CN118177790A - Gait evaluation method and system combining multi-angle monitoring feedback - Google Patents
Gait evaluation method and system combining multi-angle monitoring feedback Download PDFInfo
- Publication number
- CN118177790A CN118177790A CN202410595668.9A CN202410595668A CN118177790A CN 118177790 A CN118177790 A CN 118177790A CN 202410595668 A CN202410595668 A CN 202410595668A CN 118177790 A CN118177790 A CN 118177790A
- Authority
- CN
- China
- Prior art keywords
- gait
- monitoring
- preset
- point cloud
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Measuring devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor or mobility of a limb
- A61B5/112—Gait analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
- G06V40/25—Recognition of walking or running movements, e.g. gait recognition
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Psychiatry (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physiology (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The application provides a gait evaluation method and a gait evaluation system combined with multi-angle monitoring feedback, which relate to the technical field of image processing, wherein the method comprises the following steps: obtaining a target walking video of a target user; constructing a first monitoring image frame set belonging to a first monitoring point; constructing a gait point cloud model; obtaining a gait three-dimensional model; obtaining a target gait model; acquiring multidimensional feature information of the target gait model based on preset gait features to obtain target gait feature information; and comparing the target gait characteristic information with the preset gait characteristic information of the preset gait model to obtain a gait similarity index. The method and the device can solve the technical problems of lack of objectivity and standardization of gait evaluation results due to higher dependence on experience and skills of doctors in the prior art, realize standardized gait quantitative evaluation, and improve the objectivity and accuracy of gait evaluation, thereby assisting the technical effect of rehabilitation therapy.
Description
Technical Field
The application relates to the technical field of image processing, in particular to a gait evaluation method and system combining multi-angle monitoring feedback.
Background
Along with the development of science and technology, the demand of the rehabilitation medicine field for intelligent rehabilitation equipment is growing. In the intelligent process of rehabilitation equipment, gait evaluation plays a vital role in accurate evaluation of rehabilitation treatment effects of patients.
The existing gait evaluation method mainly depends on manual judgment of doctors, lacks objectivity and standardization, and has higher dependence on experience and skills of the doctors.
In summary, the prior art has the technical problem that the gait evaluation result lacks objectivity and standardization due to higher dependence on experience and skills of doctors.
Disclosure of Invention
The application aims to provide a gait evaluation method and a gait evaluation system combined with multi-angle monitoring feedback, which are used for solving the technical problems of lack of objectivity and standardization of gait evaluation results due to higher dependence on experience and skills of doctors in the prior art.
In view of the above, the present application provides gait evaluation methods and systems incorporating multi-angle monitoring feedback.
In a first aspect, the present application provides a gait evaluation method in combination with multi-angle monitoring feedback, the method being implemented by a gait evaluation system in combination with multi-angle monitoring feedback, wherein the method comprises: reading a preset monitoring scheme, and acquiring a target walking video of a target user by an industrial camera based on the preset monitoring scheme; extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through an industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifications; constructing a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the walking images with the stage marks; performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model; rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model; reading a preset gait feature, and acquiring multidimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information; comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, wherein the gait similarity index is used for quantitatively evaluating the gait of the target user.
In a second aspect, the present application also provides a gait evaluation system in combination with multi-angle monitoring feedback for performing the gait evaluation method in combination with multi-angle monitoring feedback as described in the first aspect, wherein the system comprises: the walking video acquisition module is used for reading a preset monitoring scheme, and the industrial camera acquires a target walking video of a target user based on the preset monitoring scheme; the monitoring point image extraction module is used for extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through the industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifiers; the gait point cloud model building module is used for building a gait point cloud model according to a gait point cloud data set built based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the multi-frame walking images with stage identification; the texture mapping module is used for performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model; the pressure rendering module is used for rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model; the multi-dimensional feature acquisition module is used for reading preset gait features, and acquiring multi-dimensional feature information of the target gait model based on the preset gait features to obtain target gait feature information; the gait quantitative evaluation module is used for comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, and the gait similarity index is used for quantitatively evaluating the gait of the target user.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
reading a preset monitoring scheme, and acquiring a target walking video of a target user by an industrial camera based on the preset monitoring scheme; extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through an industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifications; constructing a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the walking images with the stage marks; performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model; rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model; reading a preset gait feature, and acquiring multidimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information; comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, wherein the gait similarity index is used for quantitatively evaluating the gait of the target user. By constructing the gait model of the target user and performing similarity analysis with the preset gait model, standardized gait quantitative evaluation is realized, and objectivity and accuracy of the gait evaluation are improved, so that the technical effect of rehabilitation therapy is assisted.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent. It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other features of the present application will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the application or the technical solutions of the prior art, the following brief description will be given of the drawings used in the description of the embodiments or the prior art, it being obvious that the drawings in the description below are only exemplary and that other drawings can be obtained from the drawings provided without the inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a gait evaluation method combining multi-angle monitoring feedback according to the present application;
fig. 2 is a schematic structural diagram of a gait evaluation system combining multi-angle monitoring feedback according to the present application.
Reference numerals illustrate: the walking video acquisition module 11, the monitoring point image extraction module 12, the gait point cloud model construction module 13, the texture mapping module 14, the pressure rendering module 15, the multidimensional feature acquisition module 16 and the gait quantitative evaluation module 17.
Detailed Description
The gait evaluation method and system combining multi-angle monitoring feedback solve the technical problems that the gait evaluation result lacks objectivity and standardization due to higher dependence on experience and skills of doctors in the prior art. By constructing the gait model of the target user and performing similarity analysis with the preset gait model, standardized gait quantitative evaluation is realized, and objectivity and accuracy of the gait evaluation are improved, so that the technical effect of rehabilitation therapy is assisted.
In the following, the technical solutions of the present application will be clearly and completely described with reference to the accompanying drawings, and it should be understood that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application, and that the present application is not limited by the exemplary embodiments described herein. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application. It should be further noted that, for convenience of description, only some, but not all of the drawings related to the present application are shown.
Example 1
Referring to fig. 1, the present application provides a gait evaluation method combined with multi-angle monitoring feedback, wherein the method is applied to a gait evaluation system combined with multi-angle monitoring feedback, and the method specifically comprises the following steps:
Step one: reading a preset monitoring scheme, and acquiring a target walking video of a target user by an industrial camera based on the preset monitoring scheme;
Specifically, the target user refers to any patient to be subjected to gait evaluation, and the gait evaluation is performed on the patient through the method provided by the application, so as to assist in the rehabilitation treatment effect evaluation of the patient. The industrial camera is a device for video acquisition during walking of a target user, the application does not limit the model of the industrial camera, and a person skilled in the art can combine practical selection. The predetermined monitoring scheme refers to monitoring points of the industrial camera, such as the front face, the side face and the like of a target user, through video acquisition. The target walking video is an acquisition result obtained by the industrial camera for carrying out video acquisition on the target user according to a preset monitoring scheme.
Step two: extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through an industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifications;
Specifically, the preset monitoring scheme comprises a plurality of monitoring points, the target walking video comprises image frames collected at the plurality of monitoring points, and the image frames corresponding to the first monitoring points are extracted to form a first monitoring image frame set. The stage identification refers to walking stage identification information of the target user, such as starting, falling, etc., and is identified by a person skilled in the art in combination with the actual implementation.
Step three: constructing a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the walking images with the stage marks;
Specifically, any one frame of walking image is extracted from the first monitoring image frame set and recorded as a first walking image, point cloud data corresponding to the three-dimensional space is extracted from the first walking image by using the existing three-dimensional scanning technology, such as laser scanning or structured light scanning, and the like, as first point cloud data, the first point cloud data generally refers to point cloud data corresponding to each of walking images with a plurality of frames having stage marks, and the combination of the first point cloud data and the second point cloud data is used as a gait point cloud data set, which is a common technical means for a person skilled in the art and is not described herein. Further, a gait point cloud model is constructed according to the gait point cloud data set, that is, the existing three-dimensional reconstruction technology such as surface reconstruction algorithm, poisson surface reconstruction, greedy projection triangularization and the like is utilized, the gait point cloud data set is converted into a three-dimensional grid model, the gait point cloud model is obtained, that is, the gait point cloud model is a dynamic walking model of a target user constructed based on multi-frame walking images with stage identification, and walking characteristics such as the number of steps, the walking frequency and the like of the target user can be reflected.
Step four: performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model;
Specifically, when the industrial camera is used for acquiring walking videos of a target user, parameters such as resolution, focal length of a lens, visual field range and the like are used as acquisition characteristic parameters, the industrial camera is controlled to acquire images of the target user again according to the acquisition characteristic parameters, the acquired images are used for carrying out texture feature extraction, the image texture feature extraction is a common technical means for a person skilled in the art, the extracted texture features are mapped onto a gait point cloud model without unfolding, in the process, the coordinate of the texture images is ensured to be consistent with the coordinate of the gait point cloud model, and the gait point cloud model with the texture features is obtained to be used as a gait three-dimensional model.
Step five: rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model;
Specifically, the intelligent motion table is provided with a plurality of pressure monitors, the pressure monitors are existing pressure sensors, pressure acquisition is carried out on a target user through the pressure monitors, target walking pressure information is obtained, and the target walking pressure information comprises a foot starting pressure size, a foot starting pressure direction, a foot starting pressure action point, a foot falling pressure size, a foot falling pressure direction, a foot falling pressure action point, a support swing pressure size, a support swing pressure direction and a support swing pressure action point when the target user walks. Wherein, the foot lifting pressure refers to the pressure generated by the contact of the foot with the ground when one foot is lifted from the ground in the gait cycle; the pressure direction of the foot refers to the direction in which the foot applies pressure to the ground when the foot is lifted; the foot pressure action point refers to a specific position or action point of the foot applying pressure to the ground when the foot is lifted; the foot drop pressure refers to the pressure generated by the contact of a foot with the ground when the foot contacts the ground during a gait cycle; the falling pressure direction refers to the direction in which the foot applies pressure to the ground when falling; the foot pressure action point refers to a specific position or action point of the foot applying pressure to the ground when the foot falls; the magnitude of the support swing pressure refers to the amount of pressure generated by the foot in contact with the ground when one foot supports the body weight during the gait cycle; the direction of the supporting swing pressure refers to the direction of the pressure exerted by the foot on the ground in the supporting swing stage; the support swing pressure point of action refers to the specific location or point of action at which the foot applies pressure to the ground as the support swings. And rendering the target walking pressure information to the gait three-dimensional model to obtain a target gait model, and extracting gait characteristics of a target user during walking through the target gait model so as to perform gait evaluation.
Step six: reading a preset gait feature, and acquiring multidimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information;
The predetermined gait characteristics include a foot pressure magnitude, a foot pressure direction, a foot pressure point of application, a support swing pressure magnitude, a support swing pressure direction, a support swing pressure point of application, a stride frequency, and a manner of landing, wherein the stride refers to a horizontal distance that one foot is landed until the foot is landed again during walking or running; step frequency refers to the number of steps completed in unit time; the landing mode describes the posture and angle of the foot when contacting the ground, such as heel strike, full sole strike, toe strike, etc. The features contained in the preset gait features can be directly extracted from the target gait model, and the extraction result is the target gait feature information.
Step seven: comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, wherein the gait similarity index is used for quantitatively evaluating the gait of the target user.
Specifically, the predetermined gait model is constructed by combining the method for constructing the target gait model provided by the application by a person skilled in the art, that is, the target user needs to be evaluated for rehabilitation, the professional rehabilitation expert can set the rehabilitation expected states in different stages, the historic user meeting the rehabilitation expected states can construct the gait model, the predetermined gait model is obtained, and the characteristic value corresponding to the predetermined gait characteristic is extracted from the predetermined gait model to serve as the predetermined gait characteristic information. The gait similarity index is further compared with the similarity between the target gait feature information and the preset gait feature information, the gait similarity index can be obtained through the existing cosine similarity similar identification method, the higher the gait similarity index is, the closer the rehabilitation condition of the target user is to the rehabilitation expected state, the gait similarity index is taken as the gait quantitative evaluation result of the target user, the user's rehabilitation condition can be known in time conveniently, standardized automatic evaluation of gait evaluation is achieved, objectivity is improved, and rehabilitation therapy is assisted.
Further, the first step of the present application further comprises:
Reading a preset monitoring distance; determining a preset monitoring path adjacent distance according to the picture height of the industrial camera at the preset monitoring distance; determining a preset monitoring speed according to the picture width of the industrial camera at the preset monitoring distance; the predetermined monitoring scheme is determined based on the predetermined monitoring distance, the predetermined monitoring path adjacency distance, and the predetermined monitoring speed.
Specifically, the determining process of the predetermined monitoring scheme is as follows: the preset monitoring distance refers to a linear distance between the industrial camera and the target user, and can be obtained by direct measurement through the existing distance measuring instrument. And taking the preset monitoring distance as a reference, carrying out image acquisition test by an industrial camera, and determining the picture height and the picture width based on the test result. The frame height refers to the number of pixels in the vertical direction on an imaging sensor of an industrial camera, and directly affects the vertical resolution of a captured image, such as 1080 pixels, and is related to the monitoring distance of the industrial camera, so that the vertical size of a scene range that can be captured by the industrial camera at a specific distance is determined. The width of the frame refers to the number of pixels in the horizontal direction on the imaging sensor of the camera, which determines the horizontal resolution of the captured image, and similar to the height of the frame, the higher the width of the frame, the wider the level of the scene range captured by the camera, for example, the pixel width is 1920, so that a wider scene can be captured, and the width of the frame is also affected by the monitoring distance, which determines the horizontal scene range that can be covered at a specific distance.
The preset monitoring path adjacent distance refers to the width of a monitoring area covered by the industrial camera in the monitoring process, and is related to the picture height of the industrial camera and the preset monitoring distance, and the calculating method comprises the following steps: predetermined monitoring path adjacency= (frame height/monitoring distance) multiplied by tan (angle of view), where the angle of view is the angle of view of the industrial camera lens, can be obtained directly from the lens specification. The predetermined monitoring speed refers to the frequency at which the industrial camera captures images, typically expressed in frame rate, at a predetermined monitoring distance, which can be estimated by the following formula: the predetermined monitoring speed= (moving speed of target user/width of frame) multiplied by the required image fluency, the moving speed of target user is the actual speed of target user passing through the monitoring area in unit time, and the required image fluency is the expected frame rate determined according to the application scene, and is set by the person skilled in the art.
And determining a plurality of monitoring points as a preset monitoring scheme based on the preset monitoring distance, the preset monitoring path adjacent distance and the preset monitoring speed, wherein the preset monitoring distance, the preset monitoring path adjacent distance and the preset monitoring speed reflect the monitoring range of an industrial camera, and in order to ensure the omnibearing monitoring of a target user, a plurality of monitoring points such as a front side, a side, a rear side and the like are arranged around the target user according to the preset monitoring distance, the preset monitoring path adjacent distance and the monitoring range of the preset monitoring speed, so that the corresponding preset monitoring distance, the preset monitoring path adjacent distance and the preset monitoring speed under a plurality of industrial cameras can be used for omnibearing monitoring of the target user, thereby improving the monitoring comprehensiveness and laying a foundation for the subsequent gait evaluation.
Further, the third step of the present application further comprises:
Extracting a first sample set from the gait point cloud data set through the industrial personal computer, and obtaining a first point cloud model according to the first sample set; recording the gait point cloud data set from which the first sample set is removed as a first non-sample set; judging whether a first distance from first non-sample point cloud data to the first point cloud model accords with a preset distance deviation threshold value or not, wherein the first non-sample point cloud data is any one point cloud data in the first non-sample set; if yes, adding the first non-sample point cloud data to a first inner point set, and if not, adding the first non-sample point cloud data to a first outer point set; and when the data volume ratio of the first inner point set to the first outer point set reaches a preset ratio constraint, the first point cloud model is used as the gait point cloud model.
Further, the application also comprises the following steps:
Generating a reconstruction instruction when the data volume ratio of the first inner point set to the first outer point set does not reach the preset ratio constraint; the industrial personal computer extracts a second sample set from the gait point cloud data set according to the reconstruction instruction, and obtains a second point cloud model according to the second sample set; and carrying out iterative analysis on the second point cloud model until the predetermined ratio constraint is reached, and taking the point cloud model at the moment as the gait point cloud model.
Specifically, the construction process of the gait point cloud model is as follows:
And extracting a first sample set from the gait point cloud data set by the industrial personal computer, wherein the first sample set generally refers to a point cloud data set generated by a walking image under any monitoring point, and converting the first sample set into a three-dimensional grid model by utilizing the existing three-dimensional reconstruction technology, such as a surface reconstruction algorithm, a poisson surface reconstruction, greedy projection triangulation and the like, so as to obtain a first point cloud model. And removing the first sample set from the gait point cloud data set, and taking the rest point cloud data as a first non-sample set. Any one point cloud data in the first non-sample set, namely, a first distance from the first non-sample point cloud data to the first point cloud model is calculated, and specifically, the distance between the first non-sample point cloud data and a point closest to the first point cloud model can be calculated to be used as the first distance. Further, a predetermined distance deviation threshold is set by a person skilled in the art in combination with practical experience, and the predetermined distance deviation threshold refers to a distance threshold that can consider that the point cloud data belong to the same location.
If the first distance from the first non-sample point cloud data to the first point cloud model accords with a preset distance deviation threshold, adding the corresponding first non-sample point cloud data to a first inner point set, if the first non-sample point cloud data does not accord with the preset distance deviation threshold, adding the first non-sample point cloud data to a first outer point set, wherein it can be understood that the point cloud data in the first inner point set refers to a value which is the distance from the first point cloud model to the preset distance deviation threshold, namely, the more the data volume in the first inner point set, the higher the accuracy of the first point cloud model is, so that the accuracy of the first point cloud model is judged according to the data volume ratio of the first inner point set to the first outer point set, and the preset ratio constraint refers to the ratio which is set by a person skilled in the art in combination with actual experience and is considered that the accuracy of the point cloud model accords with the requirement.
Specifically, when the ratio of the data amounts of the first inner point set to the first outer point set does not reach the predetermined ratio constraint, the accuracy of the first point cloud model is not required, the point cloud model needs to be reconstructed at the moment, a reconstruction instruction is generated, the reconstruction instruction is a control instruction for controlling reconstruction of the point cloud model, the industrial personal computer extracts a second sample set from the gait point cloud data set according to the reconstruction instruction, and the second sample set refers to a set formed by any point cloud data different from the first sample set. And converting the second sample set into a three-dimensional grid model by using the existing three-dimensional reconstruction technology, such as a surface reconstruction algorithm, a poisson surface reconstruction, greedy projection triangulation and the like, so as to obtain a second point cloud model. And repeating the steps, analyzing whether the data volume ratio of the second inner point set to the second outer point set of the second point cloud model reaches the preset ratio constraint, if so, taking the second point cloud model as a gait point cloud model, if not, continuing to perform iterative construction of the point cloud model until the preset ratio constraint is reached, and taking the point cloud model reaching the preset ratio constraint as the gait point cloud model, thereby improving the accuracy of the gait point cloud model and further ensuring the accuracy of gait evaluation.
Further, the fifth step of the present application further comprises:
Acquiring a first pressure value of a first pressure monitor, wherein the first pressure monitor is any one of the plurality of pressure monitors, and the first pressure monitor corresponds to a first coordinate; and generating the target walking pressure information based on the first coordinates and the first pressure value.
Specifically, the target walking pressure information is obtained as follows: the intelligent motion platform is provided with a plurality of pressure monitors, the intelligent motion platform is arranged in a walking area of a target user, the pressure monitors refer to existing pressure sensors, and the pressure of feet to the ground is collected in the walking process of the target user through the pressure monitors. Each pressure monitor corresponds to a specific coordinate, which may be predetermined by those skilled in the art, and the first pressure monitor is generally referred to as any one of the pressure monitors, the coordinate corresponding to the first pressure monitor is extracted, the corresponding sitting mark is the first coordinate, when the target user walks, each pressure monitor captures a corresponding pressure value, and the recorded pressure value is obtained from the first pressure monitor, and the value is a pressure reading generated on the monitor when the user walks. And combining the first coordinates and the first pressure value to obtain pressure generated by starting and falling feet, pressure action coordinate points and the like at different stages in the walking process, thereby forming target walking pressure information. And the pressure data acquisition is realized, and support is provided for subsequent gait evaluation.
Further, the seventh step of the present application further comprises:
Reading a predetermined gait feature tag scheme; performing labeling treatment on the target gait feature information according to the preset gait feature label scheme to obtain a target gait feature vector; performing labeling processing on the preset gait feature information according to the preset gait feature label scheme to obtain a preset gait feature vector; and comparing the target gait feature vector with the preset gait feature vector to obtain the gait similarity index.
Specifically, the predetermined gait feature tag scheme is formulated by a person skilled in the art, and includes a series of tags or categories for describing and classifying gait features, including stride, stride frequency, starting pressure, etc., that is, instead of comparing the target gait feature information with all the information in the predetermined gait feature information, the person skilled in the art sets a required feature type as a feature tag based on actual experience, and composes the predetermined gait feature tag scheme.
And labeling the target gait feature information by using a preset gait feature tag scheme, namely, matching the feature information in the target gait feature information with each feature tag of the preset gait feature tag scheme to obtain an information set containing all the feature tags in the preset gait feature tag scheme as a target gait feature vector. And similarly, labeling the predetermined gait feature information by using a predetermined gait feature label scheme to obtain an information set containing all feature labels in the predetermined gait feature label scheme as a predetermined gait feature vector. Comparing the target gait feature vector with the predetermined gait feature vector to calculate a gait similarity index, the degree of similarity between the two vectors can be quantified by existing similarity analysis methods, such as cosine similarity, euclidean distance, etc., to generate a gait similarity index. This index is a numerical value reflecting the similarity between the target gait and the predetermined gait. The higher the gait similarity index is, the closer the rehabilitation state of the target user is to the rehabilitation state of the preset gait model is, so that gait quantitative evaluation is realized, the rehabilitation state of the user is accurately known, and rehabilitation treatment is assisted to the target user.
In summary, the gait evaluation method combining multi-angle monitoring feedback provided by the application has the following technical effects:
reading a preset monitoring scheme, and acquiring a target walking video of a target user by an industrial camera based on the preset monitoring scheme; extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through an industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifications; constructing a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the walking images with the stage marks; performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model; rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model; reading a preset gait feature, and acquiring multidimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information; comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, wherein the gait similarity index is used for quantitatively evaluating the gait of the target user. By constructing the gait model of the target user and performing similarity analysis with the preset gait model, standardized gait quantitative evaluation is realized, and objectivity and accuracy of the gait evaluation are improved, so that the technical effect of rehabilitation therapy is assisted.
Example two
Based on the same inventive concept as the gait evaluation method combined with the multi-angle monitoring feedback in the foregoing embodiment, the present application further provides a gait evaluation system combined with the multi-angle monitoring feedback, please refer to fig. 2, the system includes:
the walking video acquisition module 11 is used for reading a preset monitoring scheme, and the industrial camera acquires a target walking video of a target user based on the preset monitoring scheme;
the monitoring point image extraction module 12 is configured to extract and construct, by using an industrial personal computer, a first monitoring image frame set belonging to a first monitoring point in the target walking video, where the first monitoring point is any monitoring point in the predetermined monitoring scheme, and the first monitoring image frame set includes multiple frames of walking images with stage identifiers;
The gait point cloud model construction module 13 is configured to construct a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, where the first walking image is any one frame of image in the multi-frame walking images with stage identifiers;
The texture mapping module 14 is used for performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model;
The pressure rendering module 15 is used for rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model;
The multi-dimensional feature acquisition module 16 is used for reading a preset gait feature, and acquiring multi-dimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information;
the gait quantitative evaluation module 17 is configured to compare the target gait characteristic information with predetermined gait characteristic information of a predetermined gait model, and obtain a gait similarity index, where the gait similarity index is used for quantitatively evaluating the gait of the target user.
Further, the walking video acquisition module 11 in the system is further configured to:
Reading a preset monitoring distance;
determining a preset monitoring path adjacent distance according to the picture height of the industrial camera at the preset monitoring distance;
Determining a preset monitoring speed according to the picture width of the industrial camera at the preset monitoring distance;
the predetermined monitoring scheme is determined based on the predetermined monitoring distance, the predetermined monitoring path adjacency distance, and the predetermined monitoring speed.
Further, the walking video acquisition module 11 in the system is further configured to:
Extracting a first sample set from the gait point cloud data set through the industrial personal computer, and obtaining a first point cloud model according to the first sample set;
Recording the gait point cloud data set from which the first sample set is removed as a first non-sample set;
Judging whether a first distance from first non-sample point cloud data to the first point cloud model accords with a preset distance deviation threshold value or not, wherein the first non-sample point cloud data is any one point cloud data in the first non-sample set;
if yes, adding the first non-sample point cloud data to a first inner point set, and if not, adding the first non-sample point cloud data to a first outer point set;
and when the data volume ratio of the first inner point set to the first outer point set reaches a preset ratio constraint, the first point cloud model is used as the gait point cloud model.
Further, the gait point cloud model construction module 13 in the system is further configured to:
Generating a reconstruction instruction when the data volume ratio of the first inner point set to the first outer point set does not reach the preset ratio constraint;
The industrial personal computer extracts a second sample set from the gait point cloud data set according to the reconstruction instruction, and obtains a second point cloud model according to the second sample set;
And carrying out iterative analysis on the second point cloud model until the predetermined ratio constraint is reached, and taking the point cloud model at the moment as the gait point cloud model.
Further, the pressure rendering module 15 in the system is also configured to:
acquiring a first pressure value of a first pressure monitor, wherein the first pressure monitor is any one of the plurality of pressure monitors, and the first pressure monitor corresponds to a first coordinate;
and generating the target walking pressure information based on the first coordinates and the first pressure value.
Further, the gait quantification evaluation module 17 in the system further includes:
The predetermined gait characteristics include a foot pressure magnitude, a foot pressure direction, a foot pressure point of action, a support swing pressure magnitude, a support swing pressure direction, a support swing pressure point of action, a stride frequency, and a landing pattern.
Further, the gait quantification evaluation module 17 in the system is also configured to:
Reading a predetermined gait feature tag scheme;
performing labeling treatment on the target gait feature information according to the preset gait feature label scheme to obtain a target gait feature vector;
performing labeling processing on the preset gait feature information according to the preset gait feature label scheme to obtain a preset gait feature vector;
and comparing the target gait feature vector with the preset gait feature vector to obtain the gait similarity index.
The embodiments of the present invention are described in a progressive manner, and each embodiment focuses on the difference from the other embodiments, and the gait evaluation method and specific example combined with multi-angle monitoring feedback in the first embodiment of fig. 1 are equally applicable to the gait evaluation system combined with multi-angle monitoring feedback in the present embodiment, and by the detailed description of the gait evaluation method combined with multi-angle monitoring feedback in the present embodiment, those skilled in the art can clearly know the gait evaluation system combined with multi-angle monitoring feedback in the present embodiment, so that the details of the present invention are not described herein for brevity of the present invention. For the system disclosed in the embodiment, since the system corresponds to the method disclosed in the embodiment, the description is simpler, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalent techniques thereof, the present application is also intended to include such modifications and variations.
Claims (8)
1. The gait evaluation method combined with the multi-angle monitoring feedback is characterized by comprising the following steps of:
Reading a preset monitoring scheme, and acquiring a target walking video of a target user by an industrial camera based on the preset monitoring scheme;
Extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through an industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifications;
constructing a gait point cloud model according to a gait point cloud data set constructed based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the walking images with the stage marks;
performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model;
rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model;
Reading a preset gait feature, and acquiring multidimensional feature information of the target gait model based on the preset gait feature to obtain target gait feature information;
Comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, wherein the gait similarity index is used for quantitatively evaluating the gait of the target user.
2. The gait evaluation method in combination with multi-angle monitoring feedback according to claim 1, wherein the determining process of the predetermined monitoring scheme comprises:
Reading a preset monitoring distance;
determining a preset monitoring path adjacent distance according to the picture height of the industrial camera at the preset monitoring distance;
Determining a preset monitoring speed according to the picture width of the industrial camera at the preset monitoring distance;
the predetermined monitoring scheme is determined based on the predetermined monitoring distance, the predetermined monitoring path adjacency distance, and the predetermined monitoring speed.
3. The gait evaluation method in combination with multi-angle monitoring feedback according to claim 1, wherein the construction process of the gait point cloud model comprises:
Extracting a first sample set from the gait point cloud data set through the industrial personal computer, and obtaining a first point cloud model according to the first sample set;
Recording the gait point cloud data set from which the first sample set is removed as a first non-sample set;
Judging whether a first distance from first non-sample point cloud data to the first point cloud model accords with a preset distance deviation threshold value or not, wherein the first non-sample point cloud data is any one point cloud data in the first non-sample set;
if yes, adding the first non-sample point cloud data to a first inner point set, and if not, adding the first non-sample point cloud data to a first outer point set;
and when the data volume ratio of the first inner point set to the first outer point set reaches a preset ratio constraint, the first point cloud model is used as the gait point cloud model.
4. A gait evaluation method in combination with multi-angle monitoring feedback according to claim 3, wherein the method further comprises:
Generating a reconstruction instruction when the data volume ratio of the first inner point set to the first outer point set does not reach the preset ratio constraint;
The industrial personal computer extracts a second sample set from the gait point cloud data set according to the reconstruction instruction, and obtains a second point cloud model according to the second sample set;
And carrying out iterative analysis on the second point cloud model until the predetermined ratio constraint is reached, and taking the point cloud model at the moment as the gait point cloud model.
5. The gait evaluation method in combination with multi-angle monitoring feedback according to claim 1, wherein the intelligent exercise platform is provided with a plurality of pressure monitors, and the process of obtaining the target walking pressure information comprises the following steps:
acquiring a first pressure value of a first pressure monitor, wherein the first pressure monitor is any one of the plurality of pressure monitors, and the first pressure monitor corresponds to a first coordinate;
and generating the target walking pressure information based on the first coordinates and the first pressure value.
6. The gait evaluation method in combination with multi-angle monitoring feedback of claim 1, wherein the predetermined gait characteristics include a foot pressure magnitude, a foot pressure direction, a foot pressure point of application, a support swing pressure magnitude, a support swing pressure direction, a support swing pressure point of application, a stride frequency, and a landing pattern.
7. The gait evaluation method in combination with multi-angle monitoring feedback of claim 6, comprising:
Reading a predetermined gait feature tag scheme;
performing labeling treatment on the target gait feature information according to the preset gait feature label scheme to obtain a target gait feature vector;
performing labeling processing on the preset gait feature information according to the preset gait feature label scheme to obtain a preset gait feature vector;
and comparing the target gait feature vector with the preset gait feature vector to obtain the gait similarity index.
8. Gait evaluation system in combination with multi-angle monitoring feedback, characterized by the steps for implementing the method of any of claims 1 to 7, said system comprising:
The walking video acquisition module is used for reading a preset monitoring scheme, and the industrial camera acquires a target walking video of a target user based on the preset monitoring scheme;
The monitoring point image extraction module is used for extracting and constructing a first monitoring image frame set belonging to a first monitoring point in the target walking video through the industrial personal computer, wherein the first monitoring point refers to any monitoring point in the preset monitoring scheme, and the first monitoring image frame set comprises a plurality of walking images with stage identifiers;
The gait point cloud model building module is used for building a gait point cloud model according to a gait point cloud data set built based on first point cloud data of a first walking image, wherein the first walking image is any frame of image in the multi-frame walking images with stage identification;
The texture mapping module is used for performing texture mapping on the gait point cloud model by combining the acquisition characteristic parameters of the industrial camera to obtain a gait three-dimensional model;
the pressure rendering module is used for rendering the target walking pressure information of the target user acquired by the intelligent motion platform to the gait three-dimensional model to obtain a target gait model;
The multi-dimensional feature acquisition module is used for reading preset gait features, and acquiring multi-dimensional feature information of the target gait model based on the preset gait features to obtain target gait feature information;
The gait quantitative evaluation module is used for comparing the target gait characteristic information with preset gait characteristic information of a preset gait model to obtain a gait similarity index, and the gait similarity index is used for quantitatively evaluating the gait of the target user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410595668.9A CN118177790A (en) | 2024-05-14 | 2024-05-14 | Gait evaluation method and system combining multi-angle monitoring feedback |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410595668.9A CN118177790A (en) | 2024-05-14 | 2024-05-14 | Gait evaluation method and system combining multi-angle monitoring feedback |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118177790A true CN118177790A (en) | 2024-06-14 |
Family
ID=91393207
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410595668.9A Pending CN118177790A (en) | 2024-05-14 | 2024-05-14 | Gait evaluation method and system combining multi-angle monitoring feedback |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118177790A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120086550A1 (en) * | 2009-02-24 | 2012-04-12 | Leblanc Donald Joseph | Pedobarographic biometric system |
| CN106056050A (en) * | 2016-05-23 | 2016-10-26 | 武汉盈力科技有限公司 | Multi-view gait identification method based on adaptive three dimensional human motion statistic model |
| CN107578019A (en) * | 2017-09-13 | 2018-01-12 | 河北工业大学 | Gait recognition system and recognition method based on visual-tactile fusion |
| CN111536977A (en) * | 2020-05-19 | 2020-08-14 | 广东电网有限责任公司 | Unmanned aerial vehicle inspection route planning method and related device |
| CN116645697A (en) * | 2023-05-31 | 2023-08-25 | 银河水滴科技（北京）有限公司 | A multi-view gait recognition method, device, electronic equipment and storage medium |
-
2024
- 2024-05-14 CN CN202410595668.9A patent/CN118177790A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120086550A1 (en) * | 2009-02-24 | 2012-04-12 | Leblanc Donald Joseph | Pedobarographic biometric system |
| CN106056050A (en) * | 2016-05-23 | 2016-10-26 | 武汉盈力科技有限公司 | Multi-view gait identification method based on adaptive three dimensional human motion statistic model |
| CN107578019A (en) * | 2017-09-13 | 2018-01-12 | 河北工业大学 | Gait recognition system and recognition method based on visual-tactile fusion |
| CN111536977A (en) * | 2020-05-19 | 2020-08-14 | 广东电网有限责任公司 | Unmanned aerial vehicle inspection route planning method and related device |
| CN116645697A (en) * | 2023-05-31 | 2023-08-25 | 银河水滴科技（北京）有限公司 | A multi-view gait recognition method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | A depth camera motion analysis framework for tele-rehabilitation: Motion capture and person-centric kinematics analysis | |
| Dikovski et al. | Evaluation of different feature sets for gait recognition using skeletal data from Kinect | |
| CN111012353A (en) | A height detection method based on face key point recognition | |
| CN115105062B (en) | Hip and knee joint coordination evaluation method, device and system and storage medium | |
| JP5868426B2 (en) | How to estimate the orientation of a stationary person | |
| Seo et al. | SVM-based waist circumference estimation using Kinect | |
| CN115568823B (en) | A method, system and device for evaluating human balance ability | |
| Wang et al. | Gait analysis and validation using voxel data | |
| CN112102951B (en) | Health monitoring method and device based on human body posture estimation algorithm | |
| Coudert et al. | A method to obtain 3D foot shape deformation during the gait cycle | |
| EP3438601B1 (en) | Measurement device, measurement method, and computer program | |
| CN113749646A (en) | Monocular vision-based human body height measuring method and device and electronic equipment | |
| CN112438722B (en) | Sarcopenia assessment device and storage medium | |
| KR20160035497A (en) | Body analysis system based on motion analysis using skeleton information | |
| JP2020018365A (en) | Foot state analysis method | |
| CN118177790A (en) | Gait evaluation method and system combining multi-angle monitoring feedback | |
| KR20220076550A (en) | System and method for recommending user-customized footwear | |
| Mundt et al. | Automating Video‐Based Two‐Dimensional Motion Analysis in Sport? Implications for Gait Event Detection, Pose Estimation, and Performance Parameter Analysis | |
| CN109740458B (en) | Method and system for measuring physical characteristics based on video processing | |
| KR20220042631A (en) | Apparatus and method for gait measurement based on gait video | |
| Niknejad et al. | Equine kinematic gait analysis using stereo videography and deep learning: stride length and stance duration estimation | |
| CN114299616A (en) | Machine vision-based cow gait scoring method and system | |
| Ye et al. | Kinematics analysis multimedia system for rehabilitation | |
| Serrano et al. | Automated feet detection for clinical gait assessment | |
| JP2004167002A (en) | Instrument and method for measuring step |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |