CN116913096A - Traffic situation investigation equipment and method based on Beidou short message communication technology - Google Patents
Traffic situation investigation equipment and method based on Beidou short message communication technology Download PDFInfo
- Publication number
- CN116913096A CN116913096A CN202311180551.6A CN202311180551A CN116913096A CN 116913096 A CN116913096 A CN 116913096A CN 202311180551 A CN202311180551 A CN 202311180551A CN 116913096 A CN116913096 A CN 116913096A
- Authority
- CN
- China
- Prior art keywords
- module
- traffic
- data
- short message
- sub
- 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.)
- Granted
Links
- 238000011835 investigation Methods 0.000 title claims abstract description 65
- 238000004891 communication Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000005516 engineering process Methods 0.000 title claims abstract description 21
- 238000012544 monitoring process Methods 0.000 claims abstract description 31
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 238000000605 extraction Methods 0.000 claims description 31
- 230000011218 segmentation Effects 0.000 claims description 28
- 238000013075 data extraction Methods 0.000 claims description 11
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 claims 1
- 230000005540 biological transmission Effects 0.000 abstract description 6
- 238000011161 development Methods 0.000 abstract description 3
- 230000004927 fusion Effects 0.000 abstract description 2
- 238000012549 training Methods 0.000 description 29
- 238000013528 artificial neural network Methods 0.000 description 17
- 239000013598 vector Substances 0.000 description 14
- 230000002452 interceptive effect Effects 0.000 description 6
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 241001255830 Thema Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000009429 distress Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/44—Event detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- 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
- G08G1/0141—Measuring and analyzing of parameters relative to traffic conditions for specific applications for traffic information dissemination
-
- 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
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/024—Guidance services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/90—Services for handling of emergency or hazardous situations, e.g. earthquake and tsunami warning systems [ETWS]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The application provides traffic condition investigation equipment and method based on Beidou short message communication technology. The apparatus includes: the traffic investigation module is used for monitoring the road and acquiring traffic data; the Beidou module is used for carrying out time service and positioning information on the investigation equipment; the communication module is used for sending the acquired traffic data and time service and positioning information to the cloud server; and the analysis scheduling module is used for acquiring traffic data from the cloud server and analyzing the traffic data. According to the application, the video acquisition and event identification are integrated on the road traffic condition investigation equipment, and the Beidou positioning, time service, SOS emergency help seeking and short message communication module is carried, so that the data transmission of road interchange sites is realized in a signal-free area or in an environment with interference factors, the equipment positioning, time service, event identification and SOS emergency help seeking functions are not affected, the level of road traffic condition investigation work is improved, and the fusion application development of Beidou and traffic fields is promoted.
Description
Technical Field
The application relates to the technical field of traffic investigation, in particular to traffic condition investigation equipment and method based on Beidou short message communication technology.
Background
Traffic parameters are the most important parameters describing traffic flow characteristics, and are the most important dynamic data in road traffic management work. The change and distribution rule of traffic parameters in time and space are known, necessary data are provided for traffic planning, road construction, traffic control and management, engineering economy analysis and the like, and therefore, the road traffic condition investigation work development status becomes an important mark for representing the national or regional road management level and informatization level.
The intelligent monitoring of traffic parameters is mainly completed by intermodulation equipment, but the current intermodulation equipment still has a plurality of problems for collecting traffic parameters, such as:
1. the identification means is single, and can not identify various traffic parameters or events;
2. at present, the domestic traffic condition investigation equipment does not have an automatic time service function, and after the equipment is operated for a period of time, the system time can deviate and is hardly perceived;
3. under the environment without signal areas or with interference factors, traffic condition investigation equipment cannot be constructed and operated.
Disclosure of Invention
The traffic condition investigation equipment and the traffic condition investigation method based on the Beidou short message communication technology are integrated based on the Beidou short message communication, positioning, timing and other functions in combination with the traffic condition investigation complete equipment, so that the multi-traffic parameter investigation system with accurate, reliable, stable and quick data acquisition, positioning, timing, event identification and SOS alarm help seeking integration is realized, and the technical problems in the process can be solved.
The technical scheme for solving the technical problems is as follows:
in a first aspect, the present application provides traffic situation investigation equipment based on a beidou short message communication technology, including:
the traffic investigation module is used for acquiring traffic data through road monitoring;
the Beidou module is used for carrying out time service and positioning information on the traffic investigation module;
the communication module is used for sending the acquired traffic data and time service and positioning information to the cloud server;
and the analysis scheduling module is used for analyzing the acquired traffic data and sending a corresponding control instruction according to the analysis result.
In some embodiments, the traffic investigation module comprises:
the traffic monitoring sub-module is used for monitoring road traffic conditions;
and the data extraction sub-module is used for extracting road and vehicle data.
In some embodiments, the data extraction submodule includes:
the video segmentation unit is used for carrying out key frame segmentation on the positions of the vehicles and the lanes in the monitoring picture and the image information according to a video segmentation algorithm;
and the feature extraction unit is used for extracting vehicle feature data from the video key frames according to the feature extraction model.
In some embodiments, the communication module comprises;
the wireless communication sub-module is used for transmitting traffic data to the cloud server according to 4G or 5G or VPN communication;
and the Beidou short message sub-module is used for transmitting traffic data to the cloud server through a short message function under the condition of losing wireless signals.
In some embodiments, the analysis scheduling module comprises:
the video clustering sub-module is used for classifying the video sequences according to a clustering algorithm and the characteristic data;
the event identification sub-module is used for identifying special events for each type of video sequence;
and the alarm sub-module is used for sending an alarm to the command center according to the special event.
In a second aspect, the application provides a traffic situation investigation method based on Beidou short message communication technology, which comprises the following steps:
s1: monitoring a road and acquiring traffic data through a traffic investigation module;
s2: the time service and positioning information are carried out on the traffic investigation module through the Beidou module;
s3: the communication module is used for sending the acquired traffic data, time service and positioning information to the cloud server;
s4: and acquiring traffic data from the cloud server through the analysis scheduling module, analyzing the traffic data, and sending a corresponding control instruction according to an analysis result.
In some embodiments, the S1 comprises:
s11: the road traffic condition is monitored through the traffic monitoring sub-module;
s12: and extracting road and vehicle data through the data extraction sub-module.
In some embodiments, the S12 includes:
s121: the method comprises the steps that key frame segmentation is carried out on positions of vehicles and lanes in a monitoring picture and image information through a video segmentation unit according to a video segmentation algorithm;
s122: and extracting vehicle characteristic data from the video key frames by a characteristic extraction unit according to the characteristic extraction model.
In some embodiments, the S3 comprises:
s31: transmitting traffic data to a cloud server through a wireless communication sub-module and through 4G or 5G or VPN communication;
s32: and under the condition of losing wireless signals, the traffic data is transmitted to the cloud server through the short message function through the Beidou short message sub-module.
In some embodiments, the S4 comprises:
s41: classifying the video sequences through a video clustering sub-module according to a clustering algorithm and feature data;
s42: the event identification sub-module is used for identifying special events for each type of video sequence;
s43: and sending an alarm to the command center through the alarm sub-module according to the special event.
The beneficial effects of the application are as follows:
according to the traffic situation investigation equipment and the traffic situation investigation method based on the Beidou short message communication technology, provided by the application, the video acquisition and event recognition are integrated on the highway traffic situation investigation equipment, and the Beidou positioning, time service, SOS emergency help seeking and short message communication modules are carried, so that the data transmission of highway traffic regulation sites and the equipment positioning and time service of the traffic situation investigation sites are ensured under the environments without signal areas/interference factors (severe weather areas, unstable network signals and even without network signals), the event recognition and SOS emergency help seeking functions are not influenced, the level of highway traffic situation investigation work is improved, and the fusion application development of Beidou and traffic fields is promoted.
Drawings
FIG. 1 is a flow chart of the method of the present application;
FIG. 2 is a sub-flowchart of step S1 of the present application;
FIG. 3 is a sub-flowchart of step S12 of the present application;
FIG. 4 is a sub-flowchart of step S3 of the present application;
fig. 5 is a sub-flowchart of step S4 of the present application.
Detailed Description
The principles and features of the present application are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the application and are not to be construed as limiting the scope of the application.
In order that the above-recited objects, features and advantages of the present application can be more clearly understood, a more particular description of the application will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It is to be understood that the depicted embodiments are some, but not all, embodiments of the present application. The specific embodiments described herein are to be considered in an illustrative rather than a restrictive sense. All other embodiments, which are obtained by a person skilled in the art based on the described embodiments of the application, fall within the scope of protection of the application.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
The traffic situation investigation equipment based on the Beidou short message communication technology provided by the application comprises the following components:
the traffic investigation module is used for monitoring the road and acquiring traffic data;
the Beidou module is used for carrying out time service and positioning information on the traffic investigation module;
the communication module is used for sending the acquired traffic data and time service and positioning information to the cloud server;
and the analysis scheduling module is used for acquiring traffic data from the cloud server, analyzing the traffic data and sending a corresponding control instruction according to an analysis result.
Specifically, at present, the domestic traffic situation investigation equipment does not have an automatic time service function, after the equipment operates for a period of time, the system time can deviate, the system time is difficult to be perceived, the monitoring data time of the traffic situation investigation equipment is inconsistent with the actual occurrence time, the timeliness of data resources is very low, meanwhile, the road and freight transportation belong to the accident high-rise field, the traffic situation investigation site has the automatic SOS alarm help seeking function and the manual alarm help seeking function of the Beidou system, and the traffic situation investigation site can provide alarm help seeking service for people in distress in the areas without network signal coverage in unmanned areas and remote areas. Therefore, the Beidou module is mainly assembled on the existing traffic investigation facility (namely, the traffic investigation module), so that the time service and the positioning of the traffic data can be performed by utilizing the time service and the positioning function of the Beidou module, and the accuracy of the acquired traffic data in positioning and time can be ensured.
In addition, the main control integrated circuit of the traffic condition investigation equipment based on the Beidou short message communication technology in the scheme can adopt a domestic autonomous controllable singlechip main control board, has high-performance processing capacity and abundant peripheral resources, and is suitable for various application scenes. The hardware circuit integrates the RS232, RJ-45, MCX antenna, IO and other interface application modules, the power supply module (the power supply voltage is less than or equal to DC36V, the power supply current is less than or equal to 2A), the 4G/5G communication module and the like. In addition, the double-channel double-mode communication output is formed by integrating Beidou modules (comprising a Beidou short message communication module, a Beidou time service module and a Beidou positioning module) on an original singlechip main control board through a circuit integration technology. The Beidou short message communication module provides short message satellite communication function for traffic condition investigation equipment, and the signal receiving and transmitting error rate is less than or equal to 1 multiplied by 10 -5 The signal interruption recovery is less than or equal to 30s, and the recapture time is less than or equal to 1s. The Beidou time service module realizes a time service function, the time service precision is more than or equal to 10ns, and the time synchronization precision is less than or equal to 5ns. The Beidou positioning module realizes the positioning function, and the positioning precision is less than or equal to 10m.
In some embodiments, the traffic investigation module comprises:
the traffic monitoring sub-module is used for monitoring road traffic conditions;
and the data extraction sub-module is used for extracting road and vehicle data.
In some embodiments, the data extraction submodule includes:
the video segmentation unit is used for carrying out key frame segmentation on the positions of the vehicles and the lanes in the monitoring picture and the image information according to a video segmentation algorithm;
and the feature extraction unit is used for extracting vehicle feature data from the video key frames according to the feature extraction model.
Specifically, the monitoring of the traffic situation in the present solution is completed by means of road monitoring (i.e. traffic monitoring sub-module) in the existing traffic investigation facility (i.e. traffic investigation module), and the acquired monitoring data also needs to be extracted by using the data extraction sub-module. The data extraction sub-module mainly comprises two parts of functions, one of the functions is video segmentation, the video segmentation is mainly completed by a video segmentation unit, and the video segmentation unit can carry out key frame segmentation on the positions of vehicles and lanes and image information in a monitoring picture according to a video segmentation algorithm. In the scheme, a video segmentation algorithm based on a time domain is adopted, and a differential image is obtained according to the difference between a current frame and a previous frame, so that a key frame is obtained. The other function is feature extraction, which is mainly completed by a feature extraction unit, wherein the feature extraction unit is a neural network feature extraction model, the model is trained by using the existing data in the traffic database as a training set, and the mapping of image data and vehicle feature data such as vehicle shape, outline, color, texture, size and the like is obtained, so that the vehicle feature information can be extracted according to the key frame.
More specifically, the specific process of extracting the vehicle feature information after the feature extraction unit performs training by using the existing data in the traffic database as the training set based on the neural network feature extraction model is as follows.
Firstly, constructing a training set formed by matching vehicle image samples and vehicle characteristic description information one by utilizing the existing data in the traffic database; the training set formed by matching the vehicle image sample and the vehicle characteristic description information one by one is expressed as follows:
wherein the method comprises the steps ofIs the total amount of the sample,is the firstA sample of the vehicle image is taken,is the image height direction pixel value of the vehicle image sample,is an image width direction pixel value of the vehicle image sample;is the first of training setMaximum of corresponding vehicle image samplesVehicle characteristic description information formed by the vehicle key characteristics and the vehicle key characteristic position frame positioning vectors;is an integer set in advance; wherein the method comprises the steps of ，Is the firstIn the individual vehicle image samplesThe key features of the individual vehicles are that,is the firstIn the individual vehicle image samplesVehicle key feature position frame corresponding to each vehicle key featureA 4-dimensional vector of center point coordinates, frame height values, and frame width values. The key features of the vehicle include license plates, car lamps, driver positions, front passenger positions and other key features of the vehicle according to the requirements of traffic investigation.
Initializing each layer of neural network configuration parameters of the neural network feature extraction model; the neural network feature extraction model comprises a ResNet model convolution layer structure and a full convolution network layer. ResNet model convolutional layer structure is represented asWhereinIs the parameter tensor of the parameters of all layers of the initialized res net model convolution layer,an input vehicle image for a ResNet model convolution layer; the full convolution network layer is represented asWherein the parameter tensorIs the parameter tensor formed by the parameters of all convolution kernels of the initialized full convolution network layer, and ，a feature map representing the low resolution provided by the ResNet model convolutional layer structure; as an alternative value, wherein,。
Further, the vehicle image sample is input into a neural network feature extraction model for training, and extracted by the neural network feature extraction modelVectorized vehicle characteristic information of the vehicle image sample is formed. Specifically, a vehicle image is sampledAs the input vehicle imageInputting the ResNet model convolution layer structure of the initialized neural network feature extraction model to obtain a low-resolution feature mapThe method comprises the steps of carrying out a first treatment on the surface of the The characteristic diagramVectorizing using a fully-convoluted network layerPersonal (S) Is a convolution kernel of (2)Spliced (concate) toMap the characteristics ofDimension reduction and channel formation numberNew high level feature map of (a)I.e.
(symbol)Representing an imageA convolution operation with the convolution kernel,the method comprises the steps of carrying out a first treatment on the surface of the Map the characteristic mapAnd position coding parametersAdding the position-coding parametersInitializing toThen, the sum of the two is calculated to obtain the tensorStraightening the planar shape of (a) to change the shape (denoted as reshape)And is converted intoTensors of (2) as input to a tensor encoder, noted asI.e.。
Constructing and initializing an encoder for tensor encoding of key features of the vehicle, and a decoder for performing self-attention interactive decoding; initializing for encoder and decoderParameter tensorsParameters of encoderAnd decoderTogether form the parameter tensor. And initializing a recognition parameter tensor for recognizing key features of the vehicle in the decoder。
And inputting the feature map into the tensor encoder to obtain tensor encoding of the key features of the vehicle, and then executing self-attention interactive decoding. Specifically, the vectorized object isInput encoder, get sumIdentically shaped vehicle key feature codesI.e.
,
, Is the parametric tensor of the encoder. Will beAnd identifying parameter tensorsDecoding by cross-attention mechanismA predictor for obtaining predictive decoding characteristics for N vehicle key characteristicsI.e.
,
Predictive decoding features that are N key features of a vehicleThe tensor that is composed of the two,is the parametric tensor of the decoder.
Constructing and initializing a vehicle key feature recognition model and a vehicle key feature position frame positioning model for carrying out parallel recognition on a plurality of vehicle key features; initializing a model for each object type identificationParameter tensor in (2)And initializing a target location frame recognition modelParameter tensor in (2)。
And inputting the predictive decoding features into the vehicle key feature recognition model and the vehicle key feature position frame positioning model, and outputting each recognition model and the vehicle key feature position frame positioning model in parallel. Characterization of N vehicle key featuresRespectively through the key feature recognition model of the vehicleAnd vehicle key feature position frame positioning modelObtaining a vector set containing N vehicle key features and vehicle key feature position frames thereofI.e.. And aiming at the vehicle image sample of the training set, N vector sets output by the vehicle key feature recognition model and the vehicle key feature position frame positioning model calculate a loss function according to the deviation between the N vector sets and the vehicle feature description information of the training set, and obtain the estimation of model parameters.
Vector set output by optimizing vehicle image samples for training setsVehicle feature description information with training setThe loss function between the two, obtain parameter estimation to each model, expressed as:，,,,,, . Thus, the training process using the training set is completed.
After training, aiming at the key frames after video segmentation, tensor coding and self-attention interactive decoding of vehicle key features are carried out by adopting the neural network feature extraction model, so that parallel identification and parallel output of a plurality of vehicle key features and vehicle key feature position frames are realized, and the output is expressed as:
is comprised ofThe key features of the individual vehicles are that,then it is the firstPosition frame location vectors for key features of the individual vehicles.
In some embodiments, the communication module comprises;
the wireless communication sub-module is used for transmitting traffic data to the cloud server according to 4G or 5G or VPN communication;
the Beidou short message sub-module is used for transmitting traffic data to the cloud server through a short message function under the condition of losing wireless signals;
specifically, the communication module of the traffic situation investigation equipment is additionally provided with the module with the Beidou short message communication function on the basis of the traditional VPN communication, 4G/5G wireless communication or wired private network communication based on operators, so that the data can not influence the acquisition and transmission in the environments without signal areas/with interference factors (severe weather areas, unstable network signals and even no network signals), the length of single transmission message with regional short message communication can reach 1000 Chinese characters, the transmission of traffic situation investigation data can be met, and the guarantee is provided for the whole scene highway traffic situation investigation work. And the Beidou satellite short messages are used for information exchange and communication, so that the real-time, continuous, accurate, rapid, stable and reliable monitoring and data transmission of data can be realized.
In some embodiments, the analysis scheduling module comprises:
the video clustering sub-module is used for classifying the video sequences according to a clustering algorithm and the characteristic data;
the event identification sub-module is used for identifying special events for each type of video sequence;
and the alarm sub-module is used for sending an alarm to the command center according to the special event.
Specifically, the main function of the analysis scheduling module of the scheme is to identify and automatically alarm the events in the video. The system mainly comprises three sub-modules, namely a video clustering sub-module, an event identification sub-module and an alarm sub-module. The video clustering sub-module is used for classifying video sequences according to a clustering algorithm and feature data, and the video clustering sub-module of the scheme is based on a key frame obtained by a video segmentation unit and a preset clustering center, firstly calculates the distance between the key frame and each clustering center by using the key frame Kmeans clustering algorithm, and divides the key frame into one clustering center closest to the key frame Kmeans clustering algorithm. After the key frames are clustered, vehicle motion related parameters in different types of videos, such as the vehicle head distance and the vehicle head time distance, can be calculated, particularly, the vehicle speed of the video vehicle can be calculated according to continuous video frames, road information and monitoring position information through a centroid motion trail method, so that the vehicle head distance and the vehicle head time distance can be calculated according to the vehicle speed, and the calculation method has the characteristics of less calculation and high calculation speed; the event recognition submodule carries out event recognition by utilizing a pre-trained event recognition model, the model carries out training by utilizing videos with event type labels as training sets in advance, so that the mapping relation between video frames and special events is obtained, the special events generally comprise retrograde, traffic accidents, road danger, water accumulation on road surfaces, raining, snowing, dense fog and the like, once the special events are recognized, the event recognition submodule can send an alarm to a command center through the alarm submodule, and similarly, if a special event occurrence place is in a signal-free state, the Beidou short message submodule can be utilized to send alarm information to the command center through short message communication.
The second aspect of the present application also provides a traffic situation investigation method based on the Beidou short message communication technology, and in combination with fig. 1, namely a method flowchart of the present application, the method comprises the following steps:
s1: monitoring a road and acquiring traffic data through a traffic investigation module;
s2: the time service and the positioning information are carried out on the investigation equipment through the Beidou module;
s3: the communication module is used for sending the acquired traffic data, time service and positioning information to the cloud server;
s4: and acquiring traffic data from the cloud server through the analysis scheduling module, and analyzing the traffic data.
In some embodiments, in conjunction with fig. 2, which is a sub-flowchart of step S1 of the present application, the step S1 includes:
s11: the road traffic condition is monitored through the traffic monitoring sub-module;
s12: and extracting road and vehicle data through the data extraction sub-module.
In some embodiments, in conjunction with fig. 3, which is a sub-flowchart of step S12 of the present application, the step S12 includes:
s121: the method comprises the steps that key frame segmentation is carried out on positions of vehicles and lanes in a monitoring picture and image information through a video segmentation unit according to a video segmentation algorithm;
s122: and extracting vehicle characteristic data from the video key frames by a characteristic extraction unit according to the characteristic extraction model.
More specifically, in step S122, the specific process of extracting the vehicle feature information after training using the existing data in the traffic database as the training set based on the neural network feature extraction model is as follows.
Firstly, constructing a training set formed by matching vehicle image samples and vehicle characteristic description information one by utilizing the existing data in the traffic database; the training set formed by matching the vehicle image sample and the vehicle characteristic description information one by one is expressed as follows:
wherein the method comprises the steps ofIs the total amount of the sample,is the firstA sample of the vehicle image is taken,is the image height direction pixel value of the vehicle image sample,is an image width direction pixel value of the vehicle image sample;is the first of training setMaximum of corresponding vehicle image samplesVehicle characteristic description information formed by the vehicle key characteristics and the vehicle key characteristic position frame positioning vectors;is an integer set in advance; wherein the method comprises the steps of ，Is the firstIn the individual vehicle image samplesThe key features of the individual vehicles are that,is the firstIn the individual vehicle image samplesAnd 4-dimensional vectors consisting of coordinates of a central point of a vehicle key feature position frame corresponding to each vehicle key feature, a frame height value and a frame width value. The key features of the vehicle include license plates, car lamps, driver positions, front passenger positions and other key features of the vehicle according to the requirements of traffic investigation.
Initializing each layer of neural network configuration parameters of the neural network feature extraction model; the neural network feature extraction model comprises a ResNet model convolution layer structure and a full convolution network layer. ResNet model convolutional layer structure is represented asWhereinIs the parameter tensor of the parameters of all layers of the initialized res net model convolution layer,an input vehicle image for a ResNet model convolution layer; the full convolution network layer is represented asWherein the parameter tensorIs an initialized full convolution networkParameter tensors consisting of parameters of all convolution kernels of the complex layer, and ，a feature map representing the low resolution provided by the ResNet model convolutional layer structure; as an alternative value, wherein,。
Further, the vehicle image sample is input to a neural network feature extraction model for training, and vectorized vehicle feature information forming the vehicle image sample is extracted from the neural network feature extraction model. Specifically, a vehicle image is sampledAs the input vehicle imageInputting the ResNet model convolution layer structure of the initialized neural network feature extraction model to obtain a low-resolution feature mapThe method comprises the steps of carrying out a first treatment on the surface of the The characteristic diagramVectorizing using a fully-convoluted network layerPersonal (S) Is a convolution kernel of (2)Spliced (concate) toMap the characteristics ofDimension reduction and channel formation numberNew high level feature map of (a)I.e.
(symbol)Representing an imageA convolution operation with the convolution kernel,the method comprises the steps of carrying out a first treatment on the surface of the Map the characteristic mapAnd position coding parametersAdding the position-coding parametersInitializing toThen, the sum of the two is calculated to obtain the tensorStraightening the planar shape of (a) to change the shape (denoted as reshape)And is converted intoTensors of (2) as input to a tensor encoder, noted asI.e.。
Constructing and initializing an encoder for tensor encoding of key features of the vehicle, and a decoder for performing self-attention interactive decoding; initializing for encoder and decoderParameter tensorsParameters of encoderAnd decoderTogether form the parameter tensor. And initializing a recognition parameter tensor for recognizing key features of the vehicle in the decoder。
And inputting the feature map into the tensor encoder to obtain tensor encoding of the key features of the vehicle, and then executing self-attention interactive decoding. Specifically, the vectorized object isInput encoder, get sumIdentically shaped vehicle key feature codesI.e.
,
, Is the parametric tensor of the encoder. Will beAnd identifying parameter tensorsPredictive decoding features for N vehicle key features via a cross-attention mechanism decoderI.e.
,
Predictive decoding features that are N key features of a vehicleThe tensor that is composed of the two,is the parametric tensor of the decoder.
Constructing and initializing a vehicle key feature recognition model and a vehicle key feature position frame positioning model for carrying out parallel recognition on a plurality of vehicle key features; initializing a model for each object type identificationParameter tensor in (2)And initializing a target location frame recognition modelParameter tensor in (2)。
And inputting the predictive decoding features into the vehicle key feature recognition model and the vehicle key feature position frame positioning model, and outputting each recognition model and the vehicle key feature position frame positioning model in parallel. Characterization of N vehicle key featuresRespectively through the key feature recognition model of the vehicleAnd vehicle key feature position frame positioning modelObtaining a vector set containing N vehicle key features and vehicle key feature position frames thereofI.e.. And aiming at the vehicle image sample of the training set, N vector sets output by the vehicle key feature recognition model and the vehicle key feature position frame positioning model calculate a loss function according to the deviation between the N vector sets and the vehicle feature description information of the training set, and obtain the estimation of model parameters.
Vector set output by optimizing vehicle image samples for training setsVehicle feature description information with training setLoss between themA loss function, obtaining parameter estimates for each model, expressed as:，,,,,, . Thus, the training process using the training set is completed.
After training, aiming at the key frames after video segmentation, tensor coding and self-attention interactive decoding of vehicle key features are carried out by adopting the neural network feature extraction model, so that parallel identification and parallel output of a plurality of vehicle key features and vehicle key feature position frames are realized, and the output is expressed as:
is comprised ofThe key features of the individual vehicles are that,then it is the firstPosition frame location vectors for key features of the individual vehicles.
In some embodiments, in conjunction with fig. 4, which is a sub-flowchart of step S3 of the present application, the step S3 includes:
s31: transmitting traffic data to a cloud server through a wireless communication sub-module and through 4G or 5G or VPN communication;
s32: and transmitting traffic data to the cloud server through the Beidou short message sub-module and the short message function under the condition of losing the wireless signal.
In some embodiments, in conjunction with fig. 5, which is a sub-flowchart of step S4 of the present application, the step S4 includes:
s41: classifying the video sequences through a video clustering sub-module according to a clustering algorithm and feature data;
s42: the event identification sub-module is used for identifying special events for each type of video sequence;
s43: and sending an alarm to the command center through the alarm sub-module according to the special event.
Those skilled in the art will appreciate that while some embodiments described herein include some features but not others included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the application and form different embodiments.
Those skilled in the art will appreciate that the descriptions of the various embodiments are each focused on, and that portions of one embodiment that are not described in detail may be referred to as related descriptions of other embodiments.
Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and alterations without departing from the spirit and scope of the present application, and such modifications and alterations fall within the scope of the appended claims, which are to be construed as merely illustrative of the present application, but the scope of the application is not limited thereto, and various equivalent modifications and substitutions will be readily apparent to those skilled in the art within the scope of the present application, and are intended to be included within the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
The present application is not limited to the above embodiments, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the present application, and these modifications and substitutions are intended to be included in the scope of the present application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (10)
1. Traffic situation investigation equipment based on big dipper short message communication technique, its characterized in that includes:
the traffic investigation module is used for monitoring the road and acquiring traffic data;
the Beidou module is used for carrying out time service and positioning information on the investigation equipment;
the communication module is used for sending the acquired traffic data and time service and positioning information to the cloud server;
and the analysis scheduling module is used for acquiring traffic data from the cloud server and analyzing the traffic data.
2. The traffic situation investigation apparatus based on the beidou short message communication technology according to claim 1, wherein the traffic investigation module comprises:
the traffic monitoring sub-module is used for monitoring road traffic conditions;
and the data extraction sub-module is used for extracting road and vehicle data.
3. The traffic situation investigation apparatus based on the beidou short message communication technology according to claim 2, wherein the data extraction submodule comprises:
the video segmentation unit is used for carrying out key frame segmentation on the positions of the vehicles and the lanes in the monitoring picture and the image information according to a video segmentation algorithm;
and the feature extraction unit is used for extracting vehicle feature data from the video key frames according to the feature extraction model.
4. The traffic situation investigation equipment based on the Beidou short message communication technology according to claim 1, wherein the communication module comprises;
the wireless communication sub-module is used for transmitting traffic data to the cloud server according to 4G or 5G or VPN communication;
and the Beidou short message sub-module is used for transmitting traffic data to the cloud server through a short message function under the condition of losing wireless signals.
5. The traffic situation investigation equipment based on the Beidou short message communication technology according to claim 1, wherein the analysis scheduling module comprises:
the video clustering sub-module is used for classifying the video sequences according to a clustering algorithm and the characteristic data;
the event identification sub-module is used for identifying special events for each type of video sequence;
and the alarm sub-module is used for sending an alarm to the command center according to the special event.
6. The traffic condition investigation method based on the Beidou short message communication technology is characterized by comprising the following steps of:
s1: monitoring a road and acquiring traffic data through a traffic investigation module;
s2: the time service and the positioning information are carried out on the investigation equipment through the Beidou module;
s3: the communication module is used for sending the acquired traffic data, time service and positioning information to the cloud server;
s4: and acquiring traffic data from the cloud server through the analysis scheduling module, and analyzing the traffic data.
7. The traffic situation investigation method based on the Beidou short message communication technology according to claim 6, wherein the step S1 comprises:
s11: the road traffic condition is monitored through the traffic monitoring sub-module;
s12: and extracting road and vehicle data through the data extraction sub-module.
8. The traffic situation investigation method based on the beidou short message communication technology according to claim 7, wherein the step S12 comprises:
s121: the method comprises the steps that key frame segmentation is carried out on positions of vehicles and lanes in a monitoring picture and image information through a video segmentation unit according to a video segmentation algorithm;
s122: and extracting vehicle characteristic data from the video key frames by a characteristic extraction unit according to the characteristic extraction model.
9. The traffic situation investigation method based on the Beidou short message communication technology according to claim 6, wherein the step S3 comprises:
s31: transmitting traffic data to a cloud server through a wireless communication sub-module and through 4G or 5G or VPN communication;
s32: and transmitting traffic data to the cloud server through the Beidou short message sub-module and the short message function under the condition of losing the wireless signal.
10. The traffic situation investigation method based on the Beidou short message communication technology according to claim 6, wherein the step S4 comprises:
s41: classifying the video sequences through a video clustering sub-module according to a clustering algorithm and feature data;
s42: the event identification sub-module is used for identifying special events for each type of video sequence;
s43: and sending an alarm to the command center through the alarm sub-module according to the special event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311180551.6A CN116913096B (en) | 2023-09-13 | 2023-09-13 | Traffic situation investigation equipment and method based on Beidou short message communication technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311180551.6A CN116913096B (en) | 2023-09-13 | 2023-09-13 | Traffic situation investigation equipment and method based on Beidou short message communication technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116913096A true CN116913096A (en) | 2023-10-20 |
| CN116913096B CN116913096B (en) | 2023-12-15 |
Family
ID=88353471
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311180551.6A Active CN116913096B (en) | 2023-09-13 | 2023-09-13 | Traffic situation investigation equipment and method based on Beidou short message communication technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116913096B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006178853A (en) * | 2004-12-24 | 2006-07-06 | Aisin Aw Co Ltd | Traffic situation presenting system and traffic situation presenting method |
| CN102737511A (en) * | 2012-07-04 | 2012-10-17 | 武汉大学 | Intelligent road side system |
| CN102779406A (en) * | 2011-05-12 | 2012-11-14 | 北京维信通科技股份有限公司 | Cloud computing intelligent transportation scheduling platform based on Beidou time service technology |
| CN104504903A (en) * | 2014-12-31 | 2015-04-08 | 北京赛维安讯科技发展有限公司 | Traffic incident acquiring device and method and traffic incident monitoring system and method |
| CN105043441A (en) * | 2015-06-29 | 2015-11-11 | 北斗航天卫星应用科技集团有限公司 | Environmental monitoring system based on Beidou time service and positioning technology |
| CN205810117U (en) * | 2016-07-18 | 2016-12-14 | 保定维特瑞交通设施工程有限责任公司 | A kind of intelligent traffic control system based on Big Dipper short message |
| CN106297369A (en) * | 2016-11-04 | 2017-01-04 | 重庆两江新区临济信息技术有限公司 | A kind of vehicle monitoring system based on Beidou satellite navigation and monitoring method |
| CN205943076U (en) * | 2016-05-27 | 2017-02-08 | 武汉万集信息技术有限公司 | Detection car -mounted device blocks up |
| CN106444547A (en) * | 2016-11-28 | 2017-02-22 | 东莞职业技术学院 | Vehicular monitoring terminal and system based on GPRS communication and Beidou positioning |
| CN109523811A (en) * | 2019-01-07 | 2019-03-26 | 周书恒 | The monitoring of group's mist and early warning system based on Beidou satellite navigation system |
| CN109841062A (en) * | 2019-04-01 | 2019-06-04 | 长沙而道新能源科技有限公司 | Traffic infrastructure real-time system for monitoring and pre-warning based on Beidou big data cloud platform |
| CN109901199A (en) * | 2019-03-16 | 2019-06-18 | 毕延彬 | A kind of emergency communication rescue system based on Beidou navigation technology |
| CN112562382A (en) * | 2020-12-08 | 2021-03-26 | 特路（北京）科技有限公司 | Intelligent expressway vehicle positioning unit and method |
| CN114937367A (en) * | 2022-05-20 | 2022-08-23 | 苏州天准科技股份有限公司 | Intelligent camera system for cooperative monitoring of vehicle and road and control method |
-
2023
- 2023-09-13 CN CN202311180551.6A patent/CN116913096B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006178853A (en) * | 2004-12-24 | 2006-07-06 | Aisin Aw Co Ltd | Traffic situation presenting system and traffic situation presenting method |
| CN102779406A (en) * | 2011-05-12 | 2012-11-14 | 北京维信通科技股份有限公司 | Cloud computing intelligent transportation scheduling platform based on Beidou time service technology |
| CN102737511A (en) * | 2012-07-04 | 2012-10-17 | 武汉大学 | Intelligent road side system |
| CN104504903A (en) * | 2014-12-31 | 2015-04-08 | 北京赛维安讯科技发展有限公司 | Traffic incident acquiring device and method and traffic incident monitoring system and method |
| CN105043441A (en) * | 2015-06-29 | 2015-11-11 | 北斗航天卫星应用科技集团有限公司 | Environmental monitoring system based on Beidou time service and positioning technology |
| CN205943076U (en) * | 2016-05-27 | 2017-02-08 | 武汉万集信息技术有限公司 | Detection car -mounted device blocks up |
| CN205810117U (en) * | 2016-07-18 | 2016-12-14 | 保定维特瑞交通设施工程有限责任公司 | A kind of intelligent traffic control system based on Big Dipper short message |
| CN106297369A (en) * | 2016-11-04 | 2017-01-04 | 重庆两江新区临济信息技术有限公司 | A kind of vehicle monitoring system based on Beidou satellite navigation and monitoring method |
| CN106444547A (en) * | 2016-11-28 | 2017-02-22 | 东莞职业技术学院 | Vehicular monitoring terminal and system based on GPRS communication and Beidou positioning |
| CN109523811A (en) * | 2019-01-07 | 2019-03-26 | 周书恒 | The monitoring of group's mist and early warning system based on Beidou satellite navigation system |
| CN109901199A (en) * | 2019-03-16 | 2019-06-18 | 毕延彬 | A kind of emergency communication rescue system based on Beidou navigation technology |
| CN109841062A (en) * | 2019-04-01 | 2019-06-04 | 长沙而道新能源科技有限公司 | Traffic infrastructure real-time system for monitoring and pre-warning based on Beidou big data cloud platform |
| CN112562382A (en) * | 2020-12-08 | 2021-03-26 | 特路（北京）科技有限公司 | Intelligent expressway vehicle positioning unit and method |
| CN114937367A (en) * | 2022-05-20 | 2022-08-23 | 苏州天准科技股份有限公司 | Intelligent camera system for cooperative monitoring of vehicle and road and control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116913096B (en) | 2023-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112541440B (en) | Subway people stream network fusion method and people stream prediction method based on video pedestrian recognition | |
| EP4152204A1 (en) | Lane line detection method, and related apparatus | |
| CN113506317B (en) | Multi-target tracking method based on Mask R-CNN and apparent feature fusion | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN111008574A (en) | Key person track analysis method based on body shape recognition technology | |
| CN116824859B (en) | Intelligent traffic big data analysis system based on Internet of things | |
| Xie et al. | Deep learning-based computer vision for surveillance in its: Evaluation of state-of-the-art methods | |
| CN114372503A (en) | Cluster vehicle motion trail prediction method | |
| Masmoudi et al. | Trajectory analysis for parking lot vacancy detection system | |
| Minnikhanov et al. | Detection of traffic anomalies for a safety system of smart city | |
| CN113095277B (en) | Unmanned aerial vehicle aerial photography vehicle detection method based on target space distribution characteristics | |
| Lu et al. | A segmentation-based multitask learning approach for isolating switch state recognition in high-speed railway traction substation | |
| Mukhopadhyay et al. | A hybrid lane detection model for wild road conditions | |
| CN114048536A (en) | Road structure prediction and target detection method based on multitask neural network | |
| CN116913096B (en) | Traffic situation investigation equipment and method based on Beidou short message communication technology | |
| CN113094422B (en) | Urban road traffic flow map generation method, system and equipment | |
| CN112580633B (en) | Public transport passenger flow statistics device and method based on deep learning | |
| CN111242044B (en) | Night unmanned vehicle scene prediction method based on ConvLSTM dual-channel coding network | |
| CN113850111A (en) | Road condition identification method and system based on semantic segmentation and neural network technology | |
| Sakaino | Deepscene, deepvis, deepdist, and deepreject: Image-based visibility estimation system for uav | |
| Xu et al. | Vehicle trajectory prediction considering multi-feature independent encoding | |
| CN114863685B (en) | Traffic participant trajectory prediction method and system based on risk acceptance degree | |
| CN114863370B (en) | Complex scene high altitude parabolic identification method and system | |
| Jiang et al. | Collision Risk Prediction for Vehicles with Sensor Data Fusion through a Machine Learning Pipeline | |
| CN113538891A (en) | Intelligent vehicle counting system |
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 |