CN104318250B - Motor behavior mode identification method and system based on distributed perimeter system - Google Patents
Motor behavior mode identification method and system based on distributed perimeter system Download PDFInfo
- Publication number
- CN104318250B CN104318250B CN201410571160.1A CN201410571160A CN104318250B CN 104318250 B CN104318250 B CN 104318250B CN 201410571160 A CN201410571160 A CN 201410571160A CN 104318250 B CN104318250 B CN 104318250B
- Authority
- CN
- China
- Prior art keywords
- motor behavior
- response
- behavior pattern
- characteristic vector
- threshold value
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Eye Examination Apparatus (AREA)
Abstract
The invention discloses a kind of motor behavior mode identification method and system based on distributed perimeter system, method comprises the following steps：The noisy data of all sensing points position in the range of the whole distributed perimeter detection of collection；Operation is standardized according to the noisy data of collection, 0/1 matrix is converted into, if the noisy data of certain sensing point position is more than the threshold value of setting, the state of sensing point position this moment is designated as 1,0 is otherwise designated as；Extract the coordinate information of institute promising 1, including time and positional information calculate response range S according to coordinate information statistics, duration of response T and respond shape M；Using promising 1 coordinate information, least square fitting is carried out, and ask for slope K, so that the speed V moved, constitutes the characteristic vector F of motor behavior pattern（S, T, M, K, V）；Motor behavior pattern is determined by characteristic vector F.
Description
Technical field
The present invention relates to pattern-recognition, more particularly to a kind of motor behavior pattern-recognition side based on distributed perimeter system
Method and system.
Background technology
Circumference security protection refers in important area, such as national defence border, military base, key departments of government, oil depot coalfield, nuclear energy
Power station, solar power station, transformer station of power plant, bank, prison, museum, airport, harbour, villa community, data center, Shui Chu
Factory, chemical plant, school and other great infrastructure etc. are managed, to prevent illegal invasion destructive activity, along place circumference shape
Into safety precaution.Distributed perimeter system is a kind of a wide range of, and highly sensitive border guard system is increasingly taken seriously,
But existing technology can not be judged the behavior of motion, it is impossible to judge the behavior pattern of the direction of motion and motion.For example it is special
Sharp CN102280006A, CN102168953A and Wu Mawei's etc.《Distribution type fiber-optic based on phase sensitivity optical time domain reflection technology encloses
Column intrusion detection application study》Etc. the intrusion detection and positioning that refer to distributed perimeter system, but motor behavior it is not related to
Identification；The present invention proposes a kind of recognition methods of the motor behavior pattern based on distributed perimeter system, solves above-mentioned skill
Art can not be to operation action pattern identification, high with discrimination, wrong report is few, the features such as application value is high.
The content of the invention
Can not have for the distributed perimeter system of prior art to motor behavior (people's row, garage, aircraft, bullet etc.)
Effect differentiates, to meet effective identification to motor behavior pattern, now provides one kind and gives distributed perimeter system motor behavior mould
Formula recognition methods and system.
The technical solution adopted for the present invention to solve the technical problems is：
A kind of motor behavior mode identification method based on distributed perimeter system is provided, comprised the following steps：
The noisy data of all sensing points position in the range of the whole distributed perimeter detection of collection；
Operation is standardized according to the noisy data of collection, 0/1 matrix is converted into, if the noisy data of certain sensing point position
More than the threshold value of setting, then the state of sensing point position this moment is designated as 1, is otherwise designated as 0；
Extract the coordinate information of institute promising 1, including time and positional information, according to coordinate information statistics calculating response range
S and duration of response T, and using the edge of edge detection method detecting system response, then using template matching method to being
The template that the response shape of system used in match cognization, the shape that meets with a response M, template matching method is the mould prestored
Plate；
Using promising 1 coordinate information (including time and positional information), carry out least square fitting, and ask for tiltedly
Rate K, so that the speed V moved, constitutes the characteristic vector F (S, T, M, K, V) of motor behavior pattern；
Motor behavior pattern is determined by characteristic vector F.
In method of the present invention, the threshold value set is power spectrum, perturbation amplitude, fixed frequency band energy or entropy
Threshold value.
In method of the present invention, step determines that motor behavior pattern is specially using default by characteristic vector F
Data sample storehouse carries out similarity analysis, and motor behavior pattern is judged according to analysis result.
Present invention also offers a kind of motor behavior PRS based on distributed perimeter system, including：
Acquisition module, the noisy data for gathering all sensing points position in the range of whole distributed perimeter detection；
Matrix conversion module, for being standardized operation according to the noisy data of collection, is converted into 0/1 matrix, if certain
The noisy data of sensing point position is more than the threshold value of setting, then the state of sensing point position this moment is designated as into 1, is otherwise designated as 0；
Characteristic vector computing module, for extract promising 1 coordinate information, including time and positional information, according to seat
Mark Information Statistics and calculate response range S and duration of response T, this feature vector calculation module is additionally operable to utilize side edge detection
The edge of method detecting system response, then carries out match cognization to the response shape of system using template matching method, meets with a response
The template used in shape M, template matching method is the template prestored；This feature vector calculation module also utilizes institute promising 1
Coordinate information, least square fitting is carried out, and ask for slope K, so that the speed V moved, constitutes motor behavior mould
The characteristic vector F (S, T, M, K, V) of formula；
Motor behavior mode decision module, for determining motor behavior pattern by characteristic vector F.
In system of the present invention, the threshold value set is power spectrum, perturbation amplitude, fixed frequency band energy or entropy
Threshold value.
In system of the present invention, motor behavior mode decision module by characteristic vector F specifically for determining to move
Behavior pattern is specially that wherein slope K and the speed V of motion passes through default threshold decision, response range S, duration of response
T and response shape M carry out similarity analysis using default data sample storehouse, and motor behavior mould is judged according to analysis result
Formula.
The beneficial effect comprise that：This paper presents a kind of motor behavior pattern based on distributed perimeter system
Recognition methods, the principle based on Distributed probing extracts the characteristic parameter of various motor behavior patterns, including response shape
Shape, response range, duration, response slope and movement velocity etc., then carry out correlation using the data in feature samples storehouse
Analysis, solves the identification that above-mentioned technology can not be to operation action pattern, high with discrimination, and wrong report is few, application value height etc.
Feature.
Brief description of the drawings
Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing：
Fig. 1 is the flow chart of motor behavior mode identification method of the embodiment of the present invention based on distributed perimeter system；
Fig. 2 is the structural representation of motor behavior PRS of the embodiment of the present invention based on distributed perimeter system
Figure；
Fig. 3 is the distributed perimeter system activation profile figure under different motion behavior pattern in the embodiment of the present invention；
Fig. 4 is the echo probe distribution map in one section of investigative range of distributed circumference in one embodiment of the invention；
Fig. 5 is the echo probe distribution map in one section of investigative range of distributed circumference in another embodiment of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the present invention, not
For limiting the present invention.
Motor behavior mode identification method of the embodiment of the present invention based on distributed perimeter system, as shown in figure 1, including with
Lower step：
The noisy data of all sensing points position in the range of S1, the whole distributed perimeter detection of collection, the noisy data bag
Include the excited data of various motor behaviors；
S2, operation is standardized according to the noisy data of collection, 0/1 matrix is converted into, if the disturbance of certain sensing point position
Data are more than the threshold value of setting, then the state of sensing point position this moment are designated as into 1, are otherwise designated as 0；
S3, the coordinate information for extracting institute promising 1, including time and positional information are calculated according to coordinate information statistics and responded
Scope S and duration of response T, and using the edge of edge detection method detecting system response, then utilize template matching method
Response shape to system carries out match cognization, the shape that meets with a response M.
S4, using promising 1 coordinate information (including time and positional information), carry out least square fitting, and ask
Slope K is taken, so that the speed V moved, constitutes the characteristic vector F (S, T, M, K, V) of motor behavior pattern；
S5, motor behavior pattern determined by characteristic vector F.
Set threshold value can be the threshold value of power spectrum, perturbation amplitude, fixed frequency band energy or entropy.Such as select power
Spectrum, then carry out power Spectral Estimation calculating to the noisy data of system acquisition, then carry out threshold decision to it, such as selection power spectrum
As threshold decision, then threshold value is set to 5 × 108(mv)2During/Hz, if greater than the threshold value of the setting, then this moment of sensing point
State is 1, otherwise is zero.Threshold value herein can also for perturbation amplitude (during such as larger than 60mV, take 1, on the contrary take 0), fixed frequency
Band energy (such as larger than 4.5 × 109(mv)2When take 1, otherwise take 0), 0) etc. entropy (takes 1 during if greater than 1.34, otherwise takes；Threshold value
The main function of judgement is to judge whether disturbance, does not limit the above method.
Determine that motor behavior pattern is specially that step determines to move by characteristic vector F by characteristic vector F in step S5
Behavior pattern is specially that wherein slope K and the speed V moved are by default threshold decision, and the threshold value of specific movement velocity is set
Shown in definite opinion table 1 below.Response range S, duration of response T and response shape M carry out phase using default data sample storehouse
Like property analysis, motor behavior pattern is judged according to analysis result.
The various actions speedometer of table 1
| Behavior | People walks | Cycle | Automobile | Remarks |
| Speed | <100m/min | 250~350m/min | >500m/min |
The present invention is based on the continuous uninterrupted detectable principle of distributed perimeter system, when motor behavior pattern generating process
In, distributed perimeter system can be changed to excitation point response with motor behavior pattern.For example when people's row is along distribution
When the search coverage of perimeter security system is walked about, during the detection motion of vertical distribution formula, it is distributed in region during certain point Persistent Excitation
The exciter response of formula perimeter system is as shown in Figure 2.
Motor behavior PRS of the embodiment of the present invention based on distributed perimeter system, for realizing above-mentioned side
Method, as shown in Fig. 2 system is specifically included：
Acquisition module, the noisy data for gathering all sensing points position in the range of whole distributed perimeter detection；
Matrix conversion module, for being standardized operation according to the noisy data of collection, is converted into 0/1 matrix, if certain
The noisy data of sensing point position is more than the threshold value of setting, then the state of sensing point position this moment is designated as into 1, is otherwise designated as 0；Wherein
Threshold value of the threshold value set as power spectrum, perturbation amplitude, fixed frequency band energy or entropy.
Characteristic vector computing module, for extract promising 1 coordinate information, including time and positional information, according to seat
Mark information forms figure as shown in Figure 3, wherein x-axis coordinate representation distance, and y-axis represents the time, and black color dots represent echo probe model
Enclose；Then edge detection method is utilized, is examined such as curve-fitting method, boundary operator method (sobles operators, Canny operators)
The edge of examining system response, then carries out match cognization, identification response shape to the response shape of system using template matching method
M；And count calculating response range S and duration of response T；And (including time and position are believed using the coordinate information of institute promising 1
Breath), least square fitting is carried out, and slope K is asked for, so that the speed V moved, constitutes the feature of motor behavior pattern
Vectorial F (S, T, M, K, V)；
Motor behavior mode decision module, for determining motor behavior pattern by characteristic vector F.The module specifically for
By characteristic vector F determine motor behavior pattern be specially wherein slope K and motion speed V by default threshold decision,
Response range S, duration of response T and response shape M using default data sample storehouse carry out similarity analysis, according to divide
Analysis result judges motor behavior pattern.
In one embodiment of the present of invention, the data of actual people's row are alarmed accordingly in distributed perimeter system
Judge, the response distribution map of actual one section of distributed perimeter detection system, as shown in Figure 4；It is using edge detection method, such as bent
The edge of the detecting systems such as line fitting process, boundary operator method response, then carries out matching method meter using the intrinsic template of system
Calculate, matching result is quadrangle, while the coverage for calculating system is 39m, 40m, the duration is 25s, 30s；It is logical
Cross least square fitting slope -13.5469 and 14.1103；By slope calculate movement velocity be 85.625m/min and
83.63m/min；Constitute the characteristic vector 1 (39m, 25s, ' quadrangle ', -13,5469,85.625m/min) of motor behavior pattern
With characteristic vector 2 (40m, 30s, ' quadrangle ', 14.1103,83.63m/min).
According to table 1, with reference to the speed of various actions, judge excitation above for people's walking behaviour inspiration；While basis
The Statistical Speed of people's walking<100m/min, calculates that actual distribution formula perimeter system is using the slope of this algorithm | k | >=12, and it is real
The result that border is calculated meets notional result, illustrates the actual effectiveness that the method is present.The Statistical Speed of automobile>500m/min,
Calculate that distributed perimeter system slope calculations are | k |<2.4.Simultaneously according to response time (being more than 20s) and scope (being less than 50m),
It is shaped as the analysis such as ' quadrangle ' and judges that excitation behavior behaviour is moved along detecting optical cable.
In another embodiment of the present invention, Persistent Excitation response distribution map is as shown in figure 5, the method according to the present invention is final
It is (1.5m, 30s, ' rectangle ', ' inf ', ' 0 ') to extract characteristic vector, wherein " inf " represents infinitely great, be may determine that with this
Go out the Persistent Excitation behavior for people.
It should be appreciated that for those of ordinary skills, can according to the above description be improved or converted,
And all these modifications and variations should all belong to the protection domain of appended claims of the present invention.
Claims (6)
1. a kind of motor behavior mode identification method based on distributed perimeter system, it is characterised in that comprise the following steps：
The noisy data of all sensing points position in the range of the whole distributed perimeter detection of collection；
Operation is standardized according to the noisy data of collection, 0/1 matrix is converted into, if the noisy data of certain sensing point position is more than
The threshold value of setting, then be designated as 1 by the state of sensing point position this moment, be otherwise designated as 0；
Extract the coordinate information of institute promising 1, including time and positional information, according to coordinate information statistics calculate response range S with
Duration of response T, and using the edge of edge detection method detecting system response, then using template matching method to system
It is the template prestored to respond the template used in shape progress match cognization, the shape that meets with a response M, template matching method；
Using promising 1 coordinate information, carry out least square fitting, and ask for slope K so that the speed V moved,
Constitute the characteristic vector F of motor behavior pattern（S, T, M, K, V）；
Motor behavior pattern is determined by characteristic vector F.
2. according to the method described in claim 1, it is characterised in that the threshold value set is power spectrum, perturbation amplitude, fixation
The threshold value of frequency band energy or entropy.
3. according to the method described in claim 1, it is characterised in that step determines that motor behavior pattern has by characteristic vector F
Body is to carry out similarity analysis using default data sample storehouse, and motor behavior pattern is judged according to analysis result.
4. a kind of motor behavior PRS based on distributed perimeter system, it is characterised in that including：
Acquisition module, the noisy data for gathering all sensing points position in the range of whole distributed perimeter detection；
Matrix conversion module, for being standardized operation according to the noisy data of collection, is converted into 0/1 matrix, if certain is detected
The noisy data of point position is more than the threshold value of setting, then the state of sensing point position this moment is designated as into 1, is otherwise designated as 0；
Characteristic vector computing module, for extract promising 1 coordinate information, including time and positional information are believed according to coordinate
Breath statistics calculates response range S and duration of response T, and this feature vector calculation module is additionally operable to utilize edge detection method inspection
The edge of examining system response, then carries out match cognization, meet with a response shape using template matching method to the response shape of system
The template used in M, template matching method is the template prestored；This feature vector calculation module also using promising 1 seat
Information is marked, least square fitting is carried out, and asks for slope K, so that the speed V moved, constitutes motor behavior pattern
Characteristic vector F（S, T, M, K, V）；
Motor behavior mode decision module, for determining motor behavior pattern by characteristic vector F.
5. system according to claim 4, it is characterised in that the threshold value set is power spectrum, perturbation amplitude, fixation
The threshold value of frequency band energy or entropy.
6. system according to claim 4, it is characterised in that motor behavior mode decision module is specifically for passing through feature
Vectorial F determines that motor behavior pattern is specially that wherein slope K and the speed V of motion passes through default threshold decision, response range
S, duration of response T and response shape M carry out similarity analysis using default data sample storehouse, are sentenced according to analysis result
Disconnected motor behavior pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410571160.1A CN104318250B (en) | 2014-10-23 | 2014-10-23 | Motor behavior mode identification method and system based on distributed perimeter system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410571160.1A CN104318250B (en) | 2014-10-23 | 2014-10-23 | Motor behavior mode identification method and system based on distributed perimeter system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104318250A CN104318250A (en) | 2015-01-28 |
| CN104318250B true CN104318250B (en) | 2017-09-29 |
Family
ID=52373479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410571160.1A Active CN104318250B (en) | 2014-10-23 | 2014-10-23 | Motor behavior mode identification method and system based on distributed perimeter system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104318250B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104700624B (en) * | 2015-03-16 | 2017-07-07 | 电子科技大学 | The monitoring method of the vehicle flowrate on-line monitoring system based on phase sensitivity optical time domain reflectometer |
| CN104766099B (en) * | 2015-05-05 | 2017-12-05 | 武汉理工光科股份有限公司 | Distribution type fiber-optic circumference vibration signal processing and recognition methods based on image |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345808A (en) * | 2013-06-26 | 2013-10-09 | 武汉理工光科股份有限公司 | Fiber Bragg grating perimeter intrusion pattern recognition method and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4284288B2 (en) * | 2005-03-10 | 2009-06-24 | 株式会社東芝 | Pattern recognition apparatus and method |
-
2014
- 2014-10-23 CN CN201410571160.1A patent/CN104318250B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345808A (en) * | 2013-06-26 | 2013-10-09 | 武汉理工光科股份有限公司 | Fiber Bragg grating perimeter intrusion pattern recognition method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104318250A (en) | 2015-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108709633B (en) | Distributed optical fiber vibration sensing intelligent safety monitoring method based on deep learning | |
| CN105389917B (en) | A kind of quick method for early warning based on phase sensitive optical time domain reflectometer | |
| CN108492564A (en) | The method and system of express-road vehicle running speed are measured based on road network | |
| KR102468597B1 (en) | Worker management system for field of construction work using KIOSK and beacon signal | |
| CN110620906A (en) | Electric power construction safety management system based on thing networking | |
| EP2037426A1 (en) | Device and method for detecting suspicious activity, program, and recording medium | |
| CN110532889A (en) | Track foreign matter detecting method based on rotor unmanned aircraft and YOLOv3 | |
| CN104821056A (en) | Intelligent guarding method based on radar and video integration | |
| CN102081801A (en) | Multi-feature adaptive fused ship tracking and track detecting method | |
| CN106597360A (en) | Method for using robot to real-time and dynamically position and monitor indoor object based on RFID | |
| CN105607063A (en) | Detection method and system for low-altitude wind shear at airport | |
| CN103298156B (en) | Based on the passive multi-target detection tracking of wireless sensor network | |
| CN109076190A (en) | Detect the device and method of abnormal conditions | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN104700056A (en) | Method for detecting uniqueness of person entering coal mineral well | |
| CN106295751A (en) | Substation equipment maintenance and repair operation location and safe early warning method | |
| CN104318250B (en) | Motor behavior mode identification method and system based on distributed perimeter system | |
| CN105445731A (en) | Full-automatic radar tracking method based on high precision speed measurement | |
| CN104732236A (en) | Intelligent crowd abnormal behavior detection method based on hierarchical processing | |
| CN107578646A (en) | low slow small target detection monitoring management system and method | |
| CN102116876B (en) | Method for detecting spatial point target space-base on basis of track cataloguing model | |
| CN106706109B (en) | Vibration source identification method and system based on time domain two-dimensional characteristics | |
| CN108573280A (en) | A kind of unmanned boat independently passes through the method for bridge | |
| CN115879616A (en) | High-risk meteorological identification method and device based on power transmission line microclimate station monitoring data | |
| CN105911553B (en) | A kind of road feasible zone determines method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |