CN110491141B - Vehicle information identification system and identification method - Google Patents

Vehicle information identification system and identification method Download PDF

Info

Publication number
CN110491141B
CN110491141B CN201910780670.2A CN201910780670A CN110491141B CN 110491141 B CN110491141 B CN 110491141B CN 201910780670 A CN201910780670 A CN 201910780670A CN 110491141 B CN110491141 B CN 110491141B
Authority
CN
China
Prior art keywords
vehicle
geomagnetic
data
upper computer
neural network
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
CN201910780670.2A
Other languages
Chinese (zh)
Other versions
CN110491141A (en
Inventor
雷旭
候亚虹
张帅
杨雅麟
冯斌
赵熙
陈毅
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.)
Changan University
Original Assignee
Changan University
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 Changan University filed Critical Changan University
Priority to CN201910780670.2A priority Critical patent/CN110491141B/en
Publication of CN110491141A publication Critical patent/CN110491141A/en
Application granted granted Critical
Publication of CN110491141B publication Critical patent/CN110491141B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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
    • G08G1/015—Detecting movement of traffic to be counted or controlled with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
    • 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

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Navigation (AREA)
  • Traffic Control Systems (AREA)
  • Measuring Magnetic Variables (AREA)

Abstract

The invention discloses a vehicle information identification system and an identification method, wherein an upper computer eliminates time difference of preprocessed data of two adjacent geomagnetic sensors, and characteristic points of the data after the time difference is eliminated are aligned by adopting a DTW algorithm; the upper computer adopts a self-adaptive weighting fusion method to fuse the data after the feature points are aligned; and the upper computer classifies the vehicle types by adopting an FWOA-GA-BP neural network algorithm according to the fused data. Feature points in the vehicle data are aligned by adopting a DTW algorithm, the triaxial data of the sensor are fused by utilizing a method of self-adaptive weighted fusion of related functions, and the fused data are classified into vehicle types by adopting an FWOA-GA-BP neural network algorithm, so that the detection precision of the system and the precision of vehicle type classification are improved.

Description

Vehicle information identification system and identification method
Technical Field
The invention belongs to the field of vehicle information detection, and particularly relates to a vehicle information identification system and an identification method.
Background
In recent years, around the problems of traffic jam, long commuting time, frequent traffic accidents, energy shortage caused by automobile fuel consumption, urban air quality reduction caused by automobile exhaust emission and the like caused by the rapid increase of the number of automobiles, the development of an intelligent traffic system is the best means for solving the current traffic problems. For the vehicle information identification technology, the acquisition of the vehicle information becomes a problem which must be solved first in the intelligent transportation.
At present, the main vehicle information identification technology is the annular coil vehicle information identification technology, the microwave vehicle information identification technology and the like. The annular coil vehicle information identifier is the vehicle information identifier with the largest use amount at present. A coil wound by a plurality of turns of conducting wires is buried under the road surface, the inductance of the induction coil can be changed when ferrous metal objects such as automobiles and the like pass through the induction coil, and the internal controller calculates and judges the change of the inductance of the induction coil, so that vehicle information is recognized, but the damage to the road is large. The microwave vehicle information recognizer is a radar-based non-inductive vehicle information recognizer. The microwave detection device is arranged on the road side of a multi-lane and emits modulated microwaves to cover a detection road surface during operation, so that a microwave band covering the multi-lane is formed. When a vehicle passes through the microwave band, the electromagnetic wave is reflected, and the passing of the vehicle can be detected by extracting a difference frequency signal in the reflected wave. In addition, the conventional detection method has low detection accuracy and low vehicle type classification accuracy.
Disclosure of Invention
The present invention provides a vehicle information identification system and an identification method, which aim to solve the above technical problems.
In order to solve the technical problems, the invention solves the problems by the following technical scheme:
a vehicle information identification system comprises a geomagnetic sensor, a communication module and an upper computer;
the geomagnetic sensors are arranged under the road surface at intervals and are used for acquiring geomagnetic disturbance information;
a plurality of earth magnetism sensor intervals set up under the road surface, and earth magnetism sensor is used for gathering earth magnetism disturbance information, and earth magnetism sensor sends the earth magnetism disturbance information of gathering for the host computer through communication module, and the host computer receives earth magnetism disturbance information, and the host computer is used for carrying out analysis processes output vehicle type classification result to the earth magnetism disturbance information of receiving.
A vehicle information identification method comprising the steps of:
step 1: the geomagnetic sensor collects geomagnetic disturbance information and sends the geomagnetic disturbance information to the upper computer, and the upper computer receives the geomagnetic disturbance information and judges whether a vehicle passes through the geomagnetic sensor or not;
step 2: if a vehicle passes through the geomagnetic disturbance information, the upper computer preprocesses the geomagnetic disturbance information;
and step 3: the upper computer eliminates time difference of the preprocessed data of two adjacent geomagnetic sensors, and performs feature point alignment on the data after the time difference is eliminated by adopting a DTW algorithm;
and 4, step 4: the upper computer adopts a self-adaptive weighting fusion method to fuse the data after the feature points are aligned;
and 5: and the upper computer extracts the vehicle characteristics according to the fused data and classifies the vehicle types by adopting an FWOA-GA-BP algorithm.
Further, in step 1, the upper computer judges whether a vehicle passes through the geomagnetic sensor by using a state vehicle detection algorithm based on a dynamic threshold.
Further, in step 2, the pretreatment process comprises: firstly, performing moving average filtering processing on geomagnetic disturbance information, then performing three-axis data normalization processing, and finally performing baseline tracking processing.
Further, in the step 3, the upper computer eliminates the time difference of the preprocessed data of the two adjacent geomagnetic sensors by using a correlation coefficient method;
and (3) performing feature point alignment on the data after the time difference is eliminated by adopting a DTW algorithm, wherein the specific method comprises the following steps:
step 3.1: the time series of two adjacent geomagnetic sensors are defined as Q and C respectively, and Q is Q1,q2...qn，C＝c1,c2...cmThe time sequence lengths of two adjacent geomagnetic sensors are respectively N and M, and an N multiplied by M distance matrix is constructed by utilizing Q and C;
defining matrix elements (i, j) as qiAnd cjDistance d (q) between two pointsi,cj)，d(qi,cj)＝d(i,j)＝(qi-cj)2Wherein q isiI-th element representing a time series Q, cjRepresents the jth element of the time series C;
defining W as a regular path, specifically: w (k) ═ w1,w2,...,wk,...,wKMax (M, N) ≦ K ≦ M + N-1, and the kth element of W is defined as W (K) ═ i, j)k＝(i(k),j(k))；
Step 3.2: using d as defined in step 3.1(qi,cj) And W (k) calculating the minimum overall matching distance between two adjacent geomagnetic sensor time sequences:
Figure BDA0002176480570000031
wherein d (W (k)) ((q)) isi-cj)2；
Define f (k) as the weighting factor in the warping path, f (k) i (k) -i (k-1)]+[j(k)-j(k-1)]，i(1)＝j(1)＝1，i(K)＝n，j(K)＝m，
Figure BDA0002176480570000032
The denominator in equation 1 is constant, thus minimizing the result of equation 1, the condition satisfied is to minimize the numerator in equation 1, i.e., to minimize the numerator in equation 1
Figure BDA0002176480570000033
Minimum;
since the regular path before any point is independent of the regular path after this point,
Figure BDA0002176480570000034
to:
g (W (k)), (k)) + min g (W (k-1)) (equation 2)
And the formula 2 is a calculation method from the k-1 step to the k step, and combines the formula 2 with different search modes to find out a path with the minimum overall matching distance so as to complete the alignment of the characteristic points of the geomagnetic curves of two adjacent geomagnetic sensors.
Further, in step 5, the upper computer extracts vehicle features according to the fused data and then classifies the vehicle types by adopting an FWOA-GA-BP algorithm, specifically:
optimizing the initial weight and the threshold of the BP neural network, taking an individual in the FWOA-GA as the weight and the threshold in the BP neural network, taking the fitness value as a training error in the BP neural network, and selecting the optimal initial value and the threshold of the BP neural network; and extracting a vehicle feature vector on the fused vehicle geomagnetic disturbance curve, taking vehicle feature vector sample data as the input of the optimized BP neural network, continuously training a vehicle feature vector sample data set through an optimal initial value and a threshold value in the BP neural network, and classifying the vehicle types.
Compared with the prior art, the invention has at least the following beneficial effects: according to the vehicle information identification system, the wireless geomagnetic sensor has extremely high sensitivity when detecting the vehicle, the vehicle identification performance is excellent through slight observation on the change of the magnetic field, and the accuracy of data acquisition can be guaranteed to the greatest extent. The wireless geomagnetic sensor is not buried underground by drilling and breaking ground, but directly and firmly adhered to the ground surface, so that the damage to the ground surface is avoided, the construction difficulty is greatly reduced, the construction period is shortened, the installation cost is saved, and the equipment can be quickly and effectively put into use. The geomagnetic detection technology is not interfered by external electromagnetic waves, can be normally used in thunderstorm days, has excellent waterproof performance and can work all the day.
The invention relates to a vehicle information identification method.A host computer eliminates time difference of preprocessed data of two adjacent geomagnetic sensors, and performs characteristic point alignment on the data after the time difference is eliminated by adopting a DTW algorithm; the upper computer adopts a self-adaptive weighting fusion method to fuse the data after the feature points are aligned; and the upper computer extracts the vehicle characteristics according to the fused data and classifies the vehicle types by adopting an FWOA-GA-BP neural network algorithm. Feature points in the vehicle data are aligned by adopting a DTW algorithm, the triaxial data of the sensor are fused by utilizing a method of self-adaptive weighted fusion of related functions, the vehicle types are classified by adopting an FWOA-GA-BP neural network algorithm after the vehicle features are extracted from the fused data, and the detection precision of the system and the precision of vehicle type classification are improved.
In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic block diagram of a system configuration of a vehicle information recognition system of the present invention;
FIG. 2 is a flow chart of vehicle data processing and traffic parameter calculation for the vehicle information identification system of the present invention;
fig. 3 is a flow chart of data transmission performed by the upper computer and the data acquisition module of the vehicle information identification system of the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
A vehicle information identification system comprises a geomagnetic sensor, a communication module and an upper computer; the geomagnetic sensors are arranged under the road surface at intervals and are used for acquiring geomagnetic disturbance information; the geomagnetic disturbance information is sent to an upper computer through a communication module; the upper computer is used for analyzing and processing the geomagnetic disturbance information and outputting vehicle type classification and other vehicle flow parameter results.
As a preferred embodiment of the present invention, as shown in fig. 1, the vehicle information identification system includes a geomagnetic data acquisition terminal, a wireless transmission module, and an upper computer data processing module. The geomagnetic data acquisition terminal is formed by placing a plurality of geomagnetic sensors below different lane roads in pairs, acquiring change information of the geomagnetic field at the node sensors when a vehicle passes through the centers of the different lane roads, and sending data in the node sensors to upper computer processing software through a router after the vehicle passes through the center. The wireless transmission module is completed by adopting a router, is installed on a lamp pole or a road upright pole, and has the main functions of node data connection, data forwarding and transmission distance increase. The upper computer data processing module is used for remotely connecting and controlling the geomagnetic detector of the vehicle, identifying and preprocessing data transmitted by the router, and calculating traffic flow, lane occupancy, vehicle speed, vehicle running direction and vehicle type traffic parameters. The data transmission is completed by connecting the router with the MCU CC3200 microcontroller in the node sensor through a TCP/IP protocol.
The invention relates to a vehicle information identification method, which comprises the following steps:
step 1: the method comprises the steps that a geomagnetic sensor collects geomagnetic disturbance information of a vehicle and sends the geomagnetic disturbance information of the vehicle to an upper computer, the upper computer receives the geomagnetic disturbance information of the vehicle and judges whether the vehicle passes through the geomagnetic sensor, and specifically, the upper computer judges whether the vehicle passes through the geomagnetic sensor by using a state locomotive detection algorithm based on a dynamic threshold value;
step 2: if there is the car process, the host computer carries out the preliminary treatment to earth magnetism disturbance information, and is specific, and the preliminary treatment process includes: firstly, performing moving average filtering processing on geomagnetic disturbance information, then performing three-axis data normalization processing, and finally performing baseline tracking processing;
and step 3: the upper computer adopts a correlation coefficient method to eliminate time difference of the preprocessed data of two adjacent geomagnetic sensors, and adopts a DTW algorithm to align feature points of the data after the time difference is eliminated, and the specific method comprises the following steps:
step 3.1: the time series of two adjacent geomagnetic sensors are defined as Q and C respectively, and Q is Q1,q2...qn，C＝c1,c2...cmThe time sequence lengths of two adjacent geomagnetic sensors are respectively N and M, and an N multiplied by M distance matrix is constructed by utilizing Q and C;
defining matrix elements (i, j) as qiAnd cjDistance d (q) between two pointsi,cj)，d(qi,cj)＝d(i,j)＝(qi-cj)2Wherein q isiI-th element representing a time series Q, cjRepresents the jth element of the time series C;
defining W as a regular path, specifically: w (k) ═ w1,w2,...,wk,...,wKMax (M, N) ≦ K ≦ M + N-1, where the kth element of W is defined as Wk＝(i,j)k＝(i(k),j(k))；
Step 3.2: using d (q) as defined in step 3.1i,cj) And W (k) calculating the minimum overall matching distance between two adjacent geomagnetic sensor time sequences:
Figure BDA0002176480570000061
wherein d (W (k)) ((q)) isi-cj)2；
Define f (k) as the weighting factor in the warping path, f (k) i (k) -i (k-1)]+[j(k)-j(k-1)]，i(1)＝j(1)＝1，i(K)＝n，j(K)＝m，
Figure BDA0002176480570000062
The denominator in equation 1 is constant, thus minimizing the result of equation 1, the condition satisfied is to minimize the numerator in equation 1, i.e., to minimize the numerator in equation 1
Figure BDA0002176480570000063
Minimum;
since the regular path before any point is independent of the regular path after this point,
Figure BDA0002176480570000071
to:
g (W (k)), (k)) + min g (W (k-1)) (equation 2)
Formula 2 is a calculation method from the k-1 step to the k step, and combines the formula 2 with different search modes to find out a path with the minimum overall matching distance, so as to complete the alignment of the characteristic points of the geomagnetic curves of two adjacent geomagnetic sensors;
and 4, step 4: the upper computer adopts a self-adaptive weighting fusion method to fuse the data after the feature points are aligned;
and 5: the upper computer extracts vehicle features according to the fused data and then classifies the vehicle types by adopting an FWOA-GA-BP algorithm, and the vehicle type classification algorithm based on the FWOA-GA-BP specifically comprises the following steps:
FWOA is an optimization algorithm that introduces a feedback mechanism for the classical whale algorithm, the feedback of which is divided into two phases, namely the walk-around foraging phase, enclosing the feedback of the contraction and spiral predation phases. By utilizing a feedback mechanism among whales and introducing a feedback mechanism based on a mean individual and a standard deviation, the overall search capability and the later convergence speed of the basic whale algorithm are improved. GA is a kind of randomized search method which is obtained by means of the evolution law of biology world, and the optimal solution is searched by simulating the natural evolution process;
the FWOA-GA-BP algorithm combines the respective adequacy of the FWOA algorithm and the GA algorithm, optimizes the initial weight and the threshold of the BP neural network, takes individuals in the FWOA-GA as the weight and the threshold in the BP neural network, takes the fitness value as the training error in the BP neural network, performs the operations of selection, crossing, mutation and surrounding, bubble net and searching predation, and selects the optimal initial value and the threshold of the BP neural network; and extracting vehicle feature vectors from the fused vehicle geomagnetic disturbance curve, taking the optimal initial value and threshold value of the BP neural network as the input of vehicle feature classification, classifying the vehicle types, and selecting 500 samples in the collected vehicle data, wherein 400 training sample sets are used, and 100 testing sample sets are used. The number of the selected hidden layer neurons is 16, the initial population M is 1000 and 2000 respectively, and the maximum evolutionary times T are 1000 and 2000 respectively. The FWOA-GA-BP algorithm overcomes the defect that a BP neural network is easy to fall into local optimization, improves optimization searching capability, can reach an optimal solution more quickly, and has obvious improvement on vehicle type classification precision compared with the traditional BP neural network algorithm.
As shown in fig. 2, the upper computer software receives the data of the acquisition terminal and then analyzes the data. After receiving geomagnetic disturbance information of a vehicle collected by a geomagnetic sensor, firstly judging whether the vehicle passes by, and establishing a temporary storage area for storing vehicle information data for data playback; then, displaying and preprocessing the original data acquired by the geomagnetic sensor, wherein the preprocessing comprises data moving average filtering, triaxial data normalization and baseline tracking; secondly, extracting and counting vehicle data by each sensor node, comprehensively judging the detection result of the difference, and counting again for calculating traffic parameters such as vehicle speed, vehicle length and the like; then, eliminating time difference between sensors by adopting a correlation coefficient method, aligning characteristic points in vehicle data by adopting a DTW algorithm, and fusing triaxial data of the sensors by adopting a self-adaptive weighting fusion method; and finally, the FWOA-GA-BP neural network algorithm is adopted to classify the vehicle types, so that the vehicle classification precision is improved.
Fig. 3 shows a process of wireless data transmission based on a TCP/IP protocol according to the present invention, in which a CC3200 chip having a WIFI function is selected as a control chip, and data can be transmitted using WIFI. The CC3200 chip may play three roles in the WIFI network: station, AP, and P2P. The invention sets CC3200 as Station, which is Station and exists in terminal mode. The CC3200 microcontroller can be internally divided into three parts, namely an application microprocessor system, a WIFI network management system and a power management system. An additional MCU in the WIFI network management system is specially used for WIFI network management, and the working frequency is 480 MHz. The MCU can completely avoid the operation processing burden of the MCU in the application microcontroller system. The WIFI network management system can realize safe and quick network connection and support various internet protocols and networking modes. And the upper computer software is designed to be programmed by adopting a Labview environment.
The network protocol used by the connection between the CC3200 and the upper computer is a TCP/IP protocol, and in the connection between the upper computer and the CC3200, a listener is created through a TCP Listen.vi sub VI in the TCP protocol, the listener continuously checks a TCP connection request of a designated port of the local computer, and if the TCP connection is established, the TCP Listen.vi sub VI returns parameters such as a connection ID, an address of a remote client, a port number and the like. The connection ID is used as a unique identifier for connection between the upper computer and the sensor node, is used by two sub-VI (Read TCP. VI and Write TCP. VI) and is used for transmitting data with the sensor node, and the Close TCP. VI is used for closing the connection with the current sensor node.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention, and they should be construed as being included therein. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (5)

1. A vehicle information identification method is characterized in that a vehicle information identification system is applied, wherein the vehicle information identification system comprises a geomagnetic sensor, a communication module and an upper computer;
the system comprises a plurality of geomagnetic sensors, a communication module, an upper computer and a vehicle type classification result output module, wherein the geomagnetic sensors are arranged under a road surface at intervals and used for acquiring geomagnetic disturbance information;
the vehicle information identification method includes the steps of:
step 1: the geomagnetic sensor collects geomagnetic disturbance information and sends the geomagnetic disturbance information to the upper computer, and the upper computer receives the geomagnetic disturbance information and judges whether a vehicle passes through the geomagnetic sensor or not;
step 2: if a vehicle passes through the geomagnetic disturbance information, the upper computer preprocesses the geomagnetic disturbance information;
and step 3: the upper computer eliminates time difference of the preprocessed data of two adjacent geomagnetic sensors, and performs feature point alignment on the data after the time difference is eliminated by adopting a DTW algorithm;
and 4, step 4: the upper computer adopts a self-adaptive weighting fusion method to fuse the data after the feature points are aligned;
and 5: and the upper computer extracts the vehicle characteristics according to the fused data and classifies the vehicle types by adopting an FWOA-GA-BP algorithm.
2. A vehicle information identification method according to claim 1, characterized in that: in the step 1, the upper computer judges whether a vehicle passes through the geomagnetic sensor by using a state vehicle detection algorithm based on a dynamic threshold value.
3. A vehicle information identification method according to claim 1, characterized in that: in step 2, the pretreatment process comprises: firstly, performing moving average filtering processing on geomagnetic disturbance information, then performing three-axis data normalization processing, and finally performing baseline tracking processing.
4. A vehicle information identification method according to claim 1, characterized in that: in the step 3, the upper computer adopts a correlation coefficient method to eliminate the time difference of the preprocessed data of the two adjacent geomagnetic sensors;
and (3) performing feature point alignment on the data after the time difference is eliminated by adopting a DTW algorithm, wherein the specific method comprises the following steps:
step 3.1: the time series of two adjacent geomagnetic sensors are defined as Q and C respectively, and Q is Q1,q2...qn，C＝c1,c2...cmThe time sequence lengths of two adjacent geomagnetic sensors are respectively N and M, and an N multiplied by M distance matrix is constructed by utilizing Q and C;
defining matrix elements (i, j) as qiAnd cjDistance d (q) between two pointsi,cj)，d(qi,cj)＝d(i,j)＝(qi-cj)2Wherein q isiI-th element representing a time series Q, cjRepresents the jth element of the time series C;
defining W as a regular path, specifically: w (k) ═ w1,w2,...,wk,...,wKMax (M, N) ≦ K ≦ M + N-1, and the kth element of W is defined as W (K) ═ i, j)k＝(i(k),j(k))；
Step 3.2: using d (q) as defined in step 3.1i,cj) And W (k) calculating the minimum overall matching distance between two adjacent geomagnetic sensor time sequences:
Figure FDA0002938344780000021
wherein d (W (k)) ((q)) isi-cj)2；
Define f (k) as the weighting factor in the warping path, f (k) i (k) -i (k-1)]+[j(k)-j(k-1)]，i(1)＝j(1)＝1，i(K)＝n，j(K)＝m，
Figure FDA0002938344780000022
The denominator in equation 1 is constant, thus minimizing the result of equation 1, the condition satisfied is to minimize the numerator in equation 1, i.e., to minimize the numerator in equation 1
Figure FDA0002938344780000023
Minimum;
since the regular path before any point is independent of the regular path after this point,
Figure FDA0002938344780000024
to:
g (W (k)), (k)) + min g (W (k-1)) (equation 2)
And the formula 2 is a calculation method from the k-1 step to the k step, and combines the formula 2 with different search modes to find out a path with the minimum overall matching distance so as to complete the alignment of the characteristic points of the geomagnetic curves of two adjacent geomagnetic sensors.
5. A vehicle information identification method according to claim 1, characterized in that: in step 5, the upper computer extracts vehicle features according to the fused data and then classifies the vehicle types by adopting an FWOA-GA-BP algorithm, specifically:
optimizing the initial weight and the threshold of the BP neural network, taking an individual in the FWOA-GA as the weight and the threshold in the BP neural network, taking the fitness value as a training error in the BP neural network, and selecting the optimal initial value and the threshold of the BP neural network; and extracting a vehicle feature vector on the fused vehicle geomagnetic disturbance curve, taking vehicle feature vector sample data as the input of the optimized BP neural network, continuously training a vehicle feature vector sample data set through an optimal initial value and a threshold value in the BP neural network, and classifying the vehicle types.
CN201910780670.2A 2019-08-22 2019-08-22 Vehicle information identification system and identification method Active CN110491141B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910780670.2A CN110491141B (en) 2019-08-22 2019-08-22 Vehicle information identification system and identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910780670.2A CN110491141B (en) 2019-08-22 2019-08-22 Vehicle information identification system and identification method

Publications (2)

Publication Number Publication Date
CN110491141A CN110491141A (en) 2019-11-22
CN110491141B true CN110491141B (en) 2021-03-23

Family

ID=68551748

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910780670.2A Active CN110491141B (en) 2019-08-22 2019-08-22 Vehicle information identification system and identification method

Country Status (1)

Country Link
CN (1) CN110491141B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114141022B (en) * 2020-09-03 2023-05-23 丰图科技(深圳)有限公司 Emergency lane occupation behavior detection method and device, electronic equipment and storage medium
CN112348076B (en) * 2020-11-04 2023-06-27 长安大学 Multi-sensor data fusion method suitable for road information acquisition
CN113759333B (en) * 2021-07-14 2024-04-02 中国人民解放军空军预警学院 Wind turbine multipath echo jiggle parameter estimation method based on whale optimization algorithm

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101237876B1 (en) * 2011-05-09 2013-02-27 한국건설기술연구원 Methods and Systems which can classify vehicles class through the magnetic characteristic analysis of vehicles
CN103354032A (en) * 2013-06-22 2013-10-16 太仓博天网络科技有限公司 Wireless vehicle information detection system based on magnetoresistive sensor
CN105118297A (en) * 2015-08-19 2015-12-02 中国科学院上海微系统与信息技术研究所 System and method for detecting and identifying type of vehicle
US20170140645A1 (en) * 2015-11-06 2017-05-18 The Board Of Regents Of The University Of Oklahoma Traffic monitoring system
CN206249557U (en) * 2016-11-14 2017-06-13 苏州大学 Based on the vehicle detecting system that Magnetic Sensor and ultrasonic sensor are merged
CN107274679B (en) * 2017-06-06 2019-08-30 东莞理工学院 Vehicle identification method, device, equipment and computer readable storage medium
CN107886727B (en) * 2017-11-13 2020-04-14 深圳先进技术研究院 Geomagnetic sensor-based automobile classification method and system and electronic equipment

Also Published As

Publication number Publication date
CN110491141A (en) 2019-11-22

Similar Documents

Publication Publication Date Title
CN110491141B (en) Vehicle information identification system and identification method
CN110084094B (en) Unmanned aerial vehicle target identification and classification method based on deep learning
CN109636049A (en) A kind of congestion index prediction technique of combination road network topology structure and semantic association
CN102737510A (en) Real-time traffic condition acquisition method based on mobile intelligent terminal
CN112101088A (en) Automatic unmanned aerial vehicle power inspection method, device and system
CN112163574A (en) ETC interference signal transmitter identification method and system based on deep residual error network
CN110426037A (en) A kind of pedestrian movement track real time acquiring method under enclosed environment
CN112215427A (en) Vehicle driving track reconstruction method and system under condition of bayonet data loss
Zhang et al. Vehicle classification based on feature selection with anisotropic magnetoresistive sensor
CN111598028A (en) Method for identifying earth surface vegetation distribution based on remote sensing imaging principle
CN113596989A (en) Indoor positioning method and system for intelligent workshop
He et al. Improved waveform‐feature‐based vehicle classification using a single‐point magnetic sensor
CN105717529B (en) A kind of signal source locating method based on radio signal drive test performance number
CN113159371B (en) Unknown target feature modeling and demand prediction method based on cross-modal data fusion
Cui et al. Optimizing Internet of Things-based intelligent transportation system’s information acquisition using deep learning
CN105743756A (en) Frame detection method based on Adaboost algorithm in Wi-Fi system
CN116992305A (en) Weather forecast method and system based on big data
CN116311357A (en) Double-sided identification method for unbalanced bovine body data based on MBN-transducer model
Lin Low power consumption and reliability of wireless communication network in intelligent parking system
CN105321354A (en) Simple vehicle detection, classification and identification method based on geomagnetic sensor
CN115565388A (en) Traffic light control method based on multi-channel vehicle detection and three-dimensional feature labeling
CN112016539B (en) Signal identification method and device, electronic equipment and storage medium
CN113762043A (en) Abnormal track identification method and device
Dong et al. An anti-interference detection algorithm for parking monitoring systems
Liu et al. Multi-lane traffic flow monitoring and detection system based on video detection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant