CN110955794A - Method and device for searching associated object and electronic equipment - Google Patents
Method and device for searching associated object and electronic equipment Download PDFInfo
- Publication number
- CN110955794A CN110955794A CN201910969488.1A CN201910969488A CN110955794A CN 110955794 A CN110955794 A CN 110955794A CN 201910969488 A CN201910969488 A CN 201910969488A CN 110955794 A CN110955794 A CN 110955794A
- Authority
- CN
- China
- Prior art keywords
- record
- archive
- file
- target object
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012163 sequencing technique Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 9
- 238000003064 k means clustering Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 16
- 230000001174 ascending effect Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 238000007621 cluster analysis Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the disclosure provides a method and a device for searching a related object and electronic equipment, wherein the method comprises the following steps: determining a first archive record of a target object from an archive; determining at least one second archive record associated with the first archive record from the archive; determining a related object of the target object based on the object characteristics in all the second file records; and determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object. The technical scheme is used for solving the technical problems that the existing face control technology is only limited to searching a target person and cannot further determine related personnel frequently contacting with the target person.
Description
Technical Field
The invention relates to the technical field of face recognition, in particular to a method and a device for searching a related object and electronic equipment.
Background
Face recognition technology has been applied to various fields such as security, face recognition access control systems, airport/high-speed rail entry gates, and the like. Face searching is a basic technology in face recognition technology, and is similar to picture searching, and a face picture of a face to be recognized needs to be found out from a massive face picture database.
The existing human face control system utilizes a human face searching technology, shoots a human face through a camera in a control area, and then compares the shot human face picture with a human face picture (namely a target figure to be searched) stored in a database, thereby determining whether the shot human face is the human face of the target figure. In practical application, the face control system has a high requirement on the real-time performance of face search, that is, after a camera captures a face picture, whether the face picture is the most similar or not needs to be searched in a database in real time, so as to judge whether the captured face is the target figure to be searched. With the continuous improvement of the algorithm, the timeliness of the face control technology is gradually improved.
However, the existing face control technology is limited to searching for a target person, and if the related persons frequently contacting with the target person need to be further determined, no targeted solution is provided at present.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for searching for a related object, and an electronic device, so as to solve the technical problem that the existing face control technology is limited to search for a target person, and cannot further determine related people who often contact with the target person.
According to an aspect of the present disclosure, there is provided a method for searching an associated object, the method including: determining a first archive record of a target object from an archive; determining at least one second archive record associated with the first archive record from the archive; determining a related object of the target object based on the object characteristics in all the second file records; and determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
According to another aspect of the present disclosure, there is provided a finding apparatus of an associated object, including: the first file record query module is used for determining a first file record of the target object from the archive; a second archive record query module for determining at least one second archive record associated with the first archive record from the archive; the related object searching module is used for determining related objects related to the target object based on the object characteristics in all the second file records; and the associated object characteristic determining module is used for determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
According to yet another aspect of the present disclosure, there is provided an electronic device including: a processor; a memory for storing the processor-executable instructions; the processor is used for reading the executable instruction from the memory and executing the instruction to realize the searching method of the related object.
According to yet another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing a computer program for executing the above-described method of finding an associated object.
Compared with the prior art, the technical scheme at least has the following beneficial effects:
according to the searching method of the related object provided by the embodiment of the disclosure, the related search is performed according to the first file record of the target object, and the related object of the target object is determined according to the object characteristics in the second file record by searching at least one second file record related to the first file record information. And then, determining the object characteristics of the associated object according to the second file record corresponding to the determined associated object, and finally outputting the object characteristics of the target object and the object characteristics of the associated object of the target object. Therefore, the technical problems that in the prior art, only the target person is searched, and related persons frequently contacting with the target person cannot be further determined are solved.
Further, in the process of determining the associated object of the target object, clustering the object features in the second file record by using a clustering algorithm to determine a clustering center to which all the object features belong, and further sequencing the clustering centers according to the number of the second file records corresponding to the clustering center and the number of the first type of record information in the second file records, thereby determining the associated object of the target object according to the sequencing result. The clustering algorithm is adopted to cluster the object characteristics, and the clustering centers are further sequenced, so that the determined associated objects of the target objects are more accurate.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings needed to be used in the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present disclosure, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1A is a schematic flowchart of a method for searching an associated object according to an exemplary embodiment of the present disclosure.
Fig. 1B is a schematic diagram of an archive record of an application scenario of a method for searching an associated object according to an exemplary embodiment of the present disclosure.
FIG. 2 is a flowchart illustrating a method for searching for an associated object according to another exemplary embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a device for finding an associated object according to an exemplary embodiment of the present disclosure;
fig. 4 is a schematic structural diagram of a device for finding an associated object according to another exemplary embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of a device for finding an associated object according to yet another exemplary embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of a device for finding an associated object according to yet another exemplary embodiment of the present disclosure.
Fig. 7 is a block diagram of an electronic device of an exemplary embodiment of the present disclosure.
Detailed Description
Hereinafter, example embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of the embodiments of the present disclosure and not all embodiments of the present disclosure, with the understanding that the present disclosure is not limited to the example embodiments described herein.
Summary of the application
According to the searching method of the related object provided by the technical scheme, the related search is carried out according to the first file record of the target object, at least one second file record related to the first file record information is searched, and the related object of the target object is determined according to the object characteristics in the second file record. And then, determining the object characteristics of the associated object according to the second file record corresponding to the determined associated object, and finally outputting the object characteristics of the target object and the object characteristics of the associated object of the target object. Therefore, the technical problems that in the prior art, only the target person is searched, and related persons frequently contacting with the target person cannot be further determined are solved.
The target object in the technical solution may be a person, a vehicle, or other movable object, and is determined according to different application scenarios. For example, if the application scenario is to find a target person and an associated person closely related to the target person, the target object is a person. For another example, if the application scenario is to search for a target vehicle and an associated vehicle in close relationship with the target vehicle, the target object is a vehicle.
Exemplary method
Fig. 1A is a schematic flowchart of a method for searching an associated object according to an exemplary embodiment of the present disclosure. The embodiment can be applied to the server. As shown in fig. 1A, the method comprises the following steps:
Specifically, the archive is a database, and a plurality of archive records are included in the database, and each archive record is used for recording the acquired relevant information of an object, including the time information and the location information of the object, the image information of the object, the object features extracted from the image information, and the like. Wherein, the object can be a person, a vehicle or other movable objects according to different application scenes. For example, the archive is used for recording face information captured by one or more cameras, including time information and location information of capture, the number of the camera, a captured face image, feature information of the face image, and the like. The feature information of the face image can be extracted from the face image through a face recognition technology, and the feature information of the face image is the object feature of the object. In practical applications, the archive records in the archive can be updated in real time, that is, every time the related information of a new object is obtained, the related information of the object is added to the archive as an archive record in real time.
The target object is an object to be searched (e.g. a target person is searched) according to different specific application scenarios, and the archive records of the target object are stored in the archive. And searching the archive record (namely the first archive record) corresponding to the target object in a proper searching mode according to certain related information of the target object in the archive. For example, still taking the case that the archive records face information captured by a camera as an example, the archive of the target object is searched from the archive by means of image retrieval based on a face image of the target object.
In practical applications, since the target object may be captured by different cameras or captured by the same camera at different times, the searched archive records of the target object may include a plurality of first archive records. In addition, if a plurality of target objects exist, the first file record of each target object is determined respectively.
Specifically, as described above, each of the archive records includes various information, for example, one archive record includes time information and place information of a snapshot, a number of a camera, and the like. According to one or more information in the determined first file record, searching other file records associated with the one or more information from the archive as the second file recordRecording device。
Depending on the number of messages, the manner of searching the first archive record for the associated other archive records may also vary.
For example, if the number of information is one, the manner of searching the associated other archive records from the first archive record may be: and searching file records with the information identical to the information in the first file record from other file records as the second file record, or searching file records with the information different from the information in the first file record but meeting preset conditions between the information and the first file record from other file records as the second file record.
For another example, if the number of the information is plural, the method of searching the associated other archive records from the first archive record may be: and searching file records with the same information as the corresponding information in the first file record from other file records as the second file record, or searching file records with part of the information same as the corresponding information in the first file record and the other part of the information different but meeting preset conditions from other file records as the second file record, or searching file records with the information different from the corresponding information in the first file record and meeting preset conditions from other file records as the second file record.
In practical applications, the specific condition satisfied by the second file record associated with the first file record may be determined according to different application scenarios.
For example, the application scene is used as an associated object for determining other objects which appear at the same place and time as the target object.
Based on the application scenario, the step specifically includes determining, from the archive, a second archive record that is the same as the first archive record in terms of first type record information and satisfies a preset condition in terms of second type record information. The first type of record information is location information, the second type of record information is time information, and the preset condition may be a preset time range, that is, a difference between time in the second file record and time in the first file record satisfies the preset time range.
Further, if all the first file records related to the target object searched from the archive library include one piece of the location information, determining a second file record which is the same as the location information and the time information of which meets the preset time range from the archive library. And if all the first file records related to the target object searched from the archive library comprise a plurality of the place information, respectively determining second file records which are the same as the place information and the time information of which meets the preset time range from the archive library.
And 103, determining a related object of the target object based on the object characteristics in all the second file records.
Specifically, at least one second file record associated with the first file record determined in step 102 above may include a plurality of other objects associated with the target object, but all the other objects are not determined to be associated with the target object, and these object features need to be analyzed based on the object features in the second file record and in combination with an algorithm, so as to determine one or several other objects that are most closely related to the target object as associated objects of the target object.
In practical application, the associated object most closely related to the target object may be determined according to different application scenarios.
For example, still taking the application scenario as an example of determining another object that appears at the same location and has a time similar to that of the target object as the related object of the target object, after determining a second archive record that is the same as the first archive record in the archive and satisfies the preset condition as the second archive record, the second archive record includes archive records of a plurality of different objects, and may also include different archive records of the same object, so it is necessary to further perform cluster analysis on the object features in the second archive record by using a clustering algorithm, and use an object that appears at the same location and has a time similar to that of the target object as the related object of the target object according to the cluster analysis result. The specific process of how to perform cluster analysis on the object features in the second file record by using a clustering algorithm will be described in detail in the following embodiments.
And 104, determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
Specifically, after screening out the associated object with the target object from the object features in all the second file records according to an algorithm (e.g., a clustering algorithm), the object features of the associated object are determined from the corresponding second file records according to the determined associated object.
According to the method for searching for the associated object in the embodiment, the associated search is performed according to the first file record of the target object to obtain at least one associated second file record, the associated object of the target object is determined based on the object features in the second file record and by combining an algorithm, and the object features of the associated object are obtained, so that the associated object having a close relationship with the target object is further determined on the basis of determining the target object.
Fig. 1B is a schematic diagram of an archive record of an application scenario of a method for searching an associated object according to an exemplary embodiment of the present disclosure.
Referring to FIG. 1B, a plurality of archive records are stored in the archive (only 13 of which are shown in FIG. 1B as a schematic). Each archive record includes an object name, time information at which the object is acquired, location information, image information of the object, and object features (extracted from the image information).
The application scenario of the present embodiment is to determine an associated object that appears at the same place and at a time close to zhangsan (target object).
Specifically, after the image information of zhang san is acquired, all the archive records of zhang san (i.e., the first archive record) are searched from the archive by using an image retrieval method. Such as the archive records including number 1, number 4 and number 6 in fig. 1B.
Then, according to the location information (first type record information) and the time information (second type record information) in the first file record, a second file record which is the same as the location information in the first file record and has the time information within a preset time range (for example, 5 minutes) is searched. FIG. 1B shows the file records including number 2, number 3, number 7, number 8, number 9, number 10, number 11, and number 13. The file records with the numbers 2 and 3 are second file records which are the same as the place A in the first file record and have time information within a preset time range; the file records with the number 7, the number 8 and the number 9 are second file records which are the same as the location B in the first file record and have time information within a preset time range; the file records with the number 10, the number 11 and the number 13 are the same as the location C in the first file record, and the time information is in the second file record within the preset time range.
And further, performing cluster analysis on the object characteristics based on all the object characteristics in the second file records and by combining a clustering algorithm. The object features belonging to the same object have the same or similar object feature values, and the object feature values of different objects have larger difference, so that the object features having the same or similar object feature values are classified into one group (i.e., correspond to one cluster center), thereby determining how many cluster centers all the object features can be classified into. As shown in fig. 1B, all the object features in the second profile records may be clustered into 4 cluster centers, which correspond to lie four, wang five, liu five, and wang seven, respectively.
And then, sequencing the 4 clustering centers according to the number of times of the target object appearing in the same place to obtain a sequencing result, and selecting other objects corresponding to the clustering centers with higher times of the same place from the sequencing result as the associated objects of the target object. As shown in fig. 1B, the ranking results were that lie four and king five each appeared at the same location as the target object 3 times, and liu five and king seven each appeared at the same location as the target object 1 time. After the clustering centers are sorted according to the number of the place information (namely the first type of record information) in the second file record, the number of the place information in the second file record corresponding to the fourth item and the fifth item is the same, so that the fourth item and the fifth item can be further sorted according to the number of the second file record corresponding to the clustering centers so as to further sort the fourth item and the fifth item. Since the number of records of the second file corresponding to lie four and king five is also the same in the file shown in fig. 1B, lie four and king five are juxtaposed. In practical application, in the case of including a larger range of file records, if the numbers of the second file records corresponding to lie four and wang five are different, then lie four and wang five are further sorted according to the numbers.
Therefore, in the archive record shown in fig. 1B, lie four and wangwu are determined as the related objects of zhang three.
In other embodiments, the following steps are included after step 104: and outputting the object characteristics of the target object and the object characteristics of the associated object of the target object.
Different from the prior art, according to the method for searching the related object provided in this embodiment, after the target object is determined, the related object of the target object is further determined, so that the object feature of the target object and the object feature of the related object of the target object are finally output, and thus, while the user obtains the object feature of the target object, the user can further obtain the object feature of the related object closely related to the target object.
Fig. 2 is a flowchart illustrating a method for searching an associated object according to another exemplary embodiment of the present disclosure. On the basis of the embodiment shown in fig. 1, a clustering algorithm is used to perform clustering analysis on the object features in the second file record, and the associated object of the target object is determined according to the result of the clustering analysis.
Referring to FIG. 2, step 103 in the embodiment shown in FIG. 1 may include the following steps:
and 1031, clustering the object features in all the second file records, and determining each clustering center of the object features in the second file records.
Specifically, in the archive, the object features in the second archive record are recorded in an array form. For example, the object feature in one of the second file records is [0.2, 0.11, 0.35, -0.28, …, -0.43], the number of elements in the array is the dimension of the object feature extracted from the image of the object (for example, 128-dimensional feature points are extracted, and the number of elements in the array is 128), and each element in the array is the object feature value of each feature point.
Therefore, after the object features in all the second file records are obtained, all the object features are clustered through a clustering algorithm, so that each clustering center of the object features in the second file records is determined. Those skilled in the art understand that clustering algorithms are one of the algorithms involved in grouping data in machine learning. Within a given data set, it can be divided into several different groups by clustering algorithms. Generally, data in the same group has the same attribute or characteristic, and data in different groups have a larger attribute or characteristic difference.
Specifically, in this embodiment, the object features acquired in all the second file records are clustered, and the object features with the same or similar object feature values are classified into one group (that is, correspond to one clustering center), so as to determine how many clustering centers all the object features can be classified into.
Specifically, after the respective cluster centers are determined, the number of the second file records corresponding to each cluster center is determined respectively. That is, the object features in one or more of the second file records are classified into a cluster center, and therefore, after the cluster center is determined, the number of the second file records corresponding to the cluster center needs to be further determined.
Specifically, the step is to sort each cluster center according to the following criteria:
1) according to the number of the second file records corresponding to each clustering center;
2) and according to the number of the first type of record information in the second file record corresponding to each clustering center.
In practical application, the sorting can be performed according to the basis 1) and then according to the basis 2); or sorting according to the above-mentioned criterion 2) first and then sorting according to the above-mentioned criterion 1).
For example, taking the sorting according to the above criterion 1) first and then the sorting according to the above criterion 2) as an example, the step 1033 includes:
step 10331, sorting all the cluster centers according to the number of the second file records corresponding to each cluster center to obtain a first sorting result. The sorting mode can be ascending sorting or descending sorting.
Step 10332, determining N clustering centers meeting a preset condition from the first sorting result.
In this step, the cluster centers with a larger number of records corresponding to the second file are selected. Therefore, if the step 10331 is the first sorting result obtained by ascending sorting, the last N cluster centers in the first sorting result are selected; otherwise, if the step 10331 is to obtain the first sorting result by descending sorting, the first N cluster centers in the first sorting result are selected.
10333, if there are two or more second file records in the N clustering centers with equal numbers, further performing sorting according to the numbers of the first type of record information in the second file records corresponding to the clustering centers to obtain a second sorting result; and N is a natural number which is greater than or equal to the preset number.
Specifically, in this step, after N cluster centers meeting the preset condition in the first sorting result are selected according to the step 10332, if the number of second file records corresponding to two or more cluster centers is equal, the cluster centers are further sorted according to the number of the first type record information in the second file records corresponding to the cluster centers, so as to obtain a second sorting result. The sorting mode can be ascending sorting or descending sorting.
Step 10334, selecting a preset number of cluster centers from the second sorting result as the associated objects of the target object.
In this step, a larger number of records corresponding to the second file is selected, and the cluster center with a larger number of the first type of record information is selected under the condition that the number of the records of the second file is the same. Therefore, if the step 10333 is to obtain the second sorting result by ascending sorting, then a preset number (for example, the last 5) of cluster centers in the second sorting result are selected; on the contrary, if the second sorting result obtained in the step 10333 is sorted in a descending order, the cluster centers with the preset number (for example, the top 5) sorted in the second sorting result are selected.
For another example, taking the sorting according to the above-mentioned criterion 2) first and then the sorting according to the above-mentioned criterion 1) as an example, the step 1033 includes:
step 10335, sorting all the cluster centers according to the number of the first type of record information in the second file record corresponding to each cluster center to obtain a third sorting result. The sorting mode can be ascending sorting or descending sorting.
Step 10336, determining N cluster centers meeting a preset condition from the third sorting result.
In this step, the cluster center with a larger number of the first type of record information in the corresponding second file record is selected. Therefore, if the third sorting result obtained in the step 10335 is sorted in ascending order, the last N cluster centers in the third sorting result are selected; on the contrary, if the third sorting result obtained in the step 10335 is sorted in a descending order, the first N cluster centers in the third sorting result are selected.
10337, if the number of the first type of record information in two or more corresponding second file records in the N clustering centers is equal, further performing sorting according to the number of the second file records corresponding to the clustering centers to obtain a fourth sorting result; and N is a natural number which is greater than or equal to the preset number.
Specifically, in this step, after N cluster centers meeting the preset condition in the third sorting result are selected according to the step 10336, if the number of second file records corresponding to two or more cluster centers is equal, the cluster centers are further sorted according to the number of the first type record information in the second file records corresponding to the cluster centers, so as to obtain a fourth sorting result. The sorting mode can be ascending sorting or descending sorting.
Step 10338, selecting a preset number of cluster centers from the fourth sorting result as the associated objects of the target object.
In this step, the number of the first type of record information in the corresponding second file record is larger, and the cluster center with the larger number of the second file record is selected under the condition that the number of the first type of record information is the same. Therefore, if the step 10337 is the fourth sorting result obtained by ascending sorting, a preset number (for example, the last 5) of cluster centers in the fourth sorting result are selected; on the contrary, if the step 10337 is to obtain the fourth sorting result by descending sorting, the cluster centers with the preset number (for example, the top 5) sorted in the fourth sorting result are selected.
With continued reference to FIG. 2, step 104 in the embodiment shown in FIG. 1 may include the following steps:
Specifically, each selected clustering center corresponds to one associated object of the target object, and therefore, the object feature in the second file record corresponding to each clustering center is the object feature of the associated object.
In this embodiment, the clustering algorithm may adopt a K-means clustering algorithm or a density-based clustering algorithm. The number of clustering centers needs to be predetermined in the K-means clustering algorithm. First, the center point of each group, which is a vector having the same length as the vector of the data point corresponding to each object feature value, is randomly initialized. And classifying each data point by calculating the distance between the current data point and the central point of each group, and classifying the data point into the group corresponding to the central point with the closest distance. Based on the iterated results, the average of all data points within each class is calculated as the new center point. The above steps are repeated until the center point of each class does not change much after each iteration.
Unlike the K-means clustering algorithm, the density-based clustering algorithm does not need to determine the number of clustering centers in advance, but needs to determine the radius r and minPoints. The method specifically comprises the following steps: step 1: first the radius r and minPoints are determined. Then, starting from an arbitrary data point which is not visited, whether the number of data points contained in a circle with the data point as the center and r as the radius is larger than or equal to minPoints or not is judged, and if the number of data points is larger than or equal to minPoints, the data point is marked as centralpoint; otherwise, it is marked as noise point. And (3) repeating the step 1, if one noise point exists in a circle with a radius of a certain central point, marking the point as an edge point, and otherwise, still indicating the point as the noise point. Step 1 is repeated until all data points have been accessed.
In the embodiment, in the process of determining the associated object of the target object, the clustering algorithm is used to cluster the object features in the second file record to determine the clustering centers to which all the object features belong, and then the clustering centers are sorted according to the number of the second file records corresponding to the clustering centers and the number of the first type of record information in the second file records, so as to determine the associated object of the target object according to the sorting result.
Exemplary devices
Fig. 3 is a schematic structural diagram of a device for finding an associated object according to an exemplary embodiment of the present disclosure.
Referring to fig. 3, the finding device 3 for the associated object includes: the first archive record query module 31 is configured to determine a first archive record of the target object from the archive. A second archive record query module 32, configured to determine at least one second archive record associated with the first archive record from the archive. And the associated object searching module 33 is configured to determine an associated object associated with the target object based on the object features in all the second file records. The related object feature determination module 34 is configured to determine an object feature of the related object of the target object based on the second file record corresponding to the related object of the target object.
Specifically, a plurality of archive records are stored in the archive, and each archive record is used for recording the acquired related information of one object, including the time information and the place information of the object, the image information of the object, the object characteristics extracted from the image information and the like. The first archive record query module 31 may search an archive record (i.e., the first archive record) corresponding to the target object in a suitable search manner according to some relevant information of the target object in the archive.
In one embodiment, the second archive record query module 32 is configured to determine, from the archive, a second archive record that is identical to the first type of record information of the first archive record and satisfies a predetermined condition. The first type of record information is location information, the second type of record information is time information, and the preset condition is a preset time range.
Specifically, the second archive record query module 32 includes an archive record query unit (not shown in fig. 3) configured to search, if one of the location information is included in all first archive records related to the target object searched from the archive, a second archive record that is the same as the location information and satisfies the preset time range; if all the first file records related to the target object searched from the archive library comprise a plurality of the place information, respectively searching second file records which are the same as the place information and the time information meets the preset time range from the archive library.
Fig. 4 is a schematic structural diagram of a device for finding an associated object according to another exemplary embodiment of the present disclosure. The present embodiment is a schematic structural diagram of a specific embodiment of the related object searching module 33 and the related object feature determining module 34 based on the related object searching apparatus 3 shown in fig. 3.
Referring to the figure, the finding means 4 of the associated object comprises: a first archive record query module 41, configured to determine a first archive record of the target object from the archive. A second archive record query module 42, configured to determine at least one second archive record associated with the first archive record from the archive. And the related object searching module 43 is configured to determine, based on the object features in all the second file records, a related object related to the target object. And the associated object feature determination module 44 is configured to determine an object feature of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
Specifically, the specific implementation processes of the first file record query module 41 and the second file record query module 42 may refer to the embodiment described in fig. 3, and are not described herein again.
The associated object searching module 43 includes: and the clustering unit 431 is configured to cluster the object features in all the second file records, and determine each clustering center of the object features in the second file records. An archive record number determining unit 432, configured to determine the number of the second archive records corresponding to each cluster center. And an associated object determining unit 433, configured to select a preset number of cluster centers from all the cluster centers as associated objects of the target object according to the number of the second profile records corresponding to each cluster center and the number of the first type of record information in the corresponding second profile records.
The associated object feature determining module 44 is configured to determine, based on the object features in the second file records corresponding to the selected preset number of clustering centers, object features of associated objects of the target object.
Further, the associated object determining unit 433 includes: a first sorting unit 4331, configured to sort, according to the number of the second file records corresponding to each cluster center, all the cluster centers in a descending order according to all the cluster centers, so as to obtain a first sorting result. A first clustering center determining unit 4332, configured to determine N clustering centers meeting a preset condition from the first sorting result. A second sorting unit 4333, configured to, if two or more second file records corresponding to the N clustering centers have the same number, further sort according to the number of the first type of record information in the second file records corresponding to the clustering centers to obtain a second sorting result; and N is a natural number which is greater than or equal to the preset number. A first associated object selecting unit 4334, configured to select a preset number of cluster centers from the second sorting result as associated objects of the target object.
Fig. 5 is a schematic structural diagram of a device for finding an associated object according to yet another exemplary embodiment of the present disclosure. Different from the embodiment shown in fig. 4, this embodiment is a schematic structural diagram of another specific embodiment of the related object searching module 33 and the related object feature determining module 34 on the basis of the related object searching apparatus 3 shown in fig. 3.
In this embodiment, the searching device 5 for the associated object includes: the first archive record query module 51 is configured to determine a first archive record of the target object from the archive. A second archive record query module 52, configured to determine at least one second archive record associated with the first archive record from the archive. And the associated object searching module 53 is configured to determine an associated object associated with the target object based on the object features in all the second file records. The related object feature determining module 54 is configured to determine an object feature of the related object of the target object based on the second file record corresponding to the related object of the target object.
Specifically, the specific implementation processes of the first archive record query module 51 and the second archive record query module 52 may refer to the embodiments described in fig. 3 or fig. 4, and are not described herein again.
The related object finding module 53 includes: and the clustering unit 531 is configured to cluster the object features in all the second file records, and determine each clustering center of the object features in the second file records. An archive record number determination unit 532, configured to determine the number of the second archive records corresponding to each cluster center. The associated object determining unit 533 is configured to select a preset number of cluster centers from all the cluster centers as the associated objects of the target object according to the number of the second profile records corresponding to each cluster center and the number of the first type of record information in the corresponding second profile records.
Unlike the embodiment described in fig. 4, in this embodiment, the associated object determining unit 533 includes: a third sorting unit 5331, configured to sort all the cluster centers according to the number of the first type of record information in the second file record corresponding to each cluster center, so as to obtain a third sorting result. A second cluster center determining unit 5332, configured to determine N cluster centers meeting a preset condition from the third sorting result. A fourth sorting unit 5333, configured to, if two or more second file records in the N clustering centers have the same number of the first type of record information, further perform sorting according to the number of the second file records corresponding to the clustering centers, so as to obtain a fourth sorting result; and N is a natural number which is greater than or equal to the preset number. A second associated object selecting unit 5334, configured to select a preset number of cluster centers from the fourth sorting result as associated objects of the target object.
In this embodiment, the clustering algorithm is a K-means clustering algorithm or a density-based clustering algorithm.
Fig. 6 is a schematic structural diagram of a device for finding an associated object according to yet another exemplary embodiment of the present disclosure.
Unlike the embodiment shown in fig. 3, in this embodiment, the searching apparatus 6 for the associated object includes: the first archive record query module 61 is configured to determine a first archive record of the target object from the archive. A second archive record query module 62, configured to determine at least one second archive record associated with the first archive record from the archive. And the related object searching module 63 is configured to determine, based on the object features in all the second file records, a related object related to the target object. And the associated object feature determination module 64 is configured to determine an object feature of the associated object of the target object based on the second file record corresponding to the associated object of the target object. The finding means 6 of the associated object further comprise: an output module 65 connected to the first file record query module 61 and the associated object feature determination module 64, respectively; the output module 65 is configured to output the object feature of the target object and the object feature of the associated object of the target object.
Exemplary electronic device
Fig. 7 is a block diagram of an electronic device of an exemplary embodiment of the present disclosure.
As shown in fig. 7, the electronic device 11 includes one or more processors 111 and memory 112.
The processor 111 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 11 to perform desired functions.
In one example, the electronic device 11 may further include: an input device 113 and an output device 114, which are interconnected by a bus system and/or other form of connection mechanism (not shown).
For example, the input device 113 may be a camera (or an image acquisition module) for acquiring a target object. The input device 113 may also include, for example, a keyboard, a mouse, and the like. The output device 114 can output various information including the object feature of the target object and the object feature of the object related to the target object to the outside. The output devices 114 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, among others.
Of course, for simplicity, only some of the components of the electronic device 11 relevant to the present disclosure are shown in fig. 7, omitting components such as buses, input/output interfaces, and the like. In addition, the electronic device 11 may include any other suitable components, depending on the particular application.
Exemplary computer program product and computer-readable storage Medium
In addition to the above-described methods and apparatus, embodiments of the present disclosure may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the method of finding an associated object according to various embodiments of the present disclosure described in the "exemplary methods" section of this specification above.
The computer program product may write program code for carrying out operations for embodiments of the present disclosure in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present disclosure may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in a method of finding an associated object according to various embodiments of the present disclosure described in the "exemplary methods" section above in this specification.
The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing describes the general principles of the present disclosure in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present disclosure are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present disclosure. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the disclosure is not intended to be limited to the specific details so described.
The block diagrams of devices, apparatuses, systems referred to in this disclosure are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".
It is also noted that in the devices, apparatuses, and methods of the present disclosure, each component or step can be decomposed and/or recombined. These decompositions and/or recombinations are to be considered equivalents of the present disclosure.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the disclosure to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (12)
1. A method of finding an associated object, the method comprising:
determining a first archive record of a target object from an archive;
determining at least one second archive record associated with the first archive record from the archive;
determining a related object of the target object based on the object characteristics in all the second file records;
and determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
2. The method of claim 1, the determining at least one second archive record associated with the first archive record from the archive comprising:
and determining a second file record which is the same as the first type of record information of the first file record and meets a preset condition in the second type of record information from the archive.
3. The method of claim 2, wherein determining the associated object of the target object based on the object features in all of the second file records comprises:
clustering the object features in all the second file records, and determining each clustering center of the object features in the second file records;
determining the number of the second file records corresponding to each clustering center;
and selecting a preset number of clustering centers from all the clustering centers as the associated objects of the target object according to the number of the second file records corresponding to each clustering center and the number of the first type of record information in the corresponding second file records.
4. The method of claim 3, wherein the determining the object characteristics of the associated object of the target object based on the second profile record corresponding to the associated object of the target object comprises:
and determining the object characteristics of the associated objects of the target object based on the object characteristics in the second file records corresponding to the selected cluster centers with the preset number.
5. The method of claim 2, wherein the first type of recorded information is location information, the second type of recorded information is time information, and the predetermined condition is a predetermined time range;
the determining, from the archive, a second file record that is the same as the first type of record information of the first file record and whose second type of record information satisfies a preset condition includes:
if all first file records related to the target object searched from the archive library comprise one piece of location information, determining second file records which are the same as the location information and the time information meets the preset time range from the archive library;
and if all the first file records related to the target object searched from the archive library comprise a plurality of the place information, respectively determining second file records which are the same as the place information and the time information of which meets the preset time range from the archive library.
6. The method according to claim 3, wherein the selecting a preset number of cluster centers from all the cluster centers as the associated objects of the target object according to the number of the second file records corresponding to each cluster center and the number of the first type record information in the corresponding second file records comprises:
sequencing all the clustering centers according to the number of the second file records corresponding to each clustering center to obtain a first sequencing result;
determining N clustering centers meeting preset conditions from the first sequencing result;
if the number of two or more corresponding second file records in the N clustering centers is equal, further sorting according to the number of the first type of record information in the second file records corresponding to the clustering centers to obtain a second sorting result; wherein N is a natural number greater than or equal to the preset number;
and selecting a preset number of clustering centers from the second sequencing result as associated objects of the target object.
7. The method according to claim 3, wherein the selecting a preset number of cluster centers from all the cluster centers as the associated objects of the target object according to the number of the second file records corresponding to each cluster center and the number of the first type record information in the corresponding second file records comprises:
sequencing all the clustering centers according to the number of the first type of record information in the second file record corresponding to each clustering center to obtain a third sequencing result;
determining N clustering centers meeting preset conditions from the third sequencing result;
if the number of the first type of record information in two or more corresponding second file records in the N clustering centers is equal, further sequencing according to the number of the second file records corresponding to the clustering centers to obtain a fourth sequencing result; wherein N is a natural number greater than or equal to the preset number;
and selecting a preset number of clustering centers from the fourth sequencing result as associated objects of the target object.
8. The method of any one of claims 3 to 7, wherein the clustering algorithm is a K-means clustering algorithm or a density-based clustering algorithm.
9. The method of claim 1, wherein determining the object characteristics of the associated object of the target object based on the second profile record corresponding to the associated object of the target object further comprises:
and outputting the object characteristics of the target object and the object characteristics of the associated object of the target object.
10. A lookup apparatus for an associated object, comprising:
the first file record query module is used for determining a first file record of the target object from the archive;
a second archive record query module for determining at least one second archive record associated with the first archive record from the archive;
the related object searching module is used for determining related objects related to the target object based on the object characteristics in all the second file records;
and the associated object characteristic determining module is used for determining the object characteristics of the associated object of the target object based on the second file record corresponding to the associated object of the target object.
11. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method for finding the associated object according to any one of claims 1 to 9.
12. A non-transitory computer-readable storage medium storing a computer program for executing the method of finding a related object according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910969488.1A CN110955794A (en) | 2019-10-12 | 2019-10-12 | Method and device for searching associated object and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910969488.1A CN110955794A (en) | 2019-10-12 | 2019-10-12 | Method and device for searching associated object and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110955794A true CN110955794A (en) | 2020-04-03 |
Family
ID=69975649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910969488.1A Pending CN110955794A (en) | 2019-10-12 | 2019-10-12 | Method and device for searching associated object and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110955794A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617163A (en) * | 2013-10-18 | 2014-03-05 | 中国电子科技集团公司第十研究所 | Quick target association method based on clustering analysis |
| CN106998444A (en) * | 2017-02-14 | 2017-08-01 | 北京中科天云科技有限公司 | A kind of big data face monitoring system and device |
| CN108229335A (en) * | 2017-12-12 | 2018-06-29 | 深圳市商汤科技有限公司 | It is associated with face identification method and device, electronic equipment, storage medium, program |
| CN108875548A (en) * | 2018-04-18 | 2018-11-23 | 科大讯飞股份有限公司 | Character track generation method and device, storage medium and electronic equipment |
| CN109543078A (en) * | 2018-10-18 | 2019-03-29 | 深圳云天励飞技术有限公司 | Social relationships determine method, apparatus, equipment and computer readable storage medium |
| CN109783664A (en) * | 2018-12-28 | 2019-05-21 | 上海依图网络科技有限公司 | A kind of archiving method and device |
-
2019
- 2019-10-12 CN CN201910969488.1A patent/CN110955794A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103617163A (en) * | 2013-10-18 | 2014-03-05 | 中国电子科技集团公司第十研究所 | Quick target association method based on clustering analysis |
| CN106998444A (en) * | 2017-02-14 | 2017-08-01 | 北京中科天云科技有限公司 | A kind of big data face monitoring system and device |
| CN108229335A (en) * | 2017-12-12 | 2018-06-29 | 深圳市商汤科技有限公司 | It is associated with face identification method and device, electronic equipment, storage medium, program |
| CN108875548A (en) * | 2018-04-18 | 2018-11-23 | 科大讯飞股份有限公司 | Character track generation method and device, storage medium and electronic equipment |
| CN109543078A (en) * | 2018-10-18 | 2019-03-29 | 深圳云天励飞技术有限公司 | Social relationships determine method, apparatus, equipment and computer readable storage medium |
| CN109783664A (en) * | 2018-12-28 | 2019-05-21 | 上海依图网络科技有限公司 | A kind of archiving method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106446816B (en) | Face recognition method and device | |
| EP3709184B1 (en) | Sample set processing method and apparatus, and sample querying method and apparatus | |
| CN110175549B (en) | Face image processing method, device, equipment and storage medium | |
| US11074434B2 (en) | Detection of near-duplicate images in profiles for detection of fake-profile accounts | |
| JP2017033547A (en) | Information processing apparatus, control method therefor, and program | |
| CN110235138A (en) | System and method for appearance search | |
| WO2018171412A1 (en) | Pedestrian search method and apparatus | |
| US20130121535A1 (en) | Detection device and method for transition area in space | |
| KR102028930B1 (en) | method of providing categorized video processing for moving objects based on AI learning using moving information of objects | |
| CN108229358B (en) | Index establishing method and device, electronic equipment and computer storage medium | |
| CN110825894A (en) | Data index establishing method, data index retrieving method, data index establishing device, data index retrieving device, data index establishing equipment and storage medium | |
| CN113821657A (en) | Artificial intelligence-based image processing model training method and image processing method | |
| Hernández et al. | Video analysis system using deep learning algorithms | |
| CN113963303A (en) | Image processing method, video recognition method, device, equipment and storage medium | |
| CN106575353A (en) | Hash-based media search | |
| CN113987243A (en) | Image file gathering method, image file gathering device and computer readable storage medium | |
| JP2022043631A (en) | Information processing apparatus, information processing method, and program | |
| CN115062186A (en) | Video content retrieval method, device, equipment and storage medium | |
| CN109241316B (en) | Image retrieval method, image retrieval device, electronic equipment and storage medium | |
| CN114780606A (en) | Big data mining method and system | |
| CN112115281A (en) | Data retrieval method, device and storage medium | |
| CN112241470A (en) | Video classification method and system | |
| EP3828712A1 (en) | Data parsing method and device | |
| CN116071569A (en) | Image selection method, computer equipment and storage device | |
| CN110955794A (en) | Method and device for searching associated object 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200403 |
|
| RJ01 | Rejection of invention patent application after publication |