CN117935169B - Passenger transport hub queuing time length prediction system based on image recognition - Google Patents
Passenger transport hub queuing time length prediction system based on image recognition Download PDFInfo
- Publication number
- CN117935169B CN117935169B CN202410188480.2A CN202410188480A CN117935169B CN 117935169 B CN117935169 B CN 117935169B CN 202410188480 A CN202410188480 A CN 202410188480A CN 117935169 B CN117935169 B CN 117935169B
- Authority
- CN
- China
- Prior art keywords
- time length
- personnel
- target
- queuing
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000003550 marker Substances 0.000 claims abstract description 21
- 238000007405 data analysis Methods 0.000 claims abstract description 13
- 238000004364 calculation method Methods 0.000 claims description 32
- 230000033001 locomotion Effects 0.000 claims description 27
- 238000012937 correction Methods 0.000 claims description 25
- 238000010191 image analysis Methods 0.000 claims description 14
- 230000000877 morphologic effect Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 8
- 238000003708 edge detection Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000012935 Averaging Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000029305 taxis Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
Abstract
The invention relates to the technical field of intelligent transportation, in particular to a passenger transport hub queuing time length prediction system based on image recognition, which comprises the following steps: the data acquisition module acquires the history passing duration of any queuing channel through a history database; the data analysis module acquires a first passing duration by acquiring personnel distribution information of any queuing channel and weather information outside a passenger transport hub and correcting historical passing duration; the calibration module is used for acquiring the actual moving speed of any marker in any queuing channel, calibrating the first passing time length and acquiring a second passing time length; the determining module is used for determining a recommended channel according to the departure time and the second passing time of the target passenger; the updating module is used for collecting the target moving speed of the target passenger in the recommended channel in real time so as to update the second pass time length according to the target moving speed and obtain the final predicted time length. The invention improves the real-time performance of the queuing time prediction of the passenger transport hub.
Description
Technical Field
The invention relates to the technical field of intelligent transportation, in particular to a passenger transport hub queuing time length prediction system based on image recognition.
Background
With the rapid development of the transportation industry, the passenger flow of passenger transport hubs is increasing, and the problem of passenger queuing is also becoming more prominent. The prediction of the queuing time of the passenger transport hub becomes a key link for improving the service quality. The passenger transport hub can better schedule resources by predicting queuing time, optimize service flow and improve traveling experience of passengers.
The patent document with the Chinese patent publication number of CN114565128A discloses a passenger queuing time prediction method for a passenger transportation junction taxi based on image recognition, which comprises the following steps: step 1, dividing the predicted time into T time intervals [ T1, T2..ti-1, ti ]; step 2, obtaining the average arrival rate lambdaj of passengers in a waiting area of the taxi in the ith time interval; step 3, obtaining the average service rate mu i of the taxis in the ith time interval and the number Gi of service desks capable of providing service in unit time; step 4, judging the taxi number information T of the storage yard, and calculating the total transport capacity TT of the taxis; step 5, judging whether the taxi carrying capacity is sufficient or not; and 6, estimating the queuing time of the passengers in the waiting area of the taxi according to the queuing theory model.
In the prior art, the queuing time length prediction of the passenger transport hub is based on a large amount of historical data, the change rule of the queuing time length is found out through analyzing the historical data and a prediction model, the future queuing time length is predicted, and the trend change of the future time length cannot be accurately predicted only by relying on the historical data, so that the time length prediction result is lagged.
Disclosure of Invention
Therefore, the invention provides a passenger transport hub queuing time length prediction system based on image recognition, which can solve the problem of hysteresis of passenger transport hub queuing time length prediction by gradually calibrating and updating the historical passing time length through collecting personnel distribution and weather information of the current queuing channel, real-time collecting the moving speed of the identification personnel and real-time collecting the moving speed of the target passengers.
In order to achieve the above object, the present invention provides a passenger transportation hub queuing time length prediction system based on image recognition, including:
the data acquisition module is used for acquiring the history passing duration of any queuing channel through the history database;
The data analysis module is connected with the data acquisition module and used for acquiring personnel distribution information of any queuing channel and weather information outside a passenger transport hub, correcting the historical passing duration through the personnel distribution information and the weather information and acquiring a first passing duration;
The calibration module is connected with the data analysis module and used for acquiring the actual moving speed of any marker in any queuing channel, calibrating the first passing duration and acquiring a second passing duration;
The determining module is connected with the calibration module and is used for determining a recommended channel according to the departure time of the target passenger and the second passing time;
And the updating module is connected with the determining module and used for acquiring the target moving speed of the target passenger in the recommended channel in real time so as to update the second pass time length according to the target moving speed and acquire the final predicted time length.
Further, the data analysis module comprises a personnel acquisition unit, a weather acquisition unit and a first acquisition unit, wherein,
The personnel acquisition unit is used for acquiring personnel images of any queuing channel through the image acquisition equipment, analyzing a plurality of personnel images, judging personnel distribution conditions, and acquiring personnel calibration parameters according to the personnel distribution conditions;
the weather acquisition unit is used for determining the actual severe weather grade outside the passenger transport hub and acquiring weather calibration parameters according to the actual severe weather grade;
The first acquisition unit is connected with the personnel acquisition unit and the weather acquisition unit and used for calibrating the historical passing duration according to the personnel calibration parameters and the weather calibration parameters to acquire a first passing duration.
Further, the calibration module comprises an image acquisition unit, an image analysis unit and a speed calibration unit, wherein,
The image acquisition unit is used for acquiring the moving process of the identification personnel through the image acquisition equipment, acquiring a moving video, extracting continuous frame images in the moving video, and acquiring a plurality of moving images;
The image analysis unit is connected with the image acquisition unit and used for extracting edge contour features of the identification personnel in a plurality of moving images, determining target points on the edge contour features of the identification personnel, and performing motion trail fitting on the identification personnel according to the plurality of target points;
The speed calibration unit is connected with the image analysis unit and used for acquiring the actual moving speed of the marker according to the moving track, calculating a speed correction coefficient according to the actual moving speed and calibrating the first passing duration according to the speed correction coefficient.
Further, the personnel acquisition unit comprises a personnel acquisition subunit, a personnel analysis subunit and a personnel calculation subunit, wherein,
The personnel acquisition subunit is used for acquiring a plurality of morphological images in any queuing channel through the image acquisition equipment;
The personnel analysis subunit is connected with the personnel acquisition subunit and used for analyzing a plurality of morphological images, extracting edge contour features in a plurality of morphological images and identifying the number of old people, the number of children and the number of passengers carrying large articles in the queuing channel through the plurality of edge contour features;
The personnel calculating subunit is connected with the personnel analyzing subunit and used for calculating and obtaining personnel calibration parameters according to the number of the old people, the number of the children and the number of the passengers carrying large articles in the queuing channel.
Further, the personnel analysis subunit identifies old people and children by judging the curvature and the height of the edge profile, and identifies passengers carrying large articles by identifying the area of the edge profile of the object in the edge profile.
Further, the image analysis unit comprises a target determination subunit and a fitting subunit, wherein,
The target determination subunit is configured to obtain an edge contour feature of the identified person in the moving image through an edge detection algorithm, and determine a center point of the edge contour feature of the identified person as the target point;
The fitting subunit is connected with the target determination subunit and is used for connecting target points between adjacent frame images to acquire the motion trail of the marker.
Further, the speed calibration unit comprises a speed calculation subunit and a duration calibration subunit, wherein,
The speed calculation subunit is used for obtaining the actual movement speed of the marker by calculating the distance value between two adjacent frame images of the movement track divided by the time interval between the adjacent frame images;
The time length calibration subunit is connected with the speed calculation subunit, compares the actual moving speed with a preset moving speed, calculates a speed correction coefficient according to a comparison result, and calibrates the first passing time length according to the speed correction coefficient.
Further, the determining module comprises a duration calculating unit and a selecting unit, wherein,
The time length calculation unit is used for calculating preset waiting time length according to the departure time of the target passenger;
The selecting unit is connected with the time length calculating unit and used for sorting the second passing time lengths corresponding to the queuing channels from small to large to obtain a time length sequence, marking the queuing channels corresponding to the second passing time length smaller than the preset waiting time length in the time length sequence, and selecting the first queuing channel as a recommended channel.
Further, the updating module comprises a speed acquisition unit and an updating unit, wherein,
The speed acquisition unit is used for acquiring a plurality of images containing target passengers and target objects through the image acquisition equipment, analyzing the images to acquire a plurality of target passenger outlines and a plurality of target object outlines in the images, and calculating the target moving speed of the target passengers according to the distance change values between the target passenger outlines and the target object outlines;
the updating unit is connected with the speed acquisition unit and used for comparing the target moving speed of the target passenger with the preset moving speed, calculating an updating parameter according to a comparison result, and updating the second pass time length according to the updating parameter to obtain the final predicted time length.
Further, the data acquisition module comprises a building unit and a historical duration acquisition unit, wherein,
The establishing unit is used for establishing the historical database, and the historical database comprises queuing channel numbers and average passing time length of passengers in corresponding each period;
the historical time length obtaining unit is connected with the establishing unit and is used for averaging the average passing time length of passengers in each period corresponding to any queuing channel, and taking the average result as the historical passing time length of the queuing channel.
Compared with the prior art, the method has the advantages that the data acquisition module is arranged to establish a historical database to provide accurate and traceable historical passing time length data, reliable data support is provided for subsequent passing time length prediction, the data analysis module is arranged to correct the historical passing time length by combining personnel distribution information and weather information, influences of personnel distribution and weather on queuing time length are comprehensively considered, the accuracy of time length prediction is improved, the moving speed of the actual channel is collected by the calibration module to calibrate the historical passing time length, the accuracy of time length prediction is improved, the predicted result of the time length is more similar to the actual situation, the determination module is arranged to recommend the optimal queuing channel for the passengers according to the departure time and the predicted passing time length of the target passengers, the passengers are helped to select faster and more effective travel routes, the efficiency of the passengers is improved, the predicted passing time length is updated in time by the updating module, the accuracy and reliability of a prediction system are improved, the passengers are monitored in real time and the queuing time and the abnormal traffic situation are discovered and the safety of the hub is improved.
In particular, the real-time video data is provided by setting the image acquisition unit to acquire the moving process of the identification personnel in real time, the real-time performance and accuracy of the image data are ensured, meanwhile, continuous frame images are completely extracted, omission or deletion of the image data is avoided, the integrity of the image data is improved, the edge profile characteristics of the identification personnel are accurately extracted by setting the image analysis unit, the accuracy and the accuracy of subsequent analysis are ensured, the moving track fitting is carried out through the target points, the accuracy and the reliability of moving track description are improved, the speed correction coefficient is calculated according to the real-time moving track and the actual moving speed by setting the speed calibration unit, the real-time performance of speed calculation is enhanced, the real-time performance of calibration is enhanced, the accuracy of the first passing time length is improved, and the accuracy of the subsequent time length prediction is improved, so that the time length prediction result is more accurate and reliable.
In particular, the edge contour features of the marker are accurately acquired through the edge detection algorithm by the target determination subunit, the accuracy and the reliability of edge detection are improved, the possibility of false detection is reduced, the central point of the edge contour features of the marker is further determined to serve as a target point on the basis of acquiring the edge contour features, the accuracy and the consistency of the determination of the target point are improved, the target points between adjacent frame images are connected by the fitting subunit, the movement track of the marker is accurately acquired, the accuracy and the reliability of track fitting are enhanced, the continuity and the consistency of data are ensured by connecting the target points of the adjacent frame images, and a more reliable basis is provided for subsequent analysis and calculation.
In particular, the speed calculation subunit is arranged to calculate the distance value between two points of the adjacent frame images of the motion track and divided by the time interval between the adjacent frame images, so that the actual motion speed of the marker is obtained, the accuracy and the reliability of speed calculation are ensured, the speed correction coefficient is calculated according to the comparison result of the actual motion speed and the preset motion speed by the time length calibration subunit, the first passing time length is calibrated according to the speed correction coefficient, the accuracy and the reliability of data are improved, more accurate basic data are provided for the subsequent queuing channel length calculation and passenger flow statistics, and the reliability of the system is improved.
Drawings
FIG. 1 is a block diagram of a system for predicting queuing time of a passenger transportation hub based on image recognition according to an embodiment of the present invention;
FIG. 2 is a block diagram of a second structure of a passenger hub queuing time prediction system based on image recognition according to an embodiment of the present invention;
FIG. 3 is a third block diagram of a system for predicting queuing time of a passenger transportation hub based on image recognition according to an embodiment of the present invention;
Fig. 4 is a fourth structural block diagram of a passenger transportation hub queuing time length prediction system based on image recognition according to an embodiment of the present invention.
Detailed Description
In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In addition, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "coupled," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
Referring to fig. 1, an embodiment of the present invention provides a passenger transportation hub queuing time prediction system based on image recognition, which includes:
the data acquisition module 10 is used for acquiring the history passing duration of any queuing channel through the history database;
The data analysis module 20 is connected with the data acquisition module and is used for acquiring personnel distribution information of any queuing channel and weather information outside a passenger transport hub, correcting the historical passing duration through the personnel distribution information and the weather information and acquiring a first passing duration;
the calibration module 30 is connected with the data analysis module and is used for acquiring the actual moving speed of any identifier in any queuing channel, calibrating the first passing duration and acquiring a second passing duration;
a determining module 40, connected to the calibrating module 30, for determining a recommended channel according to the departure time of the target passenger and the second pass duration;
The updating module 50 is connected to the determining module 40, and configured to acquire the target moving speed of the target passenger in the recommended passage in real time, so as to update the second pass duration according to the target moving speed, and acquire a final predicted duration.
In particular, the queuing channels are numbered according to the embodiment of the invention, and are respectively marked as a first queuing channel P1, a second queuing channel P2 … … and an Nth queuing channel Pn,
For any queuing channel Pi, i=1, 2 … … n, calculating the final predicted time length, and setting t Terminal (A) =ji×t2, where Ji is an update parameter of the second passing time length to the final predicted time length, t2 is the second passing time length, and t Terminal (A) is the final predicted time length;
Let t2=ki×t1, where t1 is the first pass duration and Ki is the velocity correction coefficient.
Specifically, the embodiment of the invention provides accurate and traceable historical passing time length data by setting the data acquisition module to establish a historical database, provides reliable data support for subsequent passing time length prediction, combines personnel distribution information and weather information to correct the historical passing time length by setting the data analysis module, comprehensively considers the influence of personnel distribution and weather on queuing time length, so that the correction result of the historical passing time length is accurate, thereby improving the accuracy of time length prediction, and the correction of the historical passing time length by setting the correction module to acquire the moving speed of an actual channel, thereby improving the accuracy of time length prediction, leading the prediction result of the time length to be closer to the actual condition, recommending the optimal queuing channel for the passengers by setting the determination module according to the departure time and the predicted passing time length of target passengers, helping the passengers to select faster and more effective travel routes, improving the travel efficiency of the passengers, timely updating the predicted passing time length by setting the update module, improving the accuracy and reliability of a prediction system, timely finding and timely responding to abnormal conditions, and enhancing the safety of passengers and queuing efficiency of the hub.
Referring to fig. 2, the data analysis module 20 includes a person acquisition unit 21, a weather acquisition unit 22, and a first acquisition unit 23, wherein,
The personnel acquisition unit 21 is configured to acquire personnel images of any queuing channel through an image acquisition device, analyze a plurality of personnel images, determine personnel distribution conditions, and acquire personnel calibration parameters according to the personnel distribution conditions;
the weather collection unit 22 is configured to determine an actual level of weather severity outside the passenger transportation hub, and obtain weather calibration parameters according to the actual level of weather severity;
The first obtaining unit 23 is connected to the personnel collecting unit 21 and the weather collecting unit 22, and is configured to calibrate the historical passing duration according to the personnel calibration parameter and the weather calibration parameter, so as to obtain a first passing duration.
Specifically, for the first passing duration t1, t1=fi×pi×t Calendar with a display is set, where Fi is a person calibration parameter, pi is a weather calibration parameter, and t Calendar with a display is a history passing duration.
Specifically, the weather acquisition unit in the embodiment of the invention calculates the actual severe weather grade according to a plurality of real-time temperature values, a plurality of real-time humidity values and a plurality of real-time wind speed values in weather forecast by acquiring the plurality of real-time temperature values, the plurality of real-time humidity values and the plurality of real-time wind speed values;
Setting a plurality of real-time temperature values as T1, T2 and … … Tn, a plurality of real-time humidity values as h1, h2 and … … hn, and a plurality of real-time wind speed values as v1, v2 and … … vn, wherein the real-time temperature difference value is the same as the real-time temperature difference value T isT= |Ti-T Pre-preparation ｜,i=1、2、……n,T Pre-preparation is a preset temperature value, T Pre-preparation can be set to 20 ℃, and the real-time temperature difference value is obtainedT is different from the preset temperatureT Pre-preparation for comparison, preset temperature differenceT Pre-preparation is 10℃ whenT/T Pre-preparation > 1, the weather level of the temperature is three-level, when 0.5 <T/T Pre-preparation is less than or equal to 1, the weather grade of the temperature is two-level, whenT/T Pre-preparation is less than 0.5, the weather grade of the temperature is first-order, and the real-time humidity difference valueH isH= |hi-h Pre-preparation ｜,i=1、2、……n,h Pre-preparation is a preset humidity value, h Pre-preparation may be set to 45% rh, and the real-time humidity difference value is setH is different from the preset humidityH Pre-preparation comparing, presetting humidity difference valueH Pre-preparation can be set to 15% RH whenh/H Pre-preparation > 1.5, the weather level of humidity is three-level, when 0.8 < +fatherT/∆T Pre-preparation is less than or equal to 1.5, the weather grade of the humidity is second grade, and the father is equal toh/∆H Pre-preparation is less than 0.8, the weather grade of the humidity is the first grade, and the real-time wind speed difference value is the sameV is%V= |vi-v Pre-preparation ｜,i=1、2、……n,v Pre-preparation is a preset wind speed value, v Pre-preparation can be set to 17m/s, and the real-time wind speed difference is fatterV and the difference value of the preset wind speedV Pre-preparation comparing, presetting the wind speed difference valueV Pre-preparation is 7m/s whenv/V Pre-preparation > 1, the weather level of wind speed is three-level, when 0.5 <v/V Pre-preparation is less than or equal to 1, the weather level of the wind speed is two-level, whenv/V Pre-preparation is less than 0.5, the weather grade of the wind speed is first grade, when the weather grade of the temperature or the weather grade of the humidity or the weather grade of the wind speed is first grade, the actual severe weather grade is first grade, when the weather grade of the temperature, the weather grade of the humidity and the weather grade of the wind speed are both third grade, the actual severe weather grade is third grade, and the actual severe weather grade is second grade under the other conditions;
The weather calibration parameter Pi is a quotient of the actual severe weather grade divided by the preset severe weather grade when the actual severe weather grade is larger than the preset severe weather grade, and is 1 when the actual severe weather grade is smaller than or equal to the preset severe weather grade.
Specifically, the personnel distribution condition and the weather condition of the queuing channel are monitored in real time by arranging the personnel acquisition unit and the weather acquisition unit, the prediction accuracy can be improved, the real-time performance of time length prediction is enhanced, and the accuracy and the reliability of a time length prediction result are improved by arranging the first acquisition unit to calibrate the historical passing time length through the personnel calibration parameter and the weather calibration parameter.
In particular, the calibration module comprises an image acquisition unit, an image analysis unit and a speed calibration unit, wherein,
The image acquisition unit is used for acquiring the moving process of the identification personnel through the image acquisition equipment, acquiring a moving video, extracting continuous frame images in the moving video, and acquiring a plurality of moving images;
The image analysis unit is connected with the image acquisition unit and used for extracting edge contour features of the identification personnel in a plurality of moving images, determining target points on the edge contour features of the identification personnel, and performing motion trail fitting on the identification personnel according to the plurality of target points;
The speed calibration unit is connected with the image analysis unit and used for acquiring the actual moving speed of the marker according to the moving track, calculating a speed correction coefficient according to the actual moving speed and calibrating the first passing duration according to the speed correction coefficient.
Specifically, the embodiment of the invention provides real-time video data by setting an image acquisition unit to acquire the moving process of the identification personnel in real time, ensures the real-time performance and accuracy of the image data, simultaneously completely extracts continuous frame images, avoids missing or missing of the image data, improves the integrity of the image data, accurately extracts the edge profile characteristics of the identification personnel by setting the image analysis unit, ensures the accuracy and the accuracy of subsequent analysis, carries out motion track fitting through target points, improves the accuracy and the reliability of motion track description, calculates a speed correction coefficient according to the real-time motion track and the actual moving speed by setting the speed calibration unit, enhances the real-time performance of speed calculation, calculates the speed correction coefficient according to the actual moving speeds of different identification personnel and different queuing channels, enhances the real-time performance of calibration, calibrates the first passing time length, improves the accuracy of the prediction of the subsequent time length, and ensures the more accurate and reliable duration prediction result.
Referring to fig. 3, the person collecting unit 21 includes a person collecting sub-unit 211, a person analyzing sub-unit 212, and a person calculating sub-unit 213, wherein,
The personnel acquisition subunit 211 is configured to acquire a plurality of morphological images in any queuing channel through an image acquisition device;
The personnel analysis subunit 212 is connected with the personnel acquisition subunit and is used for analyzing a plurality of morphological images, extracting edge contour features in a plurality of morphological images and identifying the number of old people, the number of children and the number of passengers carrying large articles in the queuing channel through a plurality of edge contour features;
The personnel calculating subunit 213 is connected to the personnel analyzing subunit, and is configured to calculate and obtain personnel calibration parameters according to the number of the old people, the number of the children and the number of the passengers carrying the large articles in the queuing channel.
Specifically, for the personnel calibration parameter Fi, fi=α× (m1+m2)/m Total (S) +β×m3/m Total (S) is set, where m1 is the number of old people in the current queuing channel, m2 is the number of children in the current queuing channel, m Total (S) is the number of all people in the current queuing channel, α is the weight of the proportion of old people and children in the current queuing channel to the personnel calibration parameter, α may be set to 0.7, m3 is the number of passengers carrying large articles in the current queuing channel, β is the weight of the proportion of passengers carrying large articles in the current queuing channel to the personnel calibration parameter, and β may be set to 0.2;
Specifically, the personnel analysis subunit identifies old people and children by judging the curvature and the height of the edge profile, and identifies passengers carrying large articles by identifying the area of the edge profile of the object in the edge profile.
Specifically, the method for identifying the old and the young through judging the curvature and the height of the edge profile comprises the following steps:
detecting the maximum curvature of the edge profile;
Detecting the longitudinal maximum height of the edge profile;
If the maximum curvature is greater than the preset curvature and the maximum height is greater than the first preset height, the edge contour is an old man, and if the maximum curvature is less than or equal to the preset curvature and the maximum height is less than the second preset height, the edge contour is a child;
The preset curvature is the curvature average value of adults aged 30-40 years old, the first preset height is the largest height value of the height values of the aged over 65 years old, and the second preset height is the largest height value of the height values of children under 10 years old;
Identifying a passenger carrying a large item by identifying the area of the object edge profile in the edge profile comprises:
extracting edge contour features in a plurality of morphological images;
Comparing the edge contour features in the morphological images with edge contour features of preset people, and identifying the edge contour features of the people in the morphological images;
judging the area of the edge contour feature of the object adjacent to the edge contour feature of the morphological image person;
When the area of the area is larger than a preset area, representing the edge profile features of the adjacent morphological image persons corresponding to the area of the area as passengers carrying large articles in the queuing channel, and counting the number of marks as the number of passengers carrying large articles in the queuing channel;
The preset area is 1 square meter.
Specifically, the embodiment of the invention considers the curvature and the height factors of the edge profile simultaneously, so that the identification of old people and children is more accurate, the walking posture and the physical characteristics of people of different age groups can be better described, the results of the subsequent judgment of the number of the old people and the number of the children are accurate, the number of passengers carrying large articles is determined by identifying the edge profile characteristics, the accuracy and the reliability of data are improved, the personnel calibration parameters are calculated by setting personnel calculation subunits, the parameters are calibrated and optimized according to the personnel distribution characteristics of different queuing channels, and the accuracy of the subsequent duration prediction and calculation is improved.
In particular, the image analysis unit comprises a target determination subunit and a fitting subunit, wherein,
The target determination subunit is configured to obtain an edge contour feature of the identified person in the moving image through an edge detection algorithm, and determine a center point of the edge contour feature of the identified person as the target point;
The fitting subunit is connected with the target determination subunit and is used for connecting target points between adjacent frame images to acquire the motion trail of the marker.
Specifically, determining the center point of the edge profile feature of the identified person includes:
extracting inflection point key points in the edge profile features of the identified personnel;
And carrying out mean value calculation on a plurality of inflection point key points, and taking a calculation result as a center point of the edge contour feature of the marker.
Specifically, the embodiment of the invention accurately acquires the edge contour characteristics of the marker through the edge detection algorithm by setting the target determination subunit, improves the accuracy and reliability of edge detection, reduces the possibility of false detection, further determines the central point of the edge contour characteristics of the marker as the target point on the basis of acquiring the edge contour characteristics, improves the accuracy and consistency of the target point determination, accurately acquires the motion track of the marker by setting the fitting subunit to connect the target points between the adjacent frame images, enhances the accuracy and reliability of track fitting, and ensures the continuity and consistency of data by connecting the target points of the adjacent frame images, thereby providing a more reliable basis for subsequent analysis and calculation.
In particular, the speed calibration unit comprises a speed calculation subunit and a duration calibration subunit, wherein,
The speed calculation subunit is used for obtaining the actual movement speed of the marker by calculating the distance value between two adjacent frame images of the movement track divided by the time interval between the adjacent frame images;
The time length calibration subunit is connected with the speed calculation subunit, compares the actual moving speed with a preset moving speed, calculates a speed correction coefficient according to a comparison result, and calibrates the first passing time length according to the speed correction coefficient.
Specifically, the actual movement speed is V Real world , the movement speed V Pre-preparation is preset, the speed correction coefficient Ki is ki= (1- (V Real world -V Pre-preparation ）/V Pre-preparation ) in the case of V Real world ≥V Pre-preparation , and the speed correction coefficient Ki is ki= (1+ (V Pre-preparation -V Real world ）/V Pre-preparation ) in the case of V Real world ＜V Pre-preparation .
Specifically, the embodiment of the invention obtains the actual movement speed of the marker by dividing the distance value between two points of the adjacent frame images of the movement track by the time interval between the adjacent frame images by setting the speed calculation subunit, ensures the accuracy and the reliability of speed calculation, calculates the speed correction coefficient by setting the time length calibration subunit according to the comparison result of the actual movement speed and the preset movement speed, calibrates the first passing time length according to the speed correction coefficient, improves the accuracy and the reliability of data, provides more accurate basic data for the subsequent calculation of the length of a queuing channel and the statistics of the passenger flow, and improves the reliability of a system.
Specifically, the determining module comprises a duration calculating unit and a selecting unit, wherein,
The time length calculation unit is used for calculating preset waiting time length according to the departure time of the target passenger;
The selecting unit is connected with the time length calculating unit and used for sorting the second passing time lengths corresponding to the queuing channels from small to large to obtain a time length sequence, marking the queuing channels corresponding to the second passing time length smaller than the preset waiting time length in the time length sequence, and selecting the first queuing channel as a recommended channel.
Specifically, the preset waiting time can be accurately calculated according to the departure time of the target passenger by the setting time calculation unit, the calculation precision of the waiting time is ensured, the channel selection for the target passenger is more accurate, the channel selection accuracy and reliability are improved by selecting the first channel as the recommended channel by the setting selection unit, the use efficiency of the channel is improved by reasonable channel selection, and the waiting time and queuing time of the passenger are reduced.
In particular, the update module comprises a speed acquisition unit and an update unit, wherein,
The speed acquisition unit is used for acquiring a plurality of images containing target passengers and target objects through the image acquisition equipment, analyzing the images to acquire a plurality of target passenger outlines and a plurality of target object outlines in the images, and calculating the target moving speed of the target passengers according to the distance change values between the target passenger outlines and the target object outlines;
the updating unit is connected with the speed acquisition unit and used for comparing the target moving speed of the target passenger with the preset moving speed, calculating an updating parameter according to a comparison result, and updating the second pass time length according to the updating parameter to obtain the final predicted time length.
Specifically, let the target moving speed be V Order of (A) , the moving speed be V Pre-preparation , and when V Order of (A) ≥V Pre-preparation , the update parameter Ji be ji= (1- (V Order of (A) -V Pre-preparation ）/V Pre-preparation ), and when V Order of (A) ＜V Pre-preparation , the update parameter Ji be ki= (1+ (V Order of (A) -V Real world ）/V Pre-preparation ).
Specifically, the embodiment of the invention accurately acquires the images containing the target passengers and the target objects through the image acquisition equipment by the setting speed acquisition unit, acquires the outlines of the target passengers and the target objects, further calculates the moving speed of the target passengers, ensures the accuracy and the reliability of the speed, and calculates the updating parameters according to the comparison result of the moving speed of the target passengers and the preset moving speed by the setting updating unit, so that the updating parameters are accurately calculated, the accuracy and the reliability of data are improved, and more accurate basic data are provided for the subsequent prediction duration calculation.
Referring to fig. 4, the data acquisition module 10 includes a setup unit 11 and a history duration acquisition unit 12, wherein,
The establishing unit 11 is configured to establish the history database, where the history database includes queuing channel numbers and average passing time lengths of passengers in corresponding respective time periods;
The historical time length obtaining unit 12 is connected with the establishing unit, and is used for averaging the average passing time length of the passengers in each period corresponding to any queuing channel, and taking the average result as the historical passing time length of the queuing channel.
Specifically, the embodiment of the invention is responsible for establishing a historical database by setting the establishment unit, storing and integrating the passenger passing time length data of each queuing channel, ensuring the integrity and traceability of the data, providing a reliable basis for subsequent data analysis, and accurately calculating the historical passing time length of any queuing channel by averaging the average passing time length of passengers in each time length of the channel by setting the historical time length acquisition unit, ensuring the accuracy of the historical time length and providing more accurate basic data for subsequent prediction.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A passenger hub queuing time prediction system based on image recognition, comprising:
the data acquisition module is used for acquiring the history passing duration of any queuing channel through the history database;
the data analysis module is connected with the data acquisition module and used for acquiring personnel distribution information of any queuing channel and weather information outside a passenger transport hub, correcting the historical passing duration through the personnel distribution information and the weather information and acquiring a first passing duration;
The data analysis module comprises a personnel acquisition unit, a weather acquisition unit and a first acquisition unit, wherein,
The personnel acquisition unit is used for acquiring personnel images of any queuing channel through the image acquisition equipment, analyzing a plurality of personnel images, judging personnel distribution conditions, and acquiring personnel calibration parameters according to the personnel distribution conditions;
the weather acquisition unit is used for determining the actual severe weather grade outside the passenger transport hub and acquiring weather calibration parameters according to the actual severe weather grade;
The first acquisition unit is connected with the personnel acquisition unit and the weather acquisition unit and used for calibrating the historical passing duration according to the personnel calibration parameters and the weather calibration parameters to acquire a first passing duration;
The calibration module comprises an image acquisition unit, an image analysis unit and a speed calibration unit, wherein,
The image acquisition unit is used for acquiring the moving process of the identification personnel through the image acquisition equipment, acquiring a moving video, extracting continuous frame images in the moving video, and acquiring a plurality of moving images;
The image analysis unit is connected with the image acquisition unit and used for extracting edge contour features of the identification personnel in a plurality of moving images, determining target points on the edge contour features of the identification personnel, and performing motion trail fitting on the identification personnel according to the plurality of target points;
The speed calibration unit is connected with the image analysis unit and used for acquiring the actual moving speed of the marker according to the moving track, calculating a speed correction coefficient according to the actual moving speed and calibrating the first passing duration according to the speed correction coefficient;
the personnel acquisition unit comprises a personnel acquisition subunit, a personnel analysis subunit and a personnel calculation subunit, wherein,
The personnel acquisition subunit is used for acquiring a plurality of morphological images in any queuing channel through the image acquisition equipment;
The personnel analysis subunit is connected with the personnel acquisition subunit and used for analyzing a plurality of morphological images, extracting edge contour features in a plurality of morphological images and identifying the number of old people, the number of children and the number of passengers carrying large articles in the queuing channel through the plurality of edge contour features;
The personnel computing subunit is connected with the personnel analyzing subunit and is used for computing and obtaining personnel calibration parameters according to the number of the old people, the number of the children and the number of the passengers carrying large articles in the queuing channel;
The calibration module is connected with the data analysis module and used for acquiring the actual moving speed of any marker in any queuing channel, calibrating the first passing duration and acquiring a second passing duration;
The determining module is connected with the calibration module and is used for determining a recommended channel according to the departure time of the target passenger and the second passing time;
And the updating module is connected with the determining module and used for acquiring the target moving speed of the target passenger in the recommended channel in real time so as to update the second pass time length according to the target moving speed and acquire the final predicted time length.
2. The image recognition-based passenger junction queuing time prediction system according to claim 1, wherein the person analysis subunit recognizes old people and children by judging the curvature and the height of the edge profile, and recognizes passengers carrying large articles by recognizing the area of the edge profile of the object in the edge profile.
3. The image recognition-based passenger junction queuing time prediction system of claim 2, wherein the image analysis unit comprises a target determination subunit and a fitting subunit, wherein,
The target determination subunit is configured to obtain an edge contour feature of the identified person in the moving image through an edge detection algorithm, and determine a center point of the edge contour feature of the identified person as the target point;
The fitting subunit is connected with the target determination subunit and is used for connecting target points between adjacent frame images to acquire the motion trail of the marker.
4. The image recognition-based passenger hub queuing time length prediction system of claim 3, wherein the speed calibration unit comprises a speed calculation subunit and a time length calibration subunit, wherein,
The speed calculation subunit is used for obtaining the actual movement speed of the marker by calculating the distance value between two adjacent frame images of the movement track divided by the time interval between the adjacent frame images;
The time length calibration subunit is connected with the speed calculation subunit, compares the actual moving speed with a preset moving speed, calculates a speed correction coefficient according to a comparison result, and calibrates the first passing time length according to the speed correction coefficient.
5. The image recognition-based passenger junction queuing time length prediction system of claim 4, wherein the determination module comprises a time length calculation unit and a selection unit, wherein,
The time length calculation unit is used for calculating preset waiting time length according to the departure time of the target passenger;
The selecting unit is connected with the time length calculating unit and used for sorting the second passing time lengths corresponding to the queuing channels from small to large to obtain a time length sequence, marking the queuing channels corresponding to the second passing time length smaller than the preset waiting time length in the time length sequence, and selecting the first queuing channel as a recommended channel.
6. The image recognition-based passenger hub queuing time prediction system of claim 5, wherein the update module comprises a speed acquisition unit and an update unit, wherein,
The speed acquisition unit is used for acquiring a plurality of images containing target passengers and target objects through the image acquisition equipment, analyzing the images to acquire a plurality of target passenger outlines and a plurality of target object outlines in the images, and calculating the target moving speed of the target passengers according to the distance change values between the target passenger outlines and the target object outlines;
the updating unit is connected with the speed acquisition unit and used for comparing the target moving speed of the target passenger with the preset moving speed, calculating an updating parameter according to a comparison result, and updating the second pass time length according to the updating parameter to obtain the final predicted time length.
7. The image recognition-based passenger junction queuing time prediction system of claim 6, wherein the data acquisition module comprises a setup unit and a historical time acquisition unit, wherein,
The establishing unit is used for establishing the historical database, and the historical database comprises queuing channel numbers and average passing time length of passengers in corresponding each period;
the historical time length obtaining unit is connected with the establishing unit and is used for averaging the average passing time length of passengers in each period corresponding to any queuing channel, and taking the average result as the historical passing time length of the queuing channel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410188480.2A CN117935169B (en) | 2024-02-20 | Passenger transport hub queuing time length prediction system based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410188480.2A CN117935169B (en) | 2024-02-20 | Passenger transport hub queuing time length prediction system based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117935169A CN117935169A (en) | 2024-04-26 |
| CN117935169B true CN117935169B (en) | 2024-07-09 |
Family
ID=
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104899650A (en) * | 2015-05-26 | 2015-09-09 | 成都中科大旗软件有限公司 | Method for predicting tourist flow volume of tourist attraction on basis of multi-source data analysis |
| CN113888857A (en) * | 2021-10-11 | 2022-01-04 | 南京微道科技有限公司 | Public transportation management system, device and method based on Internet of vehicles |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104899650A (en) * | 2015-05-26 | 2015-09-09 | 成都中科大旗软件有限公司 | Method for predicting tourist flow volume of tourist attraction on basis of multi-source data analysis |
| CN113888857A (en) * | 2021-10-11 | 2022-01-04 | 南京微道科技有限公司 | Public transportation management system, device and method based on Internet of vehicles |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110493816B (en) | Real-time prediction method for rail transit subway station passenger flow volume | |
| CN103593974B (en) | A kind of public transport passenger capacity collection method based on locating information | |
| CN104185275A (en) | Indoor positioning method based on WLAN | |
| CN110245377B (en) | Travel scheme recommendation method and recommendation system | |
| CN111260822B (en) | Rail transit vehicle health state analysis method and terminal based on big data | |
| CN109598947B (en) | Vehicle identification method and system | |
| CN109446881B (en) | Heterogeneous data-based highway section traffic state detection method | |
| CN113628432B (en) | Intelligent early warning system for subway people flow current limiting | |
| CN108665178A (en) | A kind of building staircase passenger flow forecast method in the subway station based on AFC | |
| CN107886189B (en) | Method for deducing path travel time based on subway card swiping data | |
| CN116308958A (en) | Carbon emission online detection and early warning system and method based on mobile terminal | |
| CN103793761B (en) | A kind of recognition methods of the abnormal volume of the flow of passengers factor of hinge | |
| CN117935169B (en) | Passenger transport hub queuing time length prediction system based on image recognition | |
| CN110321982A (en) | A kind of rail traffic section volume of the flow of passengers real-time computing technique | |
| Zheng et al. | Improved iterative prediction for multiple stop arrival time using a support vector machine | |
| CN112700473B (en) | Carriage congestion degree judging system based on image recognition | |
| CN117935169A (en) | Passenger transport hub queuing time length prediction system based on image recognition | |
| Nie et al. | Use of multisensor data in reliable short-term travel time forecasting for urban roads: Dempster–Shafer approach | |
| CN110441830A (en) | Zonule Weather prediction system, method and storage medium based on automatic weather station | |
| CN113256014B (en) | Intelligent detection system for 5G communication engineering | |
| CN108665705B (en) | Traffic guidance statistical model checking method and device | |
| CN112884352B (en) | Lightning stroke fault risk assessment method for overhead transmission line | |
| CN111275482B (en) | Machine learning-based real-time dynamic track traffic sorting method | |
| CN111931968B (en) | Method for optimizing public bicycle station layout by using mobile phone GPS data | |
| CN110263993B (en) | Real-time prediction method for urban rail transit passenger traveling time |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |