MAV airborne target tracking method and system based on adaptive dynamic template

Abstract

The invention discloses an MAV airborne target tracking method and system based on an adaptive dynamic template, belongs to the technical field of target tracking, and solves the problem that tracking performance and calculation complexity are difficult to balance in MAV airborne tracking. The method comprises the following steps: setting an initial template, an adjacent template and a memory template, and inputting a current frame search feature, the initial template, the adjacent template and the memory template into an adaptive template fusion STF module to generate a final template; performing related operation on the final template and the search template to obtain a response diagram, judging a current tracking state, and updating the adjacent template and the memory template; the memory template generation module extracts and integrates key information in historical tracking results in a time series connection mode, so that a limited template feature memory can contain all historical information. And the adaptive template fusion module utilizes a similarity matrix between the template and the search features to realize dynamic adjustment of the template weight in different tracking stages. The method is suitable for MAV airborne target tracking.

Description

Technical Field

[0001] The present application relates to the field of target tracking technology, and in particular to MAV airborne target tracking. Background Art

[0002] With the development of micro-nanotechnology and system integration technology, micro unmanned aerial vehicles (MAVs) technology has made significant progress, and a series of new MAVs such as "Black Hornet 3", "MetaFly" and "Drone40" have been launched. Among them, the "BlackHornet 3" drone adopts a modular design and is equipped with infrared imaging and visible light cameras. Although it can transmit video and image data in real time, it still requires human intervention during the execution of tasks, and its autonomy and intelligence levels are relatively low. This limitation will distract the operator's attention, reduce their decision-making ability, increase the risk of misoperation, and may even cause the operator to miss opportunities or fall into dangerous situations. Therefore, improving the autonomy and intelligence level of MAVs and realizing single-target tracking with onboard processing are crucial to improving execution efficiency and response speed. More and more researchers are integrating drone systems with target tracking algorithms to improve their perception capabilities in complex environments.

[0003] However, the high maneuverability of MAVs, the complex flight environment, and the limited onboard computing resources pose challenges to MAV airborne tracking. First, the high maneuverability of MAVs often leads to dramatic changes in the camera's viewing angle, resulting in significant changes in the target's shape, position, and projection in the image. When such visual changes occur, there will be a serious mismatch between the target's current appearance model and the model learned by the tracking algorithm, which will cause tracking failure. Secondly, in complex urban or wilderness environments, challenges such as interference and occlusion of similar objects will arise. When the tracker identifies an interference object or encounters severe occlusion, erroneous information continues to accumulate in the video stream, eventually leading to tracking failure. In addition, current algorithms with strong anti-interference capabilities often also have a high number of parameters and computational complexity, while MAVs have strict volume and load restrictions, and the computing devices they can carry are very limited in computing power, storage space, and energy consumption, which further limits the application of algorithms on MAV platforms.

[0004] Although much progress has been made in the research of target tracking methods, target tracking specifically for MAVs is very limited. Existing MAV airborne tracking methods are still at the stage of correlation filtering algorithms. Correlation filtering algorithms have relatively poor robustness when faced with complex tracking scenarios such as background clutter and target occlusion. Once features similar to the target appear in the background, or the target is partially or even completely occluded, it is difficult for the tracker to identify the target from the background, resulting in tracking deviation or loss. Summary of the invention

[0005] The purpose of the present invention is to solve the problem of difficulty in balancing tracking performance and computational complexity in MAV airborne tracking, and to provide a MAV airborne target tracking method and system based on an adaptive dynamic template.

[0006] The present invention is implemented by the following technical solutions. On the one hand, the present invention provides a MAV airborne target tracking method based on an adaptive dynamic template, and the method includes:

[0007] Step 1: At the initial moment, extract features from the initial image and store the extracted features as the initial template, adjacent template, and memory template. The adjacent template refers to the features retained after state evaluation in the adjacent frame, and the memory template refers to the template that integrates all historical information through the temporal attention mechanism.

[0008] Step 2: Taking the current position as the center, intercept the search image, extract features from the search image, and obtain the search features of the current frame;

[0009] Step 3: Input the current frame search features, initial template, adjacent template, and memory template in step 2 into the adaptive template fusion STF module to generate the final template;

[0010] Step 4: Perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map;

[0011] Step 5: The evaluation module determines the current tracking status based on the response graph;

[0012] Step 6: If the tracking status is good, update the adjacent template and use the memory template generation module MTG to generate the memory template. Otherwise, proceed to track the next frame.

[0013] Step 7: Continue to perform steps 2 to 6 as the tracking progresses until the tracking ends.

[0014] Further, in step 3, the adaptive template fusion STF module specifically includes:

[0015] Keep the three input template features and the number of search feature channels unchanged, and expand the two-dimensional feature map of each channel into one dimension;

[0016] The template features are used as query and the search features are used as keys to calculate the similarity matrix;

[0017] Use global maximum pooling to filter the similarity matrix to obtain the fusion weight, and finally weight it to generate the final template.

[0018] Furthermore, the calculation formula of the adaptive template fusion STF module is:

[0019]

[0020] Among them, T0, T1, T m They represent the initial template, adjacent template, and memory template respectively, X represents the search feature, T represents the final template, and d k Indicates the number of channels, GMP indicates global maximum pooling, δ0, δ1, δ m They represent the fusion weights of the initial template, adjacent template, and memory template, μ represents the modulation coefficient, and X T represents the transpose of X.

[0021] Furthermore, in step 5, the evaluation module includes: evaluating the current tracking state by averaging the peak correlation energy and the maximum response value.

[0022] Furthermore, the evaluation criteria of the evaluation module are as follows:

[0023]

[0024] Among them, R i,j , R max , R min They represent the response value, maximum response value, and minimum response value of the i-th row and j-th column in the response diagram, respectively. avg(·) means taking the average. α and β are the modulation coefficients, both of which are 0.9.

[0025] Furthermore, in step 6, the memory template generation module MTG fuses multiple frame features in a serial manner, specifically including: the input of the memory template generation module MTG includes the memory template T at the previous moment m t-1 and the current adjacent template T1 t ;

[0026] T1 t and (T m t-1 ) T Perform matrix multiplication to obtain the similarity matrix A m , A m The element a in i,j Represents T1 t The i-th row element in T m t-1 The inner product between the elements in the jth column in , uses linear normalization Linear(·) to replace the Softmax(·) of the QKV structure in Transformer;

[0027] A m As a weight to T m t-1 Perform weighted summation and obtain the same value as T1 t associated contextual information;

[0028] Use residual connection to weight the T before and after m t Add and multiply by the modulation factor γ, which is used to adjust T m t The value range of .

[0029] Furthermore, in step 6, the update formula of the memory template is:

[0030]

[0031] Among them, t represents the current frame and t-1 represents the previous frame.

[0032] In a second aspect, the present invention provides a MAV airborne target tracking system based on an adaptive dynamic template, the system comprising:

[0033] The template setting module is used to extract features from the initial image at the initial moment, and store the extracted features as the initial template, adjacent template, and memory template. The adjacent template refers to the features retained after state evaluation in the adjacent frame, and the memory template refers to the template that integrates all historical information through the time attention mechanism.

[0034] The current frame search feature acquisition module is used to intercept the search image with the current position as the center, extract features from the search image, and obtain the current frame search features;

[0035] The template fusion module is used to input the current frame search features, initial template, adjacent template, and memory template into the adaptive template fusion STF module to generate the final template;

[0036] The target position acquisition module is used to perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map;

[0037] A tracking state judgment module is used by the evaluation module to judge the current tracking state according to the response graph;

[0038] The template update module is used to update the adjacent template if the tracking status is good, and use the memory template generation module MTG to generate the memory template, otherwise, track the next frame;

[0039] The tracking module is used to continuously execute the above modules as the tracking proceeds until the tracking ends.

[0040] In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the steps of the MAV airborne target tracking method based on an adaptive dynamic template as described above are executed.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein a plurality of computer instructions are stored in the computer-readable storage medium, and the plurality of computer instructions are used to enable a computer to execute a MAV airborne target tracking method based on an adaptive dynamic template as described above.

[0042] Beneficial effects of the present invention:

[0043] The present invention provides a lightweight airborne target tracking method based on an adaptive dynamic template. The number of parameters and computational complexity of the target tracking method based on the adaptive dynamic template are much smaller than those of the existing tracking methods, but the performance is better than the correlation filtering method, achieving a good balance between tracking performance and computational complexity. In addition, the method also introduces a temporal attention mechanism and an adaptive template generation module, which improves the adaptability and anti-interference ability of the model without increasing the number of parameters.

[0044] The present invention proposes a memory template generation module and an adaptive template fusion module. Among them, the memory template generation module adopts a temporal series method to extract and integrate key information from historical tracking results. Compared with the conventional time series fusion method, this strategy can effectively save storage space, so that the limited template feature memory can contain all historical information. The adaptive template fusion module uses the similarity matrix between the template and the search feature to realize the dynamic adjustment of the template weight at different tracking stages.

[0045] The present invention performs particularly well in dealing with complex situations such as two-dimensional plane rotation, three-dimensional rotation, and posture changes. The memory template generation module of the present invention can continuously fuse effective information of the target during the change process during tracking, thereby ensuring that stable and accurate tracking can be maintained when the target has a significant appearance difference from the initial state. The present invention also performs well in the face of occlusion problems. When the tracked object reappears after a period of partial occlusion, the adaptive fusion module dynamically adjusts the proportion of different templates, which not only enhances the adaptability of the target, but also retains the necessary anti-interference ability.

[0046] The present invention is applicable to the field of MAV airborne target tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0048] Figure 1 It is a structural schematic diagram of a MAV airborne target tracking method based on an adaptive dynamic template of the present invention. DETAILED DESCRIPTION

[0049] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0050] Embodiment 1: A method for tracking a MAV airborne target based on an adaptive dynamic template. The method includes:

[0051] Step 1: At the initial moment, extract features from the initial image and store the extracted features as the initial template, adjacent template, and memory template. The adjacent template refers to the features retained after state evaluation in the adjacent frame, and the memory template refers to the template that integrates all historical information through the temporal attention mechanism.

[0052] Step 2: Taking the current position as the center, intercept the search image, extract features from the search image, and obtain the search features of the current frame;

[0053] Step 3: Input the current frame search features, initial template, adjacent template, and memory template in step 2 into the adaptive template fusion STF module to generate the final template;

[0054] Step 4: Perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map;

[0055] Step 5: The evaluation module determines the current tracking status based on the response graph;

[0056] Step 6: If the tracking status is good, update the adjacent template and memory template, and use the memory template generation module MTG to generate the memory template. Otherwise, proceed to track the next frame.

[0057] Step 7: Continue to perform steps 2 to 6 as the tracking progresses until the tracking ends.

[0058] This implementation proposes a method for tracking MAV airborne targets based on an adaptive dynamic template. The number of parameters and computational complexity of this method are much smaller than those of existing advanced tracking methods, and its performance is better than that of the correlation filtering method, achieving a good balance between tracking performance and computational complexity. In addition, this method also introduces a temporal attention mechanism and an adaptive template generation module, which improves the adaptability and anti-interference ability of the model without increasing the number of parameters.

[0059] Implementation method 2: This implementation method further defines the MAV airborne target tracking method based on the adaptive dynamic template as described above. In this implementation method, the adaptive template fusion STF module in step 3 is further defined, specifically including:

[0060] In step 3, the adaptive template fusion STF module specifically includes:

[0061] Keep the three input template features and the number of search feature channels unchanged, and expand the two-dimensional feature map of each channel into one dimension;

[0062] The template features are used as query and the search features are used as keys to calculate the similarity matrix;

[0063] Use global maximum pooling to filter the similarity matrix to obtain the fusion weight, and finally weight it to generate the final template.

[0064] This embodiment provides an adaptive template fusion module STF, which mainly performs two steps: calculating the similarity matrix between the template features and the search features, and filtering the weights using the global maximum pooling. Taking into account the uncertainty of the tracking task itself, the module cannot predict in advance the fusion ratio of the three templates at a certain moment of tracking, so it dynamically adjusts the fusion weight according to the tracking state of the target itself.

[0065] Implementation method three, this implementation method is a further limitation of the MAV airborne target tracking method based on the adaptive dynamic template as described above. In this implementation method, the formula of the adaptive template fusion STF module calculation process in step 3 is further limited, specifically including:

[0066] The calculation formula of the adaptive template fusion STF module is:

[0067]

[0068] Among them, T0, T1, T m They represent the initial template, adjacent template, and memory template respectively, X represents the search feature, T represents the final template, and d k Indicates the number of channels, GMP indicates global maximum pooling, δ0, δ1, δ m They represent the fusion weights of the initial template, adjacent template, and memory template, μ represents the modulation coefficient, and X T represents the transpose of X.

[0069] Embodiment 4: This embodiment further limits the MAV airborne target tracking method based on the adaptive dynamic template as described above. In this embodiment, the evaluation module is further limited, specifically including:

[0070] In step 5, the evaluation module includes: evaluating the current tracking state by averaging the peak correlation energy and the maximum value of the response.

[0071] The evaluation module of this embodiment is used to determine the tracking state. The module evaluates the current tracking state through the average peak correlation energy (APCE) and the maximum response value, and screens the tracking template to reduce the introduction of erroneous information.

[0072] Embodiment 5: This embodiment further defines the MAV airborne target tracking method based on an adaptive dynamic template as described above. In this embodiment, the evaluation criteria of the evaluation module are further defined, specifically including:

[0073] The evaluation criteria of the evaluation module are as follows:

[0074]

[0075] Among them, R i,j , R max , R min They represent the response value, maximum response value, and minimum response value of the i-th row and j-th column in the response graph. avg(·) means taking the average, α and β are the modulation coefficients, both of which are 0.9, and the t in the upper right corner represents the data of the t-th frame.

[0076] This embodiment provides an evaluation criterion applicable to target tracking based on an adaptive dynamic template as described above, which is used to determine the tracking state.

[0077] Embodiment 6: This embodiment further defines the MAV airborne target tracking method based on an adaptive dynamic template as described above. In this embodiment, the memory template generation module MTG in step 6 is further defined, specifically including:

[0078] Step 6 involves the memory template generation module MTG, which fuses multi-frame features in series. The input of MTG includes the memory template T m t-1 and the current adjacent template T1 t .

[0079] T1 t and (T m t-1 ) T Perform matrix multiplication to obtain the similarity matrix A m , A m The element a in i,j Represents T1 t The i-th row element in Tm t-1 The inner product between the elements in the j-th column in is replaced by the linear normalization Linear(·) in the QKV structure of Transformer to reduce nonlinear calculations and prepare for subsequent edge device deployment.

[0080] A m As a weight to T m t-1 Perform weighted summation to obtain the same value as T1 t Associated context information to achieve effective aggregation of information between historical frame templates and current frame templates.

[0081] Since MTG aggregates information in a time-series manner, it is necessary to ensure that the T generated at each moment is m t Expressing the same spatial attributes, the residual connection is used at the end to weight the T before and after m t Add and multiply by a modulation coefficient γ, where the modulation coefficient is used to keep T m t The value range does not fluctuate greatly.

[0082] Embodiment 7: This embodiment further limits the MAV airborne target tracking method based on the adaptive dynamic template as described above. In this embodiment, the update formula of the memory template in step 6 is further limited, specifically including:

[0083] In step 6, the update formula of the memory template is:

[0084]

[0085] Among them, t represents the current frame and t-1 represents the previous frame.

[0086] This embodiment provides an update formula for a memory template. In addition, the update of the adjacent template evaluates the current tracking state through the evaluation module, and directly uses the search feature as the adjacent template if the state is good.

[0087] Embodiment 8: This embodiment is an example of a MAV airborne target tracking method based on an adaptive dynamic template as described above. Figure 1 As shown, specifically including:

[0088] Aiming at the problem of balancing the tracking performance and computational complexity in MAV airborne tracking, a MAV airborne target tracking method based on adaptive dynamic template is proposed. The specific scheme is as follows:

[0089] Step 1: At the initial moment, extract features from the initial image and store the extracted features as the initial template, adjacent template and memory template; the adjacent template refers to the features retained after state evaluation in the adjacent frame. During the tracking process, only one frame of features is retained and continuously updated as the tracking state changes; the memory template refers to the template that integrates all historical information through the temporal attention mechanism, which will be introduced in detail in the subsequent MTG module;

[0090] Step 2: Taking the current position as the center, intercept the search image, extract features from the search image, and obtain the search features of the current frame;

[0091] Step 3: Input the current frame search features, initial template, adjacent template, and memory template in step 2 into the adaptive template fusion STF module to generate the final template;

[0092] The adaptive template fusion module STF takes into account the uncertainty of the tracking task itself and cannot predict in advance the fusion ratio of the three templates at a certain moment of tracking. Therefore, the fusion weight is dynamically adjusted according to the tracking state of the target itself. This module mainly performs two steps: calculating the similarity matrix between the template features and the search features, and using the global maximum pooling to filter the weights. The specific calculation process is as follows:

[0093] Keep the three input template features and the number of search feature channels unchanged, and expand the two-dimensional feature map of each channel into one dimension;

[0094] The template features are used as query and the search features are used as keys to calculate the similarity matrix;

[0095] Use global maximum pooling to filter the similarity matrix to obtain the fusion weight, and finally weight it to generate the final template.

[0096] The above process can form the following formula:

[0097]

[0098] Among them, T0, T1, T m They represent the initial template, adjacent template, and memory template respectively, X represents the search feature, T represents the final template, and d k Indicates the number of channels, GMP indicates global maximum pooling, δ0, δ1, δ m They represent the fusion weights of the initial template, adjacent template, and memory template, μ represents the modulation coefficient, and X T represents the transpose of X.

[0099] Step 4: Perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map to infer the current target position based on the response map;

[0100] Step 5: The evaluation module determines the current tracking state according to the response graph; if the state is good, the frame is saved as a neighboring template.

[0101] The evaluation module for judging the tracking state evaluates the current tracking state through the average peak-to-correlation energy (APCE) and the maximum response value, and screens the tracking template to reduce the introduction of erroneous information. The evaluation criteria of the evaluation module are as follows:

[0102]

[0103] R max t ≥α·R max 0

[0104]

[0105] Among them, R i,j , R max , R min They represent the response value, maximum response value, and minimum response value of the i-th row and j-th column in the response diagram, respectively. avg(·) means taking the average. α and β are the modulation coefficients, both of which are 0.9.

[0106] Step 6: If the tracking status is good, update the adjacent template and the memory template, and use the memory template generation module MTG to generate the memory template. Otherwise, proceed to track the next frame.

[0107] The memory template generation module MTG fuses multiple frame features in series. The input of MTG includes the memory template T m t-1 and the current adjacent template T1 t .

[0108] T1 t and (T m t-1 ) T Perform matrix multiplication to obtain the similarity matrix A m , A m The element a in i,j Represents T1 t The i-th row element in T m t-1 The inner product between the elements in the j-th column in is replaced by the linear normalization Linear(·) in the QKV structure of Transformer to reduce nonlinear calculations and prepare for subsequent edge device deployment.

[0109] A mAs a weight to T m t-1 Perform weighted summation to obtain the same value as T1 t Associated context information to achieve effective aggregation of information between historical frame templates and current frame templates.

[0110] Since MTG aggregates information in a time-series manner, it is necessary to ensure that the T generated at each moment is m t Expressing the same spatial attributes, the residual connection is used at the end to weight the T before and after m t Add and multiply by a modulation coefficient γ, where the modulation coefficient is used to keep T m t The value range does not fluctuate greatly.

[0111] The updated memory template calculation formula is as follows:

[0112]

[0113] Among them, t represents the current frame and t-1 represents the previous frame.

[0114] The evaluation module evaluates the current tracking status and directly uses the search features as adjacent templates if the status is good.

[0115] Step 7: Continue to execute steps 2 to 6 as the tracking progresses until the tracking is completed.

[0116] The method of the present invention is compared with several tracking methods in the prior art using the OTB100 and UAV123 datasets, as shown in the following table:

[0117]

[0118] It can be seen that the performance of the present invention is better than the benchmark algorithm SiamFC and the advanced correlation filtering method. Compared with SiamFC, the present invention improves the accuracy by 4.1% and the success rate by 2.1% on the OTB100 dataset, and improves the accuracy by 4.8% and the success rate by 1.2% on the UAV123 dataset.

Claims

1. A MAV airborne target tracking method based on adaptive dynamic template, characterized in that: The method comprises: Step 1: At the initial moment, extract features from the initial image and store the extracted features as the initial template, adjacent template, and memory template. The adjacent template refers to the features retained after state evaluation in the adjacent frame, and the memory template refers to the template that integrates all historical information through the temporal attention mechanism. Step 2: Taking the current position as the center, intercept the search image, extract features from the search image, and obtain the search features of the current frame; Step 3: Input the current frame search features, initial template, adjacent template, and memory template in step 2 into the adaptive template fusion STF module to generate the final template; Step 4: Perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map; Step 5: The evaluation module determines the current tracking status based on the response graph; Step 6: If the tracking status is good, update the adjacent template and use the memory template generation module MTG to generate the memory template. Otherwise, proceed to track the next frame. Step 7: Continue to perform steps 2 to 6 as the tracking progresses until the tracking ends.

2. According to claim 1, a MAV airborne target tracking method based on adaptive dynamic template is characterized in that: In step 3, the adaptive template fusion STF module specifically includes: Keep the three input template features and the number of search feature channels unchanged, and expand the two-dimensional feature map of each channel into one dimension; The template features are used as query and the search features are used as keys to calculate the similarity matrix; Use global maximum pooling to filter the similarity matrix to obtain the fusion weight, and finally weight it to generate the final template.

3. The method for tracking an airborne target of a MAV based on an adaptive dynamic template according to claim 2, characterized in that: The calculation formula of the adaptive template fusion STF module is: Among them, T0, T1, T m They represent the initial template, adjacent template, and memory template respectively, X represents the search feature, T represents the final template, and d k Indicates the number of channels, GMP indicates global maximum pooling, δ0, δ1, δ m They represent the fusion weights of the initial template, adjacent template, and memory template, μ represents the modulation coefficient, and X T represents the transpose of X.

4. The method for tracking a MAV airborne target based on an adaptive dynamic template according to claim 1, characterized in that: In step 5, the evaluation module includes: evaluating the current tracking state by averaging the peak correlation energy and the maximum value of the response.

5. The method for tracking an airborne target of a MAV based on an adaptive dynamic template according to claim 4, characterized in that: The evaluation criteria of the evaluation module are as follows: R max t ≥α·R max 0 Among them, R i,j , R max , R min They represent the response value, maximum response value, and minimum response value in the i-th row and j-th column of the response diagram, respectively. avg(·) means taking the average. α and β are the modulation coefficients, both of which are 0.

9.

6. The method for tracking a MAV airborne target based on an adaptive dynamic template according to claim 1, characterized in that: In step 6, the memory template generation module MTG fuses multiple frame features in a serial manner, specifically including: the input of the memory template generation module MTG includes the memory template T at the previous moment m t-1 and the current adjacent template T1 t ; T1 t and (T m t-1 ) T Perform matrix multiplication to obtain the similarity matrix A m , A m The element a in i,j Represents T1 t The i-th row element in T m t-1 The inner product between the elements in the jth column in , uses linear normalization Linear(·) to replace the Softmax(·) of the QKV structure in Transformer; A m As a weight to T m t-1 Perform weighted summation and obtain the same value as T1 t associated contextual information; Use residual connection to weight the T before and after m t Add and multiply by the modulation factor γ, which is used to adjust T m t The value range of .

7. The method for tracking an airborne target of a MAV based on an adaptive dynamic template according to claim 6, characterized in that: In step 6, the update formula of the memory template is: Among them, t represents the current frame and t-1 represents the previous frame.

8. A MAV airborne target tracking system based on adaptive dynamic template, characterized in that: The system comprises: The template setting module is used to extract features from the initial image at the initial moment, and store the extracted features as the initial template, adjacent template, and memory template. The adjacent template refers to the features retained after state evaluation in the adjacent frame, and the memory template refers to the template that integrates all historical information through the time attention mechanism. The current frame search feature acquisition module is used to intercept the search image with the current position as the center, extract features from the search image, and obtain the current frame search features; The template fusion module is used to input the current frame search features, initial template, adjacent template, and memory template into the adaptive template fusion STF module to generate the final template; The target position acquisition module is used to perform correlation operations on the final template and the search template to obtain a response map, and infer the target position based on the response map; A tracking state judgment module is used by the evaluation module to judge the current tracking state according to the response graph; The template update module is used to update the adjacent template if the tracking status is good, and use the memory template generation module MTG to generate the memory template, otherwise, track the next frame; The tracking module is used to continuously execute the above modules as the tracking proceeds until the tracking ends.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor runs the computer program stored in the memory, the steps of the method according to any one of claims 1 to 7 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of computer instructions, and the plurality of computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images