CN108710822A - Personnel falling detection system based on infrared array sensor - Google Patents
Personnel falling detection system based on infrared array sensor Download PDFInfo
- Publication number
- CN108710822A CN108710822A CN201810300365.4A CN201810300365A CN108710822A CN 108710822 A CN108710822 A CN 108710822A CN 201810300365 A CN201810300365 A CN 201810300365A CN 108710822 A CN108710822 A CN 108710822A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- interference
- processing module
- data processing
- 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.)
- Granted
Links
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a personnel falling detection system based on an infrared array sensor, which is used for collecting temperature data by taking the infrared array sensor as a medium, processing, screening, extracting and training the data to obtain final falling information. The system comprises: the device comprises a data processing module, an interference data eliminating module, a feature extracting and processing module and a feature training module. The data processing module is used for processing initial data, differential data, separating foreground and background, and determining the specific position and the retention time of a person; the interference data elimination module finds out interference information elimination interference data through S-G filtering and interference characteristics; the feature extraction and processing module extracts four most obvious features of falling through methods such as center segmentation, clustering and the like; and the feature training module utilizes a random forest algorithm to train to obtain a final detection result. The invention processes each stage of infrared detection, and has high detection accuracy, low false alarm rate, low cost and wide application range.
Description
Technical field
The present invention relates to Activity recognition technical field, more particularly to a kind of fall detection system based on infrared sensor array
System.
Background technology
Currently, China's aging is on the rise.According to statistics, 60 years old 2015 and the above population reach 2.22 hundred million, account for total people
The 16.15% of mouth.The year two thousand twenty is expected, elderly population reach 2.48 hundred million.Simultaneously, hospital nursing staffs shortage, high number of row
The treatment that inconvenient patient cannot get proprietary Health care staff is moved, cannot be ensured safely.The either nurse problem of Empty nest elderly,
Or the treatment problem of handicapped personnel, their safety problem should obtain ours note that especially personnel fall after
It cannot timely give treatment to, often result in serious consequence.Therefore it is necessary to be paid attention to and developed for personnel's fall detection.
Learn that Chinese Patent Application No. is through carrying out retrieval to existing literature：201710287706.4 entitled：A kind of base
In the Falls Among Old People detection method of smart mobile phone；Technical solution recorded in it be using smart mobile phone to Falls Among Old People act into
Row acquisition process needs personnel to wear mobile phone in real time although avoiding the integration of various sensors, and threshold test
Method adaptive capacity to environment is low.Chinese Patent Application No. is：201621329569.3 entitled：A kind of human body fall detection dress
It sets；Although technical solution recorded in it is to be detected using infrared sensor array, but simple infrared image is not
The tumble situation of target can accurately be analyzed, varying environment temperature can not accurately only be judged according to fixed heat source pixel map
Tumble situation when degree transformation.
Invention content
Present invention aims at provide it is a kind of using infrared sensor array into administrative staff's temperature acquisition, each stage progress
Algorithm Analysis and data processing are finally extracted feature, are obtained training result automatic identification human behavior using random forests algorithm
It acts, to reach personnel's fall detection system of automatic detection tumble purpose.
To achieve the above object, following technical scheme is used：System of the present invention includes data acquisition module, data
Processing module, interference data exclude module, characteristic extracting module and data training module；
Data acquisition module is connected with data processing module, and data acquisition module handles data by sensor collection, obtains
Tumble information is simultaneously transmitted to data processing module by the quiescent time fallen to human body and position；Data processing module and interference number
It is connected according to module data collection module is excluded, data processing module is transmitted to interference data row after being pre-processed to tumble information
Except module；Interference data exclude module and are connect with characteristic extracting module data processing module, are used for further exclusive PCR data
It influences, solves the problems, such as that the sensor response time does not catch up with detection speed and data is caused to judge by accident when movement is too fast；Feature extraction mould
Block excludes module with interference data and is connected, and for analyzing the data after exclusive PCR, extracts the key feature of tumble；Number
It is connected with characteristic extracting module according to training module, the information that the characteristic extracting module that data training module receives transmits, to extraction
The algorithm of characteristic use random forest is trained, and obtains final result, determines whether the state of " falling ", is obtained and is most terminated
Fruit.
Further, the data processing module includes differential data module and track determining module；
Differential data module is used for the information that receiving sensor transmits, and the data received are carried out difference processing, removal
Environmental Heat Source interferes；It,BtRespectively present frame and background frames image, T are foreground gray threshold；Work as Nt=|It(x,y)-Bt(x,
y)|>Retain N when Tt, background data zero setting.
Track determining module is connected with differential data module, for substantially determining personnel's estimation and quiescent time.
Maximum temperature during track determining module is determined per frame data, uses ThIt indicates；And utilize ThFind the specific coordinate of track；Position
Abscissa is ThWith 8 quotient, ordinate ThWith 8 remainder and work as ThResidence time be more than threshold time TIt stopsWhen, then it is assumed that
There is stationary state in human body.
Further, it includes filter module, interference characteristic extraction module, interference elimination mould that the interference data, which exclude module,
Block：
Filter module is connected with data processing module, and the data transmitted to data processing module using S-G filtering are filtered
Wave processing, is used for smooth data plot, removes influence of noise；
Interference characteristic extraction module is connected with filter module, by extracting different characteristic, using peak away from kurtosis and detection
Elemental area size shared by object extracts data distracter；
Interference elimination module is connected with interference characteristic extraction module, and whether the interference characteristic for excluding to detect rejects；
Data screening is carried out if detecting, retains normal data.
Further, testing staff's elemental area size detection utilizes cluster and Da-Jin algorithm phase in the characteristic extracting module
In conjunction with method, by environment temperature, human body radiation temperature and human body temperature are polymerized to three classes, find the highest a kind of i.e. people of temperature
Temperature recycles Da-Jin algorithm to carry out size detection, judges size；
Wherein in Da-Jin algorithm：T=Max[w0*w1*(u0-u1)2]
In formula, T is all max-thresholds being worth to of traversal, i.e. the separation threshold value of foreground and background；
w0：Foreground pixel point accounts for the ratio of entire image size；
u0：The average value of foreground pixel point；
w1：Background pixel point accounts for the ratio of entire image size；
u0：The average value of background pixel point；
Wherein, in clustering method：
In formula, xiRepresent the temperature value of treated i-th of sample data；
μkRepresent the position of centre of gravity of k-th of class；
K representatives need sample to be polymerized to several classes, and k is 3 here；
J represents the sum of each class distortion degree.The distortion degree of each class is equal to such center of gravity and its internal members position
The quadratic sum of distance, when making J minimums, cluster is completed.
Further, the characteristic extracting module extraction interference data exclude the normal data that module transfer comes；Pass through
To static arrival time, it is layered the time used, size and static four features of first three area summation are extracted, obtained
To training characteristics data.
Compared with prior art, the invention has the advantages that：It is analyzed by detecting each process that human body is fallen
Handle data, so as to get data it is more practical, it is as a result more accurate；Make choosing by the feature extraction and training of random forest
The feature taken is more quickly accurate；Conventionally employed threshold value is effectively avoided by the method for the combination of cluster and Da-Jin algorithm algorithm
Error existing for method, and it is suitable for various different temperatures, breach the limitation of different threshold values under traditional different temperatures；Using red
Outer sensing array detection, facilitate it is small and exquisite, it is easy to operate, protect privacy, do not restricted by environment and spatially.
Description of the drawings
Fig. 1 is the system structure composition schematic diagram of the embodiment of the present invention；
Fig. 2 is the schematic diagram of data processing module in Fig. 1；
Fig. 3 is the schematic diagram of interference elimination module in Fig. 1；
Fig. 4 is exclusive PCR data flowchart disclosed by the embodiments of the present invention；
Fig. 5 is the defect area discriminant flow chart disclosed by the embodiments of the present invention for clustering and being combined with Da-Jin algorithm.
Specific implementation mode
The present invention will be further described below in conjunction with the accompanying drawings：
As shown in Figure 1, present system includes mainly：Data processing module 1, interference data exclude module 2, feature extraction
Module 3 and data training module 4, wherein：Data processing module 1 is responsible for processing and collects data.It interferes data to exclude module 2 to be responsible for
Removal interference data, characteristic extracting module 3 are responsible for extraction tumble feature, and data training module 4 is responsible for training data, is fallen
Information.
Data processing module 1 is connected with sensor, the data that receiving sensor transmits, and will treated that information is transmitted to is dry
It disturbs data and excludes module 2.As shown in Fig. 2, the module finds the time of human body tumble by being pre-processed to sensing data
And position, then analyzed for specific tumble time location.
Data processing module 1 includes：Differential data module 1-1 and track determining module 1-2；
Differential data module 1-1, is connected with sensor；The information transmitted for receiving sensor, and the data that will be received
Carry out difference processing, removal Environmental Heat Source interference.It,BtRespectively present frame and background frames image, T are foreground gray threshold；When
Nt=|It(x,y)-Bt(x,y)|>Retain N when Tt, background data zero setting.
Track determining module 1-2 is connected with differential data module 1-1；For substantially determining personnel motion trail and static
Time.Maximum temperature position during track determination sub-module is determined per frame data, uses ThIt indicates；And utilize ThFind the tool of track
Body coordinate；Position abscissa is ThWith 8 quotient, ordinate ThWith 8 remainder and work as ThResidence time be more than threshold time
TIt stopsWhen, then it is assumed that there is stationary state in human body.Such as ThMore than TIt stopsTime, it is 45 to obtain differential data at this time, then at this time
Coordinate just be (5,5).Just only with analysis (5,5) and its neighbouring pixel when our detection datas.
Interference data exclude module 2 and are connected with data processing module 1, are used for further exclusive PCR data influence.Such as Fig. 3
Shown, when which mainly solves mobile too fast, the sensor response time does not catch up with detection speed and data erroneous judgement is caused to ask
Topic.
Interference data exclude module 2：Filter module 2-1, interference characteristic extraction module 2-2 and interference elimination module
2-3；
Filter module 2-1 is connected with data processing module 1, since sensor detects actual temperature wave between ± 0.5 DEG C
Dynamic, the data that detect are it is possible that burr phenomena, to generate error.Influence in order to avoid error to experimental data,
The data transmitted to data processing module using S-G filtering are filtered, and keep the trend of original curve and its data true
Real value, smooth data plot eliminate influence of noise.
Interference characteristic extraction module 2-2 is connected with filter module 2-1, this module mainly utilizes abnormal data feature, extraction
Abnormal data.Under normal circumstances, since the sensor response time limits, mobile too fast action can not be detected, the data are avoided
To falling below, training has an impact, it is necessary to exclude abnormal data by the module.By extract different characteristic, using peak away from
(when abnormal data and normal data detect personnel, when there is peak value, peak is away from smaller), kurtosis (abnormal data with it is normal
When Data Detection is to personnel, kurtosis is larger) and detection object shared by elemental area size it is (abnormal since the response time limits
Often area of detection is excessive for data), extract data distracter.
Interference elimination module 2-3 is connected with interference characteristic extraction module 2-2, and whether the interference characteristic for excluding to detect
It rejects；Data screening is carried out if detecting, retains normal data.Detailed process is as shown in Figure 4.
Characteristic extracting module 3 excludes module 2 with interference data and is connected, by static arrival time, being layered the time used,
Size and first three static area summation this four features extract, and obtain the best characteristic of training effect.
And the method for often using threshold value when traditional contour of object size using infrared detection, although having reached certain
Effect, but fixed threshold value can not be suitable for the four seasons and sooner or later different environment temperatures.Face in the characteristic extracting module
Product size detection has abandoned traditional threshold detection method, is replaced using the method that cluster and Da-Jin algorithm are combined.Such as Fig. 5 institutes
Show, by environment temperature, human body radiation temperature and human body temperature are polymerized to three classes, find the highest a kind of i.e. human body temperature of temperature,
It recycles Da-Jin algorithm to carry out size detection, judges size.
Wherein in Da-Jin algorithm：T=Max[w0*w1*(u0-u1)2]
In formula, T is all max-thresholds being worth to of traversal, i.e. the separation threshold value of foreground and background；
w0：Foreground pixel point accounts for the ratio of entire image size
u0：The average value of foreground pixel point
w1：Background pixel point accounts for the ratio of entire image size
u0：The average value of background pixel point
Wherein in clustering method：
In formula, xiRepresent the temperature value of treated i-th of sample data；
μkRepresent the position of centre of gravity of k-th of class；
K representatives need sample to be polymerized to several classes, and k is 3 here；
J represents the sum of each class distortion degree.The distortion degree of each class is equal to such center of gravity and its internal members position
The quadratic sum of distance,
Data training module 4 receives the information that characteristic extracting module 3 transmits, the calculation to the characteristic use random forest of extraction
Method is trained, and obtains final result, determines whether the state of " falling ".
It is 60 ° that the infrared sensor array of the present embodiment, which uses the AMG8853 of panasonic companies, angular field of view,；Most
Big detecting distance is 7 meters；Use environment is -20 DEG C to 80 DEG C；8x8IR grid arrays (64 pixel) export.Meet a certain range
Detection environmental requirement.
The meaning of random forest training algorithm is in the present invention, falls it is possible that various features, and we can not
Differentiate that feature is more advantageous to by the method for expertise and detects this state of falling.Random forests algorithm can show instruction
Experienced each feature for training result importance size, therefore we have chosen represent the most tumble state feature carry out
Training, and given up the smaller feature of other importance, reduce training experiment data training amount, the experiment essence of raising
Degree.
Embodiment described above is only that the preferred embodiment of the present invention is described, not to the model of the present invention
It encloses and is defined, under the premise of not departing from design spirit of the present invention, technical side of the those of ordinary skill in the art to the present invention
The various modifications and improvement that case is made should all be fallen into the protection domain of claims of the present invention determination.
Claims (5)
1. a kind of personnel's fall detection system based on infrared sensor array, which is characterized in that the system comprises data to adopt
Collect module, data processing module, interference data and excludes module, characteristic extracting module and data training module；
Data acquisition module is connected with data processing module, and data acquisition module handles data by sensor collection, obtains people
Tumble information is simultaneously transmitted to data processing module by the quiescent time of body tumble and position；Data processing module and interference data row
Except module data collection module is connected, data processing module is transmitted to interference data and excludes mould after being pre-processed to tumble information
Block；Interference data exclude module and are connect with characteristic extracting module data processing module, are used for further exclusive PCR data influence,
The sensor response time does not catch up with detection speed and data is caused to judge by accident when solving the problems, such as mobile too fast；Characteristic extracting module with it is dry
It disturbs data exclusion module to be connected, for analyzing the data after exclusive PCR, extracts the key feature of tumble；Data are trained
Module is connected with characteristic extracting module, the information that the characteristic extracting module that data training module receives transmits, to the feature profit of extraction
It is trained with the algorithm of random forest, obtains final result, determine whether the state of " falling ", obtain final result.
2. a kind of personnel's fall detection system based on infrared sensor array according to claim 1, it is characterised in that：
The data processing module includes differential data module and track determining module；
Differential data module is used for the information that receiving sensor transmits, and the data received are carried out difference processing, removes environment
Heat source interferes；
Track determining module is connected with differential data module, for substantially determining personnel's estimation and quiescent time.
3. a kind of personnel's fall detection system based on infrared sensor array according to claim 1, it is characterised in that：
It includes filter module, interference characteristic extraction module, interference elimination module that the interference data, which exclude module,：
Filter module is connected with data processing module, and place is filtered to the data that data processing module transmits using S-G filtering
Reason is used for smooth data plot, removes influence of noise；
Interference characteristic extraction module is connected with filter module, by extracting different characteristic, using peak away from kurtosis and detection object
Shared elemental area size extracts data distracter；
Interference elimination module is connected with interference characteristic extraction module, and whether the interference characteristic for excluding to detect rejects；If inspection
It measures, carries out data screening, retain normal data.
4. a kind of personnel's fall detection system based on infrared sensor array according to claim 1, it is characterised in that：
The method that testing staff's elemental area size detection is combined using cluster and Da-Jin algorithm in the characteristic extracting module, by environment
Temperature, human body radiation temperature and human body temperature are polymerized to three classes, find the highest a kind of i.e. human body temperature of temperature, recycle big Tianjin
Method carries out size detection, judges size；
Wherein in Da-Jin algorithm：T=Max[w0*w1*(u0-u1)2]
In formula, T is all max-thresholds being worth to of traversal, i.e. the separation threshold value of foreground and background；
w0：Foreground pixel point accounts for the ratio of entire image size；
u0：The average value of foreground pixel point；
w1：Background pixel point accounts for the ratio of entire image size；
u0：The average value of background pixel point；
Wherein, in clustering method：
In formula, xiRepresent the temperature value of treated i-th of sample data；
μkRepresent the position of centre of gravity of k-th of class；
K representatives need sample to be polymerized to several classes；
J represents the sum of each class distortion degree.The distortion degree of each class is equal to such center of gravity and its internal members' positional distance
Quadratic sum, when making J minimums, cluster complete.
5. a kind of personnel's fall detection system based on infrared sensor array according to claim 1, it is characterised in that：
The characteristic extracting module extraction interference data exclude the normal data that module transfer comes；By to static arrival time, dividing
Time used in layer, size and static four features of first three area summation extract, and obtain training characteristics data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300365.4A CN108710822B (en) | 2018-04-04 | 2018-04-04 | Personnel falling detection system based on infrared array sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300365.4A CN108710822B (en) | 2018-04-04 | 2018-04-04 | Personnel falling detection system based on infrared array sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108710822A true CN108710822A (en) | 2018-10-26 |
| CN108710822B CN108710822B (en) | 2022-05-13 |
Family
ID=63867064
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810300365.4A Active CN108710822B (en) | 2018-04-04 | 2018-04-04 | Personnel falling detection system based on infrared array sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108710822B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111562011A (en) * | 2020-04-21 | 2020-08-21 | 上海上实龙创智慧能源科技股份有限公司 | Non-video human body temperature monitoring system |
| CN112212980A (en) * | 2020-09-29 | 2021-01-12 | 中电工业互联网有限公司 | Human body temperature detection method capable of resisting external environment temperature interference |
| CN112244818A (en) * | 2020-09-30 | 2021-01-22 | 仲恺农业工程学院 | Falling detection device and method based on human body infrared perception |
| CN112613388A (en) * | 2020-12-18 | 2021-04-06 | 燕山大学 | Personnel falling detection method based on multi-dimensional feature fusion |
| CN114533040A (en) * | 2022-01-12 | 2022-05-27 | 北京京仪仪器仪表研究总院有限公司 | Method for monitoring specific activity of personnel in fixed space |
| CN114627618A (en) * | 2022-03-30 | 2022-06-14 | 成都理想科技开发有限公司 | Method for detecting falling of old people and giving alarm |
| CN115778374A (en) * | 2022-10-14 | 2023-03-14 | 广东工业大学 | Fall detection method and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722715A (en) * | 2012-05-21 | 2012-10-10 | 华南理工大学 | Tumble detection method based on human body posture state judgment |
| CN103325080A (en) * | 2013-06-21 | 2013-09-25 | 电子科技大学 | Gerocamium intelligent nursing system and method based on Internet of Things technology |
| CN203931100U (en) * | 2013-12-30 | 2014-11-05 | 杨松 | The terminal that human body is fallen |
| JP2016067641A (en) * | 2014-09-30 | 2016-05-09 | 東京エレクトロンデバイス株式会社 | Fall detection processing apparatus and fall detection system |
| CN106228200A (en) * | 2016-10-17 | 2016-12-14 | 中北大学 | A kind of action identification method not relying on action message collecting device |
| CN106407996A (en) * | 2016-06-30 | 2017-02-15 | 华南理工大学 | Machine learning based detection method and detection system for the fall of the old |
| CN106503667A (en) * | 2016-10-26 | 2017-03-15 | 太原理工大学 | A kind of based on WISP and the fall detection method of pattern recognition |
| CN107180511A (en) * | 2017-05-11 | 2017-09-19 | 南京理工大学 | A kind of detection of Falls Among Old People and prior-warning device and method |
-
2018
- 2018-04-04 CN CN201810300365.4A patent/CN108710822B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722715A (en) * | 2012-05-21 | 2012-10-10 | 华南理工大学 | Tumble detection method based on human body posture state judgment |
| CN103325080A (en) * | 2013-06-21 | 2013-09-25 | 电子科技大学 | Gerocamium intelligent nursing system and method based on Internet of Things technology |
| CN203931100U (en) * | 2013-12-30 | 2014-11-05 | 杨松 | The terminal that human body is fallen |
| JP2016067641A (en) * | 2014-09-30 | 2016-05-09 | 東京エレクトロンデバイス株式会社 | Fall detection processing apparatus and fall detection system |
| CN106407996A (en) * | 2016-06-30 | 2017-02-15 | 华南理工大学 | Machine learning based detection method and detection system for the fall of the old |
| CN106228200A (en) * | 2016-10-17 | 2016-12-14 | 中北大学 | A kind of action identification method not relying on action message collecting device |
| CN106503667A (en) * | 2016-10-26 | 2017-03-15 | 太原理工大学 | A kind of based on WISP and the fall detection method of pattern recognition |
| CN107180511A (en) * | 2017-05-11 | 2017-09-19 | 南京理工大学 | A kind of detection of Falls Among Old People and prior-warning device and method |
Non-Patent Citations (7)
| Title |
|---|
| CHEN JINGJING等: "Comparative Study on Automatic Lung Parenchyma Segmentation of CT Data Using Improved OTSU and FCM Methods", 《航天医学与医学工程》 * |
| WEI-HAN CHEN等: "A Fall Detection System Based on Infrared Array Sensors with Tracking Capability for the Elderly at Home", 《2015 17TH INTERNATIONAL CONFERENCE ON E-HEALTH NETWORKING, APPLICATION & SERVICES》 * |
| YUN LI等: "Improving automatic sound-based fall detection using iVAT clustering and GA-based feature selection", 《34TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE EMBS》 * |
| 吴科艳等: "基于邻域一致性和DBPSO的跌倒检测特征集优化算法", 《计算机与现代化》 * |
| 张祥军: "基于iOS平台的人体跌倒监测***的设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 杨任兵等: "红外图像中基于多特征提取的跌倒检测算法研究", 《红外技术》 * |
| 王宝玉: "运动人体目标跟踪及异常行为识别", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111562011A (en) * | 2020-04-21 | 2020-08-21 | 上海上实龙创智慧能源科技股份有限公司 | Non-video human body temperature monitoring system |
| CN112212980A (en) * | 2020-09-29 | 2021-01-12 | 中电工业互联网有限公司 | Human body temperature detection method capable of resisting external environment temperature interference |
| CN112244818A (en) * | 2020-09-30 | 2021-01-22 | 仲恺农业工程学院 | Falling detection device and method based on human body infrared perception |
| CN112613388A (en) * | 2020-12-18 | 2021-04-06 | 燕山大学 | Personnel falling detection method based on multi-dimensional feature fusion |
| CN114533040A (en) * | 2022-01-12 | 2022-05-27 | 北京京仪仪器仪表研究总院有限公司 | Method for monitoring specific activity of personnel in fixed space |
| CN114533040B (en) * | 2022-01-12 | 2024-04-09 | 北京京仪仪器仪表研究总院有限公司 | Method for monitoring specific activity of personnel in fixed space |
| CN114627618A (en) * | 2022-03-30 | 2022-06-14 | 成都理想科技开发有限公司 | Method for detecting falling of old people and giving alarm |
| CN114627618B (en) * | 2022-03-30 | 2024-02-06 | 成都理想科技开发有限公司 | Method for detecting falling of old people and giving alarm |
| CN115778374A (en) * | 2022-10-14 | 2023-03-14 | 广东工业大学 | Fall detection method and system |
| CN115778374B (en) * | 2022-10-14 | 2023-07-04 | 广东工业大学 | Fall detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108710822B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108710822A (en) | Personnel falling detection system based on infrared array sensor | |
| CN111209848B (en) | Real-time falling detection method based on deep learning | |
| CN102302370B (en) | Method and device for detecting tumbling | |
| CN108960056A (en) | A kind of fall detection method based on posture analysis and Support Vector data description | |
| CN109215806A (en) | A kind of public place health monitoring systems and method based on recognition of face | |
| CN110378179B (en) | Subway ticket evasion behavior detection method and system based on infrared thermal imaging | |
| CN106203274A (en) | Pedestrian's real-time detecting system and method in a kind of video monitoring | |
| CN107392885A (en) | A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism | |
| CN108564577A (en) | Solar cell segment grid defect inspection method based on convolutional neural networks | |
| CN104269025B (en) | Wearable single node feature and the position choosing method of monitoring is fallen down towards open air | |
| CN112581450B (en) | Pollen detection method based on expansion convolution pyramid and multi-scale pyramid | |
| CN108509938A (en) | A kind of fall detection method based on video monitoring | |
| CN116602663B (en) | Intelligent monitoring method and system based on millimeter wave radar | |
| CN112183438B (en) | Image identification method for illegal behaviors based on small sample learning neural network | |
| CN109461163A (en) | A kind of edge detection extraction algorithm for magnetic resonance standard water mould | |
| CN107563364A (en) | The discriminating conduct of the fingerprint true and false and fingerprint identification method based on sweat gland | |
| CN108764251A (en) | A kind of workpiece surface crack detection method and system | |
| CN109003275A (en) | The dividing method of weld defect image | |
| CN115953666B (en) | Substation site progress identification method based on improved Mask-RCNN | |
| CN116563887B (en) | Sleeping posture monitoring method based on lightweight convolutional neural network | |
| CN111753877B (en) | Product quality detection method based on deep neural network migration learning | |
| CN114358194A (en) | Gesture tracking based detection method for abnormal limb behaviors of autism spectrum disorder | |
| CN112801000A (en) | Household old man falling detection method and system based on multi-feature fusion | |
| CN113947796A (en) | Human body temperature trend detection method and device based on identity recognition | |
| CN112132157A (en) | Raspberry pie-based gait face fusion recognition method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |