Method and device for constructing living body abnormal body position auxiliary judgment model and storage medium

Abstract

The embodiment of the invention relates to the technical field of auxiliary judgment, and discloses a method and equipment for constructing an auxiliary judgment model for an abnormal body position of a living body and a storage medium, and the method comprises the following steps: performing data analysis on video data of the living body in different emotions or different disease states to obtain a plurality of three-dimensional skeletons with continuous time; calculating the three-dimensional information of the three-dimensional skeleton to obtain a plurality of dynamic motion parameters, and marking the dynamic motion parameters as corresponding emotion or disease states; a living body behavior feature training set is established; and by taking the kinetic parameters as input and the corresponding emotion or disease state as output, training a living body abnormal body position auxiliary judgment model through the living body behavior feature training set to obtain a single living body abnormal body position auxiliary judgment model corresponding to the current species, so that the single living body abnormal body position auxiliary judgment model corresponding to the current species can be obtained through the analysis process. The emotion or disease state is connected with the behavior characteristics, so that the emotion or disease state of the living body can be analyzed and determined through the abnormal body position.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of auxiliary judgment, and specifically relate to a method, device and storage medium for constructing an auxiliary judgment model for abnormal body positions of living bodies. Background Art

[0002] Humans or animals will involuntarily exhibit some abnormal postures and body position characteristics due to pain, fear, developmental disorders, and some disease states. In animal experimental groups, the existing evaluation of forced body positions and abnormal postures of animals requires experienced experimental personnel to apply different degrees of stimuli and describe according to the responses of animals to nociceptive stimuli, which highly depends on experience and is difficult to quantify. In addition, for some forced body positions caused by specific reasons, such as visceral pain, addiction withdrawal reactions, and abnormal postures caused by congenital scoliosis, there is currently a lack of objective quantitative evaluation methods. Summary of the Invention

[0003] In view of the above problems, the embodiments of the present invention provide a method for constructing an auxiliary judgment model for abnormal body positions of living bodies, which is used to solve the technical problem of the lack of effective quantitative evaluation methods for abnormal body positions of living bodies in the prior art.

[0004] According to one aspect of the embodiments of the present invention, there is provided a method for constructing an auxiliary judgment model for abnormal body positions of living bodies, the method comprising:

[0005] Step S1, for the same species, obtain video data of the living body in different emotions or different disease states for a first preset duration;

[0006] Step S2, perform data analysis on the video data to obtain a plurality of temporally continuous three-dimensional skeletons;

[0007] Step S3, calculate three-dimensional information of the three-dimensional skeletons to obtain a plurality of kinetic motion parameters, and label the kinetic motion parameters with the corresponding emotions or disease states;

[0008] Step S4, repeatedly execute steps S1 - S3 to obtain a plurality of kinetic motion parameters labeled with the corresponding emotions and a plurality of kinetic motion parameters labeled with the corresponding disease states, and establish a training set of living body behavior characteristics based on this;

[0009] Step S5, use the kinetic parameters as inputs and the corresponding emotions or disease states as outputs, and train an auxiliary judgment model for abnormal body positions of living bodies through the training set of living body behavior characteristics to obtain a single-living-body auxiliary judgment model corresponding to the current species.

[0010] In an optional manner, the step of performing data analysis on the video data to obtain a plurality of temporally continuous three-dimensional skeletons further includes:

[0011] Step S21: Conduct behavioral analysis on the video data to obtain multiple live three-dimensional skeletons arranged in chronological order.

[0012] Step S22: Extract pose features from the multiple three-dimensional skeletons and adjust the format to obtain multiple three-dimensional skeletons with unified size and orientation, and the time of the multiple three-dimensional skeletons is continuous.

[0013] In an optional manner, after the step of using the kinetic parameters as input and the corresponding emotion or the corresponding disease state as output to train the live abnormal body position auxiliary judgment model to obtain the single live abnormal body position auxiliary judgment model corresponding to the current species, the method further includes:

[0014] Replace the species type for collecting the video data, repeat steps S1 - S5, and establish a single live abnormal body position auxiliary judgment model corresponding to another species;

[0015] Combine the single live abnormal body position auxiliary judgment models corresponding to different species to obtain a multi-live abnormal body position auxiliary judgment model.

[0016] In an optional manner, after the step of using the kinetic parameters as input and the corresponding emotion or the corresponding disease state as output to train the live abnormal body position auxiliary judgment model through the live behavior feature training set to obtain the single live abnormal body position auxiliary judgment model corresponding to the current species, the method further includes:

[0017] Step S6: Repeat steps S1 - S3 to obtain multiple kinetic motion parameters marked as the corresponding emotion and multiple kinetic motion parameters marked as the corresponding disease state, and establish a live behavior feature verification set therewith. Each set of kinetic motion parameters in the live behavior feature verification set is different from those in the live behavior feature training set;

[0018] Step S7: Verify the trained single live abnormal body position auxiliary judgment model according to the live behavior feature verification set until the loss function meets the preset threshold.

[0019] In an optional manner, the kinetic parameters include posture, the time interval of the last occurrence of the same posture in the current video data, the cumulative occurrence times of the posture, the duration of the posture, the order and frequency of switching between different postures.

[0020] In an optional manner, the parameters of the posture include body length, body height, body bending degree, body shaking amplitude, and body curling degree.

[0021] In an alternative manner, the multi-living-body abnormal posture auxiliary judgment model includes:

[0022] A data preprocessing layer, configured to obtain video data of a living body to be analyzed, obtain real-time kinetic parameters according to the video data, and call the single-living-body abnormal posture auxiliary judgment model corresponding to the species according to the species of the living body to be analyzed; and,

[0023] The single-living-body abnormal posture auxiliary judgment model;

[0024] The single-living-body abnormal posture auxiliary judgment model includes:

[0025] An input layer, configured to obtain the species of the living body to be analyzed and real-time kinetic parameters;

[0026] A hidden layer, configured to input the kinetic parameters into the called single-living-body abnormal posture auxiliary judgment model;

[0027] An output layer, configured to output corresponding emotions or disease states according to the single-living-body abnormal posture auxiliary judgment model.

[0028] In an alternative manner, the emotion includes any one of fear and tension, and the disease state includes any one of mechanical pain, cold and heat pain, chemical pain, visceral pain, migraine, neuropathic pain, addiction withdrawal reaction, congenital scoliosis, and bone disease or muscle development disease.

[0029] According to another aspect of the embodiments of the present invention, there is provided a device for constructing a living body abnormal posture auxiliary judgment model, including:

[0030] A processor, a memory, a communication interface, and a communication bus, where the processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0031] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations of the method for constructing the living body abnormal posture auxiliary judgment model as described above.

[0032] According to still another aspect of the embodiments of the present invention, there is provided a storage medium, where at least one executable instruction is stored in the storage medium, and when the executable instruction runs on a device for constructing a living body abnormal posture auxiliary judgment model / apparatus, the device for constructing a living body abnormal posture auxiliary judgment model / apparatus is caused to execute the operations of the method for constructing the living body abnormal posture auxiliary judgment model as described above.

[0033] The present invention constructs a brand-new correspondence method from video data to a three-dimensional skeleton, analyzes kinetic motion parameters with data characteristics based on the three-dimensional skeleton, and correlates the kinetic motion parameters with the emotional or disease state of a living body. Thus, a training set of living body behavior characteristics is formed with the kinetic parameters as the input and the corresponding emotional or disease state as the output. Then, the living body abnormal posture auxiliary judgment model is trained based on the above-mentioned training set of living body behavior characteristics to obtain a single-living-body abnormal posture auxiliary judgment model corresponding to the current species. Therefore, through the above data analysis and model training process, the emotional or disease state can be connected with the behavior characteristics, and a single-living-body abnormal posture auxiliary judgment model for determining the emotional or disease state of a living body through abnormal posture analysis can be obtained, thereby solving the technical problem in the prior art of lacking an effective quantitative evaluation method for the abnormal posture of a living body.

[0034] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to be able to understand the technical means of the embodiment of the present invention more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of the embodiment of the present invention more obvious and understandable, the specific embodiments of the present invention are specifically described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings are only used to illustrate the embodiments and are not considered as a limitation to the present invention. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0036] Figure 1 FIG. shows a schematic flowchart of the first embodiment of the method for constructing a living body abnormal posture auxiliary judgment model provided by the present invention;

[0037] Figure 2 FIG. shows a schematic diagram of the three-dimensional skeleton reconstruction process in the method for constructing a living body abnormal posture auxiliary judgment model provided by the present invention;

[0038] Figure 2 -a shows a schematic diagram of video data collected from multiple angles in the method for constructing a living body abnormal posture auxiliary judgment model provided by the present invention;

[0039] Figure 2 -b shows a schematic diagram of determining two-dimensional spatial coordinates according to the video data in the method for constructing a living body abnormal posture auxiliary judgment model provided by the present invention;

[0040] Figure 2 -c shows a schematic diagram of the two-dimensional spatial coordinates extracted in the method for constructing a living body abnormal posture auxiliary judgment model provided by the present invention;

[0041] Figure 2-d shows a schematic diagram of the reconstructed three-dimensional skeleton in the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0042] Figure 2 -e shows a three-dimensional schematic diagram of the three-dimensional skeleton in the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0043] Figure 2 -f shows a schematic diagram of the conversion of each posture of a living body in the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0044] Figure 3 shows a schematic flowchart of a second embodiment of the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0045] Figure 4 shows a schematic flowchart of a third embodiment of the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0046] Figure 5 shows a schematic diagram of the structure of an auxiliary judgment model for abnormal body positions of a living body in the method for constructing an auxiliary judgment model for abnormal body positions of a living body provided by the present invention;

[0047] Figure 6 shows a schematic diagram of the structure of an embodiment of a construction device for an auxiliary judgment model for abnormal body positions of a living body provided by the present invention. Detailed implementation manners

[0048] Hereinafter, exemplary embodiments of the present invention will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0049] The following describes the method for judging abnormal body positions in the prior art:

[0050] When humans or animals are in states of fear, pain, and some developmentally abnormal disease states, they will exhibit forced postures or abnormal postures and body postures, such as the forced postures during addiction withdrawal pain, the abnormal body postures in muscle dysplasia, and the abnormal postures and body postures caused by pain. For example, pain is an unpleasant feeling and emotional experience, which can be simply divided into acute pain and chronic pain. Acute pain is usually related to actual or potential tissue damage and can be further divided into mechanical pain (crush injury), thermal pain (scald), and chemical pain (chemical reagent burn); while chronic pain is usually pain that lasts longer than the normal healing time, or persistent pain caused by non-healing, and pain that recurs after remission. However, regardless of the type of pain, according to the location, nature, and severity of the pain, humans or animals will exhibit a certain degree of abnormal postures or body postures to relieve the pain. Especially for patients who cannot speak, including animals, accurately and objectively quantifying the characteristics of postures and body postures has important guiding value for pain management after injury, disease, or surgery.

[0051] However, due to the technical limitations of video information extraction, the current analysis of animal postures is still very rough, and there is basically no mature method to quantitatively analyze abnormal animal body postures.

[0052] What this application provides is a method, device, and storage medium for constructing an auxiliary judgment model for abnormal body positions of living bodies. By analyzing and processing image data and using the processed data for model training, emotions or disease states can be connected with behavioral characteristics, so that an auxiliary judgment model for abnormal body positions of a single living body for determining the emotions or disease states of a living body through abnormal body position analysis can be obtained, thereby solving the technical problem in the prior art that there is a lack of effective quantitative evaluation means for abnormal body positions of living bodies.

[0053] In an implementation scenario, this solution is implemented based on a camera component and a computer component.

[0054] Among them, the camera component is used to obtain video data, and the computer component is used to execute the method for constructing an auxiliary judgment model for abnormal body positions of living bodies.

[0055] Figure 1 The flowchart of the first embodiment of the method for constructing an auxiliary judgment model for abnormal body positions of a living body according to the present invention is shown. The method for constructing an auxiliary judgment model for abnormal body positions of a living body is executed by a device for constructing an auxiliary judgment model for abnormal body positions of a living body. As Figure 1 shown, the method includes the following steps:

[0056] The method for constructing the auxiliary judgment model for abnormal body positions of a single living body includes:

[0057] Step S1: For the same species, obtain video data of the living body in different emotional or different disease states for a first preset duration;

[0058] The above process can be achieved by using multiple high-definition cameras to separately capture the spontaneous behaviors of the same species from multiple angles. At this time, different living organisms have different emotions or different disease states. By changing the acquisition object, the acquisition of different emotions or different disease states can be realized. The duration of the video data collected at this time can be set as needed. For example, the data acquisition time for a single living organism is between 15 minutes and 60 minutes, and multiple-angle acquisitions can be performed. Reference can be made to the four perspectives shown in Figure 2 Figure -a.

[0059] For different species, based on morphological analysis, when each species faces different emotions or different disease states, due to differences in physiological structures, there are slight differences in the characteristics of the external environment they perceive, and their behavioral tendencies will show obvious species limitations. For example, in a pain state, mice tend to curl up more, while humans tend to cover the corresponding pain area more. There are significant differences in their manifestation forms and video data. Therefore, only obtaining video data of the same species in different emotions or different disease states can ensure the consistency of the data.

[0060] In addition, when the species is an animal, the above steps can be carried out by constructing animal experimental models in different states under the premise of conforming to animal ethics, including fear, using visual, olfactory or pain stimuli that animals instinctively fear. Taking mice as an example, for vision, a shadow stimulus approaching from the upper field of view; for olfaction, the smell of cat or fox urine; for pain, foot shock; pain, limb injury: animal fracture model, visceral pain: inflammation model, by injecting pro-inflammatory factors or performing visceral physical injury surgery, addiction withdrawal: giving animals a certain dose of addictive drugs, developmental skeletal and muscle diseases: constructing related gene-deficient animals through gene editing technology, etc. By the above solutions, experimental animals in different emotions or different disease states are constructed, and relevant video data is collected from them.

[0061] When the species is a human, with the consent of the acquisition object, different emotions or different disease states of the current acquisition object can be defined by collecting personal statements and doctor-assisted judgments, and relevant video data is collected from them.

[0062] Optionally, the emotion includes any one of fear and tension, and the disease state includes any one of mechanical pain, cold and heat pain, chemical pain, visceral pain, migraine, neuropathic pain, addiction withdrawal reaction, congenital scoliosis, and skeletal or muscle development diseases.

[0063] Step S2: Analyze the video data to obtain multiple three-dimensional skeletons that are continuous in time;

[0064] Since in a video, a living body is generally in a continuous motion state in the video data, therefore, its three-dimensional skeleton also continues to change over time, that is, one frame of image will obtain the information of a three-dimensional skeleton. There is a temporal sequence relationship among multiple three-dimensional skeletons corresponding to a video data after data analysis.

[0065] Further, referring to Figure 3 As shown, the step of performing data analysis on the video data to obtain multiple temporally continuous three-dimensional skeletons further includes:

[0066] Step S21: Perform ethological analysis on the video data to obtain multiple three-dimensional skeletons of living bodies arranged in chronological order;

[0067] In the above process, methods for multi-site tracking of the animal body such as fine ethological analysis programs can be used, such as Behavior Atlas (a new intelligent animal behavior precise analysis system), Moseq (action sequencing algorithm / behavior analysis algorithm that can recognize three-dimensional mouse body language (referred to as "syllables")), LEAP (behavior analysis model), Deeplabcut (a pose learning tool for pose estimation and motion analysis), EthoVision (an animal movement trajectory tracking system), etc., to automatically track multiple key body sites on the animal. The specific process of the above fine ethological analysis program is: Referring to Figure 2 As shown in -b, first, according to the position information calibrated by the camera, calculate the two-dimensional spatial coordinate information of the key body sites of the animal body. Referring to Figure 2 -c and Figure 2 As shown in -d, subsequently, integrate the spatial coordinate information of the key body sites of multiple images, calculate the coordinate information of the key points in three-dimensional space, and reconstruct the three-dimensional skeleton of the animal.

[0068] Step S22: Extract pose features from multiple three-dimensional skeletons and perform format adjustment to obtain multiple three-dimensional skeletons with unified size and orientation, and the multiple three-dimensional skeletons are temporally continuous.

[0069] Among them, the process of pose feature extraction is to extract the information of the three-dimensional skeleton reconstructed in each frame. For example, extract the three-dimensional spatial coordinates of 16 body key points of the animal in each frame. More body key points can also be set. Centered on the back point, align the three-dimensional skeleton of the animal; rotate the three-dimensional skeleton spatial coordinate information to unify the orientation of the three-dimensional skeleton; scale the skeleton according to the size ratio of the animal in the real three-dimensional space, and scale the mouse skeleton to the same scale size, and then obtain Figure 2 the three-dimensional stereogram shown in -e.

[0070] Step S3: Calculate the three-dimensional information of the three-dimensional skeleton to obtain a plurality of kinetic motion parameters, and label the kinetic motion parameters with corresponding emotions or disease states.

[0071] Among them, calculating the three-dimensional information of the three-dimensional skeleton to obtain a plurality of kinetic motion parameters is for the information of each reconstructed three-dimensional skeleton. Therefore, labeling the corresponding emotions or disease states is also for each frame of video image. The labeling process is carried out frame by frame by experimenters or users according to the process of obtaining experimental animals in different emotions or different disease states, or is carried out frame by frame according to the self-report of the collected individuals and the auxiliary judgment of doctors, and the labels are corresponding to the kinetic motion parameters.

[0072] Optionally, the kinetic parameters include posture, the time interval between the last appearance of the same posture in the current video data, the cumulative appearance times of the posture, the duration of the posture, the order and frequency of switching between different postures.

[0073] Among them, the posture also refers to the action. When counting, the number of times action A appears is a 0 / 1 value, that is, it increases by one when it appears, regardless of the duration. After this action ends, the next action B will appear. But when action A appears again, the cumulative appearance times of action A + 1. Each previous statistic requires a fixed observation time window, such as an hour of observation time between 8 and 12 in the morning.

[0074] Optionally, the parameters of the posture include body length, body height, body bending degree, body shaking amplitude, and body curling degree.

[0075] Reference Figure 2 As shown in -f, the postures include running, walking, rising, right turning, hunching, climb up, grooming, jumping, rearing, pausing, sniffing, left turning, stepping.

[0076] Among the above parameters of the posture, the specific definition of the posture is:

[0077] Body length: In three-dimensional space, the straight-line distance from the nose to the tail of the mouse.

[0078] Body height: The height of the mouse's nose, neck, and back from the ground.

[0079] Degree of body curvature: Calculate the angle formed by three points: "nose - neck - back" and "neck - back - base of the tail". The smaller the degree of the angle, the greater the degree of body curvature of the mouse.

[0080] Amplitude of body tremors: Fix a point on the back of the mouse at the same position, and calculate the displacement of the rest of the mouse's body in three-dimensional space. Then calculate the number and frequency of movement cycles.

[0081] Degree of body curling: Calculate the distance from each body site to the body's center of gravity.

[0082] Time and frequency of the last posture appearance, duration, and frequency of switching between different postures.

[0083] Step S4: Repeat steps S1 - S3 to obtain multiple kinetic motion parameters marked with corresponding emotions and multiple kinetic motion parameters marked with corresponding disease states, and establish a training set of in-vivo behavior characteristics based on this.

[0084] Step S5: Use the kinetic parameters as input and the corresponding emotion or disease state as output, and train the in-vivo abnormal body position auxiliary judgment model through the training set of in-vivo behavior characteristics to obtain the single in-vivo abnormal body position auxiliary judgment model corresponding to the current species.

[0085] In the above solution, by constructing a new correspondence between video data and three-dimensional skeletons, kinetic motion parameters with data characteristics are analyzed based on the three-dimensional skeletons, and the kinetic motion parameters are corresponded to the emotions or disease states of the in-vivo body, thereby constituting a training set of in-vivo behavior characteristics with the kinetic parameters as input and the corresponding emotion or disease state as output. And based on the above training set of in-vivo behavior characteristics, the in-vivo abnormal body position auxiliary judgment model is trained to obtain the single in-vivo abnormal body position auxiliary judgment model corresponding to the current species. Thus, through the above data analysis and model training process, emotions or disease states can be connected with behavior characteristics, and a single in-vivo abnormal body position auxiliary judgment model for determining the emotions or disease states of the in-vivo body through abnormal body position analysis can be obtained, thereby solving the technical problem in the prior art of lacking effective quantitative evaluation means for the abnormal body positions of in-vivo bodies.

[0086] Based on the above solution, the above single in-vivo abnormal body position auxiliary judgment model is not only for scientific research and drug development, nor is it only used in mouse experiments. Any animal protection scenario that can use video recording, including hospitals, pet hospitals, zoos, wildlife protection, etc., can use the algorithm of the present invention to monitor the health status of animals or humans.

[0087] It can be used to assist in identifying the general range of potential abnormal emotions or disease states of animals, providing a data basis for doctors or experimenters to more quickly judge the states of humans or animals. It includes different pain templates such as mechanical pain, cold and heat pain, chemical pain, visceral pain, migraine, and neuropathic pain, as well as spontaneous abnormal postures of animal models including addiction withdrawal reactions, congenital scoliosis, and some bone or muscle development diseases.

[0088] It can also identify the emotional or disease state of animals without using external stimuli, only by observing the characteristics of their spontaneous behavioral postures. It can observe the abnormal postures shown by animals in specific emotions and diseases in a natural state. Without external stimuli causing changes in the animals' emotions, it can better reflect the true spontaneous manifestations of animals' experiences such as fear, addiction withdrawal, and pain in a real state. By digitizing the various posture characteristics of animals through calculation, it can more accurately describe the posture characteristics of animals, providing a quantifiable comparison index for the evaluation of the degree of fear, pain, and addiction withdrawal pain, etc.

[0089] In an optional embodiment, after the step of using the kinetic parameters as input and the corresponding emotion or the corresponding disease state as output to train the auxiliary judgment model for abnormal body positions of living bodies to obtain the auxiliary judgment model for abnormal body positions of a single living body corresponding to the current species, it further includes:

[0090] Replace the species type for collecting the video data, repeat steps S1 - S5, and establish an auxiliary judgment model for abnormal body positions of a single living body corresponding to another species;

[0091] Combine the auxiliary judgment models for abnormal body positions of a single living body corresponding to different species to obtain an auxiliary judgment model for abnormal body positions of multiple living bodies.

[0092] Through the above solution, for each species, a corresponding single-living-body abnormal posture auxiliary judgment model can be trained separately, so that for each species, fine behavior analysis can be carried out. It is to further extract posture features for the actions of unsupervised clustering discrimination. It can not only provide dynamic posture kinetics data, but also more objectively divide actions based on posture features more meaningfully, realizing objective and digital quantitative analysis indicators for each species. Thus, for each species, not only the pain state of animals, skeletal and muscle development diseases, but also any disease state showing abnormal spontaneous behavior characteristics can be extracted and calculated through the solution constructed in this application. Moreover, by training a single-living-body abnormal posture auxiliary judgment model for supervision for each species, such as cats, dogs, pigs, cows, sheep, monkeys, etc., other wild animals and humans, thus by constructing a single-living-body abnormal posture auxiliary judgment model corresponding to the disease models of each species strain, it has important reference value for many fields such as basic scientific research, drug development, and animal health status detection.

[0093] In an alternative embodiment, referring to Figure 4 as shown, after the step of using the kinetic parameters as input and the corresponding emotion or the corresponding disease state as output, and training the single-living-body abnormal posture auxiliary judgment model through the living body behavior feature training set to obtain the single-living-body abnormal posture auxiliary judgment model corresponding to the current species, the following steps are further included:

[0094] Step S6: Repeat steps S1 - S3 to obtain multiple kinetic motion parameters marked as corresponding emotions and multiple kinetic motion parameters marked as corresponding disease states, and establish a living body behavior feature verification set therewith. Each set of kinetic motion parameters of the living body behavior feature verification set is different from that of the living body behavior feature training set;

[0095] Since each set of kinetic motion parameters of the living body behavior feature verification set is different from that of the living body behavior feature training set, the living body behavior feature verification set and the living body behavior feature training set can be made to collect different living bodies by replacing the collected living body objects, so as to achieve the purpose of different kinetic motion parameters.

[0096] Step S7: Verify the trained single-living-body abnormal posture auxiliary judgment model according to the living body behavior feature verification set until the loss function meets the preset threshold.

[0097] Through the above solution, the verification of the single-living-body abnormal posture auxiliary judgment model can be realized, ensuring the accuracy of the auxiliary judgment of the single-living-body abnormal posture auxiliary judgment model during subsequent use.

[0098] Optionally, the multi-living-body abnormal posture auxiliary judgment model includes:

[0099] A data preprocessing layer for obtaining video data of a living body to be analyzed, obtaining real-time kinetic parameters based on the video data, and calling a single-living-body abnormal body position auxiliary judgment model corresponding to the species of the living body to be analyzed; and,

[0100] The single-living-body abnormal body position auxiliary judgment model;

[0101] Reference Figure 5 As shown, the single-living-body abnormal body position auxiliary judgment model includes:

[0102] An input layer for obtaining the species of the living body to be analyzed and real-time kinetic parameters;

[0103] A hidden layer for inputting the kinetic parameters into the called single-living-body abnormal body position auxiliary judgment model;

[0104] An output layer for outputting corresponding emotions or disease states according to the single-living-body abnormal body position auxiliary judgment model.

[0105] Among them, the steps of obtaining video data of the living body to be analyzed and obtaining real-time kinetic parameters based on the video data are implemented with reference to steps S1 - S3. The single-living-body abnormal body position auxiliary judgment model called according to the species of the living body to be analyzed is constructed in the construction method and verified by a living body behavior feature verification set.

[0106] In the above implementation, as Figure 5 shown, the input layer has two dimensions, namely x_axis and y_axis, the hidden layer is 50-dimensional, and the output layer is a 1*4 matrix, thereby constructing a fully connected neural network model with the input being kinetic parameters and the output being emotions or disease states in one-to-one correspondence, thus providing an effective auxiliary judgment model for further analyzing the precise high-dimensional correspondence relationship between animal states and abnormal behavioral phenotypes and postures. It should be noted that the matrices of the input layer and the output layer can be transformed according to needs at this time, and the number of layers and dimensions of the hidden layer can also be transformed according to the actual number of action combinations, so as to obtain a single-living-body behavior logic abnormality auxiliary judgment model applicable to each species.

[0107] Figure 6 Fig. shows a schematic structural diagram of an embodiment of the construction device of the living body abnormal body position auxiliary judgment model of the present invention. The specific implementation of the construction device of the living body abnormal body position auxiliary judgment model is not limited in the specific embodiments of the present invention.

[0108] As Figure 6As shown in the figure, the device for constructing the living body abnormal posture auxiliary judgment model may include: a processor 402, a communications interface 404, a memory 406, and a communication bus 408.

[0109] Among them: The processor 402, the communications interface 404, and the memory 406 communicate with each other through the communication bus 408. The communications interface 404 is used to communicate with network elements of other devices such as clients or other servers. The processor 402 is used to execute the program 410, and specifically can execute the relevant steps in the above embodiments of the method for constructing the living body abnormal posture auxiliary judgment model.

[0110] Specifically, the program 410 may include program code, and the program code includes computer-executable instructions.

[0111] The processor 402 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the device for constructing the living body abnormal posture auxiliary judgment model may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0112] The memory 406 is used to store the program 410. The memory 406 may be a high-speed RAM memory, or may also include a non-volatile memory, such as at least one disk memory.

[0113] The program 410 can specifically be called by the processor 402 to enable the device for constructing the living body abnormal posture auxiliary judgment model to execute the operations of the above method for constructing the living body abnormal posture auxiliary judgment model.

[0114] It should be noted that since the device for constructing the living body abnormal posture auxiliary judgment model of the present application can implement all the embodiments of the method for constructing the living body abnormal posture auxiliary judgment model, the device for constructing the living body abnormal posture auxiliary judgment model of the present application has all the beneficial effects of the method for constructing the living body abnormal posture auxiliary judgment model, and will not be elaborated here.

[0115] An embodiment of the present invention provides a storage medium storing at least one executable instruction. When the executable instruction runs on a device for constructing a living body abnormal posture auxiliary judgment model, it causes the device for constructing the single living body abnormal posture auxiliary judgment model to execute the method for constructing a living body abnormal posture auxiliary judgment model in any of the above method embodiments.

[0116] It should be noted that since the storage medium of the present application can implement all embodiments of the method for constructing a living body abnormal posture auxiliary judgment model, the storage medium of the present application has all the beneficial effects of the method for constructing a living body abnormal posture auxiliary judgment model, which will not be elaborated herein.

[0117] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. In addition, embodiments of the present invention are not directed to any specific programming language.

[0118] In the specification provided herein, a large number of specific details are set forth. However, it can be understood that embodiments of the present invention may be practiced without these specific details. Similarly, in order to streamline the present invention and assist in understanding one or more of the various inventive aspects, in the above description of the exemplary embodiments of the present invention, the various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. Among them, the claims following the specific implementation manner are hereby expressly incorporated into the specific implementation manner, where each claim itself is a separate embodiment of the present invention.

[0119] Those skilled in the art can understand that the modules in the devices in the embodiments can be adaptively changed and disposed in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive.

[0120] It should be noted that the above embodiments illustrate the present invention rather than limit the present invention, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention can be implemented by means of hardware including several different elements and by means of a suitably programmed computer. In a unit claim listing several devices, several of these devices can be embodied by the same item of hardware. The use of the words first, second, and third, etc. does not denote any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A method for constructing an auxiliary judgment model for abnormal body positions of living bodies, characterized in that, The method includes: Step S1: For the same species, obtain video data of a living body in different emotional states or different disease states for a first preset duration; Step S2: Analyze the video data to obtain a plurality of temporally continuous three-dimensional skeletons; Step S3: Calculate the three-dimensional information of the three-dimensional skeletons to obtain a plurality of kinetic motion parameters, and label the kinetic motion parameters with the corresponding emotional states or disease states; Step S4: Repeat steps S1 - S3 to obtain a plurality of kinetic motion parameters labeled with the corresponding emotions and a plurality of kinetic motion parameters labeled with the corresponding disease states, and establish a training set of living body behavior characteristics based on this; Step S5: Use the kinetic parameters as input and the corresponding emotional state or disease state as output, and train the single living body abnormal posture auxiliary judgment model through the training set of living body behavior characteristics to obtain the single living body abnormal posture auxiliary judgment model corresponding to the current species.

2. The method for constructing a living body abnormal posture auxiliary judgment model according to claim 1, wherein The step of analyzing the video data to obtain a plurality of temporally continuous three-dimensional skeletons further includes: Step S21: Conduct ethological analysis on the video data to obtain a plurality of three-dimensional skeletons of the living body arranged in chronological order; Step S22: Extract the pose features of the plurality of three-dimensional skeletons and adjust the format to obtain a plurality of three-dimensional skeletons with unified size and orientation, and the plurality of three-dimensional skeletons are temporally continuous.

3. The method for constructing an in-vivo abnormal body position auxiliary judgment model according to claim 1 or 2, characterized in that, After the step of using the kinetic parameters as input and the corresponding emotional state or disease state as output to train the single living body abnormal posture auxiliary judgment model to obtain the single living body abnormal posture auxiliary judgment model corresponding to the current species, it further includes: Change the species type for collecting the video data, repeat steps S1 - S5, and establish a single living body abnormal posture auxiliary judgment model corresponding to another species; Combine the single living body abnormal posture auxiliary judgment models corresponding to different species to obtain a multi-living body abnormal posture auxiliary judgment model.

4. The method for constructing a living body abnormal body position auxiliary judgment model according to claim 1 or 2, characterized in that, After the step of using the kinetic parameters as input and the corresponding emotional state or disease state as output to train the single living body abnormal posture auxiliary judgment model through the training set of living body behavior characteristics to obtain the single living body abnormal posture auxiliary judgment model corresponding to the current species, it further includes: Step S6: Repeat steps S1 - S3 to obtain a plurality of kinetic motion parameters labeled with the corresponding emotions and a plurality of kinetic motion parameters labeled with the corresponding disease states, and establish a verification set of living body behavior characteristics based on this. Each set of kinetic motion parameters in the verification set of living body behavior characteristics is different from that in the training set of living body behavior characteristics; Step S7: Verify the trained single living body abnormal posture auxiliary judgment model according to the verification set of living body behavior characteristics until the loss function meets the preset threshold.

5. The method for constructing an in-vivo abnormal body position auxiliary judgment model according to claim 1, wherein, The kinetic parameters include posture, the time interval between the last occurrence of the same posture in the current video data, the cumulative occurrence times of the posture, the duration of the posture, the order and frequency of switching between different postures.

6. The method for constructing the living body abnormal body position auxiliary judgment model according to claim 5, characterized in that, The parameters of the posture include body length, body height, body bending degree, body jitter amplitude, and body curling degree.

7. The method for constructing a living body abnormal body position auxiliary judgment model according to claim 2, wherein The multi-living-body abnormal posture auxiliary judgment model includes: A data preprocessing layer, configured to obtain video data of a living body to be analyzed, obtain real-time kinetic parameters according to the video data, and call the single-living-body abnormal posture auxiliary judgment model corresponding to the species according to the species of the living body to be analyzed; and The single-living-body abnormal posture auxiliary judgment model; The single-living-body abnormal posture auxiliary judgment model includes: An input layer, configured to obtain the species of the living body to be analyzed and real-time kinetic parameters; A hidden layer, configured to input the kinetic parameters into the called single-living-body abnormal posture auxiliary judgment model; An output layer, configured to output a corresponding emotion or disease state according to the single-living-body abnormal posture auxiliary judgment model.

8. The method for constructing a living body abnormal posture auxiliary judgment model according to claim 2, wherein, The emotion includes any one of fear and tension, and the disease state includes any one of mechanical pain, cold and heat pain, chemical pain, visceral pain, migraine, neuralgia, addiction withdrawal reaction, congenital scoliosis, and bone disease or muscle development disease.

9. An apparatus for constructing a living abnormal body position auxiliary judgment model, characterized in that, It includes: A processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus; The memory is used to store at least one executable instruction, and the executable instruction causes the processor to execute the operations of the method for constructing the living-body abnormal posture auxiliary judgment model according to any one of claims 1-6.

10. A storage medium, characterized in that, At least one executable instruction is stored in the storage medium. When the executable instruction runs on the device for constructing the living-body abnormal posture auxiliary judgment model, the device for constructing the living-body abnormal posture auxiliary judgment model executes the operations of the method for constructing the living-body abnormal posture auxiliary judgment model according to any one of claims 1-6.

Images