WO2023213095A1

WO2023213095A1 - Data archiving method and apparatus

Info

Publication number: WO2023213095A1
Application number: PCT/CN2022/143523
Authority: WO
Inventors: 石小华; 陈毅飞; 潘家平
Original assignee: 深圳云天励飞技术股份有限公司
Priority date: 2022-05-06
Filing date: 2022-12-29
Publication date: 2023-11-09
Also published as: CN115269494A

Abstract

A data archiving method and apparatus. The method comprises: determining a set of archive events to be processed, wherein the set of archive events to be processed comprises a plurality of archive events, which are archived and arranged in chronological order for the first time, and the matching degrees/degree between each of the plurality of archive events and a first archive cover event and/or a second archive cover event are/is greater than or equal to a preset matching degree (101); and archiving the plurality of archive events again, so as to obtain a first target archiving event stream and a second target archiving event stream, thereby completing an archiving operation on the plurality of archive events, wherein the first target archiving event stream is a set of archive events corresponding to a first person, and the second target archiving event stream is a set of archive events corresponding to a second person (102). In the method, a plurality of archive events are archived again, such that an archiving operation on the archive events is completed, and the method is conducive to the improvement of archiving accuracy.

Description

Data archiving method and device

Technical field

This application claims priority to the Chinese patent application filed with the China Patent Office on May 6, 2022, with application number 202210487324.7 and the invention name "Data Archiving Method and Device", the entire content of which is incorporated into this application by reference.

The present invention relates to the field of computer vision technology, and specifically to a data archiving method and device.

Background technique

Archiving technology is one of the core technologies of AI visual intelligence. By archiving a large number of captured pictures into specific real-name personnel files, the event flow of specific personnel files can be obtained. Relevant staff can conduct case business investigation based on the event flow. Most of the current archiving technologies are based on face feature comparison methods for face archiving, or combined with human event archiving. However, for highly similar face (for example: twins) archives, there are great challenges for traditional archiving. This situation is not realistic. Under such circumstances, the probability that two highly similar personnel file archiving events will be mistakenly attributed to each other is very high.

Technical solutions

The embodiment of the present invention provides a data archiving method and device, which is beneficial to improving archiving accuracy.

The first aspect of the embodiment of the present invention provides a data archiving method, including:

Determine a set of archive events to be processed, wherein the set of archive events to be processed includes a plurality of archive events, and the plurality of archive events are first archived and arranged in chronological order, and each of the archives in the plurality of archive events is The matching degree between the event and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree;

Perform secondary archiving on the multiple archive events to obtain a first target archive event stream and a second target archive event stream to complete the archiving operation for the multiple archive events, wherein the first target archive event stream is a set of file events corresponding to the first person, and the second target file event stream is a set of file events corresponding to the second person.

A second aspect of the embodiment of the present invention provides a data archiving device, including:

Determining unit, used to determine a set of archive events to be processed, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are obtained by first filing in chronological order. Among the multiple archive events, The matching degree between each of the file events and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree;

The secondary archiving unit is used to perform secondary archiving on the plurality of archival events to obtain a first target archiving event stream and a second target archiving event stream to complete the archiving operation for the plurality of archival events, wherein: The first target archive event stream is a set of archive events corresponding to the first person, and the second target archive event stream is a set of archive events corresponding to the second person.

In a third aspect, the present invention provides an electronic device, including: a processor and a memory; and one or more programs, the one or more programs being stored in the memory and configured to be processed by the The program includes instructions for some or all of the steps described in the first aspect.

In a fourth aspect, embodiments of the present invention provide a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, wherein the computer program causes the computer to execute the first aspect of the embodiment of the present invention. instructions for some or all of the steps described in .

In a fifth aspect, embodiments of the present invention provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause the computer to execute the present invention. Some or all of the steps described in the first aspect of the embodiment of the invention. The computer program product may be a software installation package.

Implementing the embodiments of the present invention has the following beneficial effects:

It can be seen that through the embodiment of the present invention, a set of archive events to be processed is determined, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are obtained by first filing in chronological order. The matching degree between each of the multiple file events and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree; perform secondary archiving of the multiple file events, A first target archive event stream and a second target archive event stream are obtained to complete the archive operation for the plurality of archive events, wherein the first target archive event stream is a set of archive events corresponding to the first person, and the The second target file event stream is a set of file events corresponding to the second person. In this way, multiple archival events can be archived twice to complete the archiving operation of the archival events, which will help reduce the probability of archiving errors in archiving events, improve the archiving accuracy, and improve the accuracy when restoring the real archives in the future. Rate.

Description of the drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. Those of ordinary skill in the art can also obtain other drawings based on these drawings without exerting creative efforts.

Figure 1A is a schematic system architecture diagram of a data archiving method provided by an embodiment of the present invention;

Figure 1B is a schematic flow diagram of an embodiment of a data archiving method provided by an embodiment of the present invention;

Figure 2 is a schematic flow diagram of an embodiment of a data archiving method provided by an embodiment of the present invention;

Figure 3 is a schematic structural diagram of a data archiving device provided by an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Embodiments of the invention

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of the present invention.

The terms "first", "second", "third" and "fourth" in the description, claims and drawings of the present invention are used to distinguish different objects, rather than describing a specific sequence. . Furthermore, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes Other steps or units inherent to such processes, methods, products or devices.

Reference herein to "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those skilled in the art understand, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments.

In order to better understand a data archiving method, related devices and electronic equipment provided by embodiments of the present invention, the system architecture of the data archiving method applicable to embodiments of the present invention is first described below. Refer to FIG. 1A , which is a schematic diagram of the system architecture of the data archiving method provided by an embodiment of the present invention. As shown in Figure 1A, the system architecture may include one or more servers and multiple electronic devices, where:

Servers may include but are not limited to backend servers, component servers, data archiving system servers or data archiving software servers, etc. Servers may communicate with multiple electronic devices through the Internet to complete data archiving operations, and may also archive data results (for example, The first target archiving event stream and the second target archiving event stream in the embodiment of the present application) are sent to the electronic device. Of course, in the present invention, the data archiving operation can also be completed by the electronic device and sent to the server for data synchronization or storage.

The electronic devices described in the embodiments of the present invention may include camera devices, smart phones (such as Android phones, iOS phones, Windows Phones, etc.), tablets, handheld computers, laptops, mobile Internet devices (MID, Mobile Internet Devices) or wearable devices, etc. The above are only examples, not exhaustive, including but not limited to the above devices. Of course, the above The electronic device can also be a server.

The technical terms related to the present invention are explained below:

1) Real-name file: Each person has a unique ID, and each ID is associated with a real-name file, which is used to describe the basic information of the person.

2) Face feature value (feature): Each face capture picture has a unique feature value, which is used to store the basic features of the face in the picture. The comparison of feature values of different face data of the same person has a higher similarity theory.

3) Archiving: The process of analyzing all captured events and placing them under the name of a specific person's file.

In the present invention, the electronic device can determine a set of archive events to be processed, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are obtained by first filing in chronological order. Each archive in the multiple archive events The matching degree between the event and the first archive cover event and/or the second archive cover event is greater than or equal to the preset matching degree; perform secondary archiving on multiple archive events to obtain the first target archive event stream and the second target archive event stream to complete the archiving operation for multiple archive events, wherein the first target archive event stream is the archive event set corresponding to the first person, and the second target archive event stream is the archive event set corresponding to the second person; obtain the An external archive event information and a second external archive event information; coupling and comparing the first target archive event stream and the second target archive event stream with the first external archive event and the second external archive event respectively to obtain multiple coupling degrees; The first archive event corresponding to the highest coupling degree or the event to be analyzed is selected as the target archive event, the archive data corresponding to the target archive event is obtained, and the archive data is archived.

It should be noted that in the present invention, multiple may refer to two or more, which will not be described again. The system architecture of the data archiving method provided by this application is not limited to that shown in Figure 1A.

Please refer to FIG. 1B , which is a schematic flow chart of an embodiment of a data archiving method provided by an embodiment of the present invention. The data archiving method described in this embodiment includes the following steps:

101. Determine a set of archive events to be processed, wherein the set of archive events includes multiple archive events, and the archive events are first archived and arranged in chronological order, and each of the multiple archive events is associated with The matching degree between the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree.

Each of the above-mentioned archive events may correspond to archive data. The archive data may include: archive cover image, identity authentication logo, etc., wherein the archive cover image may be the face image, human body image, etc. of the user corresponding to the archive data. , not limited here.

Among them, each file event can correspond to a real-name file. The electronic device can smoothly arrange the above multiple archive events in chronological order when archiving for the first time.

Wherein, the first file cover event and the second file cover event can be preset in the electronic device, and the first file cover event and the second file cover event are different.

Wherein, the above-mentioned first file cover event and the second file cover event may constitute the same archive library, and the data corresponding to the first file cover event and the data corresponding to the second file cover event are highly similar user profile data, for example, the One file cover event can correspond to the older sister of the twins, and the second file cover event can correspond to the younger sister of the twins. The above-mentioned first file cover event and/or the second file cover event may be a face image, a human body image, etc.

In a possible example, determining the set of archive events to be processed includes the following steps:

A1. Obtain a capture event set, where the capture event set includes multiple capture events;

A2. Match each of the plurality of snapshot events with the facial feature value of the first file cover event and the second file cover event, respectively, to obtain multiple matching degrees;

A3. According to the multiple matching degrees, archive the multiple capture events into the first file cover event corresponding set and the second file cover event corresponding set to obtain a first event stream and a second event stream. , wherein the matching degree corresponding to any archive event in the first event stream and/or the second event stream is greater than the preset matching degree;

A4. Determine the archive event set according to the first event stream after archiving and the second preset stream after archiving.

Wherein, the above-mentioned capture event set may include multiple capture events, and each capture event may correspond to capture data. The capture data may include at least one of the following: facial feature value feature, capture time, location (latitude and longitude, GPS information, etc.) ), human body attributes (top color, bottom color, clothing style, etc.), etc., are not limited here.

Specifically, the facial feature value corresponding to each capture event can be matched with the facial feature value corresponding to the first file cover event to obtain multiple first matching degrees; the facial feature value corresponding to each capture event can be matched The value is matched with the facial feature value corresponding to the second file cover event to obtain multiple second matching degrees; select multiple first matching degrees that are greater than or equal to the preset matching degree (can be set by the user or the system defaults, There is no limitation here. The preset matching degree can be set to a high similarity threshold (such as 0.89, 0.92, 0.95, etc.). The capture events corresponding to the first matching degree are divided into the set corresponding to the first file cover event, and we get The first event stream after archiving. The archived events in this first event stream can be sorted in chronological order.

Further, the capture events corresponding to the second matching degree that is greater than or equal to the preset matching degree among the plurality of second matching degrees can be selected and divided into a set corresponding to the second archive cover event to obtain the second event stream after archiving. Archived events in the second event stream can be sorted in chronological order. Finally, the first event stream and the second event stream can be merged to obtain a collection of archive events. In this way, multiple capture events can be archived for the first time by matching facial feature values, so that capture events corresponding to highly similar image data can be archived in the same event stream (for example, the first event stream or the second event stream). flow).

In a possible example, determining the archive event set based on the first event stream after archiving and the second event stream after archiving includes the following steps:

B1. Mix all archive events in the first event stream after archiving and the second event stream after archiving to obtain a third event stream;

B2. Arrange the archive events in the third event stream in chronological order to obtain the archive event set.

Among them, because the faces are highly similar between the corresponding data of the first file cover event and the second file cover event, when archiving multiple snapshot events for the first time, only facial feature value matching is used to complete the archiving operation. If the first file The cover event and the second file cover event correspond to the face data of a pair of twins. Therefore, the probability of archiving errors is very predictable. There will be large errors in the subsequent trajectory analysis based on the archiving results after archiving; therefore, it can be Mix all the archived events in the first event stream and the second event stream, and rearrange them in chronological order to obtain a set of archived events, which can be used to implement secondary archiving in the future, which is beneficial to reducing the probability of archived event archiving errors.

102. Perform secondary archiving on the multiple archive events to obtain a first target archive event stream and a second target archive event stream to complete the archiving operation for the multiple archive events, wherein the first target archive event stream is The event stream is a set of file events corresponding to the first person, and the second target file event stream is a set of file events corresponding to the second person.

Among them, the above-mentioned multiple archive events are only determined by facial feature value matching. If the facial data corresponding to the above-mentioned first preset event and the second preset event are highly similar, archiving errors may occur. Therefore, in order to improve the archiving accuracy, further secondary archiving analysis operations can be performed based on the archiving event set to prevent the problem of archived event stream archiving disorder for people with highly similar faces.

Wherein, the above-mentioned first person and second person are not the same person, and the first person and the second person may be persons with highly similar facial features.

In a possible example, the secondary archiving of the plurality of archive events to obtain a first target archive event stream and a second target archive event stream includes the following steps:

21. According to the chronological order, select the top archive event from the plurality of archive events as the analysis event, and select the next archive event of the analysis event as the event to be analyzed;

22. Use the analysis event as the starting point of the first target archived event stream corresponding to the first person, and analyze whether the event to be analyzed and the analyzed event are archived events of the same person;

23. If it is determined that the analysis event and the event to be analyzed are archived events of the same person, then archive the event to be analyzed into the first target archived event stream, and arrange the event to be analyzed into all Describe and analyze the events behind;

24. If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, then archive the event to be analyzed into the second target archived event stream corresponding to the second person, and The event to be analyzed serves as the starting point event of the second target archived event stream corresponding to the second person.

Among them, the above-mentioned events to be analyzed may refer to archive events that have not yet undergone secondary archiving operations, and the above-mentioned analysis events may refer to archive events that have achieved secondary archiving, and the above-mentioned analysis events may be used to analyze events to be analyzed.

Among them, multiple archive events can be used as analysis objects, and the analysis of the above multiple archive events can be started in chronological order; in specific implementation, the above multiple archive events are arranged in chronological order, and can be determined to be arranged at the front That is, the top archive event is used as the analysis event, and the analysis event is directly used as any one of the two target archive event streams (the first target archive event stream and the second target archive event stream) after archiving; for example, you can The analysis event serves as the starting point or origin event in the first target archive event stream.

Further, the next archive event of the analysis event, that is, the frontmost archive event among the multiple archive events, can be used as the event to be analyzed corresponding to the analysis event. Furthermore, the person corresponding to the first target archive event stream can be defined as The first person, similarly, defines the person corresponding to the second target archive event stream as the second person; furthermore, it can be determined that the above analysis event is the first target archive event stream corresponding to the first person, that is, the starting point of the archive event collection, and Analyze whether the event to be analyzed and the event being analyzed are file events corresponding to the same person.

Furthermore, if it is determined that the event to be analyzed and the analyzed event are not archived events of the same person, then the event to be analyzed can be determined as the starting point event of the second target archived event stream of the second person. When there is only the archive event stream of the first person, that is, when the event to be analyzed and the event to be analyzed are the archive events of the same person, the event to be analyzed is archived into the first target archived event stream, and the events to be analyzed are arranged Go to the back of the analysis event.

It can be seen that in this example, the electronic device can take multiple archive events to be archived for secondary archiving as analysis objects, select the events to be analyzed and the analysis events according to the order of the multiple archive events, and compare whether the above two events are File events for the same person to complete the first cycle or the first secondary filing process to determine the file events corresponding to the first person and/or the second person; finally, the first person can be The last archive event of the target archive event stream or the second target archive event stream is used as the starting point of the next secondary archive, which is helpful to help each subsequent archive event be archived smoothly and correctly into the corresponding target archive event stream, which is beneficial to improve Secondary archiving accuracy.

Optionally, after step 23, that is, after arranging the event to be analyzed behind the analysis event, the method further includes the following steps:

C1. Use the currently highest-ranked file event among the multiple file events as the event to be analyzed for the next secondary archive, and use the last-ranked file event in the first target archive event stream as all events to be analyzed. Describe the analysis events of the next secondary archive;

C2. During the next secondary archiving, analyze whether the analysis event and the event to be analyzed are the same archive event;

C3. If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, archive the event to be analyzed into the second target archived event stream corresponding to the second person.

Among them, after step 22, if it is determined that the event to be analyzed and the analyzed event in steps 21-24 are the same archive event, it is only necessary to change the next event to be analyzed (which is the next of multiple archive events that have not been archived twice). , that is, the archive event that is currently ranked highest) is compared with the archive event that is ranked last in the first target archive event stream of the first person (for example, it can be the event to be analyzed in steps 21-24), and determines Whether to continue to archive the currently top-ranked archive event into the first target archive event stream of the first person, if it is determined that the currently top-tier archive event (event to be analyzed) is not the same as the one in the first target archive event stream. If the archive event (analysis event) at the bottom is the archive event of the same person, the event to be analyzed can be archived to the first target archive event stream; otherwise, the event to be analyzed can be directly archived to the corresponding file of the second person. The second target archives the event stream.

Optionally, after step 24 or C3, after archiving the event to be analyzed into the second target archived event stream corresponding to the second person; the method further includes the following steps:

D1. Determine the next file event of the event to be analyzed as the file to be analyzed in the next secondary archiving process;

D2. Use the last-ranked file event among the first target archiving events as the first analysis event, and use the last-ranked file event among the second target archiving events as the second analysis event;

D3. Compare the file to be analyzed with the first analysis event and the second analysis event respectively, so as to archive the file to be analyzed into the first target archive event stream or the second target archive event stream. in the target archive event stream.

Among them, after step 22, if it is determined that the event to be analyzed and the analyzed event in steps 21-24 are not the same archive event, that is, when there are both the first target archive event stream and the second target archive event stream respectively corresponding to the first person and the second person. In the case of a two-target archive event stream, it is necessary to combine the next event to be analyzed with the last archive event in the archive event stream of the first person and the second person, that is, the first analysis event and the second analysis event respectively. Compare and then classify until all archive events in the archive event collection are classified.

It can be seen that in this example, the current top-ranked archive event among the multiple archive events that have not been archived twice can always be used as an event to be analyzed, and go to the first target archive event and/or the second target archive event stream. Comparison and analysis of the starting event or the last-ranked archive event to achieve secondary archiving of the current top-ranked archive event, which is conducive to archiving the archived event into the correct target archive event stream and is conducive to improving Archiving accuracy.

In a possible example, the above-mentioned step 22, analyzing whether the event to be analyzed and the event to be analyzed are file events of the same person, includes the following steps:

221. Determine the first time difference and the first distance difference between the analysis event and the event to be analyzed, and calculate the first time difference and the first distance difference based on the first time difference and the first distance difference. The degree of spatiotemporal coupling between the analysis event and the event to be analyzed;

222. Determine the second time difference and the second distance difference between the analysis event and the event to be analyzed, and calculate the second time difference and the second distance difference based on the second time difference and the second distance difference. Possible values of changes in human body attributes between the analysis event and the event to be analyzed;

223. If the spatio-temporal coupling degree is less than the preset threshold, or the possible value of the human body attribute change is less than the preset threshold, then determine that the analysis event and the event to be analyzed are not archival events of the same person;

224. If the spatiotemporal coupling degree is greater than or equal to the preset threshold, and the possible value of the human body attribute change is greater than or equal to the preset threshold, then determine that the analysis event and the event to be analyzed are file events of the same person. .

In specific implementation, the above-mentioned preset first calculation formula and/or the preset second calculation formula can be set by the user or the system defaults, which is not limited here. The analysis event and the event to be analyzed can be scored through the first calculation formula and the second calculation formula to obtain the spatiotemporal coupling degree and possible values of changes in human body attributes.

Specifically, the analysis event and the event to be analyzed can be scored in two dimensions (the spatiotemporal coupling degree dimension and the human body attribute change value dimension). The possible values of the spatiotemporal coupling degree or human body attribute change obtained by scoring in each dimension range from 0~ 100 points; the first calculation formula can be set as: o1=k1/(t1*d1), where k1 is a configurable coefficient, t1 is the time difference (unit: s) corresponding to the two events, and d1 is the time difference between the two events. The distance difference corresponding to the event (unit: m). It can be seen from the first calculation formula that the degree of spatiotemporal coupling is inversely proportional to the distance difference and time difference between the two events. The lower the score, the lower the probability that the two events belong to the same person. The higher the score, the higher the probability that the two events belong to the same person. Furthermore, the spatio-temporal coupling degree corresponding to the analysis event and the event to be analyzed can be determined according to the first calculation formula, the first time difference value and the distance difference value.

Furthermore, the second calculation formula can be set as: o2=(k2*t2)/d2, where d2 is the second distance difference, k2 is a configurable coefficient, and different types of human body attribute changes correspond to different k2 values. , can be set according to the actual scene. For example, since it is much easier to change tops than bottoms, the k2 value corresponding to the scene of changing the color and style of the bottoms can be set to be higher than the k2 value corresponding to the scene of changing the color and style of the tops. Among them, o2 is directly proportional to the second time difference and inversely proportional to the second distance difference. Furthermore, the possible human body attribute change values corresponding to the analysis event and the event to be analyzed can be determined based on the second calculation formula, the second time difference and the second distance difference.

It should be noted that if there is no change in human body attributes, o2 will be the highest score of 100 points.

Furthermore, the above-mentioned preset threshold can be set by the user or set by the system default, which is not limited here. The preset threshold means that in the above two dimensions, the analysis event and the event to be analyzed are the same file's standard value; for example, the preset threshold can be set to 60 points.

Furthermore, if the spatiotemporal coupling degree is less than the preset threshold, or the possible changes in human body attributes are less than the preset threshold, it is determined that the analysis event and the event to be analyzed are not file events corresponding to the same person; if the spatiotemporal coupling degree is greater than or equal to the preset threshold, And the possible value of changes in human body attributes is greater than or equal to the preset threshold, then it is determined that the analysis event and the event to be analyzed are file events corresponding to the same person.

It can be seen that in this example, two archival events can be calculated in different dimensions through two dimensions to obtain the spatiotemporal coupling degree and the possible values of human body attribute changes; and combined with the spatiotemporal coupling degree and the possible values of human body attribute changes, the two archive events can be calculated in different dimensions. Evaluate whether each archive event is an archive event corresponding to the same person, to determine whether two archive events are an archive event corresponding to the same person, which is beneficial to improving the archiving accuracy.

Optionally, in step D3, the file to be analyzed is compared with the first analysis event and the second analysis event respectively, so as to archive the file to be analyzed to the first target archiving event. In the stream or in the second target archived event stream, the following steps may be included: the first calculation formula and the second calculation formula in steps 221 to 224 may be used to respectively calculate the difference between the event to be analyzed and the first analysis event. The first spatio-temporal coupling degree and the first human body attribute change possible value, as well as the second spatio-temporal coupling degree and the second human body attribute change possible value between the event to be analyzed and the second analysis event; further, the first spatio-temporal coupling degree and the second human body attribute change possible value can be compared The degree of spatiotemporal coupling, the possible value of the first human body attribute change, the second degree of spatiotemporal coupling and the possible value of the second human body attribute change.

Further, if any calculated value between the first spatiotemporal coupling degree and the second spatiotemporal coupling degree is less than or equal to the preset threshold, and the first human body attribute change possible value is greater than the preset threshold and the second human body attribute change possible value is greater than If the threshold is preset, it is determined that the event to be analyzed and the second event to be analyzed are archive events of the same person, and the event to be analyzed can be archived into the second target archived event stream.

If any calculated value between the first human body attribute change possible value and the second human body attribute change possible value is less than or equal to the preset threshold, and the first space-time coupling degree is greater than the preset threshold and the second space-time coupling degree is greater than the preset threshold. ; then it is determined that the event to be analyzed and the first event to be analyzed are archive events of the same person, and the event to be analyzed can be archived into the first target archived event stream.

If the first spatio-temporal coupling degree, the second spatio-temporal coupling degree, the first possible human body attribute change value and the second human body attribute change possible value are all greater than the preset threshold, or the first spatio-temporal coupling degree, the second spatio-temporal coupling degree, the first human body attribute change possible value If both the possible human body attribute change value and the second human body attribute change possible value are less than or equal to the preset threshold, then the sum between the first spatiotemporal coupling degree and the second spatiotemporal coupling degree is calculated, the first result is obtained, and the first human body attribute is calculated The second result is obtained by the sum of the possible change value and the possible change value of the second human body attribute.

If the first result is greater than the second result, it will be determined that the event to be analyzed and the first event to be analyzed are archive events of the same person, and the event to be analyzed can be archived into the first target archived event stream.

If the first result is less than the second result, it is determined that the event to be analyzed and the second event to be analyzed are archive events of the same person, and the event to be analyzed can be archived into the second target archived event stream.

Optionally, after obtaining the first target archive event stream and the second target archive event stream, the following steps may also be included:

E1. Obtain the first external file event information and the second external file event information;

E2. Perform the first external archive event information and the second external archive event information with the first target archive event stream and/or the second target archive event stream to obtain multiple coupling degrees;

E3. Obtain the archive event corresponding to the highest coupling degree among multiple coupling degrees; the target external archive event information (the first external archive event information or the second external archive event information) corresponding to the archive event can be determined.

Among them, after secondary archiving of all multiple archive events, the first target archive event stream and the second target archive event stream can be output; however, it cannot be implemented to determine which real-name archive corresponds to which target archive event stream. It needs to be matched with The third party externally provides two target archive events (imsi code information or card swiping information with ID card, etc.) information for coupling comparison. After comparison, the event with the highest coupling degree of the archive event stream is obtained, and the archive identity information to which the archive event belongs can be confirmed; The more target file events provided by third parties, the higher the probability of correct files and archiving events corresponding to the target external file event information analyzed.

For example, as shown in Figure 2, it is a schematic flow chart of a data archiving method. As shown in the figure, module 1 can be used to perform the method shown in A1-A4 shown in the present invention, and module 2 can be used to perform this method. In the method shown in steps B1-B2 of the present invention, module 3 can be used to perform steps 21-step 24 and related methods of the present invention; module 4 can be used to perform the method shown in steps E1-E3 of the present invention.

Consistent with the above, the following is a device for implementing the above data archiving method, specifically as follows:

Please refer to Figure 3a, which is a schematic structural diagram of a data archiving device provided by an embodiment of the present invention. The data archiving device described in this embodiment includes: a determination unit 301 and a secondary archiving unit 302, specifically as follows:

The determining unit 301 is used to determine a set of archive events to be processed, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are obtained by first filing in chronological order. The multiple archive events The matching degree between each of the file events and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree;

The secondary archiving unit 302 is used to perform secondary archiving on the plurality of archival events to obtain a first target archiving event stream and a second target archiving event stream to complete the archiving operation for the plurality of archival events, wherein, The first target archive event stream is a set of archive events corresponding to the first person, and the second target archive event stream is a set of archive events corresponding to the second person.

In a possible example, the determination of a set of archive events to be processed; the above-mentioned determination unit 301 is specifically used to:

Obtain a collection of capture events, where the collection of capture events includes multiple capture events;

Perform face feature value matching on each of the plurality of snapshot events with the first file cover event and the second file cover event, respectively, to obtain multiple matching degrees;

According to the multiple matching degrees, the multiple capture events are respectively archived into the first archive cover event corresponding set and the second archive cover event corresponding set to obtain a first event stream and a second event stream, Wherein, the matching degree corresponding to any archive event in the first event stream and/or the second event stream is greater than the preset matching degree;

Determine the archive event set according to the first event stream after archiving and the second event stream after archiving

In a possible example, in determining the archive event set based on the first event stream after archiving and the second event stream after archiving, the above-mentioned determination unit 301 is specifically used to:

Mix all archived events in the first event stream after archiving and the second event stream after archiving to obtain a third event stream;

Arrange the archive events in the third event stream in chronological order to obtain the archive event set.

In a possible example, in terms of secondary archiving of the plurality of archive events to obtain the first target archive event stream and the second target archive event stream, the above-mentioned secondary archive unit 302 is specifically used to:

According to the chronological order, select the highest-ranked archive event from the plurality of archive events as the analysis event, and select the next archive event of the analysis event as the event to be analyzed;

Use the analysis event as the starting point of the first target archived event stream corresponding to the first person, and analyze whether the event to be analyzed and the analyzed event are archived events of the same person;

If it is determined that the analysis event and the event to be analyzed are archived events of the same person, then the event to be analyzed is archived into the first target archived event stream, and the event to be analyzed is arranged into the analysis event. behind the incident;

If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, then the event to be analyzed is archived into the second target archived event stream corresponding to the second person, and the event to be analyzed is The event serves as the starting point event of the second target archived event stream corresponding to the second person.

In a possible example, after arranging the event to be analyzed behind the analysis event, the above-mentioned secondary archiving unit 302 is also specifically used to:

The archive event currently ranked highest among the plurality of archive events is used as the event to be analyzed for the next secondary archive, and the archive event ranked last in the first target archive event stream is used as the next archive event. Analysis events of primary and secondary archives;

During the next secondary archiving, analyze whether the analysis event and the event to be analyzed are archive events of the same person;

If it is determined that the analysis event and the event to be analyzed are archived events of the same person, then the event to be analyzed is archived into the first target archived event stream, and the event to be analyzed is arranged into the first target archived event stream. The back of the archive event describing the analysis event;

If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, then the event to be analyzed is archived into the second target archived event stream corresponding to the second person.

In a possible example, after archiving the event to be analyzed into the second target archiving event stream corresponding to the second person, the secondary archiving unit 302 is further configured to:

Determine the next file event of the event to be analyzed as the file to be analyzed in the next secondary archiving process;

Use the last-ranked file event among the first target archiving events as the first analysis event, and use the last-ranked file event among the second target archiving events as the second analysis event;

Compare the file to be analyzed with the first analysis event and the second analysis event respectively, so as to archive the file to be analyzed into the first target archiving event stream or the second target archive. in the event stream.

In a possible example, in terms of analyzing whether the event to be analyzed and the event to be analyzed are archival events of the same person, the above-mentioned secondary archiving unit 302 is also specifically used to:

Determine a first time difference and a first distance difference between the analysis event and the event to be analyzed, and calculate the analysis event based on the first time difference and the first distance difference. and the spatiotemporal coupling degree between the events to be analyzed;

Determine a second time difference and a second distance difference between the analysis event and the event to be analyzed, and calculate the analysis event based on the second time difference and the second distance difference. and the possible values of human body attribute changes between the events to be analyzed;

If the spatiotemporal coupling degree is less than the preset threshold, or the possible value of the human body attribute change is less than the preset threshold, it is determined that the analysis event and the event to be analyzed are not file events of the same person;

If the spatiotemporal coupling degree is greater than or equal to the preset threshold, and the possible value of the human body attribute change is greater than or equal to the preset threshold, it is determined that the analysis event and the event to be analyzed are file events of the same person.

It can be seen that through the data archiving device described in the embodiment of the present invention, a set of archive events to be processed is determined, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are first archived in chronological order. The archive arrangement is such that the matching degree between each of the multiple archive events and the first archive cover event and/or the second archive cover event is greater than or equal to the preset matching degree; for the multiple archives The event is archived twice to obtain a first target archive event stream and a second target archive event stream to complete the archiving operation for the multiple archive events, wherein the first target archive event stream is the event stream corresponding to the first person. A set of file events, and the second target file event stream is a set of file events corresponding to the second person. In this way, multiple archival events can be archived twice to complete the archiving operation of the archival events, which will help reduce the probability of archiving errors in archiving events, improve the archiving accuracy, and improve the accuracy when restoring the real archives in the future. Rate.

Consistent with the above, please refer to FIG. 4 , which is a schematic structural diagram of an electronic device according to an embodiment of the present invention. Including the data archiving system, camera or server described in this embodiment, including: at least one input device 1000; at least one output device 2000; at least one processor 3000, such as CPU; and memory 4000, the above-mentioned input device 1000, output device 2000, processor 3000 and memory 4000 are connected through bus 5000.

The input device 1000 may be a touch panel, a physical button, or a mouse.

The above-mentioned output device 2000 may specifically be a display screen.

The above-mentioned memory 4000 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as a disk memory. The above-mentioned memory 4000 is used to store a set of program codes. The above-mentioned input device 1000, output device 2000 and processor 3000 are used to call the program code stored in the memory 4000 and perform the following operations:

The above processor 3000 is used for:

Determine a set of archive events to be processed, the set of archive events includes multiple archive events, and the archive events are arranged according to the first archive in chronological order;

Perform secondary archiving on the multiple archive events to obtain a first target archive event stream and a second target archive event stream to complete the archiving operation for the multiple archive events.

It can be seen that the electronic device described in the embodiment of the present invention determines a set of archive events to be processed, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are arranged for the first time in chronological order. It is obtained that the matching degree between each of the multiple file events and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree; the multiple file events are processed Secondary archiving to obtain the first target archiving event stream and the second target archiving event stream to complete the archiving operation for the multiple archive events, wherein the first target archive event stream is the archive event corresponding to the first person The second target file event stream is a set of file events corresponding to the second person. In this way, multiple archival events can be archived twice to complete the archiving operation of the archival events, which will help reduce the probability of archiving errors in archiving events, improve the archiving accuracy, and improve the accuracy when restoring the real archives in the future. Rate.

In a possible example, the above-mentioned processor 3000 is used to determine the set of archive events to be processed:

In a possible example, in determining the archive event set based on the first event stream after archiving and the second event stream after archiving, the above processor 3000 is used to:

In a possible example, in terms of secondary archiving of the plurality of archive events to obtain a first target archive event stream and a second target archive event stream, the above processor 3000 is used to:

In a possible example, after arranging the event to be analyzed behind the analysis event, the above processor 3000 is also used to:

The archive event currently ranked highest among the plurality of archive events is used as the event to be analyzed for the next secondary archive, and the archive event ranked last in the first target archive event stream is used as the next archive event. A secondary archived analysis event;

During the next secondary archiving, analyze whether the analysis event and the event to be analyzed are the same archive event;

In a possible example, after archiving the event to be analyzed into the second target archived event stream corresponding to the second person, the above processor 3000 is further configured to:

In a possible example, in analyzing whether the event to be analyzed and the event to be analyzed are archive events of the same person, the above processor 3000 is used to:

Embodiments of the present invention also provide a computer storage medium, wherein the computer storage medium can store a program, which when executed includes part or all of the steps of any of the data archiving methods described in the above method embodiments.

Although the present invention has been described herein in conjunction with various embodiments, those skilled in the art, in practicing the claimed invention, will understand and understand by reviewing the drawings, the disclosure, and the appended claims. Other variations of the disclosed embodiments are implemented. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit may perform several of the functions recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not mean that a combination of these measures cannot be combined to advantageous effects.

Those skilled in the art will understand that embodiments of the present invention may be provided as methods, apparatuses (devices), or computer program products. Thus, the invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the invention 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, etc.) having computer-usable program code embodied therein. The computer program is stored/distributed on a suitable medium, provided with or as part of other hardware, and may also take other forms of distribution, such as over the Internet or other wired or wireless telecommunications systems.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (equipment) and computer program products according to embodiments of the invention. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations 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 archiving device to produce a machine such that the instructions executed by the processor of the computer or other programmable data archiving device produce a use A device for realizing the functions specified in one process or multiple processes of the flowchart and/or one block or multiple blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data archiving device to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the instructions The device implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data archiving device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby executing on the computer or other programmable device. Instructions provide steps for implementing the functions specified in a process or processes of a flowchart diagram and/or a block or blocks of a block diagram.

Although the invention has been described in conjunction with specific features and embodiments thereof, it will be apparent that various modifications and combinations may be made without departing from the spirit and scope of the invention. Accordingly, the specification and drawings are intended to be illustrative only of the invention as defined by the appended claims and are to be construed to cover any and all modifications, variations, combinations or equivalents within the scope of the invention. Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the invention. In this way, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

A data archiving method, characterized by including:

Determine a set of archive events to be processed, wherein the set of archive events to be processed includes a plurality of archive events, and the plurality of archive events are first archived and arranged in chronological order, and each of the archives in the plurality of archive events is The matching degree between the event and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree;

Perform secondary archiving on the multiple archive events to obtain a first target archive event stream and a second target archive event stream to complete the archiving operation for the multiple archive events, wherein the first target archive event stream is a set of file events corresponding to the first person, and the second target file event stream is a set of file events corresponding to the second person.
The method according to claim 1, characterized in that determining the set of archive events to be processed includes:

Obtain a collection of capture events, where the collection of capture events includes multiple capture events;

Perform face feature value matching on each of the plurality of snapshot events with the first file cover event and the second file cover event, respectively, to obtain multiple matching degrees;

According to the multiple matching degrees, the multiple capture events are respectively archived into the first archive cover event corresponding set and the second archive cover event corresponding set to obtain a first event stream and a second event stream, Wherein, the matching degree corresponding to any archive event in the first event stream and/or the second event stream is greater than the preset matching degree;

The archive event set is determined based on the first event stream after archiving and the second event stream after archiving.
The method of claim 2, wherein determining the archive event set based on the first event stream after archiving and the second event stream after archiving includes:

Mix all archived events in the first event stream after archiving and the second event stream after archiving to obtain a third event stream;

Arrange the archive events in the third event stream in chronological order to obtain the archive event set.
The method according to claim 3, characterized in that the second archiving of the plurality of archival events to obtain the first target archiving event stream and the second target archiving event stream includes:

According to the chronological order, select the highest-ranked archive event from the plurality of archive events as the analysis event, and select the next archive event of the analysis event as the event to be analyzed;

Use the analysis event as the starting point of the first target archived event stream corresponding to the first person, and analyze whether the event to be analyzed and the analyzed event are archived events of the same person;

If it is determined that the analysis event and the event to be analyzed are archived events of the same person, then the event to be analyzed is archived into the first target archived event stream, and the event to be analyzed is arranged into the analysis event. behind the incident;

If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, then the event to be analyzed is archived into the second target archived event stream corresponding to the second person, and the event to be analyzed is The event serves as the starting point event of the second target archived event stream corresponding to the second person.
The method according to claim 4, characterized in that, after arranging the event to be analyzed behind the analysis event, the method further includes:

The archive event currently ranked highest among the plurality of archive events is used as the event to be analyzed for the next secondary archive, and the archive event ranked last in the first target archive event stream is used as the next archive event. A secondary archived analysis event;

During the next secondary archiving, analyze whether the analysis event and the event to be analyzed are archive events of the same person;

If it is determined that the analysis event and the event to be analyzed are archived events of the same person, then the event to be analyzed is archived into the first target archived event stream, and the event to be analyzed is arranged into the first target archived event stream. The back of the archive event describing the analysis event;

If it is determined that the analysis event and the event to be analyzed are not archived events of the same person, then the event to be analyzed is archived into the second target archived event stream corresponding to the second person.
The method according to claim 4 or 5, characterized in that, after archiving the event to be analyzed into the second target archived event stream corresponding to the second person; the method further includes:

Determine the next file event of the event to be analyzed as the file to be analyzed in the next secondary archiving process;

Use the last-ranked file event among the first target archiving events as the first analysis event, and use the last-ranked file event among the second target archiving events as the second analysis event;

Compare the file to be analyzed with the first analysis event and the second analysis event respectively, so as to archive the file to be analyzed into the first target archiving event stream or the second target archive. in the event stream.
The method according to any one of claims 4 to 6, characterized in that the analysis of whether the event to be analyzed and the analyzed event are archived events of the same person includes:

Determine a first time difference and a first distance difference between the analysis event and the event to be analyzed, and calculate the analysis event based on the first time difference and the first distance difference. and the spatiotemporal coupling degree between the events to be analyzed;

Determine a second time difference and a second distance difference between the analysis event and the event to be analyzed, and calculate the analysis event based on the second time difference and the second distance difference. and the possible values of human body attribute changes between the events to be analyzed;

If the spatiotemporal coupling degree is less than the preset threshold, or the possible value of the human body attribute change is less than the preset threshold, it is determined that the analysis event and the event to be analyzed are not file events of the same person;

If the spatiotemporal coupling degree is greater than or equal to the preset threshold, and the possible value of the human body attribute change is greater than or equal to the preset threshold, it is determined that the analysis event and the event to be analyzed are file events of the same person.
A data archiving device, characterized by including:

Determining unit, used to determine a set of archive events to be processed, wherein the set of archive events to be processed includes multiple archive events, and the multiple archive events are obtained by first filing in chronological order. Among the multiple archive events, The matching degree between each of the file events and the first file cover event and/or the second file cover event is greater than or equal to the preset matching degree;

The secondary archiving unit is used to perform secondary archiving on the plurality of archival events to obtain a first target archiving event stream and a second target archiving event stream to complete the archiving operation for the plurality of archival events, wherein: The first target archive event stream is a set of archive events corresponding to the first person, and the second target archive event stream is a set of archive events corresponding to the second person.
An electronic device, characterized by comprising a processor, a memory, a communication interface, and one or more programs, the one or more programs being stored in the memory and configured to be executed by the processor, The program includes instructions for performing the steps of the method of any one of claims 1-7.
A computer-readable storage medium, characterized by storing a computer program for electronic data exchange, wherein the computer program causes the computer to execute the method according to any one of claims 1-7.