CN117315943A - Monitoring analysis and early warning method and system for overrun transportation violations - Google Patents
Monitoring analysis and early warning method and system for overrun transportation violations Download PDFInfo
- Publication number
- CN117315943A CN117315943A CN202311602581.1A CN202311602581A CN117315943A CN 117315943 A CN117315943 A CN 117315943A CN 202311602581 A CN202311602581 A CN 202311602581A CN 117315943 A CN117315943 A CN 117315943A
- Authority
- CN
- China
- Prior art keywords
- data
- traffic
- transportation
- load
- overrun
- 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
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000012544 monitoring process Methods 0.000 title claims abstract description 34
- 238000004458 analytical method Methods 0.000 title claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 171
- 230000002159 abnormal effect Effects 0.000 claims abstract description 128
- 230000005856 abnormality Effects 0.000 claims abstract description 125
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000012216 screening Methods 0.000 claims abstract description 37
- 238000004140 cleaning Methods 0.000 claims description 56
- 238000004364 calculation method Methods 0.000 claims description 37
- 230000008859 change Effects 0.000 claims description 28
- 238000003708 edge detection Methods 0.000 claims description 18
- 230000001133 acceleration Effects 0.000 claims description 11
- 238000012937 correction Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 description 17
- 230000000875 corresponding effect Effects 0.000 description 16
- 238000004422 calculation algorithm Methods 0.000 description 15
- 238000010801 machine learning Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 12
- 238000007619 statistical method Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000006399 behavior Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 206010039203 Road traffic accident Diseases 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 4
- 230000002596 correlated effect Effects 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 2
- 238000012098 association analyses Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000005206 flow analysis Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000013468 resource allocation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 230000003252 repetitive effect Effects 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
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of transportation data processing, in particular to an overrun transportation illegal monitoring analysis and early warning method and system. The method comprises the following steps: acquiring traffic and transportation basic data, and screening the traffic and transportation basic data for data validity, so as to acquire traffic and transportation data; carrying out load abnormality detection on the traffic data so as to obtain load abnormality detection data; the method comprises the steps of obtaining historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data, obtaining first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data, so that second overrun load abnormality data are obtained. The invention can accurately identify the abnormal condition of overload, reduce the false alarm rate and improve the early warning accuracy.
Description
Technical Field
The invention relates to the technical field of transportation data processing, in particular to an overrun transportation illegal monitoring analysis and early warning method and system.
Background
The monitoring, analyzing, early warning and disposing method for the illegal behavior of overload and overrun of vehicles in road transportation is a method system for monitoring, analyzing, early warning and disposing by applying modern technological means. Some monitoring devices have limited performance and cannot accurately monitor all types of violations, especially for some potential or bursty transportation violations.
Disclosure of Invention
The invention provides an overrun transportation illegal monitoring analysis and early warning method and system for solving the technical problems, and aims to solve at least one of the technical problems.
The application provides an overrun transportation illegal monitoring analysis and early warning method, which comprises the following steps:
step S1: acquiring traffic and transportation basic data, and screening the traffic and transportation basic data for data validity, so as to acquire traffic and transportation data;
step S2: carrying out load abnormality detection on the traffic data so as to obtain load abnormality detection data;
step S3: acquiring historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data, thus acquiring first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data, thus acquiring second overrun load abnormality data;
Step S4: and carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data, thereby acquiring overrun transportation illegal monitoring early warning data.
According to the invention, through reference of historical data, the load abnormal data is correlated twice, so that the first and second overrun load abnormal conditions can be accurately identified, the false alarm rate is reduced, and the early warning accuracy is improved. Based on the load abnormality detection data, the load condition of the vehicle can be monitored in real time, and the illegal behavior of overrun transportation can be found in time, so that corresponding measures can be taken. The first overrun and the second overrun load abnormal data are used for generating early warning information, so that traffic management departments or law enforcement departments can be informed in advance, overrun transportation illegal behaviors can be handled in time, and traffic accident risks are reduced. Through data screening and an accurate early warning mechanism, a user can take measures in a targeted manner, manpower and material resources are saved, and the working efficiency is improved. The method and the device follow the preset road load limiting standard data, can ensure that the overrun transportation activities accord with regulations, and ensure the safety and smoothness of road traffic.
Preferably, step S1 is specifically:
step S11: acquiring basic data of traffic transportation;
Step S12: carrying out traffic data cleaning on the traffic basic data so as to obtain traffic cleaning data;
step S13: performing time sequence traffic scene recognition and classification according to traffic cleaning data so as to obtain time sequence traffic scene data, wherein the time sequence traffic scene data comprises specific time point data, vehicle speed data, vehicle density data and vehicle road condition data, the vehicle density data comprises high vehicle density data and low vehicle density data, and the vehicle road condition data comprises congestion road condition data and smooth road condition data;
step S14: and screening the traffic abnormal value of the traffic cleaning data according to the time sequence traffic scene data, thereby obtaining the traffic data.
According to the invention, invalid or abnormal data can be removed by cleaning the traffic data in the step S12, so that the accuracy and reliability of the data are improved. By the time sequence traffic scene recognition classification in the step S13, the traffic data can be more accurately classified into specific traffic scenes, and subsequent analysis and processing are facilitated. Abnormal value screening is carried out on traffic data through time sequence traffic scene data in the step S14, so that abnormal data can be more accurately identified and removed under a specific scene, and the accuracy of abnormal detection is improved. By screening the abnormal data, abnormal conditions in transportation can be found in time, early warning is facilitated, corresponding measures are taken, and the risk of traffic accidents is reduced. Invalid data can be removed through data cleaning and outlier screening, so that the acquisition and processing cost of resources is saved, and the data utilization efficiency is improved. Through data cleaning and outlier screening, traffic and transportation activities can be ensured to accord with regulations, and safety and smoothness of road traffic are ensured.
Preferably, the cleaning of the transportation data is performed by a transportation data cleaning calculation formula, wherein the transportation data cleaning calculation formula specifically comprises:
;
for transportation cleaning data->Order item data for basic data of transportation, < +.>For the quantity data of the traffic base data, +.>Is->Personal transportation base data, < >>Is->Traffic quality index data corresponding to the individual traffic base data,/-for>For cleaning the lower limit of the data range, +.>For cleaning the upper limit of the data range, +.>For detecting abnormal traffic data->For the data item to be detected in the transportation base data, < > for the data item to be detected in the transportation base data>For data correction factor, ++>For the vehicle driving direction angle data, +.>For traffic time change data, < >>Is traffic time data.
The invention constructs a traffic data cleaning calculation formula which uses the following steps ofThe transportation data is cleaned and processed, abnormal values and correction data can be eliminated, useful information can be extracted, and accordingly more reliable and accurate transportation data can be obtained, and support is provided for subsequent analysis and decision making. By for each underlying data itemIs the quality index corresponding to the same The logarithmic conversion can reflect the quality characteristics of the traffic data more accurately, and the logarithmic operation can convert the exponential data into the linear data, so that the difference between the data is reduced, and the abnormal data is easier to identify and process. Function->Represents an anomaly detector by means of the data item to be detected +.>Performing integral operation and correction factor->By combining the method, the existing abnormal data can be effectively identified and filtered, the interference of abnormal values in the data processing process is avoided, and the reliability of the data is improved. Correction factor->And upper and lower limit parameters (>,/>) The function in the formula is to make further corrections and control of the data. The data can be smoothed and adjusted through root number operation and integration, so that the data is ensured to be in a reasonable range, the fluctuation of the data is reduced, and the stability of the data is improved. For time sequence characteristics in traffic data, including vehicle driving direction angle +.>And traffic time variation->Sine and time derivative in the formula +.>The introduction of the terms enables the model to take the influence of time change into consideration when processing the data so as to realize more accurate reflection of the change trend of the traffic data along with the time, thereby being beneficial to long-term trend analysis and prediction. The invention carries out the advanced treatment on the traffic data in theory and actual operation, and provides a reliable basis for the subsequent data analysis, prediction and decision.
Preferably, the transportation cleansing data includes single transportation cleansing data and group transportation cleansing data, and step S13 is specifically:
step S131: extracting time sequence transportation characteristic data according to the single transportation cleaning data so as to obtain time sequence transportation characteristic data, wherein the time sequence transportation characteristic data comprises time interval data, speed change rate data, load change rate data, parking time length data and acceleration change data;
step S132: carrying out distribution feature extraction on the group transportation cleaning data so as to obtain group transportation distribution feature data;
step S133: historical traffic experience mapping is carried out on the group traffic distribution characteristic data, so that traffic scene category data are obtained;
step S134: and carrying out time sequence traffic scene identification and classification on the time sequence traffic characteristic data according to the traffic scene category data so as to obtain time sequence traffic scene data, wherein the time sequence traffic scene identification and classification carries out identification and classification through a long and short time memory network.
According to the invention, through extracting the time sequence characteristic data of the single transportation cleaning data, rich time sequence characteristic information such as time interval, speed change rate, load change rate, parking duration and acceleration change can be obtained, so that the dynamic change of the transportation process can be more comprehensively described. And carrying out distributed feature extraction on the group transportation cleaning data to obtain the overall feature of group transportation so as to mine the statistical rule behind the data, thereby realizing the advanced processing of the data. The group traffic distribution characteristic data is mapped to the historical traffic experience, and the current traffic condition can be verified and evaluated through the historical data, so that the prediction accuracy is improved. The time sequence scene recognition classification is carried out through a long and short time memory network (LSTM), the finer classification is carried out on the traffic data, and the scene recognition precision is improved. By adopting deep learning technologies such as long-short-term memory networks and the like, compared with the traditional method, the time sequence data can be better processed, and the recognition and classification effects are improved. Meanwhile, the characteristics of single traffic and group traffic are considered, so that data of different layers are comprehensively utilized, and more comprehensive information is provided. By layering different types of data, the efficiency of data processing can be effectively improved, and the analysis and prediction process is more efficient.
Preferably, step S14 is specifically:
step S141: mapping traffic parameters according to the time sequence traffic scene data to generate specific time sequence threshold setting data;
step S142: abnormal value screening is carried out on the transportation cleaning data by utilizing specific time sequence threshold setting data, so that transportation screening data are obtained;
step S143: and reconstructing traffic data of the traffic screening data according to the time sequence traffic scene data, thereby obtaining the traffic data.
According to the invention, the traffic parameter mapping is carried out according to the time sequence traffic scene data, and the specific time sequence threshold setting data is generated, so that the judgment of the abnormal value is more targeted, and the abnormal condition can be identified more accurately. Abnormal value screening is carried out on the transportation cleaning data by utilizing specific time sequence threshold setting data, so that the abnormal data can be more effectively eliminated, and the accuracy and the credibility of the data are ensured. The screened data is reconstructed according to the time sequence traffic scene data, so that the original traffic data can be restored, and further analysis and application are facilitated. By reconstructing traffic data, more comprehensive traffic conditions including information of various traffic parameters can be obtained, and comprehensive understanding of traffic conditions is facilitated. Through a series of processing steps including outlier screening and data reconstruction, the quality of traffic data can be improved, and the accuracy and reliability of the data are ensured.
Preferably, the load abnormality detection data includes regular load abnormality detection data and vehicle load abnormality detection data, and step S2 specifically includes:
step S21: acquiring traffic position rule data according to traffic position data corresponding to the traffic data;
step S22: carrying out regular load abnormality detection on the transportation data according to the transportation position regular data so as to obtain regular load abnormality detection data;
step S23: detecting vehicle damage according to the vehicle image data in the traffic data, thereby obtaining vehicle damage detection data;
step S24: and detecting abnormal running of the vehicle according to the vehicle damage detection data, thereby obtaining abnormal load detection data of the vehicle.
According to the invention, the transportation position rule data is obtained according to the transportation data, and the load abnormality detection is carried out by combining the position information and the rule information, so that the detection is more comprehensive and accurate. The regular load abnormality detection is carried out by utilizing the traffic transportation position rule data, so that the existing load abnormality condition can be rapidly identified, and the detection efficiency is improved. And the vehicle damage detection is carried out by combining the vehicle image data, so that additional auxiliary information can be provided for the load abnormality detection, and the detection accuracy is improved. The abnormal running condition of the vehicle is detected through the vehicle damage detection data, so that the existing abnormal load condition can be further screened out, and the detection precision is improved. The load abnormality detection is divided into a plurality of steps, each step has a specific function, a layering detection strategy is formed between each step, and the detection accuracy and efficiency are improved. By combining the position data, the image data and the damage detection data, different types of information are comprehensively processed, so that the detection is more comprehensive and accurate. Through a multi-level detection strategy and comprehensive utilization of different information, the false alarm rate can be reduced, and the reliability of the detection result is improved.
Preferably, in step S22, the rule load abnormality detection is processed by a rule load abnormality detection calculation formula, where the rule load abnormality detection calculation formula specifically includes:
;
for regular load abnormality detection data, < > for>For the quantity data of the traffic data, +.>Order item data of transportation data, < ->For the first->The value of the individual load data points,/">For the +.>Numerical value of each angle data point, +.>For the distance data of the vehicle>For traffic data->Is->Specific data item of individual traffic data, +.>Is->A particular data item of traffic location rule data.
The invention constructs a rule load abnormality detection calculation formula which is used for evaluating whether the load condition in the transportation process accords with rules or standards. Wherein the method comprises the steps ofPart will transport quantity->The increase of (2) is scaled on a logarithmic scale, which can maintain a controllable range of computational complexity when processing large amounts of traffic data, while for larger +.>Value, it can slow down +.>So that anomaly detection can also operate effectively for large data sets. / >The product of the load values representing all traffic data points, by taking the square root of the product of these values, takes into account the aggregate nature of the load data, making the algorithm robust to load variations between different traffic.This term is used to process angle data, and plays an important role in regular load anomaly detection, involving directionInformation on the nature, for example, in the vehicle transportation, the angle is related to the cargo distribution, the stacking mode, and the like, and the information on this aspect can be effectively extracted by the ratio of sine and cosine. />This term includes the derivative of the distance L travelled by the vehicle, taking into account dynamic changes during the travel of the vehicle, such as road conditions, accelerations, etc. The method is used for detecting the abnormality in the transportation field, and the method comprises the processing of information in various aspects such as load, angle, distance and the like, so that the algorithm has adaptability to different conditions, and plays an important role in improving transportation safety, optimizing loading strategies and the like.
Preferably, in step S23, the vehicle damage detection is performed by a vehicle damage detection model, where the vehicle damage detection model is constructed specifically as follows:
Step S231: acquiring standard vehicle image data and standard vehicle damage tag data;
step S232: performing binarization processing on the standard vehicle image data so as to obtain standard vehicle binarization image data;
step S233: edge detection is carried out on the standard vehicle binarized image data, so that standard vehicle edge detection image data are obtained;
step S234: performing normal distribution calculation according to the standard vehicle edge detection image data, so as to obtain standard vehicle normal distribution data;
step S235: and marking the normal distribution data of the standard vehicle according to the damage label data of the standard vehicle, thereby obtaining a damage detection model of the vehicle.
According to the invention, the standard vehicle image data and the standard vehicle damage label data are acquired, so that a data set containing the standard image and corresponding damage information is established, and a basis is provided for model training. The image is preprocessed through binarization processing and edge detection on the standard vehicle image data, so that the damaged part in the image is more prominent, and subsequent feature extraction is facilitated. By acquiring standard vehicle edge detection image data through edge detection, edge information in a vehicle image can be extracted, which is one of important features of damage detection. And carrying out normal distribution calculation based on the standard vehicle edge detection image data, and carrying out statistical analysis on the damaged area to provide additional characteristic information. The normal distribution data is marked through the standard vehicle damage label data, so that the model can learn the characteristics of the damaged area, and the construction of the vehicle damage detection model is realized. Compared with the prior art, the method avoids complex calculation of machine learning, and simultaneously refines the depth data behind the data through edge calculation and normal distribution calculation so as to provide accurate and reliable coarse index data instead of higher-definition data specific to a certain index value layer, thereby improving the practicability.
Preferably, step S3 is specifically:
step S31: acquiring historical traffic and transportation data, wherein the historical traffic and transportation data comprises conventional historical traffic and transportation data and abnormal historical traffic and transportation data;
step S32: carrying out data association on the load abnormal detection data according to the conventional historical traffic and transportation data and the abnormal historical traffic and transportation data, so as to obtain conventional associated traffic and transportation data and abnormal associated traffic and transportation data;
step S33: carrying out traffic and transportation confidence calculation on the conventional associated traffic and transportation data and the abnormal associated traffic and transportation data so as to acquire first overrun load abnormal data;
step S34: and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data, thereby obtaining second overrun load abnormality data.
According to the method, the conventional historical traffic data and the abnormal historical traffic data are obtained, the historical data are taken into consideration, and the comprehensiveness and the accuracy of load abnormal detection are improved. The current load abnormal data and the historical data can be compared by correlating the load abnormal detection data according to the historical data, and the existing abnormal situation can be identified. And the traffic confidence coefficient calculation is carried out on the associated traffic data, so that the reliability of the abnormal condition of the load can be quantized, and the reliability of the detection result is improved. And carrying out second overrun load abnormality association on load abnormality detection data through preset road load limiting standard data, so that the accuracy of abnormality detection is further improved. The concepts of the first overrun and the second overrun are introduced, and the overrun condition is detected in a grading manner, so that the severity of the overrun condition can be accurately judged. By comprehensively utilizing historical data, confidence coefficient calculation and overrun standard data, the anomaly detection is more accurate and reliable.
Preferably, the present invention also provides an overrun transportation violation monitoring analysis and early warning system for executing the overrun transportation violation monitoring analysis and early warning method as described above, the overrun transportation violation monitoring analysis and early warning system comprising:
the traffic data validity screening module is used for acquiring traffic and transportation basic data and carrying out data validity screening on the traffic and transportation basic data so as to acquire traffic and transportation data;
the load abnormal detection module is used for carrying out load abnormal detection on the traffic data so as to obtain load abnormal detection data;
the overrun load abnormality association module is used for acquiring historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data so as to acquire first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data so as to acquire second overrun load abnormality data;
and the early warning generation module is used for carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data so as to acquire overrun transportation illegal monitoring early warning data.
The invention has the beneficial effects that: when the basic data of traffic and transportation are acquired, the data validity screening is performed, invalid or abnormal data are eliminated, the data quality of subsequent analysis is ensured, and the analysis accuracy is improved. By adopting an advanced load abnormality detection technology and analyzing and processing traffic data, the load abnormality situation is accurately identified, and the problems of missing report and false report in the traditional method are avoided. By introducing historical traffic data and carrying out correlation analysis of first overrun and second overrun on load abnormal detection data, multi-level and omnibearing detection of abnormal conditions is realized. This can improve the accuracy and reliability of the abnormal situation. And the preset road load limiting standard data is adopted, and the second overrun load abnormality association analysis is carried out on the load abnormality detection data, so that the abnormality detection precision is further improved. Based on the first overrun and the second overrun load abnormal data, the monitoring system can timely find and alarm the overrun transportation condition by generating the overrun transportation illegal monitoring early warning data, so that the illegal investigation efficiency is forcefully improved. Through the establishment of the comprehensive monitoring and early warning system, the occurrence frequency of the overrun transportation violations can be effectively reduced, and the road traffic safety level is improved. Traffic jam and accident risk caused by overrun transportation are reduced, road traffic resource utilization is optimized, and the method has positive pushing effect on urban traffic planning and resource allocation.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting implementations made with reference to the following drawings in which:
FIG. 1 is a flow chart illustrating steps of an embodiment of an overrun transport violation monitoring analysis and early warning method;
FIG. 2 shows a step flow diagram of step S1 of an embodiment;
FIG. 3 shows a step flow diagram of step S13 of an embodiment;
FIG. 4 shows a step flow diagram of step S14 of an embodiment;
fig. 5 shows a step flow diagram of step S2 of an embodiment.
Detailed Description
The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.
It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Referring to fig. 1 to 5, the present application provides an overrun transportation violation monitoring, analyzing and early warning method, which includes the following steps:
step S1: acquiring traffic and transportation basic data, and screening the traffic and transportation basic data for data validity, so as to acquire traffic and transportation data;
specifically, for example, GPS positioning technology: the longitude and latitude coordinates of the vehicle are obtained using a global satellite positioning system (e.g., GPS, GLONASS, etc.). Sensor technology: including speed sensors, load sensors, etc., for acquiring the speed and load conditions of the vehicle. Vehicle-mounted terminal equipment: and equipment installed on the vehicle for data acquisition, transmission and processing. Data communication technology: the method is used for transmitting the acquired data to a background server, and a common 4G/5G communication technology is adopted.
Specifically, for example, the data acquisition device is subjected to device detection to obtain device detection data, and the transportation base data is screened according to the device detection data to obtain transportation data, such as the transportation base data is adjusted according to the use loss of the sensor to obtain the transportation data.
Step S2: carrying out load abnormality detection on the traffic data so as to obtain load abnormality detection data;
specifically, for example, a load sensor: and the sensor is arranged on the vehicle and used for monitoring the weight of goods in real time. Data processing algorithm: and analyzing the load data by adopting a statistical method or a machine learning algorithm, and identifying abnormal conditions. Threshold setting: and setting a proper threshold value according to historical data or road load limiting standards to perform abnormality detection.
Step S3: acquiring historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data, thus acquiring first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data, thus acquiring second overrun load abnormality data;
Specifically, for example, the system correlates historical transportation data with load anomaly detection data. For example, the system correlates load anomaly detection data with corresponding transportation events in the historical transportation data by matching time and location information. After the association is completed, the system obtains first overrun load anomaly data including which transportation events occurred load anomalies and their specifics. Load limiting criteria for the road are set in advance, and are determined according to local regulations or road characteristics. By comparing the load anomaly data with preset road load limit standard data, the system determines which transport events have load anomalies that exceed the limits of the road, thereby obtaining second overrun load anomaly data, including which transport events violate the road regulations due to load overrun.
Step S4: and carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data, thereby acquiring overrun transportation illegal monitoring early warning data.
Specifically, for example, data analysis: and processing data by using historical data and real-time data and adopting methods such as statistical analysis, machine learning and the like. Prediction model: predictive models, such as logistic regression, neural networks, etc., are built for predicting overrun traffic conditions. Early warning system: and generating early warning information according to the prediction result, and notifying related departments or drivers in a mode of short messages, mails and the like.
Specifically, for example: date: 2022 9 month 15 days, time: 14:00, license plate number: country a12345, load: 15 tons, vehicle speed: 60 km/h, GPS coordinates: (39.9042 DEG N,116.4074 DEG E), detecting by a load sensor, and finding that the load of the vehicle is abnormal; retrieving the historical data and finding transportation records of similar dates and times; comparing the similar records, and finding that the load of a plurality of transportation records exceeds a preset threshold value; according to a preset road load limiting standard, confirming that the load limit of the road where the vehicle is located is 10 tons, and the 15 tons of load exceeds the limit at the moment; generating monitoring and early warning data of the overrun transportation illegal law: early warning type: overrun transportation and early warning content: when the truck with the license plate number of national A12345 is transported by 14:00 on the 15 th 9 th 2022, the load exceeds the road load limit standard.
According to the invention, through reference of historical data, the load abnormal data is correlated twice, so that the first and second overrun load abnormal conditions can be accurately identified, the false alarm rate is reduced, and the early warning accuracy is improved. Based on the load abnormality detection data, the load condition of the vehicle can be monitored in real time, and the illegal behavior of overrun transportation can be found in time, so that corresponding measures can be taken. The first overrun and the second overrun load abnormal data are used for generating early warning information, so that traffic management departments or law enforcement departments can be informed in advance, overrun transportation illegal behaviors can be handled in time, and traffic accident risks are reduced. Through data screening and an accurate early warning mechanism, a user can take measures in a targeted manner, manpower and material resources are saved, and the working efficiency is improved. The method and the device follow the preset road load limiting standard data, can ensure that the overrun transportation activities accord with regulations, and ensure the safety and smoothness of road traffic.
Preferably, step S1 is specifically:
step S11: acquiring basic data of traffic transportation;
specifically, for example, a GPS positioning device: is installed on a vehicle and is used for acquiring the position information of the vehicle in real time, including longitude and latitude coordinates. Sensor technology: and the speed sensor, the load sensor and the like are used for acquiring basic data such as the speed, the load and the like of the vehicle. Vehicle-mounted terminal equipment: for data acquisition and transmission, typically includes a processor, memory device, etc.
Step S12: carrying out traffic data cleaning on the traffic basic data so as to obtain traffic cleaning data;
specifically, for example, a data anomaly detection algorithm: abnormal data points are identified and processed using statistical methods or machine learning algorithms. A data filter: noise and interference in the data are removed using filter techniques. Data normalization processing: the data is converted into a unified format, so that subsequent processing and analysis are convenient.
Step S13: performing time sequence traffic scene recognition and classification according to traffic cleaning data so as to obtain time sequence traffic scene data, wherein the time sequence traffic scene data comprises specific time point data, vehicle speed data, vehicle density data and vehicle road condition data, the vehicle density data comprises high vehicle density data and low vehicle density data, and the vehicle road condition data comprises congestion road condition data and smooth road condition data;
Specifically, for example, scene classification: and classifying the time-series data into scenes according to the output result of the model, wherein the scenes comprise specific time point data, vehicle speed data, vehicle density data and vehicle road condition data. Time-series traffic scene data: the identified time-series traffic scenario data is stored, including time stamps, vehicle speed, vehicle density, road conditions, and the like. Subdivision data: the vehicle density data is subdivided into a high vehicle density and a low vehicle density, and the road condition data is subdivided into congestion and smoothness. Traffic flow analysis: and (3) carrying out traffic flow analysis by using the stored time sequence data, and identifying traffic scenes with high density and low density, and congestion and smooth road conditions.
Specifically, for example, a machine learning algorithm: such as a Recurrent Neural Network (RNN) or a Convolutional Neural Network (CNN), for classifying the time series data. Characteristic engineering: and extracting characteristics in the time sequence data, such as time intervals, speed change rates and the like, and taking the characteristics as the basis of classification. Training a model: and training a model by using the marked data set to realize time sequence scene classification.
Step S14: and screening the traffic abnormal value of the traffic cleaning data according to the time sequence traffic scene data, thereby obtaining the traffic data.
Specifically, for example, a machine learning algorithm: such as a Recurrent Neural Network (RNN) or a Convolutional Neural Network (CNN), for classifying the time series data. Characteristic engineering: and extracting characteristics in the time sequence data, such as time intervals, speed change rates and the like, and taking the characteristics as the basis of classification. Training a model: and training a model by using the marked data set to realize time sequence scene classification.
According to the invention, invalid or abnormal data can be removed by cleaning the traffic data in the step S12, so that the accuracy and reliability of the data are improved. By the time sequence traffic scene recognition classification in the step S13, the traffic data can be more accurately classified into specific traffic scenes, and subsequent analysis and processing are facilitated. Abnormal value screening is carried out on traffic data through time sequence traffic scene data in the step S14, so that abnormal data can be more accurately identified and removed under a specific scene, and the accuracy of abnormal detection is improved. By screening the abnormal data, abnormal conditions in transportation can be found in time, early warning is facilitated, corresponding measures are taken, and the risk of traffic accidents is reduced. Invalid data can be removed through data cleaning and outlier screening, so that the acquisition and processing cost of resources is saved, and the data utilization efficiency is improved. Through data cleaning and outlier screening, traffic and transportation activities can be ensured to accord with regulations, and safety and smoothness of road traffic are ensured.
Preferably, the cleaning of the transportation data is performed by a transportation data cleaning calculation formula, wherein the transportation data cleaning calculation formula specifically comprises:
;
for transportation cleaning data->Order item data for basic data of transportation, < +.>For the quantity data of the traffic base data, +.>Is->Personal transportation base data, < >>Is->Traffic quality index data corresponding to the individual traffic base data,/-for>For cleaning the lower limit of the data range, +.>For cleaning the upper limit of the data range, +.>For detecting abnormal traffic data->For the data item to be detected in the transportation base data, < > for the data item to be detected in the transportation base data>For data correction factor, ++>For the vehicle driving direction angle data, +.>For traffic time change data, < >>Is traffic time data.
The invention constructs a transportation data cleaning calculation formula, which can eliminate abnormal values, correct data and extract useful information by cleaning and processing transportation data, thereby obtaining more reliable and accurate transportation data and providing support for subsequent analysis and decision. By for each underlying data itemIs the quality index corresponding to the same The logarithmic conversion can reflect the quality characteristics of the traffic data more accurately, and the logarithmic operation can convert the exponential data into the linear data, so that the difference between the data is reduced, and the abnormal data is easier to identify and process. Function->Represents an anomaly detector by means of the data item to be detected +.>Performing integral operation and correction factor->By combining the method, the existing abnormal data can be effectively identified and filtered, the interference of abnormal values in the data processing process is avoided, and the reliability of the data is improved. Correction factor->And upper and lower limit parameters (>,/>) The function in the formula is to make further corrections and control of the data. The data can be smoothed and adjusted through root number operation and integration, so that the data is ensured to be in a reasonable range, the fluctuation of the data is reduced, and the stability of the data is improved. For time sequence characteristics in traffic data, including vehicle driving direction angle +.>And traffic time variation->Sine and time derivative in the formula +.>The introduction of the terms enables the model to take the influence of time change into consideration when processing the data so as to realize more accurate reflection of the change trend of the traffic data along with the time, thereby being beneficial to long-term trend analysis and prediction. The invention carries out the advanced treatment on the traffic data in theory and actual operation, and provides a reliable basis for the subsequent data analysis, prediction and decision.
Preferably, the transportation cleansing data includes single transportation cleansing data and group transportation cleansing data, and step S13 is specifically:
step S131: extracting time sequence transportation characteristic data according to the single transportation cleaning data so as to obtain time sequence transportation characteristic data, wherein the time sequence transportation characteristic data comprises time interval data, speed change rate data, load change rate data, parking time length data and acceleration change data;
specifically, for example, using GPS data time stamps, the difference between adjacent time stamps is calculated. And calculating the change rate of the speed at adjacent time by using the speed information in the GPS data. And calculating the change rate of the load at adjacent moments by combining the load sensor data. The speed data is analyzed to identify a time period when the speed is zero. Based on the velocity data, accelerations at adjacent times are calculated.
Step S132: carrying out distribution feature extraction on the group transportation cleaning data so as to obtain group transportation distribution feature data;
specifically, the distribution characteristics of the population are described, for example, using statistical information of the population data, such as average value, variance, and the like.
Step S133: historical traffic experience mapping is carried out on the group traffic distribution characteristic data, so that traffic scene category data are obtained;
Specifically, for example, the historical data is used to map the population characteristics, and a historical experience model is established. If a large amount of historical transportation data is acquired, the data records characteristic information under different transportation scenes and the category of the transportation scenes. By utilizing methods such as machine learning or statistical analysis, a historical experience model is established in advance, the model can predict the category of the traffic scene to which the model belongs according to the characteristic data, the group traffic distribution characteristic data is input into the historical experience model, and the model can map the characteristic data to the corresponding traffic scene category according to the historical experience. After mapping, the system obtains a set of traffic scene category data, and the data indicate the traffic scene category to which each feature data belongs.
Step S134: and carrying out time sequence traffic scene identification and classification on the time sequence traffic characteristic data according to the traffic scene category data so as to obtain time sequence traffic scene data, wherein the time sequence traffic scene identification and classification carries out identification and classification through a long and short time memory network.
Specifically, the time series characteristic data is trained and classified, for example, by using an LSTM network, and different traffic scenes are identified. For training and testing using a data set containing 1000 samples, each sample comprising 10 timing features, there are 5 different traffic scenario categories. The system builds a neural network model comprising two LSTM layers and a fully connected layer, wherein the number of nodes of the hidden layer is 32. During training, the system uses a cross entropy loss function and a random gradient descent (SGD) optimizer to minimize the loss function. After multiple rounds of training, the model achieves 90% accuracy on the test set, which indicates that the model can effectively identify different traffic and transportation scenes.
According to the invention, through extracting the time sequence characteristic data of the single transportation cleaning data, rich time sequence characteristic information such as time interval, speed change rate, load change rate, parking duration and acceleration change can be obtained, so that the dynamic change of the transportation process can be more comprehensively described. And carrying out distributed feature extraction on the group transportation cleaning data to obtain the overall feature of group transportation so as to mine the statistical rule behind the data, thereby realizing the advanced processing of the data. The group traffic distribution characteristic data is mapped to the historical traffic experience, and the current traffic condition can be verified and evaluated through the historical data, so that the prediction accuracy is improved. The time sequence scene recognition classification is carried out through a long and short time memory network (LSTM), the finer classification is carried out on the traffic data, and the scene recognition precision is improved. By adopting deep learning technologies such as long-short-term memory networks and the like, compared with the traditional method, the time sequence data can be better processed, and the recognition and classification effects are improved. Meanwhile, the characteristics of single traffic and group traffic are considered, so that data of different layers are comprehensively utilized, and more comprehensive information is provided. By layering different types of data, the efficiency of data processing can be effectively improved, and the analysis and prediction process is more efficient.
Preferably, step S14 is specifically:
step S141: mapping traffic parameters according to the time sequence traffic scene data to generate specific time sequence threshold setting data;
specifically, scene features are mapped to corresponding traffic parameter ranges, such as speed, load, etc., using, for example, time-series traffic scene data. Based on history data or experience of a field expert, a threshold value at a specific timing is set for outlier detection.
Step S142: abnormal value screening is carried out on the transportation cleaning data by utilizing specific time sequence threshold setting data, so that transportation screening data are obtained;
specifically, for example, traffic wash data is compared to a set particular timing threshold, and data points that exceed the threshold are identified.
Step S143: and reconstructing traffic data of the traffic screening data according to the time sequence traffic scene data, thereby obtaining the traffic data.
Specifically, complete traffic data is reconstructed, for example, using time-series traffic scene data and the screened data.
According to the invention, the traffic parameter mapping is carried out according to the time sequence traffic scene data, and the specific time sequence threshold setting data is generated, so that the judgment of the abnormal value is more targeted, and the abnormal condition can be identified more accurately. Abnormal value screening is carried out on the transportation cleaning data by utilizing specific time sequence threshold setting data, so that the abnormal data can be more effectively eliminated, and the accuracy and the credibility of the data are ensured. The screened data is reconstructed according to the time sequence traffic scene data, so that the original traffic data can be restored, and further analysis and application are facilitated. By reconstructing traffic data, more comprehensive traffic conditions including information of various traffic parameters can be obtained, and comprehensive understanding of traffic conditions is facilitated. Through a series of processing steps including outlier screening and data reconstruction, the quality of traffic data can be improved, and the accuracy and reliability of the data are ensured.
Preferably, the load abnormality detection data includes regular load abnormality detection data and vehicle load abnormality detection data, and step S2 specifically includes:
step S21: acquiring traffic position rule data according to traffic position data corresponding to the traffic data;
specifically, accurate location information of the transportation vehicle is acquired, for example, using a Global Positioning System (GPS) or the like.
Step S22: carrying out regular load abnormality detection on the transportation data according to the transportation position regular data so as to obtain regular load abnormality detection data;
specifically, for example, load thresholds for different locations are set according to traffic location rule data. Real-time load data are acquired by means of load sensors and the like, are compared with rules, and whether abnormality exists or not is identified. For example, a section of highway is divided into three sections according to traffic location rule data: a start section, a middle section, and an end section. Each section corresponds to a different road condition and traffic condition. Starting section: based on historical data analysis, there are typically many vehicles entering the highway in the initial section, so the load will be slightly higher than in the other sections, and the system sets a relatively high load threshold, for example 15000 kg, based on this law. Intermediate section: in the middle section, the speed of the vehicle will be more stable, so the load threshold can be set at a medium level, the system sets the load threshold at 12000 kg. End section: at the end of the section, the vehicle may slow down or perform an exit operation, so the load will again be slightly higher than in the middle section, the system setting the load threshold at around 13000 kg. And according to the rule load threshold value set in the way, carrying out load abnormality detection on the traffic data according to different sections, thereby obtaining rule load abnormality detection data. If the load of a vehicle exceeds a set threshold, it may be marked as load abnormality.
Step S23: detecting vehicle damage according to the vehicle image data in the traffic data, thereby obtaining vehicle damage detection data;
specifically, the vehicle image is preprocessed, for example, using an image processing technique, to extract useful information. And constructing a vehicle damage detection model by using deep learning or a traditional computer vision algorithm. If the vehicle image is subjected to binarization processing, binarization vehicle image data are obtained, statistical distribution processing is carried out on the binarization vehicle image data, vehicle image statistical feature data are obtained, a preset vehicle damage detection model is utilized to carry out recognition calculation on the vehicle image statistical feature data, and vehicle damage detection data are obtained, wherein the vehicle damage detection model is an empirical model which is subjected to binarization according to standard vehicle image data, statistical feature extraction and labeling according to preset vehicle damage labels, or machine learning training is carried out according to standard vehicle image data, and the standard vehicle image data are training data sets in a preset database.
Step S24: and detecting abnormal running of the vehicle according to the vehicle damage detection data, thereby obtaining abnormal load detection data of the vehicle.
Specifically, data such as speed, acceleration, and the like when the vehicle is traveling is acquired, for example, using an in-vehicle sensor or the like. The driving data is analyzed using machine learning or statistical methods to identify whether abnormal behavior exists. For example, assume that during the training phase, the model learns that the average speed in a normal running state is 60km/h, the maximum acceleration is 3m/s, and the running distance is 100km. If the average speed of a certain vehicle is monitored to be more than 80km/h in real time, the acceleration is more than 5m/s, and the driving distance is only 50km, the abnormal behavior is judged.
According to the invention, the transportation position rule data is obtained according to the transportation data, and the load abnormality detection is carried out by combining the position information and the rule information, so that the detection is more comprehensive and accurate. The regular load abnormality detection is carried out by utilizing the traffic transportation position rule data, so that the existing load abnormality condition can be rapidly identified, and the detection efficiency is improved. And the vehicle damage detection is carried out by combining the vehicle image data, so that additional auxiliary information can be provided for the load abnormality detection, and the detection accuracy is improved. The abnormal running condition of the vehicle is detected through the vehicle damage detection data, so that the existing abnormal load condition can be further screened out, and the detection precision is improved. The load abnormality detection is divided into a plurality of steps, each step has a specific function, a layering detection strategy is formed between each step, and the detection accuracy and efficiency are improved. By combining the position data, the image data and the damage detection data, different types of information are comprehensively processed, so that the detection is more comprehensive and accurate. Through a multi-level detection strategy and comprehensive utilization of different information, the false alarm rate can be reduced, and the reliability of the detection result is improved.
Preferably, in step S22, the rule load abnormality detection is processed by a rule load abnormality detection calculation formula, where the rule load abnormality detection calculation formula specifically includes:
;
for regular load abnormality detection data, < > for>For the quantity data of the traffic data, +.>Order item data of transportation data, < ->For the first->The value of the individual load data points,/">For the +.>Numerical value of each angle data point, +.>For the distance data of the vehicle>For traffic data->Is->Specific data of individual traffic dataItem (S)>Is->A particular data item of traffic location rule data.
The invention constructs a rule load abnormality detection calculation formula which is used for evaluating whether the load condition in the transportation process accords with rules or standards. Wherein the method comprises the steps ofPart will transport quantity->The increase of (2) is scaled on a logarithmic scale, which can maintain a controllable range of computational complexity when processing large amounts of traffic data, while for larger +.>Value, it can slow down +.>So that anomaly detection can also operate effectively for large data sets. / >The product of the load values representing all traffic data points, by taking the square root of the product of these values, takes into account the aggregate nature of the load data, making the algorithm robust to load variations between different traffic.This term is used to process angle data, which plays an important role in regular load anomaly detection, and relates to directional information, for example, in vehicle transportation, angle is related to cargo distribution, stacking mode and the like, and information on this aspect can be effectively extracted through the ratio of sine and cosine. />This term includes the derivative of the distance L travelled by the vehicle, taking into account dynamic changes during the travel of the vehicle, such as road conditions, accelerations, etc. The method is used for detecting the abnormality in the transportation field, and the method comprises the processing of information in various aspects such as load, angle, distance and the like, so that the algorithm has adaptability to different conditions, and plays an important role in improving transportation safety, optimizing loading strategies and the like.
Preferably, in step S23, the vehicle damage detection is performed by a vehicle damage detection model, where the vehicle damage detection model is constructed specifically as follows:
Step S231: acquiring standard vehicle image data and standard vehicle damage tag data;
specifically, a standard vehicle is photographed, for example, using a high-resolution camera or a professional in-vehicle camera, to acquire image data. The damaged area is marked, and marking tools or software can be used.
Step S232: performing binarization processing on the standard vehicle image data so as to obtain standard vehicle binarization image data;
specifically, for example, an image processing algorithm such as OTSU, an adaptive threshold, or the like is applied to convert a standard vehicle image into a binary image, highlighting the lesion site.
Step S233: edge detection is carried out on the standard vehicle binarized image data, so that standard vehicle edge detection image data are obtained;
specifically, edge information in the image is detected, for example, using an image processing algorithm such as Canny, sobel, or the like, so as to highlight the outline of the damaged portion.
Step S234: performing normal distribution calculation according to the standard vehicle edge detection image data, so as to obtain standard vehicle normal distribution data;
specifically, for example, based on standard vehicle edge detection image data, the mean and standard deviation of the normal distribution are calculated using a statistical method.
Step S235: and marking the normal distribution data of the standard vehicle according to the damage label data of the standard vehicle, thereby obtaining a damage detection model of the vehicle.
Specifically, for example, according to the damage tag data, the damage-related part in the normal distribution data is marked to form a training set or a vehicle damage detection model. The method can be used for further machine learning training and also can be directly used for damage detection.
According to the invention, the standard vehicle image data and the standard vehicle damage label data are acquired, so that a data set containing the standard image and corresponding damage information is established, and a basis is provided for model training. The image is preprocessed through binarization processing and edge detection on the standard vehicle image data, so that the damaged part in the image is more prominent, and subsequent feature extraction is facilitated. By acquiring standard vehicle edge detection image data through edge detection, edge information in a vehicle image can be extracted, which is one of important features of damage detection. And carrying out normal distribution calculation based on the standard vehicle edge detection image data, and carrying out statistical analysis on the damaged area to provide additional characteristic information. The normal distribution data is marked through the standard vehicle damage label data, so that the model can learn the characteristics of the damaged area, and the construction of the vehicle damage detection model is realized. Compared with the prior art, the method avoids complex calculation of machine learning, and simultaneously refines the depth data behind the data through edge calculation and normal distribution calculation so as to provide accurate and reliable coarse index data instead of higher-definition data specific to a certain index value layer, thereby improving the practicability.
Preferably, step S3 is specifically:
step S31: acquiring historical traffic and transportation data, wherein the historical traffic and transportation data comprises conventional historical traffic and transportation data and abnormal historical traffic and transportation data;
specifically, for example, historical traffic records provided by traffic departments, carrier companies, or related institutions, including data on vehicle travel tracks, load conditions, and the like, are obtained.
Step S32: carrying out data association on the load abnormal detection data according to the conventional historical traffic and transportation data and the abnormal historical traffic and transportation data, so as to obtain conventional associated traffic and transportation data and abnormal associated traffic and transportation data;
specifically, for example, the load abnormality detection data is correlated with the regular and abnormal historical transportation data using a data correlation algorithm such as KNN, correlation rule mining, or the like, and the correlated data is found. Conventional historical traffic data: record 1, license plate number-country A123, load-10 tons, route-A to B, time-2022, 9 months and 10 days, record 2, license plate number-country B456, load-12 tons, route-B to C, time-2022, 9 months and 12 days, record 3, license plate number-country C789, load-9 tons, route-A to C, time-2022, 9 months and 14 days; abnormal historical traffic data: record 4, license plate number-country X111, load-20 tons, route-D to E, time-2022, 9 months and 11 days, record 5, license plate number-country Y222, load-18 tons, route-B to F, time-2022, 9 months and 13 days, record 6, license plate number-country Z333, load-22 tons, route-C to G, time-2022, 9 months and 15 days; detecting and recording license plate number-national A123, load-15 tons, route-A to B, time-2022 9 months and 16 days; the system correlates the detection records with the historical data by matching key information such as license plate number, load, route, etc. using a data correlation algorithm such as KNN or correlation rule mining. The results obtained: conventional associated transportation data: record 1, license plate number-country A123, load-10 tons, route-A to B, time-2022 9 months 10 days; abnormal associated traffic data: recording 6, license plate number-country Z333, load-22 tons, route-C to G, time-2022 9 months and 15 days, associating the load abnormality detection data with the historical data by the system, and finding out conventional associated traffic data and abnormal associated traffic data.
Step S33: carrying out traffic and transportation confidence calculation on the conventional associated traffic and transportation data and the abnormal associated traffic and transportation data so as to acquire first overrun load abnormal data;
specifically, the first overrun load anomaly data is identified by confidence calculations of the associated regular and anomaly transportation data, for example using statistical methods or machine learning models. Such as conventional associated traffic data and anomalous associated traffic data, include a number of characteristics such as load, speed, time, etc. The confidence level may be calculated using statistical methods or machine learning models. If the system adopts a machine learning model, a Support Vector Machine (SVM) is used to train a classifier, and the classifier can distinguish normal transportation data from abnormal transportation data according to the characteristics. Training data set: and taking the conventional associated transportation data and the abnormal associated transportation data as training sets, extracting corresponding characteristics (such as load, speed and the like) as input, and marking the characteristics as normal or abnormal. Training a model: a classification model is trained using SVM or other classification algorithms to accurately distinguish between normal and abnormal data. Confidence coefficient calculation: for each piece of associated transportation data, it is input into the trained model, resulting in a confidence score. The higher the confidence score, the more anomalous the transportation data. Setting a threshold value: according to the actual demand, a threshold value may be set, and data above the threshold value will be determined as first overrun load abnormality data.
Step S34: and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data, thereby obtaining second overrun load abnormality data.
Specifically, for example, load limiting standards for different road types, including bridges, tunnels, etc., are formulated or acquired for comparison. And comparing the load abnormal detection data with a preset road load limiting standard, and identifying second overrun load abnormal data. And comparing the load abnormality detection data with a preset road load limiting standard. For example, if a piece of transportation data passes through a bridge, it needs to be compared with the load limiting standard of the bridge. Examples: assuming that the load limit of a bridge is 10 tons and the load of a piece of transportation data is 12 tons, it can be determined that the transportation data has a second overrun load abnormality.
According to the method, the conventional historical traffic data and the abnormal historical traffic data are obtained, the historical data are taken into consideration, and the comprehensiveness and the accuracy of load abnormal detection are improved. The current load abnormal data and the historical data can be compared by correlating the load abnormal detection data according to the historical data, and the existing abnormal situation can be identified. And the traffic confidence coefficient calculation is carried out on the associated traffic data, so that the reliability of the abnormal condition of the load can be quantized, and the reliability of the detection result is improved. And carrying out second overrun load abnormality association on load abnormality detection data through preset road load limiting standard data, so that the accuracy of abnormality detection is further improved. The concepts of the first overrun and the second overrun are introduced, and the overrun condition is detected in a grading manner, so that the severity of the overrun condition can be accurately judged. By comprehensively utilizing historical data, confidence coefficient calculation and overrun standard data, the anomaly detection is more accurate and reliable.
Preferably, the present invention also provides an overrun transportation violation monitoring analysis and early warning system for executing the overrun transportation violation monitoring analysis and early warning method as described above, the overrun transportation violation monitoring analysis and early warning system comprising:
the traffic data validity screening module is used for acquiring traffic and transportation basic data and carrying out data validity screening on the traffic and transportation basic data so as to acquire traffic and transportation data;
the load abnormal detection module is used for carrying out load abnormal detection on the traffic data so as to obtain load abnormal detection data;
the overrun load abnormality association module is used for acquiring historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data so as to acquire first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data so as to acquire second overrun load abnormality data;
and the early warning generation module is used for carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data so as to acquire overrun transportation illegal monitoring early warning data.
When the basic data of traffic and transportation are acquired, the data validity screening is performed, invalid or abnormal data are eliminated, the data quality of subsequent analysis is ensured, and the analysis accuracy is improved. By adopting an advanced load abnormality detection technology and analyzing and processing traffic data, the load abnormality situation is accurately identified, and the problems of missing report and false report in the traditional method are avoided. By introducing historical traffic data and carrying out correlation analysis of first overrun and second overrun on load abnormal detection data, multi-level and omnibearing detection of abnormal conditions is realized. This can improve the accuracy and reliability of the abnormal situation. And the preset road load limiting standard data is adopted, and the second overrun load abnormality association analysis is carried out on the load abnormality detection data, so that the abnormality detection precision is further improved. Based on the first overrun and the second overrun load abnormal data, the monitoring system can timely find and alarm the overrun transportation condition by generating the overrun transportation illegal monitoring early warning data, so that the illegal investigation efficiency is forcefully improved. Through the establishment of the comprehensive monitoring and early warning system, the occurrence frequency of the overrun transportation violations can be effectively reduced, and the road traffic safety level is improved. Traffic jam and accident risk caused by overrun transportation are reduced, road traffic resource utilization is optimized, and the method has positive pushing effect on urban traffic planning and resource allocation.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. The monitoring, analyzing and early warning method for the overrun transportation violations is characterized by comprising the following steps of:
step S1, including:
step S11: acquiring basic data of traffic transportation;
step S12: carrying out traffic data cleaning on the traffic basic data so as to obtain traffic cleaning data, wherein the traffic cleaning data comprises single traffic cleaning data and group traffic cleaning data; the transportation data cleaning is carried out through a transportation data cleaning calculation formula, wherein the transportation data cleaning calculation formula specifically comprises the following steps:
;
For transportation cleaning data->Order item data for basic data of transportation, < +.>For the quantity data of the traffic base data, +.>Is->Personal transportation base data, < >>Is->Traffic quality index data corresponding to the individual traffic base data,/-for>For cleaning the lower limit of the data range, +.>For cleaning the upper limit of the data range, +.>For detecting abnormal traffic data->For the data item to be detected in the transportation base data, < > for the data item to be detected in the transportation base data>For data correction factor, ++>For the vehicle driving direction angle data, +.>For traffic time change data, < >>Is traffic and transportation time data;
step S13, including:
step S131: extracting time sequence transportation characteristic data according to the single transportation cleaning data so as to obtain time sequence transportation characteristic data, wherein the time sequence transportation characteristic data comprises time interval data, speed change rate data, load change rate data, parking time length data and acceleration change data;
step S132: carrying out distribution feature extraction on the group transportation cleaning data so as to obtain group transportation distribution feature data;
step S133: historical traffic experience mapping is carried out on the group traffic distribution characteristic data, so that traffic scene category data are obtained;
Step S134: carrying out time sequence traffic scene identification and classification on time sequence traffic characteristic data according to traffic scene type data so as to obtain time sequence traffic scene data, wherein the time sequence traffic scene identification and classification is carried out through a long-short-term memory network, the time sequence traffic scene data comprises specific time point data, vehicle speed data, vehicle density data and vehicle road condition data, the vehicle density data comprises high vehicle density data and low vehicle density data, and the vehicle road condition data comprises congestion road condition data and smooth road condition data;
step S14: screening traffic abnormal values of the traffic cleaning data according to the time sequence traffic scene data, so as to obtain traffic data;
step S2: carrying out load abnormality detection on the traffic data so as to obtain load abnormality detection data;
step S3, including:
step S31: acquiring historical traffic and transportation data, wherein the historical traffic and transportation data comprises conventional historical traffic and transportation data and abnormal historical traffic and transportation data;
step S32: carrying out data association on the load abnormal detection data according to the conventional historical traffic and transportation data and the abnormal historical traffic and transportation data, so as to obtain conventional associated traffic and transportation data and abnormal associated traffic and transportation data;
Step S33: carrying out traffic and transportation confidence calculation on the conventional associated traffic and transportation data and the abnormal associated traffic and transportation data so as to acquire first overrun load abnormal data;
step S34: carrying out second overrun load abnormality association on load abnormality detection data according to preset road load limiting standard data, thereby obtaining second overrun load abnormality data;
step S4: and carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data, thereby acquiring overrun transportation illegal monitoring early warning data.
2. The method according to claim 1, wherein step S14 is specifically:
step S141: mapping traffic parameters according to the time sequence traffic scene data to generate specific time sequence threshold setting data;
step S142: abnormal value screening is carried out on the transportation cleaning data by utilizing specific time sequence threshold setting data, so that transportation screening data are obtained;
step S143: and reconstructing traffic data of the traffic screening data according to the time sequence traffic scene data, thereby obtaining the traffic data.
3. The method according to claim 1, wherein the load abnormality detection data includes regular load abnormality detection data and vehicle load abnormality detection data, and step S2 is specifically:
Step S21: acquiring traffic position rule data according to traffic position data corresponding to the traffic data;
step S22: carrying out regular load abnormality detection on the transportation data according to the transportation position regular data so as to obtain regular load abnormality detection data;
step S23: detecting vehicle damage according to the vehicle image data in the traffic data, thereby obtaining vehicle damage detection data;
step S24: and detecting abnormal running of the vehicle according to the vehicle damage detection data, thereby obtaining abnormal load detection data of the vehicle.
4. A method according to claim 3, wherein the rule load abnormality detection in step S22 is handled by a rule load abnormality detection calculation formula, wherein the rule load abnormality detection calculation formula is specifically:
;
for regular load abnormality detection data, < > for>For the quantity data of the traffic data, +.>Order item data of transportation data, < ->For the first->The value of the individual load data points,/">For the +.>Numerical value of each angle data point, +.>For the distance data of the vehicle>For traffic data- >Is->Specific data item of individual traffic data, +.>Is->A particular data item of traffic location rule data.
5. The method according to claim 3, wherein the vehicle damage detection in step S23 is performed by a vehicle damage detection model, and the vehicle damage detection model is constructed specifically by:
step S231: acquiring standard vehicle image data and standard vehicle damage tag data;
step S232: performing binarization processing on the standard vehicle image data so as to obtain standard vehicle binarization image data;
step S233: edge detection is carried out on the standard vehicle binarized image data, so that standard vehicle edge detection image data are obtained;
step S234: performing normal distribution calculation according to the standard vehicle edge detection image data, so as to obtain standard vehicle normal distribution data;
step S235: and marking the normal distribution data of the standard vehicle according to the damage label data of the standard vehicle, thereby obtaining a damage detection model of the vehicle.
6. An overrun transportation violation monitoring analysis and early warning system for performing the overrun transportation violation monitoring analysis and early warning method of claim 1, the overrun transportation violation monitoring analysis and early warning system comprising:
The traffic data validity screening module is used for acquiring traffic and transportation basic data and carrying out data validity screening on the traffic and transportation basic data so as to acquire traffic and transportation data;
the load abnormal detection module is used for carrying out load abnormal detection on the traffic data so as to obtain load abnormal detection data;
the overrun load abnormality association module is used for acquiring historical traffic and transportation data, carrying out first overrun load abnormality association on load abnormality detection data according to the historical traffic and transportation data so as to acquire first overrun load abnormality data, and carrying out second overrun load abnormality association on the load abnormality detection data according to preset road load limiting standard data so as to acquire second overrun load abnormality data;
and the early warning generation module is used for carrying out early warning generation according to the first overrun load abnormal data and the second overrun load abnormal data so as to acquire overrun transportation illegal monitoring early warning data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311602581.1A CN117315943B (en) | 2023-11-28 | 2023-11-28 | Monitoring analysis and early warning method and system for overrun transportation violations |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311602581.1A CN117315943B (en) | 2023-11-28 | 2023-11-28 | Monitoring analysis and early warning method and system for overrun transportation violations |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117315943A true CN117315943A (en) | 2023-12-29 |
CN117315943B CN117315943B (en) | 2024-02-06 |
Family
ID=89297592
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311602581.1A Active CN117315943B (en) | 2023-11-28 | 2023-11-28 | Monitoring analysis and early warning method and system for overrun transportation violations |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117315943B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050065711A1 (en) * | 2003-04-07 | 2005-03-24 | Darwin Dahlgren | Centralized facility and intelligent on-board vehicle platform for collecting, analyzing and distributing information relating to transportation infrastructure and conditions |
US20160210857A1 (en) * | 2015-01-15 | 2016-07-21 | International Business Machines Corporation | Vehicle-based abnormal travel event detecting and reporting |
US20190122544A1 (en) * | 2017-10-20 | 2019-04-25 | Microsoft Technology Licensing, Llc | Traffic data reconciliation and brokering |
CN110148230A (en) * | 2019-05-20 | 2019-08-20 | 兴民智通(武汉)汽车技术有限公司 | A kind of vehicle load-carrying prediction technique based on LSTM neural network |
CN110232824A (en) * | 2019-05-27 | 2019-09-13 | 武汉理工大学 | A kind of non-contact vehicle overload identification early warning system |
WO2020052436A1 (en) * | 2018-09-12 | 2020-03-19 | 杭州海康威视数字技术股份有限公司 | Vehicle overload alarming method and apparatus, electronic device, and storage medium |
CN111260925A (en) * | 2020-02-24 | 2020-06-09 | 南京震坤物联网科技有限公司 | Urban bridge transfinite overload safety monitoring system based on block chain technology |
CN112687099A (en) * | 2020-12-16 | 2021-04-20 | 北京北大千方科技有限公司 | Method and device for judging overload suspected vehicle |
CN113865683A (en) * | 2021-12-01 | 2021-12-31 | 苏州博宇鑫交通科技有限公司 | Urban viaduct overweight and overload dynamic early warning method based on machine vision |
CN116863696A (en) * | 2023-07-11 | 2023-10-10 | 福建易华路信息技术有限公司 | Method for predicting and intercepting early warning of overrun vehicle route |
-
2023
- 2023-11-28 CN CN202311602581.1A patent/CN117315943B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050065711A1 (en) * | 2003-04-07 | 2005-03-24 | Darwin Dahlgren | Centralized facility and intelligent on-board vehicle platform for collecting, analyzing and distributing information relating to transportation infrastructure and conditions |
US20160210857A1 (en) * | 2015-01-15 | 2016-07-21 | International Business Machines Corporation | Vehicle-based abnormal travel event detecting and reporting |
US20190122544A1 (en) * | 2017-10-20 | 2019-04-25 | Microsoft Technology Licensing, Llc | Traffic data reconciliation and brokering |
WO2020052436A1 (en) * | 2018-09-12 | 2020-03-19 | 杭州海康威视数字技术股份有限公司 | Vehicle overload alarming method and apparatus, electronic device, and storage medium |
CN110148230A (en) * | 2019-05-20 | 2019-08-20 | 兴民智通(武汉)汽车技术有限公司 | A kind of vehicle load-carrying prediction technique based on LSTM neural network |
CN110232824A (en) * | 2019-05-27 | 2019-09-13 | 武汉理工大学 | A kind of non-contact vehicle overload identification early warning system |
CN111260925A (en) * | 2020-02-24 | 2020-06-09 | 南京震坤物联网科技有限公司 | Urban bridge transfinite overload safety monitoring system based on block chain technology |
CN112687099A (en) * | 2020-12-16 | 2021-04-20 | 北京北大千方科技有限公司 | Method and device for judging overload suspected vehicle |
CN113865683A (en) * | 2021-12-01 | 2021-12-31 | 苏州博宇鑫交通科技有限公司 | Urban viaduct overweight and overload dynamic early warning method based on machine vision |
CN116863696A (en) * | 2023-07-11 | 2023-10-10 | 福建易华路信息技术有限公司 | Method for predicting and intercepting early warning of overrun vehicle route |
Non-Patent Citations (3)
Title |
---|
张英贵等: "基于关联规则挖掘的铁路超限超重货物运输路径决策方法", 铁道科学与工程学报, vol. 18, no. 7 * |
王志雄;: "基于大数据技术的高速公路预测预警处置平台", 中国交通信息化, no. 01 * |
赵建峰;王巍;: "基于物联网技术的车辆超载远程监测***设计与实现", 公路, no. 10 * |
Also Published As
Publication number | Publication date |
---|---|
CN117315943B (en) | 2024-02-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yu et al. | Prediction of bus travel time using random forests based on near neighbors | |
US20220084395A1 (en) | System and method for determining a vehicle classification from gps tracks | |
CN109145954B (en) | Network taxi appointment travel safety evaluation method and system based on multi-source time-space data | |
CN110544373B (en) | Truck early warning information extraction and risk identification method based on Beidou Internet of vehicles | |
CN105303197A (en) | Vehicle following safety automatic assessment method based on machine learning | |
CN108932255B (en) | Vehicle comprehensive capacity analysis method and device | |
CN109493449B (en) | Freight train cargo state estimation method based on freight train GPS track data and high-speed transaction data | |
CN108305461A (en) | A kind of determination method and apparatus for evading expense suspected vehicles | |
Meng et al. | Expressway crash prediction based on traffic big data | |
Taghavi et al. | Identifying truck stops from a large stream of gps data via a hidden markov chain model | |
CN116168356A (en) | Vehicle damage judging method based on computer vision | |
CN117315943B (en) | Monitoring analysis and early warning method and system for overrun transportation violations | |
CN114973681B (en) | In-transit vehicle sensing method and device | |
CN115565373B (en) | Expressway tunnel accident real-time risk prediction method, device, equipment and medium | |
CN116092037B (en) | Vehicle type identification method integrating track space-semantic features | |
CN116882755A (en) | Method for predicting oil theft risk of trucking route based on vehicle-mounted Tbox data | |
CN114973156B (en) | Night muck car detection method based on knowledge distillation | |
CN116206451A (en) | Intelligent traffic flow data analysis method | |
CN114565155A (en) | Road safety testing method and device, electronic equipment and storage medium | |
Golze et al. | Impact analysis of accidents on the traffic flow based on massive floating car data | |
Hou | Applications of big data technology in intelligent transportation system | |
CN114201530A (en) | Suspected abnormal operation passenger car early discrimination and preventive supervision method | |
CN114119256A (en) | UBI dangerous chemical vehicle driving behavior acquisition and analysis system and premium discount method | |
CN114066305A (en) | Early warning method for high risk area where hazardous chemical substance transport vehicle abnormally resides | |
Feng et al. | Learning a precipitation indicator from traffic speed variation patterns |
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 |