CN114155714B - Motor vehicle license plate relay identification system and license plate relay identification method - Google Patents
Motor vehicle license plate relay identification system and license plate relay identification method Download PDFInfo
- Publication number
- CN114155714B CN114155714B CN202210041528.8A CN202210041528A CN114155714B CN 114155714 B CN114155714 B CN 114155714B CN 202210041528 A CN202210041528 A CN 202210041528A CN 114155714 B CN114155714 B CN 114155714B
- Authority
- CN
- China
- Prior art keywords
- license plate
- processor
- information
- data
- image
- 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
Images
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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- 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
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a motor vehicle license plate relay identification system and a license plate relay identification method, which comprise a distribution terminal, a bayonet base station and a central device, wherein the distribution terminal is used for collecting and processing image information; the traffic gate base station is fixedly arranged at each traffic gate, and is used for collecting and processing image information on one hand and transferring the information of the distributed terminal and the central equipment on the other hand; the central equipment is used for managing and coordinating all the distributed terminals and the checkpoint base stations in the system on one hand, and predicting the position of a target license plate according to the received license plate information on the other hand. The system realizes the relay recognition of the motor vehicle license plate by predicting and calculating the cross-mirror position of the target vehicle and circularly executing the relay recognition process of the license plate, can greatly improve the real-time performance of recognition and positioning of the target motor vehicle license plate, and has wide practical prospect.
Description
The invention relates to a relay identification system and a relay identification method for a motor vehicle license plate, which is a divisional application of an invention patent with the application number of 2020109704458.
Technical Field
The invention relates to the field of computer and intelligent identification, in particular to a relay identification system and a relay identification method for a motor vehicle license plate.
Background
The license plate number of the motor vehicle can be accurately obtained in real time in the whole process of the traffic system, and the method has important significance for realizing and managing the intelligent traffic system.
At present, image recognition of a license plate number of a target motor vehicle mainly depends on a city monitoring system built in a traffic management department, if a motor vehicle needs to be tracked and the license plate number of the target motor vehicle needs to be located, only license plate number information of the target motor vehicle can be input, then the license plate number information of the target motor vehicle mainly depends on computing resources of a data center to be recognized and searched from massive images, the license plate number information is recognized from massive images, so that the target motor vehicle can be searched within a long time, the target vehicle still runs in a searching time period, and after the position information of the motor vehicle of the target license plate is searched, the target motor vehicle already drives away from the searched previous position for a long distance, so that the current license plate recognition system and method cannot efficiently track the target vehicle in real time. Meanwhile, the existing monitoring bayonet density is difficult to support the requirement of obtaining high-frequency license plate information.
Disclosure of Invention
The invention discloses a relay identification system and a relay identification method for motor vehicle license plates based on cross-mirror position prediction, and aims to solve the technical problems that the existing monitoring checkpoint density is difficult to support the requirement of high-frequency license plate information acquisition and the target motor vehicle license plate cannot be positioned efficiently in real time at present, so as to realize the technical effect of identifying and positioning the target motor vehicle license plate in real time.
The technical scheme of the invention is as follows:
a relay identification system for motor vehicle license plates comprises a distribution terminal, a bayonet base station and a central device, wherein the distribution terminal is used for collecting and processing image information; the traffic gate base station is fixedly arranged at each traffic gate, and is used for collecting and processing image information on one hand and transferring the information of the distributed terminal and the central equipment on the other hand; the central equipment is used for managing and coordinating all the distributed terminals and the gate base stations in the system on one hand, and predicting the position of the target license plate according to the received license plate information on the other hand.
Furthermore, the device comprises a first camera, a first clock module, a first memory, a first positioning chip, a first processor, a first GPU and a first communication module;
the first camera is used for collecting image information and is in communication connection with the first processor;
the first clock module is used for system timing and accurate timing, is in communication connection with the first processor and can communicate data;
the first memory is used for storing data information, and the memory is in data connection with the first processor and can communicate data;
the first positioning chip is used for generating and recording the position information of the distributed terminal, and the positioning chip is in data connection with the first processor and can communicate data;
the first processor runs a processing program and is used for processing data and communication, and the processor is in data connection with the first GPU and can communicate data;
the first GPU runs an image processing program used for license plate recognition and can recognize license plate numbers in the images;
the first communication module is used for communication processing, and the communication module is in data connection with the first processor and can intercommunicate data.
Further, the bayonet base station comprises a second camera, a second clock module, a second memory, a second positioning chip, a second processor, a second GPU and a second communication module;
the second camera is used for collecting image information; the second camera is in communication connection with the second processor;
the second clock module is used for system timing and accurate timing, and is in communication connection with the second processor and can communicate data with each other;
the second memory is used for storing data information, and the second memory and the second processor have data connection and can communicate data;
the second positioning chip is used for generating and recording the position information of the distributed terminal, and the second positioning chip is in data connection with the second processor and can intercommunicate data;
the second processor runs a processing program and is used for processing data and communication, and the second processor is in data connection with the second GPU and can intercommunicate data;
the second GPU runs an image processing program used for license plate recognition and can recognize license plate numbers in the images;
the second communication module is used for communication processing, and the second communication module is connected with the second processor through data and can intercommunicate data.
Further, the central device comprises a third communication module, a clock coordinator, a storage module, a third processor, a fourth communication module, an interface module and a fourth processor;
the third communication module is used for data communication between the central equipment and the bayonet base station, and the third communication module is in data connection with the third processor and can communicate data with each other;
the clock coordinator is used for accurately timing and calibrating the time with a second clock module of the bayonet base station; the clock coordinator and the third processor are in data connection and can communicate data;
the storage module is used for storing various identification result data; the storage module and the third processor are in data connection and can communicate data;
the third processor runs a processing program, receives the target license plate information transmitted by the fourth communication module or the interface module on one hand, processes the license plate information and then transmits the processed license plate information to the fourth processor; on the other hand, the third processor calls a processing result of the fourth processor for quick relay identification; the third processor, the fourth communication module and the interface module are in data connection and can communicate data;
the fourth communication module is used for being in butt joint with the existing traffic gate monitoring system, sending a data request to the traffic gate monitoring system and receiving data returned by the traffic gate monitoring system;
the interface module is used for being in butt joint with a third-party system and receiving image data and a license plate relay identification request sent by the third-party system;
the fourth processor runs a target motor vehicle position prediction program based on image recognition and can predict the position of the target motor vehicle in the next time period according to the picture and the time of the target motor vehicle; the fourth processor and the third processor are in data connection and can intercommunicate data.
Furthermore, the communication connection mode comprises a data bus, a data line, a cable, a network cable, wifi and Bluetooth.
A vehicle license relay identification method based on the motor vehicle license relay identification system comprises the following steps that the central equipment receives image data of an existing traffic access monitoring system or image data and a vehicle license relay identification request sent by a third-party system through a fourth communication module or an interface module, wherein the image data comprises an IMG (image IMG) of a target motor vehicle license s Number plate text information Num s And a shooting time t s And location information l s Recording image data complete data packet as TC s ={IMG s ,Num s ,t s ,l s A fourth communication module or an interface module receives the data packet TC s Sending the received TC to a third processor s Sending to the storage module for storage, and on the other hand, sending TC s Sending the data to a fourth processor, and receiving the TC by the fourth processor s After information, performing cross-mirror position prediction calculation to obtain a predicted bayonet base station address set, and sending the predicted bayonet base station address set to the third processor and simultaneously sending the predicted bayonet base station address set to the storage module for storage by the fourth processor;
the third processor receives the address set of the gate base stations and then sends target license plate texts to all the gate base stations contained in the set, a second communication module in each gate base station receives the target license plate texts and then sends target license plate text information to the second processor, the second processor sends the target license plate text information to a second memory for storage on one hand and sends the text information to a second GPU on the other hand, a license plate recognition algorithm is operated in the second GPU, the license plate text information is recognized in real time according to the image information collected by a second camera, and whether the license plate text information is the target license plate or not is judged;
the second processor broadcasts target license plate text information to all distribution terminals near the gate base station through the second communication module, after the distribution terminals receive the target license plate text information, the first processor sends the target license plate text information to the first memory for storage on one hand, and sends the text information to the first GPU on the other hand, a license plate recognition algorithm is operated in the first GPU, the license plate text information is recognized in real time according to the image information collected by the first camera, whether the target license plate is judged, and if the target license plate is judged, relevant image information including but not limited to time, image information and geographic information of the distribution terminals or the gate base station is sent to the gate base station;
and after the checkpoint base station identifies the target license plate or obtains the related information of the target license plate returned by the nearby distribution terminals, the checkpoint base station sends the related image information to the central equipment.
Further, the cross-mirror position prediction calculation specific processing comprises the following steps:
s1, determining an image sampling period of cross-mirror position prediction;
s2, sampling the target recognition image according to a sampling period, and predicting and calculating the cross-mirror position information.
Further, in step S1, a reasonable image sampling period is determined, and the following method is adopted:
the central equipment broadcasts the latest piece of picture information in the storage module to all the base stations of the card ports in the system through the third communication module to serve as heartbeat information, and then waits for response signals of the base stations of the card ports; after receiving the heartbeat information, the bayonet base station immediately replies a response signal A to the central equipment on one hand, and on the other hand, the bayonet base station broadcasts the heartbeat information to the distributed terminals, waits for the response signal of the distributed terminals, and after receiving the response signal B of the distributed terminals, the bayonet base station replies a response signal B to the central equipment again;
in each system heartbeat process, the time when the center equipment broadcasts heartbeat information to the bayonet base station is taken as a timing origin, and the time when the center equipment receives the response signal A is taken as1≤i≤m 1 ,Indicating the reception of the response signal A of the ith base station, m1 indicates a weekThe number of the answer signals A of the base station of the card port received by the central equipment in the period is more than or equal to 1 and less than or equal to m1 and more than or equal to n 2 ,n 2 The number of the card interface base stations in the system; the central equipment receives the response signal B at the moment1≤j≤m 2 ,The central equipment receives response signals B of the jth bayonet base station, m2 represents the number of the response signals B of the bayonet base stations received by the central equipment in one period, and m2 is more than or equal to 1 and less than or equal to m1;
to be in the period T 1 M1 base stations of the bayonet with normal center jump are taken as the reference, and the total number of the base stations of the bayonet in the system is n 2 Taking the heart beat normality rate of the equipmentSlave cycle T again 2 Middle according toTo determine the period T by taking the heart beat normal bayonet device 2 Intrinsic device heartbeat normalityTaking an image sample periodWherein mu is a partition coefficient, the value mu is more than or equal to 1, the larger the mu is, the more accurate the prediction result is, but the larger the value mu is, the higher the requirement on the calculation resources is, and d is the count of the accumulation calculation.
Further, the fourth processor is arranged to process the image of the target license plate from the storage module according to T X Sampling in a sampling period, recording the pixel coordinates of the license plate in each frame of sampled image as (X, y), and collecting the set X i =(x i ,y i ) (i =1,2, \ 8230;, n) is inputted as a training sample to the target detector, n is the number of pixels of the target license plate, the target detector learns the training sample, andthe cyclic shift of the target sample vector X may obtain a cyclic matrix X', that is:
the discrete Fourier transform matrix of the circulant matrix can be diagonalized intoWhere F is a constant matrix used to compute the Fourier transform, H represents the matrix conjugate transpose,a Fourier transform representing the vector X, diag being a diagonal matrix;
setting a linear regression function of the training samples as f (x) i )=ω T x i Where ω is a weight coefficient, which can be determined by least squaresSolving, delta is used for controlling the structural complexity of the system, and omega = (x) is obtained H x+δ) - 1 x H y, where x = { x 1 ,x 2 ,…,x n } T T represents a transposition calculation;
introducing a kernel function under the condition of nonlinear regression, mapping the calculation of a low-dimensional space to a high-dimensional kernel space, and setting a nonlinear mapping function of a sample asObtaining an optimal solution of ridge regressionα i The coefficients of the ridge regression are taken as the coefficients,wherein α is α i I is the identity matrix, sinceIs a circulant matrix, the solution of the ridge regression can be found as Is a nuclear correlation matrixThe first line of (2), the detection sample is a sample set z obtained by the license plate coordinates in the previous frame image and the cyclic transfer of the license plate coordinates j =P j z, wherein P is a permutation matrix, and z is a license plate coordinate pixel matrix of the previous frame;
for an input license plate image, the tracker response isThe sample coordinate of the maximum value is output by the tracker as the new target coordinate Y = minf (z) j ) Predicting the position coordinates of the license plate of the next frame; continuously learning through training samples, comparing the new target coordinate Y with the actual license plate coordinate Y' of the next frame, updating ridge regression coefficients,updating iteratively until the error between the new target coordinate Y and the actual license plate coordinate Y' of the next frame is smaller than a preset error threshold;
and inquiring a position dictionary of the checkpoint base station according to the image prediction result to obtain an address set of the checkpoint base station corresponding to the cross-mirror position and the geographical position of the target vehicle possibly appearing in the next period.
The invention has the beneficial effects that:
the system performs prediction calculation on the cross-mirror position of the target vehicle, fully utilizes the information processing capacity of the distribution terminal and the bayonet base station, circularly executes the relay recognition process of the license plate, realizes relay recognition of the motor vehicle license plate, can greatly improve the real-time performance of recognition and positioning of the target motor vehicle license plate, realizes the purpose of real-time recognition and positioning of the target motor vehicle license plate from mass image information which is not based on database query, and reduces the pressure of calculation resources of a data center.
Drawings
FIG. 1 is a block diagram of a relay identification system for a motor vehicle license plate according to the present invention;
FIG. 2 is a block diagram of a distributed terminal system according to the present invention;
FIG. 3 is a block diagram of a Bayonet base station system according to the present invention;
FIG. 4 is a block diagram of a system of central facilities according to the present invention;
in the figure: 10-a distributed terminal, 101-a first camera, 102-a first clock module, 103-a first memory, 104-a first positioning chip, 105-a first processor, 106-a first GPU, 107-a first communication module; 20-bayonet base station, 201-second camera, 202-second clock module, 203-second memory, 204-second positioning chip, 205-second processor, 206-second GPU, 207-second communication module; 30-center device, 301-third communication module, 302-clock coordinator, 303-storage module, 304-third processor, 305-fourth communication module, 306-interface module, 307-fourth processor.
Detailed Description
The invention discloses a relay identification system for motor vehicle license plates, which comprises a distribution terminal, a bayonet base station and a central device, and provides a relay identification method for license plates based on a cross-mirror position prediction technology on the basis of the identification system, which mainly comprises the following steps: the central equipment receives the license plate number information which is TC and wants to relay the identification s ={IMG s ,Num s ,t s ,l s And (4) after the target license plate vehicle image, the license plate number, the time and the geographic position are obtained, position prediction is carried out on the target license plate (vehicle). The specific process is as follows:
firstly, the central equipment receives the license plate number information TC which wants to relay to identify s Predicting a position of a target license plate under a camera, and predicting a group of positions (namely, predicting a cross-mirror position, and narrowing a data range to be searched by the target license plate); then, according toPredicted locations, broadcasting the target license plate number to be identified to the gate base station devices at the locations; secondly, on one hand, the checkpoint base station equipment identifies the shot license plate information, simultaneously broadcasts the license plate number to be identified to nearby distribution terminals, the distribution terminals identify the license plate information, and sends the picture, the license plate text, the time and the position information of the target license plate vehicle to the central equipment, so that one-time relay identification is completed, and the tracking of the current position of the target license plate is completed; and the central equipment repeats the steps and enters next relay identification according to the current position to realize real-time tracking.
According to the invention, a group of checkpoint base stations with the highest possibility is screened out through cross-mirror position prediction, and identification and information reporting are completed by the checkpoint base stations and the peripheral distribution terminals thereof, so that the real-time performance and the identification efficiency of continuous tracking and identification of the license plate can be greatly improved, and the defect that the vehicle cannot be tracked in real time because the searched data is very much and the time is long because mass data search needs to be carried out by relying on a data center in the traditional traffic checkpoint license plate identification is overcome.
Specifically, the cross-mirror position prediction refers to a camera image in which a target may appear in a multi-camera system, and is one of cross-mirror identification technologies (reids) in the field of vision processing.
The technical scheme of the application is described in detail below with reference to the accompanying drawings and embodiments, and with reference to fig. 1, the relay identification system for the motor vehicle license plate is composed of the following parts:
a distribution terminal 10, a bayonet base station 20 and a central device 30.
The distribution terminal 10 is installed at various vehicle or outdoor installation points, and collects and processes image information. The distribution terminal 10 includes a first camera 101, a first clock module 102, a first memory 103, a first positioning chip 104, a first processor 105, a first GPU106, and a first communication module 107.
The first camera 101 is configured to collect image information. The camera and the first processor 105 have a communication connection, the communication connection includes but is not limited to wired and wireless connection modes such as a data bus, a data line, a cable, a network cable, wifi, bluetooth and the like, and data can be communicated;
the first clock module 102 is used for system timing and accurate timing, and the clock module is in communication connection with the first processor 105 and can communicate data with each other;
the first memory 103 is used for storing data information such as images and processing results, and the memory is in data connection with the first processor 105 and can communicate data;
the first positioning chip 104 is used for generating and recording the position information of the distributed terminal, and the positioning chip is in data connection with the first processor 105 and can communicate data with each other;
the first processor 105 runs with a processing program for processing data and communication, and the processor is in data connection with the first GPU106 and can communicate data;
the first GPU106 is an image processing module, runs an image processing program for license plate recognition, and can recognize a license plate number in an image;
the first communication module 107 is used for communication processing, and the communication module has data connection with the first processor 105 and can communicate data with each other.
The first camera 101 collects image information and sends the image information to the first processor 105, and the image information is sent to the first memory 103 for storage after being compressed by the processor, wherein the compression process includes but is not limited to h.263/264/265 video coding and the like. The first processor 105 reads image information from the first memory 103 and sends the image information to the first GPU106, the GPU recognizes a license plate number from the received image, packages and sends an effective image IMG and license plate number text information Nc to the first processor 105, and the first processor 105 reads current time Te and position information L from the first clock module 102 and the first positioning chip 104, processes the information into a complete data packet, and sends the complete data packet to the first memory 13 for storage.
The gate base station 20 is fixedly installed at each traffic gate, and is used for collecting and processing image information on one hand, and is used for information transfer of the distribution terminal 10 and the center device 30 on the other hand. Referring to fig. 3, the bayonet base station includes: the system comprises a second camera 201, a second clock module 202, a second memory 203, a second positioning chip 204, a second processor 205, a second GPU206 and a second communication module 207.
The second camera 201 is configured to collect image information. The second camera 201 is in communication connection with the second processor 205, and the communication connection includes but is not limited to wired and wireless connection modes such as a data bus, a data line, a cable, a network cable, wifi, bluetooth and the like, and data can be communicated;
the second clock module 202 is configured to perform system timing and precise timing, and the second clock module 202 and the second processor 205 have a communication connection and may communicate data with each other;
the second memory 203 is used for storing data information such as images and processing results, and the second memory 203 is in data connection with the second processor 205 and can communicate data;
the second positioning chip 204 is used for generating and recording position information of the distributed terminal, and the second positioning chip 204 is in data connection with the second processor 205 and can communicate data with each other;
the second processor 205 runs a processing program for processing data and communication, and the second processor 205 has a data connection with the second GPU206 and can communicate data with each other;
the second GPU206 is an image processing module, runs an image processing program for license plate recognition, and can recognize a license plate number in an image;
the second communication module 207 is configured to perform communication processing, and the second communication module 207 and the second processor 205 have data connection and can communicate data with each other.
The processing procedure of the bayonet base station 20 for collecting and processing image information is as follows:
the second camera 201 collects image information and sends the image information to the second processor 205, and the image information is sent to the second memory 203 for storage after being compressed by the processor, wherein the compression process includes but is not limited to h.263/264/265 video coding and the like. The second processor 205 reads image information from the second memory 203 and sends the image information to the second GPU206, the GPU recognizes a license plate number from the received image, packages and sends an effective image IMG2 and license plate number text information Nc2 to the second processor 205, and the second processor 205 reads current time Te2 and position information L2 from the second clock module 202 and the second positioning chip 204, processes the information into a complete data packet, and sends the complete data packet to the second memory 203 for storage.
The central device 30 is configured to manage and coordinate all the distributed terminals 10 and the gate base stations 20 in the system, and predict a position of a target license plate according to the received license plate information, so as to guide the relevant distributed terminals and the gate base stations more quickly and identify the target license plate. The center device 30 includes: a third communication module 301, a clock coordinator 302, a storage module 303, a third processor 304, a fourth communication module 305, an interface module 306, and a fourth processor 307.
The third communication module 301 is configured to perform data communication between the central device 30 and the bayonet base station 20. The third communication module 301 has a data connection with the third processor 304, and can communicate data;
the clock coordinator 302 is configured to precisely time and calibrate with the second clock module 202 of the bayonet base station 20. The clock coordinator 302 is in data connection with the third processor 304 and can communicate data with each other;
the storage module 303 is configured to store various types of identification result data; the storage module 303 and the third processor 304 have data connection, and can communicate data;
the third processor 304 runs a processing program, on one hand, receives the target license plate information transmitted from the fourth communication module 305 or the interface module 306, processes the license plate information, and then sends the processed license plate information to the fourth processor 307; on the other hand, the third processor 304 calls the processing result of the fourth processor 307 for fast relay identification; the third processor 304 has data connection with the fourth communication module 305 and the interface module 306, and can communicate data with each other;
the fourth communication module 305 is configured to interface with an existing traffic gate monitoring system, send a data request to the traffic gate monitoring system, and receive data returned by the traffic gate monitoring system;
the interface module 306 is used for being in butt joint with a third-party system and receiving image data and a license plate relay identification request sent by the third-party system;
the fourth processor 307 runs a target vehicle position prediction program based on image recognition, and can predict the position of the target vehicle in the next time period according to the picture and the time of the target vehicle. The fourth processor 307 and the third processor 304 have data connection, and can communicate data.
The central device 30 receives the image data of the existing traffic checkpoint monitoring system or the image data and the license plate relay identification request sent by the third party system through the fourth communication module 305 or the interface module 306, where the image data includes the picture IMG of the target vehicle license plate s License plate text information Num s And a shooting time t s And location information l s Recording image data complete data packet as TC s ={IMG s ,Num s ,t s ,l s } the fourth communication module 305 or the interface module 306 transmits the received data packet TC s The third processor 304 sends the received TC to the third processor 304 s Sending to the storage module 303 for storage and on the other hand sending TC s Sent to the fourth processor 307, and the TC received by the fourth processor 307 s After information, cross-mirror position prediction calculation is carried out, and the specific processing process is as follows:
s1, determining an image sampling period of cross-mirror position prediction;
the reasonable image sampling period is the basis for ensuring the cross-mirror position prediction calculation accuracy and the overall recognition efficiency of the recognition system, and in order to combine the actual working condition of the recognition system, the reasonable image sampling period is determined by adopting the following method:
the center device 30 broadcasts the latest piece of picture information in the storage module 303 to all the bayonet base stations 20 in the system through the third communication module 301 as heartbeat information, and then waits for a response signal of the bayonet base stations 20. After receiving the heartbeat message, the bayonet base station 20 replies an answer signal a to the central device 30 immediately, and on the other hand, the bayonet base station 20 broadcasts heartbeat information to the distribution terminal 10, waits for the response signal of the distribution terminal 10, and after receiving the response signal of the distribution terminal 10, the bayonet base station 20 replies the response signal B to the center device 30 again. In each system heartbeat process, the time when the center device 30 broadcasts heartbeat information to the bayonet base station 20 is taken as a timing origin, and the time when the center device receives the response signal a is taken as Indicating that the response signal A of the ith base station is received, m1 indicating the number of the response signals A of the base stations of the bayonets received by the central equipment in one period, m1 being more than or equal to 1 and more than or equal to n 2 N2 is the number of the card interface base stations in the system, and the time when the central equipment receives the response signal B is(Indicating that the central device 30 receives the response signal B of the jth base station of the checkpoint, m2 indicating the number of the response signals B of the base stations of the checkpoint received by the central device in one period, and m2 being greater than or equal to 1 and less than or equal to m 1).
In order to ensure the response efficiency of the system, the sampling period of the cross-mirror position prediction should be within one heartbeat period T 2 In addition, a sufficient prediction data base is provided, so that the heartbeat information sent by the central device 30 is the target identification information, and meanwhile, the accurate cross-mirror prediction information is provided. The identification system is a typical distributed system, the number of the bayonet base stations 20 is large, when the image sampling period is determined, hundreds of coverage is not needed, and good effects can be obtained in the actual working conditions of the system only by paying attention to most bayonet base stations with normal heartbeat to determine the sampling period. The invention is to set the period T 1 M1 base stations of the bayonets with normal center jumps are taken as reference, and the total number of the base stations of the bayonets in the system is n 2 Taking the heart beat normality rate of the equipmentSlave cycle T again 2 Middle according toTo determine the period T by taking the heart beat normal bayonet device 2 Normal heartbeat of a deviceTaking an image sample periodWherein mu is a partition coefficient, the value mu is more than or equal to 1, the larger the mu is, the more accurate the prediction result is, but the larger the value mu is, the higher the requirement on computing resources is, and d is the count of the accumulation calculation.
And S2, sampling the target recognition image according to a sampling period, and performing predictive calculation on the cross-mirror position information.
The fourth processor 307 processes the image containing the target license plate from the storage module 303 according to T X Sampling is carried out in a sampling period, and the pixel coordinates of the license plate in each frame of sampled image are recorded as (x, y) (the lower left corner of the image is the origin of coordinates). Set X i =(x i ,y i ) (i =1,2, \ 8230;, n) is input into the target detector as a training sample, n is the number of pixels of the target license plate, the target detector learns the training sample, and the target sample vector X is circularly shifted to obtain a circular matrix X', namely:
according to the property of the circulant matrix, the discrete Fourier transform matrix can be diagonalized intoWhere F is a constant matrix used to compute the Fourier transform, H represents the matrix conjugate transpose,representing the fourier transform of vector X, diag is a diagonal matrix.
Setting a linear regression function of the training samples as f (x) i )=ω T x i Where ω is a weight coefficient, which can be determined by least squaresSolving, delta is used for controlling the structural complexity of the system, and omega = (x) is obtained H x+δ) - 1 x H y, where x = { x 1 ,x 2 ,…,x n } T And T denotes a transposition calculation.
Introducing a kernel function under the condition of nonlinear regression, mapping the calculation of a low-dimensional space to a high-dimensional kernel space, and setting a nonlinear mapping function of a sample asObtaining an optimal solution to ridge regressionα i The coefficients of the ridge regression are taken as the coefficients,wherein α is α i I is the identity matrix, sinceIs a circulant matrix, the solution of the ridge regression can be found as Is a nuclear correlation matrixThe first row of (2). The detection sample is a sample set z obtained by the license plate coordinates in the previous frame of image and cyclic transfer of the license plate coordinates j =P j z, whereinP is a permutation matrix, and z is a license plate coordinate pixel matrix of the previous frame. For an input license plate image, the tracker response isThe sample coordinate of the maximum value is output by the tracker as the new target coordinate Y = minf (z) j ) Namely, the position coordinates of the license plate of the next frame are predicted. Continuously learning through training samples, comparing the new target coordinate Y with the actual license plate coordinate Y' of the next frame, updating ridge regression coefficients,and (4) carrying out iterative updating until the error between the new target coordinate Y and the actual license plate coordinate Y' of the next frame is less than a preset error threshold. And inquiring a position dictionary of the checkpoint base station according to the image prediction result to obtain an address set of the checkpoint base station corresponding to the cross-mirror position and the geographical position of the target vehicle possibly appearing in the next period.
This completes the predictive calculation of the cross-mirror position.
The fourth processor 307 sends the predicted bayonet base station address set to the third processor 304, and at the same time sends the predicted bayonet base station address set to the storage module 303 for storage. The first processor receives the address set of the bayonet base station and then sends target license plate texts to all bayonet base stations 20 contained in the set, the second communication module 207 in the bayonet base station 20 receives the target license plate texts and then sends the target license plate text information to the second processor 205, and the second processor 205 sends the target license plate text information to the second memory 203 for storage on one hand and sends the text information to the second GPU206 on the other hand. The second GPU206 runs a license plate recognition algorithm, which is the prior art and will not be described in detail herein, and recognizes the text information of the license plate in real time according to the image information collected by the second camera 201, and determines whether the license plate is a target license plate.
The second processor 205 further broadcasts the target license plate text information to all the distribution terminals 10 near the affiliated gate base station 20 through the second communication module 207, and after the distribution terminals receive the target license plate text information, the first processor 105 sends the target license plate text information to the first memory 103 for storage on one hand, and sends the text information to the first GPU106 on the other hand. A license plate recognition algorithm is operated in the first GPU106, the license plate recognition algorithm is the prior art, and will not be described in detail herein, and the license plate text information is recognized in real time according to the image information acquired by the first camera 101, and whether the license plate is a target license plate is judged, and if the target license plate is judged, the relevant image information including but not limited to time, image information, and geographic information of a distribution terminal or a gate base station is sent to the gate base station.
When the gate base station recognizes the target license plate or obtains the related information of the target license plate returned by the nearby distribution terminals, the gate base station sends the related image information to the center device 30.
Thus, the relay identification of the license plate is completed once.
After obtaining the relevant information of the target license plate image, the central device 30 performs predictive calculation on the cross-mirror position of the target vehicle according to the method of S1-S2, and circularly executes the relay identification process of the license plate, thereby realizing the relay identification of the motor license plate. The motor vehicle license plate relay recognition system can greatly improve the real-time performance of recognition and positioning of the target motor vehicle license plate.
It should be understood that the above are only preferred embodiments of the present invention, and any modifications made based on the present invention or the technical idea of the present invention should be included in the protection scope of the present invention.
Claims (5)
1. A relay identification system for motor vehicle license plates is characterized by comprising a distribution terminal, a bayonet base station and a central device, wherein the distribution terminal is used for collecting and processing image information; the traffic gate base station is fixedly arranged at each traffic gate, and is used for collecting and processing image information on one hand and transferring the information of the distributed terminal and the central equipment on the other hand; the central equipment is used for managing and coordinating all distributed terminals and the gate base stations in the system on one hand, and for carrying out position adjustment on the target license plate according to the received license plate information on the other handRow prediction; the method specifically comprises the following steps that the central equipment receives image data of an existing traffic access monitoring system or image data and a license plate relay identification request sent by a third-party system through a fourth communication module or an interface module, wherein the image data comprises an IMG (image-in-g) picture of a target motor vehicle license plate s Number plate text information Num s And a shooting time t s And location information l s Recording image data complete data packet as TC s ={IMG s ,Wum s ,t s ,l s A fourth communication module or an interface module receives the data packet TC s Sending the received TC to a third processor s Sending to the storage module for storage, and on the other hand, sending TC s Sending to the fourth processor, the fourth processor receiving the TC s After information is received, cross-mirror position prediction calculation is carried out, wherein the cross-mirror position prediction means that in a plurality of camera systems, a prediction target possibly appears in a picture of a camera to obtain a predicted bayonet base station address set, and a fourth processor sends the predicted bayonet base station address set to a third processor and simultaneously sends the predicted bayonet base station address set to a storage module for storage;
the third processor receives the address set of the gate base stations and then sends target license plate texts to all the gate base stations contained in the set, a second communication module in each gate base station receives the target license plate texts and then sends target license plate text information to the second processor, the second processor sends the target license plate text information to a second memory for storage on one hand and sends the text information to a second GPU on the other hand, a license plate recognition algorithm is operated in the second GPU, the license plate text information is recognized in real time according to the image information collected by a second camera, and whether the license plate text information is the target license plate or not is judged;
the second processor broadcasts target license plate text information to all distribution terminals near the gate base station through the second communication module, after the distribution terminals receive the target license plate text information, the first processor sends the target license plate text information to the first memory for storage on one hand, and sends the text information to the first GPU on the other hand, a license plate recognition algorithm is operated in the first GPU, the license plate text information is recognized in real time according to the image information collected by the first camera, whether the target license plate is judged, and if the target license plate is judged, relevant image information including time, image information and geographic information of the distribution terminals or the gate base station is sent to the gate base station;
when the checkpoint base station identifies a target license plate or obtains related information of the target license plate returned by a nearby distribution terminal, the checkpoint base station sends the related image information to the central equipment;
the cross-mirror position prediction calculation specific processing comprises the following steps:
s1, determining an image sampling period of cross-mirror position prediction;
s2, sampling the target recognition image according to a sampling period, and performing predictive calculation on the cross-lens position information;
in the step S1, a reasonable image sampling period is determined, and the following method is adopted:
the central equipment broadcasts the latest piece of picture information in the storage module to all the base stations of the card ports in the system through the third communication module to serve as heartbeat information, and then waits for response signals of the base stations of the card ports; after receiving the heartbeat information, the bayonet base station immediately replies a response signal A to the central equipment, on the other hand, the bayonet base station broadcasts the heartbeat information to the distribution terminal, waits for the response signal of the distribution terminal, and after receiving the response signal B of the distribution terminal, the bayonet base station replies a response signal B to the central equipment again;
in each system heartbeat process, the time when the center equipment broadcasts heartbeat information to the checkpoint base station is recorded as a timing origin, and the time when the center equipment receives the response signal A is recorded as Indicating the reception of the response signal A, m of the ith base station 1 Indicating the number of the central equipment received the answer signals A of the base station in one period,1≤m 1 ≤n 2 ,n 2 The number of the card interface base stations in the system; the central equipment receives the response signal B at the moment Indicating that the central equipment receives the response signal B, m of the jth base station of the checkpoint 2 M is more than or equal to 1 and represents the number of the answer signals B of the base station of the card port received by the central equipment in one period 2 ≤m 1 ;
To be in the period T 1 Normal m of heartbeat 1 The base stations of the bayonets are taken as the benchmark, and the total number of the base stations of the bayonets in the system is n 2 Taking the heart beat normality rate of the equipmentThen from cycle T 2 Middle according toTo determine the period T by taking the heart beat normal bayonet device 2 Intrinsic device heartbeat normalityTaking an image sample periodWherein mu is a partition coefficient, the value mu is more than or equal to 1, the larger the mu is, the more accurate the prediction result is, but the larger the value mu is, the higher the requirement on computing resources is, and d is the count of the accumulation calculation;
the distribution terminal comprises a first camera, a first clock module, a first memory, a first positioning chip, a first processor, a first GPU and a first communication module;
the first camera is used for collecting image information and is in communication connection with the first processor;
the first clock module is used for system timing and accurate timing, and is in communication connection with the first processor and can communicate data with each other;
the first memory is used for storing data information, and the memory is in data connection with the first processor and can communicate data;
the first positioning chip is used for generating and recording the position information of the distributed terminal, and the positioning chip is in data connection with the first processor and can communicate data;
the first processor runs a processing program and is used for processing data and communication, and the processor is in data connection with the first GPU and can communicate data;
the first GPU runs an image processing program used for license plate recognition and can recognize license plate numbers in the images;
the first communication module is used for communication processing, and the communication module is in data connection with the first processor and can intercommunicate data;
the processing process of the distributed terminal for collecting and processing the image information comprises the following steps:
image information collected by the first camera is sent to the first processor, and is sent to the first memory for storage after being compressed by the processor, wherein the compression comprises H.263/264/265 video coding; the first processor reads image information from the first memory and sends the image information to the first GPU, after recognizing license plate numbers from received images, the GPU packs effective images IMG and license plate number text information Nc and sends the packed effective images IMG and license plate number text information Nc to the first processor, after reading current time Te and position information L from the first clock module and the first positioning chip, the first processor processes the information into complete data packets and sends the complete data packets to the first memory for storage.
2. The motor vehicle license plate relay identification system of claim 1, wherein the mount base station comprises a second camera, a second clock module, a second memory, a second positioning chip, a second processor, a second GPU, and a second communication module;
the second camera is used for collecting image information; the second camera is in communication connection with the second processor;
the second clock module is used for system timing and accurate timing, and is in communication connection with the second processor and can communicate data with each other;
the second memory is used for storing data information, and the second memory and the second processor have data connection and can communicate data;
the second positioning chip is used for generating and recording the position information of the distributed terminal, and the second positioning chip is in data connection with the second processor and can intercommunicate data;
the second processor runs a processing program and is used for processing data and communication, and the second processor is in data connection with the second GPU and can intercommunicate data;
the second GPU runs an image processing program used for license plate recognition and can recognize license plate numbers in the images;
the second communication module is used for communication processing, and the second communication module and the second processor have data connection and can intercommunicate data.
3. The relay identification system for motor vehicle license plates according to claim 2, wherein the central device comprises a third communication module, a clock coordinator, a storage module, a third processor, a fourth communication module, an interface module, and a fourth processor;
the third communication module is used for data communication between the central equipment and the bayonet base station, and the third communication module is connected with the third processor through data and can communicate data with each other;
the clock coordinator is used for accurately timing and calibrating the time with a second clock module of the bayonet base station; the clock coordinator and the third processor are in data connection and can communicate data;
the storage module is used for storing various identification result data; the storage module and the third processor are in data connection and can communicate data;
the third processor runs a processing program, receives the target license plate information transmitted by the fourth communication module or the interface module on one hand, processes the license plate information and then transmits the processed license plate information to the fourth processor; on the other hand, the third processor calls the processing result of the fourth processor for quick relay identification; the third processor, the fourth communication module and the interface module are in data connection and can communicate data;
the fourth communication module is used for being in butt joint with the existing traffic gate monitoring system, sending a data request to the traffic gate monitoring system and receiving data returned by the traffic gate monitoring system;
the interface module is used for being in butt joint with a third-party system and receiving image data and a license plate relay identification request sent by the third-party system;
the fourth processor runs a target motor vehicle position prediction program based on image recognition and can predict the position of the target motor vehicle in the next time period according to the picture and the time of the target motor vehicle; the fourth processor is in data connection with the third processor and can communicate data with each other;
the processing process of the bayonet base station for collecting and processing the image information comprises the following steps:
image information acquired by the second camera is sent to the second processor, and is sent to the second storage device for storage after being compressed by the processor, wherein the compression processing comprises but is not limited to H.263/264/265 video coding and the like; the second processor reads image information from the second memory and sends the image information to the second GPU, after the GPU identifies the license plate number from the received image, the GPU packages the effective image IMG2 and the license plate number text information Nc2 and sends the packaged effective image IMG2 and the license plate number text information Nc2 to the second processor, and after the second processor reads the current time Te2 and the position information L2 from the second clock module and the second positioning chip, the information is processed into a complete data packet and sent to the second memory for storage.
4. The relay identification system for motor vehicle license plates according to any of claims 1 to 3, wherein the communication connection means comprises a data bus, a data line, a cable, a network cable, wifi or Bluetooth.
5. The relay identification system for license plates of motor vehicles of claim 4, wherein the fourth processor derives T from the image of the storage module containing the target license plate X Sampling in a sampling period, recording the pixel coordinates of the license plate in each frame of sampled image as (x, y), and collecting the x i =(x i ,y i ) (i =1,2, \ 8230;, n) is input into the target detector as a training sample, n is the number of pixels of the target license plate, the target detector learns the training sample, and the target sample vector X is circularly shifted to obtain a circular matrix X', namely:
according to the property of the circulant matrix, the discrete Fourier transform matrix can be diagonalized intoWhere F is a constant matrix used to compute the Fourier transform, H represents the matrix conjugate transpose,represents the fourier transform of vector X, diag is the diagonal matrix;
setting the linear regression function of the training sample as f (x) i )=ω T x i Where ω is a weight coefficient, which can be determined by a least squares methodSolving, delta is used for controlling the structural complexity of the system, and omega = (x) is obtained H x+δ) -1 x H y, where x = { x 1 ,x 2 ,…,x n } T T represents a transposition calculation;
introducing a kernel function in the case of nonlinear regression, mapping the calculation of a low-dimensional space to a high-dimensional space, and establishing a nonlinear mapping function of the samples asObtaining an optimal solution of ridge regressionα i The coefficients of the ridge regression are taken as the coefficients,wherein α is α i I is the identity matrix, sinceIs a circulant matrix, the solution of the ridge regression can be found as
Is a nuclear correlation matrixThe first line of (2), the detection sample is a sample set z obtained by the license plate coordinates in the previous frame image and the cyclic transfer of the license plate coordinates j =P j z, wherein P is a permutation matrix, and z is a license plate coordinate pixel matrix of the previous frame;
for an input license plate image, the tracker responds asThe sample coordinate of the maximum value is output by the tracker as the new target coordinate Y = min f (z) j ) Predicting the position coordinates of the license plate of the next frame; continuously learning through training samples, comparing the new target coordinate Y with the actual license plate coordinate Y' of the next frame, updating ridge regression coefficients,iteratively updating untilThe error between the new target coordinate Y and the actual next frame license plate coordinate Y' is less than a preset error threshold value;
and inquiring a position dictionary of the checkpoint base station according to the image prediction result to obtain an address set of the checkpoint base station corresponding to the cross-mirror position and the geographical position of the target vehicle possibly appearing in the next period.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210041528.8A CN114155714B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210041528.8A CN114155714B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
CN202010970445.8A CN112071083B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010970445.8A Division CN112071083B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114155714A CN114155714A (en) | 2022-03-08 |
CN114155714B true CN114155714B (en) | 2022-11-08 |
Family
ID=73696452
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210041528.8A Active CN114155714B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
CN202010970445.8A Active CN112071083B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010970445.8A Active CN112071083B (en) | 2020-09-15 | 2020-09-15 | Motor vehicle license plate relay identification system and license plate relay identification method |
Country Status (1)
Country | Link |
---|---|
CN (2) | CN114155714B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112700647B (en) * | 2020-12-29 | 2022-09-16 | 杭州海康威视数字技术股份有限公司 | Method and device for monitoring vehicle driving information |
CN115691143B (en) * | 2022-12-30 | 2023-04-07 | 北京码牛科技股份有限公司 | Dynamic correction method and system for data acquisition time of traffic access point equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130088480A (en) * | 2012-01-31 | 2013-08-08 | 삼성테크윈 주식회사 | Integration control system and method using surveillance camera for vehicle |
CN104616494A (en) * | 2014-12-23 | 2015-05-13 | 浙江宇视科技有限公司 | Method and device for recording and determining target object based on base station and block port |
CN107424412A (en) * | 2017-09-21 | 2017-12-01 | 程丹秋 | A kind of traffic behavior analysis system |
CN108241737A (en) * | 2016-12-23 | 2018-07-03 | 三星电子株式会社 | For providing the system of location information and supporting the electronic device and method of the system |
CN111582253A (en) * | 2020-06-19 | 2020-08-25 | 上海其高电子科技有限公司 | License plate tracking and identifying method based on event triggering |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5227803A (en) * | 1992-07-22 | 1993-07-13 | Hughes Aircraft Company | Transponder location and tracking system and method |
GB9320045D0 (en) * | 1993-09-29 | 1993-11-17 | Ventress Jonathan | Sensor security system |
US6504895B2 (en) * | 2001-01-31 | 2003-01-07 | General Electric Company | Method and system monitoring image detection |
CN101303803A (en) * | 2008-06-11 | 2008-11-12 | 北京中星微电子有限公司 | Method and system for discriminating license plate |
US20140218529A1 (en) * | 2013-02-04 | 2014-08-07 | Magna Electronics Inc. | Vehicle data recording system |
CN104751639A (en) * | 2015-03-17 | 2015-07-01 | 上海视聪网络信息技术有限公司 | Big-data-based video structured license plate recognition system and method |
CN105139659A (en) * | 2015-10-09 | 2015-12-09 | 东方网力科技股份有限公司 | Vehicle license plate recognition method and device |
US9928667B2 (en) * | 2015-12-21 | 2018-03-27 | International Business Machines Corporation | Determining vehicle occupancy using sensors |
CN108305349A (en) * | 2017-01-11 | 2018-07-20 | 宁波轩悦行电动汽车服务有限公司 | A kind of electric vehicle car rental condition monitoring system and method |
CN108734967B (en) * | 2017-04-20 | 2021-09-28 | 杭州海康威视数字技术股份有限公司 | Method, device and system for monitoring illegal vehicle |
CN107393307A (en) * | 2017-08-17 | 2017-11-24 | 南京泊众智轩网络科技有限公司 | A kind of wireless Car license recognition parking management system |
CN107945514B (en) * | 2017-12-07 | 2020-11-17 | 广州坤扬电子科技有限公司 | Method and device for acquiring and positioning information of vehicle intelligent terminal |
CN208589142U (en) * | 2018-07-04 | 2019-03-08 | 珠海市特车网络科技有限公司 | A kind of control region car monitoring system |
CN109191856A (en) * | 2018-08-17 | 2019-01-11 | 江苏信息职业技术学院 | The method of vehicle tracking system and tracking vehicle based on big data |
CN110515073B (en) * | 2019-08-19 | 2021-09-07 | 南京慧尔视智能科技有限公司 | Multi-radar cross-regional networking multi-target tracking identification method and device |
CN110472607A (en) * | 2019-08-21 | 2019-11-19 | 上海海事大学 | A kind of ship tracking method and system |
-
2020
- 2020-09-15 CN CN202210041528.8A patent/CN114155714B/en active Active
- 2020-09-15 CN CN202010970445.8A patent/CN112071083B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130088480A (en) * | 2012-01-31 | 2013-08-08 | 삼성테크윈 주식회사 | Integration control system and method using surveillance camera for vehicle |
CN104616494A (en) * | 2014-12-23 | 2015-05-13 | 浙江宇视科技有限公司 | Method and device for recording and determining target object based on base station and block port |
CN108241737A (en) * | 2016-12-23 | 2018-07-03 | 三星电子株式会社 | For providing the system of location information and supporting the electronic device and method of the system |
CN107424412A (en) * | 2017-09-21 | 2017-12-01 | 程丹秋 | A kind of traffic behavior analysis system |
CN111582253A (en) * | 2020-06-19 | 2020-08-25 | 上海其高电子科技有限公司 | License plate tracking and identifying method based on event triggering |
Non-Patent Citations (1)
Title |
---|
基于原子交互的交通卡口可视分析系统;曹帝胄;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20190715;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN114155714A (en) | 2022-03-08 |
CN112071083B (en) | 2022-03-01 |
CN112071083A (en) | 2020-12-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108513251B (en) | Positioning method and system based on MR data | |
CN114155714B (en) | Motor vehicle license plate relay identification system and license plate relay identification method | |
CN109672973B (en) | Indoor positioning fusion method based on strongest AP | |
CN103068035A (en) | Wireless network location method, device and system | |
CN111935820B (en) | Positioning implementation method based on wireless network and related equipment | |
CN111479224B (en) | High-precision track recovery method and system and electronic equipment | |
CN106792549A (en) | Indoor locating system based on WiFi fingerprints and its stop pick-up navigation system | |
CN113115216B (en) | Indoor positioning method, service management server and computer storage medium | |
CN106304330A (en) | A kind of radio frequency tomography localization method alleviating background electromagnetic wave action | |
CN111586585A (en) | Position fingerprint positioning method based on Wifi | |
CN114758364B (en) | Industrial Internet of things scene fusion positioning method and system based on deep learning | |
CN116758454A (en) | Wetland bird monitoring method and system | |
Mukhtar et al. | Machine learning-enabled localization in 5g using lidar and rss data | |
CN113009538B (en) | Vehicle traffic positioning system | |
CN112835084B (en) | Power equipment positioning method and system based on power network scene and power equipment | |
CN117545047A (en) | Resident data-based NTN connection selection method, device, equipment and medium | |
CN110596739B (en) | Building identification system and method based on satellite positioning and WiFi module | |
CN113395763B (en) | NB terminal high-precision positioning method and system based on 5G/NB-IoT cluster node information fusion | |
CN116033345A (en) | Indoor abnormal signal high-precision positioning method, system and device | |
CN111654843A (en) | Method and system for automatically updating fingerprint database and wifi positioning method and system | |
CN115765848A (en) | 5G network signal strength detection system and use method | |
CN116996760A (en) | Video data processing method and device, computer readable medium and electronic equipment | |
CN113776491A (en) | Multi-dimensional distance measurement method, MEC and distance measurement unit based on B-M2M | |
CN114513741B (en) | Method and device for positioning terminal equipment based on measurement report | |
CN111343575B (en) | Indoor positioning method and equipment based on wireless access point signal intensity distribution |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20221017 Address after: 518000 25F, Block C, NEO Lvjing Plaza, No. 6009, Shennan Avenue, Tian'an Community, Shatou Street, Futian District, Shenzhen, Guangdong Applicant after: SHENZHEN FRAMER INFORMATION TECHNOLOGY CO.,LTD. Address before: 318000 room 1215, building 1, peninsula garden, Tengda Road, Lubei street, Luqiao District, Taizhou City, Zhejiang Province Applicant before: Taizhou yuanxingke Network Technology Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |