CN114549885A - Target object clustering method and device and electronic equipment - Google Patents

Target object clustering method and device and electronic equipment Download PDF

Info

Publication number
CN114549885A
CN114549885A CN202210186853.3A CN202210186853A CN114549885A CN 114549885 A CN114549885 A CN 114549885A CN 202210186853 A CN202210186853 A CN 202210186853A CN 114549885 A CN114549885 A CN 114549885A
Authority
CN
China
Prior art keywords
target object
acquisition
designated
integrals
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210186853.3A
Other languages
Chinese (zh)
Inventor
王凯垚
陈立力
周明伟
何林强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Dahua Technology Co Ltd
Original Assignee
Zhejiang Dahua Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhejiang Dahua Technology Co Ltd filed Critical Zhejiang Dahua Technology Co Ltd
Priority to CN202210186853.3A priority Critical patent/CN114549885A/en
Publication of CN114549885A publication Critical patent/CN114549885A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The application relates to a target object clustering method and device and electronic equipment. The method comprises the following steps: acquiring target object picture frame numbers respectively corresponding to a plurality of target object files; respectively calculating a first target object acquisition integral corresponding to each target object file in a specified time period according to each target object picture frame number; screening at least one designated object acquisition integral from the first target object acquisition integrals, and determining designated object files corresponding to the designated object acquisition integrals respectively; and adjusting the clustering threshold value corresponding to each designated object archive, and clustering the target object pictures corresponding to each designated object archive according to the adjusted clustering threshold value. By the method, the designated object archives corresponding to the target objects frequently appearing in the target area are determined, the clustering threshold corresponding to the designated object archives is adjusted, and the recall rate corresponding to each target object archive is improved in the target object clustering process.

Description

Target object clustering method and device and electronic equipment
Technical Field
The present application relates to the field of image processing target object technologies, and in particular, to a target object clustering method, an apparatus, and an electronic device.
Background
With the rapid development of the society, clustering target objects becomes one of key means for target locking, the target object pictures are identified by utilizing a deep learning technology based on the collected target object pictures of the cameras in the area within a period of time, target object features in the target object pictures are extracted, similar target object pictures are clustered to form a target object archive corresponding to each target object, and the target objects are locked based on track information of the target objects in the target object archives.
At the present stage, most of the above processes are based on a deep learning technique, target object features in target object pictures are extracted, and then target object clustering is performed on the target object pictures with similarity values larger than a clustering threshold according to the target object features, but due to reasons such as acquisition angles of the target object pictures and picture definition, the situation that the similarity of the target object pictures corresponding to the same target object is slightly lower than the clustering threshold occurs, so that the target object pictures are not clustered, and if the clustering threshold is reduced according to a uniform standard for the target object pictures corresponding to all the target objects, a target object picture clustering error phenomenon is caused, so that the target object clustering accuracy is not high.
Disclosure of Invention
The application provides a target object clustering method, a target object clustering device and electronic equipment, wherein designated object archives are determined in all target object archives corresponding to a target area, and clustering threshold values corresponding to the designated object archives are adjusted, so that the recall rate corresponding to all the target object archives is improved in the target object clustering process, and further the target object clustering accuracy is improved.
In a first aspect, the present application provides a target object clustering method, where the method includes:
acquiring target object picture frame numbers respectively corresponding to a plurality of target object archives, wherein target object pictures in the target object archives are pictures collected in a target area and in a preset time period;
respectively calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number, wherein the specified time period comprises a plurality of preset time periods;
screening at least one designated object acquisition integral from the first target object acquisition integrals, and determining designated object files corresponding to the designated object acquisition integrals respectively;
and adjusting the clustering threshold value corresponding to each designated object archive, and clustering the target object pictures corresponding to each designated object archive according to the adjusted clustering threshold value.
By the method, the designated object archives in which the target objects frequently appear are determined in the target object archives corresponding to the target area, and the clustering threshold corresponding to the designated object archives is adjusted, so that the recall rate corresponding to each target object archive in the target object clustering process is improved, and the target object clustering accuracy is improved.
In a possible design, the obtaining target object picture frame numbers respectively corresponding to a plurality of target object archives includes:
acquiring a plurality of target object files corresponding to the target object pictures acquired in the target area and the designated time period;
respectively extracting corresponding first track information of the target object in each target object file within each preset time period;
and acquiring the target object picture frame number respectively corresponding to each target object file according to each first track information.
By the method, the target object picture frame number in each target object file in the target area and the designated time period is obtained, and the target object picture frame number is used for determining the target object files of the target object frequently appearing in the target area.
In a possible design, the separately extracting first trajectory information corresponding to the target object in each target object archive within each preset time period includes:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to each target object file is obtained;
converting the longitude and latitude information of the acquisition points corresponding to each second track information according to a preset code to obtain third track information;
and performing duplicate removal processing on the target object picture in each piece of third track information to obtain first track information.
The first track information obtained by the method eliminates the influence of the repeatedly acquired target object picture, is beneficial to improving the accuracy of subsequently determining the frame number of the target object picture in the target object archive, and further improves the recall rate of each target object archive in the target object clustering process.
In one possible design, calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number respectively includes:
respectively calculating a plurality of second target object acquisition integrals corresponding to each target object file according to the number of frames of each target object picture, wherein the same target object file comprises pictures acquired in a plurality of preset time periods, and one preset time period corresponds to one second target object acquisition integral;
respectively setting a weight value for each second target object acquisition integral corresponding to the same target object file to obtain third target object acquisition integrals corresponding to a plurality of preset time periods;
and summing a plurality of third target object acquisition integrals corresponding to the same target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period.
By the method, the target object acquisition integral corresponding to each target object file can be calculated, and the target object acquisition integral is used for screening out the target object files corresponding to the target objects frequently appearing in the target area.
In a possible design, the screening at least one designated object collection integral from each first target object collection integral, and determining a designated object file corresponding to each designated object collection integral respectively includes:
sorting the collection integrals of the first target objects from large to small;
taking the first M first target object acquisition integrals in the sequenced first target object acquisition integrals as designated object acquisition integrals;
and determining the designated object files corresponding to the acquisition integrals of the designated objects in the target object files.
By the method, the target object archives corresponding to the target objects frequently appearing in the target area are determined in each target object archive.
In a possible design, the adjusting the clustering threshold corresponding to each of the designated object profiles includes:
determining attenuation values respectively corresponding to the appointed object files according to the arrangement sequence of the first target object acquisition integrals corresponding to the appointed object files;
and adjusting the clustering threshold corresponding to the designated object file according to the attenuation value.
By the method, the clustering threshold corresponding to the designated object archive is adjusted, so that the recall rate of each target object archive in the target object clustering process corresponding to the target area is improved.
In a second aspect, the present application provides a target object clustering apparatus, the apparatus comprising:
the acquisition module is used for acquiring target object picture frame numbers respectively corresponding to a plurality of target object files, wherein the target object pictures in the target object files are pictures acquired in a target area and in a preset time period;
the calculation module is used for respectively calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number, wherein the specified time period comprises a plurality of preset time periods;
the screening module is used for screening at least one designated object acquisition integral from the first target object acquisition integrals and determining designated object files corresponding to the designated object acquisition integrals;
and the clustering module is used for adjusting the clustering threshold value corresponding to each designated object archive respectively and clustering the target object pictures corresponding to each designated object archive respectively according to the adjusted clustering threshold value.
In one possible design, the obtaining module is specifically configured to:
acquiring a plurality of target object files corresponding to the target object pictures acquired in the target area and the designated time period;
respectively extracting corresponding first track information of the target object in each target object file within each preset time period;
and acquiring the target object picture frame number respectively corresponding to each target object file according to each first track information.
In one possible design, the obtaining module is further configured to:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to each target object file is obtained;
converting the longitude and latitude information of the acquisition points corresponding to each second track information according to a preset code to obtain third track information;
and performing deduplication processing on the target object pictures in each piece of third track information to obtain first track information.
In one possible design, the calculation module is specifically configured to:
respectively calculating a plurality of second target object acquisition integrals corresponding to each target object file according to the number of frames of each target object picture, wherein the same target object file comprises pictures acquired in a plurality of preset time periods, and one preset time period corresponds to one second target object acquisition integral;
respectively setting a weight value for each second target object acquisition integral corresponding to the same target object file to obtain third target object acquisition integrals corresponding to a plurality of preset time periods;
and summing a plurality of third target object acquisition integrals corresponding to the same target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period.
In one possible design, the screening module is specifically configured to:
sorting the collection integrals of the first target objects from large to small;
taking the first M first target object acquisition integrals in the sequenced first target object acquisition integrals as designated object acquisition integrals;
and determining the designated object files corresponding to the acquisition integrals of the designated objects in the target object files.
In one possible design, the clustering module is specifically configured to:
determining attenuation values respectively corresponding to the appointed object files according to the arrangement sequence of the first target object acquisition integrals corresponding to the appointed object files;
and adjusting the clustering threshold corresponding to the designated object file according to the attenuation value.
In a third aspect, the present application provides an electronic device, comprising:
a memory for storing a computer program;
and the processor is used for realizing the steps of the target object clustering method when executing the computer program stored in the memory.
In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored therein, which when executed by a processor, implements the above-mentioned target object clustering method steps.
Based on the target object clustering method, the designated object archives in which the target objects frequently appear are determined from the target object archives corresponding to the target area, and the clustering threshold corresponding to the designated object archives is adjusted, so that the recall rate corresponding to each target object archive in the target object clustering process is improved, and the target object clustering accuracy corresponding to the target area is improved.
For each of the second to fifth aspects and possible technical effects of each aspect, reference is made to the above description of the possible technical effects of the first aspect or various possible solutions of the first aspect, and repeated descriptions are omitted here.
Drawings
Fig. 1 is a flowchart of a target object clustering method provided in the present application;
fig. 2 is a schematic structural diagram of a target object clustering apparatus provided in the present application;
fig. 3 is a schematic structural diagram of an electronic device provided in the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more clear, the present application will be further described in detail with reference to the accompanying drawings. The particular methods of operation in the method embodiments may also be applied to apparatus embodiments or system embodiments. It should be noted that "a plurality" is understood as "at least two" in the description of the present application. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. A is connected with B and can represent: a and B are directly connected and A and B are connected through C. In addition, in the description of the present application, the terms "first," "second," and the like are used for descriptive purposes only and are not intended to indicate or imply relative importance nor order to be construed.
The embodiments of the present application will be described in detail below with reference to the accompanying drawings.
Most of the existing target object clustering technologies are based on a deep learning technology, target object features in target object pictures are extracted, then target object clustering is performed on the target object pictures with similarity values larger than a clustering threshold value according to the target object features, but due to capture angles under different conditions, picture definition and other reasons, the similarity of the target object pictures corresponding to the same target object is sometimes slightly lower than the clustering threshold value, so that clustering is missed, and if the clustering threshold value is reduced for the target object pictures corresponding to all the target objects according to a uniform standard, a clustering error phenomenon is caused, so that clustering accuracy is not high.
In order to solve the above problems, the present application provides a target object clustering method, in which an assigned object archive in which a target object frequently appears is determined from each target object archive corresponding to a target area, and a clustering threshold corresponding to the assigned object archive is adjusted, which is helpful for improving a recall rate corresponding to each target object archive in a target object clustering process, thereby improving target object clustering accuracy corresponding to the target area. The method and the device in the embodiment of the application are based on the same technical concept, and because the principles of the problems solved by the method and the device are similar, the device and the embodiment of the method can be mutually referred, and repeated parts are not repeated.
As shown in fig. 1, a flowchart of a target object clustering method provided in the present application specifically includes the following steps:
s11, acquiring target object picture frame numbers corresponding to the plurality of target object files respectively;
in this embodiment of the application, before target object clustering is performed, a plurality of target object archives corresponding to target object pictures acquired within a target area specified time period are acquired, where a target object may be a person or a vehicle, and here, no specific limitation is made, a target area may be a geohash block, the specified time period includes a plurality of preset time periods, and if the specified time period is the previous 30 days, the preset time period may be set to be yesterday, previous day, or the tth day of the previous 30 days, and t is an integer greater than or equal to 1 and less than or equal to 30.
Further, the first trajectory information corresponding to the target object in each target object archive in each preset time period is respectively extracted, and the specific extraction method may be:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to the target object files is obtainedAnd the acquisition point can be a target object picture acquisition bayonet. If the preset time period is one day, the target object file DiThe second track information of the target object in (1) is D corresponding to j days before the current dateijThen D isijThe specific formula of (A) is as follows:
Figure BDA0003523887520000081
in the formula (1), CiAnd acquiring a corresponding bayonet for the ith target object picture, wherein i is 1, …, n.
Next, converting the latitude and longitude information of the acquisition point corresponding to each second track information according to a preset code to obtain third track information, where the preset code may be a geohash code, such as geohash7, and using the geohash code to convert the second track information D into the third track information DijConverting to obtain D 'third track information'ij，D′ijThe specific formula of (A) is as follows:
Figure BDA0003523887520000082
in the formula (2), GiIs a bayonet CiThe corresponding geohash code.
Further, the target object picture in each third track information is subjected to duplication elimination processing to obtain the first track information. For example, when the third track information D'ijIn the presence of Ci＝Ci+1When it is, it is considered that the lock is at the bayonet CiThe target object picture collection is repeated, so that the target object picture collected at the (i + 1) th time is deleted, and the file track D 'is subjected'ijThe duplicate removal is performed to obtain new track information D ″ijThe specific formula of (2) is as follows:
Figure BDA0003523887520000083
by the method, the first track information corresponding to the target object in each target object file in each preset time period can be obtained, and further, the target object picture frame number corresponding to each target object file is obtained according to each first track information, wherein the target object picture in the target object file is a picture collected in a target area and in a preset time period, for example, the target object picture frame number collected every day in a geohash block range.
S12, respectively calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number;
in the embodiment of the application, after target object picture frame numbers acquired in a target area and within a specified time period in target object files are acquired, a plurality of second target object acquisition integrals corresponding to each target object file are respectively calculated according to each target object picture frame number, wherein the same target object file comprises a plurality of pictures acquired in preset time periods, one preset time period corresponds to one second target object acquisition integral, and the target object acquisition integral indicates the frequency of target objects in the target object files appearing in the target area. The formula for calculating the second target acquisition integral is as follows:
Figure BDA0003523887520000091
in formula (4), a is the number of target object picture frames collected in the target area and in the jth preset time period in the target object archive, SjAnd collecting an integral for a second target object corresponding to the jth preset time period. In the embodiment of the present application, if the target object file DiOn day j at the geohash block GkIf a target object pictures are collected, the formula (4) can represent the archive DiDay j in the geohash block GkThe corresponding second target object collects the integral.
Further, the weight values are respectively set for the acquisition integrals of the second target objects corresponding to the same target object file, and the first target objects corresponding to a plurality of preset time periods are obtainedCollecting integrals by three target objects, wherein the setting of the weight value is related to a preset time period corresponding to the collection fraction of the second target object, if the preset time period is closer to the current time, the set weight value is larger, and the second target object collects the integrals SjThe corresponding weight value may be set to (1-j/N), then SjThe corresponding third target object acquisition integral is (1-j/N)j。
And further, summing a plurality of third target object acquisition integrals corresponding to each target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period. The calculation formula corresponding to the first target object acquisition integral is as follows:
Figure BDA0003523887520000101
in the formula (4), SallRepresenting a second target object acquisition integral SjAnd collecting integral of the corresponding first target object, wherein N is the number of preset time periods contained in the designated time period, and the value of N is an integer greater than 1.
By the method, the first target object acquisition integral corresponding to each target object file in the appointed time period can be acquired, in the calculation process, the weight value is set according to the relation between the preset time period and the current time, and the second target object acquisition integral ratio corresponding to the preset time period which is closer to the current time is increased.
S13, screening at least one designated object acquisition integral from the first target object acquisition integrals, and determining designated object files corresponding to the designated object acquisition integrals respectively;
in the embodiment of the application, after the first target object acquisition scores corresponding to the target object archives are obtained, the first target object acquisition scores are sorted from large to small, and the first M first target object acquisition scores in the sorted first target object acquisition scores are used as the designated object acquisition scores, where M is an integer greater than or equal to 1. The M first target object acquisition integrals may be first y% of the name, first 2 y% of the name, or first 3 y% of the name, where a value of y is generally 5, and may also be adjusted according to an actual situation, which is not specifically limited herein. And finally, determining appointed object files corresponding to the acquisition integrals of the appointed objects in the plurality of object files, wherein the appointed object files are the object files corresponding to the object objects which frequently appear in the object area.
And S14, adjusting the clustering threshold value corresponding to each designated object file, and clustering the target object pictures corresponding to the designated object files according to the adjusted clustering threshold value.
In the embodiment of the application, after the designated object files corresponding to the target area are determined, attenuation values corresponding to the designated object files are determined according to the arrangement sequence of the first target object acquisition integrals corresponding to the designated object files, for example, if the first target object acquisition integral sequence corresponding to the target object is the first y%, the attenuation values may be determined to be 3x, where the value of x is generally 0.01, and may be adjusted according to specific situations; if the acquisition integration sequence of the first target object corresponding to the target object is 2 y%, determining the attenuation value as 2 x; if the first target object acquisition integration order corresponding to the target object is 3 y%, the attenuation value may be determined as x, i.e. the attenuation value corresponding to the first target object acquisition score which is earlier in the arrangement order is larger. After the attenuation values corresponding to the designated object archives are determined, the target object clustering threshold values corresponding to the designated object archives are respectively adjusted according to the attenuation values.
Further, target object acquisition information corresponding to the current time period of the target area is obtained, and clustering is performed by using each target object file after the target object clustering threshold is adjusted and the target object acquisition information.
For example, the target object archive with the target object clustering threshold reduced by 3x is clustered with the target object acquisition information, the target object archive with the target object clustering threshold reduced by 2x is clustered with the target object acquisition information, the target object archive with the target object clustering threshold reduced by x is clustered with the target object acquisition information, and finally the remaining target object archive is clustered with the target object acquisition information to finally obtain the target object clustering result of the current day.
In one possible embodiment, the key value pair may be generated by the target object profile corresponding to the target region after the target object clustering and the adjusted target object clustering threshold, for example, the target object profile DiIn the geohash block GkThe key value pair after the target object clustering threshold value of (2) is decreased by x is { (G)k,Di) X, and finally, according to the key value pair, determining the designated object files frequently appearing in the target area and directly acquiring the clustering threshold corresponding to the designated object files.
Based on the target object clustering method, the designated object archives in which the target objects frequently appear are determined in the target object archives corresponding to the target area, and the clustering threshold corresponding to the designated object archives is adjusted, so that the recall rate corresponding to each target object archive is improved in the target object clustering process.
Further, in order to describe the target object clustering method provided by the present application in more detail, the method provided by the present application is described in detail through a specific application scenario. Taking the clustering of vehicles as an example, firstly, files after all vehicles at a certain intersection are clustered in the previous 30 days are obtained, track information corresponding to vehicle pictures in each file every day is obtained, and then, based on longitude and latitude information corresponding to an acquisition card port of each vehicle picture in the files, the track information corresponding to the vehicle pictures is converted through a geohash code, so that new track information is obtained. And then, carrying out track deduplication on the new track information according to a set deduplication rule, and removing repeated pictures which are continuously collected to obtain the track information after deduplication.
Further, according to the track information after the weight is removed, the number of times that the vehicle in each archive in the previous 30 days is collected in each geohash block is calculated, and the collected integral corresponding to each archive in each geohash block in each day is calculated according to the collected number of times. Then, the acquired integrals of each archive in each geohash block every day are accumulated to obtain the total acquired integral of each geohash block 30 days before each archive.
Further, the total collected integrals of each archive corresponding to each geohash block in the previous 30 days are sorted from large to small, the archives of the first 5%, the first 10% and the first 15% of the total collected integrals of each geohash block in the previous 30 days are obtained respectively, the archives are used as the frequently-occurring archives corresponding to the geohash blocks, then, the vehicle clustering threshold of the archives corresponding to the geohash is reduced, and the key value pair is generated by the archives in the clustering threshold of each geohash.
Further, vehicle picture information corresponding to all the current collection checkpoints of the crossroad is obtained, longitude and latitude information of the collection checkpoints in the vehicle picture information is converted according to the geohash codes, then frequently appearing files in each geohash block are found according to key value pairs corresponding to the previous 30 days of vehicle files, and then the frequently appearing files are clustered respectively according to the sequence of the cluster threshold adjustment amount from large to small to obtain the current vehicle cluster result. And finally, updating the key value pair according to the vehicle clustering result of the current day.
Based on the vehicle clustering method, the designated archives frequently appearing in the vehicles corresponding to the designated areas are determined, and the clustering threshold corresponding to the designated object archives is adjusted, so that the recall rate corresponding to each vehicle archive is improved in the vehicle clustering process.
Based on the same inventive concept, an embodiment of the present application further provides a target object clustering device, as shown in fig. 2, which is a schematic structural diagram of the target object clustering device in the present application, and the device includes:
an obtaining module 21, configured to obtain target object picture frame numbers corresponding to multiple target object files, where target object pictures in the multiple target object files are pictures collected in a target area and within a preset time period;
the calculation module 22 is configured to calculate, according to the number of frames of each target object picture, a first target object acquisition integral corresponding to each target object archive in a specified time period, where the specified time period includes multiple preset time periods;
the screening module 23 is configured to screen at least one designated object acquisition integral from the first target object acquisition integrals, and determine designated object files corresponding to the designated object acquisition integrals;
and the clustering module 24 is configured to adjust a clustering threshold value corresponding to each designated object archive, and cluster the target object pictures corresponding to each designated object archive according to the adjusted clustering threshold value.
In one possible design, the obtaining module 21 is specifically configured to:
acquiring a plurality of target object files corresponding to the target object pictures acquired in the target area and the designated time period;
respectively extracting corresponding first track information of the target object in each target object file within each preset time period;
and acquiring the target object picture frame number respectively corresponding to each target object file according to each first track information.
In one possible design, the obtaining module 21 is further configured to:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to each target object file is obtained;
converting the longitude and latitude information of the acquisition points corresponding to each second track information according to a preset code to obtain third track information;
and performing duplicate removal processing on the target object picture in each piece of third track information to obtain first track information.
In one possible design, the calculation module 22 is specifically configured to:
respectively calculating a plurality of second target object acquisition integrals corresponding to each target object file according to the number of frames of each target object picture, wherein the same target object file comprises pictures acquired in a plurality of preset time periods, and one preset time period corresponds to one second target object acquisition integral;
respectively setting a weight value for each second target object acquisition integral corresponding to the same target object file to obtain third target object acquisition integrals corresponding to a plurality of preset time periods;
and summing a plurality of third target object acquisition integrals corresponding to the same target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period.
In one possible design, the screening module 23 is specifically configured to:
sorting the collection integrals of the first target objects from large to small;
taking the first M first target object acquisition integrals in the sequenced first target object acquisition integrals as designated object acquisition integrals;
and determining the designated object files corresponding to the acquisition integrals of the designated objects in the target object files.
In one possible design, the clustering module 24 is specifically configured to:
determining attenuation values corresponding to the appointed object files according to the arrangement sequence of the first target object acquisition integrals corresponding to the appointed object files;
and adjusting a target object clustering threshold corresponding to the specified object file according to the attenuation value.
Based on the target object clustering device, the designated object archives in which the target objects frequently appear are determined in the target object archives corresponding to the target area, and the clustering threshold corresponding to the designated object archives is adjusted, so that the recall rate corresponding to each target object archive is improved in the target object clustering process.
Based on the same inventive concept, an embodiment of the present application further provides an electronic device, where the electronic device can implement the function of the foregoing target object clustering apparatus, and with reference to fig. 3, the electronic device includes:
at least one processor 31, and a memory 32 connected to the at least one processor 31, in this embodiment, a specific connection medium between the processor 31 and the memory 32 is not limited, and fig. 3 illustrates an example where the processor 31 and the memory 32 are connected through a bus 30. The bus 30 is shown in fig. 3 by a thick line, and the connection between other components is merely illustrative and not limited thereto. The bus 30 may be divided into an address bus, a data bus, a control bus, etc., and is shown with only one thick line in fig. 3 for ease of illustration, but does not represent only one bus or type of bus. Alternatively, the processor 31 may also be referred to as a controller, without limitation to name a few.
In the embodiment of the present application, the memory 32 stores instructions executable by the at least one processor 31, and the at least one processor 31 can execute the target object clustering method discussed above by executing the instructions stored in the memory 32. A processor 31 to implement the functions of the various modules in the apparatus shown in fig. 2.
The processor 31 is a control center of the apparatus, and may connect various parts of the entire control device by using various interfaces and lines, and perform various functions of the apparatus and process data by operating or executing instructions stored in the memory 32 and calling data stored in the memory 32, thereby performing overall monitoring of the apparatus.
In one possible design, processor 31 may include one or more processing units, and processor 31 may integrate an application processor, which primarily handles operating systems, user interfaces, application programs, and the like, and a modem processor, which primarily handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 31. In some embodiments, the processor 31 and the memory 32 may be implemented on the same chip, or in some embodiments, they may be implemented separately on separate chips.
The processor 31 may be a general-purpose processor, such as a Central Processing Unit (CPU), digital signal processor, application specific integrated circuit, field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof, that implements or performs the methods, steps, and logic blocks disclosed in embodiments of the present application. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the target object clustering method disclosed in the embodiments of the present application may be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor.
Memory 32, which is a non-volatile computer-readable storage medium, may be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. The Memory 32 may include at least one type of storage medium, and may include, for example, a flash Memory, a hard disk, a multimedia card, a card-type Memory, a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Programmable Read Only Memory (PROM), a Read Only Memory (ROM), a charged Erasable Programmable Read Only Memory (EEPROM), a magnetic Memory, a magnetic disk, an optical disk, and the like. The memory 32 is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto. The memory 32 in the embodiments of the present application may also be circuitry or any other device capable of performing a storage function for storing program instructions and/or data.
By programming the processor 31, the code corresponding to the target object clustering method described in the foregoing embodiment may be solidified into the chip, so that the chip can execute the steps of the target object clustering method of the embodiment shown in fig. 1 when running. How to program the processor 31 is well known to those skilled in the art and will not be described in detail here.
Based on the same inventive concept, the present application further provides a storage medium storing computer instructions, which when executed on a computer, cause the computer to perform the target object clustering method discussed above.
In some possible embodiments, the aspects of the target object clustering method provided in the present application may also be implemented in the form of a program product, which includes program code for causing the control apparatus to perform the steps in the target object clustering method according to various exemplary embodiments of the present application described above in this specification, when the program product is run on an apparatus.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims (14)

1. A method for clustering target objects, the method comprising:
acquiring target object picture frame numbers respectively corresponding to a plurality of target object files, wherein target object pictures in the target object files are pictures acquired in a target area and in a preset time period;
respectively calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number, wherein the specified time period comprises a plurality of preset time periods;
screening at least one designated object acquisition integral from the first target object acquisition integrals, and determining designated object files corresponding to the designated object acquisition integrals respectively;
and adjusting the clustering threshold value corresponding to each designated object archive, and clustering the target object pictures corresponding to each designated object archive according to the adjusted clustering threshold value.
2. The method of claim 1, wherein the obtaining target object picture frame numbers respectively corresponding to the plurality of target object archives comprises:
acquiring a plurality of target object files corresponding to all target object pictures acquired in the target area and the designated time period;
respectively extracting corresponding first track information of the target object in each target object file within each preset time period;
and acquiring the target object picture frame number respectively corresponding to each target object file according to each first track information.
3. The method as claimed in claim 2, wherein the separately extracting the first trajectory information corresponding to the target object in each target object archive in each preset time period comprises:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to each target object file is obtained;
converting the longitude and latitude information of the acquisition points corresponding to each second track information according to a preset code to obtain third track information;
and performing duplicate removal processing on the target object picture in each piece of third track information to obtain first track information.
4. The method of claim 1, wherein calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number comprises:
respectively calculating a plurality of second target object acquisition integrals corresponding to each target object file according to the frame number of each target object picture, wherein the same target object file comprises a plurality of pictures acquired in preset time periods, and one preset time period corresponds to one second target object acquisition integral;
respectively setting a weight value for each second target object acquisition integral corresponding to the same target object file to obtain third target object acquisition integrals corresponding to a plurality of preset time periods;
and summing a plurality of third target object acquisition integrals corresponding to the same target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period.
5. The method of claim 1, wherein the screening out at least one designated object acquisition score from each first target object acquisition score and determining a designated object profile corresponding to each designated object acquisition score comprises:
sequencing the acquisition integrals of the first target objects from large to small;
taking the first M first target object acquisition integrals in the sequenced first target object acquisition integrals as designated object acquisition integrals;
and determining the designated object files corresponding to the acquisition integrals of the designated objects in the target object files.
6. The method of claim 1, wherein the adjusting the clustering threshold corresponding to each of the designated object profiles comprises:
determining attenuation values respectively corresponding to the appointed object files according to the arrangement sequence of the first target object acquisition integrals corresponding to the appointed object files;
and adjusting the clustering threshold corresponding to the designated object file according to the attenuation value.
7. An apparatus for clustering target objects, the apparatus comprising:
the acquisition module is used for acquiring target object picture frame numbers respectively corresponding to a plurality of target object files, wherein the target object pictures in the target object files are pictures acquired in a target area and in a preset time period;
the calculation module is used for respectively calculating a first target object acquisition integral corresponding to each target object archive in a specified time period according to each target object picture frame number, wherein the specified time period comprises a plurality of preset time periods;
the screening module is used for screening at least one designated object acquisition integral from the first target object acquisition integrals and determining designated object files corresponding to the designated object acquisition integrals;
and the clustering module is used for adjusting the clustering threshold value corresponding to each designated object archive respectively and clustering the target object pictures corresponding to each designated object archive respectively according to the adjusted clustering threshold value.
8. The apparatus of claim 7, wherein the acquisition module is specifically configured to:
acquiring a plurality of target object files corresponding to the target object pictures acquired in the target area and the designated time period;
respectively extracting corresponding first track information of the target object in each target object file within each preset time period;
and acquiring the target object picture frame number respectively corresponding to each target object file according to each first track information.
9. The apparatus of claim 8, wherein the acquisition module is further to:
according to the acquisition points corresponding to the target object pictures in each target object file, second track information corresponding to each target object file is obtained;
converting the longitude and latitude information of the acquisition points corresponding to each second track information according to a preset code to obtain third track information;
and performing duplicate removal processing on the target object picture in each piece of third track information to obtain first track information.
10. The apparatus of claim 7, wherein the computing module is specifically configured to:
respectively calculating a plurality of second target object acquisition integrals corresponding to each target object file according to the number of frames of each target object picture, wherein the same target object file comprises pictures acquired in a plurality of preset time periods, and one preset time period corresponds to one second target object acquisition integral;
respectively setting a weight value for each second target object acquisition integral corresponding to the same target object file to obtain third target object acquisition integrals corresponding to a plurality of preset time periods;
and summing a plurality of third target object acquisition integrals corresponding to the same target object archive to obtain a first target object acquisition integral corresponding to each target object archive in a specified time period.
11. The apparatus of claim 7, wherein the screening module is specifically configured to:
sorting the collection integrals of the first target objects from large to small;
taking the first M first target object acquisition integrals in the sequenced first target object acquisition integrals as designated object acquisition integrals;
and determining the designated object files corresponding to the acquisition integrals of the designated objects in the target object files.
12. The apparatus of claim 7, wherein the clustering module is specifically configured to:
determining attenuation values respectively corresponding to the appointed object files according to the arrangement sequence of the first target object acquisition integrals corresponding to the appointed object files;
and adjusting the clustering threshold corresponding to the designated object file according to the attenuation value.
13. An electronic device, comprising:
a memory for storing a computer program;
a processor for implementing the method steps of any one of claims 1-6 when executing the computer program stored on the memory.
14. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1-6.
CN202210186853.3A 2022-02-28 2022-02-28 Target object clustering method and device and electronic equipment Pending CN114549885A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210186853.3A CN114549885A (en) 2022-02-28 2022-02-28 Target object clustering method and device and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210186853.3A CN114549885A (en) 2022-02-28 2022-02-28 Target object clustering method and device and electronic equipment

Publications (1)

Publication Number Publication Date
CN114549885A true CN114549885A (en) 2022-05-27

Family

ID=81679504

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210186853.3A Pending CN114549885A (en) 2022-02-28 2022-02-28 Target object clustering method and device and electronic equipment

Country Status (1)

Country Link
CN (1) CN114549885A (en)

Similar Documents

Publication Publication Date Title
CN109710780B (en) Archiving method and device
CN109815370A (en) A kind of archiving method and device
EP4209959A1 (en) Target identification method and apparatus, and electronic device
CN111783743A (en) Image clustering method and device
CN115424053B (en) Small sample image recognition method, device, equipment and storage medium
CN112489063A (en) Image segmentation method, and training method and device of image segmentation model
CN113869137A (en) Event detection method and device, terminal equipment and storage medium
CN113256683B (en) Target tracking method and related equipment
CN114549885A (en) Target object clustering method and device and electronic equipment
CN116137061A (en) Training method and device for quantity statistical model, electronic equipment and storage medium
CN113656625B (en) Method and device for determining human body space domain and electronic equipment
CN115660817A (en) Classification model training method and device
CN113505188A (en) Big data-based urban map mapping updating system and method
CN114708728A (en) Method for identifying traffic peak period, electronic equipment and storage medium
CN114663691B (en) Method and device for positioning foothold and electronic equipment
CN113111689A (en) Sample mining method, device, equipment and storage medium
CN113807445B (en) File rechecking method and device, electronic device and readable storage medium
CN117349688B (en) Track clustering method, device, equipment and medium based on peak track
CN112749625B (en) Time sequence behavior detection method, time sequence behavior detection device and terminal equipment
CN114998642A (en) Image feature extraction method and device and electronic equipment
CN116311207A (en) Fuzzy license plate recognition method and device, electronic equipment and storage medium
CN116386107A (en) Image clustering method, image clustering device and computer storage medium
CN114579782A (en) Image file gathering method, device, equipment and medium
CN114463571A (en) Vehicle picture clustering method, device, equipment and storage medium
CN116577646A (en) Intelligent high-voltage switch abnormality detection 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