An image archiving method, device, terminal and computer readable storage medium

Abstract

The application provides an image merging method and device, a terminal and a computer readable storage medium. The image merging method comprises: acquiring a plurality of image sets; wherein the image set comprises a plurality of images of the same target object, each image has a collection time and a collection position; based on the similarity between a plurality of target objects corresponding to a plurality of image sets, determining the image sets corresponding to two target objects with a similarity greater than a similarity threshold as a first image set and a second image set; according to the collection time of the image, forming a candidate trajectory image set by the image in the first image set and the image in the second image set; based on the speed between two adjacent images in the candidate trajectory image set, determining whether the first image set and the second image set are merged. The application determines whether the candidate trajectory is abnormal by the speed between two adjacent images, thereby improving the accuracy of merging the first image set and the second image set.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of portrait aggregation, in particular to an image aggregation method and device, a terminal and a computer readable storage medium. BACKGROUND

[0002] Image search, also known as image searching, refers to comparing the similarity of one or more input face or full-body photos with historical snapshot records, and finally outputting a number of historical photos in descending order of similarity. Image search is a similarity ranking, and can only find scattered and discontinuous portrait pictures in time and space. Compared with license plate recognition, it cannot accurately determine whether the faces in multiple photos are the same person, and therefore needs to be aggregated to know the complete time and space trajectory of a person.

[0003] Through portrait aggregation, the snapshot pictures form a plurality of one-person-one-file groups, and each group constitutes a person's trajectory. This improves the value of the snapshot pictures. Therefore, the accuracy of portrait aggregation is a very important indicator for evaluating the aggregation capability. SUMMARY

[0004] The technical problem solved by the present application is to provide an image aggregation method, device, terminal and computer readable storage medium, which solves the problem of low accuracy of portrait file aggregation in the prior art.

[0005] To solve the above technical problem, the first technical solution adopted by the present application is to provide an image aggregation method, which includes: obtaining a plurality of image sets; wherein the image set includes a plurality of images of the same target object, and each image has a collection time and a collection location; based on the similarity between a plurality of target objects corresponding to a plurality of image sets, determining two target objects corresponding to the image sets as the first image set and the second image set if the similarity is greater than a similarity threshold; according to the collection time of the images, forming a candidate trajectory image set from the images in the first image set and the images in the second image set; and based on the speed between adjacent two images in the candidate trajectory image set, determining whether the first image set and the second image set are aggregated.

[0006] Wherein, based on the speed between adjacent two images in the candidate trajectory image set, determining whether the first image set and the second image set are aggregated, includes: determining the speed corresponding to the adjacent two images according to the collection location and the collection time corresponding to the adjacent two images in the candidate trajectory image set; judging whether the speed corresponding to all adjacent two images in the candidate trajectory image set exceeds a speed threshold; if the speed corresponding to all adjacent two images in the candidate trajectory image set does not exceed the speed threshold, then the first image set and the second image set are aggregated; wherein the second image set and the first image set correspond to the same target object.

[0007] According to the collection positions and the collection times corresponding to the two adjacent images in the candidate trajectory image set, the speed corresponding to the two adjacent images is determined, including: the speed corresponding to the two adjacent images is calculated by formula 1.

[0008]

[0009] In formula 1, v represents the speed corresponding to the two adjacent images, with the unit of Km / h; distance represents the distance between the collection positions corresponding to the two adjacent images, with the unit of m; and t represents the time difference between the collection times corresponding to the two adjacent images, with the unit of s.

[0010] According to the collection positions and the collection times corresponding to the two adjacent images in the candidate trajectory image set, the speed corresponding to the two adjacent images is determined, including: the distance between the collection positions corresponding to the two adjacent images is calculated by formula 2.

[0011] distance = 6371000 * acos(distance_angle) (formula 2)

[0012] In formula 2, distance represents the distance between the collection positions corresponding to the two adjacent images; and distance_angle represents the distance angle between the collection devices corresponding to the two images.

[0013] The collection position includes the longitude data and the latitude data of the image collection device; according to the collection positions and the collection times corresponding to the two adjacent images in the candidate trajectory image set, the speed corresponding to the two adjacent images is determined, including: the distance angle corresponding to the two adjacent images is calculated by formula 3.

[0014]

[0015] In formula 3, distance_angle represents the distance angle between the collection devices corresponding to the two images; if distance_angle is greater than 1, 1 is taken; Lng1 represents the longitude data of the collection position of the former image in the two adjacent images; Lng2 represents the longitude data of the collection position of the latter image in the two adjacent images; Lat1 represents the latitude data of the collection position of the former image in the two adjacent images; and Lat2 represents the latitude data of the collection position of the latter image in the two adjacent images.

[0016] According to the collection positions and the collection times corresponding to the two adjacent images in the candidate trajectory image set, the speed corresponding to the two adjacent images is determined, including: the time difference between the collection times corresponding to the two adjacent images is calculated by formula 4.

[0017] t=abs((next_time-start_time) / 1000) (Formula 4)

[0018] In Formula 4: t represents the time difference between the acquisition times of two adjacent images; start_time represents the acquisition time of the previous image in two adjacent images; next_time represents the acquisition time of the next image in two adjacent images.

[0019] The process involves forming a candidate trajectory image set from images in the first image set and images in the second image set based on the image acquisition time. Prior to this, the process also includes denoising the images in the first image set and the images in the second image set respectively.

[0020] The images in the first image set and the images in the second image set are subjected to denoising processing, including: removing images in the first image set and the second image set that do not contain any of the following information: the identity data of the target object, the image acquisition time, the image acquisition location, and the image acquisition device.

[0021] The images in the first image set and the second image set each contain the identity data of the target object. Based on the image acquisition time, the images in the first image set and the second image set are combined to form a candidate trajectory image set. Before this, the process includes: removing images in the first image set and the second image set that do not contain the corresponding identity data of the target object; determining whether the merged identity data corresponding to the image contained in the first image set or the second image set is null; if the merged identity data corresponding to the image is null, then the identity data corresponding to the image is assigned the merged identity data of the image.

[0022] To solve the above-mentioned technical problems, the second technical solution adopted in this application is: to provide an image merging device, which includes: an acquisition module for acquiring multiple image sets; wherein the image set includes multiple images of the same target object, and each image has an acquisition time and acquisition position; a comparison module for determining, based on the similarity between the multiple target objects corresponding to the multiple image sets, the image sets corresponding to two target objects with a similarity greater than a similarity threshold as a first image set and a second image set; an analysis module for forming a candidate trajectory image set from the images in the first image set and the images in the second image set according to the acquisition time of the images; and a merging module for determining whether the first image set and the second image set should be merged based on the velocity between two adjacent images in the candidate trajectory image set.

[0023] To solve the above-mentioned technical problems, the third technical solution adopted in this application is: to provide a terminal, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor is used to execute program data to implement the steps in the above-mentioned image merging method.

[0024] To solve the above-mentioned technical problems, the fourth technical solution adopted in this application is: to provide a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps in the above-mentioned image merging method.

[0025] The beneficial effects of this application are as follows: Unlike existing technologies, this application provides an image merging method, apparatus, terminal, and computer-readable storage medium. The image merging method includes: acquiring multiple image sets; wherein each image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location; based on the similarity between multiple target objects corresponding to the multiple image sets, determining the image sets corresponding to two target objects with a similarity greater than a similarity threshold as a first image set and a second image set; forming a candidate trajectory image set from the images in the first image set and the second image set according to the acquisition time of the images; and determining whether the first image set and the second image set should be merged based on the velocity between adjacent images in the candidate trajectory image set. This application improves the accuracy of merging the first image set and the second image set by filtering and determining multiple image sets of target objects with a similarity greater than a similarity threshold, forming a candidate trajectory image set from the images in the multiple image sets based on the acquisition time, and judging whether there are anomalies in the candidate trajectories by the velocity between adjacent images. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart illustrating the image merging method provided in this application;

[0028] Figure 2 This is a flowchart illustrating an embodiment of the image merging method provided in this application;

[0029] Figure 3 This is a schematic diagram of a specific embodiment corresponding to step S13 provided in this application;

[0030] Figure 4 This is a schematic block diagram of an embodiment of the image merging device provided in this application;

[0031] Figure 5 This is a schematic block diagram of one embodiment of the terminal provided in this application;

[0032] Figure 6 This is a schematic block diagram of one embodiment of the computer-readable storage medium provided in this application. Detailed Implementation

[0033] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0034] In the following description, specific details such as particular system architectures, interfaces, and technologies are presented for illustrative purposes rather than for limiting purposes, in order to provide a thorough understanding of this application.

[0035] In this article, the term "and / or" simply describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "more" in this article means two or more objects.

[0036] To enable those skilled in the art to better understand the technical solution of this application, the image merging method provided in this application will be described in further detail below with reference to the accompanying drawings and specific embodiments.

[0037] Please see Figure 1 and Figure 2 , Figure 1 This is a flowchart illustrating the image merging method provided in this application; Figure 2 This is a flowchart illustrating an embodiment of the image merging method provided in this application. This embodiment provides an image merging method, and the specific steps of this method are as follows, assuming the longitude and latitude data formats and content of the acquisition device are correct.

[0038] S11: Obtain multiple image sets.

[0039] Specifically, the image set includes multiple images of the same target object, each image having an acquisition time and acquisition location.

[0040] In one embodiment, a data acquisition device is installed at a pre-defined location, and the device is assigned a number and a pre-defined model number. In other words, the pre-defined model number of the data acquisition device is its name. This data acquisition device can be an image acquisition device or a video acquisition device. In this embodiment, the image acquired by the data acquisition device contains the device's number and model number, as well as the pre-defined location of the data acquisition device corresponding to the image. The pre-defined location includes Earth's longitude and latitude data.

[0041] Multiple images of preset categories are collected by data acquisition devices installed at each checkpoint. In this embodiment, the preset categories include people. The acquired images are compared with identity images in an identity database to obtain information about the target object contained in each image. The target object information includes the ID card number and ID card type. Based on the target object information in each image, the acquired images are clustered to obtain multiple image sets. Each image set corresponds to one target object. Each image set includes multiple images containing the same target object.

[0042] Multiple image sets are preprocessed separately. Specifically, all images in each image set are sorted according to their acquisition time to generate the trajectory of the target object for each image set. The velocities of adjacent images are calculated based on their acquisition time and location. It is then determined whether the velocities of all images in the image set are less than a velocity threshold. If the velocities of the selected image and its preceding and following images are both less than the velocity threshold, the selected image is considered part of the valid trajectory. If the velocities of the selected image and its preceding or following image are not less than the velocity threshold, the selected image is not considered part of the valid trajectory and is deleted. These steps yield the valid trajectories of the target objects for each image set.

[0043] In one specific embodiment, each image set includes multiple images containing the target object, and the information accompanying the images is shown in Table 1. The ID of the most similar cluster master member (identity) is the ID of the identity image corresponding to the target object contained in the image.

[0044] Table 1

[0045]

[0046]

[0047] S12: Based on the similarity between multiple target objects corresponding to multiple image sets, determine the image sets corresponding to two target objects with a similarity greater than a similarity threshold as the first image set and the second image set.

[0048] Specifically, the similarity between target objects in each of the multiple image sets is calculated, and it is determined whether each calculated similarity is greater than a similarity threshold. If the calculated similarity is greater than the similarity threshold, it is preliminarily determined that the two target objects corresponding to the similarity are the same target object, and the image sets to which the two target objects belong are the same image sets. It is preliminarily determined that the image sets to which the two target objects belong can be merged to obtain a merged image set. Each image set corresponds to a primary file ID (dossier ID), and the merged image set also corresponds to a merged primary file ID (relation dossier ID).

[0049] In one specific embodiment, after similarity comparison, a preliminary set of images that can be combined is determined. Each image set includes multiple images containing the target object, and the information accompanying the images is shown in Table 2.

[0050] Table 2

[0051] Master ID Identity landing time Master ID after merging Gender Certificate number Age range Certificate type Whether associated with human body Whether identity landing Creation id time

[0052] To obtain good data quality, null values ​​were removed from the images in each image set beforehand.

[0053] In one embodiment, denoising processing is performed on images contained in the first image set and images contained in the second image set. Specifically, images in the first and second image sets that do not contain any of the following information: the identity data of the included target object, the image acquisition time, the image acquisition location, or the image acquisition device. In a specific embodiment, images in which any one of the following fields—"most similar cluster master member ID," "acquisition time," "acquisition device number," "acquisition device name," "Earth longitude," or "Earth latitude"—is deleted.

[0054] In one embodiment, the images in the first image set and the second image set each contain identity data of the target object. Images in the first and second image sets that do not contain corresponding identity data of the target object are removed. It is then determined whether the merged identity data corresponding to an image in the first or second image set is null. If the merged identity data corresponding to an image is null, the merged identity data of the image is assigned to the image's identity data. In a specific embodiment, images whose "Master File ID" field is null are deleted. For images whose "Merged Master File ID" field is null, the field corresponding to the image's "Master File ID" is assigned to the "Merged Master File ID" of the same image, thereby ensuring that the fields corresponding to both the "Master File ID" and the "Merged Master File ID" of each image are not null. Images whose fields corresponding to both the "Master File ID" and the "Merged Master File ID" are not null are retained.

[0055] The above steps yield the preprocessed first image set and the second image set. In Table 2, the fields corresponding to each item of all images that make up the first image set and all images that make up the second image set are not null.

[0056] S13: Based on the image acquisition time, form a candidate trajectory image set from the images in the first image set and the images in the second image set.

[0057] Please see Figure 3 , Figure 3 This is a schematic diagram of a specific embodiment corresponding to step S13 provided in this application.

[0058] Specifically, through the above steps, it can be preliminarily determined that the first image set and the second image set correspond to the same target object. All images contained in the first image set and all images contained in the second image set can be merged. Based on the acquisition time of each image, all images are sorted to form a merged candidate trajectory image set. The relevant information for each image is shown in Table 3. For example, the first image set is A, and the second image set is B.

[0059] Table 3

[0060]

[0061]

[0062] S14: Based on the velocity between two adjacent images in the candidate trajectory image set, determine whether the first image set and the second image set should be merged.

[0063] Specifically, based on the acquisition location and acquisition time of each adjacent image in the candidate trajectory image set, the velocity corresponding to each adjacent image is determined. It is then determined whether the velocities corresponding to all adjacent images in the candidate trajectory image set exceed a velocity threshold. If the velocities corresponding to all adjacent images in the candidate trajectory image set do not exceed the velocity threshold, the candidate trajectory image set is determined to be valid trajectory data, and the first and second image sets are archived. This means that the second and first image sets correspond to the same target object. The candidate trajectory image sets are then associated with the face master file ID and the most similar cluster master member ID, and the "merged master file ID" replaces the "master file ID".

[0064] If the speed of the selected image in the candidate trajectory image set exceeds the speed threshold for both the speed corresponding to the previous image and the speed corresponding to the next image, then it is determined that the selected image has an abnormal situation, and all images with the most similar cluster master member ID corresponding to the selected image are deleted.

[0065] In one embodiment, the velocity corresponding to two adjacent images is calculated using Formula 1.

[0066]

[0067] In Formula 1: v represents the velocity between two adjacent images, in km / h; distance represents the distance between the acquisition positions of two adjacent images, in meters; t represents the time difference between the acquisition times of two adjacent images, in seconds.

[0068] In one specific embodiment, the distance between the acquisition positions corresponding to two adjacent images is calculated using Formula 2.

[0069] distance=6371000*acos(distance_angle) (Formula 2)

[0070] In Formula 2: distance represents the distance between the acquisition positions corresponding to two adjacent images; distance_angle represents the distance angle between the acquisition devices corresponding to the two images.

[0071] Specifically, the acquisition location includes the longitude and latitude data of the image acquisition device. The distance angle between two adjacent images is calculated using Formula 3.

[0072]

[0073] In Formula 3: distance_angle represents the distance angle between the acquisition devices corresponding to the two images; if distance_angle is greater than 1, it is set to 1; Lng1 represents the longitude data of the acquisition position of the previous image in two adjacent images; Lng2 represents the longitude data of the acquisition position of the next image in two adjacent images; Lat1 represents the latitude data of the acquisition position of the previous image in two adjacent images; Lat2 represents the latitude data of the acquisition position of the next image in two adjacent images.

[0074] In one specific embodiment, the time difference between the acquisition times of two adjacent images is calculated using Formula 4.

[0075] t=abs((next_time-start_time) / 1000) (Formula 4)

[0076] In Formula 4: t represents the time difference between the acquisition times of two adjacent images; start_time represents the acquisition time of the previous image in two adjacent images; next_time represents the acquisition time of the next image in two adjacent images.

[0077] The image merging method provided in this embodiment includes: acquiring multiple image sets; wherein, each image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location; based on the similarity between multiple target objects corresponding to the multiple image sets, determining the image sets corresponding to two target objects with a similarity greater than a similarity threshold as a first image set and a second image set; forming a candidate trajectory image set from the images in the first image set and the second image set according to the acquisition time of the images; and determining whether the first image set and the second image set should be merged based on the velocity between two adjacent images in the candidate trajectory image set. This application improves the accuracy of merging the first image set and the second image set by screening and determining multiple image sets of target objects with a similarity greater than a similarity threshold, forming a candidate trajectory image set from the images in the multiple image sets based on the acquisition time, and judging whether there are anomalies in the candidate trajectories by the velocity between two adjacent images.

[0078] Please see Figure 2 , Figure 4 This is a schematic block diagram of an embodiment of the image merging device provided in this application.

[0079] This embodiment provides an image merging device 60, which includes an acquisition module 61, a comparison module 62, an analysis module 63, and a merging module 64.

[0080] The acquisition module 61 is used to acquire multiple image sets; wherein, the image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location;

[0081] The comparison module 62 is used to determine the image sets corresponding to two target objects with a similarity greater than a similarity threshold as the first image set and the second image set based on the similarity between multiple target objects corresponding to multiple image sets respectively;

[0082] The analysis module 63 is used to form a candidate trajectory image set from the images in the first image set and the images in the second image set according to the image acquisition time;

[0083] The merging module 64 is used to determine whether the first image set and the second image set should be merged based on the velocity between two adjacent images in the candidate trajectory image set.

[0084] The image merging device provided in this embodiment includes: an acquisition module for acquiring multiple image sets; wherein each image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location; a comparison module for determining, based on the similarity between multiple target objects corresponding to the multiple image sets, image sets corresponding to two target objects with a similarity greater than a similarity threshold as a first image set and a second image set; an analysis module for forming a candidate trajectory image set from the images in the first image set and the images in the second image set according to the acquisition time of the images; and a merging module for determining whether the first image set and the second image set should be merged based on the velocity between two adjacent images in the candidate trajectory image set. This application improves the accuracy of merging the first image set and the second image set by filtering and determining multiple image sets of target objects with a similarity greater than a similarity threshold, forming candidate trajectory image sets from the images in the multiple image sets based on the acquisition time, and judging whether there are anomalies in the candidate trajectories by the velocity between two adjacent images.

[0085] See Figure 5 , Figure 5 This is a schematic block diagram of one embodiment of the terminal provided in this application. The terminal 70 in this embodiment includes: a processor 71, a memory 72, and a computer program stored in the memory 72 and executable on the processor 71. When the computer program is executed by the processor 71, it implements the above-described image merging method. To avoid repetition, it will not be described in detail here.

[0086] See Figure 6 , Figure 6 This is a schematic block diagram of one embodiment of the computer-readable storage medium provided in this application.

[0087] The embodiments of this application also provide a computer-readable storage medium 90, which stores a computer program 901. The computer program 901 includes program instructions, and the processor executes the program instructions to implement the image merging method provided in the embodiments of this application.

[0088] The computer-readable storage medium 90 can be an internal storage unit of the computer device described in the foregoing embodiments, such as a hard disk or memory of the computer device. Alternatively, the computer-readable storage medium 90 can be an external storage device of the computer device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device.

[0089] The above are merely embodiments of this application and do not limit the scope of patent protection of this application. Any equivalent structural or procedural changes made using the content of this application’s specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of this application.

Claims

1. An image merging method, characterized in that, The image merging method includes: Acquire multiple image sets; wherein, the image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location; Based on the similarity between the target objects corresponding to the multiple image sets, the image sets corresponding to the two target objects with similarity greater than a similarity threshold are determined as the first image set and the second image set. Based on the acquisition time of the images, the images in the first image set and the images in the second image set are combined to form a candidate trajectory image set; Based on the velocity between two adjacent images in the candidate trajectory image set, determine whether the first image set and the second image set should be merged. The step of determining whether to merge the first image set and the second image set based on the velocity between two adjacent images in the candidate trajectory image set includes: Based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set, the velocity corresponding to two adjacent images is determined; Determine whether the velocities corresponding to all two adjacent images in the candidate trajectory image set exceed a velocity threshold. If the speeds corresponding to all two adjacent images in the candidate trajectory image set do not exceed the speed threshold, then the first image set and the second image set are archived; wherein the second image set and the first image set correspond to the same target object.

2. The image merging method according to claim 1, characterized in that, The step of determining the velocity corresponding to two adjacent images based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set includes: The velocity corresponding to two adjacent images is calculated using Formula 1. In Formula 1: v represents the velocity corresponding to two adjacent images, in km / h; distance represents the distance between the acquisition positions corresponding to two adjacent images, in meters; t represents the time difference between the acquisition times corresponding to two adjacent images, in seconds.

3. The image merging method according to claim 2, characterized in that, The step of determining the velocity corresponding to two adjacent images based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set includes: The distance between the acquisition positions corresponding to two adjacent images is calculated using Formula 2. distance=6371000*acos(distance_angle) (Formula 2) In Formula 2: distance represents the distance between the acquisition positions corresponding to two adjacent images; distance_angle represents the distance angle between the acquisition devices corresponding to the two images.

4. The image merging method according to claim 3, characterized in that, The acquisition location includes the longitude and latitude data of the image acquisition device; The step of determining the velocity corresponding to two adjacent images based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set includes: The distance angle between two adjacent images is calculated using Formula 3. In Formula 3: distance_angle represents the distance angle between the acquisition devices corresponding to the two images; if distance_angle is greater than 1, it is set to 1; Lng1 represents the longitude data of the acquisition position of the first image in two adjacent images; Lng2 represents the longitude data of the acquisition position of the second image in two adjacent images; Lat1 represents the latitude data of the acquisition position of the first image in two adjacent images; Lat2 represents the latitude data of the acquisition position of the second image in two adjacent images.

5. The image merging method according to claim 2, characterized in that, The step of determining the velocity corresponding to two adjacent images based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set includes: The time difference between the acquisition times of two adjacent images is calculated using Formula 4. t=abs((next_time-start_time) / 1000) (Formula 4) In Formula 4: t represents the time difference between the acquisition times of two adjacent images; start_time represents the acquisition time of the previous image among two adjacent images; next_time represents the acquisition time of the next image among two adjacent images.

6. The image merging method according to claim 1, characterized in that, The step of forming a candidate trajectory image set from the images in the first image set and the images in the second image set according to the image acquisition time further includes: The images contained in the first image set and the images contained in the second image set are respectively subjected to denoising processing.

7. The image merging method according to claim 6, characterized in that, The step of performing noise reduction processing on the images contained in the first image set and the images contained in the second image set includes: Remove images from the first image set and the second image set that do not contain any of the following information: the identity data of the target object, the acquisition time of the image, the acquisition location of the image, or the acquisition device of the image.

8. The image merging method according to claim 1 or 6, characterized in that, The images in the first image set and the second image set respectively contain the identity data of the target object; The step of forming a candidate trajectory image set from the images in the first image set and the images in the second image set according to the image acquisition time further includes: Remove images from the first image set and the second image set that do not contain the identity data corresponding to the target object; Determine whether the merged identity data corresponding to the image contained in the first image set or the second image set is empty; If the merged identity data corresponding to the image is empty, then the merged identity data of the image is assigned to the identity data corresponding to the image.

9. An image merging device, characterized in that, The image merging device includes: An acquisition module is used to acquire multiple image sets; wherein, the image set includes multiple images of the same target object, and each image has an acquisition time and acquisition location; The comparison module is used to determine, based on the similarity between the multiple target objects corresponding to the multiple image sets respectively, the image sets corresponding to two target objects with a similarity greater than a similarity threshold as the first image set and the second image set; An analysis module is used to form a candidate trajectory image set from the images in the first image set and the images in the second image set according to the acquisition time of the images; The merging module is used to determine whether the first image set and the second image set should be merged based on the velocity between two adjacent images in the candidate trajectory image set; The merging module is further configured to determine the speed corresponding to two adjacent images based on the acquisition position and acquisition time corresponding to two adjacent images in the candidate trajectory image set; and to determine whether the speed corresponding to all two adjacent images in the candidate trajectory image set exceeds a speed threshold; if the speed corresponding to all two adjacent images in the candidate trajectory image set does not exceed the speed threshold, then the first image set and the second image set are merged; wherein the second image set and the first image set correspond to the same target object.

10. A terminal, characterized in that, The terminal includes a memory, a processor, and a computer program stored in the memory and running on the processor, the processor being configured to execute program data to implement the steps of the image merging method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the image merging method as described in any one of claims 1 to 8.

Images