CN104850841B - Combination RFID and video identification a kind of old man abnormal behaviour monitoring method - Google Patents
Combination RFID and video identification a kind of old man abnormal behaviour monitoring method Download PDFInfo
- Publication number
- CN104850841B CN104850841B CN201510260138.XA CN201510260138A CN104850841B CN 104850841 B CN104850841 B CN 104850841B CN 201510260138 A CN201510260138 A CN 201510260138A CN 104850841 B CN104850841 B CN 104850841B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- image
- mtd
- background
- 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.)
- Active
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
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Closed-Circuit Television Systems (AREA)
- Alarm Systems (AREA)
Abstract
A kind of combination RFID and video identification old man's abnormal behaviour monitoring method, comprise the following steps:1) it is all monitoring regional deployment cameras, while in the visual range of each camera place a label reader, while to institute monitoring in need elders wear RFID tag, and on label storage carrier identity information;2) when label-carrier is introduced into reader region, camera gathers background image, background modeling is carried out using mixed Gauss model, and background model is constantly updated according to the newest background image collected, obtained newest background model is used for needing the picture frame for carrying out behavioral value to carry out difference, so as to obtain the foreground image of pedestrian;3) enter fashionable when reader detects label, open the unusual checking function of the region cameras, abnormal behavior detection uses the method based on template matches.The present invention realize it is round-the-clock monitoring, robustness preferably, reliability it is higher.
Description
Technical field
The invention belongs to old man's abnormal behaviour identification field, more particularly to a kind of old man's abnormal behaviour monitoring method.
Background technology
The population of more than 60 years old is generally accounted for total population proportion in the world and reaches 10%, or over-65s population accounts for total people
The proportion of mouth reaches 7% standard for entering aging society as country.According to this standard, China is since nineteen ninety-nine
Veteran form country is come into, and the ratio of aging in recent years is also increasingly being aggravated, and aging population brings many societies
Meeting problem, and with the transformation of modern young man's life and work mode, it is difficult to there is enough energy to look after old man, traditional family
Old-age provision model gradually concentrates old-age provision model to substitute by the elderly community, home for destitute etc..And for the peace of these the elderlys
Full monitoring becomes particularly important.
Video monitoring has been widely used in the daily prison for the elderly as the technical way of modern security protection
In shield, but current video monitoring system is more based on artificial monitor, it is difficult to round-the-clock effective monitoring, simultaneously for monitoring mesh
Mark lacks purpose, and it is the elderly to add guardianship, in order to which the daily life guaranteed to old man has a comprehensive prison
Shield, it is necessary to which, in more place deployment cameras, the consequence so done is exactly to cause the leakage of life of elderly person privacy, and this is
They are undesirable.
Radio frequency label (RFID) is a kind of wireless communication technology, can carry out identity knowledge to target by wireless signal
Other and data transfer, is widely used in every field in recent years, and RFID system is added in safety-protection system can be to monitoring pair
The identity information of elephant is identified, play it is autotelic be monitored, new thinking is provided for video monitoring.
The content of the invention
In order to overcome existing old man's abnormal behaviour identification method can not realize it is round-the-clock monitoring, robustness it is poor, reliable
Property poor deficiency, the invention provides it is a kind of realize round-the-clock monitoring, robustness preferably, the higher combination RFID of reliability with
Old man's abnormal behaviour monitoring method of video identification.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of combination RFID and video identification old man's abnormal behaviour monitoring method, the monitoring method include following step
Suddenly:
1) in all monitoring regional deployment cameras, read while placing a label in the visual range of each camera
Read device, while to institute monitoring in need elders wear RFID tag, and on label storage carrier identity information;
2) when label-carrier is introduced into reader region, camera is in normal monitoring state, without different
Normal behavioral value, camera collection background image, background modeling is carried out using mixed Gauss model during this period, and according to collection
The newest background image arrived constantly updates background model, and obtained newest background model is used for needing the figure of progress behavioral value
As frame carries out difference, so as to obtain the foreground image of pedestrian;
3) enter fashionable when reader detects label, the identity information for reading label storage is sent to central server, in
Central server recalls correspondence personal files from database server, and opens the unusual checking work(of the region cameras
Energy;
Abnormal behavior detection uses the method based on template matches, the old man guarded first to each needs
Carry out behavior IMAQ, gathers the image sequence of common behavior, and extracts the feature of different behaviors, is carried out with histogram of gradients
Description, ultimately forms the feature templates of different behaviors;
After camera opens unusual checking function, by the detection two field picture collected and the background mould set up before
Type carries out background difference, obtains foreground image, then with detecting that the behavioural characteristic template of target is matched in database, matches
Alarm is carried out during to abnormal behaviours such as tumbles, and be recorded in personal files;
Foreground extraction is as follows with judgment formula:
dt(x, y)=| It(x,y)-Bt(x,y)| (5)
Wherein dt(x,y)、It(x,y)、Bt(x, y) represents difference image, current frame image, the background image of t respectively
Pixel value at (x, y) place, T1For prospect decision threshold;
Assuming that detection target is A, then its template is TA1~TAh, the template of respectively A common behavior, h is common behavior
Sum, h >=5;By TAx(m, n) overlays dtTranslation search, T are carried out on (x, y)AxTo detect target A x-th of behavior correspondence
Threshold values, x represents the arbitrary value between 1~h, if dtIn (x, y) by the subgraph of template covering part be Si,j(m, n), wherein
(i, j) is subgraph top left corner pixel point in dtCoordinate in (x, y) image, M, N are respectively the maximum transverse and longitudinal coordinate value of image, use D
(i, j) represents the matching degree of template and image, and matching formula is as follows:
Wherein, T2For template matches threshold values, Si,j(m, n) and TAx(m, n) is respectively subgraph and template at (m, n) coordinate
Pixel value, when D (i, j) be less than threshold values when be then judged as that image is matched with target A x-th of behaviour template, recorded data
Storehouse, if hazardous act, then starts alarm.
Further, the step 2) in, the foundation of background model and renewal process are as follows:
If XtFor the background image of t, mixed Gauss model is set up such as to each pixel on the moment background image
Under:
Wherein, ωP, t,μP, t,The weight coefficient, average and variance of p-th of Gaussian component of t are represented respectively,The distribution function of p-th of Gaussian component of t is represented, k represents the number of Gaussian component;
Mixed Gauss model updates as follows:
ωP, t+1=(1+ α) ωP, t, (2)
μP, t+1=β μP, t+ (1- β) I (x, y, t) (3)
Wherein, α is background model weight undated parameter, and β is background model average undated parameter, and I (x, y, t) is image I
Gray value at pixel (x, y) place;
Each pixel is set up mixed Gauss model formula (1) and Gauss model newer (2), formula (3) on background image
Afterwards, in t, the mean μ and variances sigma of each pixel point model gray value are finally obtained2.In t, for a given pair
Testing image U, U (x, y) are gray values of the image U at pixel (x, y) place, with each pixel U (x, y) and corresponding moment t
Background image pixels point matched, think that the match is successful if below equation is met, (x, y) be judged as background dot;
Otherwise, (x, y) is foreground point, and matching formula is as follows:
Wherein, σ2(x, y) is variance of the adaptation function at point (x, y) place, and T is default threshold value.
Further, center control platform not display monitoring picture, when camera opens abnormal behavior detection function, in
Heart control platform detects the behavior state of object according to the result display of template matches during detection, only when the abnormal row of system detectio
For and when being alarmed, center control platform just opens on-site supervision picture.
Further, the step 1) in, the personal information of all carrier is stored in database server;It is described
Step 3) in, by all personal letters for gathering and being stored in after the feature templates to be formed coding in database server to old man's behavior
Cease in archives.
Beneficial effects of the present invention are mainly manifested in:1st, the present invention is combined using RFID tag with video monitoring
Method, is detected according to the appearance situation of target, greatly reduces the mistake that the detection and environmental change of hash are brought
Sentence, while saving overhead;2nd, the present invention picture frame of collection without detection target when camera is without behavioral value
Carry out the renewal of background model, it is possible to reduce influence of the environmental change to system robustness;3rd, the present invention is by behavioral value and examines
The personal information for surveying object is combined, and gathering behavior sequence extraction feature templates respectively for different guardianships is stored in individual
In archives, then according to the difference of detection object, template matches, judgement pair are carried out to image to be detected with different feature templates
As behavior, the erroneous judgement brought by Different Individual using unified behavioural characteristic template can be effectively reduced, while can be formed more
For perfect personal files, the reference of medical monitoring etc. is used as;4th, center control platform is only detecting old man's behavior
On-site supervision picture is just opened when occurring abnormal, it is possible to prevente effectively from guarding the influence that privacy of being lived to old man is brought comprehensively.
Brief description of the drawings
Fig. 1 is the flow chart of old man's abnormal behaviour monitoring method with reference to RFID and video identification.
Fig. 2 is the Organization Chart of monitoring system.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings.
Referring to Figures 1 and 2, a kind of old man's abnormal behaviour monitoring method of combination RFID and video identification, including following step
Suddenly
1) as shown in Fig. 2 in all monitoring regional deployment cameras, while being placed in the visual range of each camera
One label reader, at the same to monitoring in need elders wear RFID tag, and on label store carrier body
Part information, the personal information of all carrier is stored in database server.
2) when label-carrier is introduced into reader region, camera is in normal monitoring state, without different
Normal behavioral value, camera collection background image, background modeling is carried out using mixed Gauss model during this period, and according to collection
The newest background image arrived constantly updates background model, and obtained newest background model is used for needing the figure of progress behavioral value
As frame carries out difference, so as to obtain the foreground image of pedestrian.
The foundation of background model and renewal process are as follows:
If XtFor the background image of t, mixed Gauss model is set up such as to each pixel on the moment background image
Under:
Wherein, ωP, t,μP, t,The weight coefficient, average and variance of p-th of Gaussian component of t are represented respectively,The distribution function of p-th of Gaussian component of t is represented, k represents the number of Gaussian component;
Mixed Gauss model updates as follows:
ωP, t+1=(1+ α) ωP, t, (2)
μP, t+1=β μP, t+ (1- β) I (x, y, t) (3)
Wherein, α is background model weight undated parameter, and β is background model average undated parameter, and I (x, y, t) is image I
Gray value at pixel (x, y) place.
Each pixel is set up mixed Gauss model formula (1) and Gauss model newer (2), formula (3) on background image
Afterwards, in t, the mean μ and variances sigma of each pixel point model gray value are finally obtained2.In t, for a given pair
Testing image U, U (x, y) are gray values of the image U at pixel (x, y) place, with each pixel U (x, y) and corresponding moment t
Background image pixels point matched, think that the match is successful if below equation is met, (x, y) be judged as background dot;
Otherwise, (x, y) is foreground point, and matching formula is as follows:
Wherein, σ2(x, y) is variance of the adaptation function at point (x, y) place, and T is default threshold value, and 0.7 is chosen for here.
3) enter fashionable when reader detects label, the identity information for reading label storage is sent to central server, in
Central server recalls correspondence personal files from database server, and opens the unusual checking work(of the region cameras
Energy.
Abnormal behavior detection uses the method based on template matches, it is contemplated that everyone human figure feature and day
Difference between normal behavioural characteristic, the present invention proposes the method for setting up behaviour template respectively to each detection object, to institute
There is guardianship to realize targetedly to detect, improve the false drop rate of behavioral value.
Carry out behavior IMAQ to each old man for being guarded of needs first, collection contain walk, squat down, it is curved
Waist, ten a kind of image sequences of common behavior such as sit down, lie down, standing, and the feature of different behaviors is extracted, use histogram of gradients
It is described, ultimately forms the feature templates of different behaviors.Old man's behavior is gathered all after the feature templates to be formed coding
It is stored in the personal files in database server.Feature templates coding schedule such as table 1 below:
Table 1
After camera opens unusual checking function, by the detection two field picture collected and the background mould set up before
Type carries out background difference, obtains foreground image, then with detecting that the behavioural characteristic template of target is matched in database, matches
Alarm is carried out during to abnormal behaviours such as tumbles, and be recorded in personal files.Foreground extraction is as follows with judgment formula:
dt(x, y)=| It(x,y)-Bt(x,y)| (5)
Wherein dt(x,y)、It(x,y)、Bt(x, y) represents difference image, current frame image, the background image of t respectively
Pixel value at (x, y) place, T1For prospect decision threshold.
Assuming that detection target is A, then its template is TA1~TA11, a kind of respectively the ten of A templates of behavior.By TAxStack
In dtTranslation search, T are carried out on (x, y)AxTo detect the target A corresponding threshold values of x-th of behavior, x represents 1~11 (h=11)
Between arbitrary value, if dtIn (x, y) by the subgraph of template covering part be Si,j(m, n), wherein (i, j) is the subgraph upper left corner
Pixel is in dtCoordinate in (x, y) image, M, N are respectively the maximum transverse and longitudinal coordinate value of image, represented with D (i, j) template with
The matching degree of image, matching formula is as follows:
Wherein T2For template matches threshold values, Si,j(m, n) and TAx(m, n) is respectively subgraph and template at (m, n) coordinate
Pixel value, is then judged as that image is matched with target A x-th of behaviour template when D (i, j) is less than threshold values, recorded data
Storehouse, if hazardous act, then starts alarm.
In order to protect the life privacy of the elderly, center control platform not display monitoring under normal circumstances under normal circumstances
Picture, when camera opens abnormal behavior detection function, platform detects object according to the result display of template matches during detection
Behavior state, such as walk, sit down, squat down.Only when system detectio goes out the abnormal behaviours such as tumble and is alarmed, platform
On-site supervision picture is just opened, and takes rescue action in time.
The technical principle for being the specific embodiment of the present invention and being used above, if conception under this invention institute
The change of work, during the spirit that function produced by it is still covered without departing from specification and accompanying drawing, should belong to the present invention's
Protection domain.
Claims (4)
1. a kind of combination RFID and video identification old man's abnormal behaviour monitoring method, it is characterised in that:The monitoring method bag
Include following steps:
1) in all monitoring regional deployment cameras, read while placing a label in the visual range of each camera
Device, at the same to monitoring in need elders wear RFID tag, and on label store carrier identity information;
2) when label-carrier is introduced into reader region, camera is in normal monitoring state, without abnormal row
For detection, camera collection background image during this period carries out background modeling using mixed Gauss model, and according to collecting
Newest background image constantly updates background model, and obtained newest background model is used for needing the picture frame of progress behavioral value
Difference is carried out, so as to obtain the foreground image of pedestrian;
3) enter fashionable when reader detects label, read label storage identity information send to central server, in it is genuinely convinced
Business device recalls correspondence personal files from database server, and opens the unusual checking function of the region cameras;
Abnormal behavior detection uses the method based on template matches, and the old man that each needs is guarded is carried out first
Behavior IMAQ, gathers the image sequence of common behavior, and extracts the feature of different behaviors, is retouched with histogram of gradients
State, ultimately form the feature templates of different behaviors;
After camera opens unusual checking function, the background model by the detection two field picture collected with setting up before is entered
Row background difference, obtains foreground image, then with detecting that the behavioural characteristic template of target is matched in database, matches and falls
Alarm etc. is carried out during abnormal behaviour, and be recorded in personal files;
Foreground extraction is as follows with judgment formula:
dt(x, y)=| It(x,y)-Bt(x,y)| (5)
<mrow>
<msub>
<mi>d</mi>
<mi>t</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<msub>
<mi>d</mi>
<mi>t</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<msub>
<mi>T</mi>
<mn>1</mn>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<msub>
<mi>d</mi>
<mi>t</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<msub>
<mi>T</mi>
<mn>1</mn>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein dt(x,y)、It(x,y)、Bt(x, y) represents that difference image, current frame image, the background image of t exist respectively
The pixel value at (x, y) place, T1For prospect decision threshold;
Assuming that detection target is A, then its template is TA1~TAh, the template of respectively A common behavior, h is the total of common behavior
Number, h >=5;By TAx(m, n) overlays dtTranslation search, T are carried out on (x, y)AxTo detect the target A corresponding valve of x-th of behavior
Value, x represents the arbitrary value between 1~h, if dtIn (x, y) by the subgraph of template covering part be Si,j(m, n), wherein (i, j)
It is subgraph top left corner pixel point in dtCoordinate in (x, y) image, M, N are respectively the maximum transverse and longitudinal coordinate value of image, with D (i, j)
To represent the matching degree of template and image, matching formula is as follows:
<mrow>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>n</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>m</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>M</mi>
</munderover>
<msup>
<mrow>
<mo>&lsqb;</mo>
<msub>
<mi>S</mi>
<mrow>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>m</mi>
<mo>,</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msub>
<mi>T</mi>
<mrow>
<mi>A</mi>
<mi>x</mi>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>m</mi>
<mo>,</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo><</mo>
<msub>
<mi>T</mi>
<mn>2</mn>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<msub>
<mi>T</mi>
<mn>2</mn>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, T2For template matches threshold values, Si,j(m, n) and TAx(m, n) is respectively the picture of subgraph and template at (m, n) coordinate
Element value, is then judged as that image is matched with target A x-th of behaviour template when D (i, j) is less than threshold values, recorded database, such as
Fruit is hazardous act, then starts alarm.
2. old man's abnormal behaviour monitoring method of RFID and video identification is combined as claimed in claim 1, it is characterised in that:Institute
State step 2) in, the foundation of background model and renewal process are as follows:
If XtFor the background image of t, mixed Gauss model is set up to each pixel on the moment background image as follows:
<mrow>
<mi>P</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>X</mi>
<mi>t</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msubsup>
<mi>&Sigma;</mi>
<mrow>
<mi>p</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</msubsup>
<msub>
<mi>&omega;</mi>
<mrow>
<mi>p</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>X</mi>
<mi>t</mi>
</msub>
<mo>,</mo>
<msub>
<mi>&mu;</mi>
<mrow>
<mi>p</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
</msub>
<mo>,</mo>
<msubsup>
<mi>&sigma;</mi>
<mrow>
<mi>p</mi>
<mo>,</mo>
<mi>t</mi>
</mrow>
<mn>2</mn>
</msubsup>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, ωP, t,μP, t,The weight coefficient, average and variance of p-th of Gaussian component of t are represented respectively,The distribution function of p-th of Gaussian component of t is represented, k represents the number of Gaussian component;
Mixed Gauss model updates as follows:
ωP, t+1=(1+ α) ωP, t, (2)
μP, t+1=β μP, t+ (1- β) I (x, y, t) (3)
Wherein, α is background model weight undated parameter, and β is background model average undated parameter, and I (x, y, t) is image I in picture
The gray value at vegetarian refreshments (x, y) place;
Each pixel is set up after mixed Gauss model formula (1) and Gauss model newer (2), formula (3) on background image,
T, finally obtains the mean μ and variances sigma of each pixel point model gray value2, it is secondary to be measured for given one in t
Image U, U (x, y) are gray values of the image U at pixel (x, y) place, with the back of the body of each pixel U (x, y) with corresponding moment t
Scape image pixel point is matched, and thinks that the match is successful if below equation is met, (x, y) is judged as into background dot;Otherwise,
(x, y) is foreground point, and matching formula is as follows:
<mrow>
<msup>
<mi>e</mi>
<mrow>
<mo>-</mo>
<mfrac>
<msup>
<mrow>
<mo>(</mo>
<mi>U</mi>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
<mo>-</mo>
<mi>&mu;</mi>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mrow>
<mn>2</mn>
<msup>
<mi>&sigma;</mi>
<mn>2</mn>
</msup>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
</mrow>
</msup>
<mo>></mo>
<mi>T</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, σ2(x, y) is variance of the adaptation function at point (x, y) place, and T is default threshold value.
3. combining old man's abnormal behaviour monitoring method of RFID and video identification as claimed in claim 1 or 2, its feature exists
In:Center control platform not display monitoring picture, when camera opens abnormal behavior detection function, center control platform according to
The behavior state of the result display detection object of template matches, only when system detectio abnormal behaviour and is alarmed during detection
When, center control platform just opens on-site supervision picture.
4. combining old man's abnormal behaviour monitoring method of RFID and video identification as claimed in claim 1 or 2, its feature exists
In:The step 1) in, the personal information of all carrier is stored in database server;The step 3) in, by institute
Have and be stored in after the feature templates to be formed coding is gathered to old man's behavior in the personal files in database server.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510260138.XA CN104850841B (en) | 2015-05-20 | 2015-05-20 | Combination RFID and video identification a kind of old man abnormal behaviour monitoring method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510260138.XA CN104850841B (en) | 2015-05-20 | 2015-05-20 | Combination RFID and video identification a kind of old man abnormal behaviour monitoring method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104850841A CN104850841A (en) | 2015-08-19 |
CN104850841B true CN104850841B (en) | 2017-11-07 |
Family
ID=53850476
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510260138.XA Active CN104850841B (en) | 2015-05-20 | 2015-05-20 | Combination RFID and video identification a kind of old man abnormal behaviour monitoring method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104850841B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105303769A (en) * | 2015-10-30 | 2016-02-03 | 安徽云硕科技有限公司 | Omnibearing intelligent home-based care method for the aged |
CN106204573A (en) * | 2016-07-07 | 2016-12-07 | Tcl集团股份有限公司 | A kind of food control method and system of intelligent refrigerator |
CN106231265A (en) * | 2016-08-15 | 2016-12-14 | 深圳市校联宝科技有限公司 | A kind of image-pickup method and image capturing system |
CN107289586B (en) * | 2017-06-15 | 2020-06-05 | 广东美的制冷设备有限公司 | Air conditioning system, air conditioner and method for alarming falling through air conditioning system |
CN107330414A (en) * | 2017-07-07 | 2017-11-07 | 郑州轻工业学院 | Act of violence monitoring method |
CN110659397B (en) * | 2018-06-28 | 2022-10-04 | 杭州海康威视数字技术股份有限公司 | Behavior detection method and device, electronic equipment and storage medium |
CN108924509A (en) * | 2018-08-03 | 2018-11-30 | 陕西中达公路技术服务有限公司 | A kind of multi-cam transmission method based on MultiView gateway |
CN111753860B (en) * | 2019-03-27 | 2024-02-02 | 杭州海康威视数字技术股份有限公司 | Analysis anomaly detection method and device |
CN112399049A (en) * | 2020-09-22 | 2021-02-23 | 深圳英飞拓科技股份有限公司 | Gas station illegal gas filling identification method, system and equipment based on intelligent box |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101542548A (en) * | 2006-06-02 | 2009-09-23 | 传感电子公司 | Systems and methods for distributed monitoring of remote sites |
CN103021136A (en) * | 2012-12-24 | 2013-04-03 | 电子科技大学 | System and method for monitoring swimming pool drowning based on Internet of Things |
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 |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9342777B2 (en) * | 2006-07-06 | 2016-05-17 | Ricoh Company, Ltd. | Programmatic control of RFID tags |
-
2015
- 2015-05-20 CN CN201510260138.XA patent/CN104850841B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101542548A (en) * | 2006-06-02 | 2009-09-23 | 传感电子公司 | Systems and methods for distributed monitoring of remote sites |
CN103021136A (en) * | 2012-12-24 | 2013-04-03 | 电子科技大学 | System and method for monitoring swimming pool drowning based on Internet of Things |
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 |
Also Published As
Publication number | Publication date |
---|---|
CN104850841A (en) | 2015-08-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104850841B (en) | Combination RFID and video identification a kind of old man abnormal behaviour monitoring method | |
CN104881637B (en) | Multimodal information system and its fusion method based on heat transfer agent and target tracking | |
CN103325080B (en) | A kind of home for the aged based on technology of Internet of things intelligent safeguard system and method | |
CN105787472B (en) | A kind of anomaly detection method based on the study of space-time laplacian eigenmaps | |
CN103605971B (en) | Method and device for capturing face images | |
CN109670441A (en) | A kind of realization safety cap wearing knows method for distinguishing, system, terminal and computer readable storage medium | |
CN106919921B (en) | Gait recognition method and system combining subspace learning and tensor neural network | |
CN104933819B (en) | Alarm and alarm method based on recognition of face and landmark identification | |
CN107832708A (en) | A kind of human motion recognition method and device | |
CN104408406B (en) | Personnel based on frame difference method and background subtraction leave the post detection method | |
CN108875561A (en) | A kind of hidden danger repetition method of discrimination of transmission line of electricity monitoring hidden danger early warning image | |
CN105931325A (en) | Gym intelligent management system and method | |
CN109740411A (en) | Intelligent monitor system, monitoring method based on recognition of face and quickly go out alarm method | |
CN105761261A (en) | Method for detecting artificial malicious damage to camera | |
CN102945366A (en) | Method and device for face recognition | |
CN112183438B (en) | Image identification method for illegal behaviors based on small sample learning neural network | |
WO2022121498A1 (en) | Identity recognition method, model training method, apparatuses, and device and storage medium | |
CN110287856A (en) | A kind of security personnel's behavior analysis system, method and device | |
CN105005852A (en) | Image analysis based intelligent monitoring system for dormitory environment | |
CN103390151A (en) | Face detection method and device | |
CN104298988B (en) | A kind of property guard method matched based on video image local feature | |
CN114663390A (en) | Intelligent anti-pinch method, device, equipment and storage medium for automatic door | |
CN105160285A (en) | Method and system for recognizing human body tumble automatically based on stereoscopic vision | |
CN109472220A (en) | A kind of substation's worker safety helmet detection method and its system based on Faster R-CNN | |
CN106897726A (en) | The finding method and device of Missing Persons |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
EXSB | Decision made by sipo to initiate substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP01 | Change in the name or title of a patent holder | ||
CP01 | Change in the name or title of a patent holder |
Address after: 310012 floor 1, building 1, No. 223, Yile Road, Xihu District, Hangzhou City, Zhejiang Province Patentee after: Yinjiang Technology Co.,Ltd. Address before: 310012 floor 1, building 1, No. 223, Yile Road, Xihu District, Hangzhou City, Zhejiang Province Patentee before: ENJOYOR Co.,Ltd. |