CN103501330B - A kind of Weigh sensor processing method for personal monitoring - Google Patents
A kind of Weigh sensor processing method for personal monitoring Download PDFInfo
- Publication number
- CN103501330B CN103501330B CN201310455390.7A CN201310455390A CN103501330B CN 103501330 B CN103501330 B CN 103501330B CN 201310455390 A CN201310455390 A CN 201310455390A CN 103501330 B CN103501330 B CN 103501330B
- Authority
- CN
- China
- Prior art keywords
- data
- value
- user
- data set
- recognition result
- 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.)
- Expired - Fee Related
Links
Landscapes
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
A kind of Weigh sensor processing method for personal monitoring, the data gathered including input, the data that each user gathers under certain period respectively constitute a data set;Each data set is carried out respectively feature identification process, and returns recognition result as early warning foundation.The realization that any data group carries out feature identification process includes extracting the data that feature identification processes, comprehensively obtain user's real-time status value M, ask for historic state meansigma methods according to current User Status parameter X, if judging whether M meets to be unsatisfactory for, marker recognition result is abnormal, if meeting, marker recognition result is normal, adds up ratio modified chi value according to recognition result during iteration carries out feature identification process.The present invention can be identified in conjunction with the multiple data of user, avoid the deviation that single data cause, identification process is iterated measuring and calculating for the valid data in data set, makes up user error operation or monitoring leak and the error that causes, intelligent strengthens recognition effect.
Description
Technical field
The present invention relates to Computer Applied Technology field, a kind of Weigh sensor processing method for personal monitoring.
Background technology
Using sensor device to gather monitoring data is current and following wearable computing and the main development side of intelligent endowment
To, can be with situations such as monitoring management personal health, behavior, living habit, it was predicted that its integrality trend, get rid of the most ahead of time
Accident occurs.At present at personal monitoring management aspect, the many employings of China carry out what single real-time data acquisition extracted at user side
Method.Based on the single data monitoring method that presently, there are, can not accurately judge that gathered person's is whole by single data target
Body situation, urgently there is more advanced solution in this area.
Summary of the invention
The present invention discloses a kind of Weigh sensor processing method for personal monitoring, and support processes the data of large-scale consumer crowd
Feature monitoring and identification.
The technical scheme is that a kind of Weigh sensor processing method for personal monitoring, comprise the steps:
Step 1, the data that input gathers, the data that each user gathers under certain period respectively constitute a data set;
Step 2, each data set obtaining step 1 carries out feature identification process respectively, and returns recognition result and depend on as early warning
According to;The realization that the data set of any user carries out feature identification process includes following sub-step,
Step 2.1, extracts, from the data set of present period, the data that feature identification processes, including current body physiological value H,
Movable value V and geographical location information P;
Step 2.2, according to comprehensively obtaining real-time status value M of user with drag,
M=aH+bV+cP
Wherein, a, b, c are respectively weighted value shared by data in data set, and a, b, c sum is 1;
Step 2.3, takes the body physiological value of family multiple time period, movable value and geographical location information value, obtains body physiological
Meansigma methodsActivity meansigma methodsGeographical location information meansigma methods
Step 2.4, based on the model in step 2.2, willSubstitution obtains state meansigma methods
Step 2.5, sets current User Status parameter X, it is judged that whether M meetsFirst execution
During step 2.5, User Status parameter X uses default value,
If being unsatisfactory for, marker recognition result is abnormal, update anomalies result number of times Rfalse=Rfalse+1;
If meeting, marker recognition result is normal, updates normal outcome number of times Rtrue=Rtrue+1;
Step 2.6, works as RfalseOr RtrueWhen being 0, keep X value constant, work as RfalseOr RtrueWhen being not 0, modified chi value,
Realize by following equation,
Wherein,Being revised X value, Y is that recognition result adds up ratio,
Step 2.7, returns step 2.1, proceeds to process according to the data set of subsequent period.
And, use distributed parallel mode that the data set of each user is carried out feature identification process respectively.
The present invention can be identified in conjunction with the multiple data of user, it is to avoid the monitoring result error that single data cause, and identified
Journey is iterated measuring and calculating for the valid data in data set, makes up user error operation or monitoring leak and the error that causes,
Strengthen recognition effect.The data extracted in all data sets are grouped by the technical solution adopted in the present invention according to ID,
The identification processing procedure of each user is relatively independent, is suitable to system and carries out United Dispatching, distributes to different calculating resources and processes,
Last only need to be particularly well-suited to utilize distributed platform to carry out at large-scale discriminatory analysis by effective recognition result record and return
Reason.
Accompanying drawing explanation
Fig. 1 is the flow chart of the embodiment of the present invention.
Detailed description of the invention
The present invention proposes the Weigh sensor processing method for personal monitoring, describes this in detail below in conjunction with drawings and Examples
Inventive technique scheme.
The present invention can use computer software technology to realize automatic operational process.Seeing Fig. 1, the flow process of embodiment includes following step
Rapid:
Step 1, input gathers data.Each user can be gathered data respectively, teletransmission to service end is concentrated and is identified
Process.Service end can use Distributed Computing Platform technology, the unified data to all users are managed and carry out identifying processing.
User can be carried out ID distribution, as the identification code of mark Data Source, under same No. ID, the data of same period constitute one
Data set.Gather data and can include many data, concrete gather realization can use existing sensor, GPS locating module,
The equipment such as gravity sensor.Those skilled in the art can sets itself time period length, the most every 10 minutes gather and once obtain
Data constitute a data set.
Step 2, each data set obtaining step 1 carries out feature identification process respectively, and returns recognition result and depend on as early warning
According to;The data set of any user is carried out the realization of feature identification process and includes following sub-step:
Step 2.1, extracts the data that feature identification processes.In embodiment, gather and be used for identify data include body physiological
Value (such as pulse), movable value (such as walking step number), the geographical location information value distance of address (such as with), use body
Body biological value, activity information summary consider to judge the reliability of gained body physiological value, and reference geographical location determines emergency situations
Or whether there is exception.When being embodied as, those skilled in the art can sets itself for the kind of data analyzed, gather institute
After data are transferred to service end, by data all in data set according to its different characteristic identification extraction.
If the body physiological value currently collected is H, movable value is V, and geographical location information value is P.
Step 2.2, if M(H, V, P) represent the real-time status value of this ID user, it is called for short M, according to comprehensive with drag
Arrive,
M=aH+bV+cP
Wherein, a, b, c are respectively weighted value shared by data in data set, and a, b, c sum is 1, can use empirical value, such as a=0.5,
B=0.3, c=0.2.
This model is body physiological value H, movable value V and the comprehensive utilization of geographical location information value P, may during actual application
Occurring having in this three item data one or binomial is not acquired, such as user is in bed, and geographical location information is without continuous collecting.
Flow process the most provided by the present invention and model still are able to application, can simplify process, make corresponding geographical location information value P
It is 0.
Step 2.3, takes the body physiological value of family multiple time period, movable value, geographical location information value are obtained body physiological and put down
AverageActivity meansigma methodsGeographical location information meansigma methods
When being embodied as, can each time period is gathered in user's current slot starts to take predetermined time period historical data
Be averaged calculating.Those skilled in the art can predetermined time period voluntarily, such as 24 hours.When performing step 2.3 first,
Body physiological meansigma methodsActivity meansigma methodsGeographical location information meansigma methodsEqual to currently collect, i.e.
Step 2.4, based on the model in step 2.2, willSubstitution obtains state meansigma methods
Step 2.5, according to current User Status parameter X, it is judged that whether M meetsThus obtain
Recognition result RiIf M is unsatisfactory for, marker recognition result is Ri=false, represents abnormal, update anomalies result number of times
Rfalse=Rfalse+ 1, prompting or warning can be issued the user with;If meeting, labelling Ri=true, represents normal, updates normal outcome
Number of times Rtrue=Rtrue+1.When performing step 2.5 first, Rfalse、RtrueInitial value be 0.
When being embodied as, the average default value as User Status parameter X of the habit statistical according to user can be used, such as
X=12.When going to step 2.3 first, X uses default value.User (such as the whole nation) can add up on a large scale,
To adjust User Status parameter X as early as possible to being adapted to individual consumer.When successive iterations goes to step 2.3, can use upper
Once perform step 2.6 gained X value.
Step 2.6, modified chi value, realize by following equation,
Wherein,Being revised X value, Y is that recognition result adds up ratio,Try to achieveAfter, order
Especially, in the stage started, it is possible that RfalseOr Rtrue=0, now X value is constant, i.e. keeps X=X.
By modified chi value, dynamically adjust according to marker recognition result and identify the regional extent size judged.
Step 2.7, returns step 2.1, proceeds to process according to the data set of subsequent period, differentiates until stopping gathering.To this
The result that secondary iteration obtains preserves, and the result that can be obtained by current iteration and historical results return to user and check.
It is identified the continuous data group obtained under same ID processing, corresponding real-time status value M can be respectively obtained.Pass through
Iteration performs above step, can repeatedly adjust the state meansigma methods obtaining constantly updatingWith the User Status parameter constantly revised
X, can make according to formulaThe recognition result Ri obtained is more accurate.
In above flow process, each threshold value can be set flexibly, control to identify sensitivity.
When being embodied as, Distributed Computing Platform can use cluster server to realize, and it is each that step 1 is obtained by Distributed Computing Platform
The data set of user is distributed, and distributed parallel carries out data characteristics identifying processing, realizes performing step to the data of any user
2.1—2.7.So can make full use of all calculating resources, support processes data characteristics monitoring and the identification of large-scale consumer crowd.
Specific embodiment described herein is only to present invention spirit explanation for example.The skill of the technical field of the invention
Described specific embodiment can be made various amendment or supplements or use similar mode to substitute, respectively by art personnel
Threshold value can be set the most as the case may be by those skilled in the art, but the spirit or surmount without departing from the present invention is determined
The scope of justice.
Claims (2)
1. the Weigh sensor processing method for personal monitoring, it is characterised in that comprise the steps:
Step 1, the data that input gathers, the data that each user gathers under certain period respectively constitute a data set;
Step 2, each data set obtaining step 1 carries out feature identification process respectively, and returns recognition result and depend on as early warning
According to;The realization that the data set of any user carries out feature identification process includes following sub-step,
Step 2.1, extracts, from the data set of present period, the data that feature identification processes, including current body physiological value H,
Movable value V and geographical location information value P;
Step 2.2, according to comprehensively obtaining real-time status value M of user with drag,
M=aH+bV+cP
Wherein, a, b, c are respectively weighted value shared by data in data set, and a, b, c sum is 1;
Step 2.3, takes the body physiological value of family multiple time period, movable value and geographical location information value, obtains body physiological
Meansigma methodsActivity meansigma methodsGeographical location information meansigma methods
Step 2.4, based on the model in step 2.2, willSubstitution obtains state meansigma methods
Step 2.5, sets current User Status parameter X, it is judged that whether M meetsFirst execution
During step 2.5, User Status parameter X uses default value, Rfalse、RtrueInitial value be 0,
If being unsatisfactory for, marker recognition result is abnormal, update anomalies result number of times Rfalse=Rfalse+1;
If meeting, marker recognition result is normal, updates normal outcome number of times Rtrue=Rtrue+1;
Step 2.6, works as RfalseOr RtrueWhen being 0, keep X value constant, work as RfalseOr RtrueWhen being not 0, modified chi value,
Realize by following equation,
Wherein,Being revised X value, Y is that recognition result adds up ratio,
Step 2.7, returns step 2.1, proceeds to process according to the data set of subsequent period.
The most according to claim 1 for the Weigh sensor processing method of personal monitoring, it is characterised in that: use distributed parallel side
Formula carries out feature identification process respectively to the data set of each user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310455390.7A CN103501330B (en) | 2013-09-29 | 2013-09-29 | A kind of Weigh sensor processing method for personal monitoring |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310455390.7A CN103501330B (en) | 2013-09-29 | 2013-09-29 | A kind of Weigh sensor processing method for personal monitoring |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103501330A CN103501330A (en) | 2014-01-08 |
CN103501330B true CN103501330B (en) | 2017-01-04 |
Family
ID=49866500
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310455390.7A Expired - Fee Related CN103501330B (en) | 2013-09-29 | 2013-09-29 | A kind of Weigh sensor processing method for personal monitoring |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103501330B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106570320A (en) * | 2016-10-25 | 2017-04-19 | 兰州大学 | Prediction system for old people behavior health |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043894A (en) * | 2009-10-13 | 2011-05-04 | 北京大学 | Physiological data monitoring system and method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9044136B2 (en) * | 2007-02-16 | 2015-06-02 | Cim Technology Inc. | Wearable mini-size intelligent healthcare system |
-
2013
- 2013-09-29 CN CN201310455390.7A patent/CN103501330B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043894A (en) * | 2009-10-13 | 2011-05-04 | 北京大学 | Physiological data monitoring system and method |
Non-Patent Citations (4)
Title |
---|
Wearable multi-parameter remote physiological monitoring system;P.S.Pandian等;《Medical Engineering & Physics》;20081231;466-477 * |
个人远程医疗监护系统的设计与实现;张博;《中国优秀硕士学位论文全文数据库》;20120715;全文 * |
多参数人体状态监护系统的研究;李亚琼;《中国优秀硕士学位论文全文数据库》;20100915;全文 * |
智能远程健康监护系统生理参数数据分析及预报的研究;蒋贤海;《中国博士学位论文全文数据库》;20120215;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN103501330A (en) | 2014-01-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Akhavian et al. | Smartphone-based construction workers' activity recognition and classification | |
Palm et al. | Mapping migratory flyways in Asia using dynamic Brownian bridge movement models | |
JP6129028B2 (en) | Energy consumption prediction method for building power equipment | |
CN106502871B (en) | The alarm threshold dynamic configuration system and method for supervisory systems | |
CN104699985A (en) | Medical big-data acquisition and analysis system and method | |
CN108197737A (en) | A kind of method and system for establishing medical insurance hospitalization cost prediction model | |
EP2914100B1 (en) | Computer system for measuring real time position of a plurality of animals | |
Arcidiacono et al. | A threshold-based algorithm for the development of inertial sensor-based systems to perform real-time cow step counting in free-stall barns | |
CN105528754A (en) | Old people information service system based on dual neural network behavior recognition model | |
CN105675038A (en) | Device for predicting faults of instruments | |
CN110363090A (en) | Intelligent heart disease detection method, device and computer readable storage medium | |
KR102265809B1 (en) | Method and apparatus for detecting behavior pattern of livestock using acceleration sensor | |
CN106725376A (en) | Sign detection method and device | |
CN113724847A (en) | Medical resource allocation method, device, terminal equipment and medium based on artificial intelligence | |
CN112837799A (en) | Remote internet big data intelligent medical system based on block chain | |
Hasib et al. | Covid-19 prediction based on infected cases and deaths of bangladesh using deep transfer learning | |
CN103501330B (en) | A kind of Weigh sensor processing method for personal monitoring | |
CN113436295B (en) | Living body culture monitoring track drawing method, device, equipment and storage medium | |
Silva et al. | Fusion of clinical, self-reported, and multisensor data for predicting falls | |
CN110136841A (en) | Disease incidence prediction technique, device and computer readable storage medium | |
Seera et al. | A hybrid FMM-CART model for human activity recognition | |
Webb et al. | Developing spatially‐explicit weighting factors to account for bias associated with missed GPS fixes in resource selection studies | |
Al Jassmi et al. | Automatic recognition of labor activity: A machine learning approach to capture activity physiological patterns using wearable sensors | |
CN113688125B (en) | Abnormal value detection method and device based on artificial intelligence, electronic equipment and medium | |
CN113724891A (en) | Hospital epidemic situation monitoring method, device and related equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170104 Termination date: 20180929 |