CN107783995A - A kind of destination object search method and device - Google Patents
A kind of destination object search method and device Download PDFInfo
- Publication number
- CN107783995A CN107783995A CN201610734166.5A CN201610734166A CN107783995A CN 107783995 A CN107783995 A CN 107783995A CN 201610734166 A CN201610734166 A CN 201610734166A CN 107783995 A CN107783995 A CN 107783995A
- Authority
- CN
- China
- Prior art keywords
- image
- information
- temporal information
- positional
- time threshold
- 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.)
- Granted
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
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Computational Linguistics (AREA)
- Closed-Circuit Television Systems (AREA)
- Processing Or Creating Images (AREA)
Abstract
The embodiments of the invention provide a kind of destination object search method and device, methods described to include：The comparison picture for including destination object is obtained, retrieval is more than the first image of predetermined threshold with the similarity for comparing picture in monitoring image, and obtains positional information and temporal information corresponding to each first image；Obtain known position information and temporal information pair；Positional information and temporal information are filtered with the known position information and temporal information to unmatched first image, the second image after being filtered., can be by positional information and temporal information pair that known target object occurs in the embodiment of the present invention, the result obtained to retrieval filters, so as to improve the accuracy of destination object retrieval result.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to a kind of destination object search method and device.
Background technology
, can be by being adopted in area to be monitored placement of images collecting device, and to each image with the development of monitoring technology
Collection equipment is configured, so as to get the image of the condition of satisfaction.For example, when setting image capture device to capture face,
Whether the image capture device face will occurs in Real time identification its monitor area, obtain including people if it is, capturing
The image of face.Or it is to capture vehicle that can also set image capture device, in this case, the image capture device will
Whether there is vehicle in its monitor area of Real time identification, if it is, capturing the image for obtaining including vehicle.
After monitoring image is obtained, it is generally the case that needs are retrieved in substantial amounts of monitoring image, are met
The image of specified conditions.For example, during police solves a case, it usually needs specific target is retrieved in substantial amounts of monitoring image
Object, such as people, vehicle, to obtain useful clue to solve the case.
Existing destination object search method, by providing a comparison picture for including destination object, then a large amount of
Monitoring image in retrieval compare image of the picture similarity more than threshold value with this, and using the image retrieved as comprising target
The image of object.
But for different objects, such as people, vehicle, there is similitude in itself.That is, it is understood that there may be multiple
The higher object of similarity, so as to cause to include the image of different objects, its similarity also can be higher.Therefore, existing target
Object search method, retrieval obtains in substantial amounts of monitoring image and the comparison picture similarity comprising destination object is more than threshold value
Image, and using the image retrieved as during the image comprising destination object, there may be not in the image for retrieving to obtain
Image comprising destination object, so as to cause retrieval result accuracy relatively low.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of destination object search method and device, to improve destination object inspection
The accuracy of hitch fruit.Concrete technical scheme is as follows：
In a first aspect, the embodiments of the invention provide a kind of destination object search method, methods described includes：
The comparison picture for including destination object is obtained, retrieves in monitoring image and is more than with the similarity for comparing picture
First image of predetermined threshold, and obtain positional information and temporal information corresponding to each first image；
Obtain known position information and temporal information pair；
Positional information and temporal information are entered with the known position information and temporal information to unmatched first image
Row filtering, the second image after being filtered.
Alternatively, it is described acquisition known position information and temporal information to including：
Receive location sets and the period of user's input；
By each position and the initial time of the period, the combination for terminating time composition, it is defined as known position information
With temporal information pair.
Alternatively, it is described to positional information and temporal information with the known position information and temporal information to unmatched
First image, which carries out filtering, to be included：
For each first image, the temporal information and each known position information and temporal information of first image are calculated
The difference of centering temporal information；
According to the time threshold between each position information being locally stored, obtain the positional information of first image with it is each known
Time threshold between positional information and temporal information center position information；
Judge whether each difference is respectively less than corresponding time threshold, if it is, filtering first image.
Alternatively, it is described acquisition known position information and temporal information to including：
The each position of user's input and corresponding each moment are received, and using each position and corresponding each moment as known bits
Confidence ceases and temporal information pair；Or
Show described first image；The selection instruction of user's input is received, and by position corresponding to each 3rd image chosen
Confidence ceases and temporal information is as known position information and temporal information pair；Or
Described first image is shown, the selection instruction of user's input is received, it is determined that position corresponding to each 3rd image chosen
Confidence ceases and temporal information；And each position of user's input and corresponding each moment are received, by corresponding to each 3rd image
Positional information and temporal information, and each position of user's input and corresponding each moment believe as known position information and time
Breath pair.
Alternatively, the displaying described first image includes：
Show the first image of predetermined quantity.
Alternatively, it is described to positional information and temporal information with the known position information and temporal information to unmatched
First image, which carries out filtering, to be included：
For each first image, the temporal information and each known position information and temporal information of first image are calculated
The difference of centering temporal information；
According to the time threshold between each position information being locally stored, obtain the positional information of first image with it is each known
Time threshold between positional information and temporal information center position information；
Judge whether each difference is all higher than corresponding time threshold, if not, filtering first image.
Alternatively, storing the process of the time threshold between each position information includes：
For each two positional information, monitoring image corresponding to described two positional informations, and each monitoring image are obtained
Corresponding temporal information；
In monitoring image corresponding to described two positional informations, identification includes the monitoring image of same target；
For the monitoring image comprising same target, the time of each monitoring image of the corresponding one of positional information of calculating
The difference of information and the temporal information of each monitoring image of another corresponding positional information；
According to the difference being calculated, determine the time threshold between described two positional informations and be stored in local.
Alternatively, it is described according to the difference being calculated, determine that the time threshold between described two positional informations includes：
Minimum value in the difference is defined as the time threshold between described two positional informations；
According to the difference, the time threshold between described two positional informations is determined using quartile method；Or
According to the difference, the time threshold between described two positional informations is determined using slope diagnostic method.
Alternatively, methods described also includes：
At set time intervals, the time threshold between each position information being locally stored is periodically updated.
Alternatively, it is described filtered after the second image after, methods described also includes：
According to the positional information and temporal information of each second image, the motion of the destination object is shown on the electronic map
Track.
Second aspect, the embodiments of the invention provide a kind of destination object to retrieve device, and described device includes：
Processing module, for obtaining the comparison picture for including destination object, retrieval and the comparison chart in monitoring image
The similarity of piece is more than the first image of predetermined threshold, and obtains positional information and temporal information corresponding to each first image；
First acquisition module, for obtaining known position information and temporal information pair；
Filtering module, for positional information and temporal information with the known position information and temporal information to mismatching
The first image filtered, the second image after being filtered.
Alternatively, first acquisition module includes：
Receiving submodule, for receiving location sets and the period of user's input；
Determination sub-module, for the combination for forming each position and the initial time of the period, termination time, it is determined that
For known position information and temporal information pair.
Alternatively, the filtering module, including：
First calculating sub module, for for each first image, calculate the temporal information of first image with it is each
Know the difference of positional information and temporal information centering temporal information；
First acquisition submodule, for according to the time threshold between each position information being locally stored, obtaining first figure
Time threshold between the positional information of picture and each known position information and temporal information center position information；
First filter submodule, for judging whether each difference is respectively less than corresponding time threshold, if it is, filtering
First image.
Alternatively, first acquisition module, including：
First implementation sub-module, for each position for receiving user's input and corresponding each moment；And by each position and right
Each moment answered is as known position information and temporal information pair；Or
Second implementation sub-module, for showing described first image；The selection instruction of user's input is received, and will be chosen
Positional information and temporal information are as known position information and temporal information pair corresponding to each 3rd image；Or
3rd implementation sub-module, for showing described first image, the selection instruction of user's input is received, it is determined that choose
Positional information and temporal information corresponding to each 3rd image；And each position of user's input and corresponding each moment are received, by institute
Positional information and temporal information corresponding to each 3rd image, and each position of user's input and corresponding each moment are stated as
Know positional information and temporal information pair.
Alternatively, second implementation sub-module, the first image specifically for showing predetermined quantity.
Alternatively, the filtering module includes：
Second calculating sub module, for for each first image, calculate the temporal information of first image with it is each
Know the difference of positional information and temporal information centering temporal information；
Second acquisition submodule, for according to the time threshold between each position information being locally stored, obtaining first figure
Time threshold between the positional information of picture and each known position information and temporal information center position information；
Second filter submodule, for judging whether each difference is all higher than corresponding time threshold, if not, filter this
One image.
Alternatively, described device also includes：
Second acquisition module, for for each two positional information, obtaining monitoring figure corresponding to described two positional informations
Picture, and temporal information corresponding to each monitoring image；
Identification module, in monitoring image corresponding to described two positional informations, identification to include the prison of same target
Control image；
Computing module, for for the monitoring image comprising same target, calculating each of corresponding one of positional information
The difference of the temporal information of monitoring image and the temporal information of each monitoring image of another corresponding positional information；
Determining module, for according to the difference being calculated, determining the time threshold between described two positional informations simultaneously
It is stored in local.
Alternatively, the determining module, believe specifically for the minimum value in the difference is defined as into described two positions
Time threshold between breath；
According to the difference, the time threshold between described two positional informations is determined using quartile method；Or
According to the difference, the time threshold between described two positional informations is determined using slope diagnostic method.
Alternatively, described device also includes：
Update module, at set time intervals, periodically update between each position information being locally stored when
Between threshold value.
Alternatively, described device also includes：
Display module, for the positional information and temporal information according to each second image, on the electronic map described in displaying
The movement locus of destination object.
The embodiments of the invention provide a kind of destination object search method and device, by image matching method in monitoring figure
After retrieval obtains and compared picture similarity more than the image of predetermined threshold as in, it can believe according further to known position
Breath and temporal information pair, the obtained image of retrieval is filtered, filtered out with known position information and temporal information to not
The image matched somebody with somebody.In destination object retrieving, the retrieval result for being clearly present mistake may be retrieved, such as when destination object goes out
During present two positions, it is less than in the time difference that two positions occur in actual conditions, and destination object therefrom arrive by a position
The minimum time of another position.In this case, when the positional information and temporal information pair that known target object occurs, i.e.,
The result that can be obtained to retrieval filters, and unmatched image filtering will be fallen with known position information and temporal information, from
And the accuracy of destination object retrieval result can be improved.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the flow chart of destination object search method provided in an embodiment of the present invention；
Fig. 2 is another flow chart of destination object search method provided in an embodiment of the present invention；
Fig. 3 is the structural representation that destination object provided in an embodiment of the present invention retrieves device.
Embodiment
In order to improve the accuracy of destination object retrieval result, the embodiments of the invention provide a kind of destination object retrieval side
Method and device.
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
It should be noted that in the case where not conflicting, the feature in embodiment and embodiment in the present invention can phase
Mutually combination.Describe the present invention in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
In order to improve the accuracy of destination object retrieval result, the embodiments of the invention provide a kind of destination object retrieval side
Method process, as shown in figure 1, the process may comprise steps of：
S101, the comparison picture for including destination object is obtained, retrieved in monitoring image and compare the similar of picture to described
Degree is more than the first image of predetermined threshold, and obtains positional information and temporal information corresponding to each first image.
Method provided in an embodiment of the present invention can apply to electronic equipment.Specifically, the electronic equipment can be desk-top
Computer, portable computer, intelligent mobile terminal, server etc..
In embodiments of the present invention, electronic equipment can retrieve the image for including destination object in monitoring image.Specifically
Ground, electronic equipment can obtain the comparison picture comprising destination object, and then retrieval and the comparison chart in monitoring image first
The similarity of piece is more than the first image of predetermined threshold.Above-mentioned predetermined threshold is such as can be 60%, 70%, 80%.
Electronic equipment can receive the comparison picture comprising destination object of user's input.Carrying out destination object retrieval
When, it is generally the case that destination object is known.Also, user can provide the comparison picture comprising destination object, and input
In electronic equipment.
For example, user can directly obtain the comparison picture comprising destination object；Or user can be according to known mesh
The feature of object is marked, obtains including the comparison picture of destination object, and input in electronic equipment.
And, it is generally the case that after the image capture device of monitor area collects monitoring image, electron can be sent and set
It is standby, or other storage devices.Therefore, electronic equipment can be set according to the storage region of monitoring image from local or other storages
Standby middle acquisition monitoring image.
Wherein, electronic equipment can obtain all monitoring images that it can get, and be examined in all monitoring images
Rope is more than the first image of predetermined threshold with comparing picture similarity.
Or in some cases, user may already know that the time range and/or regional extent that needs are retrieved.Also,
In order to improve recall precision, the alternative condition that electronic equipment can be set according to user, e.g., user select the region model that need to be retrieved
Enclose, and/or time range etc., the monitoring image for meeting the alternative condition that user sets is obtained from all monitoring images, and
The first image that picture similarity is more than predetermined threshold is retrieved and compared in the monitoring image of acquisition.
It should be noted that the of picture similarity more than predetermined threshold is retrieved and compared to electronic equipment from monitoring image
The process of one image, prior art can be used, e.g., can be obtained using existing image matching method with comparing picture phase
Like first image of the degree more than predetermined threshold, or, it can be modeled to comparing picture with monitoring image, convert it into
Model data, and then be compared according to model data, the embodiment of the present invention is to this process without repeating.
Alternatively, when electronic equipment is compared by comparing picture with the model data of monitoring image, obtains and compare
When picture similarity is more than the first image of predetermined threshold, in order to improve recall precision, electronic equipment can be in advance to each monitoring
Image is modeled, and obtains the model data of each monitoring image.In this case, when being retrieved, electronic equipment can be straight
The model data for getting each monitoring image is obtained, and is only modeled to comparing picture, you can by comparing picture and monitoring figure
The first image that picture similarity is more than predetermined threshold is retrieved to obtain and compared to the model data of picture.
After retrieving the first image, electronic equipment can further obtain positional information and time corresponding to each first image
Information.Under normal circumstances, when image capture device collects monitoring image and is sent to electronic equipment or other storage devices, electricity
Sub- equipment or storage device can preserve positional information corresponding to each monitoring image and time letter simultaneously when preserving monitoring image
Breath.For example, the positional information of each monitoring image and temporal information can be stored in predetermined storage location.
As one embodiment of the present invention, above-mentioned positional information can be that the IMAQ for gathering each monitoring image is set
The location of standby information, at the time of above-mentioned temporal information can be that image capture device gathers each monitoring image.
Therefore, after electronic equipment retrieves the first image, each first image pair can be obtained at corresponding storage location
The positional information and temporal information answered.
S102, obtain known position information and temporal information pair.
In embodiments of the present invention, after electronic equipment retrieves the first image, in order to improve destination object retrieval result
Accuracy, known positional information and temporal information pair can be obtained, and according to known positional information and temporal information pair, it is right
First image is filtered.Pair of above-mentioned known position information and temporal information to as known positional information and temporal information
It should be related to.That is, each position that known destination object occurs, and it appears in the corresponding time of each position.
In one embodiment of the present invention, electronic equipment can be believed according to each object known position information and time is included
The database of breath pair, and/or information is inputted according to user, to determine the known position information of destination object and temporal information pair.
For example, electronic equipment can locally build a database in advance, and occurred according to each object having determined
Each position, and appear in the time of each position, preserve known position information and temporal information pair corresponding to each object.And
And the identification information of each object, characteristic information etc. can also be preserved.
When electronic equipment carries out destination object retrieval, its can according to the identification information of destination object, characteristic information etc.,
The object of matching is searched in the database locally preserved.If finding, by the known location of the object preserved in database
Information and temporal information are to the known position information and temporal information pair as destination object.
Or user can determine one or more positions that destination object occurs, and time corresponding to each position.This
In the case of kind, electronic equipment can determines known position information and temporal information pair according to these positions and time.
Specifically, electronic equipment can receive the positional information and temporal information of user's input, and according to the positional information
Known position information and temporal information pair are determined with temporal information.
In one implementation, user can determine that certain moment destination object appears in certain specific opening position.Such as, use
Family can determine 2016.3.12,2：01：00, destination object appears in P1 opening positions；2016.3.12 2：34：00, destination object
Appear in P2 opening positions etc..In this case, known each position and corresponding each moment can be inputted electronic equipment by user
In, as user can input positional information and temporal information as shown in the table：
| Positional information | Temporal information |
| P1 | 2016.3.12 2：01：00 |
| P2 | 2016.3.12 2：34：00 |
So as to, electronic equipment can receive each position of user's input and corresponding each moment, and by each position and correspondingly
Each moment as known position information and temporal information pair.
In another implementation, user may only determine location sets and the period of destination object appearance.Such as, use
Family determines destination object in 2016.3.12,2：01：00-2016.3.12,3：34：In P1, P2, P3 opening position in 00 this period
Occurred.In this case, user can input known location sets and period in electronic equipment, as user can be with defeated
Enter 2016.3.12,2：01：00-2016.3.12,3：34：00 and P1, P2, P3.
So as to which electronic equipment can receive location sets and the period of user's input.Further, electronic equipment can be with
By each position and the initial time of period, the combination for terminating time composition, it is defined as known position information and temporal information pair.
For example, when the location sets and period that electronic equipment receives are respectively P1, P2, P3 and 2016.3.12,2：
01：00-2016.3.12,3：34：When 00, its confirmable known position information and temporal information pair can be with as shown in the table：
| Positional information | Temporal information |
| P1 | 2016.3.12 2：01：00 |
| P1 | 2016.3.12 3：34：00 |
| P2 | 2016.3.12 2：01：00 |
| P2 | 2016.3.12 3：34：00 |
| P3 | 2016.3.12 2：01：00 |
| P3 | 2016.3.12 3：34：00 |
In another implementation, in some cases, user may not determine the positional information that destination object occurs,
And/or temporal information.In this case, electronic equipment can show the first image, with the first image according to displaying, Yi Jiyong
The selection at family determines known position information and temporal information pair.
For example, electronic equipment can show the first all images.Or in order to improve determine known position information and when
Between information pair efficiency, electronic equipment can show the first image of predetermined quantity, such as 50,60,100.
Alternatively, in order to further improve the efficiency of determination known position information and temporal information pair, electronic equipment exists
During all or part of first image of displaying, according to each first image and the similarity for comparing picture, according to similarity from big to small
Order, show the first image.The first bigger image of similarity, wherein the possibility comprising destination object is bigger, therefore, is pressed
According to similarity size order, the first image is shown, user can be made quickly to find the image comprising destination object, so as to improve really
Determine the efficiency of known position information and temporal information pair.
After electronic equipment shows the first image, user may determine that in each first image whether include destination object, and select
In include the image of destination object.And then electronic equipment can receive the selection instruction of user's input, and each image that will be chosen
Corresponding positional information and temporal information are as known position information and temporal information pair.
It should be noted that when obtaining known position information and temporal information pair, electronic equipment can be according to above-mentioned one
Kind or various ways obtain known position information and temporal information pair.Such as, electronic equipment can be according to the data locally preserved
Positional information and temporal information corresponding to each 3rd image that storehouse, the positional information of user's input and temporal information, user choose
In one kind, to determine known position information and temporal information pair；Or electronic equipment can also be according to the data locally preserved
Positional information and temporal information corresponding to each 3rd image that storehouse, the positional information of user's input and temporal information, user choose
In two or three, common to determine known position information and temporal information pair, the present invention is not limited this.
S103, to positional information and temporal information with the known position information and temporal information to unmatched first figure
As being filtered, the second image after being filtered.
Electronic equipment gets known position information and temporal information to rear, can to positional information and temporal information with
Know that positional information and temporal information filter to unmatched first image, the second image after being filtered.
Each position that the destination object that known position information and temporal information pair, as user determine occurs and to it is corresponding when
Carve, according to the known position information and temporal information pair, may further determine that whether included in each first image is target
Object.
For example, electronic equipment can preset the path length and the relation of shortest time used that destination object is passed by.
Specifically, the maximal rate for the destination object that electronic equipment can be set according to user, to determine that the destination object is passed by one section
Shortest time used in path, and locally preserving each path length and the corresponding relation of shortest time.
In this case, electronic equipment can be directed to each first image, according to each known position information and temporal information pair,
Determine whether to filter first image.
For example, being directed to any first image, electronic equipment can be determined between first image and any known location
Path length, and according to the path length, destination object is searched from the positional information of first image to the known location place
The shortest time needed.And then the difference of temporal information corresponding to the temporal information and the known location of first image can be calculated
Value, and judge whether the difference is less than the shortest time determined, if it is, showing that what is included in the first image is unlikely to be target
Object.In this case, electronic equipment can fall first image filtering.
After the second image after being filtered, electronic equipment can be believed according to the positional information of each second image and time
Breath, determine the motion conditions of destination object.For example, during police solves a case, can according to the positional information of the second image and
Temporal information, the position that destination object occurred is determined, and appear in the time of each position；Or target can also be determined
The movement locus of object, so as to obtain useful clue to solve the case.
The embodiments of the invention provide a kind of destination object search method, is examined by image matching method in monitoring image
Rope obtain and compare picture similarity more than predetermined threshold image after, can according further to known positional information and when
Between information pair, the obtained image of retrieval is filtered, filtered out with known position information and temporal information to unmatched figure
Picture.In destination object retrieving, the retrieval result for being clearly present mistake may be retrieved, such as when destination object appears in two
During individual position, it is less than in the time difference that two positions occur in actual conditions, and another is arrived in a position to destination object therefrom
The minimum time of position.In this case, when the positional information and temporal information pair that known target object occurs, you can to inspection
The result that rope obtains is filtered, and unmatched image filtering will be fallen with known position information and temporal information, so as to
Improve the accuracy of destination object retrieval result.
As a kind of embodiment of the embodiment of the present invention, electronic equipment filtered after the second image after, may be used also
With the movement locus of the positional information and temporal information according to each second image, on the electronic map displaying destination object, so that
User is intuitive to see the mass motion track of destination object, further instructs user accurately to analyze destination object.
For example, electronic equipment can according to the positional information and temporal information of each second image, according to time order and function order,
The position of each second image is connected, you can obtain the movement locus of destination object.
Alternatively, electronic equipment can show target on the electronic map according to the positional information of the second all images
The movement locus of object.Or in some cases, user may need to show destination object in the range of specific region ,/or it is special
Movement locus in the range of fixing time.In this case, electronic equipment can show each second image, so that user selects it to need
The image to be shown, so as to the positional information for each second image that electronic equipment can be chosen according to user, show destination object
Movement locus.
As a kind of embodiment of the embodiment of the present invention, in order to improve destination object recall precision, electronic equipment can be with
Time threshold between each position information is locally stored in advance, according to the time threshold, to be filtered to the first image.
It is appreciated that known position information and time that each position and corresponding each moment that are inputted according to user determine are believed
After breath pair, or each first image of electronic equipment displaying, the positional information of image and temporal information is chosen to determine according to user
Know positional information and temporal information pair, be accurately known positional information and temporal information pair.
By this known position information and temporal information pair, when being filtered to the first image, as long as the first image
Positional information and temporal information and one of known position information and temporal information are to mismatching, you can determine the first image not
Meet search condition, electronic equipment can fall first image filtering.
And the known position information and temporal information pair that the location sets and period for user's input determine, being possible
Positional information and temporal information pair.
By this known position information and temporal information pair, when being filtered to the first image, as long as the first image
Positional information and temporal information are with one of known position information and temporal information to matching, you can determine that the first image may
Meet search condition, electronic equipment can not fall first image filtering.The only positional information of the first image and time letter
Breath, to mismatching, can just determine that the first image does not meet search condition with all known position information and temporal information,
Electronic equipment can fall first image filtering.
Therefore, different sides can be used for different types of known position information and temporal information pair, electronic equipment
Formula filters to the first image.
Introduce first and be directed to accurately known positional information and temporal information pair, electronic equipment is filtered to the first image
Method.
Specifically, electronic equipment first against each first image, can calculate the temporal information of first image with it is every
The difference of individual known position information and temporal information centering temporal information.
In embodiments of the present invention, electronic equipment can be directed to each first figure when being filtered to each first image
Picture, calculate the temporal information and each known position information and the difference of temporal information centering temporal information of first image.
For example, when the positional information and temporal information of the first image are P1,2016.3.3,3：33：33, it is known that position letter
Breath and temporal information are to as shown in the table：
| Positional information | Temporal information |
| P1 | 2016.3.3 4：00：00 |
| P2 | 2016.3.3 3：00：00 |
| P3 | 2016.3.3 4：30：30 |
When the temporal information of the first image is calculated with each known position information and temporal information centering in electronic equipment
Between the difference of information be respectively：
| Positional information pair | Time difference |
| |T1-T1| | 26 points 27 seconds |
| |T1-T2| | 33 points 33 seconds |
| |T1-T3| | 56 points 27 seconds |
Wherein, | T1-T1 | it is the time used in destination object from P1 opening positions to P1 opening positions, | T1-T2 | it is destination object
From the time used in P1 opening positions to P2 opening positions, | T1-T3 | it is the time used in destination object from P1 opening positions to P3 opening positions.
Obtain the first image and after the difference of each known position information and temporal information centering temporal information, electronic equipment can
With further according to the time threshold between each position information that is locally stored, obtain the positional information of first image with it is each known
Time threshold between positional information and temporal information center position information.
For example, electronic equipment can be according to the identification information of each position, the time threshold between each two position is locally stored
Value.And then electronic equipment can be each from what is be locally stored according to the positional information of the first image, and each known position information
The positional information of the first image is searched in time threshold between positional information with each known position information and temporal information to middle position
Time threshold between confidence breath.
Still by taking above-mentioned example as an example, electronic equipment can be obtained between P1 and P1, between P1 and P2, and between P1 and P3
Time threshold, its obtain time threshold can be with as shown in the table：
| Positional information pair | Time threshold |
| T1-1 | 0 |
| T1-2 | 30 minutes |
| T1-3 | 50 minutes |
The temporal information and each known position information and the difference of temporal information centering temporal information of the first image are obtained,
And first image positional information and each known position information and temporal information center position information between time threshold after, electricity
Sub- equipment may determine that whether each difference is all higher than corresponding time threshold, if not, filtering first image.
For example, in above-mentioned example, the time 26 used in destination object from P1 opening positions to P1 opening positions, which divides to be more than for 27 seconds, to be corresponded to
Time threshold 0；Time 33 used in destination object from P1 opening positions to P2 opening positions, which divides, is more than corresponding time threshold 30 for 33 seconds
Minute；Time 56 used in destination object from P1 opening positions to P3 opening positions, which divides, is more than corresponding time threshold for 27 seconds 50 minutes, respectively
Time difference is all higher than corresponding time threshold, and therefore, the first image does not filter.
When such as T1-2 being 35 minutes, the time 33 used in destination object from P1 opening positions to P2 opening positions divide be less than within 33 seconds it is corresponding
Time threshold 35 minutes, in this case, electronic equipment can fall the first image filtering.
It is described below and the first image is filtered for possible known position information and temporal information pair, electronic equipment
Method.
Specifically, electronic equipment first against each first image, can calculate the temporal information of first image with it is every
The difference of individual known position information and temporal information centering temporal information.
In embodiments of the present invention, electronic equipment can be directed to each first figure when being filtered to each first image
Picture, calculate the temporal information and each known position information and the difference of temporal information centering temporal information of first image.
For example, when the positional information and temporal information of the first image are P1,2016.3.12,3：33：00, it is known that position
Information and temporal information are to as shown in the table：
When the temporal information of the first image is calculated with each known position information and temporal information centering in electronic equipment
Between the difference of information be respectively：
| Positional information pair | Time difference |
| |T1-T1| | 32 minutes 1 hour |
| |T1-T1| | 1 minute |
| |T1-T2| | 32 minutes 1 hour |
| |T1-T2| | 1 minute |
| |T1-T3| | 32 minutes 1 hour |
| |T1-T3| | 1 minute |
Wherein, | T1-T1 | it is the time used in destination object from P1 opening positions to P1 opening positions, | T1-T2 | it is destination object
From the time used in P1 opening positions to P2 opening positions, | T1-T3 | it is the time used in destination object from P1 opening positions to P3 opening positions.
Obtain the first image and after the difference of each known position information and temporal information centering temporal information, electronic equipment can
With further according to the time threshold between each position information that is locally stored, obtain the positional information of first image with it is each known
Time threshold between positional information and temporal information center position information.
For example, electronic equipment can be according to the identification information of each position, the time threshold between each two position is locally stored
Value.And then electronic equipment can be each from what is be locally stored according to the positional information of the first image, and each known position information
The positional information of the first image is searched in time threshold between positional information with each known position information and temporal information to middle position
Time threshold between confidence breath.
Still by taking above-mentioned example as an example, electronic equipment can be obtained between P1 and P1, between P1 and P2, and between P1 and P3
Time threshold, its obtain time threshold can be with as shown in the table：
| Positional information pair | Time threshold |
| T1-1 | 0 |
| T1-2 | 2 hours |
| T1-3 | 50 minutes |
The temporal information and each known position information and the difference of temporal information centering temporal information of the first image are obtained,
And first image positional information and each known position information and temporal information center position information between time threshold after, electricity
Sub- equipment may determine that whether each difference is respectively less than corresponding time threshold, if it is, filtering first image.
For example, in above-mentioned example, 32 minutes time 1 hour used in destination object from P1 opening positions to P1 opening positions was more than pair
The time threshold 0 answered；Another 1 minute time is less than corresponding time threshold 0；Destination object is from P1 opening positions to P2 positions place
It was less than corresponding time threshold 2 hours with 32 minutes time 1 hour；It is small that another 1 minute time is less than corresponding time threshold 2
When；32 minutes time 1 hour used in destination object from P1 opening positions to P3 opening positions was more than corresponding time threshold 50 minutes；Separately
One 1 minute time was less than corresponding time threshold 50 minutes, and existence time difference is more than the situation of corresponding time threshold, therefore,
First image does not filter.
In above-mentioned example, when known positional information and temporal information are to for shown in following table：
| Positional information | Temporal information |
| P2 | 2016.3.12 2：01：00 |
| P2 | 2016.3.12 3：34：00 |
| P3 | 2016.3.12 2：01：00 |
| P3 | 2016.3.12 3：34：00 |
And time threshold is as shown in the table：
At this point it is possible to it is corresponding to determine that 32 minutes time 1 hour used in destination object from P1 opening positions to P2 opening positions was less than
Time threshold 2 hours；Another 1 minute time is less than corresponding time threshold 2 hours；Destination object is from P1 opening positions to P3 positions
Put place and be less than corresponding time threshold 3 hour 20 minutes with 32 minutes time 1 hour；Another 1 minute time is less than corresponding
Time threshold 3 hour 20 minutes, i.e., all time differences are respectively less than corresponding time threshold, in this case, filter the first figure
Picture.
In this programme, electronic equipment can be according to the time threshold between each position information locally preserved, to the first image
Filtered, so as to improve the accuracy of destination object retrieval result.
As a kind of embodiment of the embodiment of the present invention, in order to improve the accuracy filtered to the first image, electricity
Time threshold between each position information of sub- equipment storage can determine according to historical data.It can be set according to each IMAQ
The monitoring image of standby collection, and the positional information of each monitoring image and temporal information determine.
Specifically, as shown in Fig. 2 destination object search method provided in an embodiment of the present invention, can also include following step
Suddenly：
S201, for each two positional information, obtain monitoring image corresponding to described two positional informations, and each monitoring
Temporal information corresponding to image.
In embodiments of the present invention, electronic equipment can be directed to each two positional information, obtain two positional informations pair
The monitoring image answered, and temporal information corresponding to each monitoring image.
Specifically, electronic equipment can be directed to two positional informations, obtain the IMAQ at the two positional informations and set
All monitoring images of standby collection, and the temporal information of each monitoring image.Or in order to improve the determination of time threshold effect
Rate, electronic equipment can obtain the monitoring image in the scheduled time of the image capture device collection of the two opening positions, such as obtain
Take the monitoring image of nearest 5 days, 10 days or 1 month, and the temporal information of each monitoring image.
For example, when electronic equipment calculates the time threshold between P1 positions and P2 positions, the monitoring of its P1 opening position obtained
Image information can be with as shown in the table：
| Object | Temporal information |
| A | 8：00：00 |
| B | 9：00：00 |
| C | 10：00：00 |
| A | 11：00：00 |
| D | 12：00：00 |
The monitoring image information for the P2 opening positions that electronic equipment obtains can be with as shown in the table：
| Object | Temporal information |
| A | 15：00：00 |
| A | 16：00：00 |
| C | 17：00：00 |
| B | 18：00：00 |
| D | 19：00：00 |
S202, in monitoring image corresponding to described two positional informations, identification includes the monitoring image of same target.
After getting the monitoring image at two positional informations, electronic equipment can identify in the monitoring image of acquisition
Monitoring image comprising same target.
For example, electronic equipment can be by way of existing any image recognition, to the monitoring figure of two opening positions
As being compared two-by-two, identify in each two image whether include same target, and then identify the monitoring comprising same target
Image, the embodiment of the present invention is to this process without repeating.
Still by taking above-mentioned example as an example, electronic equipment can identify P1 opening positions 8：00：00 monitoring image, P2 opening positions
15：00：00 monitoring image and P2 opening positions 16：00：00 monitoring image is the monitoring image comprising same target.P1
Opening position 9：00：00 monitoring image and P2 opening positions 18：00：00 monitoring image is the monitoring figure comprising same target
Picture.P1 opening positions 10：00：00 monitoring image and P2 opening positions 17：00：00 monitoring image is to include same target
Monitoring image.P1 opening positions 11：00：00 monitoring image, P2 opening positions 15：00：00 monitoring image and P2 opening positions
16：00：00 monitoring image is the monitoring image comprising same target.P1 opening positions 12：00：00 monitoring image and P2
Opening position 19：00：00 monitoring image is the monitoring image comprising same target.
S203, for the monitoring image comprising same target, calculate each monitoring image for corresponding to one of positional information
Temporal information and the temporal information of each monitoring image of another corresponding positional information difference.
After identifying the monitoring image comprising same target, electronic equipment can be directed to the monitoring figure comprising same target
Picture, calculate the temporal information for each monitoring image for corresponding to one of positional information and each monitoring of another corresponding positional information
The difference of the temporal information of image.
For example, in the above example, for P1 opening positions 8：00：00 monitoring image, P2 opening positions 15：00：00 prison
Control image and P2 opening positions 16：00：00 monitoring image, electronic equipment can be directed to P1 opening positions 8：00：00 monitoring figure
Picture, calculate itself and P2 opening positions 15：00：00 monitoring image and P2 opening positions 16：00：The time letter of 00 monitoring image
The difference of breath is respectively 7 hours and 8 hours.
For P1 opening positions 9：00：00 monitoring image and P2 opening positions 18：00：00 monitoring image, electronic equipment
P1 opening positions 9 can be directed to：00：00 monitoring image, calculate itself and P2 opening positions 18：00：The time letter of 00 monitoring image
The difference of breath is 9 hours.
For P1 opening positions 10：00：00 monitoring image and P2 opening positions 17：00：00 monitoring image, electronics are set
It is standby to be directed to P1 opening positions 10：00：00 monitoring image, calculate itself and P2 opening positions 17：00：The time of 00 monitoring image
The difference of information is 7 hours.
For P1 opening positions 11：00：00 monitoring image, P2 opening positions 15：00：00 monitoring image and P2 positions
Place 16：00：00 monitoring image, electronic equipment can be directed to P1 opening positions 11：00：00 monitoring image, calculate itself and P2 positions
Put place 15：00：00 monitoring image and P2 opening positions 16：00：The difference of the temporal information of 00 monitoring image is respectively 4
Hour and 5 hours.
For P1 opening positions 12：00：00 monitoring image and P2 opening positions 19：00：00 monitoring image, electronics are set
It is standby to be directed to P1 opening positions 12：00：00 monitoring image, calculate itself and P2 opening positions 19：00：The time of 00 monitoring image
The difference of information is 7 hours.
S204, according to the difference being calculated, determine the time threshold between described two positional informations and be stored in this
Ground.
After the difference that two opening positions are respectively included between the monitoring image of same target is calculated, electronic equipment can enter one
Step ground determines the time threshold between two positional informations and is stored in local according to the difference being calculated.
In one implementation, the minimum value in the difference being calculated can be defined as two positions by electronic equipment
Time threshold between information.
Such as, in above-mentioned example, minimum value in each difference can be defined as between P1 positions and P2 positions by electronic equipment for 4 hours
Time threshold.
It is appreciated that the minimum value in the difference being calculated is object needed for from a position to another position
The minimum time.Minimum value in the difference being calculated is defined as the time threshold between two positional informations, can be ensured
According to the time threshold, when being filtered to the first image, the first image filtering comprising destination object will not be fallen, ensured
Filter the integrality of result.
In another implementation, when data volume is larger, such as data volume, can be according to meter more than 1000,2000
Obtained difference, the time threshold between two positional informations is determined using quartile method, to improve the accurate of time threshold
Property.
Specifically, electronic equipment can arrange the difference being calculated from small to large, and by first quartile, also
It is the number in a quarter opening position, as the time threshold between two positional informations.
In another implementation, electronic equipment can also according to the difference being calculated, using slope diagnostic method come
Determine the time threshold between two positional informations.It is appreciated that, it is generally the case that the more difference of occurrence number, it can most represent
Most situation, also best embody situation about occurring in practical application.Therefore, two positions are determined using slope diagnostic method
Between time threshold, can be effectively prevented from some due to not meeting the difference of actual conditions caused by cause specific being defined as
Time threshold, so as to improve the accuracy of time threshold.
Specifically, electronic equipment can be using the time as transverse axis, and quantity is the longitudinal axis, makees curve according to the difference being calculated,
And then determine the section of maximum slope on the curve, that is to say, that the quantity larger period, and during by the starting of the period
Between be used as time threshold.
Alternatively, over time, time of the destination object used in from a position to another position may
Change, so as to which the time threshold between two positions be able to may also change.Therefore, in embodiments of the present invention, in order to improve two
The accuracy of time threshold between position, electronic equipment can at set time intervals, such as 1 day, 3 days, 1 week or 1 month
Deng periodically updating the time threshold between each position information being locally stored.
In this programme, electronic equipment can according to corresponding to each position information monitoring image, accurately determine every confidence
Time threshold between breath, the first image is filtered according further to the time threshold between each position information, so as to
Improve the accuracy of destination object retrieval result.
Corresponding to above method embodiment, the embodiment of the present invention additionally provides corresponding device embodiment.
Fig. 3 retrieves device for a kind of destination object provided in an embodiment of the present invention, and described device includes：
Processing module 310, for obtaining the comparison picture for including destination object, retrieve in monitoring image and compared with described
The similarity of picture is more than the first image of predetermined threshold, and obtains positional information and temporal information corresponding to each first image；
First acquisition module 320, for obtaining known position information and temporal information pair；
Filtering module 330, for positional information and temporal information with the known position information and temporal information to not
First image of matching is filtered, the second image after being filtered.
The embodiments of the invention provide a kind of destination object to retrieve device, is examined by image matching method in monitoring image
Rope obtain and compare picture similarity more than predetermined threshold image after, can according further to known positional information and when
Between information pair, the obtained image of retrieval is filtered, filtered out with known position information and temporal information to unmatched figure
Picture.In destination object retrieving, the retrieval result for being clearly present mistake may be retrieved, such as when destination object appears in two
During individual position, it is less than in the time difference that two positions occur in actual conditions, and another is arrived in a position to destination object therefrom
The minimum time of position.In this case, when the positional information and temporal information pair that known target object occurs, you can to inspection
The result that rope obtains is filtered, and unmatched image filtering will be fallen with known position information and temporal information, so as to
Improve the accuracy of destination object retrieval result.
As a kind of embodiment of the embodiment of the present invention, first acquisition module 420 includes：
Receiving submodule (not shown), for receiving location sets and the period of user's input；
Determination sub-module (not shown), for by each position and the initial time of the period, terminate time structure
Into combination, be defined as known position information and temporal information pair.
As a kind of embodiment of the embodiment of the present invention, the filtering module 330, including：
First calculating sub module (not shown), for for each first image, calculating the time of first image
Information and each known position information and the difference of temporal information centering temporal information；
First acquisition submodule (not shown), for according to the time threshold between each position information that is locally stored,
Obtain the time threshold between the positional information of first image and each known position information and temporal information center position information；
First filter submodule (not shown), for judging whether each difference is respectively less than corresponding time threshold
Value, if it is, filtering first image.
As a kind of embodiment of the embodiment of the present invention, first acquisition module 420 includes：
First implementation sub-module (not shown), for each position for receiving user's input and corresponding each moment；And
Using each position and corresponding each moment as known position information and temporal information pair；Or
Second implementation sub-module (not shown), for showing described first image；The selection for receiving user's input refers to
Order, and using positional information and temporal information corresponding to each 3rd image chosen as known position information and temporal information pair；
Or
3rd implementation sub-module (not shown), for showing described first image, the selection for receiving user's input refers to
Order, it is determined that positional information and temporal information corresponding to each 3rd image chosen；And receive each position and correspondingly of user's input
Each moment, by positional information and temporal information corresponding to each 3rd image, and each position of user's input and correspondingly
Each moment as known position information and temporal information pair.
As a kind of embodiment of the embodiment of the present invention, second implementation sub-module, specifically for showing predetermined number
First image of amount.
As a kind of embodiment of the embodiment of the present invention, the filtering module 430 includes：
Second calculating sub module (not shown), for for each first image, calculating the time of first image
Information and each known position information and the difference of temporal information centering temporal information；
Second acquisition submodule (not shown), for according to the time threshold between each position information that is locally stored,
Obtain the time threshold between the positional information of first image and each known position information and temporal information center position information；
Second filter submodule (not shown), for judging whether each difference is all higher than corresponding time threshold, such as
Fruit is no, filters first image.
As a kind of embodiment of the embodiment of the present invention, described device also includes：
Second acquisition module (not shown), for for each two positional information, obtaining described two positional informations
Corresponding monitoring image, and temporal information corresponding to each monitoring image；
Identification module (not shown), in monitoring image corresponding to described two positional informations, identification to include
The monitoring image of same target；
Computing module (not shown), for for the monitoring image comprising same target, calculating and corresponding to one of them
The difference of the temporal information of each monitoring image of positional information and the temporal information of each monitoring image of another corresponding positional information
Value；
Determining module (not shown), for according to the difference being calculated, determining between described two positional informations
Time threshold and be stored in local.
As a kind of embodiment of the embodiment of the present invention, the determining module, specifically for by the difference most
Small value is defined as the time threshold between described two positional informations；
According to the difference, the time threshold between described two positional informations is determined using quartile method；Or
According to the difference, the time threshold between described two positional informations is determined using slope diagnostic method.
As a kind of embodiment of the embodiment of the present invention, described device also includes：
Update module (not shown), at set time intervals, periodically updating everybody that be locally stored
Time threshold between confidence breath.
As a kind of embodiment of the embodiment of the present invention, described device also includes：
Display module (not shown), for the positional information and temporal information according to each second image, electronically
The movement locus of the destination object is shown on figure.
For device embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, it is related
Part illustrates referring to the part of embodiment of the method.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply and deposited between these entities or operation
In any this actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to
Nonexcludability includes, so that process, method, article or equipment including a series of elements not only will including those
Element, but also the other element including being not expressly set out, or it is this process, method, article or equipment also to include
Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that
Other identical element also be present in process, method, article or equipment including the key element.
Each embodiment in this specification is described by the way of related, identical similar portion between each embodiment
Divide mutually referring to what each embodiment stressed is the difference with other embodiment.It is real especially for system
For applying example, because it is substantially similar to embodiment of the method, so description is fairly simple, related part is referring to embodiment of the method
Part explanation.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.It is all
Any modification, equivalent substitution and improvements made within the spirit and principles in the present invention etc., are all contained in protection scope of the present invention
It is interior.
Claims (20)
1. a kind of destination object search method, it is characterised in that methods described includes：
The comparison picture for including destination object is obtained, retrieval and the similarity for comparing picture are more than predetermined in monitoring image
First image of threshold value, and obtain positional information and temporal information corresponding to each first image；
Obtain known position information and temporal information pair；
Positional information and temporal information were carried out with the known position information and temporal information to unmatched first image
Filter, the second image after being filtered.
2. according to the method for claim 1, it is characterised in that the acquisition known position information and temporal information are to bag
Include：
Receive location sets and the period of user's input；
By each position and the initial time of the period, terminate the combination that the time forms, be defined as known position information and when
Between information pair.
3. according to the method for claim 2, it is characterised in that described to positional information and temporal information and the known bits
Confidence ceases and temporal information carries out filtering to unmatched first image and included：
For each first image, temporal information and each known position information and the temporal information centering of first image are calculated
The difference of temporal information；
According to the time threshold between each position information being locally stored, the positional information of first image and each known location are obtained
Time threshold between information and temporal information center position information；
Judge whether each difference is respectively less than corresponding time threshold, if it is, filtering first image.
4. according to the method for claim 1, it is characterised in that the acquisition known position information and temporal information are to bag
Include：
The each position of user's input and corresponding each moment are received, and using each position and corresponding each moment as known bits confidence
Breath and temporal information pair；Or
Show described first image；The selection instruction of user's input is received, and position corresponding to each 3rd image chosen is believed
Breath and temporal information are as known position information and temporal information pair；Or
Described first image is shown, receives the selection instruction of user's input, it is determined that position corresponding to each 3rd image chosen is believed
Breath and temporal information；And each position of user's input and corresponding each moment are received, by position corresponding to each 3rd image
Information and temporal information, and each position of user's input and corresponding each moment are as known position information and temporal information
It is right.
5. according to the method for claim 4, it is characterised in that the displaying described first image includes：
Show the first image of predetermined quantity.
6. according to the method for claim 4, it is characterised in that described to positional information and temporal information and the known bits
Confidence ceases and temporal information carries out filtering to unmatched first image and included：
For each first image, temporal information and each known position information and the temporal information centering of first image are calculated
The difference of temporal information；
According to the time threshold between each position information being locally stored, the positional information of first image and each known location are obtained
Time threshold between information and temporal information center position information；
Judge whether each difference is all higher than corresponding time threshold, if not, filtering first image.
7. the method according to claim 3 or 6, it is characterised in that the process of the time threshold between storage each position information
Including：
For each two positional information, monitoring image corresponding to described two positional informations is obtained, and each monitoring image corresponds to
Temporal information；
In monitoring image corresponding to described two positional informations, identification includes the monitoring image of same target；
For the monitoring image comprising same target, the temporal information of each monitoring image of the corresponding one of positional information of calculating
With the difference of the temporal information of each monitoring image of another corresponding positional information；
According to the difference being calculated, determine the time threshold between described two positional informations and be stored in local.
8. according to the method for claim 7, it is characterised in that it is described according to the difference being calculated, determine described two
Time threshold between individual positional information includes：
Minimum value in the difference is defined as the time threshold between described two positional informations；
According to the difference, the time threshold between described two positional informations is determined using quartile method；Or
According to the difference, the time threshold between described two positional informations is determined using slope diagnostic method.
9. according to the method for claim 7, it is characterised in that methods described also includes：
At set time intervals, the time threshold between each position information being locally stored is periodically updated.
10. according to the method for claim 1, it is characterised in that it is described filtered after the second image after, the side
Method also includes：
According to the positional information and temporal information of each second image, the motion rail of the destination object is shown on the electronic map
Mark.
11. a kind of destination object retrieves device, it is characterised in that described device includes：
Processing module, the comparison picture of destination object include for obtaining, retrieval and the picture that compares in monitoring image
Similarity is more than the first image of predetermined threshold, and obtains positional information and temporal information corresponding to each first image；
First acquisition module, for obtaining known position information and temporal information pair；
Filtering module, for positional information and temporal information and the known position information and temporal information to unmatched the
One image is filtered, the second image after being filtered.
12. device according to claim 11, it is characterised in that first acquisition module includes：
Receiving submodule, for receiving location sets and the period of user's input；
Determination sub-module, for the combination for forming each position and the initial time of the period, termination time, it is defined as
Know positional information and temporal information pair.
13. device according to claim 12, it is characterised in that the filtering module, including：
First calculating sub module, for for each first image, calculating the temporal information of first image and each known bits
Confidence ceases and the difference of temporal information centering temporal information；
First acquisition submodule, for according to the time threshold between each position information being locally stored, obtaining first image
Time threshold between positional information and each known position information and temporal information center position information；
First filter submodule, for judge each difference whether be respectively less than corresponding to time threshold, if it is, filter this
One image.
14. device according to claim 11, it is characterised in that first acquisition module, including：
First implementation sub-module, for each position for receiving user's input and corresponding each moment；And by each position and corresponding
Each moment is as known position information and temporal information pair；Or
Second implementation sub-module, for showing described first image；Receive the selection instruction of user's input, and will choose each the
Positional information and temporal information are as known position information and temporal information pair corresponding to three images；Or
3rd implementation sub-module, for showing described first image, the selection instruction of user's input is received, it is determined that choose each the
Positional information and temporal information corresponding to three images；And each position of user's input and corresponding each moment are received, will be described each
Positional information and temporal information corresponding to 3rd image, and each position of user's input and corresponding each moment are as known bits
Confidence ceases and temporal information pair.
15. device according to claim 14, it is characterised in that second implementation sub-module, it is pre- specifically for showing
First image of fixed number amount.
16. device according to claim 14, it is characterised in that the filtering module includes：
Second calculating sub module, for for each first image, calculating the temporal information of first image and each known bits
Confidence ceases and the difference of temporal information centering temporal information；
Second acquisition submodule, for according to the time threshold between each position information being locally stored, obtaining first image
Time threshold between positional information and each known position information and temporal information center position information；
Second filter submodule, for judging whether each difference is all higher than corresponding time threshold, if not, filtering first figure
Picture.
17. the device according to claim 13 or 16, it is characterised in that described device also includes：
Second acquisition module, for for each two positional information, obtaining monitoring image corresponding to described two positional informations, with
And temporal information corresponding to each monitoring image；
Identification module, in monitoring image corresponding to described two positional informations, identification to include the monitoring figure of same target
Picture；
Computing module, for for the monitoring image comprising same target, calculating each monitoring of corresponding one of positional information
The difference of the temporal information of image and the temporal information of each monitoring image of another corresponding positional information；
Determining module, for according to the difference being calculated, determining the time threshold between described two positional informations and storage
In local.
18. device according to claim 17, it is characterised in that the determining module, specifically for by the difference
Minimum value be defined as time threshold between described two positional informations；
According to the difference, the time threshold between described two positional informations is determined using quartile method；Or
According to the difference, the time threshold between described two positional informations is determined using slope diagnostic method.
19. device according to claim 17, it is characterised in that described device also includes：
Update module, at set time intervals, periodically updating the time threshold between each position information being locally stored
Value.
20. device according to claim 11, it is characterised in that described device also includes：
Display module, for the positional information and temporal information according to each second image, the target is shown on the electronic map
The movement locus of object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610734166.5A CN107783995B (en) | 2016-08-26 | 2016-08-26 | Target object retrieval method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610734166.5A CN107783995B (en) | 2016-08-26 | 2016-08-26 | Target object retrieval method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107783995A true CN107783995A (en) | 2018-03-09 |
| CN107783995B CN107783995B (en) | 2020-09-04 |
Family
ID=61439771
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610734166.5A Active CN107783995B (en) | 2016-08-26 | 2016-08-26 | Target object retrieval method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107783995B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110119456A (en) * | 2019-05-14 | 2019-08-13 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
| CN110134810A (en) * | 2019-05-14 | 2019-08-16 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
| CN110309341A (en) * | 2019-07-09 | 2019-10-08 | 重庆紫光华山智安科技有限公司 | Image search method, device, equipment and storage medium |
| CN110309878A (en) * | 2019-06-28 | 2019-10-08 | 深圳市商汤科技有限公司 | A kind of information processing method and device, storage medium |
| CN110442742A (en) * | 2019-07-31 | 2019-11-12 | 深圳市商汤科技有限公司 | Retrieve method and device, processor, electronic equipment and the storage medium of image |
| CN111881322A (en) * | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Target searching method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102843547A (en) * | 2012-08-01 | 2012-12-26 | 安科智慧城市技术（中国）有限公司 | Intelligent tracking method and system for suspected target |
| CN104239445A (en) * | 2014-09-01 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Method and device for representing search results |
| CN105335387A (en) * | 2014-07-04 | 2016-02-17 | 杭州海康威视系统技术有限公司 | Retrieval method for video cloud storage system |
| US20160154882A1 (en) * | 2012-04-23 | 2016-06-02 | Sri International | Classification, search and retrieval of complex video events |
| CN105788271A (en) * | 2016-05-17 | 2016-07-20 | 厦门市美亚柏科信息股份有限公司 | Method and apparatus for identifying target moving object through trajectory matching |
-
2016
- 2016-08-26 CN CN201610734166.5A patent/CN107783995B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160154882A1 (en) * | 2012-04-23 | 2016-06-02 | Sri International | Classification, search and retrieval of complex video events |
| CN102843547A (en) * | 2012-08-01 | 2012-12-26 | 安科智慧城市技术（中国）有限公司 | Intelligent tracking method and system for suspected target |
| CN105335387A (en) * | 2014-07-04 | 2016-02-17 | 杭州海康威视系统技术有限公司 | Retrieval method for video cloud storage system |
| CN104239445A (en) * | 2014-09-01 | 2014-12-24 | 百度在线网络技术（北京）有限公司 | Method and device for representing search results |
| CN105788271A (en) * | 2016-05-17 | 2016-07-20 | 厦门市美亚柏科信息股份有限公司 | Method and apparatus for identifying target moving object through trajectory matching |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110119456A (en) * | 2019-05-14 | 2019-08-13 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
| CN110134810A (en) * | 2019-05-14 | 2019-08-16 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
| CN110309878A (en) * | 2019-06-28 | 2019-10-08 | 深圳市商汤科技有限公司 | A kind of information processing method and device, storage medium |
| CN110309878B (en) * | 2019-06-28 | 2022-02-22 | 深圳市商汤科技有限公司 | Information processing method and device and storage medium |
| CN110309341A (en) * | 2019-07-09 | 2019-10-08 | 重庆紫光华山智安科技有限公司 | Image search method, device, equipment and storage medium |
| CN110309341B (en) * | 2019-07-09 | 2021-04-16 | 重庆紫光华山智安科技有限公司 | Image retrieval method, device, equipment and storage medium |
| CN110442742A (en) * | 2019-07-31 | 2019-11-12 | 深圳市商汤科技有限公司 | Retrieve method and device, processor, electronic equipment and the storage medium of image |
| CN111881322A (en) * | 2020-09-28 | 2020-11-03 | 成都睿沿科技有限公司 | Target searching method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107783995B (en) | 2020-09-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107783995A (en) | A kind of destination object search method and device | |
| CN107370983B (en) | method and device for acquiring track of video monitoring system | |
| CN101281593B (en) | Method and system for researching intelligent video monitoring case | |
| CN105208528B (en) | A kind of system and method for identifying with administrative staff | |
| WO2016107463A1 (en) | Transaction risk detection method and device | |
| CN105791022A (en) | Congestion degree detection and early-warning system | |
| CN109447150A (en) | A kind of plants ' aesthetics method, apparatus, electronic equipment and storage medium | |
| CN111222373B (en) | Personnel behavior analysis method and device and electronic equipment | |
| CN114862946B (en) | Location prediction method, system, device, and medium | |
| CN107543547A (en) | A kind of air navigation aid, apparatus and system | |
| CN108566620A (en) | A kind of indoor orientation method based on WIFI | |
| CN105630842A (en) | Method and device for establishing air quality information database and determining pollution information | |
| CN111383246A (en) | Scroll detection method, device and equipment | |
| CN108572974A (en) | A kind of vehicle pictures search method and device | |
| CN113190700B (en) | Face snapshot, screening and storage method and system for real estate transaction | |
| CN111259789A (en) | Face recognition intelligent security monitoring management method and system | |
| CN112288050B (en) | Abnormal behavior identification method and device, terminal equipment and storage medium | |
| CN108540424A (en) | Monitoring system, monitoring device and monitoring method | |
| CN116912517A (en) | Method and device for detecting camera view field boundary | |
| CN110264722A (en) | The screening technique and system of warping apparatus in information collecting device | |
| CN116028660A (en) | Weight value-based image data screening method, system and medium | |
| CN109871421A (en) | A kind of geography information model stocking system | |
| CN114863364A (en) | Security detection method and system based on intelligent video monitoring | |
| CN113961826A (en) | Camera searching method and device, intelligent terminal and computer readable storage medium | |
| CN110781797B (en) | Labeling method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |