A method, device, readable storage medium and electronic device for around-the-stick testing

Abstract

The application provides a method and device for around-the-pole test, a computer readable storage medium and an electronic device. The method comprises: determining an expected area sequence in response to a trigger timing start event; identifying images to determine a first area where a ball object is located and a second area where a ball carrier object is located; performing a violation process in response to a change in the first area determined by adjacent two frames of images, and the changed first area being a first expected area in the expected area sequence; performing a violation process in response to a change in the second area determined by adjacent two frames of images, and the changed second area being a second expected area in the expected area sequence; and confirming a test result indicating that the around-the-pole test is passed in response to the ball carrier object and the ball object entering an end area, and a first area sequence composed of non-repeated first areas in sequence and a second area sequence composed of non-repeated second areas in sequence both meeting the expected area sequence.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of video playing, and more particularly, to a winding pole test method, device, readable storage medium and electronic equipment. BACKGROUND

[0002] The S-shaped winding pole is a common ball training test item in some ball sports. At present, when the S-shaped winding pole test is performed, the artificial test method is usually adopted, and the test scores of the test personnel are artificially judged by the invigilators. This traditional artificial test and invigilation method has a large subjective factor, and cannot accurately evaluate the position between the ball and the ball personnel, so the possibility of misjudgment is high, and the accuracy of the test result is low. SUMMARY

[0003] The present application provides a winding pole test method, device, computer readable storage medium and electronic equipment to solve the technical problem of low accuracy of the test result of the winding pole test in the prior art.

[0004] According to a first aspect of the present application, a winding pole test method is provided, comprising:

[0005] In response to a trigger timing start event, an expected region sequence is determined based on the departure direction of the ball object;

[0006] The real-time collected image is identified to determine the first region where the ball object is located and the second region where the ball object is located;

[0007] In response to the change of the first region determined by the adjacent two frames of images, and the changed first region is not the first expected region in the expected region sequence, a violation processing is performed, and the first expected region is determined based on the first region before the change;

[0008] In response to the change of the second region determined by the adjacent two frames of images, and the changed second region is not the second expected region in the expected region sequence, a violation processing is performed, and the second expected region is determined based on the second region before the change;

[0009] In response to the ball object and the ball object entering the end region, and the first region sequence composed of non-repeated first regions in sequence and the second region sequence composed of non-repeated second regions in sequence both conform to the expected region sequence, a test result for indicating that the winding pole test is passed is confirmed.

[0010] Optionally, the violation processing comprises:

[0011] In response to the test rule allowing retesting, determining a violation handling when the ball object corresponds to a first valid area and the ball carrier object corresponds to a second valid area;

[0012] Based on the sequence order of the first valid area and the second valid area in the expected area sequence, the target valid area is determined in sequence;

[0013] Prompt the ball carrier object to carry the ball object back to the target valid area, and update the first expected area and the second expected area based on the target valid area.

[0014] Optionally, the violation handling includes:

[0015] In response to the ball object not being in the test area and the test rule not allowing retesting, a test result indicating an out-of-bounds violation is determined;

[0016] In response to the test rule not allowing retesting, a test result indicating a around-the-stick violation is determined.

[0017] Optionally, the real-time image recognition determines the first area where the ball object is located and the second area where the ball carrier object is located, including:

[0018] Obtain images captured by at least two image acquisition devices arranged at predetermined positions;

[0019] Identify the ball object and the ball carrier object in the image to determine the first spatial information corresponding to the ball object and the second spatial information corresponding to the ball carrier object;

[0020] Based on the first spatial information corresponding to the ball object, the first projection information of the ball object is determined, and based on the first projection information, the first area where the ball object is located is determined;

[0021] Based on the second spatial information corresponding to the ball carrier object, the second projection information of the ball carrier object is determined, and based on the second projection information, the second area where the ball carrier object is located is determined.

[0022] Optionally, the second projection information is used to determine the second area where the ball carrier object is located, including:

[0023] In the case that the second projection information corresponds to at least two divided areas, the projection proportion corresponding to each divided area is determined;

[0024] Based on the projection proportion corresponding to each divided area, the second area where the ball carrier object is located is determined among the at least two divided areas.

[0025] Optionally, the ball object and the ball-carrying object in the image are identified, and first spatial information corresponding to the ball object and second spatial information corresponding to the ball-carrying object are determined, including:

[0026] The ball object, the ball-carrying object, and a stand object in the image are identified, and original spatial information corresponding to the ball object, original spatial information corresponding to the ball-carrying object, and third spatial information corresponding to the stand object are determined.

[0027] The original spatial information corresponding to the ball object is corrected based on the annotation data of the stand object and the third spatial information, and the first spatial information corresponding to the ball object is determined.

[0028] The original spatial information corresponding to the ball-carrying object is corrected based on the annotation data of the stand object and the third spatial information, and the second spatial information corresponding to the ball-carrying object is determined.

[0029] Optionally, when the real-time collected image is identified, the method further includes:

[0030] The posture information and the spatial information of the ball-carrying object are determined.

[0031] The motion parameters of the ball-carrying object are determined based on the posture information and the spatial information of the continuous frames.

[0032] An analysis report is generated based on the test result, the posture information, the motion parameters, and historical data.

[0033] According to a second aspect of the present application, a pole-winding test device is provided, including:

[0034] A sequence determination module is configured to determine an expected region sequence based on a starting direction of a ball-carrying object in response to a trigger timing start event.

[0035] An image recognition module is configured to identify a real-time collected image, and determine a first region where a ball object is located and a second region where the ball-carrying object is located.

[0036] A first processing module is configured to perform a violation processing in response to a change in the first region determined based on two adjacent frames of images, and the changed first region is not a first expected region in the expected region sequence, and the first expected region is determined based on the first region before the change.

[0037] a second processing module, configured to, in response to the second region determined based on the two adjacent frames of images changing and the changed second region being different from a second expected region in the expected region sequence, perform a violation processing, the second expected region being determined based on the second region before the change;

[0038] a result determining module, configured to, in response to the dribble object and the ball object both entering the ending region, and a first region sequence composed of non-repeated first regions in sequence and a second region sequence composed of non-repeated second regions in sequence both conforming to the expected region sequence, confirm a test result indicating that the around-the-pole test is passed.

[0039] According to a third aspect of the present application, a computer readable storage medium is provided, the storage medium storing a computer program for executing the above-mentioned around-the-pole test method.

[0040] According to a fourth aspect of the present application, an electronic device is provided, the electronic device comprising:

[0041] a processor;

[0042] a memory for storing executable instructions of the processor;

[0043] the processor is configured to read the executable instructions from the memory and execute the instructions to implement the above-mentioned around-the-pole test method.

[0044] Compared with the prior art, the around-the-pole test method, device, computer readable storage medium and electronic device provided by the present application have at least the following beneficial effects:

[0045] The technical scheme of the present application determines the expected region sequence according to the departure direction of the ball-carrying object after triggering the timing start event, the expected region sequence is composed of correct expected regions in sequence, and is used to indicate the correct ball-carrying route. The real-time collected images are recognized, and the first region where the ball object is located and the second region where the ball-carrying object is located are continuously recognized. When the first region determined by the adjacent two frames of images changes, and the changed first region is not the first expected region in the expected sequence, it indicates that the route of the ball object has a problem, and therefore, the violation processing is performed. The first expected region is determined based on the changed first region, and is the correct expected region that the ball object should reach. When the second region determined by the adjacent two frames of images changes, and the changed second region is not the second expected region in the expected sequence, it indicates that the route of the ball-carrying object has a problem, and therefore, the violation processing is performed. The second expected region is determined based on the changed second region, and is the correct expected region that the ball-carrying object should reach. When the ball-carrying object and the ball object both enter the end region, and the first region sequence composed of non-repeated first regions in sequence and the second region sequence composed of non-repeated second regions in sequence both meet the expected region sequence, it indicates that the ball-carrying object carries the ball object according to the correct route during the ball-carrying process, and therefore, the test result indicating that the around-pole test is passed is confirmed. In the technical scheme of the present application, the automatic around-pole test can be realized, the test efficiency is high, the participation of human factors is avoided in the test process, the accuracy is high, and the test result accuracy is further ensured by monitoring the ball-carrying object and the ball object. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical scheme of the present application, the drawings needed to be used in the description of the present application will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative labor.

[0047] Figure 1 is a flowchart of the around-pole test method provided by an exemplary embodiment of the present application Figure 1 ;

[0048] Figure 2 is a test region plane diagram in the around-pole test method provided by an exemplary embodiment of the present application

[0049] Figure 3 is a partial flowchart of the around-pole test method provided by an exemplary embodiment of the present application Figure 1 ;

[0050] Figure 4 is a partial flowchart of the around-pole test method provided by an exemplary embodiment of the present application Figure 2 ;

[0051] Figure 5 is a local flowchart of a method for testing around a rod provided by an exemplary embodiment of the present application Figure 3 ;

[0052] Figure 6 is a flowchart of a method for testing around a rod provided by an exemplary embodiment of the present application Figure 2 ;

[0053] Figure 7 is a structural diagram of a device for testing around a rod provided by an exemplary embodiment of the present application

[0054] Figure 8 is a structural diagram of an electronic device provided by an exemplary embodiment of the present application DETAILED DESCRIPTION

[0055] The technical solutions in the present application will be described clearly and completely below in conjunction with the drawings in the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0056] Exemplary method

[0057] Figure 1 is a flowchart of a method for testing around a rod provided by an exemplary embodiment of the present application, applied to an analysis terminal or a server, which can be a cloud server, and the method comprises at least the following steps:

[0058] Step 11, in response to a timing start event, determining an expected region sequence based on a departure direction of the ball object.

[0059] The timing start event is used to indicate the start of the test around the rod. Specifically, the timing start event can be a start instruction given by an invigilator. In one possible application scenario, the invigilator gives the start instruction through a mobile terminal, the mobile terminal reports the timing start event to the analysis terminal after receiving the start instruction, and the analysis terminal responds after receiving the timing start event. The timing start event can also be a position change of a ball object such as a football located in the departure region. In one possible application scenario, an image acquisition device performs real-time image acquisition, identifies the ball object in the real-time acquired images, and triggers the timing start event when the position of the football located in the departure region changes between two adjacent images. The timing start event can also be a specified time. In one possible application scenario, a specified time is set in advance, and the timing start event is triggered when the specified time is reached.

[0060] The dribbling object is a testee, such as a student. After the timing starts, the dribbling object starts to leave the starting area. Different starting directions of the dribbling object determine different dribbling routes of the dribbling object, and thus different expected area sequences, which are composed of correct expected areas in sequence and used to indicate correct dribbling routes.

[0061] As shown in the example, Figure 2 two vertical intersecting directions, to achieve stereoscopic vision of the student and the football. The areas are divided according to the extensions of the vertical rods in the X and Y directions, to obtain 10 areas, i.e., 1A, 1B, 2A, 2B, 3A, 3B, 4A, 4B, 5A, and 5B. If the student starts to run around the right side of the first vertical rod, the expected area sequence corresponding to the whole process is 1B, 1A, 2A, 2B, 3B, 3A, 4A, 4B, 5B, and 5A. If the student starts to run around the left side of the first vertical rod, the expected area sequence corresponding to the whole process is 1A, 1B, 2B, 2A, 3A, 3B, 4B, 4A, 5A, and 5B.

[0062] Step 12, identifying the real-time collected images to determine the first area where the ball object is located and the second area where the dribbling object is located.

[0063] Specifically, the real-time collected images are identified to identify the first area where the ball object, such as a football or a basketball, is located and the second area where the dribbling object is located. The first area and the second area are one or more of the pre-divided areas or out-of-area areas. The first area and the second area can be the same or different. For example, when the dribbling object starts to run around the right side of the first vertical rod, the dribbling object and the ball object in the collected images can both be in the 1B area, and thus the first area and the second area are the same. As the dribbling object advances with the ball, the ball object can have entered the 1A area while the dribbling object is still in the 1B area, and thus the first area and the second area are different.

[0064] In some embodiments, the step 12 of identifying the real-time collected images to determine the first area where the ball object is located and the second area where the dribbling object is located includes:

[0065] Step 121, acquiring images collected by at least two image acquisition devices arranged at preset positions.

[0066] Specifically, the images are acquired by using at least two image acquisition devices arranged at preset positions, such as Figure 2As shown, one camera is installed in each of the XY two directions of the top plane of the test area, and image acquisition is performed synchronously using the two cameras, wherein each camera can capture images at a speed of 30 fps (30 frames per second).

[0067] In step 122, the ball object and the ball carrier object in the image are identified, and the first spatial information corresponding to the ball object and the second spatial information corresponding to the ball carrier object are determined.

[0068] Specifically, after obtaining the images synchronously collected by the at least two image collection devices, image recognition is performed on at least two images at the same time to determine the first spatial information of the ball object, which is used to indicate the position information of the ball object in the three-dimensional space; and the second spatial information corresponding to the ball carrier object is determined, which is used to indicate the position information of the ball carrier object in the three-dimensional space.

[0069] In some embodiments, step 122 identifies the ball object and the ball carrier object in the image, and determines the first spatial information corresponding to the ball object and the second spatial information corresponding to the ball carrier object, including:

[0070] In step 1221, the ball object, the ball carrier object, and the stand object in the image are identified, and the original spatial information corresponding to the ball object, the original spatial information corresponding to the ball carrier object, and the third spatial information corresponding to the stand object are determined.

[0071] Specifically, when the image is identified, not only the ball object and the ball carrier object in the image are identified, but also the stand object in the image is identified, to obtain the original spatial information of the ball object in the three-dimensional space, the original spatial information of the ball carrier object in the three-dimensional space, and the third spatial information of the stand object in the three-dimensional space.

[0072] In step 1222, the original spatial information corresponding to the ball object is corrected based on the annotation data of the stand object and the third spatial information, and the first spatial information corresponding to the ball object is determined.

[0073] In step 1223, the original spatial information corresponding to the ball carrier object is corrected based on the annotation data of the stand object and the third spatial information, and the second spatial information corresponding to the ball carrier object is determined.

[0074] Specifically, since the standing pole object has measurability, the length of the standing pole object is determined in advance, and the coordinate information of the standing pole object on the test ground is determined, and the length information and the coordinate information of the standing pole object are taken as the labeling data of the standing pole object. After determining the labeling data of the standing pole object, the original spatial information corresponding to the ball object is corrected by using the labeling data and the third spatial information to obtain the first spatial information corresponding to the ball object; and the original spatial information corresponding to the ball carrier object is corrected by using the labeling data and the third spatial information to determine the second spatial information corresponding to the ball carrier object.

[0075] Exemplarily, the correction coefficient is determined based on the labeling data and the third spatial information, such as dividing the coordinate value of the third spatial information by the coordinate value of the labeling data to obtain the correction coefficient, and the original spatial information corresponding to the ball object is further corrected based on the correction coefficient to obtain the first spatial information corresponding to the ball object, and the original spatial information corresponding to the ball carrier object is corrected based on the correction coefficient to determine the second spatial information corresponding to the ball carrier object.

[0076] In the embodiment, after the original spatial information of the ball object and the original spatial information of the ball carrier object are corrected by the labeling data of the standing pole object and the third spatial information, the accuracy of the first spatial information and the second spatial information can be further improved, which provides a prerequisite for accurately determining the first region and the second region subsequently.

[0077] In a possible application scenario, as shown in Figure 3 The image frame captured by the camera 1 is obtained, and the image frame captured by the camera 2 is obtained, the two obtained image frames are time-synchronized, the football (i.e. the ball object), the student's torso (i.e. the ball carrier object) and the standing pole in the image frame captured by the camera 1 are identified; the football, the student's torso and the standing pole in the image frame captured by the camera 2 are identified, so as to perform three-dimensional positioning based on the binocular principle, to determine the original spatial information of the football and the original spatial information of the student's torso, and to correct by using the known standing pole position to obtain the first spatial information of the football and the second spatial information of the student's torso. The accuracy of the first spatial information and the second spatial information is relatively high.

[0078] In step 123, the first projection information of the ball object is determined based on the first spatial information corresponding to the ball object, and the first region where the ball object is located is determined based on the first projection information.

[0079] Specifically, in order to accurately determine the first region where the ball object is located, after determining the first spatial information corresponding to the ball object, the first projection information on the test ground is determined according to the first spatial information, the first projection information is used to indicate the projection coordinates of the ball object on the test ground, and the partition region containing the first projection information is further determined, and the partition region is taken as the first region where the ball object is located.

[0080] When the first projection information corresponds to at least two partition regions, the projection proportion of the first projection information corresponding to each partition region is determined, and based on the projection proportion of each partition region, the first region where the ball object is located is determined from the at least two partition regions. For example, when the projection proportion of the ball object in a certain partition region is more than 80%, the partition region is considered as the first region where the ball object is located.

[0081] Step 124, determining the second projection information of the ball object based on the second spatial information corresponding to the ball object, and determining the second region where the ball object is located based on the second projection information.

[0082] Specifically, in order to accurately determine the second region where the ball object is located, after determining the second spatial information corresponding to the ball object, the second projection information on the test ground is determined according to the second spatial information, the second projection information is used to indicate the projection coordinates of the ball object on the test ground, and the partition region containing the second projection information is further determined, and the partition region is taken as the second region where the ball object is located.

[0083] In some embodiments, the step 124 of determining the second region where the ball object is located based on the second projection information comprises:

[0084] Step 1241, when the second projection information corresponds to at least two partition regions, the projection proportion of each partition region is determined.

[0085] Step 1242, based on the projection proportion of each partition region, the second region where the ball object is located is determined from the at least two partition regions.

[0086] Specifically, because the volume of the ball object is large, the second projection information may correspond to at least two partition regions, therefore the projection proportion of each partition region is determined, and then the second region where the ball object is located is determined from the at least two partition regions according to the projection proportion of each partition region.

[0087] Exemplarily, the projection proportion can be calculated by the projection area. When the second projection information of the dribbling object is determined according to the second spatial information corresponding to the dribbling object, the projection of the vertical direction of the trunk of the dribbling object on the ground can be a shadow area on the ground, the distribution of the shadow area in each divided region is determined, the shadow areas in each divided region are compared, and the projection proportion corresponding to each divided region is obtained. When the projection proportion corresponding to a divided region is greater than 80%, the divided region is considered as the second region where the dribbling object is located. When each projection proportion is less than 80%, the divided region with the largest projection proportion is determined as the second region where the dribbling object is located.

[0088] In a possible application scenario, because the dribbling object may be too close to the stand during the dribbling around the stand, the projection area of the trunk of the dribbling object on the ground may cross different regions. Therefore, the dribbling object is allowed to be in 1 to 3 regions in the calculation. When the dribbling object corresponds to 2 to 3 regions, the proportion of the dribbling object in each region is calculated to determine the second region. It should be noted that due to detection errors, the trunk of the dribbling object may be projected onto all four regions in the calculation process, but considering the physical barrier of the stand, the trunk of the dribbling object cannot actually be projected onto four regions. Therefore, the occlusion relationship between the trunk of the dribbling object and the stand is used to remove one region that does not conform to the actual situation. After removing the region that does not conform, the projection proportion corresponding to each divided region in the second projection information is calculated to accurately determine the second region where the dribbling object is located.

[0089] Step 13, in response to the first region determined according to the adjacent two images changing and the changed first region being a first expected region in the expected region sequence, performing a violation processing, the first expected region being determined based on the first region before the change.

[0090] Specifically, because the image acquisition device acquires images in real time, new images are continuously acquired, and the acquired images are all identified to identify the first region where the ball object is located. When the first regions determined according to adjacent two images change, it is judged whether the changed first region is a first expected region. The first expected region is determined in the expected region sequence according to the first region before the change, and is used to indicate the correct expected region. Therefore, when the changed first region is not the first expected region in the expected region sequence, it indicates that the route of the dribbling object has a problem, and therefore a violation processing needs to be performed.

[0091] Exemplarily, as Figure 2As shown, images captured by two cameras at the same time are identified to determine the first region where the ball is located, such as 1B. According to the expected region sequence, the first expected region corresponding to 1B is 1A. When the ball is dribbling, it rolls forward and there is a moment when the ball leaves the 1B region. At this time, when the images are identified, the first region where the ball is located in the two adjacent frames changes. If the changed first region is 2B, and the changed first region 2B is not the first expected region 1A, then the ball dribbling object has a route problem and needs to be dealt with as a violation.

[0092] In some embodiments, the method further includes responding to a change in a first region determined from two adjacent frames, and if the changed first region is the first expected region in the expected region sequence, then no violation is detected, and the pole-driving test continues, or step 14 continues.

[0093] In some embodiments, step 13 includes handling violations, including:

[0094] Step 131: In response to the test rules allowing retesting, when determining the violation handling, the first valid area corresponding to the ball object and the second valid area corresponding to the dribbling object.

[0095] The first valid area is used to indicate the last correct area for the ball before the violation; the second valid area is used to indicate the last correct area for the dribbling player before the violation.

[0096] Specifically, when handling violations, first determine whether the testing rules allow retesting. If retesting is allowed, determine the first valid area corresponding to the ball and the second valid area corresponding to the dribbling object.

[0097] Step 132: Based on the sequence order of the first effective region and the second effective region in the expected region sequence, the region with the earlier sequence is determined as the target effective region.

[0098] Specifically, after determining the first effective region and the second effective region, the sequence order of the first effective region in the expected region sequence and the sequence order of the second effective region in the expected region sequence are determined. Then, in the first effective region and the second effective region, the one with the earlier sequence order is determined as the target effective region. The target effective region is the retreat region of the dribbling object and the ball object.

[0099] Step 133: Prompt the dribbling object to return to the target effective area with the ball, and update the first expected area and the second expected area based on the target effective area.

[0100] Specifically, after the target effective area is determined, the dribbling object is prompted to carry the ball object back to the target effective area, and the first expected area and the second expected area are updated based on the target effective area, so that the subsequent test process can proceed normally.

[0101] Further, when the first area or the second area is identified, the first area and the second area are recorded, and then the area after the target effective area in the recording content can be deleted, so that the subsequent test process can proceed normally.

[0102] For example, as shown in Figure 2 the student and the football are currently in area 1A, and the new expected area should be 2A. The student mistakenly kicks the football into area 2B, which does not meet the sequence team requirements, so the student is required to return the ball to the target effective area. At this time, the second effective area of the student is 1A, and the first effective area of the football is 1A. Therefore, the target effective area 1A is determined, and the first expected area of the ball object is set to 2A, and the second expected area of the dribbling object is set to 2A. The first expected area of the ball object and the second expected area of the dribbling object are effectively returned, and the subsequent test is ensured to proceed normally.

[0103] In some embodiments, the violation processing includes:

[0104] Step 134, in response to the test rule not allowing retesting, determining a test result for indicating a around-the-pole violation.

[0105] Specifically, in the case where the test rule does not allow retesting, the dribbling object stops continuing the dribbling test, and the analysis on the dribbling object is no longer continued. After the dribbling object leaves the test site from the end area, if the timing start event is triggered again, step 11 is re-executed.

[0106] In some embodiments, the violation processing includes:

[0107] Step 135, in response to the ball object not being in the test area and the test rule not allowing retesting, determining a test result for indicating an out-of-bounds violation.

[0108] Specifically, when the ball object is not in the test area, if the test rule does not allow retesting, a test result for indicating an out-of-bounds violation is determined. Further, the dribbling object stops continuing the dribbling test, and the analysis on the dribbling object is no longer continued. After the dribbling object leaves the test site from the end area, if the timing start event is triggered again, step 11 is re-executed.

[0109] It should be noted that, in the case that the ball object is not in the test area determined by the two adjacent frames of images, the first area changes, and the changed first area is not the first expected area in the expected area sequence, one of the possible cases, so if the test rule allows retesting, step 131 is performed.

[0110] In some embodiments, in response to not detecting the ball object or the ball-carrying object in the current image, the reason can be that the object is blocked or not in the shooting range, at this time, detection will be performed on the image at the next time point, and when the number of detection failures accumulates to a set threshold, it will be handled as an exception, and a prompt information is generated to prompt the relevant staff to perform equipment maintenance.

[0111] In a possible implementation, a current frame image is obtained, a first area of a ball object in the current frame image is determined, a first expected area in an expected area sequence is determined according to the first area; a next frame image is obtained, a first area of a ball object in the next frame image is determined, and a next first expected area in the expected area sequence is determined according to the first area. When the first area changes, if the changed first area is the first expected area (here, it is not the next first expected area, but the first expected area corresponding to the current frame image), no violation is detected, and if the changed first area is not the first expected area, a violation is detected, and a violation handling is performed. When the violation handling is performed, when the test rule allows retesting, a target valid area is determined, so that the ball-carrying object carries the ball object back to the target valid area, and the next first expected area is rolled back to the first expected area corresponding to the target valid expected area, and subsequent detection is continued.

[0112] In step 14, in response to the second area determined by the two adjacent frames of images changing, and the changed second area being not a second expected area in the expected area sequence, a violation handling is performed, and the second expected area is determined based on the second area before the change.

[0113] Specifically, because the image acquisition device acquires images in real time, new images are continuously acquired, and the acquired images are all identified to identify the second area where the ball-carrying object is located. If the second area determined by the two adjacent frames of images changes, it is judged whether the changed second area is a second expected area, and the second expected area is determined in the expected area sequence according to the second area before the change, which is used to indicate a correct expected area. Therefore, when the changed second area is not the second expected area in the expected area sequence, it indicates that the route of the ball-carrying object has a problem, and therefore a violation handling needs to be performed.

[0114] Exemplarily, as Figure 2As shown, the images captured by the two cameras at the same time are identified to determine the second region where the dribbling object is located, such as 1B. According to the expected region sequence, it can be known that the first expected region corresponding to 1B is 1A. The dribbling object moves forward before dribbling. There is a moment when the dribbling object leaves the 1B region, that is, when the images are identified, the second regions where the dribbling objects are located in the adjacent two frames of images change. If the changed second region is 2B, the changed second region 2B is not the second expected region 1A, then the dribbling object is in violation when dribbling, and needs to be handled as a violation.

[0115] In some embodiments, the method further comprises, in response to the change of the second region determined by the adjacent two frames of images, and the changed second region being a second expected region in the expected region sequence, no violation is detected, and the around-the-pole test continues.

[0116] It should be noted that step 13 is used to monitor the ball object, and when the route of the ball object has a problem, there may be a violation of dribbling. Step 14 is used to monitor the dribbling object, and when the route of the dribbling object has a problem, there may be a violation of dribbling. Further, steps 13 and 14 can be executed sequentially, that is, the route of the ball object is determined, and then the route of the dribbling object is determined. Of course, steps 13 and 14 can also be executed simultaneously, or step 14 is executed after step 13, and the present embodiment does not make specific limitations thereon.

[0117] In some embodiments, the violation handling in step 14 includes:

[0118] Step 141, in response to determining that the violation handling is allowed to retest according to the test rule, determining the first valid region corresponding to the ball object and the second valid region corresponding to the dribbling object.

[0119] Wherein, the first valid region is used to indicate the last correct region of the ball object before the violation; and the second valid region is used to indicate the last correct region of the dribbling object before the violation.

[0120] Specifically, when the violation is handled, it is first determined whether the test rule allows retesting. When retesting is allowed, the first valid region corresponding to the ball object and the second valid region corresponding to the dribbling object are determined respectively.

[0121] Step 142, based on the sequence order of the first valid region and the second valid region in the expected region sequence, determining the target valid region in sequence first.

[0122] Specifically, after determining the first effective area and the second effective area, the sequence order of the first effective area in the expected area sequence and the sequence order of the second effective area in the expected area sequence are determined, and then in the first effective area and the second effective area, the one with the sequence order first is determined as the target effective area, which is the retreat area of the dribbling object and the ball object.

[0123] Step 143, prompting the dribbling object to carry the ball object back to the target effective area, and updating the first expected area and the second expected area based on the target effective area.

[0124] Specifically, after determining the target effective area, the dribbling object is prompted to carry the ball object back to the target effective area, and the first expected area and the second expected area are updated based on the target effective area, so that the subsequent test process can proceed normally.

[0125] Further, when the first area or the second area is identified, the first area and the second area are recorded, and then the area after the target effective area in the recording content can be deleted to ensure that the subsequent test process can proceed normally.

[0126] For example, as shown in Figure 2 , the student and the football are currently in area 1A, and the expected area of the new entry should be 2A. The student first kicks the football into the 2A area, and at this time the expected area of the football moves to 2B in sequence, while the student directly runs into the 2B area from the 1A area, which does not meet the sequence requirements, so the student is required to return the ball to the target effective area. At this time, the second effective area of the student is 1A, and the first effective area of the football is 2A. Since 1A is in the earlier sequence in the sequence, the target effective area 1A is determined, and the first expected area of the ball object is set to 2A, the second expected area of the dribbling object is set to 2A, and the test is restarted.

[0127] In some embodiments, the violation processing includes:

[0128] Step 144, in response to the test rule not allowing retesting, determining a test result indicating a pole bypass violation.

[0129] Specifically, in the case where the test rule does not allow retesting, a test result indicating a pole bypass violation is determined. Further, in the case where retesting is not allowed, the dribbling object stops continuing the dribbling test, and the analysis on the dribbling object is no longer continued. After the dribbling object leaves the test site from the end area, if the timing start event is triggered again, step 11 is executed again.

[0130] In a possible implementation, a current frame image is acquired, a second region of the dribbling object in the current frame image is determined, a second expected region in the expected region sequence is determined according to the second region; a next frame image is acquired, a second region of the dribbling object in the next frame image is determined, and a next second expected region in the expected region sequence is determined according to the second region. When the second region changes, if the changed second region is the second expected region (here, not the next second expected region, but the second expected region corresponding to the current frame image), no violation is detected, and if the changed second region is not the second expected region, a violation is detected, and a violation processing is performed. When the violation processing is performed, when the test rule allows retesting, a target valid region is determined, so that the dribbling object carries the ball object back to the target valid region, and the next second expected region is rolled back to the second expected region corresponding to the target valid expected region, and subsequent detection is continued.

[0131] Exemplarily, as shown in Figure 4 When the violation detection is performed, the ball object can be mainly used, and when two images acquired at the same time are detected for violation, the following specific steps exist:

[0132] Step 1, new images captured by two cameras are acquired, target positioning is performed on the new images, wherein the target includes a football and a student, that is, a first region where the ball object is located and a second region where the dribbling object is located are determined; if the football is successfully positioned, that is, the first region of the football is determined, step 2 is performed; if the football detection is abnormal, that is, the football is not detected, step 3 is performed.

[0133] Step 2, if it is indicated according to the first region that the football is in the court, information is recorded, the first region is recorded, and step 4 is performed; if it is indicated according to the first region that the football is out of the court, whether the test rule allows retry is further judged, if the test rule does not allow retry, it is determined that the out-of-court violation ends the current process; if the test rule allows retry, the student carries the football back to the target valid region, and a next new image is acquired to reposition the target and continue the test.

[0134] Step 3, if the student is successfully positioned, that is, the second region of the student is determined, information is recorded, the first region is recorded, and step 4 is further performed; if the student positioning fails, it is determined that the detection is abnormal, and in the case that the number of abnormal times reaches a threshold, the positioning is abnormal and is exited. In the case that the number of abnormal times does not reach the threshold, a next new image is acquired to reposition the target and continue the test.

[0135] Step 4: Check if the soccer ball or student has entered a new area. Specifically, check if the first area determined by two adjacent frames has changed, or if the second area determined by two adjacent frames has changed. If the soccer ball or student has not entered a new area, acquire the next frame and re-locate the target, continuing the test. If the soccer ball or student has entered a new area, proceed to Step 5.

[0136] Step 5: Determine whether the football area conforms to the sequence, i.e., the expected area sequence, i.e., whether the change of the first area conforms to the expected area sequence. If the football area does not conform to the sequence, i.e., the expected area sequence, then proceed to step 6; if the football area conforms to the sequence, i.e., the expected area sequence, then the expected football area is updated according to the sequence, and step 7 is executed.

[0137] Step 6: Determine whether the rules allow retry. If retry is not allowed, it is judged as a violation of the pole-driving rule. If retry is allowed, the student carries the soccer ball back to the target valid area, and the soccer ball student expected area is backed up. That is, the first expected area and the second expected area are updated using the target valid area, and the next frame of new image is obtained to relocate the target and continue the test.

[0138] Step 7: Determine whether the student region conforms to the sequence queue, that is, whether the change in the second region conforms to the second expected region. If the student region does not conform to the sequence queue, i.e., the expected region sequence, then proceed to step 6; if the student region conforms to the sequence queue, i.e., the expected region sequence, then the student expected region is updated according to the sequence queue, and step 8 is executed.

[0139] Step 8: Determine whether both the soccer ball and the student are within the expected area. If they are, it is determined that no violation has been detected. If neither the soccer ball nor the student is within the expected area, then obtain the next frame of the new image to re-locate the target and continue the test.

[0140] By following the steps above, precise and automated monitoring can be performed on dribbling objects and ball-like objects in two frames of images acquired at the same time, with high efficiency and accuracy.

[0141] Step 15: In response to the fact that both the dribbling object and the ball object have entered the end region, and the first region sequence consisting of non-repeating first regions in sequence and the second region sequence consisting of non-repeating second regions in sequence both conform to the expected region sequence, confirm the test result used to indicate that the dribbling test has passed.

[0142] Specifically, when the ball object and the dribbling object both enter the ending area, it indicates that the dribbling of the dribbling object is completed, and a first area sequence composed of non-repeated first areas in sequence and a second area sequence composed of non-repeated second areas in sequence are determined, and the first area sequence and the second area sequence are compared with the expected area sequence. In the case that the first area sequence and the second area sequence both conform to the expected area sequence, it indicates that the dribbling object has taken the correct dribbling route in the dribbling process, and there is no illegal dribbling, so the test result for indicating that the around-the-pole test is passed is confirmed.

[0143] It should be noted that when the first area where the ball object is located and the second area where the dribbling object is located are identified, the first area and the second area are recorded, so there may be repeated first areas and repeated second areas. When the first area sequence is composed, the first areas are arranged in sequence according to the order of recording of each first area, and the repeated first areas are deleted, so that the first area sequence composed of non-repeated first areas in sequence is obtained. When the second area sequence is composed, the second areas are arranged in sequence according to the order of recording of each second area, and the repeated second areas are deleted, so that the second area sequence composed of non-repeated second areas in sequence is obtained.

[0144] Exemplarily, the expected area sequence is 1B, 1A, 2A, 2B, 3B, 3A, 4A, 4B, 5B, 5A, if the first area sequence and the second area sequence are both 1B, 1A, 2A, 2B, 3B, 3A, 4A, 4B, 5B, 5A, the dribbling object dribbles according to the correct route, so as to determine the test result for indicating that the around-the-pole test is passed.

[0145] It should be noted that if the dribbling object has illegal dribbling in the dribbling process and is allowed to be retested, after the dribbling object carries the ball object back to the target valid area, the relevant illegal record is deleted, and the dribbling object subsequently completes the dribbling according to the correct route, at this time, because the relevant illegal record has been deleted, the dribbling object can also obtain the test result for indicating that the around-the-pole test is passed.

[0146] In some embodiments, when the real-time collected images are identified, the method further comprises:

[0147] Step 16, determining the posture information and the spatial information of the dribbling object.

[0148] Specifically, when the image is recognized, a pose recognition algorithm can be used for recognition, the pose recognition algorithm is used to construct the connection between the human body key points and the human body center thereof, the human body center vector can be predicted through a convolutional neural network first, and the human body key points are pointed to the human body center thereof, then the distance of the human body center pointed by the human body key points is analyzed to complete the grouping of the human body key points, the estimation of the human body pose is realized, and the pose information of the dribbling object is obtained.

[0149] In a possible implementation, when the pose information of the dribbling object is acquired, human joints are recognized first, the human joints include knees, ankles, and hip, shoulder, elbow, and wrist joints, and the human body posture is recognized based on the joint positions, including leg, foot, torso, arm angle, etc.

[0150] Further, when the image is recognized, the spatial information of the moving object can also be determined, the spatial information is used to indicate the coordinate position of the moving object in the three-dimensional space.

[0151] Step 17, determining the motion parameter of the dribbling object based on the pose information and the spatial information of the continuous frames.

[0152] Specifically, after the pose information and the spatial information of the dribbling object are determined, the pose information and the spatial information of the dribbling object in the continuous frame images are compared, and the motion parameter can be determined. Exemplarily, the motion parameter includes but is not limited to the step length of the moving object, the step frequency of the moving object, the kicking action of the moving object, the ball distance, the arm swing of the moving object, etc.

[0153] In a possible implementation, as shown in FIG. 13, Figure 5 In the frame calculation, interference information needs to be filtered to avoid using data with low confidence and large error in position positioning calculation due to occlusion, shooting angle, etc. In addition, in a single frame picture, a multi-person face regression algorithm is used to detect the multi-person face in the picture, and in continuous multiple frames, the face features and the student position in the last frame are used to ensure that the analysis and detection are not disturbed by other people in the background lens, and the accuracy of the data is improved.

[0154] In a possible implementation, referring to the gait recognition mechanism of machine vision, a skeleton model based on the torso and leg joints is used, the torso position is moved forward, the step length is determined based on the landing position of the ankle joint, and the step frequency is calculated according to the time of each step.

[0155] In a possible implementation, when the kicking action of the foot and the ball distance are determined, the determination method is that the foot and the football position are continuously close, and the football speed appears instantaneous acceleration.

[0156] In one possible implementation, when determining arm swing, the presence of a significant angle between the arm and the torso is used to detect whether the arm is swinging, the amplitude of the arm swing is calculated, and it is determined whether the arm swing is synchronized with the footsteps.

[0157] Step 18: Generate an analysis report based on the test results, the posture information, the motion parameters, and historical data.

[0158] After acquiring the test results, attitude information, and motion parameters, an analysis report is generated by combining historical data. This report provides data on the pole maneuvering process and offers improvement suggestions based on historical data. Furthermore, the analysis report is sent to the target terminal for display.

[0159] In some embodiments, the analysis report includes at least two parts: an objective data section and an exercise recommendation section.

[0160] The objective data section describes the movement of the object around the cones, including: total time, segmented timing of the cones; number of ball touches and ball contact points; running distance, ball distance, number of changes of direction and angle of change of direction; cadence, stride length, and arm swing amplitude; changes in ball speed and running speed; changes in the distance between the person and the ball, and the average, maximum, and variance; and screenshots of key milestones such as starting, navigating the cones, and reaching the finish line.

[0161] The exercise suggestion section uses data analysis to obtain a more ideal exercise model based on the performance of different participants in the test and the corresponding objective motion parameters. This model includes reasonable speed allocation, route allocation, and body posture. Based on this, a stratified analysis is conducted for participants of different skill levels, and improvement suggestions are provided that match their skill level.

[0162] For example, such as Figure 6 As shown, the timing automatically starts when the dribbling object begins dribbling. The analysis terminal acquires images captured in real time by at least two cameras and uses at least four detection algorithms to perform comprehensive detection on the images. The four detection algorithms include target detection, posture detection, face detection, and coordinate calculation. After comprehensive detection of the images, the detection dataset of the person and ball position, human posture, and dribbling parameters is obtained. The calculation ends automatically after both the person and ball reach the end area. The recorded person and ball positions, i.e., the first area sequence and the second area sequence, are compared with the expected area sequence to determine the violation. An analysis report is generated based on the violation determination result, human posture, motion parameters, and historical data.

[0163] In the above embodiment, after triggering the timing start event, according to the starting direction of the ball object, the expected region sequence is determined, the expected region sequence is composed of correct expected regions in sequence, and is used to indicate the correct balling route. The real-time collected images are identified, and the first region where the ball object is located and the second region where the balling object is located are continuously identified. When the first region determined by the adjacent two frames of images changes, and the changed first region is not the first expected region in the expected sequence, it indicates that the route of the ball object has a problem, and therefore a violation is handled. The first expected region is determined based on the changed first region, and is the correct expected region that the ball object should reach. When the second region determined by the adjacent two frames of images changes, and the changed second region is not the second expected region in the expected sequence, it indicates that the route of the balling object has a problem, and therefore a violation is handled. The second expected region is determined based on the changed second region, and is the correct expected region that the balling object should reach. When the balling object and the ball object both enter the end region, and the first sequence composed of non-repeated first regions in sequence and the second region sequence composed of non-repeated second regions in sequence both meet the expected region sequence, it indicates that the balling object and the ball object complete the balling according to the correct route in the balling process, and therefore the test result used to indicate that the pole test is passed is confirmed. In the technical scheme of the application, the automatic pole test can be realized, the test efficiency is high, the participation of human factors is avoided in the test process, and the accuracy is high. Therefore, by monitoring the balling object and the ball object, the accuracy of the test result is further ensured, and the ability of detecting the posture of the balling object is possessed. The motion parameters can be identified, the balling object student can be provided with the suggestion for improving the motion skill, the setting of the rules is supported, more application scenarios are applicable, and the requirements of more regions are met. The detection device is simple and can be directly and quickly deployed on the existing site. The detection device is not installed in the test region and is not easy to be damaged in the test.

[0164] Exemplary apparatus

[0165] Based on the same concept as the method embodiment of the application, the pole test device is also provided in the embodiment of the application.

[0166] Figure 7 The structure schematic diagram of the pole test device provided by the exemplary embodiment of the application is shown, which comprises:

[0167] The sequence determination module 71 is used to determine the expected region sequence based on the starting direction of the balling object in response to the triggering of the timing start event;

[0168] The image identification module 72 is used to identify the real-time collected images, and determine the first region where the ball object is located and the second region where the balling object is located;

[0169] The first processing module 73 is configured to, in response to the first region determined based on two adjacent images changing and the changed first region not being a first expected region in the expected region sequence, perform a violation processing, the first expected region being determined based on the first region before the change;

[0170] The second processing module 74 is configured to, in response to the second region determined based on two adjacent images changing and the changed second region not being a second expected region in the expected region sequence, perform a violation processing, the second expected region being determined based on the second region before the change;

[0171] The result determining module 75 is configured to, in response to the ball object and the player object both entering the end region and a first region sequence composed of non-repeated first regions in sequence and a second region sequence composed of non-repeated second regions in sequence both meeting the expected region sequence, confirm a test result indicating that the around-the-pole test is passed.

[0172] In an example embodiment of the present application, the first processing module or the second processing module comprises:

[0173] The region determining unit is configured to, in response to the test rule allowing retesting, determine a first effective region corresponding to the ball object and a second effective region corresponding to the player object when the violation processing is performed.

[0174] The target determining unit is configured to determine a target effective region in sequence based on sequence orders of the first effective region and the second effective region in the expected region sequence.

[0175] The rollback processing unit is configured to prompt the player object to carry the ball object back to the target effective region and update the first expected region and the second expected region based on the target effective region.

[0176] In an example embodiment of the present application, the first processing module or the second processing module comprises:

[0177] The violation determining unit is configured to, in response to the ball object not being in the test region and the test rule not allowing retesting, determine a test result indicating an out-of-bounds violation; and in response to the test rule not allowing retesting, determine a test result indicating an around-the-pole violation.

[0178] In an example embodiment of the present application, the image recognition module comprises:

[0179] The image acquisition unit is configured to acquire images collected by at least two image acquisition devices in real time, the at least two image acquisition devices being respectively arranged at preset positions.

[0180] An image recognition unit is configured to recognize the ball object and the ball carrier object in the image, determine first spatial information corresponding to the ball object and second spatial information corresponding to the ball carrier object.

[0181] A first region determination unit is configured to determine first projection information of the ball object based on the first spatial information corresponding to the ball object, and determine a first region where the ball object is located based on the first projection information.

[0182] A second region determination unit is configured to determine second projection information of the ball carrier object based on the second spatial information corresponding to the ball carrier object, and determine a second region where the ball carrier object is located based on the second projection information.

[0183] In an example embodiment of the present application, the second region determination unit is further configured to, in a case where the second projection information corresponds to at least two divided regions, determine a projection proportion corresponding to each divided region respectively, and determine the second region where the ball carrier object is located from the at least two divided regions based on the projection proportion corresponding to each divided region respectively.

[0184] In an example embodiment of the present application, the image recognition unit comprises:

[0185] A recognition subunit is configured to recognize the ball object, the ball carrier object, and a vertical rod object in the image, determine original spatial information corresponding to the ball object, original spatial information corresponding to the ball carrier object, and third spatial information corresponding to the vertical rod object.

[0186] A correction subunit is configured to correct the original spatial information corresponding to the ball object based on annotation data of the vertical rod object and the third spatial information, and determine first spatial information corresponding to the ball object; and correct the original spatial information corresponding to the ball carrier object based on the annotation data of the vertical rod object and the third spatial information, and determine second spatial information corresponding to the ball carrier object.

[0187] In an example embodiment of the present application, the image recognition module is further configured to determine posture information and spatial information of the ball carrier object.

[0188] The method further comprises a parameter determination module configured to determine motion parameters of the ball carrier object based on the posture information and the spatial information of the continuous frames.

[0189] A report determination module is configured to generate an analysis report based on the test result, the posture information, the motion parameters, and historical data.

[0190] Exemplary electronic device

[0191] Figure 8 Fig. 1 illustrates a block diagram of an electronic device according to an embodiment of the present application.

[0192] As Figure 8 shown, the electronic device 80 includes one or more processors 81 and a memory 82.

[0193] The processor 81 can be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities and can control other components in the electronic device 80 to perform desired functions.

[0194] The memory 82 can include one or more computer program products that can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory, for example, can include random access memory (RAM), cache memory, and / or the like. The non-volatile memory, for example, can include read only memory (ROM), hard disk, flash memory, and / or the like. One or more computer program instructions can be stored on the computer-readable storage media, which the processor 81 can run to implement the around-the-pole testing method of various embodiments of the present application described above and / or other desired functions.

[0195] In one example, the electronic device 80 can further include an input device 83 and an output device 84, which are interconnected through a bus system and / or other form of connection mechanism (not shown).

[0196] Of course, for simplicity, Figure 8 only some of the components of the electronic device 80 related to the present application are shown in the figure, and components such as buses, input / output interfaces, and the like are omitted. In addition, the electronic device 80 can include any other appropriate components according to specific application cases.

[0197] Exemplary computer program product and computer readable storage medium

[0198] In a sixth aspect, in addition to the methods and devices described above, embodiments of the present application can also be computer program products including computer program instructions that, when run by a processor, cause the processor to perform the steps of the around-the-pole testing method according to various embodiments of the present application described in the above "Exemplary Methods" section of the specification.

[0199] The computer program product can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. The embodiments of the present application are not limited by the

[0200] In addition, an embodiment of the present application can also be a computer readable storage medium, having stored thereon computer program instructions which, when executed by a processor, cause the processor to perform the steps described in the above "Exemplary Method" section of the present specification for the method of around-the-pole testing according to various embodiments of the present application.

[0201] The computer readable storage medium can be any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0202] The above describes the basic principles of the present application in conjunction with specific embodiments, but it should be noted that the advantages, benefits, effects and the like mentioned in the present application are only examples and are not limiting, and these advantages, benefits, effects and the like cannot be considered as mandatory for each embodiment of the present application. In addition, the above specific details of the application are only for the purpose of example and understanding, and are not limiting, and the above details do not limit the application to be necessarily implemented with the above specific details.

[0203] The block diagrams of the devices, apparatuses, equipment, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be as shown in the block diagrams. These devices, apparatuses, equipment, systems can be connected, arranged, configured in any manner as will be appreciated by those skilled in the art. Words such as "including," "containing," "comprising," and the like are to be construed in an inclusive fashion, indicating open-ended groups and that "consisting of only" and "consisting essentially of only" are not intended to be limiting. The words "or" and "and" as used herein, unless otherwise indicated, are to be interpreted as "and / or." The word "comprising" as used herein is to be interpreted as "comprising, but not limited to." The word "or" as used herein is to be interpreted as "and / or." The word "including" as used herein is to be interpreted as "including, but not limited to."

[0204] It is also to be noted that in the devices, apparatuses and methods of the present application, the various components or steps can be split and / or recombined. These splits and / or re-combinations are to be considered as equivalents of the present application.

[0205] The above description of the aspects of the application is provided to enable any person skilled in the art to make or use the application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0206] The above description has been given for the purpose of illustration and description. Furthermore, this description does not purport to be exhaustive or to limit the embodiments of the application to the precise forms disclosed. Although various example aspects and embodiments have been discussed above, those of ordinary skill in the art will appreciate a variety of modifications, alternatives, permutations, additions, and sub-combinations.

Claims

1. A method of around-the-stalk testing, characterized by, The method comprises the following steps: in response to a triggering timing start event, determining an expected region sequence based on a starting orientation of a ball object; identifying real-time collected images to determine a first region where the ball object is located and a second region where the ball object is located; in response to a change in the first region determined from adjacent two frames of images, and the changed first region is not a first expected region in the expected region sequence, performing a violation processing, the first expected region being determined based on the first region before the change; in response to a change in the second region determined from adjacent two frames of images, and the changed second region is not a second expected region in the expected region sequence, performing a violation processing, the second expected region being determined based on the second region before the change; in response to the ball object and the ball object both entering an end region, and a first region sequence composed of non-repeated first regions in sequence and a second region sequence composed of non-repeated second regions in sequence both conforming to the expected region sequence, confirming a test result indicating that the around-the-pole test is passed; The method comprises the following steps: acquiring images collected by at least two image collection devices in real time, the at least two image collection devices being arranged at preset positions respectively; identifying the ball object and the ball object in the images to determine first spatial information corresponding to the ball object and second spatial information corresponding to the ball object; determining first projection information of the ball object based on the first spatial information corresponding to the ball object, and determining the first region where the ball object is located based on the first projection information; determining second projection information of the ball object based on the second spatial information corresponding to the ball object, and determining the second region where the ball object is located based on the second projection information; The method comprises the following steps: in the case that the second projection information corresponds to at least two divided regions, determining a projection proportion corresponding to each divided region respectively; based on the projection proportion corresponding to each divided region respectively, determining the second region where the ball object is located from the at least two divided regions.

2. The method of claim 1, wherein, The method comprises the following steps: in response to a test rule allowing retesting, determining a first effective region corresponding to the ball object and a second effective region corresponding to the ball object when performing a violation processing; based on the sequence order of the first effective region and the second effective region in the expected region sequence, determining a target effective region in sequence first; prompting the ball object to carry the ball object back to the target effective region, and updating the first expected region and the second expected region based on the target effective region.

3. The method of claim 1, wherein, The method comprises the following steps: in response to the ball object being out of the test region and the test rule not allowing retesting, determining a test result indicating an out-of-bounds violation; In response to the test rule not allowing retesting, a test result indicating a pole-winding violation is determined.

4. The method of claim 1, wherein, The identifying the ball object and the ball-carrying object in the image, determining first spatial information corresponding to the ball object and second spatial information corresponding to the ball-carrying object, comprises: The ball object, the ball-carrying object, and a standing-pole object in the image are identified, and original spatial information corresponding to the ball object, original spatial information corresponding to the ball-carrying object, and third spatial information corresponding to the standing-pole object are determined. The original spatial information corresponding to the ball object is corrected based on the label data of the standing-pole object and the third spatial information, and the first spatial information corresponding to the ball object is determined. The original spatial information corresponding to the ball-carrying object is corrected based on the label data of the standing-pole object and the third spatial information, and the second spatial information corresponding to the ball-carrying object is determined.

5. The method of claim 1, wherein, When the real-time collected image is identified, the method further comprises: Determining the posture information and spatial information of the ball-carrying object; Determining the motion parameters of the ball-carrying object based on the posture information and spatial information of the continuous frames; Generating an analysis report based on the test result, the posture information, the motion parameters, and historical data.

6. A pole testing arrangement, characterised by Comprise: A sequence determination module configured to determine an expected region sequence based on a starting direction of a ball-carrying object in response to a trigger timing start event; An image recognition module configured to identify a real-time collected image, and determine a first region where a ball object is located and a second region where the ball-carrying object is located; A first processing module configured to perform a violation processing in response to a change in the first region determined by two adjacent frames of images, and the changed first region is not a first expected region in the expected region sequence, the first expected region being determined based on the first region before the change; A second processing module configured to perform a violation processing in response to a change in the second region determined by two adjacent frames of images, and the changed second region is not a second expected region in the expected region sequence, the second expected region being determined based on the second region before the change; A result determination module configured to confirm a test result indicating that the pole-winding test is passed in response to the ball-carrying object and the ball object both entering an end region, and a first region sequence composed of non-repeated first regions in sequence and a second region sequence composed of non-repeated second regions in sequence both meeting the expected region sequence. The image recognition module comprises: An image acquisition unit configured to acquire images collected by at least two image acquisition devices in real time, the at least two image acquisition devices being respectively arranged at preset positions; An image recognition unit configured to identify the ball object and the ball-carrying object in the image, and determine first spatial information corresponding to the ball object and second spatial information corresponding to the ball-carrying object. A first region determining unit is configured to determine first projection information of the ball object based on first spatial information corresponding to the ball object, and determine a first region where the ball object is located based on the first projection information. A second region determining unit is configured to determine second projection information of the ball carrier based on second spatial information corresponding to the ball carrier, and determine a second region where the ball carrier is located based on the second projection information. The second region determining unit is further configured to, in a case where the second projection information corresponds to at least two divided regions, determine a projection proportion corresponding to each divided region respectively, and determine the second region where the ball carrier is located from the at least two divided regions based on the projection proportion corresponding to each divided region respectively. 7.A computer readable storage medium, the storage medium storing a computer program, the computer program being configured to execute the method of any one of claims 1-5. 8.An electronic device, comprising: a processor; a memory for storing executable instructions of the processor; the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method of any one of claims 1-5.

Images