A bone testing method based on visual recognition technology

Abstract

The present application discloses a skeletal testing method based on visual recognition technology, which aims to realize the identity recognition of the tester based on the tester's limb features and the user's facial features. Under normal circumstances, it is difficult for the tester to always face the image acquisition device when exercising. On the one hand, the tester himself will move, and the limbs may block the face. On the other hand, the image acquisition device may not be able to move with the tester, which may lead to the situation where the tester turns his back to the image acquisition device. The method in this specification examines the tester's limbs and face respectively. If the regularity shown by the two during the test is matched, it indicates that the recognition result for the face is credible, and the possibility of fraud is low. The manual to automatic conversion during the test process is realized, which is conducive to improving the test efficiency.

Description

Technical Field

[0001] The present application relates to the field of identity authentication technology based on image processing, and in particular to a bone testing method based on visual recognition technology. Background Art

[0002] The development of technology has provided great convenience for production and life, allowing the work that originally needed to be done manually to be completed by machines and programs, which saves labor on the one hand and improves efficiency on the other. However, in some scenarios, the use of manual labor still dominates. For example, in scenarios such as physical training and physical fitness tests. On the one hand, it is necessary to determine the tester's score, and on the other hand, it is necessary to verify the identity of the tester to avoid cheating. This puts higher demands on the detection means. In addition, if the test item is a long-distance run or a physical limit test, it will take a long time to test. If there are many testers, it is difficult to allocate labor to them one by one. It can be seen that how to adapt to the test scenario has become a problem that needs to be solved urgently. Summary of the invention

[0003] An embodiment of the present application provides a bone testing method based on visual recognition technology to at least partially solve the above technical problems.

[0004] The present application embodiment adopts the following technical solutions:

[0005] In a first aspect, an embodiment of the present application provides a bone testing method based on visual recognition technology, the method comprising:

[0006] Obtain a video captured of a tester; the tester includes a head and limbs;

[0007] A first number of first target frame images are extracted from the video; the first target frame images are images containing a first specified number of reference points, and the relative positions between the reference points meet a preset condition; the reference points are feature points on the limbs; the first number is negatively correlated with the number of second target frame images, and the second target frame images are images containing target points in the video that are sufficient to achieve face-based identity recognition; the target points are feature points on the face of the tester;

[0008] For each of the reference points, fitting a first motion trajectory of the reference point based on the first target frame image;

[0009] Using, among the frame images contained in the video, an image except the first target frame image, in which the position and posture of the reference point in the image to which it belongs matches the first motion trajectory, as a third target frame image;

[0010] Based on the target point on the second target frame image, identify the tester to obtain a first identification result;

[0011] Fitting a second motion trajectory of the head based on the first target frame image, the second target frame image, and the target point on the third target frame image;

[0012] If the regularity of the second motion trajectory over time matches the regularity of the first motion trajectory over time, the first recognition result is determined to be a valid result.

[0013] In an optional embodiment of the present specification, the preset condition includes: no overlap in a shooting direction of the first target frame image to which the reference point belongs.

[0014] In an optional embodiment of this specification, the method further includes:

[0015] In the case where the first target frame image is included in each segment contained in the video, the preset conditions include: for at least one of the reference points, the reference point is included in both the previous frame image and the next frame image of the first target frame image to which the reference point belongs; and the segments are obtained by dividing the video according to time.

[0016] In an optional embodiment of this specification, the method further includes:

[0017] In the case that the first target frame image is not included in at least one segment contained in the video, the preset condition includes: for at least one of the reference points, the reference point is included in a frame image before or after the first target frame image to which the reference point belongs; and the segments are obtained by dividing the video according to time.

[0018] In an optional embodiment of the present specification, the feature points on the limbs include: feature points corresponding to the tester's joints, feature points corresponding to the tester's skeleton endpoints, and feature points corresponding to the tester's skeleton midpoints.

[0019] In an optional embodiment of this specification, the method further includes:

[0020] If a segment does not include the first target frame image, determining the segment as a target segment;

[0021] Based on the target point included in the target segment, the identity of the tester is identified to obtain a second identification result;

[0022] If the similarity between the second recognition result and the first recognition result is greater than a preset first similarity threshold, the first recognition result is determined to be a valid result.

[0023] In an optional embodiment of this specification, the method further includes:

[0024] If a segment does not include the first target frame image, determining the segment as a target segment;

[0025] Based on the target point contained in the target segment, the identity of the tester is identified. If no identification result can be obtained, manual identification prompt information is generated and displayed.

[0026] In a second aspect, an embodiment of the present application further provides a bone testing device based on visual recognition technology, the device comprising:

[0027] The video acquisition module is configured to: acquire a video collected from a tester; the tester includes a head and limbs;

[0028] The first target frame image determination module is configured to: extract a first number of first target frame images from the video; the first target frame image is an image containing a first specified number of reference points, and the relative positions between the reference points meet a preset condition; the reference points are feature points on the limbs; the first number is negatively correlated with the number of second target frame images, the second target frame image is an image containing target points in the video that are sufficient to realize face-based identity recognition; the target points are feature points on the face of the tester;

[0029] A first motion trajectory generating module is configured to: for each of the reference points, fit a first motion trajectory of the reference point based on the first target frame image;

[0030] A third target frame image determination module is configured to: use, among the frame images contained in the video, an image other than the first target frame image, in which the position of the reference point contained in the image to which it belongs matches the first motion trajectory, as a third target frame image;

[0031] A first recognition result generating module is configured to: perform identity recognition on the tester based on the target point on the second target frame image to obtain a first recognition result;

[0032] A second motion trajectory generating module is configured to: fit a second motion trajectory of the head based on the first target frame image, the second target frame image, and the target point on the third target frame image;

[0033] The determination module is configured to determine that the first recognition result is a valid result if the regularity exhibited by the second motion trajectory over time matches the regularity exhibited by the first motion trajectory over time.

[0034] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0035] Processor; and

[0036] A memory arranged to store computer executable instructions which, when executed, cause the processor to perform the method steps described in the first aspect.

[0037] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method steps described in the first aspect.

[0038] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0039] The method in this specification aims to realize the identity recognition of the tester based on the tester's limb features and the user's facial features. Under normal circumstances, it is difficult for the tester to always face the image acquisition device when exercising. On the one hand, the tester himself will move, and the limbs may block the face. On the other hand, the image acquisition device may not be able to move with the tester, which may lead to the tester's back to the image acquisition device. This provides conditions for fraud and may cause the test results to not match the tester. The method in this specification examines the tester's limbs and face respectively. If the regularity of the two during the test is matched, it indicates that the recognition result for the face is credible and the possibility of fraud is low. As a result, the manual to automatic process is achieved, which is conducive to improving the test efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0041] Figure 1 A schematic diagram of a process of a bone testing method based on visual recognition technology provided in an embodiment of this specification;

[0042] Figure 2 It is a schematic diagram of the structure of an electronic device in an embodiment of this specification. DETAILED DESCRIPTION

[0043] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, and this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0044] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0045] The serial numbers of the components in this document, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning. The "connection" and "coupling" mentioned in this application, unless otherwise specified, include direct and indirect connections (couplings).

[0046] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0047] The method in this specification is applicable to the scenario of testing a tester. The tester in this specification includes a head and limbs. The part above the neck is the head, and the part below the neck is the limbs. The face is on the head. The part used to implement facial recognition is mainly on the face.

[0048] The test scenarios in this specification may include the following test items: sit-ups, push-ups, pull-ups, bent-arm hangs, running, etc. The test scenarios may be daily training scenarios, physical education test scenarios, sports team test scenarios, military training scenarios, etc.

[0049] like Figure 1 As shown, the skeleton testing method based on visual recognition technology in this specification includes the following steps:

[0050] S100: Acquire a video captured of a tester; the tester includes a head and limbs.

[0051] The video in this specification can be captured by an image acquisition device, which can be a separate device or a part of the bone testing device based on visual recognition technology in this specification. In an optional embodiment of this specification, the video can be a video captured during the entire test process of the tester; or it can be a video clipped from the video captured during the entire test process.

[0052] Due to the limitations of shooting conditions and test environment, the video may not contain the entire body of the tester, and the video may not always be shot towards the tester's face. For example, the tester is running around the playground, and the image acquisition device is set at one end of the playground. The distance between the tester and the image acquisition device varies, and the tester's arms swing back and forth, which may block the tester's face.

[0053] S102: Extracting a first number of first target frame images from the video.

[0054] The method in this specification can at least verify the identity of the user. Imagine that in an application scenario such as long-distance running, if the entire long-distance running video is analyzed, the amount of data to be processed is very large. The first target frame image obtained in this step can, on the one hand, identify the regularity of the tester's limb movement, and on the other hand, can effectively reduce the amount of data processing.

[0055] Corresponding to the first target frame, the method in this specification also relates to a second target frame image. The second target frame image is an image in which the target points contained in the video are sufficient to realize face-based identity recognition; the target points are feature points on the face of the tester. What kind of image is sufficient to realize face-based identity recognition depends on the specific facial recognition technology. In the related technology, the technical means that can realize face-based identity recognition based on the user's facial image are applicable to this specification when conditions permit.

[0056] The feature points in this specification are points that can identify the features of the tester. Feature points located on the face are called target points, and feature points located on the limbs are reference points. Target points are points that constitute facial features, contours, and positions. Reference points are points corresponding to the tester's joints, points corresponding to the tester's skeletal endpoints, and points corresponding to the tester's skeletal midpoints.

[0057] The first target frame is an image that contains a first specified number of reference points, and the relative positions between the reference points meet a preset condition. "Contains" means "at least", and the number of reference points contained in a first target frame may be greater than the first specified number (optionally, a preset value). The preset condition can be determined according to actual business needs.

[0058] The first number in this specification is negatively correlated with the number of second target frame images. If every frame in the video is a second target frame, user identification will be very easy. However, due to facial occlusion and other conditions, not all video frames are second target frames. The fewer the second target frames, the lower the credibility of the user identification result.

[0059] It should be noted that a certain frame in the video may be both the first target frame and the second target frame. The first target frames in the video may be greater than the first number, that is, this step may only select the first number of first target frame images from many first target frames to avoid consuming too much data processing resources for processing the reference points.

[0060] In an optional embodiment of the present specification, the first specified number is positively correlated with the number of testers included in the video. In the case where there are more than one tester included in the video, the method of the present specification can be implemented for each tester separately.

[0061] S104: For each of the reference points, fitting a first motion trajectory of the reference point based on the first target frame image.

[0062] In the related art, any technical means that can realize the fitting of motion trajectories based on feature points are applicable to this specification when conditions permit. Taking the left elbow as an example, the first motion trajectory of the reference point belonging to the left elbow is the line connecting the positions of the left elbow at different times fitted during the test process.

[0063] For example, when the tester is running, the swing direction and amplitude of the left and right arms of the tester are different, but the rhythmic patterns of the two movements are the same. For example, the elbow reference point of the left arm reaches the highest position at time T1, and at the same time, the elbow reference point of the right arm also reaches the highest position. If it is not the same tester, it is difficult to show such a high synchronization rate in the first target frame image.

[0064] S106: Using, among the frame images contained in the video, an image excluding the first target frame image and containing a reference point whose position in the image to which it belongs matches the first motion trajectory as a third target frame image.

[0065] If the first motion trajectory contains four motion trajectories of reference points corresponding to the left elbow, right elbow, left knee and right knee, respectively, then the obtained third target frame image is an image whose reference points can match the four motion trajectories respectively. If the reference point corresponding to the left elbow of a non-first target frame image is located on the first motion trajectory corresponding to the left elbow, and the reference points corresponding to the right elbow, left knee and right knee are also located on the first motion trajectories corresponding to each other, then the non-first target frame image is the third target frame image.

[0066] The information used for trajectory determination and identity recognition shown by the first target frame image can be represented by the obtained first motion trajectory. If other images in the video except the first target frame image can verify the information represented by the first motion trajectory, it will be helpful to improve the accuracy of the verification. The third target frame image determined in this step matches the first target frame image at least in terms of physical behavior.

[0067] The third target frame is not the first target frame, indicating that the third target frame may be an image containing a smaller number of reference points. Considering that a frame of image is neither the first target frame nor the second target frame, the frame of image may be an intermediate state between the image acquisition device being able to capture a clear image of the tester's face and the tester's limb posture changing during the user's posture change. In other words, the third target frame may be an image containing a connection change between the features of the head and limbs. If the third target frame can also reflect the same change pattern of the head following the movement of the tester as the limbs, it indicates that the facial information of the tester identified based on the video is likely to be credible.

[0068] S108: Based on the target points on the second target frame image, identify the tester to obtain a first identification result.

[0069] The first recognition result is a set of data indicating who the tester is. In the related art, any technical means capable of determining the identity of the tester based on the target point is applicable to this specification if conditions permit.

[0070] S110: Fitting a second motion trajectory of the head based on the first target frame image, the second target frame image, and target points on the third target frame image.

[0071] The first target frame image reflects the movement of the limbs at each moment of the first motion trajectory. If the movement of the head reflected in it matches the movement of the limbs, it means that the head and the limbs maintain the same movement law, and the possibility of cheating is low. The second target frame image actually captures the user's face. Based on the relative position relationship between the face and the head, the second target frame image can determine the movement law of the head to a certain extent, but the number of second target frame images may not be sufficient. The third target frame image can characterize the transition between the head and the limbs to a certain extent. The second motion trajectory thus obtained can coordinate the relationship between different body parts of the tester to a certain extent, but if there is cheating, the second motion trajectory will deviate from the first motion trajectory.

[0072] For example, if the pupil, nose, and mouth are respectively used as target points, there should be three fitted curves. The second motion trajectory in this specification can be the result of fitting the curves fitted based on target points at different positions of the face. The fitting result is related to the selection of the target points and the actual movement of the head. In some cases, the fitted second motion trajectory is the motion trajectory of the center of mass of the head.

[0073] S112: If the regularity of the second motion trajectory over time matches the regularity of the first motion trajectory over time, determining that the first recognition result is a valid result.

[0074] The regularity in this specification can be represented by the distance of the tester's body part from the ground over time, or by the degree of recurrence of the regularity of the tester's body part in a certain time period in other time periods.

[0075] Exemplarily, if the similarity between the regularity exhibited by the second motion trajectory and the regularity exhibited by the first motion trajectory is greater than a preset second similarity threshold, the regularities match; otherwise, the regularities do not match.

[0076] In the case of regular matching, the probability of cheating is low, and the first recognition result is more likely to be a valid result.

[0077] The method in this specification aims to realize the identity recognition of the tester based on the tester's limb features and the user's facial features. Under normal circumstances, it is difficult for the tester to always face the image acquisition device when exercising. On the one hand, the tester himself will move, and the limbs may block the face. On the other hand, the image acquisition device may not be able to move with the tester, which may lead to the tester's back to the image acquisition device. This provides conditions for fraud and may cause the test results to not match the tester. The method in this specification examines the tester's limbs and face respectively. If the regularity of the two during the test is matched, it indicates that the recognition result for the face is credible and the possibility of fraud is low. As a result, the manual to automatic process is achieved, which is conducive to improving the test efficiency.

[0078] The preset conditions satisfied by the first target frame image in an optional embodiment of the present specification are now described.

[0079] In an optional embodiment of the present specification, the preset condition includes: no overlap in the shooting direction of the first target frame image to which the reference point belongs. As the relative positional relationship between the tester and the image acquisition device changes, there may be a phenomenon in which the reference points overlap in the shooting direction. For example, the feature points corresponding to the left elbow and the right wrist overlap, and this overlap hinders the recognition of the feature points and may cause confusion, which is not conducive to determining the first motion trajectory. Such an image should be excluded as the first target frame image.

[0080] In other optional embodiments, the preset conditions also include other contents. Specifically, the video is first divided to obtain a number of segments arranged continuously over time. Optionally, the duration of each segment is the same. In the case where the first target frame image is included in each segment contained in the video (at this time, it indicates that the tester's limb information is distributed more evenly in the video. If there is cheating, it is difficult to locate which segment it occurred in), the preset conditions include: the reference point is also included in the previous frame image and the next frame image of the first target frame image to which the reference point belongs (indicating that the limb movement law shown in the first target frame video is continuous in time, and the movement law of the tester's limbs shown by it is more realistic).

[0081] When the first target frame image is not included in at least one segment of the video (in this case, it indicates that the tester's limb information is unevenly distributed in the video. If cheating occurs, it is possible to locate the specific segment in which it occurs), the preset conditions include: the reference point is included in the previous frame image or the next frame image of the first target frame image to which the reference point belongs (since the first target frame itself is unevenly distributed, the densely distributed part can show the real movement law of the tester's limb, and the real movement law is knowable, so there is no need to set more stringent conditions).

[0082] To further achieve the purpose of verification, in a further optional embodiment of the present specification, if a segment does not contain the first target frame, the segment is determined as a target segment (if fraud exists, the target segment is a segment with a higher probability of fraud).

[0083] The identity of the tester is identified based on the target points contained in the target segment to obtain a second identification result. If the similarity between the second identification result and the first identification result is greater than a preset first similarity threshold (the preset value may be an empirical value or may be positively correlated with the number of second target frame images), and the regularity of the second motion trajectory over time matches the regularity of the first motion trajectory over time, the first identification result is determined to be a valid result. If no identification result can be obtained, indicating that there is a risk of fraud, manual identification prompt information is generated and displayed.

[0084] Furthermore, this specification also provides a bone testing device based on visual recognition technology, the device comprising:

[0085] The video acquisition module is configured to: acquire a video collected from a tester; the tester includes a head and limbs;

[0086] The first target frame image determination module is configured to: extract a first number of first target frame images from the video; the first target frame image is an image containing a first specified number of reference points, and the relative positions between the reference points meet a preset condition; the reference points are feature points on the limbs; the first number is negatively correlated with the number of second target frame images, the second target frame image is an image containing target points in the video that are sufficient to realize face-based identity recognition; the target points are feature points on the face of the tester;

[0087] A first motion trajectory generating module is configured to: for each of the reference points, fit a first motion trajectory of the reference point based on the first target frame image;

[0088] A third target frame image determination module is configured to: use, among the frame images contained in the video, an image other than the first target frame image, in which the position of the reference point contained in the image to which it belongs matches the first motion trajectory, as a third target frame image;

[0089] A first recognition result generating module is configured to: perform identity recognition on the tester based on the target point on the second target frame image to obtain a first recognition result;

[0090] A second motion trajectory generating module is configured to: fit a second motion trajectory of the head based on the first target frame image, the second target frame image, and the target point on the third target frame image;

[0091] The determination module is configured to determine that the first recognition result is a valid result if the regularity exhibited by the second motion trajectory over time matches the regularity exhibited by the first motion trajectory over time.

[0092] The device can execute the method in any of the aforementioned embodiments and can obtain the same or similar technical effects, which will not be described in detail here.

[0093] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 2At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0094] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0095] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0096] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a bone testing device based on visual recognition technology at the logical level. The processor executes the program stored in the memory and is specifically used to execute any of the aforementioned bone testing methods based on visual recognition technology.

[0097] The above application Figure 1The bone testing method based on visual recognition technology disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or instructions in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0098] The electronic device may also perform Figure 1 A skeleton testing method based on visual recognition technology is implemented Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0099] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple applications, any one of the aforementioned bone testing methods based on visual recognition technology is executed.

[0100] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0101] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0102] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0104] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0105] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0106] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0107] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0108] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0109] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A bone testing method based on visual recognition technology, characterized in that: The method comprises: Obtain a video captured of a tester; the tester includes a head and limbs; A first number of first target frame images are extracted from the video; the first target frame images are images containing a first specified number of reference points, and the relative positions between the reference points meet a preset condition; the reference points are feature points on the limbs; the first number is negatively correlated with the number of second target frame images, and the second target frame images are images containing target points in the video that are sufficient to achieve face-based identity recognition; the target points are feature points on the face of the tester; For each of the reference points, fitting a first motion trajectory of the reference point based on the first target frame image; Using, among the frame images contained in the video, an image except the first target frame image, in which the position and posture of the reference point in the image to which it belongs matches the first motion trajectory, as a third target frame image; Based on the target point on the second target frame image, identify the tester to obtain a first identification result; Fitting a second motion trajectory of the head based on the first target frame image, the second target frame image, and the target point on the third target frame image; If the regularity of the second motion trajectory over time matches the regularity of the first motion trajectory over time, the first recognition result is determined to be a valid result.

2. The method according to claim 1, characterized in that: The preset condition includes: no overlap in the shooting direction of the first target frame image to which the reference point belongs.

3. The method according to claim 1, characterized in that: The method further comprises: In the case where the first target frame image is included in each segment contained in the video, the preset conditions include: for at least one of the reference points, the reference point is included in both the previous frame image and the next frame image of the first target frame image to which the reference point belongs; and the segments are obtained by dividing the video according to time.

4. The method according to claim 1, characterized in that: The method further comprises: In the case that the first target frame image is not included in at least one segment contained in the video, the preset condition includes: for at least one of the reference points, the reference point is included in a frame image before or after the first target frame image to which the reference point belongs; and the segments are obtained by dividing the video according to time.

5. The method according to claim 1, characterized in that: The feature points on the limbs include: feature points corresponding to the tester's joints, feature points corresponding to the tester's skeleton endpoints, and feature points corresponding to the tester's skeleton midpoints.

6. The method according to claim 4, characterized in that: The method further comprises: If a segment does not include the first target frame image, determining the segment as a target segment; Based on the target point included in the target segment, the identity of the tester is identified to obtain a second identification result; If the similarity between the second recognition result and the first recognition result is greater than a preset first similarity threshold, the first recognition result is determined to be a valid result.

7. The method according to claim 4, characterized in that: The method further comprises: If a segment does not include the first target frame image, determining the segment as a target segment; Based on the target point contained in the target segment, the identity of the tester is identified. If no identification result can be obtained, manual identification prompt information is generated and displayed.

8. A bone testing device based on visual recognition technology, characterized in that: The device comprises: The video acquisition module is configured to: acquire a video collected from a tester; the tester includes a head and limbs; The first target frame image determination module is configured to: extract a first number of first target frame images from the video; the first target frame image is an image containing a first specified number of reference points, and the relative positions between the reference points meet a preset condition; the reference points are feature points on the limbs; the first number is negatively correlated with the number of second target frame images, the second target frame image is an image containing target points in the video that are sufficient to realize face-based identity recognition; the target points are feature points on the face of the tester; A first motion trajectory generating module is configured to: for each of the reference points, fit a first motion trajectory of the reference point based on the first target frame image; A third target frame image determination module is configured to: use, among the frame images contained in the video, an image other than the first target frame image, in which the position of the reference point contained in the image to which it belongs matches the first motion trajectory, as a third target frame image; A first recognition result generating module is configured to: perform identity recognition on the tester based on the target point on the second target frame image to obtain a first recognition result; A second motion trajectory generating module is configured to: fit a second motion trajectory of the head based on the first target frame image, the second target frame image, and the target point on the third target frame image; The determination module is configured to determine that the first recognition result is a valid result if the regularity exhibited by the second motion trajectory over time matches the regularity exhibited by the first motion trajectory over time.

9. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, enables the electronic device to execute any one of the methods of claims 1 to 7.

Images