A venue adaptive detection method and apparatus

Abstract

The application provides a field self-adaptive detection method and device, the method comprises the following steps: rotating by a preset adjustment angle according to a preset direction; performing marker identification on a central region of a scene image to obtain a marker identification result; performing first marker filtering according to the marker identification result, and recording the rotation angle corresponding to the filtered marker; when the cumulative rotation angle of the holder rotation module reaches the rotation upper limit angle, a plurality of markers and corresponding rotation angles are obtained; if the number of markers is greater than or equal to a marker threshold, the included angle value between each two markers is calculated, and two markers meeting a preset angle range are selected from the included angle value, the rotation angle of the two markers is recorded, and the maximum deflection angle is calculated; by the above method, the maximum deflection angle of the holder when shooting a sports competition can be determined, so that the field self-adaptive detection is completed.

Description

Technical Field

[0001] This invention relates to the field of intelligent shooting technology, specifically to a site-adaptive detection method and device. Background Technology

[0002] Determining the maximum left and right deflection angle of a gimbal in real-world tracking shooting scenarios is a challenge. In past cases, developers have often adopted two approaches: no maximum deflection angle or a constant maximum deflection angle. With the no maximum deflection angle approach, the gimbal is easily affected by the tracked target and may detach from the scene. For example, in shooting a football match, the gimbal might lose focus due to tracking a player leaving the field and fail to automatically return to the starting position. The constant maximum deflection angle approach is simple and efficient, but it also has three major problems: first, the initial shooting position must be fixed, and the gimbal position cannot be changed during shooting; second, the maximum deflection angle cannot be altered by external factors (such as human collisions or strong winds); and third, the angular error from each gimbal rotation can accumulate over long shooting periods, leading to angular deviation. Summary of the Invention

[0003] To address the aforementioned constraints and limitations, this invention proposes a site-adaptive detection method and device. By identifying and filtering landmarks in the scene, the maximum deflection angle of the gimbal during sports competitions can be calculated.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A site adaptive detection method, characterized in that the method includes:

[0006] S11. The gimbal rotation module rotates in a preset direction and adjusts to a preset angle.

[0007] S12. Perform landmark recognition in the central region of the scene image to obtain the landmark recognition result; if no landmark is recognized, return to S11.

[0008] S13. Based on the marker recognition results, perform the first marker filtering, record the rotation angle corresponding to the filtered marker, and return to S11.

[0009] The first marker filtering refers to selecting markers that conform to the first preset filtering rules;

[0010] S14. When the cumulative rotation angle of the gimbal rotation module reaches the upper limit angle, obtain several markers and their corresponding rotation angles; if the number of markers is greater than or equal to the marker threshold, then execute S15.

[0011] S15. Calculate the angle between each pair of markers, select two markers that fit within the preset angle range, record their rotation angles, and calculate the maximum deflection angle.

[0012] S16. When the gimbal is performing a tracking task, if the maximum deflection angle deviation occurs, the maximum deflection angle correction will be performed.

[0013] A site-adaptive detection device, the device comprising a gimbal rotation module, a gimbal operating system module, and an image acquisition module.

[0014] The gimbal rotation module is the rotating structure of the gimbal, which can rotate at a corresponding angle according to the control commands of the adjustment control module.

[0015] The gimbal operating system module includes a target detection module and an adjustment and control module.

[0016] The target detection module is used to detect landmarks in scene images.

[0017] The adjustment and control module is used to calculate the maximum deflection angle based on the marker detection results, and can also send control commands to the gimbal rotation module and the image acquisition module based on the marker detection results.

[0018] The image capture module is used to capture scene images.

[0019] Compared with the prior art, the present invention has the following advantages:

[0020] (1) The maximum deflection angle of the gimbal can be determined in real time when shooting sports competitions in actual tracking shooting scenarios;

[0021] (2) It has wide adaptability and is suitable for sports with field boundaries and obvious features at the boundaries;

[0022] (3) Real-time correction to avoid cumulative deviations or other external factors causing the gimbal to move and cause the screen angle to shift.

[0023] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the steps of a site adaptive detection method provided in an embodiment of the present invention.

[0025] Figure 2 A flowchart illustrating the steps of a maximum deflection angle correction method provided in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of a preset angle range in a football field, provided as an embodiment of the present invention.

[0027] Figure 4 This is a structural diagram of a site adaptive detection device provided in an embodiment of the present invention. Detailed Implementation

[0028] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. To further understand the present invention, the present invention will be further described in detail below with reference to the preferred embodiments.

[0029] One aspect of the present invention is a field adaptive detection method, which is applicable to sports activities with field boundaries and obvious features at the boundaries, and can determine the maximum deflection angle of the gimbal when shooting sports activities in actual tracking and shooting scenarios; the maximum deflection angle consists of the maximum left deflection angle and the maximum right deflection angle.

[0030] refer to Figure 1 The method includes the following steps:

[0031] S11. The gimbal rotation module rotates in a preset direction and adjusts to a preset angle.

[0032] S12. The target recognition module performs landmark recognition in the central area of ​​the scene image and obtains the landmark recognition result; if no landmark is recognized, it returns to S11.

[0033] The scene images were captured by the image acquisition module;

[0034] The marker recognition results include whether a marker was detected, the marker's size characteristics, and the rotation angle.

[0035] S13. Based on the marker recognition results, perform the first marker filtering, record the rotation angle corresponding to the filtered marker, and return to S11.

[0036] The first marker filtering refers to selecting markers that conform to the first preset filtering rules;

[0037] S14. When the cumulative rotation angle of the gimbal rotation module reaches the upper limit of rotation angle, a number of markers and their corresponding rotation angles are obtained. If the number of markers is less than the marker threshold, the maximum deflection angle is ±90°. If the number of markers is greater than or equal to the marker threshold, then S15 is executed.

[0038] S15. Calculate the angle between each pair of markers, select two markers that fit within the preset angle range, record their rotation angles, and calculate the maximum deflection angle.

[0039] The method for calculating the maximum deflection angle is as follows:

[0040] α left β left -s

[0041] α right =β right -s

[0042] Where α left α right These represent the maximum leftward deflection angle and the maximum rightward deflection angle, respectively; β left β right These represent the rotation angles of the left and right signs, respectively; s is the preset correction angle.

[0043] S16. When the gimbal is performing a tracking task, if the maximum deflection angle deviation occurs, the maximum deflection angle correction will be performed.

[0044] refer to Figure 2 The maximum deflection angle correction includes the following steps:

[0045] S161, The gimbal rotation module continuously tracks the rotation, and the target recognition module identifies the markers;

[0046] S162. After identifying the marker, perform a second marker filtering operation to obtain the current rotation angles corresponding to the left and right markers;

[0047] The second marker filtering refers to selecting markers that conform to the second preset filtering rules;

[0048] S163. Calculate the current maximum deflection angle based on the current rotation angle and update the position; return to S11.

[0049] As one embodiment, the target recognition module can achieve marker recognition through a target detection algorithm.

[0050] The target detection algorithm may be any one or a combination of Faster R-CNN, YOLO, SSD, RetinaNet, Mask R-CNN, Cascade R-CNN, and CenterNet.

[0051] The above-mentioned target detection algorithms are all mature technical solutions. Those skilled in the art can use the above-mentioned target detection algorithms to successfully implement the marker recognition function in S12 and S161 based on the description of the embodiments, and will not be described in detail here.

[0052] As one embodiment, the preset direction in S11 is clockwise, and the preset adjustment angle is 10 degrees.

[0053] The upper limit of the rotation angle is 360 degrees, which is one full rotation.

[0054] The preset correction parameter has a value range of 0-10 degrees.

[0055] As one embodiment, the method can be applied to a football field to perform adaptive field detection during football play. In this case, the marker is the goal.

[0056] The first preset filtering rule refers to:

[0057] (1) Does the length-to-width ratio of the goal match the current football field?

[0058] For example, if the current match is played on a FIFA standard pitch, the goal's length-to-width ratio is 1:3; if the current match is played on an indoor football pitch, the goal's length-to-width ratio is 1:2. When performing pitch adaptation testing, the parameters should be adjusted according to the match pitch.

[0059] (2) Whether the goal is located on the horizontal sideline of the field.

[0060] If all the above conditions are met, the marker (goal) conforms to the first preset filtering rule. At this time, the rotation angle corresponding to the marker (goal) should be recorded.

[0061] The threshold value for the marker is 2, which means that there should be two goals on the football field.

[0062] The preset angle range is determined based on the location of the gimbal, such as... Figure 3 As shown. The gimbal can be placed in the following positions. Figure 3 Within the semicircular area marked in the middle, the upper limit (maximum angle) of the preset angle range is the angle between the two goals when the gimbal is placed at the bottom of the arc (point A) (i.e., max); the lower limit (minimum angle) of the preset angle range is the angle between the two goals when the gimbal is placed at the top of the arc (point B) (i.e., min).

[0063] The second preset filtering rule refers to:

[0064] (1) Does the length-to-width ratio of the goal match the current football field?

[0065] (2) Whether the difference between the current rotation angle of the goal and the recorded previous rotation angle is less than the preset deviation threshold.

[0066] If all the above conditions are met, then the marker (goal) conforms to the second preset filtering rule. At this time, the marker (goal) is either the original left-side marker or the original right-side marker in the previous record.

[0067] As one embodiment, the method can also be applied to basketball courts to perform adaptive court detection in five-a-side basketball. In this case, the marker is a basketball hoop.

[0068] The first preset filtering rule refers to:

[0069] (1) Does the height of the basketball hoop match the current basketball court?

[0070] (2) Whether the basketball hoop is located on the sideline of the court.

[0071] If all the above conditions are met, the marker (basketball hoop) conforms to the first preset filtering rule. At this time, the rotation angle corresponding to the marker (basketball hoop) should be recorded.

[0072] The threshold value for the marker is 2, which means that there should be 2 basketball hoops in the basketball court.

[0073] The preset angle range is determined based on the location of the gimbal, and the method for obtaining the preset angle range of a football field is the same.

[0074] The preset correction parameter has a value range of 0-10 degrees.

[0075] The second preset filtering rule refers to:

[0076] (1) Does the height of the basketball hoop match the current basketball court?

[0077] (2) Whether the difference between the current rotation angle of the basketball hoop and the recorded previous rotation angle is less than the preset deviation threshold.

[0078] If all the above conditions are met, then the marker (basketball hoop) conforms to the second preset filtering rule. At this time, the marker (basketball hoop) is either the original left-side marker or the original right-side marker in the previous record.

[0079] It should be noted that, apart from the aforementioned football and basketball fields, the method described above can be used for adaptive field detection in any sport where there are field boundaries and obvious features (markers) at the boundaries, requiring only the adjustment of the corresponding parameters according to the field.

[0080] Another aspect of the present invention is a site-adaptive detection device, with reference to Figure 4 As shown, the device includes a gimbal rotation module, a gimbal operating system module, and an image acquisition module.

[0081] The gimbal rotation module is the rotating structure of the gimbal, which can rotate at a corresponding angle according to the control commands of the adjustment control module.

[0082] The gimbal operating system module includes a target detection module and an adjustment control module, and is a software system for detection, identification, and control; specifically,

[0083] The target detection module is used to detect landmarks in the scene image;

[0084] The adjustment and control module is used to calculate the maximum deflection angle based on the marker detection results. At the same time, it can send control commands to the gimbal rotation module and the image acquisition module based on the marker detection results to control the corresponding modules.

[0085] The image capture module is used to capture scene images and record videos, and can be implemented by the shooting module of a camera, video camera, or mobile terminal.

[0086] As one embodiment, the method described in this invention can be implemented in software and / or a combination of software and hardware, for example, using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device.

[0087] The method described in this invention can be implemented as a software program, which can be executed by a processor to achieve the steps or functions described above. Similarly, the software program (including associated data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices.

[0088] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0089] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0090] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A site adaptive detection method, characterized in that, The method includes: S11. The gimbal rotation module rotates in a preset direction and adjusts to a preset angle. S12. Perform landmark recognition in the central region of the scene image to obtain the landmark recognition result; if no landmark is recognized, return to S11. S13. Based on the marker recognition results, perform the first marker filtering, record the rotation angle corresponding to the filtered marker, and return to S11. The first marker filtering refers to selecting markers that conform to the first preset filtering rules; S14. When the cumulative rotation angle of the gimbal rotation module reaches the upper limit angle, obtain several markers and their corresponding rotation angles; if the number of markers is greater than or equal to the marker threshold, then execute S15. S15. Calculate the included angle between any two markers, select two markers that fit within the preset angle range, record their rotation angles, and calculate the maximum deflection angle; wherein, the preset angle range is determined based on the location of the gimbal; when performing field adaptive detection on a football field, the preset angle range is determined in the following way: the upper limit of the preset angle range is the included angle between the two goals when the gimbal is placed at the bottom of the arc; the lower limit of the preset angle range is the included angle between the two goals when the gimbal is placed at the top of the arc. The maximum deflection angle consists of the maximum leftward deflection angle and the maximum rightward deflection angle. The method for calculating the maximum deflection angle is as follows: in α left , α right These represent the maximum leftward deflection angle and the maximum rightward deflection angle, respectively. β left , β right These are the rotation angles of the left and right signs, respectively. s Preset correction angle; S16. When the gimbal performs a tracking task, if a maximum deflection angle deviation occurs, a maximum deflection angle correction is performed; the maximum deflection angle correction includes the following steps: S161, The gimbal rotation module continuously tracks the rotation, and the target recognition module identifies the markers; S162. After identifying the marker, perform a second marker filtering operation to obtain the current rotation angles corresponding to the left and right markers; the second marker filtering refers to filtering out markers that meet the second preset filtering rules. S163. Calculate the current maximum deflection angle based on the current rotation angle and update the position; return to S11; The second preset filtering rule refers to the requirement that the marker must simultaneously meet the following conditions: (1) Whether the size of the markers at the site boundary matches the current site; (2) Whether the difference between the current rotation angle corresponding to the marker and the recorded previous rotation angle is less than the preset deviation threshold.

2. The method according to claim 1, characterized in that, The scene images were captured by the image acquisition module; The marker recognition results include whether a marker was detected, the marker's size characteristics, and the rotation angle. The target recognition module identifies landmarks through target detection algorithms; The target detection algorithm may be any one or a combination of Faster R-CNN, YOLO, SSD, RetinaNet, Mask R-CNN, Cascade R-CNN, and CenterNet.

3. The method according to claim 1, characterized in that, When performing adaptive field detection on a football field, the first preset filtering rule requires that the markers simultaneously meet the following two conditions: (1) Does the length-to-width ratio of the goal match the current football field? (2) Whether the goal is located on the horizontal sideline of the field.

4. The method according to claim 1, characterized in that, The preset direction is clockwise, and the preset adjustment angle is 10 degrees. The upper limit of the rotation angle is 360 degrees.

5. The method according to claim 1, characterized in that, S14 also includes: if the number of markers is less than the marker threshold, then the maximum deflection angle is... .

6. The method according to claim 1, characterized in that, When performing field-adaptive detection on a football field, the marker threshold is 2.

7. A site-adaptive detection device, characterized in that... , The device includes a gimbal rotation module, a gimbal operating system module, and an image acquisition module; The gimbal rotation module is the rotating structure of the gimbal, which can rotate at a corresponding angle according to the control command of the adjustment control module. The gimbal operating system module includes a target detection module and an adjustment and control module; The target detection module is used to detect landmarks in the scene image; The adjustment and control module is used to calculate the maximum deflection angle based on the marker detection results, and can also send control commands to the gimbal rotation module and the image acquisition module based on the marker detection results. When the gimbal performs a tracking task, if the maximum deflection angle deviates, the maximum deflection angle correction is performed. The maximum deflection angle correction includes: the gimbal rotation module continuously tracks the rotation, and the target recognition module identifies the marker. After the marker is identified, a second marker filtering operation is performed to obtain the current rotation angle corresponding to the left and right markers; the current maximum deflection angle is calculated based on the current rotation angle and the position is updated. The second marker filtering refers to selecting markers that conform to the second preset filtering rule; the second preset filtering rule means that the markers must simultaneously meet the following conditions: (1) Whether the size of the markers at the site boundary matches the current site; (2) Whether the difference between the current rotation angle corresponding to the marker and the recorded previous rotation angle is less than the preset deviation threshold; The image capture module is used to capture scene images.

Images