CN105632187B - Low-power consumption curb parking detection method based on geomagnetic sensor network - Google Patents

Low-power consumption curb parking detection method based on geomagnetic sensor network Download PDF

Info

Publication number
CN105632187B
CN105632187B CN201511028573.6A CN201511028573A CN105632187B CN 105632187 B CN105632187 B CN 105632187B CN 201511028573 A CN201511028573 A CN 201511028573A CN 105632187 B CN105632187 B CN 105632187B
Authority
CN
China
Prior art keywords
signal
disturbance
threshold value
parking
vehicle
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
Application number
CN201511028573.6A
Other languages
Chinese (zh)
Other versions
CN105632187A (en
Inventor
朱红梅
于峰崎
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Priority to CN201511028573.6A priority Critical patent/CN105632187B/en
Publication of CN105632187A publication Critical patent/CN105632187A/en
Application granted granted Critical
Publication of CN105632187B publication Critical patent/CN105632187B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/042—Detecting movement of traffic to be counted or controlled using inductive or magnetic detectors
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
    • G08G1/145—Traffic control systems for road vehicles indicating individual free spaces in parking areas where the indication depends on the parking areas

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

A kind of low-power consumption curb parking detection method based on geomagnetic sensor network, the following steps are included: disturbing signal of the geomagnetic sensor node acquisition parking lot vehicle to earth magnetism, and the signal of acquisition is pre-processed, to selecting parking detection algorithm according to treated signal strength, the lower state machine detection algorithm of complexity is selected if signal strength is lower than choice threshold value, the judgement of parking space state is obtained according to the result that jumps of state machine, otherwise relative complex cross-correlation test method is selected, the state of parking stall is judged according to the correlation of signal and the K-means reference model clustered.The method that above-mentioned curb parking vehicle checking method uses the lower state machine testing of complexity to combine with the higher cross-correlation test of accuracy, this method has taken into account power consumption while guaranteeing detection accuracy, node service life is extended, to lay a good foundation for intelligent parking management.

Description

Low-power consumption curb parking detection method based on geomagnetic sensor network
[technical field]
The present invention relates to vehicle testing techniques more particularly to a kind of low-power consumption curb parkings based on geomagnetic sensor network Detection method.
[background technique]
Curb parking detection is the pith for realizing intelligent transportation, and current main detection method has: ultrasound, feel line The means such as circle, video and geomagnetic sensor.Wherein, although ultrasound detection accuracy is higher, be not suitable in parking lay-by Installation and maintenance；Ground induction coil detection performance is stablized, but installation and maintenance need to cut road surface, interferes normal traffic；Video Detection information amount is abundant and intuitive, but needs bigger storage capacity and computing capability, and detection performance is highly prone to light According to the influence of condition and rain and snow weather.In comparison, geomagnetic sensor has low-power consumption, low cost, high sensitivity, corpusculum The advantages that accumulating, being easily integrated and install and is easy to maintain.The region that geomagnetic sensor is placed in vehicle process can be adopted Collect vehicle to the disturbing signal of earth magnetism, disturbing signal can be used to detect presence, vehicle, speed of vehicle etc. after being handled Traffic information.
Ground magnetic field intensity is considered uniform and stable in several kilometer ranges, when passing through containing a large amount of ferromagnetic vehicles Biggish disturbance will be generated to earth's magnetic field, the variation of external magnetic field causes geomagnetic sensor output valve to change.The letter of output The presence of vehicle can be used to detect number after treatment, vehicle driving speed can be estimated by laying two geomagnetic sensor nodes Degree, waveform and vehicle itself construction and the iron content quantity of magnetism of output signal have much relations, therefore may be implemented according to this feature Vehicle cab recognition.Current existing algorithm is broadly divided into two kinds: state machine testing based on threshold value and based on the mutual of similitude Close detection method.
State machine detection algorithm based on threshold value, main thought are that vehicle detection process is divided into different states, when State is jumped after meeting corresponding condition, to obtain parking space state.Wherein, state machine jumps condition and relies primarily on its prison Signal is surveyed compared with given threshold.Its advantage is that method is simple, low in energy consumption, calculation amount is small, convenient for real on sensor node It is existing, but the disadvantage is that threshold value is empirical value, it is difficult to guarantee its objectivity, when signal disturbance intensity is smaller, is easy to appear mistake Sentence or fails to judge.The background signal of curb parking is complicated, and there are the interference signals of road travel, and noise is relatively low, this simple State machine approach detection accuracy in curb parking detection is difficult to ensure.
Cross-correlation test method based on similitude can effectively filter with the incoherent noise signal of reference signal so that Useful signal is enhanced, to improve Detection accuracy.The detection accuracy of cross-correlation test is by selected reference signal shadow Sound is larger, and the reference signal that existing cross-correlation test uses has a geomagnetic field variation model and Gauss curve fitting curve etc., but this A little researchs are not suitable for parking lot vehicle detection primarily directed to moving vehicle, therefore the present invention passes through K-means clustering algorithm Find with the actual maximally related reference signal of vehicle disturbing signal, while balanced power consumption and performance calculate state machine testing Method combines to obtain a kind of curb parking vehicle checking method of low-power consumption with cross-correlation test algorithm.
[summary of the invention]
In view of this, it is necessary to provide a kind of parking lay-by vehicle checking methods of low-power consumption.The geomagnetic sensor Node acquires parking lot vehicle to the disturbing signal of earth magnetism, and pre-processes to the signal of acquisition, believes according to treated Number intensity selects parking detection method, if signal strength is higher than choice threshold value (TH_OP) selects the lower shape of complexity State machine examination method of determining and calculating obtains the judgement of parking space state according to the result that jumps of state machine, otherwise selects relative complex cross-correlation Detection method judges the state of parking stall according to the correlation of signal and the K-means reference model clustered.
In a preferred embodiment, to acquisition signal pre-treatment step include the following steps: the acquisition signal into Row gaussian filtering process；The filtered signal carries out modulus calculating.Wherein gaussian filtering uses following formula:
F (x)=exp (- xs 2/(2*σ2))
Wherein, σ determines the width of Gaussian function, xsThe signal after the x-axis smothing filtering, y-axis and z-axis use with The identical method of x-axis carries out gaussian filtering.
It calculates modulus value and uses following formula:
Wherein, BX、BYAnd BZSignal after respectively indicating the gaussian filtering of x, y and z axes, BX0、BY0And BZ0Respectively indicate x, y With z-axis baseline value.
In a preferred embodiment, the selection curb parking vehicle detecting algorithm includes the following steps: to pre-process Afterwards the modulus value of signal with choice threshold value (TH_OP) be compared, if it exceeds choice threshold value (TH_OP) then select complexity compared with Otherwise low state machine detection method selects the higher cross-correlation test algorithm of accuracy.
In a preferred embodiment, detection method includes the following steps for the low complex degree state machine:
(1) all parameters and baseline are initialized, setting sign of flag _ Initial is 1 after the completion of initialization, and is jumped to Step (2)；
(2) parking space state is idle (PS=0), and carries out baseline update in the step, and baseline more new formula is as follows:
In formula, Bi0(t-1) be previous moment i axis baseline value, i ∈ { x, y, z }；Bi(t) be current i axis gaussian filtering Signal afterwards；α is forgetting factor, and range is [0-1]；PS (t-1) is the parking space state of previous moment；
When detection signal has a disturbance, and disturbing signal is more than that setting is driven into threshold value (TH_AR) and then be arranged and drives into disturbance and mark Will (Fluctuation_Arrival=1), go to step (3)；
(3) record is more than and drives into driving into for threshold value (TH_AR) to disturb number (Cnt_Arrival), in the event of disturbance Lower than driving into threshold value (TH_AR) and being then arranged, to drive into disturbance mark be 0 (Fluctuation_Arrival=0) and to go to step (4), (5) otherwise are gone to step when vehicle drives into when disturbance number is more than to set and drive into frequency threshold value (N_Arrival)；
(4) it resets and drives into disturbance number Cnt_Arrival, and record and drive into disturbance number (Cnt_ without vehicle NoArrival), if disturbance is again above threshold value (TH_AR) is driven into, being once again set up and driving into disturbance mark is 1 (Fluctuation_Arrival=1), go to step (3)；It otherwise, is more than that setting nothing is sailed when no vehicle drives into disturbance number Indegree threshold value (N_noArrival), then go to step (2)；
(5) parking stall is occupied (setting mark PS=1), if disturbance is more than that vehicle is driven out to threshold value (TH_DP), jumps step Suddenly (6)；
(6) setting vehicle is driven out to disturbance mark (Fluctuation_departure=1), and records and be driven out to disturbance number (Cnt_Departure), being driven out to disturbance mark lower than setting if being driven out to threshold value (TH_DP) if disturbance is 0 (Fluctuation_departure=0), and (5) are gone to step；Otherwise, when be driven out to disturbance number (Cnt_Departure) Frequency threshold value (N_Departure) then jump procedure (2) are driven out to more than setting；
In a preferred embodiment, the high-precision cross-correlation test method the following steps are included:
(1) reference signal that signal and K-means cluster obtain after the pretreatment carries out cross-correlation calculation；
Wherein, SD (t) is modulus value after pretreatment,For modulus value mean value, R (t) is reference signal,It is equal for reference signal Value, CC (τ) indicate the cross-correlation coefficient of two signals, and in [- 1,1] range, when two signal cross-correlation degree are higher, It more levels off to 1, when two signals are completely uncorrelated, more levels off to 0.
(2) cross-correlation calculation result CC (τ) is compared with setting cross-correlation threshold value (TH_CO), and parking space state updates public Formula is as follows
Wherein, whether Event (t) expression state changes, and indicates to change when being 1, indicates that state is constant when being 0； TH_CO To set cross-correlation threshold value；CC (τ) is cross-correlation calculation result.
In a preferred embodiment, the judgement parking stall last state uses following formula:
Wherein, PS (t)=0 indicates that parking stall is idle, and PS (t)=1 indicates that parking stall is occupied.
In a preferred embodiment, the reference signal acquisition methods the following steps are included: the preprocessed signal into Row intercepting process；The intercept signal carries out interpolation processing；Signal is normalized after the interpolation；At the normalization Signal carries out K-means cluster after reason, and obtained cluster centre is reference signal.
Signal carries out indicatrix interception after the pretreatment, static part is removed, using following formula；
SD_A=SD (t), { (t1-td)≤t≤(t2+td)}
SD_D=SD (t), { (t3-td)≤t≤(t4+td)}
Wherein, SD (t) is signal modulus value after pretreatment, t1It is the time that vehicle reaches, t2Be after parking signal from unstable State is changed into the time point of stable state；Equally, t3It is the time that vehicle leaves, t4It is that vehicle leaves rear signal from unstable shape State is changed into the time point of stable state, tdIt is time delay parameter.
The indicatrix of described pair of interception carries out interpolation processing, keeps indicatrix length identical, interpolation method includes following Step:
(1) indicatrix length M, and matched curve y=f (x) are calculated；
(2) N number of point (x is evenly distributed on the section of length M1,x2,…,xN, N is the regular length after interpolation)；
(3) according to matched curve calculate N point corresponding to functional value (y1,y2,…,yN)；
(4) Y=[y1,y2,…,yN] it is indicatrix after interpolation；
The indicatrix is normalized, and it is 1 that signal, which is normalized to maximum value, in the range that minimum value is -1, Using following formula:
Wherein, YmaxAnd YminThe maximum value 1 and minimum value -1, X for respectively indicating normalization range are original signal, xmaxWith xminRespectively indicate the maximum value and minimum value in original signal.
In a preferred embodiment, the characteristic signal progress k-means cluster of described pair of extraction includes the following steps:
(1) k cluster center of mass point is chosen, is μ1,μ2,…,μk∈Rn。
(2) each sample class is calculated using following formula:
(3) mass center of each class j is recalculated using following formula:
(2) and (3) are repeated until convergence；
Wherein, K is known cluster numbers, C(i)That class representative sample I nearest with distance in k class, C(i)Value be 1 One into k, mass center μjRepresent the central point of same class sample.
In a preferred embodiment, the reference signal is using cluster mass center.
[Detailed description of the invention]
Fig. 1 is a kind of curb parking vehicle checking method stream of the parking lay-by vehicle checking method of low-power consumption of the present invention Cheng Tu；
Fig. 2 is a kind of state machine overhaul flow chart of the parking lay-by vehicle checking method of low-power consumption of the present invention；
Fig. 3 is a kind of cross-correlation test flow chart of the parking lay-by vehicle checking method of low-power consumption of the present invention；
Fig. 4 is a kind of reference signal detecting flow chart of the parking lay-by vehicle checking method of low-power consumption of the present invention；
Fig. 5 is a kind of original Geomagnetic signal signal of acquisition of the parking lay-by vehicle checking method of low-power consumption of the present invention Figure；
Fig. 6 is a kind of characteristic signal screenshot schematic diagram of the parking lay-by vehicle checking method of low-power consumption of the present invention；
Fig. 7 is that a kind of AMR continuous collecting Geomagnetic signal of the parking lay-by vehicle checking method of low-power consumption of the present invention shows It is intended to；
Fig. 8 is a kind of testing result exemplary diagram of the parking lay-by vehicle checking method of low-power consumption of the present invention；
Fig. 9 is a kind of system layout diagram of the parking lay-by vehicle checking method of low-power consumption of the present invention.
[specific embodiment]
It is illustrated below in conjunction with specific embodiment and attached drawing.
As shown in Figure 1, a kind of parking lay-by method for detecting parking stalls, comprising the following steps:
S110, acquisition signal carry out gaussian filtering, and gaussian filtering uses following formula:
F (x)=exp (- xs 2/(2*σ2))
Wherein, σ determines the width of Gaussian function, xsIt is the x-axis component of the acquisition Geomagnetic signal, y-axis and z axis are adopted Gaussian filtering is carried out with method identical with x-axis.
S120, signal carries out modulus calculating after gaussian filtering, using following formula:
Wherein, BX、BYAnd BZSignal after respectively indicating the gaussian filtering of x, y and z axes, BX0、BY0And BZ0Respectively indicate x, y With z-axis baseline value.
S130, modulus value are compared with choice threshold value (TH_OP), if it exceeds choice threshold value (TH_OP) then selects complexity Lower state machine detection method is spent, the higher cross-correlation test algorithm of accuracy is otherwise selected.
S140, low complex degree state machine testing.The step includes the following steps, as shown in Figure 2:
S141 initializes all parameters and baseline, and setting sign of flag _ Initial is 1 after the completion of initialization, and is jumped To step S142；
S142, parking space state are idle (PS=0), and carry out baseline update in the step, and baseline more new formula is as follows:
In formula, Bi0(t-1) be previous moment i axis baseline value, i ∈ { x, y, z }；Bi(t) be current i axis gaussian filtering Signal afterwards；α is forgetting factor, and range is [0-1]；PS (t-1) is the parking space state of previous moment；
There is disturbance when detecting signal, and disturbing signal drives into threshold value (TH_AR) more than setting and disturbance mark is then arranged (Fluctuation_Arrival=1), go to step S143；
S143: record be more than drive into threshold value (TH_AR) drive into disturbance number (Cnt_Arrival), in the event of disturbing It moves lower than threshold value (TH_AR) is driven into, is then once again set up and drives into disturbance mark as 0 (Fluctuation_Arrival=0) and jump Step S144 is gone to, is more than otherwise that frequency threshold value (N_ is driven into setting when vehicle drives into disturbance number (Cnt_Arrival) Go to step S145 when Arrival)；
S144: disturbance number Cnt_Arrival is driven into clearing, and records and drive into disturbance number (Cnt_ without vehicle NoArrival), it is 1 (Fluctuation_ that disturbance mark is driven into setting if disturbance is again above threshold value (TH_AR) is driven into Arrival=1), go to step S143；It otherwise, is more than setting when no vehicle drives into disturbance number (Cnt_noArrival) Without frequency threshold value (N_noArrival) is driven into, then go to step S142；
S145: parking stall is occupied (setting mark PS=1), if disturbance is more than that vehicle is driven out to threshold value (TH_DP), jumps Step S146；
S146: setting vehicle is driven out to disturbance mark (Fluctuation_departure=1), and records and be driven out to disturbance time Number (Cnt_Departure), if disturbance is driven out to threshold value (TH_DP) lower than setting, it is 0 that setting, which is driven out to disturbance mark, (Fluctuation_departure=0), go to step S145；It otherwise, is more than that setting is driven out to number when being driven out to disturbance number Threshold value (N_Departure) then jump procedure S142；
S150: high-precision cross-correlation test.The step includes the following steps.As shown in Figure 3.
S151: reference signal is obtained, is included the following steps.As shown in Figure 4.
S1511: indicatrix interception is carried out, the original Geomagnetic signal of node acquisition is as shown in figure 5, its corresponding feature is bent Line interception is as shown in fig. 6, remove static part, using following formula；
SD_A=SD (t), { (t1-td)≤t≤(t2+td)}
SD_D=SD (t), { (t3-td)≤t≤(t4+td)}
Wherein, SD (t) is signal modulus value after pretreatment, t1It is the time that vehicle reaches, t2Be after parking signal from unstable State is changed into the time point of stable state；Equally, t3It is the time that vehicle leaves, t4It is that vehicle leaves rear signal from unstable shape State is changed into the time point of stable state, tdIt is time delay parameter.
The indicatrix of S1512: described pair interception carries out interpolation processing, keeps indicatrix length identical, interpolation method packet Include following steps.
(1) indicatrix length M, and matched curve y=f (x) are calculated；
(2) N number of point (x is evenly distributed on the section of length M1,x2,…,xN, N is the regular length after interpolation)；
(3) according to matched curve calculate N point corresponding to functional value (y1,y2,…,yN)；
(4) Y=[y1,y2,…,yN] it is indicatrix after interpolation；
S1513: the indicatrix is normalized, and it is 1 that signal, which is normalized to maximum value, the model that minimum value is -1 In enclosing, using following formula:
Wherein, YmaxAnd YminThe maximum value 1 and minimum value -1, X for respectively indicating normalization range are original signal, xmaxWith xminRespectively indicate the maximum value and minimum value in original signal.
S1514: cluster obtain reference signal the following steps are included:
(1) k cluster center of mass point is chosen, is μ1,μ2,…,μk∈Rn。
(2) each sample class is calculated using following formula:
(3) mass center of each class j is recalculated using following formula:
(2) and (3) are repeated until convergence；
Wherein, K is known cluster numbers, C(i)That class representative sample I nearest with distance in k class, C(i)Value be 1 One into k, mass center μjRepresent the central point of same class sample.
S152: the reference signal that signal and K-means cluster obtain after pretreatment carries out cross-correlation calculation, using following meter Calculate formula:
Wherein, SD (t) is modulus value after pretreatment,For modulus value mean value, R (t) is reference signal,It is equal for reference signal Value, CC (τ) indicate the cross-correlation coefficient of two signals, and in [- 1,1] range, when two signal cross-correlation degree are higher, It more levels off to 1, when two signals are completely uncorrelated, more levels off to 0.
S153: cross-correlation calculation result CC (τ) is compared with setting cross-correlation threshold value (TH_CO), and parking space state updates Formula is as follows
Wherein, whether Event (t) expression state changes, and indicates to change when being 1, indicates that state is constant when being 0； TH_CO To set cross-correlation threshold value；CC (τ) is cross-correlation calculation result.The original Geomagnetic signal that parking sensor node is detected is as schemed Shown in 7, testing result is as shown in Figure 8.This method interior joint is laid as shown in Fig. 9.
S154: judge parking stall last state using following formula:
Wherein, PS (t)=0 indicates that parking stall is idle, and PS (t)=1 indicates that parking stall is occupied.
The above is a preferred embodiment of the present invention, it is noted that for those skilled in the art For, without departing from the principles of the present invention, it can also make several improvements and retouch, these improvements and modifications It should be regarded as protection scope of the present invention.

Claims (8)

1. a kind of low-power consumption curb parking detection method based on geomagnetic sensor network, comprising the following steps:
Geomagnetic sensor node acquires curb parking vehicle to the disturbing signal of earth magnetism；
The disturbing signal of acquisition is pre-processed, preprocessed signal is obtained；
By the modulus value of signal after pretreatment and choice threshold value comparison, and the state machine detection method or accurate for selecting complexity low Spend high cross-correlation test algorithm；
If the modulus value of signal is higher than choice threshold value after pretreatment, scheme is jumped according to state machine and judges current parking space state； Otherwise the correlation for calculating preprocessed signal and the K-means reference signal clustered, by cross-correlation calculation result and mutually It closes threshold value to be compared to judge whether parking space state changes, then thinks that parking space state changes more than cross-correlation threshold value, Think according to parking stall previous moment state to judge parking stall last state after parking space state changes；
Detection method includes the following steps for the low state machine of the complexity:
(1) all parameters and baseline are initialized, setting sign of flag _ Initial is 1 after the completion of initialization, and is gone to step (2)；
(2) parking space state is the free time, and carries out baseline update in the step, and baseline more new formula is as follows:
In formula, Bi0(t-1) be previous moment i axis baseline value, i ∈ { x, y, z }；Bi(t) be current i axis gaussian filtering after Signal；α is forgetting factor, and range is [0-1]；PS (t-1) is the parking space state of previous moment；
There is disturbance when detecting signal, and disturbing signal drives into threshold value more than setting and is then arranged drive into disturb indicate to be 1, jumps to Step (3)；
(3) record be more than drive into threshold value drive into disturbance number, in the event of disturbance lower than threshold value is driven into, then setting, which is driven into, disturbs Dynamic mark is 0 and gos to step (4), is more than otherwise that setting jumps to step when driving into frequency threshold value when vehicle drives into disturbance number Suddenly (5)；
(4) reset and drive into disturbance number, and record and drive into disturbance number without vehicle, if disturbance again above driving into threshold value, then It is 1 that disturbance mark is driven into secondary setting, and go to step (3)；It otherwise, is more than that setting is secondary without driving into when no vehicle drives into disturbance number Number threshold value, then go to step (2)；
(5) parking stall is occupied, and mark PS=1 is arranged, if disturbance is more than that vehicle is driven out to threshold value, jump procedure (6)；
(6) it is 1 that setting vehicle, which is driven out to disturbance mark, and records and be driven out to disturbance number, if disturbance is driven out to threshold value lower than setting, if Setting and being driven out to disturbance mark is 0, then go to step (5)；It otherwise, is more than that setting is driven out to frequency threshold value and then jumps when being driven out to disturbance number Go to step (2).
2. the method according to claim 1, wherein the signal to acquisition pre-processes specifically:
The disturbing signal of acquisition is subjected to gaussian filtering process, the modulus value of filtered signal is finally calculated, obtains preprocessed signal.
3. according to the method described in claim 2, it is characterized in that, the gaussian filtering process uses following formula:
F (x)=exp (- xs 2/(2*σ2))
Wherein, σ determines the width of Gaussian function, xsThe x-axis signal of acquisition, y-axis and z-axis using method identical with x-axis into Row gaussian filtering.
4. according to the method described in claim 2, it is characterized in that, calculating the modulus value of signal after gaussian filtering process using following Formula:
Wherein, BX、BYAnd BZSignal after respectively indicating the gaussian filtering of x, y and z axes, BX0、BY0And BZ0Respectively indicate x, y and z Axis baseline value.
5. the method according to claim 1, wherein the reference signal acquisition methods the following steps are included:
(1) preprocessed signal intercepting process；
(2) interpolation processing is carried out to the signal of interception；
(3) signal is normalized after interpolation processing；
(4) K-means cluster is carried out to the signal after normalized, obtained cluster centre is adopted as reference signal.
6. according to the method described in claim 5, it is characterized in that, the signal intercepting process is as follows using formula:
SD_A=SD (t), (t1-td)≤t≤(t2+td)
SD_D=SD (t), (t3-td)≤t≤(t4+td)
Wherein, SD (t) is signal modulus value after pretreatment, t1It is the time that vehicle reaches, t2Be after parking signal from unstable state It is changed into the time point of stable state, t3It is the time that vehicle leaves, t4It is that vehicle leaves rear signal and is changed into surely from unstable state Determine the time point of state, tdIt is time delay parameter.
7. according to the method described in claim 5, it is characterized in that, the interpolation processing the following steps are included:
(1) indicatrix length M, and matched curve y=f (x) are calculated；
(2) N number of point, x are evenly distributed on the section that length is M1, x2..., xN；N is the regular length after interpolation；
(3) according to matched curve calculate N number of point corresponding to functional value, y1, y2..., yN；
(4) Y=[y1, y2..., yN] it is indicatrix after interpolation.
8. according to the method described in claim 5, it is characterized in that, the normalization uses following formula:
Wherein, YmaxAnd YminThe maximum value 1 and minimum value -1, x for respectively indicating normalization range are original signal, xmaxAnd xminPoint It Biao Shi not maximum value and minimum value in original signal.
CN201511028573.6A 2015-12-31 2015-12-31 Low-power consumption curb parking detection method based on geomagnetic sensor network Active CN105632187B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201511028573.6A CN105632187B (en) 2015-12-31 2015-12-31 Low-power consumption curb parking detection method based on geomagnetic sensor network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201511028573.6A CN105632187B (en) 2015-12-31 2015-12-31 Low-power consumption curb parking detection method based on geomagnetic sensor network

Publications (2)

Publication Number Publication Date
CN105632187A CN105632187A (en) 2016-06-01
CN105632187B true CN105632187B (en) 2019-05-31

Family

ID=56047060

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201511028573.6A Active CN105632187B (en) 2015-12-31 2015-12-31 Low-power consumption curb parking detection method based on geomagnetic sensor network

Country Status (1)

Country Link
CN (1) CN105632187B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106327913A (en) * 2016-08-31 2017-01-11 广州地理研究所 Parking lot management method and device based on automobile electronic identification
CN106652479A (en) * 2016-11-22 2017-05-10 广西大学 Geomagnetic sensing-based wireless low-power consumption parking space detection system
CN106781531A (en) * 2016-12-21 2017-05-31 迈锐数据（北京）有限公司 A kind of method for detecting parking stalls and device
CN106530817A (en) * 2016-12-21 2017-03-22 迈锐数据（北京）有限公司 Parking space detection method and device
CN109308814A (en) * 2017-07-28 2019-02-05 南宁富桂精密工业有限公司 Method for detecting parking stalls, parking sensor and computer readable storage medium
CN107610481B (en) * 2017-08-31 2019-09-17 深圳先进技术研究院 Stop detection method, device and terminal device and computer readable storage medium
CN109979203A (en) * 2019-04-24 2019-07-05 西安中星测控有限公司 A kind of parking stall monitoring method and device based on earth magnetism monitoring

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100476900C (en) * 2007-04-17 2009-04-08 中国科学院计算技术研究所 Detecting method and device for communication flow
KR100951692B1 (en) * 2008-04-04 2010-04-07 (주)에스엔알 System and method for managing the admittance to parking place based on geomagnetic sensor
CN103117003B (en) * 2013-01-17 2015-04-01 杭州快泊信息技术有限公司 Self-calibration threshold self-adjusting parking space detection method
CN103794058B (en) * 2014-03-05 2016-04-13 武汉慧联无限科技有限公司 Based on vehicle checking method and the device of state machine
CN203882450U (en) * 2014-04-24 2014-10-15 纳米新能源（唐山）有限责任公司 Parking-stall state detector
CN104361765B (en) * 2014-11-24 2017-01-11 天津理工大学 Parking space detecting system based on magnetoresistive sensor and ZigBee

Also Published As

Publication number Publication date
CN105632187A (en) 2016-06-01

Similar Documents

Publication Publication Date Title
CN105632187B (en) Low-power consumption curb parking detection method based on geomagnetic sensor network
CN106960580B (en) A kind of method for detecting parking stalls based on geomagnetic sensor
CN103925923B (en) A kind of earth magnetism indoor locating system based on adaptive particle filter device algorithm
Duarte et al. Vehicle classification in distributed sensor networks
CN102436644B (en) Unstructured road detection method based on adaptive edge registration
CN102881171B (en) Vehicle detecting method, vehicle detecting system and vehicle path planning system
CN109544932A (en) A kind of city road network flow estimation method based on GPS data from taxi Yu bayonet data fusion
CN105716604A (en) Mobile robot indoor positioning method and system based on geomagnetic sequences
CN103020978A (en) SAR (synthetic aperture radar) image change detection method combining multi-threshold segmentation with fuzzy clustering
CN105513354A (en) Video-based urban road traffic jam detecting system
CN103559508B (en) A kind of based on continuous Adaboost video vehicle detection method
CN206249557U (en) Based on the vehicle detecting system that Magnetic Sensor and ultrasonic sensor are merged
CN103761748A (en) Method and device for detecting abnormal behaviors
CN106017342A (en) Three-dimensional-live-action-model-based automatic detection and system for building changing
CN110456320A (en) A kind of ULTRA-WIDEBAND RADAR personal identification method based on free space gait temporal aspect
CN106028446A (en) Indoor parking lot location method
CN115342814A (en) Unmanned ship positioning method based on multi-sensor data fusion
CN107490789B (en) Trailer-mounted radar signal target detection method and device based on CFAR detection
CN109239553A (en) A kind of clustering method based on local density of partial discharge pulse
CN108197829A (en) A kind of Wetland Waterbirds group habitat suitability assessment technique
CN106384507A (en) Travel time real-time estimation method based on sparse detector
CN105590330B (en) Indoor plane drawing generating method, server and system based on intelligent perception
Liu et al. Pavement surface defect recognition method based on vehicle system vibration data and feedforward neural network
CN107749174B (en) Subway interference resistant parking detection method, device, equipment and storage medium
CN106770151A (en) A kind of method and device of quick detection raman spectral signal

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