Sitting posture health assessment method and device, equipment, storage medium and program product

Abstract

The embodiment of the invention provides a sitting posture health assessment method and device, equipment, a storage medium and a program product. The method comprises the following steps: acquiring a sitting posture image of a driver and pressure distribution time sequence data of a driver seat; performing feature extraction on the sitting posture image and the pressure distribution time sequence data to generate a multi-dimensional sitting posture feature vector; and inputting the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, and performing feature comparison and score calculation through a healthy sitting posture database to obtain a score result of the driver. According to the invention, sitting posture assessment is realized through multi-modal data, the comprehensiveness and accuracy of assessment are improved, and meanwhile, accurate sitting posture correction guidance and health risk early warning based on a sitting posture assessment result are supported.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of driver sitting posture monitoring, and in particular to a sitting posture health evaluation method and device, equipment, a storage medium and a program product. BACKGROUND

[0002] In the modern automobile driving scene, the driver keeps a bad sitting posture for a long time, which can easily cause a series of health problems such as back pain, cervical strain, and blood circulation disorders. Especially in the long-distance driving or high-frequency short-distance commuting working conditions, the muscle tension and skeletal stress imbalance problems caused by the bad sitting posture will continue to worsen, not only greatly reducing the driving comfort, but also possibly distracting the driving attention due to physical fatigue, thereby burying the driving safety hazards. For example, if the seat backrest angle is not adjusted properly, it will cause abnormal physiological curvature of the spine, uneven stress on the vertebral body, and long-term induction of lumbar disc herniation. Placing the arms too high or too low will cause the muscles of the shoulder and neck to be in a constant state of tension, causing swelling, stiffness, and even chronic strain.

[0003] Although existing vehicles are generally equipped with basic adjustment functions such as seat height and backrest angle, such functions can only meet the basic seating needs of the driver, and cannot real-time perceive and monitor the dynamic changes of the sitting posture during driving, nor can they provide accurate sitting posture correction guidance and health risk warnings based on the monitored sitting posture data.

[0004] Therefore, there is an urgent need for a driver sitting posture health evaluation scheme that adapts to the driving scene to provide personalized sitting posture adjustment suggestions for the driver, thereby effectively alleviating physical fatigue during driving and improving driving safety and comfort. SUMMARY

[0005] The sitting posture health evaluation method, device, equipment, storage medium and program product provided by the embodiments of the present application realize accurate and comprehensive evaluation of the sitting posture by collecting multi-modal data, provide a reliable basis for sitting posture adjustment suggestions, and help to improve the comfort and safety of the driver during driving.

[0006] In a first aspect, the embodiments of the present application provide a sitting posture health evaluation method, comprising: acquiring a sitting posture image of a driver and pressure distribution time series data of a driver's seat; performing feature extraction on the sitting posture image and the pressure distribution time series data to generate a multi-dimensional sitting posture feature vector; inputting the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, and performing feature comparison and score calculation through a health sitting posture database to obtain a score result of the driver.

[0007] In a possible implementation, the multi-dimensional sitting posture feature vector includes a visual feature vector and a pressure correlation vector, and the method of feature extraction on the sitting posture image and the pressure distribution time series data to generate the multi-dimensional sitting posture feature vector includes: preprocessing the sitting posture image; extracting a body key point feature of the driver in the preprocessed sitting posture image, and generating the visual feature vector based on the extracted body key point feature; preprocessing the pressure distribution time series data; determining a pressure dynamic feature and a pressure static feature based on the preprocessed pressure distribution time series data, and generating the pressure correlation vector based on the pressure dynamic feature and the pressure static feature; the pressure dynamic feature is used to represent the change of the pressure, and the pressure static feature includes the uniformity of the pressure distribution and the position of the pressure center point.

[0008] In a possible implementation, the multi-dimensional sitting posture feature vector is input into a pre-trained sitting posture evaluation model, and feature comparison and score calculation are performed on the health sitting posture database to obtain a score result of the driver, including: inputting the visual feature vector and the pressure correlation vector into the sitting posture evaluation model; determining a sitting posture angle of the driver based on the visual feature vector, determining a sitting posture stability based on the visual feature vector and the pressure correlation vector, and determining a pressure distribution feature based on the pressure correlation vector via the sitting posture evaluation model; comparing the sitting posture angle, the sitting posture stability and the pressure distribution feature with corresponding reference features of each sitting posture sample in the health sitting posture database; and determining and outputting the score result of the driver based on the comparison result.

[0009] In a possible implementation, the comparison result includes a difference proportion of each feature being compared; the difference proportion is a proportion of a difference value of two features being compared in the reference feature; and the score result of the driver is determined and output based on the comparison result, including: for each feature in the sitting posture angle, the sitting posture stability and the pressure distribution feature, the difference proportion of the feature is mapped to a corresponding score value based on the difference proportion of the feature and a score value mapping relationship corresponding to the feature; and the scores mapped by each feature in the sitting posture angle, the sitting posture stability and the pressure distribution feature are output.

[0010] In a possible implementation, the score result further includes a health suggestion; and the method further includes:

[0011] generating the health suggestion based on the number of features with scores less than a preset score among the scores mapped by each feature in the sitting posture angle, the sitting posture stability and the pressure distribution feature; and storing the health suggestion and the scores mapped by each feature in the sitting posture angle, the sitting posture stability and the pressure distribution feature to a sitting posture record database.

[0012] In a possible implementation, the method further includes: generating and pushing a health report of the driver based on a statistical result of newly added data in a current statistical period in the sitting posture record database.

[0013] In a possible implementation, the obtaining of the sitting posture image of the driver and the pressure distribution time sequence data of the driver seat includes: triggering, by a hardware clock, a camera arranged in the cab and a pressure sensor embedded in the driver seat to perform sampling, so as to obtain the time-aligned sitting posture image and the pressure distribution time sequence data.

[0014] In a second aspect, the embodiments of the present application provide a sitting posture health assessment device, which includes: a multi-modal data acquisition module configured to acquire a sitting posture image of a driver and pressure distribution time sequence data of a driver seat; a sitting posture feature generation module configured to perform feature extraction on the sitting posture image and the pressure distribution time sequence data, and generate a multi-dimensional sitting posture feature vector; and a sitting posture assessment module configured to input the multi-dimensional sitting posture feature vector into a pre-trained sitting posture assessment model, perform feature comparison and score calculation through a healthy sitting posture database, and obtain a score result of the driver.

[0015] In a third aspect, the embodiments of the present application provide an electronic device, which includes: a memory and a processor; the memory stores computer execution instructions; and the processor executes the computer execution instructions stored in the memory, so that the processor executes the first aspect and / or various possible implementation manners of the first aspect.

[0016] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores computer execution instructions; and the computer execution instructions are executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0017] In a fifth aspect, the embodiments of the present application provide a computer program product, which includes a computer program; and the computer program is executed by a processor to implement the first aspect and / or various possible implementation manners of the first aspect.

[0018] The sitting posture health assessment method, device, equipment, storage medium and program product provided by the embodiments of the present application acquire the sitting posture image of the driver and the pressure distribution time sequence data of the driver seat, perform feature extraction on the sitting posture image and the pressure distribution time sequence data, generate a multi-dimensional sitting posture feature vector, input the multi-dimensional sitting posture feature vector into a pre-trained sitting posture assessment model, perform feature comparison and score calculation through a healthy sitting posture database, and obtain a score result of the driver. The multi-dimensional sitting posture feature vector generated by the sitting posture image and the pressure distribution time sequence data makes up for the limitation of a single data source, realizes real-time perception and monitoring of dynamic changes of the sitting posture in the driving process, avoids subjective bias through feature comparison and score calculation through the healthy sitting posture database, improves the accuracy and objectivity of the sitting posture health assessment, and helps to provide accurate sitting posture correction guidance and health risk warning based on the score result of the sitting posture score. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate embodiments consistent with the application and, together with the description, further serve to explain the principles of the application.

[0020] Figure 1 A scene schematic diagram of a driver's bad sitting posture provided for an embodiment of the application;

[0021] Figure 2 A flowchart of a sitting posture health assessment method provided for an embodiment of the application Figure 1 ;

[0022] Figure 3 A flowchart of a sitting posture health assessment method provided for an embodiment of the application Figure 2 ;

[0023] Figure 4 A structure schematic diagram of a sitting posture health assessment device provided for an embodiment of the application;

[0024] Figure 5 A structure schematic diagram of an electronic device provided for an embodiment of the application.

[0025] The above drawings have shown the specific embodiments of the application, which will be described in more detail hereinafter. These drawings and the written description are not intended to restrict the scope of the inventive concept by any means, but to illustrate the inventive concept to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0026] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The following description is made with reference to the accompanying drawings in which like reference numerals represent like elements, unless the context of use indicates otherwise. The following description of exemplary embodiments is not representative of all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as detailed in the appended claims.

[0027] “Multiple” refers to two or more, and other quantifiers are similar. “And / or” describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist together, and B exists alone. The character “ / ” generally represents that the associated objects before and after are in an “or” relationship.

[0028] The terms "first", "second", "third", "fourth" and the like in the description and in the claims of the present application, and above-mentioned drawings, if any, are used as identifiers to distinguish between similar objects, and are not necessarily intended to describe a particular sequential or chronological order. It is to be understood that the use of the terms so-termed "first", "second", "third", "fourth" and the like, if any, in the description and in the claims of the present application is interchangeable under appropriate circumstances, such that the embodiments of the application described herein are capable of embodiment in other sequences than the one depicted or otherwise specified herein. Furthermore, the terms "comprise", "comprising", "have", "having", "include", "including" and any variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises, has, includes or contains a list of steps or elements, but not only those, is not to be confined to the steps or elements so listed, but can include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

[0029] It should be noted that in the embodiments of the present application, the words "exemplary" or "for example" are used to mean example, illustration, or instance, and not "preferred" over other embodiments or designs. The use of these terms in the context of an embodiment or design in the specification is not related to a preference of one embodiment or design over another. Indeed, the use of any of these terms is intended to present one or more embodiments or designs in a particular fashion, but not to exclude other embodiments or designs.

[0030] Figure 1 A scene diagram of a driver's bad sitting posture is provided in the embodiments of the present application, as shown in Figure 1 In the modern automobile driving scene, the driver maintains a bad sitting posture for a long time, which can easily cause a series of health problems such as back pain, cervical strain, and blood circulation disorder. Especially in the long-distance driving or high-frequency short-distance commuting working conditions, the muscle tension and skeletal stress imbalance problems caused by the bad sitting posture will continue to worsen, not only greatly reducing the driving comfort, but also possibly distracting the driving attention due to physical fatigue, thereby burying the hidden danger of driving safety. For example, if the seat backrest angle is not adjusted properly, it will cause abnormal physiological curvature of the spine, uneven stress on the vertebral body, and long-term induction of lumbar disc herniation. The arm placement position is too high or too low, so that the muscles of the shoulder and neck are continuously in a stretched state, causing acidification, stiffness, and even chronic strain.

[0031] In combination with the above-mentioned scene, it can be known that the existing vehicle is generally equipped with basic adjustment functions such as seat height, backrest angle, etc., but such functions can only meet the basic seating needs of the driver, and cannot realize real-time sensing and monitoring of the dynamic changes of the sitting posture during driving, nor can they provide accurate sitting posture correction guidance and health risk warning in combination with the monitored sitting posture data.

[0032] The application provides a sitting posture health evaluation method, which combines the sitting posture image of the driver and the pressure distribution time sequence data of the driver seat, and performs feature comparison and score calculation through a healthy sitting posture database, thereby solving the problems in the prior art that the dynamic changes of the sitting posture during driving cannot be perceived and monitored in real time, and the sitting posture data cannot be combined for monitoring.

[0033] The technical solutions of the application and how the technical solutions solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the application will be described below with reference to the drawings.

[0034] Figure 2 A flowchart of a sitting posture health evaluation method provided by an embodiment of the application Figure 1 As shown in Figure 2 , the method comprises the following steps.

[0035] S201, acquiring a sitting posture image of a driver and pressure distribution time sequence data of a driver seat.

[0036] The sitting posture image of the driver is a visual form of the body posture of the driver on the vehicle seat. For example, it can be a picture, a video frame, a photo, etc.

[0037] The pressure distribution time sequence data is the sitting force point distribution data of the driver recorded in time sequence.

[0038] In this step, the sitting posture image of the driver and the pressure distribution time sequence data of the driver seat are acquired through a multi-modal data acquisition module mounted on the vehicle.

[0039] The multi-modal data acquisition module comprises a vehicle-mounted visual unit and a sensor system. The vehicle-mounted visual unit can be a camera, and the sensor system comprises a pressure sensor.

[0040] For example, the sitting posture image can be obtained by a camera installed in the driver's cabin to capture the photo of the driver in real time. The sitting posture image of the driver can also be obtained by acquiring the video stream of the driver during driving and through the video frames in the video stream.

[0041] Preferably, the sitting force point distribution data of the driver can be monitored in real time by a pressure sensor integrated in the driver seat, and the pressure distribution time sequence data of the driver seat can be obtained in time sequence.

[0042] Exemplarily, a driving seat image can also be acquired through a camera, a surface indentation state and a wrinkle form of the driving seat can be acquired through image recognition, a pressure distribution data of the driving seat can be calculated in combination with the sitting posture image and a biomechanics simulation model, and a pressure distribution time sequence data of the driving seat can be obtained in time sequence.

[0043] S202, feature extraction is performed on the sitting posture image and the pressure distribution time sequence data to generate a multi-dimensional sitting posture feature vector.

[0044] The feature extraction refers to extracting information describing the sitting posture from the sitting posture image and the pressure distribution time sequence data.

[0045] The multi-dimensional sitting posture feature vector may, for example, be a visual feature vector extracted from the sitting posture image, such as a trunk inclination angle, a joint angle, a head forward inclination angle, a hand position, a foot position, etc.

[0046] For example, it can also be a pressure distribution uniformity, a pressure center point position, a pressure change rate, etc. extracted from the pressure distribution time sequence data.

[0047] In this step, information describing the sitting posture, i.e., the multi-dimensional sitting posture feature vector, is generated by performing feature extraction on the sitting posture image and the pressure distribution time sequence data.

[0048] Specifically, a segmented image is obtained by performing image segmentation processing on the sitting posture image, and a trunk inclination angle, a joint angle, a head forward inclination angle, a hand position, a foot position, etc. are extracted from the segmented image through a machine learning algorithm or a geometric structure simulation. The segmented image refers to an image in which the driver is separated from the background. The pressure distribution uniformity, the pressure center point position, the pressure change rate, etc. are obtained by performing feature extraction on the pressure distribution time sequence data.

[0049] Exemplarily, the multi-dimensional sitting posture feature vector, such as the trunk inclination angle, the joint angle, the head forward inclination angle, the hand position, the foot position, etc. can be calculated by performing geometric structure simulation on the segmented image.

[0050] Exemplarily, the segmented image can also be subjected to body key point feature detection through a deep learning model, and the body key point features are associated to generate a skeleton sketch, and the sitting posture feature vector, such as the trunk inclination angle, the joint angle, the head forward inclination angle, the hand position, the foot position, etc. is extracted from the skeleton sketch. The deep learning model may, for example, be a convolutional neural network, a recurrent neural network, a Transformer model, etc.

[0051] Preferably, the sitting posture image and the pressure distribution time sequence data can be preprocessed first, and feature extraction can be performed based on the preprocessed sitting posture image and the pressure distribution time sequence data to improve data quality.

[0052] S203, input the multi-dimensional sitting posture feature vector into the pre-trained sitting posture evaluation model, perform feature comparison and score calculation through the healthy sitting posture database, and obtain a score result of the driver.

[0053] The sitting posture evaluation model is used for scoring the sitting posture of the driver. The sitting posture evaluation model can be constructed based on a deep learning model or a rule engine.

[0054] The healthy sitting posture database is used for storing sitting posture samples, i.e., healthy sitting posture parameters. The healthy sitting posture database can be stored in a local database of the vehicle or obtained through a cloud server. For example, the healthy sitting posture database includes ideal sitting posture angle ranges, such as trunk-thigh angles (100°, 110°) and elbow joint angles (90°, 120°).

[0055] In this step, the multi-dimensional sitting posture feature vector is input into the pre-trained sitting posture evaluation model, the healthy sitting posture parameters stored in the healthy sitting posture database are compared, and based on the comparison results of each sitting posture feature vector and the scoring rules, the scores of each sitting posture feature vector of the driver are obtained, and based on the sum of the scores, the score result of the driver is determined.

[0056] The sitting posture health evaluation method provided in the embodiments of the present application obtains the sitting posture image of the driver and the pressure distribution time series data of the driver's seat, extracts features from the sitting posture image and the pressure distribution time series data, generates a multi-dimensional sitting posture feature vector, inputs the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, performs feature comparison and score calculation through a healthy sitting posture database, and obtains a score result of the driver. The multi-dimensional sitting posture feature vector generated by the sitting posture image and the pressure distribution time series data compensates for the limitations of a single data source, realizes real-time perception and monitoring of the dynamic changes of the sitting posture during driving, performs feature comparison and score calculation through the healthy sitting posture database, avoids subjective bias, improves the accuracy and objectivity of the sitting posture health evaluation, and helps to provide accurate sitting posture correction guidance and health risk warning based on the score result of the sitting posture score.

[0057] Figure 3 The sitting posture health evaluation method provided in the embodiments of the present application provides a flowchart Figure 2 In the embodiments, the multi-dimensional sitting posture feature vector includes a visual feature vector and a pressure correlation vector, and the comparison result includes a difference ratio of each feature being compared; the difference ratio is a ratio of a difference between the two features being compared to a reference feature. In the embodiments, the difference ratio is used to determine the score of each feature being compared. Figure 3 The sitting posture health evaluation method is described in detail based on the embodiments, as shown in Figure 4 The method includes:

[0058] S301, trigger the camera deployed in the cab and the pressure sensor embedded in the seat of the driving position to sample through the hardware clock synchronization, and obtain time-aligned sitting posture image and pressure distribution time series data.

[0059] If there is a time offset in the processing of the sitting posture image and the pressure distribution time series data, the accuracy of feature matching will be affected.

[0060] In this step, through the hardware clock synchronization mechanism, the sampling frequency of the camera and the pressure sensor is ensured to be consistent, and time-aligned sitting posture image and pressure distribution time series data are obtained.

[0061] Trigger the camera deployed in the cab and the pressure sensor embedded in the seat of the driving position to sample through the hardware clock synchronization, which eliminates the problem of time offset caused by asynchronous data collection and ensures the matching of feature extraction.

[0062] S302, pre-process the sitting posture image, extract the body key point features of the driver in the pre-processed sitting posture image, and generate a visual feature vector based on the extracted body key point features.

[0063] Among them, the pre-processing of the sitting posture image is image preprocessing, which can include denoising, brightness adjustment, contrast adjustment, image cropping, etc.

[0064] The body key point feature refers to the position of the skeletal joint or body part representing the human sitting posture state, which can be the head, neck, elbow, ankle, wrist, body center line, etc.

[0065] The visual feature vector refers to the attribute used to quantitatively describe the sitting posture state, such as trunk inclination angle, joint angle, head forward inclination angle, hand position, foot position, etc.

[0066] In this step, after pre-processing the sitting posture image through image processing technology, the pre-processed sitting posture image is processed through image segmentation technology to obtain a segmented image, the body key point features are detected through a deep learning model, the body key point features are associated, and the visual feature vector is extracted from the skeleton diagram.

[0067] Specifically, after pre-processing the sitting posture image through image processing technology, such as denoising, brightness adjustment, contrast adjustment, image cropping, etc., the pre-processed sitting posture image is obtained. Based on the key positions such as head, neck, elbow, ankle, wrist, and body center line, the body key point features of the driver are extracted from the pre-processed sitting posture image. The skeleton diagram is constructed by connecting the body key point features, and the visual feature vector is extracted from the skeleton diagram through a multi-layer convolution kernel, i.e. trunk inclination angle, joint angle, head forward inclination angle, hand position, foot position, etc.

[0068] For example, the extracted visual feature vector can be, for example, trunk inclination angle = 85°, head forward inclination angle = 15°, and the like.

[0069] S303, pre-process the pressure distribution time series data, determine the pressure dynamic feature and the pressure static feature based on the pre-processed pressure distribution time series data, and generate a pressure correlation vector based on the pressure dynamic feature and the pressure static feature.

[0070] The pre-processing of the pressure distribution time series data, i.e. data pre-processing, may, for example, include normalization processing, missing value repair, and abnormal value repair, etc.

[0071] The pressure dynamic feature is used to represent the change of pressure, and the pressure dynamic feature changes with time sequence, including pressure change rate, pressure center point trajectory, etc.

[0072] The pressure static feature refers to the pressure feature at a specific time, including the uniformity of pressure distribution and the position of pressure center point.

[0073] The pressure correlation vector refers to the vector obtained by correlating the pressure dynamic feature and the pressure static feature, which may be, for example, seat pressure uniform distribution, backrest pressure distribution, etc.

[0074] In this step, after pre-processing the pressure distribution time series data, the pre-processed pressure distribution time series data is feature extracted to obtain the pressure dynamic feature and the pressure static feature. The pressure dynamic feature and the pressure static feature are correlated to generate pressure correlation vectors such as seat pressure uniform distribution and backrest pressure distribution.

[0075] For example, during pre-processing, missing values can be repaired by interpolation, using the last valid value, deleting the record, etc. based on the characteristics of the signal itself.

[0076] The visual feature vector is generated from the sitting posture image, and the pressure correlation vector is generated from the pressure distribution time series data. The visual feature vector and the pressure correlation vector are combined to provide a data basis for subsequent sitting posture evaluation, avoid misjudgment due to errors of a single data source, and improve the accuracy of sitting posture health evaluation.

[0077] S304, input the visual feature vector and the pressure correlation vector into the sitting posture evaluation model.

[0078] S305, determine the sitting posture angle of the driver based on the visual feature vector via the sitting posture evaluation model, determine the sitting posture stability based on the visual feature vector and the pressure correlation vector, and determine the pressure distribution feature based on the pressure correlation vector.

[0079] The sitting posture angle includes a seat angle, a back angle, etc. The sitting posture stability refers to the stability of the sitting posture of the driver. The pressure distribution features include uniform seat pressure distribution, back pressure distribution, etc.

[0080] Through the sitting posture evaluation model, the sitting posture angle of the driver, such as the seat angle and the back angle, is determined based on the visual feature vector. The stability of the sitting posture of the driver, i.e., the sitting posture stability, is determined in combination with the visual feature vector and the pressure correlation vector. The pressure distribution features, such as the uniform seat pressure distribution and the back pressure distribution, are determined through the pressure correlation vector.

[0081] S306, compare the sitting posture angle, the sitting posture stability, and the pressure distribution features with the corresponding reference features of each sitting posture sample in the healthy sitting posture database.

[0082] The seat angle, the back angle, the sitting posture stability, the uniform seat pressure distribution, the back pressure distribution, etc. are compared with the corresponding reference features of each sitting posture sample in the healthy sitting posture database to obtain a comparison result. The comparison result includes the difference proportion of each feature being compared. The difference proportion is the proportion of the difference between the two features being compared to the reference feature.

[0083] By comparing the sitting posture angle, the sitting posture stability, and the pressure distribution features with the corresponding reference features of each sitting posture sample in the healthy sitting posture database, comprehensive sitting posture health evaluation is achieved.

[0084] S307, for each feature in the sitting posture angle, the sitting posture stability, and the pressure distribution features, the difference proportion of the feature is mapped to the corresponding score based on the difference proportion of the feature and the score mapping relationship corresponding to the feature.

[0085] For each feature in the sitting posture angle, the sitting posture stability, and the pressure distribution features, the difference proportion of the feature is mapped to the corresponding score based on the difference proportion of the feature and the score mapping relationship corresponding to the feature.

[0086] For example, if the feature is uniform seat pressure distribution, the difference proportion is more than 50%, the score = 0; the difference proportion is between 10% and 50%, the score = 0.5; the difference proportion is less than 10%, the score = 1. When the difference is 30%, the corresponding score of the feature is 0.5.

[0087] S308, output the scores of each feature in the sitting posture angle, the sitting posture stability, and the pressure distribution features.

[0088] The scores of each feature in the sitting posture angle, the sitting posture stability, and the pressure distribution features are taken as the scoring result, and the scoring result is output.

[0089] The score result is determined by the score mapped by the difference proportion of each feature, ensuring the objectivity and standardization of the generated score result.

[0090] In a possible implementation, the score result further includes a health suggestion; and the method further includes:

[0091] The health suggestion is generated based on the number of features whose mapped scores are less than the preset score among the sitting posture angle, the sitting posture stability, and the pressure distribution features; and the health suggestion and the mapped scores of the features among the sitting posture angle, the sitting posture stability, and the pressure distribution features are stored in the sitting posture record database.

[0092] When the mapped score of each feature is less than the preset score, it indicates that the feature does not conform to the healthy sitting posture, and if the driver maintains the unhealthy sitting posture for a long time, various negative health effects will be caused. According to the features whose mapped scores are less than the preset score among the features, the corresponding health suggestion is generated. The health suggestion may be, for example, suggesting to lower the seat height by 2 cm.

[0093] The health suggestion and the mapped scores of the features among the sitting posture angle, the sitting posture stability, and the pressure distribution features are stored in the sitting posture record database. When the driver maintains the same sitting posture for a long time, the health suggestion, for example, suggesting the driver to adjust the sitting posture once every 30 minutes, can also be generated.

[0094] Optionally, the individualized health sitting posture database can also be generated in combination with the individual features of the driver, for example, for the driver with lower height, the ideal range of the seat height is 40-50 cm.

[0095] The health suggestion is generated by the sitting posture health assessment, improving the adaptability of the health suggestion to the sitting posture of the driver, reducing the health risks caused by improper sitting posture, and effectively relieving the physical fatigue in the driving process while improving the driving safety and comfort.

[0096] In a possible implementation, the method further includes: generating and pushing a health report of the driver based on the statistical result of the newly added data in the current statistical period in the sitting posture record database.

[0097] The current statistical period can be the last 7 days, the last 30 days, or the like, and can also be counted monthly.

[0098] The health report can include the basic information of the driver, the summary of the sitting posture assessment in the current statistical period, the summary and frequency analysis of the health suggestion, the improvement suggestion and the subsequent plan, and the like.

[0099] By generating the health report, the cognition of the driver on the health problem is enhanced, thereby promoting the driver to actively perform the sitting posture health management.

[0100] The method for evaluating the health of a sitting posture provided in the embodiments of the present application comprises the following steps: obtaining a sitting posture image of a driver and pressure distribution time series data of a driver's seat; preprocessing the sitting posture image; extracting body key point features of the driver in the preprocessed sitting posture image; generating a visual feature vector based on the extracted body key point features; preprocessing the pressure distribution time series data; determining pressure dynamic features and pressure static features based on the preprocessed pressure distribution time series data; generating a pressure correlation vector based on the pressure dynamic features and the pressure static features; inputting the visual feature vector and the pressure correlation vector into a sitting posture evaluation model; determining a sitting posture angle of the driver based on the visual feature vector via the sitting posture evaluation model; determining a sitting posture stability based on the visual feature vector and the pressure correlation vector; and determining pressure distribution features based on the pressure correlation vector. The difference proportion of each feature in the sitting posture angle, the sitting posture stability and the pressure distribution features, and the score mapping relationship corresponding to the feature are used to map the difference proportion of the feature to a corresponding score, and the scores mapped by each feature in the sitting posture angle, the sitting posture stability and the pressure distribution features are output. The sitting posture image generates a visual feature vector, the pressure distribution time series data generates a pressure correlation vector, and the visual feature vector and the pressure correlation vector are combined to provide a data basis for subsequent sitting posture evaluation, avoid misjudgment caused by errors of a single data source, and improve the accuracy of the sitting posture evaluation. The sitting posture angle, the sitting posture stability and the pressure distribution features are compared with corresponding reference features of each sitting posture sample in a healthy sitting posture database, and comprehensive sitting posture health evaluation is realized. The scores mapped by the difference proportions of each feature are used to determine a score result, and the objectivity and standardization of the generated score result are ensured.

[0101] Figure 4 A structural schematic diagram of a device for evaluating the health of a sitting posture provided in the embodiments of the present application is shown in FIG. 4, and the device for evaluating the health of a sitting posture 40 provided in the embodiments comprises the following components. Figure 5

[0102] A multi-modal data acquisition module 401 is configured to obtain a sitting posture image of a driver and pressure distribution time series data of a driver's seat.

[0103] A sitting posture feature generation module 402 is configured to extract features from the sitting posture image and the pressure distribution time series data, and generate a multi-dimensional sitting posture feature vector.

[0104] A sitting posture evaluation module 403 is configured to input the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, perform feature comparison and score calculation through a healthy sitting posture database, and obtain a score result of the driver.

[0105] ​In a possible implementation, the multi-dimensional sitting posture feature vector includes a visual feature vector and a pressure correlation vector, and the sitting posture feature generation module 402 is further configured to: pre-process the sitting posture image; extract a body key point feature of the driver in the pre-processed sitting posture image, and generate the visual feature vector based on the extracted body key point feature; pre-process the pressure distribution time series data; determine a pressure dynamic feature and a pressure static feature based on the pre-processed pressure distribution time series data, and generate the pressure correlation vector based on the pressure dynamic feature and the pressure static feature; the pressure dynamic feature is used to represent the change of the pressure, and the pressure static feature includes uniformity of the pressure distribution and a position of a pressure center point.

[0106] In a possible implementation, the sitting posture evaluation module 403 is further configured to: input the visual feature vector and the pressure correlation vector into a sitting posture evaluation model; determine, via the sitting posture evaluation model, a sitting posture angle of the driver based on the visual feature vector, determine sitting posture stability based on the visual feature vector and the pressure correlation vector, and determine a pressure distribution feature based on the pressure correlation vector; compare the sitting posture angle, the sitting posture stability, and the pressure distribution feature with corresponding reference features of each sitting posture sample in a healthy sitting posture database; and determine a score result of the driver based on a comparison result and output the score result.

[0107] In a possible implementation, the comparison result includes a difference proportion of each feature being compared, the difference proportion is a proportion of a difference value of two features being compared in the reference feature,

[0108] In a possible implementation, the sitting posture evaluation module 403 is further configured to: for each feature of the sitting posture angle, the sitting posture stability, and the pressure distribution feature, map the difference proportion of the feature to a corresponding score value based on the difference proportion of the feature and a score value mapping relationship corresponding to the feature; and output the score values of each feature of the sitting posture angle, the sitting posture stability, and the pressure distribution feature.

[0109] In a possible implementation, the score result further includes a health suggestion, and the sitting posture evaluation module 403 is further configured to: generate the health suggestion based on a number of features of the sitting posture angle, the sitting posture stability, and the pressure distribution feature whose mapped score values are less than a preset score value; and store the health suggestion and the mapped score values of each feature of the sitting posture angle, the sitting posture stability, and the pressure distribution feature to a sitting posture record database.

[0110] In a possible implementation, the sitting posture evaluation module 403 is further configured to: generate and push a health report of the driver based on a statistical result of newly added data in a current statistical period in the sitting posture record database.

[0111] In a possible implementation, the multi-modal data acquisition module 401 is further configured to trigger, by using a hardware clock, the camera disposed in the cab and the pressure sensor embedded in the seat of the driver's seat to perform sampling, so as to obtain time-aligned sitting posture images and pressure distribution time sequence data.

[0112] The sitting posture health assessment device provided in this embodiment can execute the method provided in the method embodiment, and has similar implementation principles and technical effects. Details are not described herein again.

[0113] Figure 5 A structural schematic diagram of an electronic device is provided in this embodiment. As shown in the figure, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, the memory 502, and the communication component 503 are connected through a bus 504. ​

[0114] In the implementation process, the at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that the at least one processor 501 executes the method described above.

[0115] The specific implementation process of the processor 501 can refer to the method embodiments described above, and has similar implementation principles and technical effects. Details are not described herein again.

[0116] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the application can be directly embodied as execution completed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0117] The memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), for example, at least one disk memory.

[0118] ​The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0119] The present application also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described above.

[0120] The present application also provides a computer readable storage medium, which stores computer execution instructions, and when a processor executes the computer execution instructions, the method described above is implemented.

[0121] The readable storage medium described above can be realized by any type of volatile or non-volatile storage device or their combination, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0122] An exemplary readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium, and can write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0123] The division of units is only a logical functional division, and in actual implementation, there can be another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0124] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0125] In addition, each functional unit in various embodiments of the application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0126] If the function is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the various embodiment methods of the application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0127] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium. The program executes to perform the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, and various media that can store program codes.

[0128] Finally, it should be noted that those skilled in the art, after considering the specification and practicing the application disclosed herein, will easily think of other embodiments of the application. The application is intended to cover any variations, uses, or adaptations of the application that follow the general principles of the application and include common knowledge or conventional technical means in the art that are not disclosed by the application, and is not limited to the precise structure described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the application is only limited by the appended claims.

Claims

1. A sitting posture health assessment method, characterized in that, The method comprises: obtaining a sitting posture image of a driver and time-series data of pressure distribution of a driver seat; extracting features from the sitting posture image and the time-series data of pressure distribution to generate a multi-dimensional sitting posture feature vector; inputting the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, performing feature comparison and score calculation through a healthy sitting posture database to obtain a score result of the driver.

2. The method of claim 1, wherein, The multi-dimensional sitting posture feature vector comprises a visual feature vector and a pressure correlation vector; the feature extraction from the sitting posture image and the time-series data of pressure distribution to generate a multi-dimensional sitting posture feature vector comprises: preprocessing the sitting posture image; extracting body key point features of the driver in the preprocessed sitting posture image, and generating a visual feature vector based on the extracted body key point features; preprocessing the time-series data of pressure distribution; determining pressure dynamic features and pressure static features based on the preprocessed time-series data of pressure distribution, and generating a pressure correlation vector based on the pressure dynamic features and the pressure static features; The pressure dynamic features are used to represent changes in pressure, and the pressure static features comprise uniformity of pressure distribution and position of a pressure center point.

3. The method of claim 2, wherein, The inputting of the multi-dimensional sitting posture feature vector into the pre-trained sitting posture evaluation model, the performing of feature comparison and score calculation through the healthy sitting posture database to obtain the score result of the driver comprises: inputting the visual feature vector and the pressure correlation vector into the sitting posture evaluation model; determining, via the sitting posture evaluation model, a sitting posture angle of the driver based on the visual feature vector, determining sitting posture stability based on the visual feature vector and the pressure correlation vector, and determining pressure distribution features based on the pressure correlation vector; comparing the sitting posture angle, the sitting posture stability and the pressure distribution features with corresponding reference features of each sitting posture sample in the healthy sitting posture database; determining the score result of the driver based on the comparison result and outputting the score result.

4. The method of claim 3, wherein, The comparison result comprises a difference proportion of each feature being compared; the difference proportion is a proportion of a difference value of two features being compared to a reference feature; The determining of the score result of the driver based on the comparison result and the outputting of the score result comprise: for each feature of the sitting posture angle, the sitting posture stability and the pressure distribution features, mapping a difference proportion of the feature to a corresponding score value based on the difference proportion of the feature and a score value mapping relationship corresponding to the feature; outputting the score values of each feature of the sitting posture angle, the sitting posture stability and the pressure distribution features.

5. The method of claim 4, wherein, The score result further comprises a health suggestion; the method further comprises: generating the health suggestion based on a number of features of the score values of each feature of the sitting posture angle, the sitting posture stability and the pressure distribution features being less than a preset score value; storing the health suggestion and the score values of each feature of the sitting posture angle, the sitting posture stability and the pressure distribution features to a sitting posture record database.

6. The method of claim 5, wherein, The method further comprises: Based on the statistical result of the newly added data in the current statistical period in the sitting posture record database, a health report of the driver is generated and pushed.

7. The method according to any one of claims 1 to 6, characterized in that, The sitting posture image of the driver and the pressure distribution time series data of the driver's seat are obtained, including: The camera deployed in the cab and the pressure sensor embedded in the driver's seat are triggered to sample by the hardware clock synchronization, to obtain the time-aligned sitting posture image and pressure distribution time series data.

8. A sitting posture health assessment device, characterized by, Including: A multi-modal data acquisition module is configured to acquire the sitting posture image of the driver and the pressure distribution time series data of the driver's seat. A sitting posture feature generation module is configured to extract features from the sitting posture image and the pressure distribution time series data to generate a multi-dimensional sitting posture feature vector. A sitting posture evaluation module is configured to input the multi-dimensional sitting posture feature vector into a pre-trained sitting posture evaluation model, compare features and calculate scores through a health sitting posture database, and obtain the score result of the driver.

9. An electronic device, comprising: Including: A memory and a processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory, so that the processor executes the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to realize the method of any one of claims 1-7.

11. A computer program product, characterised in that, The computer program is executed by the processor to realize the method of any one of claims 1-7.

Images