be mounted into storage section 308 as needed.
It should be noted that framework as shown in Figure 3 is only a kind of optional implementation, it, can root during concrete practice
The component count amount and type of above-mentioned Fig. 3 are selected, are deleted, increased or replaced according to actual needs；It is set in different function component
It sets, separately positioned or integrally disposed and other implementations, such as the separable setting of GPU313 and CPU301 or can also be used
GPU313 is integrated on CPU301, the separable setting of communication unit, can also be integrally disposed on CPU301 or GPU313, etc..
These interchangeable embodiments each fall within protection scope disclosed in the disclosure.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be tangibly embodied in machine readable
Computer program on medium, computer program include the program code for method shown in execution flow chart, program code
It may include the corresponding instruction of corresponding execution method and step provided by the embodiments of the present application, for example, the related view based on multiple cases
Frequently, the corresponding suspect's image collection of each case is obtained；At least one set of be overlapped is obtained from multiple suspect's image collections to dislike
Doubt people's image, the corresponding at least two suspect's image collections of every group of coincidence suspect's image；By every group of coincidence suspect's image pair
At least two cases answered carry out processing of combining related cases.In such embodiments, which can pass through communications portion
309 are downloaded and installed from network, and/or are mounted from detachable media 311.In the computer program by central processing list
When member (CPU) 301 is executed, the above-mentioned function of limiting in the present processes is executed.
The present processes and device may be achieved in many ways.For example, can by software, hardware, firmware or
Software, hardware, firmware any combination realize the present processes and device.The said sequence of the step of for the method
Merely to be illustrated, the step of the present processes, is not limited to sequence described in detail above, special unless otherwise
It does not mentionlet alone bright.In addition, in some embodiments, also the application can be embodied as to record program in the recording medium, these programs
Including for realizing according to the machine readable instructions of the present processes.Thus, the application also covers storage for executing basis
The recording medium of the program of the present processes.
The description of the present application is given for the purpose of illustration and description, and is not exhaustively or by the application
It is limited to disclosed form.Many modifications and variations are obvious for the ordinary skill in the art.It selects and retouches
Embodiment is stated and be the principle and practical application in order to more preferably illustrate the application, and those skilled in the art is enable to manage
Solution the application is to design various embodiments suitable for specific applications with various modifications.
Claims (10)
- The method 1. one kind is combined related cases characterized by comprisingBased on the associated video of multiple cases, the corresponding suspect's image collection of each case is obtained；At least one set of coincidence suspect's image is obtained from multiple suspect's image collections, coincidence suspect figure described in every group As corresponding at least two suspect's image collections；Corresponding at least two case of coincidence suspect's image described in every group is subjected to processing of combining related cases.
- 2. the method according to claim 1, wherein the associated video based on multiple cases, obtains each The corresponding suspect's image collection of the case, comprising:Recognition of face is carried out at least two field pictures in the associated video of each case respectively, obtaining at least one includes Suspect's image of face；Suspect's image of the acquisition is based on case classification, obtains the corresponding suspect's image collection of each case.
- 3. method according to claim 1 or 2, which is characterized in that described to be obtained from multiple suspect's image collections Obtain at least one set of coincidence suspect's image, comprising:The multiple suspect's image collection is clustered, at least one cluster result, each cluster knot are obtained It include at least two suspect's images in fruit, at least two suspect's images in the cluster result correspond to the different suspicion Doubt people's image collection；Using each cluster result as suspect's image is overlapped described in one group, at least one set of coincidence suspect figure is obtained Picture.
- 4. according to the method described in claim 3, it is characterized in that, described carry out the multiple suspect's image collection Cluster, obtains at least one cluster result, comprising:Face characteristic extraction is carried out to suspect's images all in the multiple suspect's image collection, it is special to obtain multiple faces Sign；All suspect's images in the multiple suspect's image collection are clustered based on the face characteristic, are obtained at least One cluster result.
- 5. method according to claim 1 or 2, which is characterized in that described to be obtained from multiple suspect's image collections Obtain at least one set of coincidence suspect's image, comprising:Respectively by each suspect's image and other described suspect's image collections in each suspect's image collection In each suspect's image match, obtain at least one matching result, include at least in each matching result Two suspect's images, at least two suspect's images in the matching result correspond to different suspect's image sets It closes；Using each matching result as suspect's image is overlapped described in one group, at least one set of coincidence suspect figure is obtained Picture.
- 6. according to the method described in claim 5, it is characterized in that, described respectively will be in each suspect's image collection Each suspect's image is matched with each suspect's image in suspect's image collection described in other, is obtained at least One matching result, comprising:Obtain respectively each suspect's image in first suspect's image collection with it is each in second suspect's image collection Similarity between suspect's image, the first suspect image collection are in multiple suspect's image collections Any one, the second suspect image collection is in multiple suspect's image collections in addition to first suspect schemes All suspect's image collections outside image set conjunction；The similarity is greater than or equal at least two suspect's images of preset value as matching result.
- The device 7. one kind is combined related cases characterized by comprisingGather obtaining unit, for the associated video based on multiple cases, obtains the corresponding suspect's image of each case Set；It is overlapped suspect's unit, for obtaining at least one set of coincidence suspect's image from multiple suspect's image collections, The corresponding at least two suspect's image collections of coincidence suspect's image described in every group；It combines related cases processing unit, for being gone here and there simultaneously corresponding at least two case of coincidence suspect's image described in every group Case processing.
- 8. a kind of electronic equipment, which is characterized in that including processor, the processor includes as claimed in claim 6 combines related cases Device.
- 9. a kind of electronic equipment characterized by comprising memory, for storing executable instruction；And processor, for being communicated with the memory to execute the executable instruction to complete claim 1 to 5 times The operation for method of combining related cases described in meaning one.
- 10. a kind of computer storage medium, for storing computer-readable instruction, which is characterized in that described instruction is held The operation for method of combining related cases described in 1 to 5 any one of perform claim requirement when row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810922364.3A CN109271859A (en) | 2018-08-14 | 2018-08-14 | It combines related cases method and apparatus, electronic equipment, computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810922364.3A CN109271859A (en) | 2018-08-14 | 2018-08-14 | It combines related cases method and apparatus, electronic equipment, computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109271859A true CN109271859A (en) | 2019-01-25 |
Family
ID=65153307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810922364.3A Pending CN109271859A (en) | 2018-08-14 | 2018-08-14 | It combines related cases method and apparatus, electronic equipment, computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109271859A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947824A (en) * | 2019-03-21 | 2019-06-28 | 赖雪英 | Ask lonely data application processing system and its method |
| CN110008362A (en) * | 2019-03-05 | 2019-07-12 | 青岛海信网络科技股份有限公司 | A kind of case classifying method and device |
| CN110619064A (en) * | 2019-08-29 | 2019-12-27 | 苏州千视通视觉科技股份有限公司 | Case studying and judging method and device based on deep learning |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127241A (en) * | 2016-06-17 | 2016-11-16 | 中国电子科技集团公司第二十八研究所 | One is combined related cases sorting technique and categorizing system of combining related cases |
| CN106709468A (en) * | 2016-12-31 | 2017-05-24 | 北京中科天云科技有限公司 | City region surveillance system and device |
| CN108154171A (en) * | 2017-12-20 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of character recognition method, device and electronic equipment |
| CN108229321A (en) * | 2017-11-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | Human face recognition model and its training method and device, equipment, program and medium |
| CN108229335A (en) * | 2017-12-12 | 2018-06-29 | 深圳市商汤科技有限公司 | It is associated with face identification method and device, electronic equipment, storage medium, program |
| CN108228742A (en) * | 2017-12-15 | 2018-06-29 | 深圳市商汤科技有限公司 | Face duplicate checking method and apparatus, electronic equipment, medium, program |
| CN108229419A (en) * | 2018-01-22 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | For clustering the method and apparatus of image |
-
2018
- 2018-08-14 CN CN201810922364.3A patent/CN109271859A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106127241A (en) * | 2016-06-17 | 2016-11-16 | 中国电子科技集团公司第二十八研究所 | One is combined related cases sorting technique and categorizing system of combining related cases |
| CN106709468A (en) * | 2016-12-31 | 2017-05-24 | 北京中科天云科技有限公司 | City region surveillance system and device |
| CN108229321A (en) * | 2017-11-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | Human face recognition model and its training method and device, equipment, program and medium |
| CN108229335A (en) * | 2017-12-12 | 2018-06-29 | 深圳市商汤科技有限公司 | It is associated with face identification method and device, electronic equipment, storage medium, program |
| CN108228742A (en) * | 2017-12-15 | 2018-06-29 | 深圳市商汤科技有限公司 | Face duplicate checking method and apparatus, electronic equipment, medium, program |
| CN108154171A (en) * | 2017-12-20 | 2018-06-12 | 北京奇艺世纪科技有限公司 | A kind of character recognition method, device and electronic equipment |
| CN108229419A (en) * | 2018-01-22 | 2018-06-29 | 百度在线网络技术（北京）有限公司 | For clustering the method and apparatus of image |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008362A (en) * | 2019-03-05 | 2019-07-12 | 青岛海信网络科技股份有限公司 | A kind of case classifying method and device |
| CN109947824A (en) * | 2019-03-21 | 2019-06-28 | 赖雪英 | Ask lonely data application processing system and its method |
| CN110619064A (en) * | 2019-08-29 | 2019-12-27 | 苏州千视通视觉科技股份有限公司 | Case studying and judging method and device based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Exploring background-bias for anomaly detection in surveillance videos | |
| CN108229335A (en) | It is associated with face identification method and device, electronic equipment, storage medium, program | |
| CN112491796B (en) | Intrusion detection and semantic decision tree quantitative interpretation method based on convolutional neural network | |
| CN109635146B (en) | Target query method and system based on image characteristics | |
| US20100287161A1 (en) | System and related techniques for detecting and classifying features within data | |
| CN102163290A (en) | Method for modeling abnormal events in multi-visual angle video monitoring based on temporal-spatial correlation information | |
| CN109271859A (en) | It combines related cases method and apparatus, electronic equipment, computer storage medium | |
| CN109872541A (en) | A kind of information of vehicles analysis method and device | |
| CN112183162A (en) | Face automatic registration and recognition system and method in monitoring scene | |
| CN112507860A (en) | Video annotation method, device, equipment and storage medium | |
| CN103279744A (en) | Multi-scale tri-mode texture feature-based method and system for detecting counterfeit fingerprints | |
| Chen et al. | Identifying tampering operations in image operator chains based on decision fusion | |
| Wang et al. | Mutuality-oriented reconstruction and prediction hybrid network for video anomaly detection | |
| Zhang et al. | A Multiple Instance Learning and Relevance Feedback Framework for Retrieving Abnormal Incidents in Surveillance Videos. | |
| CN116824488A (en) | Target detection method based on transfer learning | |
| Li et al. | A Siamese inception architecture network for person re-identification | |
| KR101170676B1 (en) | Face searching system and method based on face recognition | |
| CN115643153A (en) | Alarm correlation analysis method based on graph neural network | |
| CN109614893B (en) | Intelligent abnormal behavior track identification method and device based on situation reasoning | |
| CN115842645A (en) | UMAP-RF-based network attack traffic detection method and device and readable storage medium | |
| CN105320704B (en) | Trans-regional similar vehicle search method and device | |
| CN114168948A (en) | Network security situation comprehensive analysis method | |
| CN113343010A (en) | Image identification method, electronic equipment and computer readable storage medium | |
| CN112651996A (en) | Target detection tracking method and device, electronic equipment and storage medium | |
| CN115442149B (en) | Data intrusion analysis method and server based on deep learning |
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: 20190125 |
|
| RJ01 | Rejection of invention patent application after publication |