Sitting posture detection method, device, electronic device and storage medium

Abstract

The present application provides a sitting posture detection method, device, electronic device, and storage medium, relating to the field of data processing technology. The method comprises: using a distance sensor to measure the distance between a subject to be detected and the sitting posture detection device; when the distance is within a preset normal distance range, detecting skeletal points of a preset body part of the subject to be detected in a target image of the subject to be detected captured by a camera, and determining the sitting posture type of the subject to be detected based on the position information of the different detected skeletal points. The application of the sitting posture detection solution provided by the present application can improve the efficiency of sitting posture detection.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a sitting posture detection method, device, electronic device, and storage medium. Background Art

[0002] Sitting posture detection is gaining increasing attention. By performing posture detection on a subject, it is possible to determine whether the subject exhibits poor sitting posture, such as tilting the head or leaning forward, allowing for correction if necessary. For example, the subject could be a child. By performing posture detection on a child, the child's sitting posture can be determined, helping to correct poor posture and fostering healthy sitting habits from an early age, thus benefiting their growth and development.

[0003] Conventional technology typically requires installing multiple sensors at various locations on a chair, such as the armrests and seat cushions, to determine the subject's sitting posture based on their measurements. However, since each sensor generates a single measurement result, multiple sensors generate multiple results. This requires processing a large amount of data when using these multiple measurement results to determine the subject's sitting posture, resulting in low posture detection efficiency. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a sitting posture detection method, device, electronic device, and storage medium to improve the efficiency of sitting posture detection. The specific technical solution is as follows:

[0005] In a first aspect, an embodiment of the present application provides a sitting posture detection method, which is applied to a sitting posture detection device, wherein the sitting posture detection device includes a ranging sensor and a camera, and the method includes:

[0006] Measuring the distance between the object to be measured and the sitting posture detection device using the distance measuring sensor;

[0007] When the distance is within a preset normal distance range, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera are detected, and the sitting posture type of the object to be measured is determined based on the position information of the different detected skeletal points.

[0008] In one embodiment of the present application, determining the sitting posture type of the subject to be measured based on the detected position information of different skeletal points includes:

[0009] Calculating object sitting posture data reflecting the sitting posture of the object to be measured based on position information of skeletal points of a preset body part of the object to be measured;

[0010] The calculated object sitting posture data is compared with preset standard sitting posture data, and the sitting posture type of the object to be measured is determined according to the comparison result, wherein the standard sitting posture data is used to reflect the standard sitting posture of the object to be measured.

[0011] In one embodiment of the present application, before comparing the calculated sitting posture data of the subject with preset standard sitting posture data and determining the sitting posture type of the subject to be measured based on the comparison result, the method further includes:

[0012] A target sub-distance range in which the distance is located is determined in each sub-distance range included in the normal distance range, and target sitting posture data corresponding to the target sub-distance range is determined in each preset group of target sitting posture data as standard sitting posture data, wherein each group of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

[0013] In one embodiment of the present application, when the distance is within a preset normal distance range, detecting skeletal points of a preset body part of the subject to be measured in a target image of the subject to be measured captured by the camera, and determining the sitting posture type of the subject to be measured based on position information of different detected skeletal points includes:

[0014] When the distance is within a preset normal distance range, the target image of the object to be measured captured by the camera is input into a pre-trained sitting posture detection model to obtain the sitting posture type of the object to be measured output by the sitting posture detection model, wherein the sitting posture detection model is trained using the image of the sample object as input and the sitting posture type of the sample object as annotation information.

[0015] In one embodiment of the present application, the sitting posture detection device is deployed on a desktop.

[0016] When the distance is within a preset normal distance range, detecting skeletal points of a preset body part of the subject to be measured in a target image of the subject to be measured captured by the camera, and determining the sitting posture type of the subject to be measured based on position information of different detected skeletal points, including:

[0017] Detecting the target image of the object to be measured captured by the camera to obtain a first pixel point where a preset facial feature point of the object to be measured is located and a second pixel point where an adjacent desktop boundary is located, wherein the adjacent desktop boundary is the boundary of the desktop closest to the object to be measured;

[0018] Calculating the height between the facial feature point and the plane where the desktop is located based on the position of the first pixel point and the position of the second pixel point;

[0019] If the distance is within the preset normal distance range and the height is within the preset normal height range, the bone points of the preset body parts of the object to be measured in the target image are detected, and the sitting posture type of the object to be measured is determined based on the position information of the different detected bone points.

[0020] In one embodiment of the present application, the camera in the sitting posture detection device is directed horizontally toward the object to be detected;

[0021] The calculating, based on the position of the first pixel point and the position of the second pixel point, the height between the facial feature point and the plane where the desktop is located, includes:

[0022] Counting the number of first pixel points between the first pixel point and a central pixel point in a pixel column direction of the image to be measured, wherein the central pixel point is located at the center of the image to be measured;

[0023] Based on a pre-obtained conversion relationship between the number of pixels and length, calculating a first height between the facial feature point and a horizontal plane where the camera is located according to the first number of pixels;

[0024] Counting the number of second pixel points between the second pixel point and the central pixel point in the direction of the pixel column;

[0025] Based on the conversion relationship, a second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels;

[0026] The sum of the first height and the second height is calculated to obtain the height between the facial feature point and the plane where the desktop is located.

[0027] In one embodiment of the present application, the conversion relationship is expressed as follows:

[0028] L=D*tan(N*R+θ)

[0029] Wherein, L represents the length, D represents the distance between the sitting posture detection device and the object to be measured, N represents the number of pixels, R represents the radian value corresponding to the preset unit pixel, and θ represents the preset radian value error.

[0030] In one embodiment of the present application, after detecting the skeletal points of a preset body part of the subject to be tested in the target image of the subject to be tested captured by the camera, and determining the sitting posture type of the subject to be tested based on the position information of different detected skeletal points, the method further includes:

[0031] If the obtained sitting posture type belongs to an unhealthy sitting posture type, the unhealthy sitting posture image of the object to be measured captured by the camera is recorded.

[0032] In one embodiment of the present application, after detecting the skeletal points of a preset body part of the subject to be tested in the target image of the subject to be tested captured by the camera, and determining the sitting posture type of the subject to be tested based on the position information of different detected skeletal points, the method further includes:

[0033] If the obtained sitting posture type belongs to an unhealthy sitting posture type, a sitting posture reminder message is generated to remind the subject to be measured to correct the sitting posture.

[0034] In a second aspect, an embodiment of the present application further provides a sitting posture detection device, the device comprising a ranging sensor, a camera, and a processor, wherein:

[0035] The distance measuring sensor is used to measure the distance between the object to be measured and itself, and send the measured distance to the processor;

[0036] The camera is used to capture a target image of the object to be measured and send the captured target image to the processor;

[0037] The processor is used to execute any method step described in the first aspect above.

[0038] In a third aspect, an embodiment of the present application further provides a sitting posture detection device, which is applied to a sitting posture detection device. The sitting posture detection device includes a ranging sensor and a camera. The device includes:

[0039] a distance measurement module, configured to measure the distance between the object to be measured and the sitting posture detection device using the distance measurement sensor;

[0040] A sitting posture determination module is used to detect the bone points of the preset body parts of the object to be tested in the target image of the object to be tested captured by the camera when the distance is within a preset normal distance range, and determine the sitting posture type of the object to be tested based on the position information of the different detected bone points.

[0041] In one embodiment of the present application, the sitting posture determination module includes:

[0042] a skeleton point detection submodule, configured to detect skeleton points of a preset body part of the object to be measured in a target image of the object to be measured captured by the camera when the distance is within a preset normal distance range;

[0043] a data calculation submodule, configured to calculate object sitting posture data reflecting the sitting posture of the object to be measured based on position information of skeletal points of preset body parts of the object to be measured;

[0044] The data comparison submodule is used to compare the calculated object sitting posture data with the preset standard sitting posture data, and determine the sitting posture type of the object to be measured based on the comparison result, wherein the standard sitting posture data is used to reflect the standard sitting posture of the object to be measured.

[0045] In one embodiment of the present application, the device further comprises:

[0046] A data selection module is used to compare the calculated object sitting posture data with the preset standard sitting posture data, and before determining the sitting posture type of the object to be measured based on the comparison result, determine the target sub-distance range in which the distance is located in each sub-distance range included in the normal distance range, and determine the target sitting posture data corresponding to the target sub-distance range in each preset group of target sitting posture data as standard sitting posture data, wherein each group of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

[0047] In one embodiment of the present application, the sitting posture determination module is specifically configured to:

[0048] When the distance is within a preset normal distance range, the target image of the object to be measured captured by the camera is input into a pre-trained sitting posture detection model to obtain the sitting posture type of the object to be measured output by the sitting posture detection model, wherein the sitting posture detection model is trained using the image of the sample object as input and the sitting posture type of the sample object as annotation information.

[0049] In one embodiment of the present application, the sitting posture detection device is deployed on a desktop.

[0050] The sitting posture determination module includes:

[0051] an image detection submodule, configured to detect a target image of the object to be measured captured by the camera, and obtain a first pixel point where a preset facial feature point of the object to be measured is located, and a second pixel point where an adjacent desktop boundary is located, wherein the adjacent desktop boundary is the boundary of the desktop closest to the object to be measured;

[0052] a height calculation submodule, configured to calculate the height between the facial feature point and the plane where the desktop is located based on the position of the first pixel point and the position of the second pixel point;

[0053] The sitting posture determination submodule is used to detect the bone points of the preset body parts of the object to be tested in the target image if the distance is within the preset normal distance range and the height is within the preset normal height range, and determine the sitting posture type of the object to be tested based on the position information of the different detected bone points.

[0054] In one embodiment of the present application, the camera in the sitting posture detection device is directed horizontally toward the object to be detected;

[0055] The height calculation submodule is specifically used to:

[0056] Counting the number of first pixel points between the first pixel point and a central pixel point in a pixel column direction of the target image to be tested, wherein the central pixel point is located at the center of the target image to be tested;

[0057] Based on a pre-obtained conversion relationship between the number of pixels and length, calculating a first height between the facial feature point and a horizontal plane where the camera is located according to the first number of pixels;

[0058] Counting the number of second pixel points between the second pixel point and the central pixel point in the direction of the pixel column;

[0059] Based on the conversion relationship, a second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels;

[0060] The sum of the first height and the second height is calculated to obtain the height between the facial feature point and the plane where the desktop is located.

[0061] In one embodiment of the present application, the conversion relationship is expressed as follows:

[0062] L=D*tan(N*R+θ)

[0063] Among them, L represents length, D represents the distance between the sitting posture detection device of the image acquisition device and the object to be measured, N represents the number of pixels, R represents the radian value corresponding to the preset unit pixel point, and θ represents the preset radian value error.

[0064] In one embodiment of the present application, the device further comprises:

[0065] An image acquisition module is used to detect the skeletal points of preset body parts of the object to be tested in the target image of the object to be tested captured by the camera, and after determining the sitting posture type of the object to be tested based on the position information of the different skeletal points detected, if the obtained sitting posture type belongs to an unhealthy sitting posture type, record the unhealthy sitting posture image of the object to be tested captured by the camera.

[0066] In one embodiment of the present application, the device further comprises:

[0067] The information generation module is used to detect the skeletal points of the preset body parts of the object to be tested in the target image of the object to be tested captured by the camera, and after determining the sitting posture type of the object to be tested based on the position information of the different detected skeletal points, if the obtained sitting posture type belongs to an unhealthy sitting posture type, generate sitting posture reminder information to remind the object to be tested to correct its sitting posture.

[0068] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0069] Memory for storing computer programs;

[0070] The processor is configured to implement any of the method steps described in the first aspect when executing a program stored in the memory.

[0071] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the method steps described in the first aspect is implemented.

[0072] In a sixth aspect, an embodiment of the present application further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the method steps described in the first aspect above.

[0073] Beneficial effects of the embodiments of the present application:

[0074] As can be seen from the above, when applying the solution provided by the embodiment of the present application to perform sitting posture detection, the distance between the object to be measured and the sitting posture detection device is first measured. When the measured distance is within the normal distance range, it means that the sitting posture of the object to be measured may be normal. In order to further determine the sitting posture type of the object to be measured, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera can be detected, and the sitting posture type of the object to be measured can be determined based on the position information of the different skeletal points detected. In this way, the sitting posture detection of the object to be measured can be realized by only using the distance measured by the distance measuring sensor and the target image captured by the camera. The amount of data required is relatively small, thereby speeding up the data processing speed. It can be seen that the sitting posture detection efficiency can be improved by applying the sitting posture detection solution provided by the embodiment of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0076] Figure 1 A flowchart of the first sitting posture detection method provided in an embodiment of the present application;

[0077] Figure 2a A flowchart of a second sitting posture detection method provided in an embodiment of the present application;

[0078] Figure 2b The first target image provided in the embodiment of the present application;

[0079] Figure 2c The second target image provided in the embodiment of the present application;

[0080] Figure 2d The third target image provided in the embodiment of the present application;

[0081] Figure 2e The fourth target image provided in the embodiment of the present application;

[0082] Figure 3 A flowchart of a third sitting posture detection method provided in an embodiment of the present application;

[0083] Figure 4 A flowchart of the fourth sitting posture detection method provided in an embodiment of the present application;

[0084] Figure 5 A flowchart of the fifth sitting posture detection method provided in an embodiment of the present application;

[0085] Figure 6a A flowchart of a sixth sitting posture detection method provided in an embodiment of the present application;

[0086] Figure 6b A schematic diagram of a sitting posture detection scenario provided in an embodiment of the present application;

[0087] Figure 7 A flowchart of the seventh sitting posture detection method provided in an embodiment of the present application;

[0088] Figure 8 A flowchart of an eighth sitting posture detection method provided in an embodiment of the present application;

[0089] Figure 9 A flowchart of a ninth sitting posture detection method provided in an embodiment of the present application;

[0090] Figure 10 A schematic structural diagram of a first sitting posture detection device provided in an embodiment of the present application;

[0091] Figure 11 A schematic structural diagram of a second sitting posture detection device provided in an embodiment of the present application;

[0092] Figure 12 A schematic structural diagram of a first sitting posture detection device provided in an embodiment of the present application;

[0093] Figure 13 A schematic structural diagram of a second sitting posture detection device provided in an embodiment of the present application;

[0094] Figure 14 A schematic structural diagram of a third sitting posture detection device provided in an embodiment of the present application;

[0095] Figure 15 A schematic structural diagram of a fourth sitting posture detection device provided in an embodiment of the present application;

[0096] Figure 16 A schematic structural diagram of a fifth sitting posture detection device provided in an embodiment of the present application;

[0097] Figure 17 A schematic structural diagram of a sixth sitting posture detection device provided in an embodiment of the present application;

[0098] Figure 18 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0099] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field based on this application are within the scope of protection of this application.

[0100] See also Figure 1 , Figure 1 This is a flow chart of the first sitting posture detection method provided in an embodiment of the present application. The above method is applied to a sitting posture detection device, which includes a ranging sensor and a camera.

[0101] The distance measuring sensor and the camera may be placed side by side horizontally or vertically.

[0102] The above method includes the following steps S101-S102.

[0103] Step S101: using a distance measuring sensor to measure the distance between the object to be measured and the sitting posture detection device.

[0104] The subjects to be tested may be children, adults or other subjects who need to correct their sitting posture.

[0105] The distance measuring sensor can be any one of various sensors such as an infrared distance measuring sensor, a laser distance measuring sensor, etc.

[0106] Specifically, the distance measuring sensor can measure the distance between itself and the object to be measured, and the sitting posture detection device can obtain the distance and use the distance as the distance between the sitting posture detection device and the object to be measured.

[0107] In one embodiment of the present application, the distance may be measured using a ranging sensor in either of the following two implementations.

[0108] In a first implementation, a distance measuring sensor may be used to measure the horizontal distance between the object to be measured and the sensor itself as the above distance.

[0109] In the second implementation, a distance measuring sensor may be used to measure the distance between a specific body part of the object to be measured and the object itself as the above distance.

[0110] For example, the above-mentioned specific body part can be the chest, or the abdomen or other body parts, and the embodiments of the present application are not limited to this.

[0111] Step S102: When the distance is within a preset normal distance range, the skeleton points of the preset body parts of the object to be tested are detected in the target image of the object to be tested captured by the camera, and the sitting posture type of the object to be tested is determined based on the position information of the detected different skeleton points.

[0112] What those skilled in the art can understand is that in actual application scenarios, when the object to be measured sits on a chair and is adjacent to a table top, in a standard sitting posture, the distance between the object to be measured and the boundary of its nearest adjacent table top is usually the distance of one fist. Therefore, based on the above-mentioned principle of the distance of one fist, the distance range between the object to be measured and the boundary of its nearest adjacent table top can be determined, and then the distance range can be added to the horizontal distance between the above-mentioned sitting posture detection equipment and the above-mentioned boundary to obtain the above-mentioned normal distance range.

[0113] The horizontal distance between the sitting posture detection device and the boundary may be actually measured after the sitting posture detection device is installed, or may be preset before installation.

[0114] For example, a person's fist is usually 5cm-8cm. If the horizontal distance between the sitting posture detection device and the boundary is 40cm, it can be calculated that the minimum distance of the normal distance range is 40+5=45cm, and the maximum distance is 40+8=48cm, that is, the normal distance range is 45cm-48cm.

[0115] The above-mentioned preset body parts may be eyes, chin, shoulders and the like.

[0116] The above-mentioned sitting posture types can be divided into standard sitting posture and bad sitting posture. The above-mentioned bad sitting posture can also be divided into various sitting postures such as tilting the head, lowering the head, leaning forward, shrugging shoulders, etc.

[0117] Specifically, after obtaining the distance between the above-mentioned object to be measured and the sitting posture detection device, it can be determined whether the distance is within the above-mentioned normal distance range. If not, it means that the object to be measured may have an unhealthy sitting posture such as leaning forward or backward; if yes, it means that the sitting posture of the object to be measured may be normal. At this time, the target image of the object to be measured captured by the camera is obtained, and the bone points of the preset body parts of the object to be measured in the target image are detected, and the position information of different bone points in the above image is determined. The relative positions between different bone points are determined based on the position information of each bone point. Since the relative positions between different bone points in different sitting postures are often different, the sitting posture type of the object to be measured can be determined based on the position information of different bone points.

[0118] There are two situations when the above-mentioned camera captures the image of the object to be measured.

[0119] In the first case, the camera can continuously capture images of the object to be measured.

[0120] In this case, when it is determined that the above-mentioned distance is within the normal distance range, the judgment moment at which this judgment result is obtained can be determined, thereby determining a collection time period of a preset length including the judgment moment, and selecting at least one image from the images collected by the above-mentioned camera within this collection time period as the target image.

[0121] Among them, the above-mentioned collection time period can be determined by taking the above-mentioned judgment moment as the earliest moment in the time period and the above-mentioned preset duration as the duration of the time period, or by taking the above-mentioned judgment moment as the latest moment in the time period and the above-mentioned preset duration as the duration of the time period, or by taking the above-mentioned judgment moment as any middle moment in the time period and the above-mentioned preset duration as the duration of the time period.

[0122] For example, the preset duration may be 0.1s, 0.5s, or other durations.

[0123] In the second case, the camera may capture an image of the object to be measured after receiving a capture instruction.

[0124] In this case, after determining that the distance is within the normal distance range, a camera acquisition instruction can be generated to instruct the camera to acquire an image of the object to be measured, thereby obtaining a target image of the object to be measured acquired by the camera after receiving the camera acquisition instruction.

[0125] In one embodiment of the present application, when detecting the skeletal points of a preset body part of the object to be measured in the above-mentioned image, the features of the skeletal points of the preset body part can be obtained in advance. After obtaining the above-mentioned target image, the features of the object to be measured in the above-mentioned target image are detected, and then the detected features are matched with the pre-obtained features. The features that are successfully matched are determined from the detected features, and the skeletal points of the preset body part of the object to be measured in the target image are determined based on the determined features.

[0126] The implementation method of determining the sitting posture type of the object to be tested based on the position information of different bone points can be found in the following Figure 2a Steps S103B-S103C in the illustrated embodiment are not described in detail here.

[0127] In addition, you can also use the subsequent Figure 4 The sitting posture detection model mentioned in the illustrated embodiment detects the above-mentioned skeleton points and determines the sitting posture type of the object to be detected based on the position information of different skeleton points, which will not be described in detail here.

[0128] As can be seen from the above, when applying the solution provided by the embodiment of the present application to perform sitting posture detection, the distance between the object to be measured and the sitting posture detection device is first measured. When the measured distance is within the normal distance range, it means that the sitting posture of the object to be measured may be normal. In order to further determine the sitting posture type of the object to be measured, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera can be detected, and the sitting posture type of the object to be measured can be determined based on the position information of the different skeletal points detected. In this way, the sitting posture detection of the object to be measured can be realized by only using the distance measured by the distance measuring sensor and the target image captured by the camera. The amount of data required is relatively small, thereby speeding up the data processing speed. It can be seen that the sitting posture detection efficiency can be improved by applying the sitting posture detection solution provided by the embodiment of the present application.

[0129] The following describes an implementation method for determining the sitting posture type of the subject to be measured based on the position information of different skeletal points mentioned in the above step S102.

[0130] In one embodiment of the present application, see Figure 2a , provides a flow chart of a second sitting posture detection method. In this embodiment, the above step S102 can be implemented by the following steps S102A-S102C.

[0131] Step S102A: When the distance is within a preset normal range, detecting the skeleton points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera.

[0132] This step is similar to the method of detecting skeleton points in the above step S102, and will not be repeated here.

[0133] Step S102B: Calculating object sitting posture data reflecting the sitting posture of the object to be measured based on the position information of the skeletal points of the preset body parts of the object to be measured.

[0134] The above-mentioned object sitting posture data reflects the sitting posture of the object to be measured at the time of acquisition of the target image, and the above-mentioned object sitting posture data may include relative position information between skeletal points of different preset body parts of the object to be measured.

[0135] For example, the above-mentioned object sitting posture data may include the angle between the straight line where the bone points of the two eyes of the object to be measured are located and the horizontal plane, the height between the chin bone point and the shoulder bone point, etc.

[0136] Specifically, after detecting the above-mentioned skeleton points, the relative position information between different preset body parts of the object to be measured can be calculated based on the position information of each skeleton point in the target image, and the calculated relative position information can be used as the object sitting posture data reflecting the sitting posture of the object to be measured.

[0137] Taking the above-mentioned object sitting posture data including the angle between the eye skeleton point and the horizontal plane as an example, in this case, the straight line where the eye skeleton point is located can be first determined in the target image, and then the angle between the straight line and the pixel row direction of the image can be measured as the angle between the eye skeleton point and the horizontal plane.

[0138] In addition, the above-mentioned object sitting posture data can also include the angle between the ear bone points and the horizontal plane, the angle between the shoulder bone points and the horizontal plane, etc. The method of calculating these two angles is the same as the method of calculating the angle between the eye bone points and the horizontal plane.

[0139] In another embodiment of the present application, the subject's sitting posture data includes the heights between different skeletal points. In this case, a conversion relationship between a unit pixel and a length in the pixel column direction of the image can be pre-obtained. Then, the number of pixels between different skeletal points in the pixel column direction can be counted. Based on the pre-obtained conversion relationship and the counted number of pixels, the heights between different skeletal points can be calculated.

[0140] For example, the conversion relationship between the above-mentioned unit pixel and length can be one pixel corresponding to 0.1 cm. If the number of pixel points between the chin bone point and the shoulder bone point is counted as 120 in the pixel column direction, then the height between the chin and the shoulder of the object to be measured can be calculated to be 0.1*120=10 cm.

[0141] Step S102C: Compare the calculated sitting posture data of the subject with the preset standard sitting posture data, and determine the sitting posture type of the subject to be measured based on the comparison result.

[0142] The standard sitting posture data is used to reflect the standard sitting posture of the subject to be measured. The standard sitting posture data may include: the range of relative position information between skeletal points of different preset body parts under the standard sitting posture.

[0143] For example, the standard sitting posture data may include a range of heights between a chin bone point and a shoulder bone point.

[0144] The above-mentioned standard sitting posture data may be set manually or calculated based on a normal sitting posture image of the subject to be measured.

[0145] Specifically, when comparing the sitting posture data of the subject with the standard sitting posture data, the same type of data included in the two types of sitting posture data may be compared, and then the sitting posture type of the subject to be measured may be determined based on the comparison result.

[0146] For example, the angle between the straight line where the skeletal points of the two eyes of the object to be measured are located and the horizontal plane included in the object sitting posture data can be compared with the angle range between the straight line where the skeletal points of the two eyes of the object to be measured included in the standard sitting posture data and the horizontal plane. If the above-mentioned angle included in the object sitting posture data is not within the above-mentioned angle range included in the standard sitting posture data, it is considered that the object to be measured has an unhealthy sitting posture with a tilted head.

[0147] For another example, the height between the chin bone point and the shoulder bone point of the object to be measured included in the object sitting posture data can be compared with the height range between the chin bone point and the shoulder bone point of the object to be measured included in the standard sitting posture data. If the above-mentioned height included in the object sitting posture data is smaller than the minimum value in the above-mentioned height range included in the standard sitting posture data, it is considered that the object to be measured has an unhealthy sitting posture with the head lowered. If the above-mentioned height included in the object sitting posture data is greater than the maximum value in the above-mentioned height range included in the standard sitting posture data, it is considered that the object to be measured has an unhealthy sitting posture with the head tilted back.

[0148] In addition, since the above two types of sitting posture data usually include relative position information between multiple different bone points, multiple comparison results can be obtained when comparing the two types of sitting posture data, and then the sitting posture type of the object to be tested can be comprehensively determined based on the multiple comparison results obtained.

[0149] Figure 2b-Figure 2e These are target images of four subjects in different sitting postures provided in the embodiments of the present application. There are multiple bone points in the above four images, and the body parts corresponding to these bone points are the nose, eyes, ears, left and right shoulders, and the elbows and wrists of the left and right arms. Figure 2b The target image shows that the subject has an unhealthy sitting posture with his head down. Figure 2c The target image shows that the subject has an unhealthy sitting posture with his head lowered or tilted. Figure 2d The subject in the target image has an unhealthy sitting posture with lowered head and uneven shoulders. Figure 2e In the target image shown, the subject has an unhealthy sitting posture with their head tilted.

[0150] As can be seen from the above, when the solution provided in the embodiment of the present application is used for sitting posture detection, the object sitting posture data is calculated based on the position information of the above-mentioned skeleton points, and then the object sitting posture data is compared with the standard sitting posture data, and the sitting posture type of the object to be measured is determined according to the comparison result. Since the object sitting posture data reflects the sitting posture of the object to be measured at the time of acquisition of the target image, and the standard sitting posture data reflects the standard sitting posture of the object to be measured, comparing the object sitting posture data with the standard sitting posture data can be understood as comparing the sitting posture of the object to be measured at the time of acquisition of the target image with the standard sitting posture. Therefore, by comparing the object sitting posture data and the standard sitting posture data, the sitting posture type of the object to be measured can be accurately determined according to the comparison results.

[0151] In one embodiment of the present application, see Figure 3 , provides a flow chart of a third sitting posture detection method. In this embodiment, before comparing the object sitting posture data with the standard sitting posture data, the above method further includes the following step S103.

[0152] Step S103: determining a target sub-distance range in each sub-distance range included in the normal distance range, and determining target sitting posture data corresponding to the target sub-distance range in each set of preset target sitting posture data as standard sitting posture data.

[0153] Each set of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

[0154] When the distance between the sitting posture detection device and the object to be measured is any distance within the above-mentioned normal distance range, it can be considered that the distance between the object to be measured and the boundary of its nearest adjacent desktop meets the "one fist" requirement. However, when the distance between the sitting posture detection device and the object to be measured is different, the standard sitting posture of the object to be measured is usually different, and the sitting posture data reflecting the standard sitting posture of the object to be measured is also different.

[0155] Specifically, the target sitting posture data corresponding to different sub-distance ranges can be obtained in advance, and then the target sub-distance range in which the distance measured by the above-mentioned ranging sensor is located is determined. The target sitting posture data corresponding to the target sub-distance range is determined from the obtained target sitting posture data, and the determined target sitting posture data is used as the standard sitting posture data.

[0156] For example, if the above-mentioned normal distance range is 40cm-70cm, the normal distance range includes sub-distance range A, sub-distance range B, and sub-distance range C, and the range values ​​of each sub-distance range are 40cm-50cm, 50cm-60cm, and 60cm-70cm respectively. Sub-distance range A corresponds to target sitting posture data A, sub-distance range B corresponds to target sitting posture data B, and sub-distance range C corresponds to target sitting posture data C. At this time, if the distance measured by the above-mentioned ranging sensor is 44cm, the target sub-distance range can be sub-distance range A, and the target sitting posture data A corresponding to sub-distance range A can be used as the standard sitting posture data.

[0157] The target sitting posture data corresponding to each sub-distance range can be set manually or calculated based on the normal sitting posture image of the object to be measured. At this time, the normal sitting posture image is obtained when the distance between the object to be measured and the sitting posture detection device is within the sub-distance range.

[0158] As can be seen from the above, when the solution provided in the embodiment of the present application is used for sitting posture detection, the target sub-distance range in which the above distance is located is determined in each sub-distance range included in the normal distance range, and the target sitting posture data corresponding to the target sub-distance range is determined in each set of preset target sitting posture data as standard sitting posture data. Since the distance between the sitting posture detection device and the object to be measured is different, the standard sitting posture of the object to be measured is usually different, and the standard sitting posture data reflecting the standard sitting posture of the object to be measured is also different. By determining the above-mentioned target sub-distance range and determining the target sitting posture data corresponding to the target sub-distance range, more accurate standard sitting posture data reflecting the standard sitting posture of the object to be measured can be obtained, thereby comparing the object sitting posture data with the target sitting posture data as the standard sitting posture data, and based on the comparison results, the sitting posture type of the object to be measured can be more accurately determined.

[0159] The specific implementation method of implementing the above step S102 using the sitting posture detection model is described below.

[0160] In one embodiment of the present application, see Figure 4 , provides a flow chart of a fourth sitting posture detection method. In this embodiment, the above step S102 can be implemented by the following step S102D.

[0161] Step S102D: When the distance is within a preset normal distance range, the target image of the object to be measured captured by the camera is input into a pre-trained sitting posture detection model to obtain the sitting posture type of the object to be measured output by the sitting posture detection model.

[0162] Among them, the sitting posture detection model is trained by taking the image of the sample object as input and the sitting posture type of the sample object as annotation information.

[0163] Specifically, the above-mentioned sitting posture detection model can be a model with a two-layer network layer structure, wherein the first network layer is used to detect the skeletal points of the preset body parts of the object to be tested in the target image, and the second network layer is used to determine the sitting posture type of the object to be tested based on the position information of the different skeletal points detected. Since the sitting posture detection model is trained using the image of the sample object as the numerical input and the sitting posture type of the sample object as the annotation information, the first network layer in the trained sitting posture detection model can learn the skeletal features of the skeletal points of the preset body parts of the sample object, and the second network layer can learn the relationship features between the position information of different skeletal points and the sitting posture type. In this way, after the target image is input into the sitting posture detection model, the first network layer in the sitting posture detection model can detect the skeletal points of the preset body parts of the object to be tested in the target image based on the learned skeletal features, and the second network layer can determine the sitting posture type of the object to be tested based on the learned relationship features and the position information of different skeletal points in the target image.

[0164] As can be seen from the above, when applying the solution provided in the embodiment of the present application to perform sitting posture detection, the target image is input into the sitting posture detection model. Since the sitting posture detection model is a pre-trained model, the sitting posture detection model learns the features of determining the sitting posture type based on the input image. In this way, after the target image is input into the sitting posture detection model, the sitting posture detection model can accurately determine the sitting posture type of the object to be tested based on the learned features.

[0165] In one embodiment of the present application, see Figure 5 , a flow chart of a fifth sitting posture detection method is provided. In this embodiment, the sitting posture detection device is deployed on a desktop, and the sitting posture detection device includes a camera that can capture the boundary of the desktop and the image of the object to be detected.

[0166] In the embodiment of the present application, the above step S102 can be implemented through the following steps S102E-S102G.

[0167] Step S102E: Detect the target image of the object to be measured captured by the camera to obtain a first pixel point where a preset facial feature point of the object to be measured is located and a second pixel point where an adjacent desktop boundary is located.

[0168] The adjacent desktop boundary is the boundary of the desktop that is closest to the object to be measured.

[0169] The facial feature points may be any one or more of a plurality of feature points such as the center point of the forehead, the tip of the nose, and the chin.

[0170] Specifically, a target image of the object to be measured captured by a camera may be first obtained, and then the target image may be detected to obtain the first pixel point and the second pixel point.

[0171] When obtaining the above-mentioned target image, an image captured within a time period that is relatively close to the ranging moment of the above-mentioned ranging sensor can be obtained. Since the interval between the acquisition moment of the target image and the ranging moment of the ranging sensor is relatively close, it can be considered that the sitting posture of the object to be measured has not changed. In this way, the above-mentioned distance and target image can be regarded as being obtained when the object to be measured is in the same sitting posture.

[0172] Detecting the first pixel point where the preset facial feature point of the object to be tested is located in the image to be tested can be achieved by using existing target detection technology, which will not be described in detail here.

[0173] In one embodiment of the present application, when detecting the second pixel point where the boundary of an adjacent desktop is located in the image to be tested, the area where the object to be tested is located and the area where the desktop is located can be detected first, and the boundary line between the area where the object to be tested is located and the area where the desktop is located can be determined, and the pixel point in the boundary line can be determined as the above-mentioned second pixel point.

[0174] In another embodiment of the present application, since the camera is typically placed directly in front of the object to be measured, the desktop boundary closest to the object to be measured is typically perpendicular to the camera's orientation. Consequently, in the image to be measured captured by the camera, the straight line containing the aforementioned adjacent desktop boundary is typically parallel to the pixel row direction. Based on this principle, after the desktop boundary area is detected in the image to be measured, the pixels in the boundary area parallel to the pixel row direction can be used as the second pixel point of the adjacent desktop boundary.

[0175] Step S102F: Calculate the height between the facial feature point and the plane where the desktop is located based on the position of the first pixel point and the position of the second pixel point.

[0176] In one embodiment of the present application, the height between the facial feature points and the plane where the desktop is located can be calculated by any one of the following two implementation methods.

[0177] In the first implementation method, the conversion relationship between a single pixel point in the pixel column direction of the image and the actual height can be obtained in advance, and then the number of pixels between the first pixel point and the second pixel point in the pixel column direction is counted. According to the conversion relationship, the counted number of pixels is converted into height, and the converted height is used as the height between the facial feature point and the plane where the desktop is located.

[0178] When counting the number of pixels between the first pixel point and the second pixel point in the pixel column direction, the average value of the row numbers of the pixel rows where each second pixel point is located can be first calculated, and then the calculated average value can be subtracted from the row number of the pixel row where the first pixel point is located to obtain the number of pixels between the first pixel point and the second pixel point in the pixel column direction.

[0179] In the second implementation, you can Figure 6a In the illustrated embodiment, steps S102F1 - S102F5 calculate the height between the facial feature points and the plane where the desktop is located, which will not be described in detail here.

[0180] In addition, there can be multiple first pixel points mentioned above, each first pixel point corresponds to a facial feature point of the object to be measured. Based on the position of each first pixel point and the position of the second pixel point, the height between the facial feature point corresponding to the first pixel point and the plane where the desktop is located can be calculated.

[0181] Step S102G: If the distance is within the preset normal distance range and the height is within the preset normal height range, the bone points of the preset body parts of the object to be tested in the target image are detected, and the sitting posture type of the object to be tested is determined based on the position information of the different detected bone points.

[0182] Among them, the above-mentioned normal height range can be an artificially set range.

[0183] Specifically, if the above distance is within the normal distance range, it means that the subject to be measured does not have an unhealthy sitting posture such as leaning forward or backward; if the above height is within the normal height range, it means that the subject to be measured does not have an unhealthy sitting posture such as bending over or hunching over, and it can be considered that the subject to be measured has an upright body.

[0184] When the above two conditions are met, the target image of the object to be measured captured by the camera is obtained; when either one of the above two conditions or both are not met, it is considered that the object to be measured has an unhealthy sitting posture.

[0185] Furthermore, since there can be multiple facial feature points and thus multiple calculated heights, a normal height range corresponding to each facial feature point can be preset. A determination can then be made as to whether the height between the facial feature point and the surface of the desktop lies within the normal height range for that facial feature point. After performing the aforementioned processing for each facial feature point, the number of facial feature points whose corresponding heights fall within the normal height range is counted. If this number exceeds a preset threshold and the distance is within the normal distance range, the target image of the subject captured by the camera is obtained.

[0186] From the above, it can be seen that when the solution provided in the embodiment is used for sitting posture detection, if the above distance is not within the normal distance range and / or the above height is not within the normal height range, it can be determined that the sitting posture of the object to be measured is an unhealthy sitting posture. Therefore, the above distance and the above height can be used to preliminarily determine whether the sitting posture of the object to be measured is an unhealthy sitting posture, which can improve the efficiency of sitting posture detection.

[0187] In one embodiment of the present application, after measuring the distance between the object to be measured and the sitting posture detection device using a ranging sensor, it is possible to first determine whether the distance is within a normal distance range. If so, a target image of the object to be measured captured by the camera is obtained, and based on the target image, the height between the facial feature points and the plane where the desktop is located is calculated. If the height is within a normal height range, the bone points of the preset body parts of the object to be measured are further detected in the target image, and the sitting posture type of the object to be measured is determined based on the position information of different bone points.

[0188] When calculating the height between the facial feature points and the plane where the desktop is located, in addition to the method mentioned in step S102F above, the following method can also be used: Figure 6a In the embodiment shown, steps S102F1 - S102F5 are implemented.

[0189] In one embodiment of the present application, see Figure 6a, a flow chart of a sixth sitting posture detection method is provided. In this embodiment, the camera in the sitting posture detection device is horizontally oriented toward the object to be detected, and the above step S102F can be implemented by the following steps S102F1-S102F5.

[0190] Step S102F1: Counting the number of first pixel points between the first pixel point and the center pixel point in the pixel column direction of the image to be tested.

[0191] The central pixel is located at the center of the image to be measured.

[0192] Specifically, the pixel row number where the first pixel point is located and the pixel row number where the middle pixel point is located can be determined in the image to be tested, and then the two determined pixel row numbers are subtracted, and the result obtained is the number of the first pixel points.

[0193] For example, if the first pixel is located at the 20th pixel row of the image to be tested and the center pixel is located at the 320th pixel row of the image to be tested, then subtracting the two pixel rows, the number of first pixels obtained is 320-20=300.

[0194] Step S102F2: Based on the pre-obtained conversion relationship between the number of pixels and the length, a first height between the facial feature point and the horizontal plane where the camera is located is calculated according to the first number of pixels.

[0195] Specifically, the first number of pixels is the number of pixels between the first pixel and the center pixel, and the center pixel is the pixel located at the center of the image to be measured. Since the camera is facing the object to be measured in the horizontal direction, the above-mentioned first number of pixels reflects the height between the facial feature points in the image to be measured and the horizontal plane where the camera is located. Then, based on the pre-obtained conversion relationship between the number of pixels and the length, the first pixel format can be converted into the height between the facial feature points of the object to be measured and the horizontal plane where the camera is located in the actual scene.

[0196] Step S102F3: Count the number of second pixel points between the second pixel point and the center pixel point in the pixel column direction.

[0197] This step is similar to the above step S102F1 and will not be described in detail here.

[0198] Step S102F4: Based on the conversion relationship, the second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels.

[0199] This step is similar to the above step S102F2 and will not be described in detail here.

[0200] Step S102F5: Calculate the sum of the first height and the second height to obtain the height between the facial feature point and the plane where the desktop is located.

[0201] The first height is the height between the facial feature points of the object to be measured and the horizontal plane where the camera is located. The second height is the height between the horizontal plane where the camera is located and the plane where the desktop is located. The first height and the second height are added together, and the result of the addition is the height between the facial feature points of the object to be measured and the plane where the desktop is located.

[0202] Figure 6b A schematic diagram of a sitting posture detection scenario. Figure 6b In the figure, h1 represents the first height, and h2 represents the second height. By adding h1 and h2, the height between the facial feature points of the object to be measured and the plane where the desktop is located can be obtained.

[0203] It can be seen from the above that when the solution provided in the embodiment of the present application is used for sitting posture detection, based on the conversion relationship between the number of pixels and the length, the above-mentioned first height and second height can be accurately calculated according to the number of first pixels and the number of second pixels, and then the first height and the second height are added together to accurately calculate the height between the facial feature points of the object to be measured and the plane where the desktop is located.

[0204] In one embodiment of the present application, the above conversion relationship is expressed as follows:

[0205] L=D*tan(N*R+θ)

[0206] Wherein, L represents the length, D represents the distance between the sitting posture detection device and the object to be measured, N represents the number of pixels, R represents the radian value corresponding to the preset unit pixel, and θ represents the preset radian value error.

[0207] In one embodiment of the present application, the above R and θ can be obtained by calibration.

[0208] The specific calibration method is: after installing the sitting posture detection device and the calibration object, first measure the horizontal distance D1 and height L1 between the sitting posture detection device and the calibration object, and then in the image of the calibration object captured by the camera, count the number of pixels N1 between the pixel point where the calibration object is located and the pixel point at the center of the image in the direction of the image pixel column, substitute the obtained D1, L1, and N1 into the above expressions to obtain a first set of expressions, and then change the position of the sitting posture detection device or the calibration object, and measure the horizontal distance D2 and height L2 between the sitting posture detection device and the calibration object again, and in the image of the calibration object re-captured by the camera, count the number of pixels N2 between the pixel point where the calibration object is located and the pixel point at the center of the image in the direction of the image pixel column, substitute the obtained D2, L2, and N2 into the above formula to obtain a second set of expressions, and by adjusting the position of the sitting posture detection device or the calibration object multiple times, you can obtain multiple sets of expressions, and calculate R and θ based on these multiple sets of expressions.

[0209] It can be seen from the above that when the solution provided in the embodiment of the present application is used to perform sitting posture detection, when the above-mentioned expression is used to calculate the above-mentioned first height, N represents the number of the above-mentioned first pixel points, and N*R+θ represents the angle between the straight line where the sitting posture detection device and the facial feature points are located and the horizontal plane where the camera is located. Therefore, the above-mentioned first height can be accurately calculated using the above-mentioned expression. Similarly, the above-mentioned second height can also be accurately calculated using the above-mentioned expression. By adding the first height and the second height, the height between the facial feature points of the object to be measured and the plane where the desktop is located can be accurately calculated, thereby improving the accuracy of sitting posture detection.

[0210] In one embodiment of the present application, see Figure 7 , provides a flow chart of a seventh sitting posture detection method. In this embodiment, after obtaining the sitting posture type of the object to be detected, the following step S104 is also included.

[0211] Step S104: If the obtained sitting posture type belongs to an unhealthy sitting posture type, an unhealthy sitting posture image of the subject to be measured captured by a camera is obtained.

[0212] Specifically, it can be determined whether the obtained sitting posture type belongs to an improper sitting posture type. If so, the improper sitting posture image captured by the camera is obtained. If not, it means that the sitting posture of the subject to be measured is a standard sitting posture, and no operation is performed.

[0213] In one embodiment of the present application, after determining that the sitting posture type of the object to be measured is an unhealthy sitting posture type, the camera can be instructed to capture an image of the object to be measured, and the image captured by the camera can be used as the unhealthy sitting posture image of the object to be measured. In addition, the above-mentioned target image can also be directly used as the unhealthy sitting posture image of the object to be measured.

[0214] In the sitting posture detection scheme provided by the above embodiment, when the subject to be tested has an unhealthy sitting posture, an image of the subject's unhealthy sitting posture can be collected. This is conducive to the subsequent analysis of the subject's sitting posture habits based on the collected unhealthy sitting posture images, and forming a sitting posture curve report of the subject to be tested, so as to better correct the subject's unhealthy sitting posture.

[0215] In one embodiment of the present application, see Figure 8 , provides a flow chart of an eighth sitting posture detection method. In this embodiment, after obtaining the sitting posture type of the object to be detected, the following step S105 is also included.

[0216] Step S105: If the obtained sitting posture type belongs to an unhealthy sitting posture type, a sitting posture reminder message is generated to remind the subject to correct the sitting posture.

[0217] The sitting posture reminder information may be one or more of voice broadcast information, text display information or other information used for reminder.

[0218] Specifically, it can be determined whether the obtained sitting posture type belongs to an unhealthy sitting posture type. If so, a sitting posture reminder message is generated to remind the subject to be tested to correct the sitting posture; if not, it means that the sitting posture of the subject to be tested is a standard sitting posture and no sitting posture reminder message needs to be generated.

[0219] In one embodiment of the present application, the above-mentioned sitting posture detection device may include a voice broadcast unit in addition to a ranging sensor and a camera, and the above-mentioned sitting posture reminder information may be voice broadcast information, so that after the sitting posture reminder information is generated, it can be played by the above-mentioned voice broadcast unit.

[0220] In addition, the above-mentioned sitting posture detection device may further include one or more of the following devices:

[0221] Intercom unit, used to realize intercom function;

[0222] Voice unit, used to implement voice function;

[0223] Infrared unit, used to supplement infrared light so that the camera can capture images at night;

[0224] Communication network interface, used to achieve network connection.

[0225] As can be seen from the above, in the sitting posture detection solution provided in the embodiment of the present application, when the subject to be tested has an unhealthy sitting posture, sitting posture reminder information can be generated. In this way, when the subject to be tested has an unhealthy sitting posture, the subject to be tested can be reminded to correct the sitting posture, so that the subject to be tested can always maintain a standard sitting posture.

[0226] In one embodiment of the present application, see Figure 9, a flow chart of a ninth sitting posture detection method is provided. In this embodiment, the above method is applied to a sitting posture detection device, which includes a ranging sensor and a camera. The above camera is deployed on a desktop, and the camera is horizontally facing the object to be measured. The above method includes the following steps S901-S913.

[0227] Step S901: using a distance measuring sensor to measure the distance between the object to be measured and the sitting posture detection device.

[0228] This step is the same as above Figure 1 Step S101 in the illustrated embodiment is similar and will not be described again here.

[0229] Step S902: Detect the target image of the object to be measured captured by the camera to obtain the first pixel point where the preset facial feature point of the object to be measured is located and the second pixel point where the adjacent desktop boundary is located, wherein the adjacent desktop boundary is the boundary of the desktop closest to the object to be measured.

[0230] This step is the same as above Figure 5 Step S102E in the illustrated embodiment is similar and will not be described again here.

[0231] Step S903: Counting the number of first pixel points between the first pixel point and the center pixel point in the pixel column direction of the image to be tested, wherein the center pixel point is located at the center of the image to be tested.

[0232] Step S904: Based on the pre-obtained conversion relationship between the number of pixels and the length, a first height between the facial feature point and the horizontal plane where the camera is located is calculated according to the first number of pixels.

[0233] Step S905: Counting the number of second pixel points between the second pixel point and the center pixel point in the pixel column direction.

[0234] Step S906: Based on the conversion relationship, a second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels.

[0235] Step S907: Calculate the sum of the first height and the second height to obtain the height between the facial feature point and the plane where the desktop is located.

[0236] The above steps S903-S907 are respectively Figure 6a Steps S102F1-S102F5 in the illustrated embodiment are similar and will not be described in detail here.

[0237] Step S908: If the distance is within the preset normal distance range and the height is within the preset normal height range, then detecting the skeleton points of the preset body part of the object to be measured in the target image.

[0238] Step S909: Calculating object sitting posture data reflecting the sitting posture of the object to be measured based on the position information of the skeleton points of the preset body parts of the object to be measured.

[0239] The above steps S908 and S909 are respectively Figure 2a Steps S102A and S102B in the illustrated embodiment are similar and will not be described in detail here.

[0240] Step S910: determining a target sub-distance range in each sub-distance range included in the normal distance range, and determining target sitting posture data corresponding to the target sub-distance range in each set of preset target sitting posture data as standard sitting posture data.

[0241] Each set of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

[0242] This step is the same as above Figure 3 Step S103 in the illustrated embodiment is similar and will not be described in detail here.

[0243] Step S911: Compare the calculated sitting posture data of the subject with the preset standard sitting posture data, and determine the sitting posture type of the subject to be measured based on the comparison result.

[0244] This step is the same as above Figure 2a Step S102C in the illustrated embodiment is similar and will not be described again here.

[0245] Step S912: If the obtained sitting posture type belongs to an unhealthy sitting posture type, an unhealthy sitting posture image of the subject to be measured captured by a camera is obtained.

[0246] This step is the same as above Figure 7 Step S104 in the illustrated embodiment is similar and will not be described again here.

[0247] Step S913: If the obtained sitting posture type belongs to an unhealthy sitting posture type, a sitting posture reminder message is generated to remind the subject to correct the sitting posture.

[0248] This step is the same as above Figure 8 Step S105 in the illustrated embodiment is similar and will not be described again here.

[0249] Corresponding to the above-mentioned sitting posture detection method, an embodiment of the present application also provides a sitting posture detection device.

[0250] See also Figure 10 , provides a structural schematic diagram of a sitting posture detection device 100, the device comprising a ranging sensor 1001, a camera 1002, and a processor 1003, wherein:

[0251] The distance measuring sensor 1001 is used to measure the distance between the object to be measured and itself, and send the measured distance to the processor 1003;

[0252] The camera 1002 is used to capture a target image of the object to be measured and send the captured target image to the processor 1002;

[0253] The processor 1003 is configured to execute any one of the steps of the sitting posture detection method in the above method embodiment.

[0254] As can be seen from the above, when applying the solution provided by the embodiment of the present application to perform sitting posture detection, the distance between the object to be measured and the sitting posture detection device is first measured. When the measured distance is within the normal distance range, it means that the sitting posture of the object to be measured may be normal. In order to further determine the sitting posture type of the object to be measured, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera can be detected, and the sitting posture type of the object to be measured can be determined based on the position information of the different skeletal points detected. In this way, the sitting posture detection of the object to be measured can be realized by only using the distance measured by the distance measuring sensor and the target image captured by the camera. The amount of data required is relatively small, thereby speeding up the data processing speed. It can be seen that the sitting posture detection efficiency can be improved by applying the sitting posture detection solution provided by the embodiment of the present application.

[0255] In another embodiment of the present application, see Figure 11 , provides a structural diagram of a second sitting posture detection device. In addition to the distance measuring sensor 1001, the camera 1002, and the processor 1003, the device also includes the following components:

[0256] Intercom unit 1004, used to implement intercom function;

[0257] Voice unit 1005, used to implement voice function;

[0258] Infrared unit 1006, used to supplement infrared light to facilitate the camera to capture images at night;

[0259] The communication network interface 1007 is used to achieve network connection.

[0260] In addition, in addition to the distance measuring sensor 1001, the camera 1002, and the processor 1003, the above-mentioned sitting posture detection device can also only include one or more devices among the above-mentioned intercom unit 1004, the voice unit 1005, the infrared unit 1006, and the communication network interface 1007.

[0261] Corresponding to the above-mentioned sitting posture detection method, an embodiment of the present application also provides a sitting posture detection device.

[0262] See also Figure 12 , provides a structural schematic diagram of a first sitting posture detection device, the device is applied to a sitting posture detection device, the sitting posture detection device includes a distance measuring sensor and a camera, the device includes:

[0263] The distance measurement module 1201 is configured to measure the distance between the object to be measured and the sitting posture detection device using the distance measurement sensor;

[0264] The sitting posture determination module 1202 is used to detect the bone points of the preset body parts of the object to be tested in the target image of the object to be tested captured by the camera when the distance is within the preset normal distance range, and determine the sitting posture type of the object to be tested based on the position information of the different detected bone points.

[0265] As can be seen from the above, when applying the solution provided by the embodiment of the present application to perform sitting posture detection, the distance between the object to be measured and the sitting posture detection device is first measured. When the measured distance is within the normal distance range, it means that the sitting posture of the object to be measured may be normal. In order to further determine the sitting posture type of the object to be measured, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera can be detected, and the sitting posture type of the object to be measured can be determined based on the position information of the different skeletal points detected. In this way, the sitting posture detection of the object to be measured can be realized by only using the distance measured by the distance measuring sensor and the target image captured by the camera. The amount of data required is relatively small, thereby speeding up the data processing speed. It can be seen that the sitting posture detection efficiency can be improved by applying the sitting posture detection solution provided by the embodiment of the present application.

[0266] In one embodiment of the present application, see Figure 13 , provides a structural diagram of a second sitting posture detection device. In this embodiment, the sitting posture determination module 1202 includes:

[0267] a skeleton point detection submodule 1202A, configured to detect skeleton points of a preset body part of the subject to be measured in a target image of the subject to be measured captured by the camera when the distance is within a preset normal distance range;

[0268] The data calculation submodule 1202B is configured to calculate the object sitting posture data reflecting the sitting posture of the object to be measured based on the position information of the skeletal points of the preset body parts of the object to be measured;

[0269] The data comparison submodule 1202C is used to compare the calculated object sitting posture data with the preset standard sitting posture data, and determine the sitting posture type of the object to be measured based on the comparison result, wherein the standard sitting posture data is used to reflect the standard sitting posture of the object to be measured.

[0270] As can be seen from the above, when the solution provided in the embodiment of the present application is used for sitting posture detection, the object sitting posture data is calculated based on the position information of the above-mentioned skeleton points, and then the object sitting posture data is compared with the standard sitting posture data, and the sitting posture type of the object to be measured is determined according to the comparison result. Since the object sitting posture data reflects the sitting posture of the object to be measured at the time of acquisition of the target image, and the standard sitting posture data reflects the standard sitting posture of the object to be measured, comparing the object sitting posture data with the standard sitting posture data can be understood as comparing the sitting posture of the object to be measured at the time of acquisition of the target image with the standard sitting posture. Therefore, by comparing the object sitting posture data and the standard sitting posture data, the sitting posture type of the object to be measured can be accurately determined according to the comparison results.

[0271] In one embodiment of the present application, see Figure 14 , provides a structural schematic diagram of a third sitting posture detection device. In this embodiment, the device further includes:

[0272] The data selection module 1203 is used to compare the calculated object sitting posture data with the preset standard sitting posture data, and before determining the sitting posture type of the object to be measured based on the comparison result, determine the target sub-distance range in which the distance is located in each sub-distance range included in the normal distance range, and determine the target sitting posture data corresponding to the target sub-distance range in each preset group of target sitting posture data as the standard sitting posture data, wherein each group of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

[0273] As can be seen from the above, when the solution provided in the embodiment of the present application is used for sitting posture detection, the target sub-distance range in which the above distance is located is determined in each sub-distance range included in the normal distance range, and the target sitting posture data corresponding to the target sub-distance range is determined in each set of preset target sitting posture data as standard sitting posture data. Since the distance between the sitting posture detection device and the object to be measured is different, the standard sitting posture of the object to be measured is usually different, and the standard sitting posture data reflecting the standard sitting posture of the object to be measured is also different. By determining the above-mentioned target sub-distance range and determining the target sitting posture data corresponding to the target sub-distance range, more accurate standard sitting posture data reflecting the standard sitting posture of the object to be measured can be obtained, thereby comparing the object sitting posture data with the target sitting posture data as the standard sitting posture data, and based on the comparison results, the sitting posture type of the object to be measured can be more accurately determined.

[0274] In one embodiment of the present application, the sitting posture determination module 1202 is specifically configured to:

[0275] When the distance is within a preset normal distance range, the target image of the object to be measured captured by the camera is input into a pre-trained sitting posture detection model to obtain the sitting posture type of the object to be measured output by the sitting posture detection model, wherein the sitting posture detection model is trained using the image of the sample object as input and the sitting posture type of the sample object as annotation information.

[0276] As can be seen from the above, when applying the solution provided in the embodiment of the present application to perform sitting posture detection, the target image is input into the sitting posture detection model. Since the sitting posture detection model is a pre-trained model, the sitting posture detection model learns the features of determining the sitting posture type based on the input image. In this way, after the target image is input into the sitting posture detection model, the sitting posture detection model can accurately determine the sitting posture type of the object to be tested based on the learned features.

[0277] In one embodiment of the present application, see Figure 15 , provides a structural schematic diagram of a fourth sitting posture detection device. In this embodiment, the sitting posture detection device is deployed on a desktop,

[0278] The sitting posture determination module 1202 includes:

[0279] An image detection submodule 1202D is configured to detect a target image of the object to be measured captured by the camera, and obtain a first pixel point where a preset facial feature point of the object to be measured is located, and a second pixel point where an adjacent desktop boundary is located, wherein the adjacent desktop boundary is the boundary of the desktop closest to the object to be measured;

[0280] a height calculation submodule 1202E, configured to calculate the height between the facial feature point and the plane where the desktop is located based on the position of the first pixel point and the position of the second pixel point;

[0281] The sitting posture determination submodule 1202F is used to detect the bone points of the preset body parts of the object to be measured in the target image if the distance is within the preset normal distance range and the height is within the preset normal height range, and determine the sitting posture type of the object to be measured based on the position information of the different detected bone points.

[0282] From the above, it can be seen that when the solution provided in the embodiment is used for sitting posture detection, if the above distance is not within the normal distance range and / or the above height is not within the normal height range, it can be determined that the sitting posture of the object to be measured is an unhealthy sitting posture. Therefore, the above distance and the above height can be used to preliminarily determine whether the sitting posture of the object to be measured is an unhealthy sitting posture, which can improve the efficiency of sitting posture detection.

[0283] In one embodiment of the present application, the camera in the sitting posture detection device is directed horizontally toward the object to be detected;

[0284] The height calculation submodule 1202E is specifically configured to:

[0285] Counting the number of first pixel points between the first pixel point and a central pixel point in a pixel column direction of the target image to be tested, wherein the central pixel point is located at the center of the target image to be tested;

[0286] Based on a pre-obtained conversion relationship between the number of pixels and length, calculating a first height between the facial feature point and a horizontal plane where the camera is located according to the first number of pixels;

[0287] Counting the number of second pixel points between the second pixel point and the central pixel point in the direction of the pixel column;

[0288] Based on the conversion relationship, a second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels;

[0289] The sum of the first height and the second height is calculated to obtain the height between the facial feature point and the plane where the desktop is located.

[0290] It can be seen from the above that when the solution provided in the embodiment of the present application is used for sitting posture detection, based on the conversion relationship between the number of pixels and the length, the above-mentioned first height and second height can be accurately calculated according to the number of first pixels and the number of second pixels, and then the first height and the second height are added together to accurately calculate the height between the facial feature points of the object to be measured and the plane where the desktop is located.

[0291] In one embodiment of the present application, the conversion relationship is expressed as follows:

[0292] L=D*tan(N*R+θ)

[0293] Among them, L represents length, D represents the distance between the sitting posture detection device of the image acquisition device and the object to be measured, N represents the number of pixels, R represents the radian value corresponding to the preset unit pixel point, and θ represents the preset radian value error.

[0294] It can be seen from the above that when the solution provided in the embodiment of the present application is used to perform sitting posture detection, when the above-mentioned expression is used to calculate the above-mentioned first height, N represents the number of the above-mentioned first pixel points, and N*R+θ represents the angle between the straight line where the sitting posture detection device and the facial feature points are located and the horizontal plane where the camera is located. Therefore, the above-mentioned first height can be accurately calculated using the above-mentioned expression. Similarly, the above-mentioned second height can also be accurately calculated using the above-mentioned expression. By adding the first height and the second height, the height between the facial feature points of the object to be measured and the plane where the desktop is located can be accurately calculated, thereby improving the accuracy of sitting posture detection.

[0295] In one embodiment of the present application, see Figure 16 , provides a structural schematic diagram of a fifth sitting posture detection device. In this embodiment, the device further includes:

[0296] The image acquisition module 1204 is used to detect the skeletal points of the preset body parts of the object to be tested in the target image of the object to be tested captured by the camera, and determine the sitting posture type of the object to be tested based on the position information of the different skeletal points detected. If the obtained sitting posture type belongs to an unhealthy sitting posture type, the unhealthy sitting posture image of the object to be tested captured by the camera is recorded.

[0297] As can be seen from the above, in the sitting posture detection solution provided in the embodiment of the present application, when the subject to be tested has a bad sitting posture, the bad sitting posture image of the subject to be tested can be collected. This is conducive to the subsequent analysis of the sitting posture habits of the subject to be tested based on the collected bad sitting posture image, and forming a sitting posture curve report of the subject to be tested, so as to better correct the bad sitting posture of the subject to be tested.

[0298] In one embodiment of the present application, see Figure 17 , provides a structural schematic diagram of a sixth sitting posture detection device. In this embodiment, the device further includes:

[0299] The information generation module 1205 is used to detect the skeletal points of the preset body parts of the object to be tested in the target image of the object to be tested captured by the camera, and determine the sitting posture type of the object to be tested based on the position information of the different skeletal points detected. If the obtained sitting posture type belongs to an unhealthy sitting posture type, sitting posture reminder information is generated to remind the object to be tested to correct its sitting posture.

[0300] As can be seen from the above, in the sitting posture detection solution provided in the embodiment of the present application, when the subject to be tested has an unhealthy sitting posture, sitting posture reminder information can be generated. In this way, when the subject to be tested has an unhealthy sitting posture, the subject to be tested can be reminded to correct the sitting posture, so that the subject to be tested can always maintain a standard sitting posture.

[0301] The present application also provides an electronic device, such as Figure 18As shown, it includes a processor 1801, a communication interface 1802, a memory 1803 and a communication bus 1804, wherein the processor 1801, the communication interface 1802, and the memory 1803 communicate with each other through the communication bus 1804.

[0302] Memory 1803, used for storing computer programs;

[0303] The processor 1801 is configured to execute the program stored in the memory 1803 to implement the following steps:

[0304] Measuring the distance between the object to be measured and the sitting posture detection device using the distance measuring sensor;

[0305] When the distance is within a preset normal distance range, the skeletal points of the preset body parts of the object to be measured in the target image of the object to be measured captured by the camera are detected, and the sitting posture type of the object to be measured is determined based on the position information of the different detected skeletal points.

[0306] Other solutions for implementing sitting posture detection by the processor 1801 executing the program stored in the memory 1803 are the same as those mentioned in the aforementioned method embodiment and will not be repeated here.

[0307] The communication bus mentioned in the electronic device mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0308] The communication interface is used for communication between the above electronic device and other devices.

[0309] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0310] The above-mentioned 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 gate or transistor logic devices, and discrete hardware components.

[0311] In another embodiment provided in the present application, a computer-readable storage medium is further provided, wherein a computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of any of the above-mentioned sitting posture detection methods are implemented.

[0312] In another embodiment provided by the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute any sitting posture detection method in the above embodiments.

[0313] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0314] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0315] Each embodiment in this specification is described in a related manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device, apparatus, electronic device, computer-readable storage medium, and computer program product embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For related portions, reference can be made to the descriptions of the method embodiments.

[0316] The above description is only a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application are included in the scope of protection of the present application.

Claims

1. A sitting posture detection method, characterized in that: The method is applied to a sitting posture detection device, the sitting posture detection device including a distance measuring sensor and a camera, and the sitting posture detection device is deployed on a desktop. The method includes: Measuring the distance between the object to be measured and the sitting posture detection device using the distance measuring sensor; Detecting the target image of the object to be measured captured by the camera to obtain a first pixel point where a preset facial feature point of the object to be measured is located and a second pixel point where an adjacent desktop boundary is located, wherein the adjacent desktop boundary is the boundary of the desktop closest to the object to be measured; Calculating the height between the facial feature point and the plane where the desktop is located based on the position of the first pixel point and the position of the second pixel point; If the distance is within a preset normal distance range, and the height is within a preset normal height range, the skeletal points of the preset body parts of the object to be measured in the target image are detected, and based on the position information of the skeletal points of the preset body parts of the object to be measured, the object sitting posture data reflecting the sitting posture of the object to be measured is calculated; the calculated object sitting posture data is compared with the preset standard sitting posture data, and the sitting posture type of the object to be measured is determined according to the comparison result, wherein the standard sitting posture data is used to reflect the standard sitting posture of the object to be measured, and the standard sitting posture data includes: the range of relative position information between the skeletal points of different preset body parts under the standard sitting posture, and the object sitting posture data includes the relative position information between the skeletal points of different preset body parts of the object to be measured; If the distance is not within the normal distance range and / or the height is not within the normal height range, determining that the sitting posture of the subject to be measured is an unhealthy sitting posture; Before comparing the calculated sitting posture data of the subject with preset standard sitting posture data and determining the sitting posture type of the subject to be measured based on the comparison result, the method further includes: A target sub-distance range in which the distance is located is determined in each sub-distance range included in the normal distance range, and target sitting posture data corresponding to the target sub-distance range is determined in each preset group of target sitting posture data as standard sitting posture data, wherein each group of target sitting posture data corresponds to a sub-distance range, and the target sitting posture data is used to reflect the standard sitting posture of the object to be measured when the distance between the object to be measured and the sitting posture detection device is within the corresponding sub-distance range.

2. The method according to claim 1, characterized in that When the distance is within a preset normal distance range, detecting skeletal points of a preset body part of the subject to be measured in a target image of the subject to be measured captured by the camera, and determining the sitting posture type of the subject to be measured based on position information of different detected skeletal points, including: When the distance is within a preset normal distance range, the target image of the object to be measured captured by the camera is input into a pre-trained sitting posture detection model to obtain the sitting posture type of the object to be measured output by the sitting posture detection model, wherein the sitting posture detection model is trained using the image of the sample object as input and the sitting posture type of the sample object as annotation information.

3. The method according to claim 1, characterized in that The camera in the sitting posture detection device is directed horizontally toward the object to be detected; The calculating, based on the position of the first pixel point and the position of the second pixel point, the height between the facial feature point and the plane where the desktop is located, includes: Counting the number of first pixel points between the first pixel point and a central pixel point in a pixel column direction of the image to be measured, wherein the central pixel point is located at the center of the image to be measured; Based on a pre-obtained conversion relationship between the number of pixels and length, calculating a first height between the facial feature point and a horizontal plane where the camera is located according to the first number of pixels; Counting the number of second pixel points between the second pixel point and the central pixel point in the direction of the pixel column; Based on the conversion relationship, a second height between the plane where the desktop is located and the horizontal plane where the camera is located is calculated according to the second number of pixels; The sum of the first height and the second height is calculated to obtain the height between the facial feature point and the plane where the desktop is located.

4. The method according to claim 3, characterized in that The conversion relationship is expressed as follows: L=D*tan(N*R+θ) Wherein, L represents the length, D represents the distance between the sitting posture detection device and the object to be measured, N represents the number of pixels, R represents the radian value corresponding to the preset unit pixel, and θ represents the preset radian value error.

5. The method according to any one of claims 1 to 4, characterized in that After detecting the skeletal points of the preset body parts of the subject to be measured in the target image of the subject to be measured captured by the camera, and determining the sitting posture type of the subject to be measured based on the position information of different detected skeletal points, the method further includes: If the obtained sitting posture type belongs to an unhealthy sitting posture type, the unhealthy sitting posture image of the object to be measured captured by the camera is recorded.

6. The method according to any one of claims 1 to 4, characterized in that After detecting the skeletal points of the preset body parts of the subject to be measured in the target image of the subject to be measured captured by the camera, and determining the sitting posture type of the subject to be measured based on the position information of different detected skeletal points, the method further includes: If the obtained sitting posture type belongs to an unhealthy sitting posture type, a sitting posture reminder message is generated to remind the subject to be measured to correct the sitting posture.

7. A sitting posture detection device, characterized in that: The device includes a ranging sensor, a camera, and a processor, wherein: The distance measuring sensor is used to measure the distance between the object to be measured and itself, and send the measured distance to the processor; The camera is used to capture a target image of the object to be measured and send the captured target image to the processor; The processor is configured to execute the method steps described in any one of claims 1-6.

8. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the method steps described in any one of claims 1 to 6 when executing a program stored in a memory.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps of any one of claims 1 to 6 are implemented.

Images