WO2023123213A1

WO2023123213A1 - Hand pressing depth measurement method and apparatus

Info

Publication number: WO2023123213A1
Application number: PCT/CN2021/143103
Authority: WO
Inventors: 焦旭
Original assignee: 焦旭
Priority date: 2021-12-30
Filing date: 2021-12-30
Publication date: 2023-07-06
Also published as: CN114556447A

Abstract

A hand pressing depth measurement method, comprising the following steps: reading a target video, the target video being a video acquired by a fixed photographing device for a hand pressing action, and a marker being worn on a hand (S21); performing marker detection on the target video to obtain a tracking frame of the marker (S22); tracking the tracking frame to obtain a highest point and a lowest point of the marker in a pressing direction (S23); and obtaining a hand pressing depth according to coordinates of the highest point and coordinates of the lowest point (S24). The problems of complex structure, high cost and low accuracy of a non-contact measurement device are solved. Moreover, a corresponding apparatus, a device, and a medium are provided.

Description

Hand pressing depth detection method and device

technical field

The invention belongs to the technical field of image processing, and in particular relates to a hand pressing depth detection method and detection device, a computer-readable medium and electronic equipment.

Background technique

An estimated 330,000 deaths from out-of-hospital cardiac arrest (OHCA) occur each year in Canada and the United States. In treated OHCA, overall survival is poor, with discharge rates ranging from 3.0% to 16.3%. The difference in patient survival rates can be attributed in part to the following five important components: rapid emergency medical system (ENS) access; early cardiopulmonary resuscitation (CPR); early defibrillation; early advanced life support (ACLS); effective post-resuscitative care . In recent years, after the communities and hospitals have vigorously strengthened the above-mentioned links, the survival rate has only slightly improved. In recent years, it has been recognized that the quality, frequency, and timeliness of CPR during cardiac resuscitation are critical to survival in cardiac arrest. Related studies have found that the increase in compression depth during CPR is closely related to the degree of improvement in survival rate, and when the compression depth is 4.03-5.53cm (peak value 4.56cm), the patient's survival rate is the highest. However, in the actual process of cardiac resuscitation, it is difficult to evaluate the compression depth in cardiac resuscitation due to the lack of tools for measuring depth. The traditional contact measurement has high precision and good stability, but it is not widely used due to factors such as volume, quality, installation conditions, structure, and inconvenient operation; Distance measurement is carried out with the patient through contact instruments.

Contents of the invention

In order to solve the above-mentioned shortcoming of the prior art, the present invention proposes a hand pressing depth detection method, comprising the following steps:

Reading the target video, the target video is a video captured by a fixed shooting device for the pressing action of the hand, and the hand is worn with a marker;

Perform marker detection on the target video, and obtain a tracking frame of the marker;

Tracking the tracking frame to obtain the highest point and the lowest point of the marker in the pressing direction;

According to the coordinates of the highest point and the coordinates of the lowest point, the pressing depth of the hand is obtained.

Further, the coordinates of the highest point are the coordinates of the upper edge of the marker when the marker is at the highest point, and the coordinates of the lowest point are the coordinates of the marker when the marker is at the lowest point. The coordinates of the lower edge.

Further, the step of performing marker detection on the target video includes:

Generating a sequence of video frames from the target video in chronological order;

Converting the video frame image to the HSV color space, in the HSV color space, according to the target color, perform binarization processing, segment the video frame image, and obtain the target color image;

Use N convolution kernels to convolve the target color image respectively, and add the outputs of each convolution kernel, and count the point with the largest kernel function response, and whether the area where the point with the largest statistical response is located is greater than the threshold, if it is greater than the threshold , determine that there is a target, and set a tracking frame according to the area of the target color image, where N is an integer greater than or equal to 3.

Further, the first target color and the second target color are spaced apart from each other on the marker, and the target color image includes the first target color image and the second target color image; before performing convolution processing, it also includes, Dilation processing and erosion processing are performed on the second target color image, and an intersection operation is performed on the first target color image.

Further, setting the tracking frame according to the area of the target color image includes: setting the tracking frame according to the union area of the first target color image and the second target color image.

Further, the first target color is white, and the second target color is green.

Further, the step of tracking the markers includes: tracking the markers using KCF algorithm.

Further, the detection method further includes, judging that the hand pressing depth information satisfies a pressing threshold, and displaying prompt information.

In the second aspect, an embodiment of the present invention provides a hand pressing depth detection device, including:

The reading module is used to read the target video, the target video is a video captured by the fixed shooting device for the pressing action of the hand, and the hand is worn with a marker;

A detection module, configured to perform marker detection on the target video, and obtain a tracking frame of the marker;

A tracking module, which tracks the tracking frame and obtains the highest point and the lowest point of the marker in the pressing direction;

The ranging module is configured to obtain the hand pressing depth according to the coordinates of the highest point and the lowest point.

In a third aspect of the present invention, an electronic device is provided, including:

one or more processors;

a storage device on which one or more programs are stored,

When the one or more programs are executed by the one or more processors, the one or more processors implement any one of the methods described above.

A fourth aspect of the present invention provides a computer-readable medium on which a computer program is stored, wherein when the program is executed by a processor, any one of the above-mentioned methods is implemented.

The invention provides a method for measuring distance by using image information. Firstly, marker detection is performed, and then tracking is performed to obtain the position of the highest point and the lowest point, and the pressed distance information is obtained according to the difference between the positions of the highest point and the lowest point. . In particular, the square video information in the embodiment of the present invention can be collected by a monocular camera device, for example, by a mobile phone, which reduces the performance requirements of the device used. Through the detection of compression depth information, it can quickly judge whether the compression depth reaches the standard, thereby improving the quality of CPR.

Description of drawings

The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the accompanying drawings:

Fig. 1 is a schematic diagram of the system architecture of the hand pressing depth detection method and detection device operation in some examples of the present invention;

Fig. 2 is a schematic flow chart of a hand pressing depth detection method in some examples of the present invention;

FIG. 3 is a schematic diagram of a video shot in a hand pressing depth detection method in some embodiments of the present invention;

Fig. 4 is a schematic flow chart of detecting markers in the hand pressing depth detection method in some embodiments of the present invention;

Fig. 5 is a schematic flow chart of detecting markers in the hand pressing depth detection method in some embodiments of the present invention;

Fig. 6 is a schematic flow chart of the tracking steps in the hand pressing depth detection method in some embodiments of the present invention;

Fig. 7 is a schematic flowchart of a hand pressing depth detection method in other embodiments of the present invention;

Fig. 8 is a system schematic diagram of a detection device implemented based on the detection method in the above-mentioned drawings in some embodiments of the present invention;

Fig. 9 is a schematic structural diagram of a computer system in which a hand pressing depth detection method or a detection device operates in some embodiments of the present invention.

Detailed ways

In order to understand the above-mentioned purpose, features and advantages of the present invention more clearly, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described here. Therefore, the protection scope of the present invention is not limited by the specific details disclosed below. EXAMPLE LIMITATIONS.

Fig. 1 shows an exemplary system architecture 100 to which the embodiment of the hand pressing depth detection method or detection device according to the embodiment of the present application can be applied.

As shown in FIG. 1 , a system architecture 100 may include

terminal devices

101 , 102 , 103 , a network 104 and a server 105 . The network 104 is used as a medium for providing communication links between the

terminal devices

101 , 102 , 103 and the server 105 . Network 104 may include various connection types, such as wires, wireless communication links, or fiber optic cables, among others.

Users can use

terminal devices

101 , 102 , 103 to interact with server 105 via network 104 to receive or send data (such as video) and the like. Various communication client applications can be installed on the

terminal devices

101, 102, 103, such as video playback software, video processing applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social networking platform software, etc.

The

terminal devices

101, 102, and 103 may be hardware or software. When the

terminal devices

101, 102, 103 are hardware, they may be various electronic devices with display screens and supporting data transmission, including but not limited to smart phones, tablet computers, laptop computers and desktop computers, etc. When the

terminal devices

101, 102, 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (such as software or software modules for providing distributed services), or as a single software or software module. No specific limitation is made here.

The server 105 may be a server that provides various services, such as a background server that provides support for videos displayed on the

terminal devices

101 , 102 , 103 . The background server can analyze and process the received data such as slicing requests, and feed back the processing results (such as indexed slices or slicing sequences) to electronic devices (such as terminal devices) connected in communication with it.

It should be noted that the hand pressing depth detection method provided in the embodiment of the present application can be executed by the server 105 , and correspondingly, the pressing depth detection device can be set in the server 105 . In addition, the hand pressing depth detection method provided in the embodiment of the present application can also be executed by the

terminal devices

101, 102, 103, and accordingly, the hand pressing depth detection device can also be set in the

terminal devices

101, 102, 103.

It should be noted that the server may be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software or software modules (such as software or software modules for providing distributed services), or as a single software or software module. No specific limitation is made here.

It should be understood that the numbers of terminal devices, networks and servers in Fig. 1 are only illustrative. According to the implementation needs, there can be any number of terminal devices, networks and servers. When the electronic device on which the live stream playback method runs does not need to perform data transmission with other electronic devices, the system architecture may only include the electronic devices on which the hand pressing depth detection method runs (such as

terminal devices

101, 102, 103 or server 105).

The embodiment of the present invention obtains the position information of the highest point and the lowest point through the visual tracking technology, and then obtains the compression depth through the difference of the position information, so that the detection efficiency and accuracy are effectively guaranteed. In the actual implementation process, the front CCD camera equipped with the mobile phone is used to shoot the pressing action, and after digital image processing, the pressing depth information is measured and obtained.

There are many factors that affect the ranging accuracy, but the influence of hardware factors can be limited by selecting high-quality hardware such as high-resolution CCD cameras and high-sampling frequency image acquisition cards, and various environmental factors. Because the styles of mobile phones are different, the method of improving the ranging accuracy of the system through software algorithms is relatively the most effective way. The usual image ranging process needs to add a calibration object (with obvious color characteristics and known length and width) in the scene, and use this calibration object to calibrate the camera and correct the image, so as to obtain the distance accuracy of the image and obtain the measurement distance of the target. However, it is difficult to set a calibration object whose length and width are known during heart compressions.

The embodiment of the present invention adopts the idea of combining the traditional scale measurement, extracts the wristband when pressed through digital image processing, and completes the distance measurement with guaranteed speed and accuracy through the vertical distance of the movement of the wristband through algorithm-assisted calculation. The overall scheme is shown in Figure 2.

The front pressing action is captured by the front camera of the mobile phone, as shown in Figure 3. Process the captured video into an image sequence. In order to ensure the speed of distance measurement, it is necessary to detect and track the target through the captured image, detect the specific position at the highest point of the press and the highest point, and complete the distance measurement. Assuming that the target is a bracelet, then detect the bracelet from a series of images to obtain the position of the bracelet in each image in the image coordinates, and determine the position of the bracelet with the highest point of coordinates and the lowest point of coordinates, and the obtained position The distance between them is the compression depth. Pass the target in each image through the recognition algorithm to identify the bracelet on the wrist (the bracelet is the target), and pass the position information [x, y, w, h] of the bracelet to the tracking module , x is the X-axis coordinate, y is the Y-axis coordinate, w is the width, and h is the height. The embodiment of the present invention uses the KCF algorithm to track the target, track the movement of the wrist and record the coordinate information of the target frame. The extreme point in the motion process is extracted through the recorded data, and the picture of the frame is input into the distance measurement module to complete the final distance measurement. Judge whether the compression depth of 5cm is reached, and display it in the video in real time.

Specifically, the detection method includes the following steps:

S1. The position of the detection marker (bracelet)

After inputting the video, first input the video stream into the program in time sequence, and process the picture into a fixed size picture, the default is 960*540, and the subsequent processing process is shown in Figure 4.

S11. Since the bracelet is selected as a bracelet structure with green and white spaced apart from each other, the color features are obvious. Therefore, the image is converted to the HSV color space to facilitate color screening and extraction. The hue and brightness of its color are directly selected as the threshold, and the target segmentation and detection are realized through binarization. The HSV space model and other color spaces are represented by circles through the H dimension.

S12. Extract white and green color features through the hsv space, expand and corrode the green area, and perform an intersection operation with the white area to further narrow the target space.

S13. Design a classifier structure with two layers of kernel functions, respectively set three convolution kernels to perform convolution and add them together, and count the point with the largest kernel function response. The value of the rule response determines whether there is a target, and at the same time sets the size of the border according to the internal green area. The ratio of the border is set to 2:5 according to the ratio of the bracelet. The specific process is shown in Figure 5. The picture is input into three After the convolution kernel is added, input to the kernel function, view the response, count the area response, and then set the border size.

S14. In order to obtain a relatively stable tracking effect, the delineated target should be selected as large as possible. Therefore, further, the embodiment of the present invention selects the union of white and green areas to ensure a large target, and returns the tracking frame [x, y, H, W].

S2, tracking markers

In order to cope with the high-speed movement of the wristband in the fast pressing video. The embodiment of the present invention proposes a method using a correlation kernel operation, which can ensure a good tracking effect when the target is only slightly deformed and moves fast, and can effectively extract the trajectory of the target moving up and down. The flow of the KCF algorithm is shown in Figure 6. The steps to be executed include a pre-computation, b training and updating the model, and c finding the target position; first execute step a, then execute step b for the first frame of image, and for the image after the second frame image, loop through steps c and b. The measure of the similarity value of two signals is measured by correlation. If the two signals are more similar, the higher the correlation value is. In the application of tracking, the filter template is used to maximize the response when it acts on the tracking target. , the location of the maximum response value is the location of the target. The characteristics of KCF: Simple implementation, good effect and fast speed. A large number of samples are generated by cyclic matrix displacement to solve the problem, and the calculation speed in the frequency domain is extremely fast through the derivation of the discrete Fourier transform. In the case of lack of samples, it is detected by a simple method, and the core operation is connected to track, which ensures the generalization ability of the track. Using the KCF algorithm has the following advantages.

1) Detection: use cyclic matrix + Fourier change to calculate the response graph, the original O(n^3) algorithm only needs O(n*log(n))

2) Training: use the nature of the cyclic matrix to train in the frequency domain

3) Kernel regression speed up: For the kernel function, it can also be converted to the frequency domain for training and detection, greatly improving the speed

4) The special kernel function is further accelerated: for the Gaussian kernel, the polynomial kernel can further use the circulant matrix to calculate the circulant matrix of the kernel function

Tracking can quickly track the experimental goals. By inputting the video of manual pressing into the program, the distance it moves can be displayed stably.

S3. Detection distance

In the chest compression video, the target moves fast, and the distance is small and the accuracy is high. Therefore, in order to ensure the ranging module, it is necessary to improve the reference system. The following factors need to be considered in the detection distance:

A. The measurement displacement of 5cm is very small. Even if the camera is placed very close, the pixels occupied by the beacon in the image are still very small, and the feature extraction of the beacon is difficult.

B. Each person's chest compression action is different, and the difference is very large, which will also affect the accuracy, and additional algorithms need to be used to make up for it.

C. Since the camera cannot guarantee absolute verticality and horizontality when shooting, the angle change of pressing will cause deviation in the measured vertical distance

D. The pixel of the camera itself may cause a loss of precision, and the error E can be obtained by the following formula.

Through mathematical derivation, it can be seen that the measurement of the actual distance and the pixel value during shooting will cause the accuracy error during the experiment, and increasing the length of the reference distance can effectively reduce the error in the experiment.

Based on the above derivation, in the actual distance measurement, we test the wristband on the upper edge of the highest point and the lower edge of the lowest point to ensure the maximum moving distance of the reference and reduce the error caused by the distance measurement. At the same time, a characteristic pattern logo is added to the bracelet to ensure the accuracy of positioning.

The embodiments of the present invention can use low-cost equipment to implement distance measurement through the marker tracking method, thereby reducing equipment requirements. It overcomes the shortcomings of the traditional monocular ranging method that requires a fixed reference object, has low precision, and cannot adapt to long-distance measurement occasions.

Referring to FIG. 7 , another embodiment of the present invention provides a mobile phone pressing depth detection method, including the following steps:

S21. Read the target video. The target video is a video captured by a fixed shooting device for the pressing action of a hand, and a marker is worn on the hand. When shooting, the subject to be pressed lies flat on the ground, the position of the shooting equipment is almost at the same height as the pressing position, and the video is shot and recorded from the side of the pressing action.

S22. Perform marker detection on the target video, and obtain the tracking frame of the marker; specifically, generate a sequence of video frames from the target video in chronological order; convert the video frame images into HSV color space, and In the HSV color space, according to the target color, binarization processing is performed, the video frame image is segmented, and the target color image is obtained; N convolution kernels are used to convolve the target color image respectively, and each convolution kernel The output is added together, the point with the largest statistical kernel function response, whether the area where the point with the largest statistical response is located is greater than the threshold value, if greater than the threshold value, it is judged that there is a target, and the tracking frame is set according to the area of the target color image, where N is greater than An integer equal to 3.

The first target color is white, and the second target color is green. The color contrast is obvious, which is good for detection.

S23. Track the tracking frame, and obtain the highest point and the lowest point of the marker in the pressing direction; in the embodiment of the present invention, the KCF method can be used for tracking, and other methods can also be used for tracking, and detection during the tracking process After reaching the extremum point, return the coordinates of the highest point and the lowest point.

S24. Obtain the hand pressing depth according to the coordinates of the highest point and the lowest point. In order to ensure the detection accuracy, a larger range will be taken as the pressing depth. Specifically, the coordinates of the highest point are the coordinates of the upper edge of the marker when the marker is at the highest point, and the coordinates of the lowest point is the coordinate of the lower edge of the marker when the marker is at the lowest point.

In order to prompt the user, the detection method further includes, S25, judging that the hand pressing depth information satisfies the pressing threshold, and displaying prompt information. For example, if the compression depth meets the compression threshold, it will display qualified, or display green.

The hand pressing depth detection method in the embodiment of the present invention obtains the pressing depth by using the coordinates of the uppermost edge at the highest point and the coordinates of the lowermost edge at the lowest point, fully considering the measurement error and ensuring the accuracy of the measurement. The measurement by the tracking method reduces the performance requirement of the ranging equipment and reduces the cost of method implementation.

As shown in FIG. 8, according to the above-mentioned method embodiment, the embodiment of the present invention also provides a hand pressing depth detection device 300, including:

The reading module 310 is used to read the target video, the target video is a video captured by the fixed shooting device for the pressing action of the hand, and the hand is worn with a marker;

A detection module 320, configured to perform marker detection on the target video, and obtain a tracking frame of the marker;

The tracking module 330 is configured to track the tracking frame, and obtain the highest point and the lowest point of the marker in the pressing direction;

The ranging module 340 is configured to obtain the hand pressing depth according to the coordinates of the highest point and the lowest point.

The specific execution steps of the above modules have been described in detail in the corresponding steps in the live stream playing method, and will not be repeated here.

Referring now to FIG. 9 , it shows a schematic structural diagram of a computer system 800 suitable for implementing the control device of the embodiment of the present application. The control device shown in FIG. 9 is only an example, and should not limit the functions and scope of use of this embodiment of the present application.

As shown in FIG. 9 , a computer system 800 includes a central processing unit (CPU) 801 that can be programmed according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage section 808 into a random-access memory (RAM) 803 Instead, various appropriate actions and processes are performed. In the RAM 803, various programs and data required for the operation of the system 800 are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to the bus 804 .

The following components are connected to the I/O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 808 including a hard disk, etc. and a communication section 809 including a network interface card such as a LAN card, a modem, or the like. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I/O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 810 as necessary so that a computer program read therefrom is installed into the storage section 808 as necessary.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, where the computer program includes program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via communication portion 809 and/or installed from removable media 811 . When the computer program is executed by the central processing unit (CPU) 801, the above-mentioned functions defined in the method of the present application are performed.

It should be noted that the computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable Programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, however, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, in which computer-readable program codes are carried. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. . Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out the operations of this application can be written in one or more programming languages, or combinations thereof, including object-oriented programming languages—such as Python, Java, Smalltalk, C++, and Includes conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (such as through an Internet service provider). Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or portion of code that contains one or more logical functions for implementing specified executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by a dedicated hardware-based system that performs the specified functions or operations , or may be implemented by a combination of dedicated hardware and computer instructions.

The units involved in the embodiments described in the present application may be implemented by means of software or by means of hardware. The described units may also be set in a processor, for example, it may be described as: a processor includes an acquiring unit, a dividing unit, a determining unit and a selecting unit. Wherein, the names of these units do not constitute a limitation on the unit itself under certain circumstances, for example, the acquisition unit may also be described as "a unit for acquiring picture book images to be processed".

As another aspect, the present application also provides a computer-readable medium. The computer-readable medium may be included in the electronic device described in the above embodiments; it may also exist independently without being assembled into the electronic device. middle. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device: reads the target video, and the target video is pressed by the fixed shooting device against the hand Actions are collected to obtain the video, and the hands are worn with markers; marker detection is performed on the target video, and the tracking frame of the marker is acquired; the tracking frame is tracked, and the marker is acquired The highest point and the lowest point in the pressing direction; according to the coordinates of the highest point and the lowest point, the pressing depth of the hand is obtained.

The above description is only a preferred embodiment of the present application and an illustration of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solution formed by the specific combination of the above-mentioned technical features, and should also cover the technical solution formed by the above-mentioned technical features or Other technical solutions formed by any combination of equivalent features. For example, a technical solution formed by replacing the above-mentioned features with technical features with similar functions disclosed in (but not limited to) this application.

Claims

A hand pressing depth detection method is characterized in that, comprising the steps of:

Reading the target video, the target video is a video captured by a fixed shooting device for the pressing action of the hand, and the hand is worn with a marker;

Perform marker detection on the target video, and obtain a tracking frame of the marker;

Tracking the tracking frame to obtain the highest point and the lowest point of the marker in the pressing direction;

According to the coordinates of the highest point and the coordinates of the lowest point, the pressing depth of the hand is obtained.
The hand pressing depth detection method according to claim 1, wherein the coordinates of the highest point are the coordinates of the upper edge of the marker when the marker is at the highest point; the coordinates of the lowest point is the coordinate of the lower edge of the marker when the marker is at the lowest point.
The hand pressing depth detection method according to claim 1 or 2, wherein the step of performing marker detection on the target video comprises:

Generating a sequence of video frames from the target video in chronological order;

Converting the video frame image to the HSV color space, in the HSV color space, according to the target color, perform binarization processing, segment the video frame image, and obtain the target color image;

Use N convolution kernels to convolve the target color image respectively, and add the outputs of each convolution kernel, and count the point with the largest kernel function response, and whether the area where the point with the largest statistical response is located is greater than the threshold, if it is greater than the threshold , determine that there is a target, and set a tracking frame according to the area of the target color image, where N is an integer greater than or equal to 3.
The method for detecting the depth of hand pressing according to claim 3, wherein the first target color and the second target color are spaced apart from each other on the marker, and the target color image includes the first target color image and the second target color image. The second target color image: before performing convolution processing, it also includes performing expansion processing and erosion processing on the second target color image, and performing an intersection operation with the first target color image.
The method for detecting the depth of hand pressing according to claim 4, wherein setting the tracking frame according to the region of the target color image comprises: according to the union of the first target color image and the second target color image Region sets the tracking bounding box.
The method for detecting the depth of hand pressing according to claim 4, wherein the first target color is white, and the second target color is green.
The method for detecting the depth of hand pressing according to claim 1 or 2, characterized in that the step of tracking the markers includes: tracking the markers using the KCF algorithm.
The method for detecting hand pressing depth according to claim 1 or 2, further comprising: judging that the hand pressing depth information satisfies a pressing threshold, and displaying prompt information.
A hand pressing depth detection device is characterized in that it comprises:

The reading module is used to read the target video, the target video is a video captured by the fixed shooting device for the pressing action of the hand, and the hand is worn with a marker;

A detection module, configured to perform marker detection on the target video, and obtain a tracking frame of the marker;

A tracking module, which tracks the tracking frame and obtains the highest point and the lowest point of the marker in the pressing direction;

The ranging module is configured to obtain the hand pressing depth according to the coordinates of the highest point and the lowest point.
An electronic device comprising:

one or more processors;

a storage device on which one or more programs are stored,

When the one or more programs are executed by the one or more processors, the one or more processors are made to implement the method according to any one of claims 1-8.
A computer-readable medium, on which a computer program is stored, wherein when the program is executed by a processor, the method according to any one of claims 1-8 is implemented.