Health assessment method and device for insured object, electronic equipment and storage medium
Through video data processing, combined with bone, body shape and appearance assessment, the problem of inaccurate health assessment of insured subjects is solved, and a more accurate and comprehensive health assessment is achieved.
Patent Information
- Application Number
- CN202411217695.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-29
- Publication Date
- 2024-12-17
AI Technical Summary
In the insurance scenario in the financial field, it is difficult for insurance institutions to accurately evaluate the health status of the insured subjects, mainly due to the subjectivity of the information filled in by the insurance applicant and the limited amount of frontal information, resulting in inaccurate health assessment.
By obtaining video data of the target insured subjects, target detection, bone positioning, posture assessment, body obesity assessment and appearance hair assessment were carried out, and comprehensive health assessment was conducted in combination with bone posture score, obesity score and hair score.
It improves the accuracy and comprehensiveness of the health status assessment of insured subjects and provides more objective health assessment results.
Smart Images
Figure CN119153092A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the fields of financial technology and artificial intelligence technology, and in particular to a health assessment method and device, electronic device and storage medium for an insured person. Background Art
[0002] In insurance scenarios in the financial field, it is usually necessary to conduct a health assessment of the insured. For example, when insuring an animal (such as a pet), the health status of the animal needs to be assessed.
[0003] At present, insurance institutions usually evaluate the health status of the insured based on the insured information and the frontal photo of the insured filled in by the insurance applicant. However, the insured information filled in by the insurance applicant is often based on the subjective filling of the insurance applicant, resulting in inaccurate filled information. At the same time, the amount of information in the frontal photo of the insured is limited, resulting in an inability to accurately evaluate the health status of the insured. Therefore, how to improve the accuracy of the assessment of the health status of the insured has become a technical problem that needs to be solved urgently. Summary of the invention
[0004] The main purpose of the embodiments of the present application is to propose a health assessment method and device, electronic device and storage medium for an insured object, aiming to improve the accuracy of the assessment of the health status of animals.
[0005] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application proposes a health assessment method for an insured person, the method comprising:
[0006] Acquire current video data containing a target insured object, wherein the target insured object includes an animal;
[0007] Performing target detection according to the current video data to obtain a target image of the target insured object;
[0008] Performing skeleton positioning on the target image to obtain skeleton key points of the object;
[0009] Based on the current video data and the object's skeletal key points, a skeletal posture assessment is performed on the target insured object to obtain a skeletal posture score;
[0010] Based on the current video data and the object's skeletal key points, assess the body obesity of the target insured person to obtain an obesity score;
[0011] Based on the current video data, performing an appearance hair assessment on the target insured person to obtain a hair score;
[0012] A health assessment of the target insured person is performed based on the skeletal posture score, the obesity score and the appearance hair score.
[0013] In some embodiments, the skeletal posture assessment of the target insured object based on the current video data and the object's skeletal key points to obtain a skeletal posture score includes:
[0014] Determining the actual limb proportions of the target insured person based on the object's skeletal key points;
[0015] Based on the actual limb proportions and the preset target limb proportions, a limb proportion assessment is performed on the target insured person to obtain a limb proportion score;
[0016] Determining the elbow angle of the target insured subject based on the skeleton key points of the subject;
[0017] Based on the elbow joint angle, evaluating the elbow joint angle of the target insured person to obtain an elbow joint angle score;
[0018] Based on the current video data and the object's skeletal key points, a gait assessment is performed on the target insured person to obtain a gait score;
[0019] The limb proportion score, the elbow angle score and the gait score are combined to obtain the skeletal posture score.
[0020] In some embodiments, the step of evaluating the body obesity of the target insured subject based on the current video data and the object's skeletal key points to obtain an obesity score includes:
[0021] Get the body shape training data set;
[0022] Based on the body shape training data set, a preset body shape assessment model is trained to obtain a body shape obesity assessment model;
[0023] Based on the current video data and the object skeleton key points, the target insured object is processed for contour clarity to obtain the object posture key points;
[0024] Determining the target body shape characteristics of the target insured object based on the object posture key points;
[0025] The target body shape feature is input into the body shape obesity assessment model to perform obesity assessment to obtain the obesity score.
[0026] In some embodiments, the performing of appearance hair assessment on the target insured person based on the current video data to obtain an appearance hair score includes:
[0027] Get the hair training dataset;
[0028] Based on the hair training data set, a preset appearance assessment model is trained to obtain an appearance hair assessment model;
[0029] Based on the current video data, feature extraction is performed on the target insured person to obtain hair features;
[0030] The hair features are input into the appearance hair evaluation model to perform appearance evaluation to obtain the appearance hair score.
[0031] In some embodiments, performing target detection according to the current video data to obtain a target image of the target insured object includes:
[0032] Performing video segmentation on the current video data to obtain a video image frame sequence;
[0033] Performing image preprocessing on the video image frame sequence to obtain a target image frame sequence;
[0034] Performing target detection on the target image frame sequence to obtain a detected object boundary box and a detected object confidence;
[0035] Filtering out a target bounding box with the highest confidence level of the detected object from the detected object bounding box;
[0036] Based on the target bounding box, the target image is determined.
[0037] In some embodiments, performing image preprocessing on the video image frame sequence to obtain a target image frame sequence includes:
[0038] Performing a size transformation on the video image frame sequence to obtain a first image frame sequence, wherein the first image frame sequence includes image pixels;
[0039] Normalizing the image pixels to obtain normalized pixels, and converting the first image frame sequence into a second image frame sequence based on the normalized pixels;
[0040] Performing data enhancement on the second image frame sequence to obtain a third image frame sequence;
[0041] The third image frame sequence is format converted to obtain the target image frame sequence.
[0042] In some embodiments, the health assessment of the target insured person based on the skeletal posture score, the obesity score and the appearance hair score includes:
[0043] Determining a skeletal target score of the target insured subject based on a preset first weight and the skeletal posture score;
[0044] Determining a target obesity score for the target insured person based on a preset second weight and the obesity score;
[0045] Determining a target appearance score of the target insured person based on a preset third weight and the appearance hair score;
[0046] The bone target score, the obesity target score and the appearance target score are combined to obtain a health score.
[0047] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a health assessment device for an insured person, the device comprising:
[0048] A video acquisition module, used to acquire current video data containing a target insured object, wherein the target insured object includes an animal;
[0049] A target detection module, used to perform target detection based on the current video data to obtain a target image of the target insured object;
[0050] A key point extraction module is used to perform skeleton positioning on the target image to obtain the object skeleton key points;
[0051] A skeleton assessment module, used to perform skeleton posture assessment on the target insured object based on the skeleton key points of the object to obtain a skeleton posture score;
[0052] An obesity assessment module, used to assess the body obesity of the target insured person based on the current video data and the object's skeletal key points to obtain an obesity score;
[0053] An appearance assessment module, configured to perform an appearance hair assessment on the target insured person based on the current video data to obtain an appearance hair score;
[0054] A health status assessment module is used to perform a health assessment on the target insured person based on the skeletal posture score, the obesity score and the appearance hair score.
[0055] To achieve the above objectives, a third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect is implemented.
[0056] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes 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, the method described in the first aspect is implemented.
[0057] The health assessment method and device, electronic device and storage medium of the insured object proposed in the present application can improve the comprehensiveness of observation of the target insured object by acquiring the current video data containing the target insured object. Furthermore, based on the object skeletal key points of the target insured object in the current video data, the skeletal posture of the target insured object is evaluated to obtain a skeletal posture score, which can achieve the assessment of the skeletal health status of the target insured object. Based on the current video data and the object skeletal key points, the body obesity of the target insured object is evaluated to obtain an obesity score, which can achieve the assessment of the target insured object's health status in terms of body shape. Based on the current video data, the appearance hair of the target insured object is evaluated to obtain a hair score, which can achieve the assessment of the target insured object's health status in terms of appearance. Finally, based on the skeletal posture score, the obesity score and the appearance hair score, the health assessment of the target insured object is performed, which can improve the comprehensiveness of the health assessment of the target insured object, thereby improving the assessment accuracy of the health status of the target insured object. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 It is a flow chart of the health assessment method for the insured subject provided in the embodiment of the present application;
[0059] Figure 2 yes Figure 1 Flow chart of step S102 in FIG.
[0060] Figure 3 yes Figure 2 Flow chart of step S202 in FIG.
[0061] Figure 4 yes Figure 1 Flow chart of step S104 in FIG.
[0062] Figure 5 yes Figure 1 Flow chart of step S105 in FIG.
[0063] Figure 6 yes Figure 1 Flow chart of step S106 in FIG.
[0064] Figure 7 It is a reference image of the annotation standard of the appearance hair in the hair training data set provided in the embodiment of the present application;
[0065] Figure 8 yes Figure 1 Flow chart of step S107 in FIG.
[0066] Fig. 9 It is a structural schematic diagram of a health assessment device for an insured subject provided in an embodiment of the present application;
[0067] Fig.10 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION
[0068] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0069] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.
[0070] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0071] First, some nouns involved in this application are analyzed:
[0072] Skeleton Keypoints: Skeleton Keypoints are important feature points used to represent the posture of the target insured object in computer vision and action recognition. These keypoints correspond to the main joints and parts of the target insured object and can be used to analyze and describe the posture, movement and action of the target insured object.
[0073] In insurance scenarios in the financial field, it is usually necessary to conduct a health assessment of the insured. For example, when insuring an animal (such as a pet), the health status of the animal needs to be assessed.
[0074] At present, insurance institutions usually evaluate the health status of the insured based on the insured information and the frontal photo of the insured filled in by the insurance applicant. However, the insured information filled in by the insurance applicant is often based on the subjective filling of the insurance applicant, resulting in inaccurate filled information. At the same time, the amount of information in the frontal photo of the insured is limited, resulting in an inability to accurately evaluate the health status of the insured. Therefore, how to improve the accuracy of the assessment of the health status of the insured has become a technical problem that needs to be solved urgently.
[0075] Based on this, the embodiments of the present application provide a health assessment method and device for an insured subject, an electronic device and a storage medium, aiming to improve the accuracy of the assessment of the health status of the insured subject.
[0076] The health assessment method and device, electronic device and storage medium for the insured provided in the embodiments of the present application are specifically described through the following embodiments. First, the health assessment method for the insured in the embodiments of the present application is described.
[0077] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.
[0078] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.
[0079] The health assessment method for the insured provided in the embodiment of the present application relates to the fields of financial technology and artificial intelligence technology. The health assessment method for the insured provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or it can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured to provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms and other basic cloud computing services. Cloud servers; software can be applications that implement the health assessment method for the insured, etc., but are not limited to the above forms.
[0080] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.
[0081] Figure 1 is an optional flow chart of a health assessment method for an insured subject provided in an embodiment of the present application, the method is applied to a server, Figure 1 The method may include but is not limited to steps S101 to S107.
[0082] Step S101, obtaining current video data containing a target insured object, wherein the target insured object includes an animal;
[0083] Step S102, performing target detection according to the current video data to obtain a target image of the target insured object;
[0084] Step S103, performing skeleton positioning on the target image to obtain the object skeleton key points;
[0085] Step S104, based on the current video data and the object's skeleton key points, a skeleton posture assessment is performed on the target insured object to obtain a skeleton posture score;
[0086] Step S105, based on the current video data and the object's skeletal key points, assess the body obesity of the target insured person to obtain an obesity score;
[0087] Step S106, based on the current video data, performing an appearance hair assessment on the target insured person to obtain a hair score;
[0088] Step S107, performing a health assessment on the target insured person based on the skeletal posture score, obesity score and appearance hair score.
[0089] In steps S101 to S107 shown in the embodiment of the present application, first, target detection is performed on the acquired current video data containing the target insured object to obtain a target image containing the target insured object, and then the skeleton of the target image is located to obtain the object skeleton key points of the target insured object. Secondly, based on the object skeleton key points, the skeleton posture of the above-mentioned target insured object is evaluated to obtain a skeleton posture score. Based on the above-mentioned current video data and the object skeleton key points, the body obesity of the above-mentioned target insured object is evaluated to obtain an obesity score. Based on the above-mentioned current video data, the appearance hair of the above-mentioned target insured object is evaluated to obtain a hair score. Finally, based on the skeleton posture score, obesity score and appearance hair score of the target insured object, the health status of the target insured object is evaluated. Therefore, the present application can improve the comprehensiveness of observation of the target insured object by acquiring the current video data containing the target insured object. Furthermore, based on the object skeletal key points of the target insured object in the current video data, the skeletal posture of the target insured object is evaluated to obtain a skeletal posture score, which can achieve an assessment of the skeletal health status of the target insured object. Based on the current video data and the object skeletal key points, the body obesity of the target insured object is evaluated to obtain an obesity score, which can achieve an assessment of the target insured object's body health status. Based on the current video data, the target insured object's appearance hair is evaluated to obtain a hair score, which can achieve an assessment of the target insured object's appearance health status. Finally, based on the skeletal posture score, the obesity score and the appearance hair score, the target insured object is evaluated for health, which can improve the comprehensiveness of the health assessment of the target insured object, thereby improving the accuracy of the assessment of the target insured object's health status.
[0090] In step S101 of some embodiments, the target insured object refers to the object protected by the insurance, and the target insured object may be an animal, wherein animals include pets, livestock, etc. The current video data may be the current dynamic video and the current static video of the target insured object, wherein the current dynamic video may include a video of the object walking, a video of the object running, etc., and the current static video may include a video of the object standing still, a video of the object sleeping, etc. The current video data of the target insured object is obtained by shooting the target insured object in a walking state, a running state, and a static state. Compared with image data, continuous current video data can better reflect the health status of the target insured object. Therefore, obtaining the current video data of the target insured object can improve the accuracy of the health assessment of the insured object.
[0091] In one application scenario, when the target insured object is a pet, the pet owner can record the pet's behavior, and then crop the pet's walking video, running video, still video, etc. from the recorded video to obtain the pet's current video data.
[0092] In step S102 of some embodiments, the target image refers to an image frame of the current video data containing the target insured object. When the current video data of the target insured object is obtained, the target insured object of the consecutive image frames in the current video data can be detected and extracted using a pre-built target detection model, and the target object containing the target insured object can be obtained.
[0093] It should be noted that the algorithms used by the above-mentioned pre-built target detection model may include a target detection algorithm, a tracking algorithm, etc., wherein the target detection algorithm may adopt a YOLOv8 (You On ly Look Once) algorithm, and the tracking algorithm may adopt a SORT (Simple On line and Real time Tracking) algorithm.
[0094] See also Figure 2 In some embodiments, step S102 may include but is not limited to steps S201 to S205:
[0095] Step S201, performing video segmentation on the current video data to obtain a video image frame sequence;
[0096] Step S202, performing image preprocessing on the video image frame sequence to obtain a target image frame sequence;
[0097] Step S203, performing target detection on the target image frame sequence to obtain a detection object boundary box and a detection object confidence;
[0098] Step S204, selecting a target bounding box with the highest confidence level from the detected object bounding boxes;
[0099] Step S205: determining the target image based on the target bounding box.
[0100] In step S201 of some embodiments, by performing video segmentation processing on the current video data, video image frames for each time step in the current video data can be obtained, and then the video image frames are arranged and combined according to the time step of each video image frame to obtain a video image frame sequence.
[0101] In step S202 of some embodiments, after obtaining the video image frame sequence, in order to ensure that target detection can be performed on each video image frame in the video image frame sequence, it is necessary to perform image preprocessing on the video image frame sequence before inputting the video image frame sequence into a pre-built target detection model to ensure that each target image frame in the target image frame sequence can meet the input conditions of the target detection model.
[0102] For details, see Figure 3 In some embodiments, step S202 may include but is not limited to steps S301 to S304:
[0103] Step S301, resizing a video image frame sequence to obtain a first image frame sequence, wherein the first image frame sequence includes image pixels;
[0104] Step S302, normalizing the image pixels to obtain normalized pixels, and converting the first image frame sequence into a second image frame sequence based on the normalized pixels;
[0105] Step S303, performing data enhancement on the second image frame sequence to obtain a third image frame sequence;
[0106] Step S304: convert the format of the third image frame sequence to obtain a target image frame sequence.
[0107] In step S301 of some embodiments, a first image frame sequence is obtained by resizing each video image frame in the video image frame sequence, so that the size of each image frame in the first image frame sequence can meet the input size of the pre-built target detection model. It should be noted that the first image frame sequence is image data, and therefore the first image frame sequence includes image pixels.
[0108] In detail, the present application can adjust the size of each video image frame in the video image frame sequence to the input size of the pre-built target detection model through a preset size transformation function.
[0109] In step S302 of some embodiments, the first image frame sequence is converted into a second image frame sequence by normalizing the image pixels in the first image frame sequence, thereby reducing the computational difficulty of the pre-built target detection model.
[0110] In detail, the present application may use a normalization formula to normalize the image pixels in the first image frame sequence to obtain normalized pixels, wherein the normalization formula is expressed as follows:
[0111]
[0112] Wherein, (x, y) represents the position of the image pixel in the first image frame sequence, p′(x, y) represents the pixel value of the normalized pixel, p(x, y) represents the pixel value of the image pixel, min(p) represents the minimum pixel value among the image pixels, and max(p) represents the maximum pixel value among the image pixels.
[0113] Furthermore, the present application may also use a normalization formula to normalize the image pixels in the first image frame sequence to obtain normalized pixels, wherein the normalization formula is expressed as follows:
[0114]
[0115] Where (x, y) represents the position of the image pixel in the first image frame sequence, p′(x, y) represents the pixel value of the normalized pixel, p(x, y) represents the pixel value of the image pixel, μ p Represents the average value of the pixel value of the image, σ p Indicates the standard deviation of the pixel values of the image.
[0116] In step S303 of some embodiments, in order to improve the generalization ability and robustness of the pre-built target detection model, data enhancement may be performed on the image frames in the second image frame sequence to obtain a third image frame sequence.
[0117] In detail, the present application can achieve data enhancement of image frames in the second image frame sequence by flipping, rotating, cropping, and other operations on the image frames in the second image frame sequence, wherein the number of times the data of the image frames in the second image frame sequence are enhanced can be one or more times.
[0118] In step S304 of some embodiments, by converting the format of the image frames in the third image frame sequence, it is possible to ensure that the pre-built target detection model can read the third image frame sequence.
[0119] In steps S301 to S304 shown in the present embodiment, by performing resizing, normalization, data enhancement and format conversion on the video image frames, a target image frame sequence capable of performing target detection is obtained, which can ensure that the target image containing the target insured object is detected from the current video data, and the target insured object is accurately positioned, thereby improving the accuracy of the assessment of the health status of the target insured object.
[0120] In step S203 of some embodiments, a pre-built target detection model is used to detect the target insured object in the target image frame sequence to obtain a detection object bounding box and a detection object confidence of the detection object bounding box, so that image frames that may contain the target insured object can be preliminarily screened out from the target image frame sequence.
[0121] In steps S204 and S205 of some embodiments, the detection object confidence is used to indicate the credibility that the object in the detection object bounding box is the target insured object. Therefore, when the detection object confidence of each detection object bounding box is obtained, the target bounding box with the largest detection object confidence is filtered out from the detection object bounding box, and the image in the target bounding box is used as the target image.
[0122] In steps S201 to S205 shown in the present embodiment, video segmentation is performed on the current video data to obtain a video image frame sequence, and then image preprocessing is performed on the video image frame sequence to obtain a target image frame sequence. Furthermore, target detection is performed on the target image frame sequence to determine the target image, so that continuous target images of the target insured object within a period of time can be obtained, thereby facilitating the assessment of the health status of the target insured object and improving the accuracy of the health assessment of the target insured object.
[0123] In step S103 of some embodiments, after obtaining the target image of the target insured object, the skeleton of the target insured object in each target image is searched and located, and the position information of the skeleton of the target insured object can be obtained. Based on the position information, the object skeleton key points are generated.
[0124] In detail, the present application can use the trained skeletal key point positioning model to perform skeletal positioning on the target image to obtain the skeletal key points of the object. Therefore, before this, the key points of the pre-stored current video data of the insured object can be marked to obtain a training data set for skeletal positioning, wherein the key points marked in the current video data of the insured object can be the head, torso, limbs, joints and other parts. Furthermore, after obtaining the training data set for skeletal positioning, the pre-constructed skeletal key point positioning model is trained using the training data set to obtain a trained skeletal key point positioning model, wherein the algorithm used by the skeletal key point positioning model is a posture estimation algorithm.
[0125] In step S104 of some embodiments, after obtaining the object skeleton key points of the target insured object, the skeleton posture of the target insured object can be evaluated based on the object skeleton key points, thereby obtaining a skeleton posture score of the target insured object.
[0126] See also Figure 4 In some embodiments, step S104 may include but is not limited to steps S401 to S406:
[0127] Step S401, determining the actual limb proportions of the target insured person based on the object's skeleton key points;
[0128] Step S402, based on the actual limb proportions and the preset target limb proportions, the limb proportions of the target insured person are evaluated to obtain a limb proportion score;
[0129] Step S403, determining the elbow joint angle of the target insured person based on the object skeleton key points;
[0130] Step S404, based on the elbow joint angle, evaluating the elbow joint angle of the target insured person to obtain an elbow joint angle score;
[0131] Step S405, based on the current video data and the object's skeleton key points, a gait assessment is performed on the target insured person to obtain a gait score;
[0132] Step S406, combining the limb proportion score, elbow angle score and gait score to obtain a skeletal posture score.
[0133] In step S401 and step S402 of some embodiments, the actual limb proportions of the target insured object can be obtained by comparing the distances between the key points of the bones of the target insured object between the forelimbs and hind limbs, and then the optimal limb proportions of the target insured object can be obtained through information query to obtain the target limb proportions. Furthermore, the actual limb proportions are compared and evaluated with the target limb proportions to obtain a limb proportion score for the target insured object.
[0134] In detail, the present application can calculate the limb proportion score of the target insured person by the following formula:
[0135] Score proportion =1-|Ratio front / back -Ratio ideal |
[0136] Among them, Score proportion Indicates the limb ratio score of the target insured person, Ratio front / back Indicates the actual limb ratio of the target insured person, Ratio ideal Indicates the ideal limb proportions of the target insured person.
[0137] In step S403 and step S404 of some embodiments, the elbow joint angle of the target insured subject can be obtained by calculating the angle formed by the object bone key points at the elbow joint when the target insured subject is standing, and then the elbow joint angle of the target insured subject is evaluated to obtain the elbow joint angle score of the target insured subject.
[0138] In detail, the present application can evaluate the elbow angle of the target insured person according to the pre-acquired elbow angle scoring rule to obtain the elbow angle score, wherein the elbow angle scoring rule can be a mapping function between the elbow angle and the elbow angle score. Further, the present application can calculate the elbow angle score of the target insured person using the following formula:
[0139] Score angle =f(θ elbow )
[0140] Among them, Score angle represents the elbow angle score of the target insured person, f(θ elbow ) represents the mapping function between elbow angle and elbow angle score.
[0141] It should be noted that, generally speaking, the higher the elbow angle score of the target insured subject, the closer the elbow angle of the target insured subject is to the normal elbow angle, and the lower the elbow angle score of the target insured subject, the more the elbow angle of the target insured subject deviates from the normal elbow angle.
[0142] In step S405 of some embodiments, based on the current dynamic video in the current video data, motion analysis is performed on the target insured object to obtain a skeletal motion change pattern of the object skeletal key points of the target insured object over time, and then based on the skeletal motion change pattern, the motion symmetry and motion coherence of the target insured object are evaluated using a three-dimensional convolutional neural network and a long short-term memory network to obtain a motion symmetry score and a motion coherence score, and then the gait periodicity of the target insured object is evaluated using Fourier transform and a convolutional neural network architecture for processing time series data to obtain a gait periodicity score. Based on the motion symmetry score, the motion coherence score and the gait periodicity score, the gait score of the target insured object can be obtained. It should be noted that the better the motion symmetry of the target insured object, the higher the motion symmetry score, the higher the motion coherence, the higher the motion coherence score, the more regular the gait periodicity, and the higher the gait periodicity score.
[0143] In detail, the present application may use the following formula to calculate the gait score of the target insured person:
[0144] Score motion =g(H 3D-CNN+LSTM )+h(H FT+TCN )
[0145] Among them, Score motion represents the gait score of the target insured person, H 3D-CNN+LSTMrepresents the motion symmetry index and motion coherence index of the target insured object output by the above three-dimensional convolutional neural network and long short-term memory network, g() represents the function of converting the motion symmetry index and motion coherence index into the motion symmetry score and motion coherence score, H FT+TCN represents the gait periodicity index of the target insured person output by the above Fourier transform and convolutional neural network architecture for processing time series data, and h() represents a function for converting the gait periodicity index into a gait periodicity score.
[0146] In step S406 of some embodiments, the score of the target insured person at the skeletal level, i.e., the skeletal posture score, is obtained by performing addition and combination operations on the limb proportion score, the elbow angle score, and the gait score.
[0147] Specifically, the skeletal posture score of the target insured person can be calculated using the following formula:
[0148] Score skeletal =Score proportion +Score angle +Score motion
[0149] Among them, Score skeletal Indicates the skeleton posture score of the target insured person, Score proportion Indicates the limb proportion score of the target insured person, Score angle Indicates the elbow angle score of the target insured person, Score motion Indicates the gait score of the target insured person.
[0150] In steps S401 to S404 shown in the present embodiment, the limb proportion score, elbow angle score and gait score of the target insured subject are calculated respectively according to the current video data of the target insured subject and the object's skeletal key points, and then the above-mentioned limb proportion score, elbow angle score and gait score are added and combined to obtain a skeletal posture score, which can improve the comprehensiveness of the assessment of the target insured subject at the skeletal development level, thereby improving the assessment accuracy of the skeletal development assessment of the target insured subject.
[0151] In step S105 of some embodiments, the body obesity of the target insured object is evaluated based on the skeletal key points of the object to obtain an obesity score for the target insured object, which can realize the evaluation of body obesity in the health assessment dimension, increase the diversity of the health assessment of the target insured object, and thus improve the accuracy of the health assessment of the target insured object.
[0152] See also Figure 5In some embodiments, step S105 may include but is not limited to steps S501 to S505:
[0153] Step S501, obtaining a body shape training data set;
[0154] Step S502, training a preset body shape assessment model based on the body shape training data set to obtain a body shape obesity assessment model;
[0155] Step S503, based on the current video data and the object skeleton key points, the target insured object is processed for contour clarity to obtain the object posture key points;
[0156] Step S504, determining the target body shape features of the target insured object based on the object posture key points;
[0157] Step S505: input the target body shape features into a body shape obesity assessment model to perform obesity assessment and obtain an obesity score.
[0158] In step S501 and step S502 of some embodiments, the body shape training data set may be photo data of the insured person with body shape parameters annotated. The pre-constructed body shape assessment model is trained using the body shape training data set to obtain a body shape obesity assessment model. Specifically, the body shape training data set may be segmented to obtain a body shape test data set and a body shape verification data set. The pre-constructed body shape assessment model is trained using the body shape test data set, and then the body shape verification data is used to verify the body shape assessment model trained with the body shape test data set to obtain a body shape obesity assessment model.
[0159] In step S503 of some embodiments, after obtaining the object skeleton key points, an edge detection algorithm is used to perform contour sharpening processing on the target insured object in the current video data to obtain the object posture key points of the target insured object, wherein the edge detection algorithm can be a Canny edge detection algorithm (Canny Edge Detection, Kenny edge detection algorithm).
[0160] In step S504 of some embodiments, the body parameters of the target insured object, such as body length, bones, waist circumference, etc., are calculated based on the above-mentioned object posture key points, so as to obtain the target body characteristics of the target insured object.
[0161] In step S505 of some embodiments, after the target body shape characteristics of the target insured subject are obtained, the body shape characteristics are input into the trained body obesity assessment model for calculation, so as to obtain the score of the target insured subject in terms of body obesity, i.e., the obesity score. It should be noted that the body shape training data set can be based on the veterinary BCS (Body Condition Scoring) to annotate body shape parameters, and therefore, the obesity score output by the body obesity assessment model is related to the BCS, wherein 1 point represents the thinnest, 9 points represents the fattest, and 5 points represents the ideal obesity score.
[0162] In steps S501 to S505 illustrated in the present embodiment, a preset body shape assessment model is trained using a body shape training data set to obtain a body shape obesity assessment model, and then the target body shape features of the target insured subject extracted from the current video data are input into the above-mentioned body shape obesity assessment model to obtain an obesity score for the target insured subject, which can clarify the health assessment results of the target insured subject in terms of body shape obesity, increase the diversity of the health assessment of the target insured subject, and thus improve the accuracy of the health assessment of the target insured subject.
[0163] In step S106 of some embodiments, the appearance hair of the target insured subject is evaluated based on the current video data of the target insured subject to obtain a hair score of the target insured subject, thereby enabling evaluation of the appearance hair in the health assessment dimension.
[0164] See also Figure 6 In some embodiments, step S106 may include but is not limited to steps S601 to S604:
[0165] Step S601, obtaining a hair training data set;
[0166] Step S602, training a preset appearance evaluation model based on the hair training data set to obtain an appearance hair evaluation model;
[0167] Step S603, based on the current video data, feature extraction is performed on the target insured person to obtain hair features;
[0168] Step S604: input the hair features into the appearance hair evaluation model to perform appearance evaluation to obtain an appearance hair score.
[0169] In step S601 and step S602 of some embodiments, the hair training data set may be photo data of the insured person with appearance hair parameters annotated, wherein the appearance hair parameters include hair gloss, hair density, hair elasticity, hair cleanliness, skin condition, etc. Specifically, the method of obtaining the hair training data set and then training the preset appearance evaluation model according to the hair training data set to obtain the appearance hair evaluation model is similar to the method of obtaining the body shape training data set and then training the preset body shape evaluation model according to the body shape training data set to obtain the body shape obesity evaluation model, so it is not repeated here.
[0170] Specifically, the labeling standards for the appearance hair parameters of the target insured person can refer to Figure 7 , Figure 7 It is an optional reference diagram of the labeling standard of the appearance hair parameters of the target insured object provided in the embodiment of the present application, which can provide labeling criteria for the appearance hair parameters, thereby improving the consistency of the evaluation of the appearance hair of the target insured object.
[0171] In step S603 of some embodiments, by detecting the hair texture, hair color and hair sparse area of the target insured person in the current video data, the data such as the hair gloss, hair density, hair elasticity, hair cleanliness, skin condition, etc. of the target insured person can be obtained, and the above data can be combined to obtain hair characteristics.
[0172] In detail, the present application can use the gray-level co-occurrence matrix to detect the texture features of the hair of the target insured object in the current video data to obtain a hair texture evaluation result, detect the hair color of the target insured object in the current video data to obtain a hair color evaluation result, and then perform edge detection on the sparse hair area of the target insured object in the current video data to obtain a hair sparseness evaluation result. By combining the above-mentioned hair texture evaluation results, hair color evaluation results and hair sparseness evaluation results, the hair features of the target insured object can be obtained.
[0173] In step S604 of some embodiments, the method of inputting hair features into an appearance hair assessment model for appearance assessment to obtain an appearance hair score is similar to the method of inputting target body shape features into a body shape obesity assessment model for obesity assessment to obtain an obesity score, so it is not repeated here.
[0174] In steps S601 to S604 shown in the present embodiment, a preset appearance assessment model is trained using a hair training data set to obtain an appearance hair assessment model, and then the hair features of the target insured subject extracted from the current video data are input into the above-mentioned appearance hair assessment model to obtain an appearance hair score for the target insured subject, which can clarify the health assessment result of the target insured subject in terms of appearance and hair, increase the diversity of the health assessment of the target insured subject, and thus improve the accuracy of the health assessment of the target insured subject.
[0175] In step S107 of some embodiments, after obtaining the skeletal posture score, obesity score and appearance hair score of the target insured subject, in order to improve the diversity and accuracy of the health assessment of the target insured subject, the health assessment result of the target insured subject can be determined based on the score ratio of the skeletal posture score, obesity score and appearance hair score.
[0176] See also Figure 8 In some embodiments, step S107 may include but is not limited to steps S801 to S804:
[0177] Step S801, determining a skeletal target score of a target insured person based on a preset first weight and a skeletal posture score;
[0178] Step S802, determining the target obesity score of the target insured person based on the preset second weight and the obesity score;
[0179] Step S803, determining the appearance target score of the target insured person based on the preset third weight and the appearance hair score;
[0180] Step S804, combining the bone target score, obesity target score and appearance target score to obtain a health score.
[0181] In steps S801 to S803 of some embodiments, the first weight, the second weight, and the third weight are all pre-set weight parameters for performing health assessment on the target insured person, wherein the sum of the first weight, the second weight, and the third weight is 1. According to the first weight, the actual score of the skeletal posture score in the health assessment of the target insured person, i.e., the skeletal target score, can be determined; according to the second weight, the actual score of the obesity score in the health assessment of the target insured person, i.e., the obesity target score, can be determined; according to the third weight, the actual score of the appearance hair score in the health assessment of the target insured person, i.e., the appearance target score, can be determined.
[0182] It is important to know that in order to unify the standards for the skeletal target score, obesity target score and appearance target score and to avoid incorrect scoring due to a certain score being too high, it is necessary to normalize the skeletal posture score, obesity score and appearance hair score before calculating the skeletal target score, obesity target score and appearance target score of the target insured person. The normalized range of the skeletal posture score, obesity score and appearance hair score can be 0-10.
[0183] In step S804 of some embodiments, after determining the bone target score, obesity target score and appearance target score of the target insured person, the bone target score, obesity target score and appearance target score are added and combined to obtain the health score of the target insured person.
[0184] In steps S801 to S804 shown in the present embodiment, the actual values of the skeletal posture score, obesity score and appearance hair score in the health score are calculated according to the preset first weight, second weight and third weight, and the skeletal target score, obesity target score and appearance target score are obtained. The skeletal target score, obesity target score and appearance target score are then combined to achieve a multi-faceted health assessment of the target insured person, thereby improving the accuracy of the health assessment of the target insured person.
[0185] The present application can improve the comprehensiveness of observation of the target insured object by acquiring current video data containing the target insured object. Furthermore, based on the object skeletal key points of the target insured object in the current video data, a skeletal posture assessment is performed on the target insured object to obtain a skeletal posture score, which can enable assessment of the skeletal health status of the target insured object. Based on the current video data and the object skeletal key points, the body obesity of the target insured object is assessed to obtain an obesity score, which can enable assessment of the body health status of the target insured object. Based on the current video data, the appearance hair of the target insured object is assessed to obtain a hair score, which can enable assessment of the appearance health status of the target insured object. Finally, based on the skeletal posture score, the obesity score and the appearance hair score, a health assessment is performed on the target insured object, which can improve the comprehensiveness of the health assessment of the target insured object, thereby improving the accuracy of the assessment of the health status of the target insured object.
[0186] See also Fig. 9 The embodiment of the present application also provides a health assessment device for an insured person, which can implement the above-mentioned health assessment method for an insured person, and the device includes:
[0187] The video acquisition module 901 is used to acquire current video data containing a target insured object, wherein the target insured object includes an animal;
[0188] The target detection module 902 is used to perform target detection based on the current video data to obtain a target image of the target insured object;
[0189] The key point extraction module 903 is used to locate the skeleton of the target image and obtain the key points of the skeleton of the object;
[0190] The skeleton assessment module 904 is used to assess the skeleton posture of the target insured object based on the skeleton key points of the object to obtain a skeleton posture score;
[0191] The obesity assessment module 905 is used to assess the body obesity of the target insured person based on the current video data and the object's skeletal key points to obtain an obesity score;
[0192] The appearance evaluation module 906 is used to evaluate the appearance and hair of the target insured person based on the current video data to obtain an appearance and hair score;
[0193] The health status assessment module 907 is used to conduct health assessment on the target insured person based on the skeletal posture score, obesity score and appearance hair score.
[0194] The specific implementation of the health assessment device for the insured subject is substantially the same as the specific implementation of the health assessment method for the insured subject described above, and will not be described in detail herein.
[0195] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned health assessment method for the insured object when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.
[0196] See also Fig.10 , Fig.10 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:
[0197] The processor 1001 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;
[0198] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the health assessment method of the insured object in the embodiment of this application;
[0199] Input / output interface 1003, used to implement information input and output;
[0200] The communication interface 1004 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WI FI, Bluetooth, etc.);
[0201] A bus 1005 , which transmits information between various components of the device (e.g., the processor 1001 , the memory 1002 , the input / output interface 1003 , and the communication interface 1004 );
[0202] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .
[0203] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the health assessment method for the insured object.
[0204] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0205] The health assessment method for an insured object, the health assessment device for an insured object, the electronic device and the storage medium provided in the embodiments of the present application can improve the comprehensiveness of observation of the target insured object by acquiring current video data containing the target insured object. Furthermore, based on the object skeletal key points of the target insured object in the current video data, a skeletal posture assessment is performed on the target insured object to obtain a skeletal posture score, which can achieve an assessment of the skeletal health status of the target insured object. Based on the current video data and the object skeletal key points, the body obesity of the target insured object is assessed to obtain an obesity score, which can achieve an assessment of the body health status of the target insured object. Based on the current video data, the appearance hair of the target insured object is assessed to obtain a hair score, which can achieve an assessment of the appearance health status of the target insured object. Finally, based on the skeletal posture score, the obesity score and the appearance hair score, the health assessment of the target insured object is performed, which can improve the comprehensiveness of the health assessment of the target insured object, thereby improving the accuracy of the assessment of the health status of the target insured object.
[0206] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0207] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.
[0208] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0209] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.
[0210] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0211] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0212] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0213] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0214] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0215] If the integrated unit is implemented in the form of a software functional 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 solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.
[0216] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.
Claims
1. A health assessment method for an insured person, characterized in that: The method comprises: Acquire current video data containing a target insured object, wherein the target insured object includes an animal; Performing target detection according to the current video data to obtain a target image of the target insured object; Performing skeleton positioning on the target image to obtain skeleton key points of the object; Based on the current video data and the object's skeletal key points, a skeletal posture assessment is performed on the target insured object to obtain a skeletal posture score; Based on the current video data and the object's skeletal key points, assess the body obesity of the target insured person to obtain an obesity score; Based on the current video data, performing an appearance hair assessment on the target insured person to obtain a hair score; A health assessment of the target insured person is performed based on the skeletal posture score, the obesity score and the appearance hair score.
2. The method according to claim 1, characterized in that The step of performing a skeletal posture assessment on the target insured object based on the current video data and the object's skeletal key points to obtain a skeletal posture score includes: Determining the actual limb proportions of the target insured person based on the object's skeletal key points; Based on the actual limb proportions and the preset target limb proportions, a limb proportion assessment is performed on the target insured person to obtain a limb proportion score; Determining the elbow angle of the target insured subject based on the skeleton key points of the subject; Based on the elbow joint angle, evaluating the elbow joint angle of the target insured person to obtain an elbow joint angle score; Based on the current video data and the object's skeletal key points, a gait assessment is performed on the target insured person to obtain a gait score; The limb proportion score, the elbow angle score and the gait score are combined to obtain the skeletal posture score.
3. The method according to claim 1, characterized in that The step of evaluating the body obesity of the target insured person based on the current video data and the object's skeletal key points to obtain an obesity score includes: Get the body shape training data set; Based on the body shape training data set, a preset body shape assessment model is trained to obtain a body shape obesity assessment model; Based on the current video data and the object skeleton key points, the target insured object is processed for contour clarity to obtain the object posture key points; Determining the target body shape characteristics of the target insured object based on the object posture key points; The target body shape feature is input into the body shape obesity assessment model to perform obesity assessment to obtain the obesity score.
4. The method according to claim 1, characterized in that: The performing of appearance hair assessment on the target insured person based on the current video data to obtain an appearance hair score includes: Get the hair training dataset; Based on the hair training data set, a preset appearance assessment model is trained to obtain an appearance hair assessment model; Based on the current video data, feature extraction is performed on the target insured person to obtain hair features; The hair features are input into the appearance hair evaluation model to perform appearance evaluation to obtain the appearance hair score.
5. The method according to claim 1, characterized in that The performing target detection according to the current video data to obtain the target image of the target insured object includes: Performing video segmentation on the current video data to obtain a video image frame sequence; Performing image preprocessing on the video image frame sequence to obtain a target image frame sequence; Performing target detection on the target image frame sequence to obtain a detected object boundary box and a detected object confidence; Filtering out a target bounding box with the highest confidence level of the detected object from the detected object bounding box; Based on the target bounding box, the target image is determined.
6. The method according to claim 5, characterized in that The performing image preprocessing on the video image frame sequence to obtain a target image frame sequence includes: Performing a size transformation on the video image frame sequence to obtain a first image frame sequence, wherein the first image frame sequence includes image pixels; Normalizing the image pixels to obtain normalized pixels, and converting the first image frame sequence into a second image frame sequence based on the normalized pixels; Performing data enhancement on the second image frame sequence to obtain a third image frame sequence; The third image frame sequence is format converted to obtain the target image frame sequence.
7. The method according to any one of claims 1 to 6, characterized in that: The health assessment of the target insured person based on the skeletal posture score, the obesity score and the appearance hair score includes: Determining a skeletal target score of the target insured subject based on a preset first weight and the skeletal posture score; Determining a target obesity score for the target insured person based on a preset second weight and the obesity score; Determining a target appearance score of the target insured person based on a preset third weight and the appearance hair score; The bone target score, the obesity target score and the appearance target score are combined to obtain a health score.
8. A health assessment device for an insured person, characterized in that: The device comprises: A video acquisition module, used to acquire current video data containing a target insured object, wherein the target insured object includes an animal; A target detection module, used to perform target detection based on the current video data to obtain a target image of the target insured object; A key point extraction module is used to perform skeleton positioning on the target image to obtain the object skeleton key points; A skeleton assessment module, used to perform skeleton posture assessment on the target insured object based on the skeleton key points of the object to obtain a skeleton posture score; An obesity assessment module, used to assess the body obesity of the target insured person based on the current video data and the object's skeletal key points to obtain an obesity score; An appearance assessment module, configured to perform an appearance hair assessment on the target insured person based on the current video data to obtain an appearance hair score; A health status assessment module is used to perform a health assessment on the target insured person based on the skeletal posture score, the obesity score and the appearance hair score.
9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the health assessment method for the insured subject as described in any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the health assessment method for an insured subject according to any one of claims 1 to 7 is implemented.