CN115251904A - Analysis method and device based on multiple motion sensors - Google Patents
Analysis method and device based on multiple motion sensors Download PDFInfo
- Publication number
- CN115251904A CN115251904A CN202210835054.4A CN202210835054A CN115251904A CN 115251904 A CN115251904 A CN 115251904A CN 202210835054 A CN202210835054 A CN 202210835054A CN 115251904 A CN115251904 A CN 115251904A
- Authority
- CN
- China
- Prior art keywords
- data
- analysis
- output data
- label
- motion
- 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
Images
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
- A61B5/1116—Determining posture transitions
- A61B5/1117—Fall detection
-
- 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 OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B2503/00—Evaluating a particular growth phase or type of persons or animals
- A61B2503/04—Babies, e.g. for SIDS detection
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Dentistry (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Heart & Thoracic Surgery (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Pure & Applied Mathematics (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Medical Informatics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Physiology (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides an analysis method and device based on multiple motion sensors, wherein the method comprises the following steps: acquiring output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are arranged at each part of a human body; performing local linear embedding analysis on the preprocessed output data to obtain analysis data; constraining the analysis data based on the Laplace feature mapping; dividing the constrained analysis data by utilizing t-distributed random neighbor embedding, removing irrelevant output data according to a division result, and reserving relevant output data; the analysis data of the relevant output data is imported to the motion sensor. Therefore, irrelevant data are removed through local linear embedding analysis, laplace feature mapping and t-distribution random neighbor embedding, only effective information is kept for analysis, the analysis accuracy can be improved, and the method is suitable for preventing infants from falling down.
Description
Technical Field
The invention relates to the technical field of sensors, in particular to an analysis method based on multiple motion sensors and an analysis device based on the multiple motion sensors.
Background
The health and safety problems of infants are always the focus of increasing attention of people, in the Chinese injury prevention report written by the organization of the ministry of health, the falling is the primary reason that infants are accidentally injured, after the infant falls, on one hand, the infant falls to cause personal injury, and on the other hand, the infant falls to cause secondary injury due to the fact that timely protection measures are not provided and timely rescue is obtained, and the normal activity and healthy growth of the infants are seriously influenced.
In the related art, patent application publication No. CN20120209092 describes a human body gait evaluation system and method, which mainly analyze the gait of a human body by collecting three-dimensional acceleration signals, but the acceleration signals only cannot calculate the overall state of the human body, and the accuracy is limited, so that the system and method are not suitable for preventing infants from falling down.
Disclosure of Invention
The invention provides the following technical scheme for solving the problems of low accuracy and unsuitability for preventing infants from falling down in the related technology.
Although related research on the aspect of fall prevention exists in the existing research and development technologies, much more about how to avoid secondary injury after falling, few researchers concentrate on protecting infants from the source and avoiding primary injury.
To this end, an embodiment of a first aspect of the present invention provides a multi-motion sensor based analysis method, including: acquiring output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are arranged on each part of a human body; performing local linear embedding analysis on the preprocessed output data to obtain analysis data; constraining the analysis data based on a Laplace eigenmap; dividing the constrained analysis data by utilizing t-distributed random neighbor embedding, removing irrelevant output data according to a division result, and reserving relevant output data; importing analysis data of the correlated output data into the motion sensor.
In addition, the multi-motion sensor based analysis method according to the above-described embodiment of the present invention may also have the following additional technical features.
According to one embodiment of the invention, preprocessing the output data comprises: and sequentially cleaning and screening the output data, and then carrying out category marking to obtain a category label of each output data.
According to one embodiment of the invention, the local linear embedding analysis is performed on the preprocessed output data, and comprises the following steps: converting each category label into label data to determine the label data and a plurality of label data nearby the label data; determining a distance between the tag data and a number of tag data in its vicinity; obtaining a weight parameter of the tag data through minimization processing of the distance; and performing dimension reduction analysis on all the label data according to the weight parameters.
According to one embodiment of the invention, constraining the analysis data based on the laplacian eigenmaps comprises: establishing a candidate area according to the similarity between the label data; calculating a similarity parameter between two label data in the candidate area; connecting the two label data when the similarity parameter is greater than or equal to a threshold value; when the similarity parameter is smaller than the threshold value, keeping the two label data unconnected; estimating a distance between two of the tag data when the two tag data are connected, the distance having a loss function of:
wherein, yiIs the i-th regular term, y, of the label datajIs the ith negative term, V, of the tag datai,jAs tag data yiAnd yjT is yiAnd yjInterval therebetween, L is yiAnd yjY is a set of regular terms of the label data; and constraining the loss function according to the similarity parameter.
According to an embodiment of the present invention, rejecting irrelevant output data and retaining relevant output data according to a partitioning result includes: minimizing the dispersion between two adjacent output data to obtain a feature vector; the correlated output data are grouped together according to the feature vector, and the uncorrelated output data are in discrete states.
According to one embodiment of the invention, importing analysis data of the correlated output data into the motion sensor comprises: determining a weight parameter of the correlated output data and importing the weight parameter of the correlated data to the motion sensor.
The embodiment of the second aspect of the invention provides an analysis device based on multiple motion sensors, which comprises: the processing module is used for acquiring output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are arranged at each part of a human body; the analysis module is used for carrying out local linear embedding analysis on the preprocessed output data to obtain analysis data; a constraint module for constraining the analysis data based on a laplacian eigenmap; the dividing module is used for dividing the constrained analysis data by utilizing t-distributed random neighbor embedding, eliminating irrelevant output data according to a dividing result and retaining relevant output data; and the importing module is used for importing the analysis data of the related data into the motion sensor.
According to the technical scheme of the embodiment of the invention, irrelevant data is removed through local linear embedding analysis, laplace feature mapping and t-distribution random neighbor embedding, only effective information is reserved for analysis, the analysis accuracy can be improved, and the method is suitable for preventing infants from falling down.
Drawings
FIG. 1 is a flow chart of a multi-motion sensor based analysis method according to an embodiment of the present invention.
FIG. 2 is a diagram illustrating the distribution of tag data in an example of the present invention.
FIG. 3 is a block diagram of a multi-motion sensor based analysis device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
FIG. 1 is a flow chart of a multi-motion sensor based analysis method according to an embodiment of the present invention.
As shown in fig. 1, the multi-motion sensor based analysis method includes the following steps S1 to S5.
The method comprises the following steps of S1, collecting output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are installed on all parts of a human body.
The plurality of motion sensors may be, for example, speed sensors, position sensors, posture sensors, or the like, and may be mounted on a part where the motion of the person can be detected without affecting the normal movement of the person, such as shoes, trousers, or arms of a person who is an infant.
Specifically, when the human body moves, the output data of each motion sensor is collected and preprocessed, such as cleaning and screening, to obtain the preprocessed output data.
And S2, carrying out local linear embedding analysis on the preprocessed output data to obtain analysis data.
Specifically, in order to perform dimensionality reduction on the preprocessed output data, local linear embedding analysis is performed on the preprocessed output data to obtain analysis data.
And S3, constraining the analysis data based on the Laplace feature mapping.
Specifically, after the local linear embedding analysis is performed, in order to improve the analysis accuracy, constraints are added to the analysis data based on the laplacian eigenmap to obtain the constrained analysis data, so that the influence of errors is avoided, and the analysis precision is improved.
And S4, dividing the constrained analysis data by utilizing t-distribution random neighbor embedding, eliminating irrelevant output data according to a division result, and keeping relevant output data.
Specifically, after the constrained analysis data are obtained, the constrained analysis data are divided by using t-distribution random neighbor embedding, irrelevant output data (namely, dissimilar data) are removed from all the output data according to the division result, and relevant output data (namely, similar data) are reserved, so that the operation amount and the storage amount are reduced, only effective information is reserved for analysis, and the analysis accuracy is improved.
And S5, importing the analysis data of the related output data into the motion sensor.
Specifically, after the relevant output data is obtained, the analysis data of the relevant output data is selected from the analysis data, and the selected analysis data is imported into the plurality of motion sensors.
Based on the above description, the embodiment of the invention performs local linear embedding analysis on the preprocessed data, constrains the analysis data based on laplacian feature mapping, divides the constrained data by combining t-distribution random neighbor embedding, eliminates irrelevant data, only retains effective information for analysis, increases the analysis accuracy, improves the accuracy, and can be suitable for preventing infants from falling down.
According to the analysis method based on the multi-motion sensor, irrelevant data are embedded and removed through local linear embedding analysis, laplace feature mapping and t-distribution random neighbors, only effective information is reserved for analysis, the analysis accuracy can be improved, and the method is suitable for preventing infants from falling down.
In one embodiment of the present invention, the preprocessing of the output data may include: and cleaning and screening the output data in sequence, and then carrying out category marking to obtain a category label of each output data.
Specifically, the output data of multiple motion sensors may be washed, filtered, and then category-labeled to output a category label for each output data.
In one embodiment of the present invention, performing the local linear embedding analysis on the preprocessed output data may include: converting each category label into label data; determining tag data and a plurality of tag data nearby the tag data; determining the distance between the tag data and a plurality of tag data nearby the tag data; obtaining a weight parameter of the label data through minimization processing of the distance; and performing dimension reduction analysis on all the label data according to the weight parameters.
Specifically, each category tag is converted into tag data, the tag data and a plurality of tag data in the vicinity thereof are determined, and a certain tag data point x is identifiediSelecting a number of points x in its vicinityjBy wijRepresents xiAnd xjThe corresponding weight parameter is obtained by the minimization process of the distance, as shown in fig. 2.
For a certain data point xiThe identification can be performed through linear combination of all points around the point, the linear representation of all points has the minimum distance with the actual point, the weight parameters between the points are obtained, and dimension reduction analysis is performed by using the obtained weight parameters.
By x in FIG. 2iThe point is a central point, and the more densely the adjacent points are distributed, the more the weight of the central point is, namely, the multiple motion transmission is performedThe higher the correlation for sensor related data information and vice versa the lower.
In one embodiment of the present invention, constraining the analysis data based on the laplacian eigenmaps may include: establishing a candidate area according to the similarity between the label data; calculating a similarity parameter between two label data in the candidate area; when the similarity parameter is greater than or equal to the threshold value, connecting the two label data; when the similarity parameter is smaller than the threshold value, keeping the data of the two labels unconnected; when two label data are connected, estimating the distance between the two label data, wherein the loss function of the distance is as follows:
wherein, yiIs the i-th regular term, y, of the label datajIs the ith negative term, V, of the tag datai,jAs tag data yiAnd yjT is yiAnd yjInterval time therebetween, L is yiAnd yjY is a set of regular terms of the label data; and constraining the loss function according to the similarity parameter.
Wherein, the threshold value can be calibrated according to actual specific requirements.
Specifically, a candidate area is established according to the similarity between the label data, a similarity parameter between the two label data in the candidate area is calculated, whether the similarity parameter is larger than or equal to a threshold value or not is judged, and if yes, the two label data are connected; if not, the two tag data are left unconnected. When two tag data are connected, the distance between two points can be approximately calculated according to the connection on the candidate area, and the loss function is formula (1).
Then, the following constraints are added to the loss function:
If the dim y is M,Span{y1,y2,…yN}＝RM (2)
where R is the weight and N is the regularization of the label dataTotal number of items, M is a similarity parameter between two tag data, RMIs the weight similarity of the tag data.
After the constraint is performed according to the formula (2), the data with similar weights are gathered together, so that the data are prevented from being extremely discrete due to the influence of random distribution. Therefore, the influence of errors can be avoided, and the analysis precision is improved.
In an embodiment of the present invention, the removing irrelevant output data and retaining relevant output data according to the division result may include: minimizing the dispersion between two adjacent output data to obtain a feature vector; the correlated output data are grouped together according to the feature vector, and the uncorrelated output data are in discrete states.
Specifically, the constrained analysis data is divided by using t-distribution random neighbor embedding, the dispersion between two data points (a certain point and adjacent points thereof) is minimized to obtain a feature vector, clustering is performed according to the feature vector so as to enable similar data to be clustered together, and irrelevant data is in a discrete state.
Further, importing analysis data of the correlated output data into the motion sensor may include: determining a weight parameter of the correlated output data and importing the weight parameter of the correlated data to the motion sensor.
Specifically, the weight parameters of the relevant data are introduced into a plurality of motion sensors, so that the data output by the motion sensors can reflect three-dimensional parameters more, and the analysis accuracy is improved.
In summary, the embodiment of the invention performs local linear embedding analysis on the preprocessed data, constrains the analyzed data based on laplacian feature mapping, divides the constrained data by combining t-distribution random neighbor embedding, and eliminates irrelevant data, so that the method is suitable for infant falling, can protect infants from the source, and avoids primary injury.
Corresponding to the analysis method based on multiple motion sensors in the above embodiment, the invention also provides an analysis device based on multiple motion sensors.
Fig. 3 is a block schematic diagram of a multi-motion sensor based analysis apparatus according to an embodiment of the present invention.
As shown in fig. 3, the multi-motion sensor based analysis apparatus includes: a processing module 10, an analysis module 20, a constraint module 30, a partitioning module 40 and an import module 50.
The processing module 10 is configured to collect output data of a plurality of motion sensors and preprocess the output data, where the plurality of motion sensors are installed at various parts of a human body; the analysis module 20 is configured to perform local linear embedding analysis on the preprocessed output data to obtain analysis data; the constraint module 30 is configured to constrain the analysis data based on a laplacian eigenmap; the dividing module 40 is configured to divide the constrained analysis data by using t-distributed random neighbor embedding, and reject irrelevant output data and retain relevant output data according to a dividing result; an import module 50 is used to import the analysis data of the relevant data into the motion sensor.
It should be noted that, for the specific implementation and implementation principle of the analysis apparatus based on multiple motion sensors, reference may be made to the specific implementation of the analysis method based on multiple motion sensors, and details are not described here again to avoid redundancy.
According to the analysis device based on the multi-motion sensor, irrelevant data are removed through local linear embedding analysis, laplace feature mapping and t-distribution random neighbor embedding, only effective information is reserved for analysis, the analysis accuracy can be improved, and the analysis device is suitable for preventing infants from falling down.
In the description of the present invention, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to imply that the number of technical features indicated is significant. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The meaning of "plurality" is two or more unless specifically limited otherwise.
In the description of the specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Moreover, various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without being mutually inconsistent.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments. In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (7)
1. A multi-motion sensor based analysis method, comprising:
acquiring output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are arranged at each part of a human body;
performing local linear embedding analysis on the preprocessed output data to obtain analysis data;
constraining the analytical data based on a Laplace eigenmap;
dividing the constrained analysis data by utilizing t-distributed random neighbor embedding, removing irrelevant output data according to a division result, and reserving relevant output data;
importing analysis data of the correlated output data into the motion sensor.
2. The multi-motion sensor based analysis method of claim 1, wherein pre-processing the output data comprises:
and sequentially cleaning and screening the output data, and then carrying out category marking to obtain a category label of each output data.
3. The multi-motion-sensor based analysis method of claim 2, wherein performing a local linear embedding analysis on the pre-processed output data comprises:
converting each of the category labels into label data;
determining the tag data and a plurality of tag data nearby the tag data;
determining a distance between the tag data and a number of tag data in its vicinity;
obtaining a weight parameter of the tag data through minimization processing of the distance;
and performing dimension reduction analysis on all the label data according to the weight parameters.
4. The multi-motion sensor based analysis method of claim 2, wherein constraining the analysis data based on a laplacian eigenmap comprises:
establishing a candidate area according to the similarity between the label data;
calculating a similarity parameter between two label data in the candidate area;
connecting the two label data when the similarity parameter is greater than or equal to a threshold value;
when the similarity parameter is smaller than the threshold value, keeping the two label data unconnected;
estimating a distance between two of the tag data when the two tag data are connected, the distance having a loss function of:
wherein, yiIs the i-th regular term, y, of the label datajIs a markIth negative rule item, V, of signature datai,jAs tag data yiAnd yjT is yiAnd yjInterval time therebetween, L is yiAnd yjY is a set of regular terms of the label data;
and constraining the loss function according to the similarity parameter.
5. The multi-motion-sensor based analysis method of claim 2, wherein culling irrelevant output data and preserving relevant output data according to the partitioning result comprises:
minimizing the dispersion between two adjacent output data to obtain a feature vector;
the correlated output data are grouped together according to the feature vector, and the uncorrelated output data are in discrete states.
6. The multi-motion-sensor based analysis method of claim 4, wherein importing analysis data of the correlated output data into the motion sensor comprises:
determining a weight parameter of the correlated output data and importing the weight parameter of the correlated data to the motion sensor.
7. A multi-motion sensor based analysis device, comprising:
the processing module is used for acquiring output data of a plurality of motion sensors and preprocessing the output data, wherein the plurality of motion sensors are arranged at each part of a human body;
the analysis module is used for carrying out local linear embedding analysis on the preprocessed output data to obtain analysis data;
a constraint module for constraining the analysis data based on a laplacian eigenmap;
the dividing module is used for dividing the constrained analysis data by utilizing t-distributed random neighbor embedding, eliminating irrelevant output data according to a dividing result and retaining relevant output data;
and the importing module is used for importing the analysis data of the related data into the motion sensor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210835054.4A CN115251904A (en) | 2022-07-15 | 2022-07-15 | Analysis method and device based on multiple motion sensors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210835054.4A CN115251904A (en) | 2022-07-15 | 2022-07-15 | Analysis method and device based on multiple motion sensors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115251904A true CN115251904A (en) | 2022-11-01 |
Family
ID=83765404
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210835054.4A Pending CN115251904A (en) | 2022-07-15 | 2022-07-15 | Analysis method and device based on multiple motion sensors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115251904A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316057A (en) * | 2017-06-07 | 2017-11-03 | 哈尔滨工程大学 | Based on the nuclear power unit method for diagnosing faults being locally linear embedding into K nearest neighbor classifiers |
| CN109684604A (en) * | 2018-12-06 | 2019-04-26 | 北京航空航天大学 | A kind of city dynamic analysing method of the non-negative tensor resolution based on context-aware |
| CN111027487A (en) * | 2019-12-11 | 2020-04-17 | 山东大学 | Behavior recognition system, method, medium and device based on multi-convolution kernel residual network |
| CN111938644A (en) * | 2019-05-16 | 2020-11-17 | 上海宽带技术及应用工程研究中心 | Early diagnosis method, device and system for Alzheimer disease based on LLE |
| CN112861929A (en) * | 2021-01-20 | 2021-05-28 | 河南科技大学 | Image classification method based on semi-supervised weighted migration discriminant analysis |
| US20220076003A1 (en) * | 2020-09-04 | 2022-03-10 | Hitachi, Ltd. | Action recognition apparatus, learning apparatus, and action recognition method |
-
2022
- 2022-07-15 CN CN202210835054.4A patent/CN115251904A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107316057A (en) * | 2017-06-07 | 2017-11-03 | 哈尔滨工程大学 | Based on the nuclear power unit method for diagnosing faults being locally linear embedding into K nearest neighbor classifiers |
| CN109684604A (en) * | 2018-12-06 | 2019-04-26 | 北京航空航天大学 | A kind of city dynamic analysing method of the non-negative tensor resolution based on context-aware |
| CN111938644A (en) * | 2019-05-16 | 2020-11-17 | 上海宽带技术及应用工程研究中心 | Early diagnosis method, device and system for Alzheimer disease based on LLE |
| CN111027487A (en) * | 2019-12-11 | 2020-04-17 | 山东大学 | Behavior recognition system, method, medium and device based on multi-convolution kernel residual network |
| US20220076003A1 (en) * | 2020-09-04 | 2022-03-10 | Hitachi, Ltd. | Action recognition apparatus, learning apparatus, and action recognition method |
| CN112861929A (en) * | 2021-01-20 | 2021-05-28 | 河南科技大学 | Image classification method based on semi-supervised weighted migration discriminant analysis |
Non-Patent Citations (1)
| Title |
|---|
| 顾昊昱: "基于流形学习的过程监测方法研究", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑)》, no. 3, 15 March 2022 (2022-03-15), pages 029 - 418 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103429150B (en) | For monitoring the monitoring device of physiological signal | |
| Achour et al. | Unsupervised automated monitoring of dairy cows’ behavior based on Inertial Measurement Unit attached to their back | |
| JP5848551B2 (en) | Learning device, learning device control method, detection device, detection device control method, and program | |
| CN108814618B (en) | Motion state identification method and device and terminal equipment | |
| KR101796055B1 (en) | Method and device for monitoring brain status by constructing multiple brain networks | |
| KR20170107778A (en) | The method and system for diagnosing skin disease | |
| CA2992368C (en) | System and methods for adaptive noise quantification in dynamic biosignal analysis | |
| CN107428495A (en) | More camera device load estimations | |
| Cong Phi Khanh et al. | The new design of cows' behavior classifier based on acceleration data and proposed feature set | |
| Songdechakraiwut et al. | Topological learning and its application to multimodal brain network integration | |
| KR20190000195A (en) | Management System for Treatment of Neurological Disorder and Method thereof | |
| US11494702B2 (en) | Task estimation method and information processing apparatus | |
| CN114495241A (en) | Image identification method and device, electronic equipment and storage medium | |
| JP2008086479A (en) | Physical activity measuring system | |
| CN116115239B (en) | Embarrassing working gesture recognition method for construction workers based on multi-mode data fusion | |
| CN115251904A (en) | Analysis method and device based on multiple motion sensors | |
| US20090012921A1 (en) | Method for identifying a person's posture | |
| CN113171102A (en) | ECG data classification method based on continuous deep learning | |
| CN117898707A (en) | Gait recognition method and system for lumbar disc herniation based on inertial motion unit | |
| KR101483218B1 (en) | Activity diagnosis apparatus | |
| CN117281061A (en) | Livestock weight measurement method, device, system and computer equipment | |
| KR20230029547A (en) | Method and device for determining object detections of an image | |
| Iqbal et al. | Brain hemorrhage detection using hybrid machine learning algorithm | |
| CN112784890A (en) | Human body falling detection method and system by means of multi-sensor fusion | |
| CN114444603B (en) | Adaptive gait division method and device, storage medium and terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |