A secondary target tracking method, medium, and system

Abstract

The application discloses a secondary target tracking method, comprising the following steps: acquiring a plurality of to-be-recognized images containing a primary target in a continuous time period; selecting any to-be-recognized image as a first to-be-recognized image, and setting a secondary target capture area in the first to-be-recognized image as a first area; preliminarily identifying the secondary target in the first area according to a first setting standard to obtain a suspected secondary target in the first to-be-recognized image; acquiring a suspected secondary target in a second to-be-recognized image, wherein the second to-be-recognized image is a to-be-recognized image acquired after the first to-be-recognized image; comparing the first to-be-recognized image and the second to-be-recognized image on the suspected secondary target to complete confirmation of the suspected secondary target and obtain a final secondary target; and switching the tracking target from the primary target to the final secondary target to complete subsequent tracking. The secondary target tracking method can improve the tracking success rate of the secondary target.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the technical field of tracking air targets, and particularly relates to a secondary target tracking method, medium and system. BACKGROUND

[0002] The air target tracking technology has a relatively wide application in military defense or other related fields, and the tracking of a secondary target (such as a missile) after the separation of a primary target (such as an airplane) in the air is more critical. However, in reality, the identification of the secondary target will be affected by other objects in the flight background, resulting in poor identification accuracy. When other objects are mistakenly identified as the secondary target and tracked, the tracking task will fail and other negative effects will be caused. In the prior art, before starting the tracking target switching, only a preliminary identification is performed on the suspected secondary target, and thus the accuracy of identifying the secondary target will be obviously insufficient, and the system will select incorrectly and the secondary target tracking will fail. Therefore, how to improve the success rate of the secondary target tracking has become a problem to be solved in the field. SUMMARY

[0003] In view of the above defects or deficiencies in the prior art, it is expected to provide a secondary target tracking method, medium and system which can improve the tracking success rate.

[0004] The specific technical solutions are as follows:

[0005] First aspect

[0006] The present application provides a secondary target tracking method, comprising the following steps:

[0007] In a continuous time period, a plurality of to-be-identified images containing a primary target are acquired;

[0008] Any of the to-be-identified images is selected and recorded as a first to-be-identified image, and a secondary target capture region in the first to-be-identified image is set and recorded as a first region;

[0009] In the first region, a preliminary identification of the secondary target is performed according to a first set standard to acquire a suspected secondary target in the first to-be-identified image;

[0010] A suspected secondary target in a second to-be-identified image is acquired, wherein the second to-be-identified image is the to-be-identified image acquired after the first to-be-identified image;

[0011] The first to-be-identified image and the second to-be-identified image are compared in terms of the suspected secondary target to complete the confirmation of the suspected secondary target and acquire a final secondary target;

[0012] Switching the tracking target from the primary target to the final secondary target to complete subsequent tracking.

[0013] As a further limitation of the present application, the second to-be-identified image is not adjacent to the first to-be-identified image in the sequence of to-be-identified images obtained;

[0014] The secondary target tracking method further comprises the following steps:

[0015] Selecting a to-be-identified image between the first to-be-identified image and the second to-be-identified image in the acquisition sequence, denoted as a third to-be-identified image;

[0016] According to the position of the suspected secondary target in the first to-be-identified image, setting a secondary target capture area in the third to-be-identified image, denoted as a second area;

[0017] In the second area, re-identify the secondary target with a second set standard to obtain the suspected secondary target in the third to-be-identified image, wherein the second set standard is more accurate than the first set standard in identifying the secondary target.

[0018] As a further limitation of the present application, the preliminary identification of the secondary target in the first area with the first set standard specifically comprises the following steps:

[0019] According to the size of the first area, set a corresponding filter template;

[0020] Convolve the first area with the filter template to obtain a convolution image of the first area;

[0021] Preliminarily identify the secondary target in the convolution image of the first area with the first set standard.

[0022] As a further limitation of the present application, the preliminary identification of the secondary target with the first set standard to obtain the suspected secondary target in the first to-be-identified image specifically comprises the following steps:

[0023] Set a first threshold value for identifying the suspected secondary target by grayscale, and a second threshold value for identifying the edge of the suspected secondary target by grayscale;

[0024] Detect the grayscale of each pixel point in the first area in turn, and compare the first detected grayscale value of each pixel point with the first threshold value;

[0025] When the first detected grayscale value of any pixel point is found to be greater than the first threshold value, take the pixel point as the starting point, and detect the grayscale value of other pixel points adjacent to it to obtain a second detected grayscale value;

[0026] comparing the second detection gray value with the second threshold value, when the second detection gray value of other pixel points is greater than the second threshold value and less than the first threshold value, determining that the pixel point is an edge pixel point;

[0027] According to the above steps, all edge pixel points in the first to-be-identified image are obtained, and a suspected secondary target in the first to-be-identified image is further obtained.

[0028] As a further limitation of the present application, the third to-be-identified image is set according to the position of the suspected target in the first to-be-identified image, and is recorded as a second region, which specifically includes the following steps:

[0029] The first position of the primary target centroid in the first to-be-identified image and the second position of the primary target centroid in the third to-be-identified image are obtained respectively;

[0030] The first position and the second position are corresponded to determine a third region in the third to-be-identified image, which has the same size as the first region position;

[0031] The center of the third region is taken as the center to demarcate the second region, wherein the size of the second region is smaller than that of the third region.

[0032] As a further limitation of the present application, the first to-be-identified image and the second to-be-identified image are compared with respect to the suspected secondary target to complete the confirmation of the suspected secondary target and obtain a final secondary target, which specifically includes the following steps:

[0033] The first relative distance, the third relative distance and the second relative distance between the suspected secondary target and the primary target centroid in the first to-be-identified image, the third to-be-identified image and the second to-be-identified image are obtained respectively;

[0034] The relative sizes of the first relative distance, the third relative distance and the second relative distance are judged, and when the above three satisfy that the first relative distance, the third relative distance and the second relative distance increase in turn, the confirmation of the suspected secondary target is completed.

[0035] Second aspect

[0036] The present application provides a medium for tracking a secondary target, which is used to realize the secondary target tracking method as described above.

[0037] Third aspect

[0038] The application provides a system for tracking a secondary target, for implementing the secondary target tracking method as described above.

[0039] The application has the advantages that:

[0040] In the present solution, after obtaining a suspected secondary target, a plurality of to-be-identified images are continuously collected by obtaining a primary target, and a preliminary identification is performed on the secondary target in the previously collected to-be-identified image to obtain a suspected secondary target; and the suspected secondary target in the subsequently collected to-be-identified image is compared with the above suspected secondary target to confirm that it is indeed a secondary target separated from the primary target. Thus, the preliminary identification of only the suspected target to confirm that it is a secondary target separated from the primary target as in the prior art is avoided, the confirmation of the secondary target is more rigorous, the identification accuracy of the secondary target is improved, and the success rate of tracking the secondary target is improved. BRIEF DESCRIPTION OF DRAWINGS

[0041] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings:

[0042] Figure 1 A step flowchart of the secondary target tracking method provided by the present embodiment; DETAILED DESCRIPTION

[0043] The application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely intended to explain the related application, but not to limit the application. In addition, it should be noted that only the parts related to the application are shown in the drawings for ease of description.

[0044] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and embodiments. EMBODIMENT

[0045] Reference should be made to Figure 1 A secondary target tracking method provided by the present embodiment includes the following steps:

[0046] S1: In a continuous time period, a plurality of to-be-identified images containing a primary target are obtained;

[0047] S2: Select any of the to-be-identified images, denoted as a first to-be-identified image, and set a secondary target capture region in the first to-be-identified image, denoted as a first region;

[0048] S3: preliminarily identifying a secondary target in the first region according to a first set standard to obtain a suspected secondary target in the first to-be-identified image;

[0049] S4: obtaining a suspected secondary target in a second to-be-identified image, wherein the second to-be-identified image is the to-be-identified image obtained after the first to-be-identified image;

[0050] S5: comparing the first to-be-identified image and the second to-be-identified image for the suspected secondary target to complete confirmation of the suspected secondary target and obtain a final secondary target;

[0051] S6: switching the tracking target from the primary target to the final secondary target to complete subsequent tracking.

[0052] In the scheme, after obtaining a suspected secondary target, a plurality of to-be-identified images are continuously collected according to the primary target, and a secondary target in a to-be-identified image collected in advance is preliminarily identified to obtain a suspected secondary target; and the suspected secondary target in a to-be-identified image collected later is compared with the above suspected secondary target to confirm that it is indeed a secondary target separated from the primary target, so that the preliminary identification of the suspected target as in the prior art, that is, the confirmation that it is a secondary target separated from the primary target, is avoided, the confirmation of the secondary target is more rigorous, the identification accuracy of the secondary target is improved, and the success rate of tracking the secondary target is improved.

[0053] In the preferred embodiment for further improving the success rate of tracking the secondary target, the second to-be-identified image is not adjacent to the first to-be-identified image in the sequence of to-be-identified images obtained;

[0054] The secondary target tracking method further includes the following steps:

[0055] A to-be-identified image between the first to-be-identified image and the second to-be-identified image in the acquisition sequence is selected, denoted as a third to-be-identified image;

[0056] According to the position of the suspected secondary target in the first to-be-identified image, a secondary target capture region in the third to-be-identified image is set, denoted as a second region;

[0057] In the second region, a secondary target is identified again according to a second set standard to obtain a suspected secondary target in the third to-be-identified image, wherein the second set standard has higher identification accuracy for the secondary target than the first set standard.

[0058] Since the preliminary identification of the secondary target is performed in the first to-be-identified image and the confirmation of the secondary target is performed in the second to-be-identified image, a step of re-identifying the suspected secondary target in the third to-be-identified image is added between the two steps, so that the accuracy of the judgment of the suspected secondary target is further improved, the influence of other objects in the background on the identification of the secondary target is avoided, and the success rate of tracking the secondary target is further improved.

[0059] In the preferred embodiment of further improving the success rate of tracking the secondary target, the preliminary identification of the secondary target in the first region according to the first set standard specifically includes the following steps:

[0060] According to the size of the first region, a filter template corresponding to the first region is set;

[0061] The first region is convolved with the filter template to obtain a convolved image of the first region;

[0062] The secondary target in the convolved image of the first region is preliminarily identified according to the first set standard.

[0063] According to the above steps, the convolution operation of the filter template and the first region can effectively reduce the gray scale of the background objects in the first region, and further effectively highlight the suspected secondary target, so that the accuracy of the judgment of the suspected secondary target is further improved, and thus the success rate of tracking the secondary target is further improved.

[0064] In the preferred embodiment of the preliminary identification of the secondary target in the first to-be-identified image, the preliminary identification of the secondary target according to the first set standard to obtain the suspected secondary target in the first to-be-identified image specifically includes the following steps:

[0065] A first threshold value for identifying the suspected secondary target by gray scale and a second threshold value for identifying the edge of the suspected secondary target by gray scale are set;

[0066] The gray scale of each pixel point in the first region is detected in sequence, and the first detected gray scale value of each pixel point is compared with the first threshold value;

[0067] When the first detected gray scale value of any pixel point is greater than the first threshold value, the pixel point is taken as a starting point, and the gray scale values of other pixel points adjacent to the starting point are detected to obtain second detected gray scale values;

[0068] The second detected gray scale values are compared with the second threshold value, and when the second detected gray scale value of the other pixel point is greater than the second threshold value and less than the first threshold value, the pixel point is determined as an edge pixel point.

[0069] According to the above steps, all edge pixels in the first image to be identified are obtained, and a suspected secondary target in the first image to be identified is further obtained.

[0070] In a preferred embodiment for further identifying the suspected secondary target, the third image to be identified is set as a secondary target capture area according to the position of the suspected target in the first image to be identified, denoted as a second area, which includes the following steps:

[0071] The first position of the primary target centroid in the first image to be identified and the second position of the primary target centroid in the third image to be identified are obtained respectively.

[0072] The first position and the second position are corresponded to determine a third area in the third image to be identified, which has the same size as the first area.

[0073] The second area is demarcated with the center of the third area as the center, and the size of the second area is smaller than that of the third area.

[0074] By demarcating the third area corresponding to the first area in the third image to be identified, and demarcating the second area with a smaller size in the third area, the area for re-identifying the secondary target can be further reduced, and the determination of the secondary target as a target to be tracked can be further improved when re-identified.

[0075] In a preferred embodiment for improving the speed of confirming the secondary target, the first image to be identified and the second image to be identified are compared for the suspected secondary target to complete the confirmation of the suspected secondary target and obtain the final secondary target, which includes the following steps:

[0076] The first relative distance, the third relative distance and the second relative distance between the suspected secondary target and the primary target centroid in the first image to be identified, the third image to be identified and the second image to be identified are obtained respectively.

[0077] The relative sizes of the first relative distance, the third relative distance and the second relative distance are judged, and when the above three satisfy that the first relative distance, the third relative distance and the second relative distance increase in turn, the confirmation of the suspected secondary target is completed.

[0078] According to the above steps, by comparing the distance between the secondary target and the primary target in the adjacent or sequentially obtained to-be-identified image, whether the distance is gradually increasing, the influence of the interfering objects in the background of the to-be-identified image can be excluded, and it is confirmed that the suspected secondary target is the identity of the to-be-tracked secondary target separated from the primary target. Of course, more samples can be collected according to the above judgment principle, and it is confirmed whether the suspected secondary target is a to-be-tracked secondary target. In this way, the confirmation speed is faster.

[0079] In order to better illustrate the above principle, the following specific embodiments are used for specific description:

[0080] Step 1, separate target capture area filtering processing, after the tracking system stably tracks the target, the target is located in the image center area, and the target information is as follows:

[0081] The size of the target: (TarSizeX, TarSizeY);

[0082] The centroid coordinate information: (TarX, TarY);

[0083] In the range of the target directly below the 60*60 area, that is, taking (TarX, TarY+ TarSizeY / 2+30) as the center, the 60*60 area is used as a separate target capture area A, and a 5*5 size filter template d is used. As shown in equation (1), the template convolution operation is performed on the area A to obtain the area filter image B, and step 2 is executed. (1)

[0084] Step 2, perform area separation target preliminary extraction, define the gray detection threshold TH1 and the edge detection threshold TH2, let TH1=300 and TH2=200, and the separate target extraction specific operation is: scan the area filter image B pixel by pixel, and the pixel gray of the area image exceeding TH1 is used as the candidate target area starting point. According to the edge detection threshold TH2, the edge of each candidate target area is tracked, and the size (MiSizeX, MiSizeY), gray MiGrey, and coordinate information (MiX, MiY) of each candidate target are counted while tracking the edge. The candidate target with a size greater than 1*1 range is retained. All candidate targets are sorted in descending order of gray, and the top 8 targets are retained as suspected targets according to the descending order of gray. The number of suspected targets is YN, if YN=0, the tracking result of this time is no separate target, and step 1 is executed for the next frame image; if YN>0, the target with the largest gray is taken as the separate target result of this frame, and step 3 is executed.

[0085] Step 3, a target capture area is calculated, a 30*30 area A1 centered on the last frame target coordinates (TarX, TarY) is taken as a tracking range of the current frame image, a 5*5 filter template d is used, and a template convolution operation is performed on the area A1 as shown in equation (1) to obtain a filter image B1 of the area.

[0086] Step 4, TH1 = MiGrey / 2, TH2 = 200, a separated target extraction is performed, the number of suspected targets is YN, if YN = 0, the tracking result of this time is no separated target, and the next frame image re-executes step 1; if YN > 0, step 5 is executed.

[0087] Step 5, for the YN suspected targets, the pitch coordinates (i.e. the distance between the secondary target and the primary target) are sorted, and the largest pitch coordinate is selected as the target result of the current frame. Then, a trajectory judgment is performed, and the pitch coordinate of the current frame result is compared with that of the last frame result. If the pitch coordinate continuously increases for three frames, the target is confirmed as a separated target, and step 6 is executed. If the pitch coordinate increases and the number of frames is less than three, the next frame image executes step 3 to continue the separated target confirmation process. If the pitch coordinate is not greater than the last frame result, the trajectory association fails, and it is confirmed that the target is not a separated target, and step 1 is re-executed.

[0088] Step 6, after the separated target is confirmed to exist, the tracking output is switched, the original target information is stopped, the output information is updated to the separated target information, and the tracking system turntable is controlled to track the separated target. Embodiment

[0089] The embodiment provides a medium for tracking a secondary target, which is used to implement the secondary target tracking method. Embodiment

[0090] The embodiment provides a system for tracking a secondary target, which is used to implement the secondary target tracking method.

[0091] The above description is merely preferred embodiments of the present application and a description of the principles of the applied technology. It should be understood by those skilled in the art that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and also covers other technical solutions formed by any combination of the above technical features or equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features disclosed in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. A method of tracking a secondary target, characterized by, The method comprises the following steps: acquiring a plurality of to-be-recognized images containing a primary target within a continuous time period; selecting any of the to-be-recognized images as a first to-be-recognized image, and setting a secondary target capture region in the first to-be-recognized image as a first region; performing preliminary identification of a secondary target in the first region according to a first set standard to obtain a suspected secondary target in the first to-be-recognized image; obtaining a suspected secondary target in a second to-be-recognized image, wherein the second to-be-recognized image is a to-be-recognized image acquired after the first to-be-recognized image; comparing the first to-be-recognized image and the second to-be-recognized image with respect to the suspected secondary target to complete confirmation of the suspected secondary target and obtain a final secondary target; switching a tracking target from the primary target to the final secondary target to complete subsequent tracking; the second to-be-recognized image is not adjacent to the first to-be-recognized image in a sequence of to-be-recognized images acquired; the secondary target tracking method further comprises the following steps: selecting a to-be-recognized image between the first to-be-recognized image and the second to-be-recognized image in an acquisition sequence as a third to-be-recognized image; setting a secondary target capture region in the third to-be-recognized image as a second region according to a position of the suspected secondary target in the first to-be-recognized image; performing re-identification of a secondary target in the second region according to a second set standard to obtain a suspected secondary target in the third to-be-recognized image, wherein the second set standard is stronger than the first set standard in terms of identification accuracy of a secondary target; the setting of the secondary target capture region in the third to-be-recognized image as the second region according to the position of the suspected secondary target in the first to-be-recognized image specifically comprises the following steps: respectively obtaining a first position of a primary target centroid in the first to-be-recognized image and a second position of the primary target centroid in the third to-be-recognized image; corresponding the first position and the second position to determine a third region in the third to-be-recognized image with a same size as the first region; centering on a center of the third region, the second region is demarcated, wherein the size of the second region is smaller than that of the third region; the comparing of the first to-be-recognized image and the second to-be-recognized image with respect to the suspected secondary target to complete confirmation of the suspected secondary target and obtain a final secondary target specifically comprises the following steps: respectively obtaining a first relative distance, a third relative distance and a second relative distance between the suspected secondary target and the primary target centroid in the first to-be-recognized image, the third to-be-recognized image and the second to-be-recognized image; judging relative sizes of the first relative distance, the third relative distance and the second relative distance, and when the three distances satisfy a condition that the first relative distance, the third relative distance and the second relative distance increase in turn, the suspected secondary target is confirmed.

2. The method of claim 1, wherein, The preliminary identification of the secondary target in the first region according to the first set standard comprises the following steps: According to the size of the first region, a filter template corresponding to the first region is set; The first region is subjected to convolution operation with the filter template to obtain a convolution image of the first region; The secondary target in the convolution image of the first region is preliminarily identified according to the first set standard.

3. The method of claim 1, wherein, The preliminary identification of the secondary target according to the first set standard to obtain the suspected secondary target in the first to-be-identified image comprises the following steps: A first threshold for identifying the suspected secondary target by gray scale and a second threshold for identifying the edge of the suspected secondary target by gray scale are set; The first detection gray scale value of each pixel point in the first region is compared with the first threshold; When the first detection gray scale value of any pixel point is greater than the first threshold, the pixel point is taken as a starting point, and the gray scale value of other pixel points adjacent to the starting point is detected to obtain a second detection gray scale value; The second detection gray scale value is compared with the second threshold, and when the second detection gray scale value of other pixel points is greater than the second threshold and less than the first threshold, the pixel point is determined as an edge pixel point; All edge pixel points in the first to-be-identified image are obtained according to the above steps, and then the suspected secondary target in the first to-be-identified image is obtained.

4. A medium for tracking a secondary target, characterized by A secondary target tracking method is implemented.

5. A system for tracking a secondary target, the system comprising: A secondary target tracking method is implemented.

Images