Gait analysis method and apparatus based on animal posture images, and device and medium

Abstract

The present application relates to the technical field of gait analysis based on animal posture images. Disclosed are a gait analysis method and apparatus based on animal posture images, and a device and a medium. The method comprises: acquiring an animal image set captured by a target camera device in a target environment; extracting skeleton information from the animal image set, so as to obtain an animal posture image set; performing gait analysis on the animal posture image set on the basis of the animal posture image set, a standard animal posture image database and a trained deep learning network, so as to obtain gait analysis data, wherein the deep learning network comprises a feature analysis model and a feature comparison model; and performing data processing on the basis of the gait analysis data, so as to obtain an animal state report. By means of the present invention, accurate and efficient gait analysis can be performed on an animal image set captured in a target environment, thereby obtaining an animal state report with a better analysis effect.

Description

Gait analysis method, device, equipment and medium based on animal posture image Technical Field

[0001] The present invention relates to the technical field of gait analysis based on animal posture images, and in particular to a gait analysis method, device, equipment and medium based on animal posture images. Background Art

[0002] Coordinated limb movements require the brain to integrate multiple sensory information and mobilize multiple body systems, including the sensory system, motor system, respiratory system, circulatory system, and metabolic system. Humans and animals exhibit specific gait characteristics when moving. Abnormalities in the function of any of the above systems may cause humans and animals to exhibit a certain degree of motor impairment, leading to gait abnormalities. For example, in diseases such as progressive muscular dystrophy, amyotrophic lateral sclerosis, and Parkinson's disease, gait characteristics will change. Therefore, early screening of subjects and animals with abnormal motor function through gait analysis is of great value in the discovery, diagnosis, and treatment of movement disorders associated with developmental or degenerative brain diseases.

[0003] Summary of the Invention

[0004] Based on this, it is necessary to address the technical problem of how the existing technology determines the state of an animal through gait analysis, and propose a gait analysis method, device, equipment and medium based on animal posture images.

[0005] In a first aspect, a gait analysis method based on an animal posture image is provided, the method comprising:

[0006] Obtaining a set of animal images captured by a target camera device in a target environment;

[0007] Extracting skeleton information from the animal image set to obtain an animal posture image set;

[0008] Performing gait analysis on the animal posture image set according to the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model;

[0009] Data processing is performed based on the gait analysis data to obtain an animal status report.

[0010] In a second aspect, a gait analysis device based on an animal posture image is provided, the device comprising:

[0011] An acquisition module is used to acquire a set of animal images taken by a target camera device in a target environment;

[0012] an extraction module, configured to extract skeleton information from the animal image set to obtain an animal posture image set;

[0013] a gait analysis module, configured to perform gait analysis on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model;

[0014] The data processing module is used to perform data processing based on the gait analysis data to obtain an animal status report.

[0015] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned gait analysis method based on animal posture images when executing the computer program.

[0016] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned gait analysis method based on animal posture images are implemented.

[0017] The present invention proposes a gait analysis method based on animal posture images, which obtains an animal image set taken by a target camera device in a target environment, then extracts skeleton information from the animal image set to obtain an animal posture image set, and then performs gait analysis on the animal posture image set based on the animal posture image set, a standard animal posture image database and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model. Finally, data processing is performed based on the gait analysis data to obtain an animal status report. The method can perform accurate and efficient gait analysis on the animal image set taken in the target environment, thereby obtaining an animal status report with better analysis effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] in:

[0020] FIG1 is a diagram illustrating an application environment of a gait analysis method based on animal posture images in one embodiment;

[0021] FIG2 is a flow chart of a gait analysis method based on animal posture images according to one embodiment;

[0022] FIG3 is a mouse skeleton image of a gait analysis method based on animal posture images in one embodiment;

[0023] FIG4 is a block diagram of a gait analysis device based on animal posture images according to one embodiment;

[0024] FIG5 is a block diagram of a computer device according to an embodiment;

[0025] FIG6 is a structural block diagram of a computer device in another embodiment. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] The gait analysis method based on animal posture images provided in an embodiment of the present invention can be applied in an application environment such as that shown in FIG. 1 , wherein a client 110 communicates with a server 120 via a network. The server 120 can receive a set of animal images captured by a target camera device in a target environment through the client 110, then extract skeleton information from the animal image set to obtain a set of animal posture images. The server 120 then performs gait analysis on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data. The deep learning network includes a feature analysis model and a feature comparison model. Finally, the server 120 processes the gait analysis data to obtain an animal status report. The method can accurately and efficiently perform gait analysis on the set of animal images captured in the target environment, thereby obtaining an animal status report with improved analysis results. The client 110 can be, but is not limited to, various personal computers, laptops, smartphones, tablet computers, and portable wearable devices. The server 120 can be implemented as a standalone server or a server cluster consisting of multiple servers. The present invention is described in detail below through specific embodiments.

[0028] Please refer to FIG2 , which is a flow chart of a gait analysis method based on animal posture images provided by an embodiment of the present invention, including the following steps:

[0029] Step S101: obtaining a set of animal images captured by a target camera device in a target environment;

[0030] The target camera device can be a high-definition camera device, for example, fixed in a target environment. The target environment can be an open field environment, an experimental environment, etc. The animal image set refers to a series of animal images captured by the target camera device in the target environment. The target object can be a pet cat or dog, an experimental mouse, a wild monkey, or other animals.

[0031] Step S102: extracting skeleton information from the animal image set to obtain an animal posture image set;

[0032] For example, for each animal image in the animal image set, generate an animal posture image.

[0033] As shown in FIG3 , the animal in the animal posture image is a skeleton image of a mouse.

[0034] As an example, for each animal image, the 2D spatial coordinates of 16 animal neural key points are extracted for each frame. Using a camera calibration file, the 2D spatial coordinates of the animal neural key points from four perspectives are combined to construct a 3D spatial skeleton of the animal, thereby obtaining an animal posture image, where the body key points include at least the four paws of the animal in the animal image. The corresponding animal posture images of each animal image are aligned and the body orientation is unified. The animal skeletons in the animal posture images are scaled to the same scale. The horizontal plane of the animal skeleton in the animal posture images is corrected so that the animal skeleton in the proposed animal posture image moves on the same horizontal base plane. Each corrected animal posture image is then used as an animal posture image set.

[0035] Step S103: performing gait analysis on the animal posture image set according to the animal posture image set, the standard animal posture image database and the trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model.

[0036] Among them, the feature analysis model is a mathematical model, which is used to analyze the gait characteristics of the animal posture image set. In one implementation, the feature comparison model can be obtained based on convolutional neural network training. In another implementation, the feature comparison model can be obtained based on Transformer model training. The standard animal posture image database is constructed based on a large amount of standard animal posture sample data.

[0037] In this embodiment, a trained deep learning network is used to perform gait analysis on a set of animal posture images based on a standard animal posture image database to obtain gait analysis data.

[0038] Step S104: performing data processing based on the gait analysis data to obtain an animal status report.

[0039] In this embodiment, the precise gait analysis data can be processed to obtain an animal status report with better accuracy and effect.

[0040] The gait analysis method based on animal posture images proposed in this embodiment obtains a set of animal images captured by a target camera device in a target environment, then extracts skeleton information from the animal image set to obtain an animal posture image set. Gait analysis is then performed on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data. The deep learning network includes a feature analysis model and a feature comparison model. Finally, data processing is performed based on the gait analysis data to obtain an animal status report. This method can accurately and efficiently perform gait analysis on a set of animal images captured in a target environment, thereby obtaining an animal status report with improved analysis results.

[0041] Furthermore, in a preferred embodiment, the step of performing gait analysis on the animal posture image set based on the animal posture image set, the standard animal posture image database, and the trained deep learning network to obtain gait analysis data includes:

[0042] Step S201: extracting, from the standard animal posture image database, historical animal posture image sets, abnormal animal posture image sets, and normal animal posture image sets corresponding to the species information of the animal image sets according to the species information of the animal image sets;

[0043] Step S202: Based on the animal posture image set, the historical animal posture image set, the abnormal animal posture image set, the normal animal posture image set and the trained deep learning network, gait analysis is performed on the animal posture image set to obtain gait analysis data.

[0044] Among them, species information refers to the animal species of the animals in the animal posture image set, and the animal species can be white mice, black mice, cats, dogs, etc. The historical animal posture image set refers to the animal posture image set corresponding to the species information obtained by the target camera equipment in the target environment in the past. The abnormal animal posture image set refers to the animal posture image set under different abnormal conditions collected by the experimenter. The abnormal state can include abnormal emotional state, disease state, such as anxiety, depression, pain, spinal cord injury and other nerve damage, neuropathic pain, arthritis, stroke, Parkinson's disease, cerebellar ataxia, brain trauma and peripheral nerve damage, etc., and abnormal bone and muscle development. The normal animal posture image set refers to the movement posture image set of animals in a healthy state.

[0045] Furthermore, in a preferred embodiment, the gait analysis data includes first feature data, second feature data, third feature data, and fourth feature data, and the step of performing gait analysis on the animal posture image set, the historical animal posture image set, the abnormal animal posture image set, the normal animal posture image set, and the trained deep learning network to obtain the gait analysis data includes:

[0046] Step S301: inputting the animal posture image set into the feature analysis model in the deep learning network to obtain first feature data;

[0047] Step S302: inputting the historical animal posture image set into the feature analysis model in the deep learning network to obtain second feature data;

[0048] Step S303: inputting the abnormal animal posture image set into the feature analysis model in the deep learning network to obtain third feature data;

[0049] Step S304: inputting the normal animal posture image set into the feature analysis model in the deep learning network to obtain fourth feature data;

[0050] In a preferred embodiment, the first characteristic data, the second characteristic data, the third characteristic data and the fourth characteristic data all include at least step length, stride, step width, step frequency, gait cycle, support phase, swing phase, leg lifting height, force analysis data, etc. Among them, stride length refers to the distance traveled between the heel of one foot touching the ground and the heel of the opposite foot touching the ground when walking; stride length refers to the distance traveled by one foot taking a step forward when walking; step width refers to the distance between the center points of the left and right feet when walking; cadence refers to the number of steps taken per minute when walking; gait cycle refers to the process from the heel of one foot touching the ground to the foot touching the ground again when walking; the support phase refers to the stage when the foot is always in contact with the ground during walking, including single support phase and multiple support phases; the swing phase refers to the stage when the swing phase is always out of contact with the ground during walking, from the toe of one lower limb leaving the ground to the heel of the same side touching the ground; the leg lift height refers to the height of the center of the foot lifted off the ground when walking; the force analysis data refers to the data obtained by force analysis of each limb landing of the animal based on the weight of the animal, the speed and acceleration during movement, and the landing direction of the four legs.

[0051] Furthermore, in a preferred embodiment, the animal status report includes a historical status report and a current status report, and the step of performing data processing based on the gait analysis data to obtain the animal status report includes:

[0052] Step S401: performing data processing based on the first feature data, the second feature data, and a feature comparison model to determine a historical status report;

[0053] Step S402: performing data processing based on the first feature data, the third feature data, the fourth feature data, and a feature comparison model to determine a current status report.

[0054] Furthermore, in a preferred embodiment, the historical status report includes portrait data, and the step of performing data processing based on the first feature data, the second feature data, and the feature comparison model to determine the historical status report includes:

[0055] Step S501: inputting the first feature data into a first coding model in a feature comparison model for coding to obtain a first coding vector;

[0056] Step S502: inputting the second feature data into a second coding model in a feature comparison model for coding to obtain a second coding vector;

[0057] Step S503: Inputting the first code vector and the second code vector into a first similarity comparison model in a feature comparison model to obtain a similarity between the first code vector and the second code vector as a first similarity;

[0058] Step S504: If the first similarity satisfies a preset similarity condition, portrait data is generated according to the first feature data and the second feature data.

[0059] The first similarity comparison model is a mathematical model for calculating the similarity between vectors. Specifically, the first similarity comparison model uses cosine similarity. The portrait data is information that abstracts the characteristic data of the animal.

[0060] For example, if the first similarity is greater than a threshold, then the preset similarity condition is satisfied, and portrait data is generated based on the first feature data and the second feature data. Specifically, when generating the portrait, the difference between the first feature data and the second feature data is compared to obtain difference data, and the portrait is generated based on the first feature data, the second feature data, and the difference data to obtain the portrait data. For example, if the difference data includes a step size difference, if the step size in the first feature data is 10 and the step size in the second feature data is 20, then the step size difference between the first feature data and the second feature data is 10.

[0061] Furthermore, in a preferred embodiment, the step of performing gait analysis based on the first feature data, the third feature data, the fourth feature data, and the feature comparison model to determine the current status report includes:

[0062] Step S601: inputting the third feature data into a third coding model in a feature comparison model for encoding to obtain a third coding vector;

[0063] Step S602: inputting the fourth feature data into a fourth coding model in the feature comparison model for coding to obtain a fourth coding vector;

[0064] Step S603: Inputting the first encoding vector and the third encoding vector into a second similarity comparison model in a feature comparison model to obtain a similarity between the first encoding vector and the third encoding vector as a second similarity;

[0065] Step S604: Inputting the first code vector and the fourth code vector into a second similarity comparison model in a feature comparison model to obtain a similarity between the first code vector and the fourth code vector as a third similarity;

[0066] Step S605: If the second similarity is greater than the third similarity, a matching query is performed on the abnormal animal posture image set corresponding to the second similarity based on the standard detection database to determine the current status report, wherein the detection database includes an association table, which records the mapping relationship between pathological characteristics, physical condition, conditioning strategy and the abnormal animal posture image set.

[0067] Specifically, a matching query is performed on the abnormal animal posture image set corresponding to the second similarity based on a standard detection database to obtain pathological characteristics, physical condition, and conditioning strategies, and the pathological characteristics, physical condition, and conditioning strategies are reported as current status.

[0068] In another implementation, if the second similarity is less than the third similarity, a matching query is performed on the abnormal animal posture image set corresponding to the third similarity based on a standard detection database to determine the current status report, wherein the detection database includes an association table, which records the mapping relationship between pathological characteristics, physical condition, conditioning strategy and the abnormal animal posture image set.

[0069] Referring to FIG4 , in one embodiment, a gait analysis device based on animal posture images is provided, the device comprising: an acquisition module 10 for acquiring a set of animal images captured by a target camera device in a target environment;

[0070] An extraction module 20 is used to extract skeleton information from the animal image set to obtain an animal posture image set;

[0071] a gait analysis module 30 for performing gait analysis on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model;

[0072] The data processing module 40 is used to process the gait analysis data to obtain an animal status report.

[0073] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in FIG5 . The computer device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client via a network connection. When the computer program is executed by the processor, it implements the functions or steps of the server side of a gait analysis method based on animal posture images.

[0074] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as shown in FIG6 . The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When the computer program is executed by the processor, the client-side functions or steps of a gait analysis method based on animal posture images are implemented.

[0075] In one embodiment, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are implemented:

[0076] Obtaining a set of animal images captured by a target camera device in a target environment;

[0077] Extracting skeleton information from the animal image set to obtain an animal posture image set;

[0078] Performing gait analysis on the animal posture image set according to the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model;

[0079] Data processing is performed based on the gait analysis data to obtain an animal status report.

[0080] The gait analysis method based on animal posture images proposed in this embodiment obtains a set of animal images captured by a target camera device in a target environment, then extracts skeleton information from the animal image set to obtain an animal posture image set. Gait analysis is then performed on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data. The deep learning network includes a feature analysis model and a feature comparison model. Finally, data processing is performed based on the gait analysis data to obtain an animal status report. This method can accurately and efficiently perform gait analysis on a set of animal images captured in a target environment, thereby obtaining an animal status report with improved analysis results.

[0081] In one embodiment, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0082] Obtaining a set of animal images captured by a target camera device in a target environment;

[0083] Extracting skeleton information from the animal image set to obtain an animal posture image set;

[0084] Performing gait analysis on the animal posture image set according to the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model;

[0085] Data processing is performed based on the gait analysis data to obtain an animal status report.

[0086] The gait analysis method based on animal posture images proposed in this embodiment obtains a set of animal images captured by a target camera device in a target environment, then extracts skeleton information from the animal image set to obtain an animal posture image set. Gait analysis is then performed on the animal posture image set based on the animal posture image set, a standard animal posture image database, and a trained deep learning network to obtain gait analysis data. The deep learning network includes a feature analysis model and a feature comparison model. Finally, data processing is performed based on the gait analysis data to obtain an animal status report. This method can accurately and efficiently perform gait analysis on a set of animal images captured in a target environment, thereby obtaining an animal status report with improved analysis results.

[0087] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can be found in the relevant descriptions of the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0088] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0089] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0090] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A gait analysis method based on animal pose images, characterized in that, the gait analysis method based on animal pose images includes: Obtaining an animal image set captured by a target camera device in a target environment; Extracting skeleton information from the animal image set to obtain an animal pose image set; Performing gait analysis on the animal pose image set according to the animal pose image set, a standard animal pose image database, and a trained deep learning network to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model; Performing data processing based on the gait analysis data to obtain an animal status report.

2. The gait analysis method based on animal pose images according to claim 1, characterized in that, the step of performing gait analysis on the animal pose image set according to the animal pose image set, a standard animal pose image database, and a trained deep learning network to obtain gait analysis data includes: According to the species information corresponding to the animal image set, extracting the historical animal pose image set, abnormal animal pose image set, and normal animal pose image set corresponding to the species information in the standard animal pose image database; Performing gait analysis on the animal pose image set according to the animal pose image set, historical animal pose image set, abnormal animal pose image set, normal animal pose image set, and a trained deep learning network to obtain gait analysis data.

3. The gait analysis method based on animal pose images according to claim 2, characterized in that, the gait analysis data includes first feature data, second feature data, third feature data, and fourth feature data, and the step of performing gait analysis on the animal pose image set according to the animal pose image set, historical animal pose image set, abnormal animal pose image set, normal animal pose image set, and a trained deep learning network to obtain gait analysis data includes: Inputting the animal pose image set into the feature analysis model in the deep learning network to obtain first feature data; Inputting the historical animal pose image set into the feature analysis model in the deep learning network to obtain second feature data; Inputting the abnormal animal pose image set into the feature analysis model in the deep learning network to obtain third feature data; Inputting the normal animal pose image set into the feature analysis model in the deep learning network to obtain fourth feature data.

4. The gait analysis method based on animal pose images according to claim 3, characterized in that, the first feature data includes at least stride length, step amplitude, step width, step frequency, gait cycle, stance phase, swing phase, leg lift height, and force analysis data.

5. The gait analysis method based on animal pose images according to claim 4, characterized in that, the animal status report includes a historical status report and a current status report, and the step of performing data processing based on the gait analysis data to obtain an animal status report includes: Perform data processing based on the first feature data, the second feature data, and the feature comparison model to determine a historical status report; Perform data processing based on the first feature data, the third feature data, the fourth feature data, and the feature comparison model to determine a current status report.

6. The gait analysis method based on animal pose images according to claim 5, wherein, The historical status report includes portrait data. The step of performing data processing based on the first feature data, the second feature data, and the feature comparison model to determine the historical status report includes: Input the first feature data into the first encoding model in the feature comparison model for encoding to obtain a first encoding vector; Input the second feature data into the second encoding model in the feature comparison model for encoding to obtain a second encoding vector; Input the first encoding vector and the second encoding vector into the first similarity comparison model in the feature comparison model to obtain the similarity between the first encoding vector and the second encoding vector as the first similarity; If the first similarity meets the preset similarity condition, generate portrait data according to the first feature data and the second feature data.

7. The gait analysis method based on animal pose images according to claim 6, wherein, The step of performing data processing based on the first feature data, the third feature data, the fourth feature data, and the feature comparison model to determine the current status report includes: Input the third feature data into the third encoding model in the feature comparison model for encoding to obtain a third encoding vector; Input the fourth feature data into the fourth encoding model in the feature comparison model for encoding to obtain a fourth encoding vector; Input the first encoding vector and the third encoding vector into the second similarity comparison model in the feature comparison model to obtain the similarity between the first encoding vector and the third encoding vector as the second similarity; Input the first encoding vector and the fourth encoding vector into the second similarity comparison model in the feature comparison model to obtain the similarity between the first encoding vector and the fourth encoding vector as the third similarity; If the second similarity is greater than the third similarity, perform a matching query on the abnormal animal pose image set corresponding to the second similarity based on the standard detection database to determine the current status report, wherein the detection database includes an association table that records the mapping relationship between pathological features, physical condition, conditioning strategies, and the abnormal animal pose image set.

8. A gait analysis device based on animal pose images, wherein, The gait analysis device based on animal pose images includes: An acquisition module for acquiring an animal image set captured by a target camera device in a target environment; An extraction module for extracting skeleton information from the animal image set to obtain an animal pose image set; A gait analysis module, configured to perform gait analysis on the animal pose image set according to the animal pose image set, the standard animal pose image database, and the trained deep learning network, so as to obtain gait analysis data, wherein the deep learning network includes a feature analysis model and a feature comparison model; A data processing module, configured to perform data processing based on the gait analysis data to obtain an animal state report.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein, when the processor executes the computer program, the steps of the gait analysis method based on animal pose images according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, storing a computer program, wherein, when the computer program is executed by a processor, the steps of the gait analysis method based on animal pose images according to any one of claims 1 to 7 are implemented.

Images