Target re-identification method and device, electronic equipment and storage medium

Abstract

Embodiments of the present application relate to a target re-identification method and device, electronic equipment and storage medium. The method comprises: obtaining first and second images captured by a camera of a target vehicle; determining a disappearing target based on the first and second images, wherein the disappearing target is a target included in the first image and not included in the second image; obtaining an image of a to-be-identified target captured by the camera to obtain a to-be-identified image; and performing target re-identification on the to-be-identified target based on the determined disappearing target and the to-be-identified image. Thus, it can be determined whether a target in the all-around view environment of the target vehicle is a disappearing target in the all-around view environment of the target vehicle, thereby achieving target re-identification of the target in the all-around view environment of the target vehicle and improving the accuracy of target re-identification in the scenario of re-identifying the target in the all-around view environment of the target vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and particularly relates to a target re-identification method and device, electronic equipment and storage medium. BACKGROUND

[0002] Target re-identification, including pedestrian re-identification and vehicle re-identification, is a technology for judging whether a specific target exists in an image or a video sequence by using computer vision technology.

[0003] In the prior art, in the case of determining a specific target in an image to be identified, the correlation degree between the image to be identified and the image of the given specific target is low, resulting in low recognition accuracy of target re-identification in some scenarios. SUMMARY

[0004] In view of this, in order to solve the above-mentioned part or all technical problems, the embodiments of the present application provide a target re-identification method and device, electronic equipment and storage medium.

[0005] In a first aspect, the embodiments of the present application provide a target re-identification method, which comprises:

[0006] obtaining a first image and a second image of a target vehicle collected by a camera of the target vehicle, wherein the collection time of the second image is later than the collection time of the first image;

[0007] determining a disappeared target based on the first image and the second image, wherein the disappeared target is a target contained in the first image and not contained in the second image;

[0008] obtaining an image of a target to be identified collected by the camera, to obtain an image to be identified;

[0009] performing target re-identification on the target to be identified based on the determined disappeared target and the image to be identified.

[0010] In one possible implementation, the target to be identified is determined in the following manner:

[0011] obtaining a target image collected by the camera;

[0012] determining the position of the target in the target image relative to the image collection area of the camera, to obtain a target position;

[0013] determining whether the target in the target image is determined as the target to be identified based on the target position.

[0014] In one possible implementation, the determination of whether the target in the target image is determined as the target to be identified based on the target position comprises:

[0015] When the target location represents the boundary location of the image acquisition area, it is determined that there is no need to identify the target in the target image as the target to be identified;

[0016] If the target location does not represent the boundary location, the target in the target image is identified as the target to be identified.

[0017] In one possible implementation, determining the disappearing target based on the first image and the second image includes:

[0018] Based on the first image and the second image, the disappearing target and its feature data are determined;

[0019] The target identifier and feature data of the disappeared target are associated and added to a preset association table, wherein the preset association table represents the association relationship between the target identifier and feature data of the disappeared target; and

[0020] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0021] Determine whether the preset association table includes target feature data, wherein the target feature data is feature data that matches the target to be identified;

[0022] If the target feature data is included in the preset association table, the target identifier of the disappeared target associated with the target feature data in the preset association table is determined as the target identifier of the target to be identified.

[0023] In one possible implementation, the feature data includes at least one of the following:

[0024] The target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, wherein the re-identification confidence is negatively correlated with the target disappearance duration.

[0025] In one possible implementation, the step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0026] Determine whether the disappeared target and the target to be identified are the same target, so as to perform target re-identification on the target to be identified.

[0027] In one possible implementation, the target vehicle has two or more cameras; and

[0028] The step of determining the disappearing target based on the first image and the second image includes:

[0029] Based on the first image and the second image, a disappearing target is determined, and the disappearance type of the disappearing target is determined, wherein the disappearance type indicates whether the disappearance location of the disappearing target is located within the shared field of view of the two cameras; and

[0030] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0031] Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

[0032] Secondly, embodiments of this application provide a target re-identification device, the device comprising:

[0033] The first acquisition unit is used to acquire a first image and a second image captured by the camera of the target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image;

[0034] A determining unit is configured to determine a disappearing target based on the first image and the second image; wherein the disappearing target is a target contained in the first image but not contained in the second image;

[0035] The second acquisition unit is used to acquire the image of the target to be identified captured by the camera, and obtain the image to be identified;

[0036] The identification unit is used to perform target re-identification of the target to be identified based on the determined disappeared target and the image to be identified.

[0037] In one possible implementation, the target to be identified is determined in the following manner:

[0038] Acquire the target image captured by the camera;

[0039] The position of the target in the target image relative to the image acquisition area of ​​the camera is determined to obtain the target position;

[0040] Based on the target location, determine whether to identify the target in the target image as the target to be identified.

[0041] In one possible implementation, determining whether to identify a target in the target image as a target to be identified based on the target location includes:

[0042] When the target location represents the boundary location of the image acquisition area, it is determined that there is no need to identify the target in the target image as the target to be identified;

[0043] If the target location does not represent the boundary location, the target in the target image is identified as the target to be identified.

[0044] In one possible implementation, determining the disappearing target based on the first image and the second image includes:

[0045] Based on the first image and the second image, the disappearing target and its feature data are determined;

[0046] The target identifier and feature data of the disappeared target are associated and added to a preset association table, wherein the preset association table represents the association relationship between the target identifier and feature data of the disappeared target; and

[0047] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0048] Determine whether the preset association table includes target feature data, wherein the target feature data is feature data that matches the target to be identified;

[0049] If the target feature data is included in the preset association table, the target identifier of the disappeared target associated with the target feature data in the preset association table is determined as the target identifier of the target to be identified.

[0050] In one possible implementation, the feature data includes at least one of the following:

[0051] The target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, wherein the re-identification confidence is negatively correlated with the target disappearance duration.

[0052] In one possible implementation, the step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0053] Determine whether the disappeared target and the target to be identified are the same target, so as to perform target re-identification on the target to be identified.

[0054] In one possible implementation, the target vehicle has two or more cameras; and

[0055] The step of determining the disappearing target based on the first image and the second image includes:

[0056] Based on the first image and the second image, a disappearing target is determined, and the disappearance type of the disappearing target is determined, wherein the disappearance type indicates whether the disappearance location of the disappearing target is located within the shared field of view of the two cameras; and

[0057] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0058] Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

[0059] Thirdly, embodiments of this application provide an electronic device, including:

[0060] Memory, used to store computer programs;

[0061] A processor is configured to execute a computer program stored in the memory, wherein, when the computer program is executed, it implements the method of any embodiment of the target re-identification method of the first aspect of this application.

[0062] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the method of any embodiment of the target re-identification method of the first aspect described above.

[0063] Fifthly, embodiments of this application provide a computer program comprising computer-readable code, which, when executed on a device, causes a processor in the device to implement the method of any embodiment of the target re-identification method of the first aspect described above.

[0064] The target re-identification method provided in this application can acquire a first image and a second image captured by a camera of a target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image. Then, based on the first image and the second image, a vanishing target is determined; wherein the vanishing target is a target included in the first image but not included in the second image. Then, an image of the target to be identified, captured by the camera, is acquired to obtain an image to be identified. Subsequently, based on the determined vanishing target and the image to be identified, target re-identification is performed on the target to be identified. Thus, it is possible to determine whether a target in the target vehicle's surrounding environment is a vanishing target in the target vehicle's surrounding environment, thereby achieving target re-identification in the target vehicle's surrounding environment and improving the accuracy of target re-identification in scenarios involving target re-identification in the target vehicle's surrounding environment. Attached Figure Description

[0065] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0066] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0067] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0068] Figure 1 A flowchart illustrating a target re-identification method provided in an embodiment of this application;

[0069] Figure 2 A flowchart illustrating another target re-identification method provided in this application embodiment;

[0070] Figures 3A-3C This is an application scenario diagram of a target re-identification method provided in an embodiment of this application;

[0071] Figure 4 This is a schematic diagram of the structure of a target re-identification device provided in an embodiment of this application;

[0072] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0073] Various exemplary embodiments of this application will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this application.

[0074] Those skilled in the art will understand that the terms "first" and "second" in the embodiments of this application are only used to distinguish different steps, devices or modules, and do not represent any specific technical meaning, nor do they indicate the logical order between them.

[0075] It should also be understood that in this embodiment, "multiple" can refer to two or more, and "at least one" can refer to one, two or more.

[0076] It should also be understood that any component, data or structure mentioned in the embodiments of this application can generally be understood as one or more unless explicitly defined or given contrary guidance in the context.

[0077] Furthermore, the term "and / or" in this application is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this application generally indicates that the preceding and following related objects have an "or" relationship.

[0078] It should also be understood that the description of the various embodiments in this application emphasizes the differences between the various embodiments, and the similarities or similarities can be referred to each other. For the sake of brevity, they will not be described in detail.

[0079] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the scope of this application and its application or use.

[0080] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0081] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0082] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. To facilitate understanding of the embodiments of this application, the application will be described in detail below with reference to the accompanying drawings and embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0083] To address the technical problem of low accuracy in target re-identification in existing technologies, this application provides a target re-identification method that can improve the accuracy of target re-identification in scenarios where targets in the panoramic environment of a target vehicle are re-identified.

[0084] Figure 1This is a flowchart illustrating a target re-identification method provided in an embodiment of this application. This method can be applied to one or more electronic devices such as smartphones, laptops, desktop computers, portable computers, and servers. Furthermore, the execution entity of this method can be hardware or software. When the execution entity is hardware, it can be one or more of the aforementioned electronic devices. For example, a single electronic device can execute this method, or multiple electronic devices can cooperate with each other to execute this method. When the execution entity is software, this method can be implemented as multiple software programs or software modules, or as a single software program or software module. No specific limitations are imposed here.

[0085] like Figure 1 As shown, the method specifically includes:

[0086] Step 101: Acquire the first image and the second image captured by the camera of the target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image.

[0087] In this embodiment, the target vehicle can be any vehicle.

[0088] The target vehicle's cameras can be one or more cameras installed on the target vehicle. For example, the target vehicle may have six cameras. These cameras can be used to capture images of the target vehicle from the following directions: directly in front, directly behind, left front, left rear, right front, and right rear.

[0089] The first image and the second image can be two different images captured by the aforementioned camera. For example, the first image and the second image can be images, video clips, etc.

[0090] The first and second images can be obtained from the same camera or from different cameras.

[0091] Step 102: Based on the first image and the second image, determine the disappearing target; wherein the disappearing target is: a target contained in the first image and not contained in the second image.

[0092] In this embodiment, the disappearing target can be a person or a vehicle.

[0093] As an example, target detection can be performed first on the first image to identify targets contained within it. Then, for each target contained in the first image, it can be determined whether that target is present in the second image. If the target is not present in the second image, it can be identified as a vanished target.

[0094] Step 103: Obtain the image of the target to be identified captured by the camera to obtain the image to be identified.

[0095] In this embodiment, the image to be identified can be the image of the target to be identified captured by the camera.

[0096] Here, the first image, the second image, and the image to be identified can all be obtained by the same or different cameras of the target vehicle.

[0097] In some cases, the target to be identified can be determined in the following ways:

[0098] First, a third image captured by the camera is acquired. The acquisition time of the third image is after the acquisition time of the second image.

[0099] Subsequently, targets contained in the third image but not in the second image are identified as targets to be identified.

[0100] Step 104: Based on the determined disappeared target and the image to be identified, perform target re-identification on the target to be identified.

[0101] In this embodiment, it can be determined whether the missing target is a target to be identified in the image to be identified, so as to perform target re-identification on the target to be identified.

[0102] In some optional implementations of this embodiment, the target to be identified is determined in the following manner:

[0103] First, acquire the target image captured by the camera.

[0104] The target image can be any image captured by the aforementioned cameras.

[0105] Next, the position of the target in the target image relative to the image acquisition area of ​​the camera is determined to obtain the target position.

[0106] The target location can be the position of the target in the target image relative to the image acquisition area of ​​the camera.

[0107] Then, based on the target location, it is determined whether to identify the target in the target image as the target to be identified.

[0108] For example, if the target location indicates that the target in the target image is located at a preset location in the aforementioned image acquisition area, then the target in the target image can be identified as the target to be identified; if the target location indicates that the target in the target image is not located at the preset location in the aforementioned image acquisition area, then it is not necessary to identify the target in the target image as the target to be identified.

[0109] It is understood that, in the above optional implementation methods, the position of the target in the target image relative to the image acquisition area of ​​the camera can be used to determine whether the target in the target image is identified as the target to be identified. In this way, it is not necessary to re-identify all targets in the image acquired by the camera, thus improving the efficiency of re-identifying objects that need to be re-identified.

[0110] In some application scenarios of the above-mentioned optional implementation methods, the following method can be used to determine whether to identify the target in the target image as the target to be identified based on the target location:

[0111] If the target location represents the boundary of the image acquisition area, it is determined that there is no need to identify the target in the target image as the target to be identified; if the target location does not represent the boundary location, the target in the target image is identified as the target to be identified.

[0112] It is understandable that if a target in the image is located at the boundary of the image acquisition area, the target usually entered the camera's field of view recently (e.g., just entered the camera's field of view). In this scenario, the target is usually not a vanishing target and therefore does not need to be considered as a target to be identified. However, if a target in the image is not located at the boundary of the image acquisition area, the target has usually already entered the camera's field of view, and due to some reason (e.g., obstruction), the target was not tracked. In this scenario, the target is likely a vanishing target and can therefore be considered as a target to be identified. Thus, in the above application scenarios, targets with a high probability of being vanishing targets can be used as targets to be identified and then re-identified, thereby further improving the accuracy of target re-identification.

[0113] The aforementioned boundary region can be determined based on the moving speed of the target to be identified. For example, the area of ​​the boundary region can be positively correlated with the moving speed of the target to be identified.

[0114] In some optional implementations of this embodiment, the target to be identified can be re-identified based on the determined disappeared target and the image to be identified in the following manner:

[0115] Determine whether the disappeared target and the target to be identified are the same target, so as to perform target re-identification on the target to be identified.

[0116] Specifically, it can be determined whether the disappeared target and the target to be identified are the same target by calculating the similarity between the feature data of the disappeared target and the feature data of the target to be identified.

[0117] Alternatively, the images of the vanished target and the target to be identified can be input into a pre-trained re-identification model to determine whether the vanished target and the target to be identified are the same target. The re-identification model can be used to determine whether two targets contained in the two input images are the same target.

[0118] It is understood that in the above optional implementation methods, target re-identification of the target to be identified is achieved by determining whether the disappeared target and the target to be identified are the same target. This can further improve the accuracy of target re-identification.

[0119] In some optional implementations of this embodiment, the target vehicle has two or more cameras.

[0120] Based on this, the disappearing target can be determined using the first image and the second image in the following manner:

[0121] Based on the first image and the second image, the disappearing target and the disappearance type of the disappearing target are determined.

[0122] The disappearance type indicates whether the disappearance location of the target is located within the shared field of view of the two cameras.

[0123] Here, the vanishing position can refer to the last position of the vanishing target in the image captured by the camera before it disappeared.

[0124] Based on this, further, the target to be identified can be re-identified using the determined disappeared target and the image to be identified, in the following manner:

[0125] Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

[0126] As an example, the disappearance types of each identified disappearing target can be statistically analyzed. Therefore, when re-identifying the target to be identified, the position of the target relative to the image acquisition area of ​​the camera can be determined first. Then, based on this position, the last appearance position of the target in the image before disappearance can be determined. Furthermore, based on the last appearance position of the target in the image before disappearance, the disappearance type of the target is determined, where the disappearance type indicates whether the disappearance position of the target is located within the shared field of view of the two cameras. Then, among the statistically obtained disappearing targets, those with the same disappearance type as the target to be identified are identified. Further, for the identified disappearing targets, it is determined whether the disappearing target and the target to be identified are the same target, in order to perform target re-identification of the target to be identified.

[0127] It is understood that, among the above optional implementation methods, the target to be identified can be re-identified based on the disappearance type, the determined disappearance target, and the target to be identified, thereby improving the efficiency and accuracy of target re-identification.

[0128] The target re-identification method provided in this application can acquire a first image and a second image captured by a camera of a target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image. Then, based on the first image and the second image, a vanishing target is determined; wherein the vanishing target is a target included in the first image but not included in the second image. Then, an image of the target to be identified, captured by the camera, is acquired to obtain an image to be identified. Subsequently, based on the determined vanishing target and the image to be identified, target re-identification is performed on the target to be identified. Thus, it is possible to determine whether a target in the target vehicle's surrounding environment is a vanishing target in the target vehicle's surrounding environment, thereby achieving target re-identification in the target vehicle's surrounding environment and improving the accuracy of target re-identification in scenarios involving target re-identification in the target vehicle's surrounding environment.

[0129] Figure 2 This is a flowchart illustrating another target re-identification method provided in an embodiment of this application. Figure 2 As shown, the method specifically includes:

[0130] Step 201: Acquire a first image and a second image captured by the camera of the target vehicle, wherein the second image is acquired after the first image is acquired.

[0131] In this embodiment, step 201 and Figure 1 Step 101 in the corresponding embodiment is basically the same, and will not be repeated here.

[0132] Step 202: Based on the first image and the second image, determine the disappearing target and the feature data of the disappearing target; wherein, the disappearing target is: a target included in the first image and not included in the second image.

[0133] In this embodiment, the disappearing target can be a person who has disappeared or a target that has disappeared.

[0134] As an example, target detection can be performed first on the first image to identify targets contained within it. Then, for each target contained in the first image, it can be determined whether that target is present in the second image. If the target is not present in the second image, it can be identified as a vanished target.

[0135] The feature data of a vanishing target can represent the characteristics of the vanishing target. As an example, the feature data may include at least one of the following: the target's color, the target's length, the target's width, the target's height, the target's type, etc.

[0136] Step 203: Add the target identifier and feature data of the disappeared target to a preset association table, wherein the preset association table represents the association between the target identifier and feature data of the disappeared target.

[0137] Step 204: Obtain the image of the target to be identified captured by the camera to obtain the image to be identified.

[0138] In this embodiment, step 204 and Figure 1 Step 103 in the corresponding embodiment is basically the same, and will not be repeated here.

[0139] Step 205: Determine whether the preset association table includes target feature data, wherein the target feature data is feature data that matches the target to be identified.

[0140] In this embodiment, the target feature data can be feature data in a preset association table whose similarity to the feature data of the target to be identified is greater than or equal to a preset similarity threshold.

[0141] Step 206: If the target feature data is included in the preset association table, the target identifier of the disappeared target associated with the target feature data in the preset association table is determined as the target identifier of the target to be identified.

[0142] In some optional implementations of this embodiment, the feature data includes at least one of the following:

[0143] The target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, wherein the re-identification confidence is negatively correlated with the target disappearance duration.

[0144] It is understood that, among the above optional implementation methods, the target to be identified can be re-identified based on at least one of the target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, thereby further improving the accuracy of target re-identification.

[0145] It should be noted that, in addition to the contents described above, this embodiment may also include... Figure 1 The corresponding technical features described in the corresponding embodiments, thereby achieving Figure 1 For details on the technical effectiveness of the target re-identification method shown, please refer to [link / reference]. Figure 1 The relevant descriptions are presented concisely and will not be elaborated upon here.

[0146] The target re-identification method provided in this application maintains and updates relevant information of disappeared targets by using a preset association table. In this way, by dynamically updating the preset association table, the accuracy of target re-identification can be improved, and a better re-identification decision can be obtained.

[0147] The embodiments of this application are described below by way of example. However, it should be noted that the embodiments of this application may have the features described below, but the following description does not constitute a limitation on the protection scope of the embodiments of this application.

[0148] Existing target re-identification schemes do not address target re-identification in a panoramic environment, nor do they track targets that are lost and reappear without cause in a panoramic environment, nor do they dynamically update the re-identification table (i.e., the aforementioned pre-set association table) to obtain re-identification decisions.

[0149] This method includes the following:

[0150] 1. Within the specified re-id range (i.e., the aforementioned image acquisition area, or a portion thereof), a process for re-identifying vehicle panoramic targets is proposed, specifically including:

[0151] 1) Determine the instantaneously lost ID(s), that is, the aforementioned disappeared target, and add the targets that meet the requirements to the table, that is, the aforementioned preset association table.

[0152] 2) Identify the id(s) that are added without reason, which are the targets to be identified above, and perform re-identification matching according to the table.

[0153] 2. In this method, the re-identification and discrimination attributes (i.e. the feature data mentioned above) include position, velocity, length, width and height, and high-level features.

[0154] Re-identification can effectively solve two types of problems: 1. Vehicle targets (i.e., the targets mentioned above, such as disappearing targets or targets to be identified) are tracked in a designated camera (i.e., the cameras mentioned above) but disappear and reappear without reason (the problem of vehicle ID flickering in a single camera). 2. The problem of vehicle targets crossing cameras through non-overlapping fields of view (i.e., non-common viewing areas) and the ID changing in adjacent cameras.

[0155] Please refer to Figures 3A-3C , Figures 3A-3C This is an application scenario diagram of a target re-identification method provided in an embodiment of this application.

[0156] 1. Determine the re-id range:

[0157] The rule is that re-iding is only required when there is a target (instantaneously lost or unexpectedly added) within the re-id range x and re-id range y of the vehicle coordinate system. Figure 3B As shown in the figure. In this method, the re-identification range in the x-direction is set to 50 meters, and the re-identification range in the y-direction is set to 20 meters. Targets added to the table due to loss or added to the table must be within the specified range. Re-identification of distant targets outside the specified range is not considered in this scheme.

[0158] 2. Determine the ID(s) to be deleted:

[0159] There are three possible reasons why the original IDs of the six cameras in the pan may disappear without explanation.

[0160] 1) The tracked vehicle is moving away from the camera and crossing the overlapping field of view (i.e., the shared field of view mentioned above).

[0161] 2) The tracked vehicle is moving away from the camera and crossing through a non-overlapping field of view.

[0162] 3) The ID of the vehicle tracked by the single camera is lost, either within the range of re-id or outside the range of re-id.

[0163] There are two situations where re-id is needed, such as... Figure 3A As shown. Therefore, it is only necessary to save the IDs of those that disappear without reason in the following two situations, and organize and save the targets that disappear without reason to a table.

[0164] 1) Within the re-id range, the ID of the vehicle tracked by the single camera is lost.

[0165] 2) The tracking vehicle is moving away from the camera and crossing through a non-overlapping field of view.

[0166] Specific implementation method: Within the range, compare the existing camera target ID(s) in the current frame (i.e., the second image mentioned above) with all camera target IDs-before(s) in the previous frame (i.e., the first image mentioned above) to determine the ID that disappeared without cause, i.e., the disappeared target.

[0167] 3. Identify id(s) that are added without cause:

[0168] This step is consistent with confirming the ID(s) to be deleted. Within the range, the existing camera target IDs in the current frame (i.e., the third image mentioned above) are compared with all camera target IDs before in the previous frame (i.e., the second image mentioned above) to identify any IDs that have been added without cause, thereby obtaining the target to be identified.

[0169] 4. Dynamically update the re-identification table:

[0170] 1) Velocity update (updates based on x / y acceleration; if no velocity update occurs, it moves at a constant speed in the specified direction using the x / y velocity at the moment it disappears).

[0171] 2) Position update (updates based on x / y velocity).

[0172] 3) Camera prediction update (distinguishing between boundary disappearance targets and targets that disappear without cause in a single camera), which is the disappearance type mentioned above.

[0173] 4) Update the disappearance time (define the disappearance time as 10).

[0174] 5) Confidence update (the longer the disappearance time, the lower the confidence).

[0175] 5. Re-identification Decision Indicators and Re-identification:

[0176] The five key decision indicators for re-identification are as follows:

[0177] 1) The model's detection output is incorporated into the re-identification decision (vehicle attributes, head / tail discrimination, occlusion level). The model's input can be an image, and its output can include the location of the target's bounding box and the target's category.

[0178] 2) The introduction of the target vehicle's position status allows for the prediction of the next frame's position based on the current position and relative displacement.

[0179] 3) Introduction of vehicle length, width and height (the length, width and height of a vehicle are constant attributes).

[0180] 4) Introduction of basic features and high-level vehicle feature vectors (1. Vehicle color information; 2. HOG feature operator in the case of no occlusion).

[0181] 5) Model-based decision-making for re-identifying feature vectors:

[0182] The matching process is as follows Figure 3C As shown, the vehicle to be re-identified is matched by iterating through all vehicles in the re-identification table. The matching method is a weighted summation method. The matching coefficient is set based on experience.

[0183] For example, we can define [location distance, vehicle attribute difference, length, width, and height difference, basic feature difference, feature distance] = [d1, d2, d3, d4, d5]. Here, the location distance and length, width, and height are calculated using Euclidean distance, while the feature distance is calculated using cosine distance, with corresponding matching coefficients of [γ1, γ2, γ3, γ4, γ5]. The evaluation metric is: Σ(1-d(i))*γ(i), where i represents an integer from 1 to 5.

[0184] It should be noted that, in addition to the contents described above, this embodiment may also include the technical features described in the above embodiments, thereby achieving the technical effect of the target re-identification method shown above. For details, please refer to the above description. For the sake of brevity, it will not be elaborated here.

[0185] The target re-identification method provided in this application embodiment can determine the lost ID(s) (i.e., the aforementioned disappeared target), add the target that meets the requirements to the table, determine the added ID(s), i.e. the aforementioned target to be identified, and perform re-identification matching by referring to the table. This method can achieve relatively accurate target re-identification in two video frames with an interval of about 5-10 frames (i.e., the time between the aforementioned second image and the third image may include 5-10 video frames). Furthermore, it can be applied to target re-identification in the following situations: the vehicle target is lost and then reappears in the designated camera (the vehicle ID flashes in a single camera), and the vehicle target crosses cameras through a non-overlapping field of view, and the ID changes in adjacent cameras.

[0186] Figure 4 This is a schematic diagram of the structure of a target re-identification device provided in an embodiment of this application. Specifically, it includes:

[0187] The first acquisition unit 401 is used to acquire a first image and a second image captured by the camera of the target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image.

[0188] The determining unit 402 is configured to determine a disappearing target based on the first image and the second image; wherein the disappearing target is a target that is included in the first image but not included in the second image;

[0189] The second acquisition unit 403 is used to acquire the image of the target to be identified captured by the camera, and obtain the image to be identified;

[0190] The identification unit 404 is used to perform target re-identification of the target to be identified based on the determined disappeared target and the image to be identified.

[0191] In one possible implementation, the target to be identified is determined in the following manner:

[0192] Acquire the target image captured by the camera;

[0193] The position of the target in the target image relative to the image acquisition area of ​​the camera is determined to obtain the target position;

[0194] Based on the target location, determine whether to identify the target in the target image as the target to be identified.

[0195] In one possible implementation, determining whether to identify a target in the target image as a target to be identified based on the target location includes:

[0196] When the target location represents the boundary location of the image acquisition area, it is determined that there is no need to identify the target in the target image as the target to be identified;

[0197] If the target location does not represent the boundary location, the target in the target image is identified as the target to be identified.

[0198] In one possible implementation, determining the disappearing target based on the first image and the second image includes:

[0199] Based on the first image and the second image, the disappearing target and its feature data are determined;

[0200] The target identifier and feature data of the disappeared target are associated and added to a preset association table, wherein the preset association table represents the association relationship between the target identifier and feature data of the disappeared target; and

[0201] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0202] Determine whether the preset association table includes target feature data, wherein the target feature data is feature data that matches the target to be identified;

[0203] If the target feature data is included in the preset association table, the target identifier of the disappeared target associated with the target feature data in the preset association table is determined as the target identifier of the target to be identified.

[0204] In one possible implementation, the feature data includes at least one of the following:

[0205] The target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, wherein the re-identification confidence is negatively correlated with the target disappearance duration.

[0206] In one possible implementation, the step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0207] Determine whether the disappeared target and the target to be identified are the same target, so as to perform target re-identification on the target to be identified.

[0208] In one possible implementation, the target vehicle has two or more cameras; and

[0209] The step of determining the disappearing target based on the first image and the second image includes:

[0210] Based on the first image and the second image, a disappearing target is determined, and the disappearance type of the disappearing target is determined, wherein the disappearance type indicates whether the disappearance location of the disappearing target is located within the shared field of view of the two cameras; and

[0211] The step of re-identifying the target based on the determined disappeared target and the image to be identified includes:

[0212] Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

[0213] The target re-identification device provided in this embodiment can be as follows: Figure 4 The target re-identification device shown can execute all the steps of the target re-identification methods described above, thereby achieving the technical effects of the target re-identification methods described above. For details, please refer to the relevant descriptions above. For the sake of brevity, it will not be elaborated here.

[0214] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 The illustrated electronic device 500 includes at least one processor 501, a memory 502, at least one network interface 504, and other user interfaces 503. The various components in the electronic device 500 are coupled together via a bus system 505. It is understood that the bus system 505 is used to implement communication between these components. In addition to a data bus, the bus system 505 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 5The general designated all buses as Bus System 505.

[0215] The user interface 503 may include a display, keyboard, or clicking device (e.g., mouse, trackball, touchpad, or touchscreen).

[0216] It is understood that the memory 502 in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 502 described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0217] In some implementations, memory 502 stores elements, executable units or data structures, or subsets thereof, or extended sets thereof: operating system 5021 and application program 5022.

[0218] The operating system 5021 includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 5022 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this application embodiment can be included in application program 5022.

[0219] In this embodiment, by calling the program or instructions stored in memory 502, specifically the program or instructions stored in application program 5022, processor 501 executes the method steps provided in each method embodiment, including, for example:

[0220] Acquire a first image and a second image captured by a camera on the target vehicle, wherein the second image is acquired after the first image is acquired.

[0221] Based on the first image and the second image, a disappearing target is determined; wherein, the disappearing target is: a target contained in the first image and not contained in the second image;

[0222] The image of the target to be identified is obtained by acquiring the image captured by the camera.

[0223] Based on the identified disappeared target and the image to be identified, the target to be identified is re-identified.

[0224] The methods disclosed in the embodiments of this application can be applied to or implemented by processor 501. Processor 501 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit of the hardware in processor 501 or by instructions in the form of software. The processor 501 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software units in the decoding processor. The software units may be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 502. Processor 501 reads the information in memory 502 and, in conjunction with its hardware, completes the steps of the above method.

[0225] It is understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described above in this application, or combinations thereof.

[0226] For software implementation, the techniques described herein can be implemented by units that perform the functions described above. The software code can be stored in memory and executed by a processor. The memory can be implemented within the processor or external to the processor.

[0227] The electronic device provided in this embodiment may be as follows: Figure 5 The electronic device shown can execute all the steps of the target re-identification methods described above, thereby achieving the technical effects of the target re-identification methods described above. For details, please refer to the relevant descriptions above. For the sake of brevity, it will not be elaborated here.

[0228] This application also provides a storage medium (computer-readable storage medium). This storage medium stores one or more programs. The storage medium may include volatile memory, such as random access memory; it may also include non-volatile memory, such as read-only memory, flash memory, hard disk, or solid-state drive; and it may also include combinations of the above types of memory.

[0229] When one or more programs in the storage medium can be executed by one or more processors to implement the target re-identification method executed on the electronic device side.

[0230] The processor described above is used to execute the target re-identification program stored in the memory to implement the following steps of the target re-identification method executed on the electronic device side:

[0231] Acquire a first image and a second image captured by a camera on the target vehicle, wherein the second image is acquired after the first image is acquired.

[0232] Based on the first image and the second image, a disappearing target is determined; wherein, the disappearing target is: a target contained in the first image and not contained in the second image;

[0233] The image of the target to be identified is obtained by acquiring the image captured by the camera.

[0234] Based on the identified disappeared target and the image to be identified, the target to be identified is re-identified.

[0235] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0236] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0237] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0238] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A target re-identification method, characterized in that, The method includes: Acquire a first image and a second image captured by a camera on the target vehicle, wherein the second image is acquired after the first image is acquired. Based on the first image and the second image, a disappearing target is determined; wherein, the disappearing target is: a target contained in the first image and not contained in the second image; The image of the target to be identified is obtained by acquiring the image captured by the camera. Based on the identified disappeared target and the image to be identified, the target to be identified is re-identified; The target vehicle has two or more cameras; and The step of determining the disappearing target based on the first image and the second image includes: Based on the first image and the second image, a disappearing target is determined, and the disappearance type of the disappearing target is determined, wherein the disappearance type indicates whether the disappearance location of the disappearing target is located within the shared field of view of the two cameras; and The step of re-identifying the target based on the determined disappeared target and the image to be identified includes: Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

2. The method according to claim 1, characterized in that, The target to be identified is determined in the following manner: Acquire the target image captured by the camera; The position of the target in the target image relative to the image acquisition area of ​​the camera is determined to obtain the target position; Based on the target location, determine whether to identify the target in the target image as the target to be identified.

3. The method according to claim 2, characterized in that, The step of determining whether to identify a target in the target image as a target to be identified based on the target location includes: When the target location represents the boundary location of the image acquisition area, it is determined that there is no need to identify the target in the target image as the target to be identified; If the target location does not represent the boundary location, the target in the target image is identified as the target to be identified.

4. The method according to claim 1, characterized in that, The step of determining the disappearing target based on the first image and the second image includes: Based on the first image and the second image, the disappearing target and its feature data are determined; The target identifier and feature data of the disappeared target are associated and added to a preset association table, wherein the preset association table represents the association relationship between the target identifier and feature data of the disappeared target; and The step of re-identifying the target based on the determined disappeared target and the image to be identified includes: Determine whether the preset association table includes target feature data, wherein the target feature data is feature data that matches the target to be identified; If the target feature data is included in the preset association table, the target identifier of the disappeared target associated with the target feature data in the preset association table is determined as the target identifier of the target to be identified.

5. The method according to claim 4, characterized in that, Feature data includes at least one of the following: The target disappearance speed, target disappearance location, target disappearance type, target disappearance duration, and re-identification confidence, wherein the re-identification confidence is negatively correlated with the target disappearance duration.

6. The method according to any one of claims 1-5, characterized in that, The step of re-identifying the target based on the determined disappeared target and the image to be identified includes: Determine whether the disappeared target and the target to be identified are the same target, so as to perform target re-identification on the target to be identified.

7. A target re-identification device, characterized in that, The device includes: The first acquisition unit is used to acquire a first image and a second image captured by the camera of the target vehicle, wherein the acquisition time of the second image is after the acquisition time of the first image; A determining unit is configured to determine a disappearing target based on the first image and the second image; wherein the disappearing target is a target contained in the first image but not contained in the second image; The second acquisition unit is used to acquire the image of the target to be identified captured by the camera, and obtain the image to be identified; The identification unit is used to perform target re-identification of the target to be identified based on the determined disappeared target and the image to be identified; The target vehicle has two or more cameras; and The step of determining the disappearing target based on the first image and the second image includes: Based on the first image and the second image, a disappearing target is determined, and the disappearance type of the disappearing target is determined, wherein the disappearance type indicates whether the disappearance location of the disappearing target is located within the shared field of view of the two cameras; and The step of re-identifying the target based on the determined disappeared target and the image to be identified includes: Based on the disappearance type, the determined disappearance target, and the target to be identified, the target to be identified is re-identified.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing a computer program stored in the memory, wherein when the computer program is executed, it implements the method described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-6.

Images