CN117690094B - Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium - Google Patents

Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN117690094B
CN117690094B CN202410139043.1A CN202410139043A CN117690094B CN 117690094 B CN117690094 B CN 117690094B CN 202410139043 A CN202410139043 A CN 202410139043A CN 117690094 B CN117690094 B CN 117690094B
Authority
CN
China
Prior art keywords
elevator
image
car
people
elevator door
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202410139043.1A
Other languages
Chinese (zh)
Other versions
CN117690094A (en
Inventor
周旭
顾月江
高鹏
于青松
张波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
General Elevator Co ltd
Original Assignee
General Elevator Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by General Elevator Co ltd filed Critical General Elevator Co ltd
Priority to CN202410139043.1A priority Critical patent/CN117690094B/en
Publication of CN117690094A publication Critical patent/CN117690094A/en
Application granted granted Critical
Publication of CN117690094B publication Critical patent/CN117690094B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/761—Proximity, similarity or dissimilarity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/48—Matching video sequences

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Software Systems (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Algebra (AREA)
  • Health & Medical Sciences (AREA)
  • Operations Research (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Indicating And Signalling Devices For Elevators (AREA)

Abstract

The invention relates to the field of image data processing, in particular to an elevator passenger flow volume data statistical method, an apparatus, an electronic device and a storage medium, wherein the statistical method comprises the following steps: acquiring a first image in the car and a second image of the car door opening area; detecting the number of heads in the first image; generating a first person number set; calculating the number of times of opening and closing the elevator door every day, calculating the number of people newly entering the elevator car after opening and closing the elevator door every time, and calculating the first customer flow of each day in the elevator car; the detection model detects the number of heads in the second image; generating a second population set; calculating the number of lost people outside the car; calculating the number of times of opening and closing the elevator door every day, and calculating a second passenger flow rate outside the elevator car every day; and obtaining the sum of the first customer flow and the second customer flow, namely the total customer flow of each day. The invention has the effect of taking the passenger flow which does not enter the car into the statistics of the passenger flow, and improving the statistics accuracy of the passenger flow.

Description

Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium
Technical Field
The present invention relates to the field of image data processing, and in particular, to an elevator passenger flow volume data statistics method, an apparatus, an electronic device, and a storage medium.
Background
The accurate statistics of the elevator passenger flow volume is a key for realizing the accurate prediction of the elevator passenger flow volume, and has important significance for the identification of the elevator traffic flow mode and the optimization of the elevator group control dispatching method.
The traditional passenger flow volume statistical method comprises manual statistics, infrared induction, gravity induction and the like, wherein the manual statistics method has high requirement on counting personnel and is easy to cause counting errors due to personnel fatigue, the infrared induction is easy to be interfered by factors such as ambient temperature and the like, the statistical error rate is high, the gravity induction installation requirement is high, the cost is high, the stability is poor, and the uncertainty is also large.
With the continuous development of computer technology and network technology, a video-based passenger flow volume statistical method is becoming the mainstream, and according to images acquired by a camera, image data are analyzed and processed by using an algorithm, so that passenger flow volume is automatically counted. However, the current video-based passenger flow volume statistics method ignores passenger flow volume which has long waiting time and is not entered into the car in advance, so that the accuracy of passenger flow volume statistics data is reduced.
Disclosure of Invention
In order to enable passenger flow which does not enter the inside of a car to be included in statistics of the passenger flow, and improve the statistics accuracy of the passenger flow, the invention provides an elevator passenger flow data statistics method, an elevator passenger flow data statistics device, electronic equipment and a storage medium.
The invention provides an elevator passenger flow volume data statistical method, which adopts the following technical scheme:
a statistical method for elevator passenger flow data is characterized in that: comprises the following steps of
Acquiring a first image in a car and a second image of the car door opening area;
establishing a detection model, and detecting the number of heads in the first image;
generating based on the detection result of the detection modelFirst person count set of moments->，/>Wherein->Representation->Middle->Labeling information of the individuals;
acquiring elevator door opening and closing information, calculating the opening and closing times of the elevator door every day, calculating the number of people newly entering the elevator car after opening and closing the elevator door every time, and calculating the first customer flow of each day in the elevator car by summing the number of people newly entering the elevator car after opening and closing the elevator door every time;
the detection model detects the number of heads in the second image;
generating based on the detection result of the detection modelSecond population set of time of day->，/>Wherein->Representation->Middle->Labeling information of the individuals;
according to the elevator door opening and closing information, obtaining the maximum value of the number of people outside the lift car in the time period of opening and closing the elevator doorAnd minimum->By the formula: />Calculating the time from the closing of the elevator door to the re-opening of the elevator door, whereinThe number of lost people outside the lift car>；
Acquiring elevator door opening and closing information, calculating the opening and closing times of the elevator door every day, and counting the number of lost people outside the elevator car after each elevator door is closedSumming, calculating a second passenger flow per day outside the car;
and obtaining the sum of the first customer flow and the second customer flow, namely the total customer flow of each day.
In a specific embodiment, the method for establishing the detection model includes the following steps:
acquiring annotation information in the first image and the second image through an image annotation tool, and establishing a detection data set to store the annotation information, the first image and the second image;
dividing the data in the detection data set, and performing model training.
In a specific embodiment, the annotation information includes a relative abscissa starting pointRelative ordinate starting point->Relative width->Relative height->Category->，
Wherein the category is uniformly marked as 0.
In a specific embodiment, the method for calculating the number of people newly entering the car after each opening and closing of the elevator door comprises the following steps:
extracting all frame images in the stage from the opening to the closing of the elevator door in a first image, extracting the labeling information of each person in the first image, and writing the labeling information into the first person number setIn (a) and (b);
acquiring a first center point of a human headWherein->，/>；
Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the personal head rectangular frame in the image;
acquiring the center point of the elevator door，
Calculating a euclidean distance between the first center point and the elevator door center point,
if the Euclidean distance between the first central point and the elevator door central point is gradually reduced, indicating that passengers tend to leave from the elevator car; otherwise, indicating that the passenger does not have a tendency to exit from the car;
if the first center point has a tendency to leave from the car and disappears from one frame of the first image and no more appears in the subsequent images of the first time threshold, indicating that a person leaves the elevator, and the number of people leaving the elevatorFor the first number of center points to disappear and from the first set of people +.>Removing the corresponding labeling information;
if a frame of the first image does not belong to the current first person number setIs continuously present in the image before the elevator door is completely closed, indicating that someone enters the elevator, the number of people entering the elevator +.>Is the number of people newly entering the car.
In a specific embodiment, the method for determining loss of a person outside the car comprises the steps of:
extracting all frame images from a second time threshold value after the elevator door is closed to a stage of reopening in the second image, extracting the labeling information of each person in the second image, and writing the labeling information into the second people number setIn (a) and (b);
acquiring a second center point of the human headWherein->，/>；
Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the personal head rectangular frame in the image;
if the second center point disappears from one frame of the second image and no more appears in the subsequent images, indicating that people are lost and from the second people number setAnd removing the corresponding labeling information.
In one particular embodiment, the first customer flow, the second customer flow, and the total customer flow statistics are zeroed out at zero points per day.
The invention also provides an elevator passenger flow volume data statistics device, which adopts the following technical scheme:
an elevator passenger flow volume data statistics device comprises
The first image acquisition module acquires images in the car;
a second image acquisition module for acquiring an image outside the car;
one or more processors; a memory;
and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions that, when executed by the processor, cause the elevator traffic data statistics device to perform the elevator traffic data statistics method described above.
In a specific embodiment, the first image acquisition module is disposed at the top of the car interior.
The invention also provides electronic equipment, which adopts the following technical scheme:
an electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the elevator traffic data statistics method described above.
The invention also provides a non-transitory computer readable storage medium storing computer instructions, adopting the following technical scheme:
a non-transitory computer readable storage medium storing computer instructions, characterized by: the computer instructions are for causing the computer to perform the elevator traffic data statistics method described above.
In summary, the present invention includes at least one of the following beneficial technical effects:
1. the passenger flow in the elevator car and the number of lost passengers outside the elevator car are counted respectively by utilizing a machine vision technology, and the total passenger flow in each day is finally obtained by summation.
2. And the zero point of each day clears the passenger flow volume statistics, so that the passenger flow volume of each day can be conveniently and quickly obtained, and the passenger flow volume obtaining efficiency of each day is improved.
3. Through the Euclidean distance calculation between the passenger and the elevator door, the trend that the passenger leaves the elevator is judged, when the passenger has the trend that the passenger leaves the elevator, and the image of the passenger finally disappears in the first image, the passenger leaves the elevator, the accuracy of the passenger state judgment is improved, and the accuracy of the total passenger flow statistics is improved.
4. And the second image outside the elevator is acquired after the elevator door is closed for a certain time, so that passengers leaving from the elevator are prevented from appearing in the second image as much as possible, repeated statistics of the passengers is avoided as much as possible, and the accuracy of total passenger flow statistics is further improved.
Drawings
Fig. 1 is a flow chart of an elevator traffic data statistics method.
FIG. 2 is a schematic diagram showing the location of a first image acquisition module.
FIG. 3 is a schematic diagram illustrating an image acquisition range of the second image acquisition module.
Reference numerals illustrate: 1. a car; 2. a first image acquisition module; 3. an elevator door; 4. and a second image acquisition module.
Detailed Description
The present invention will be described in further detail with reference to fig. 1.
The elevator passenger flow volume data statistical method provided by the embodiment of the invention can be applied to a server or a terminal. The server may be a physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, a content delivery network (Content Delivery Network, CDN), basic cloud computing services such as big data and an artificial intelligent platform. The Terminal may be a Mobile phone, a smart phone, a notebook computer, a digital broadcast receiver, a Personal Digital Assistant (PDA), a tablet personal computer (PAD), a User Equipment (UE), a handheld device, a vehicle-mounted device, a wearable device, a computing device, or other processing device connected to a wireless modem, a Mobile Station (MS), a Mobile Terminal (Mobile Terminal), or the like, which is not limited herein.
Referring to fig. 1, the elevator traffic data statistics method includes the steps of:
s100, constructing a detection data set and a detection model.
A110, collecting at least 30 (including 30) images of scenes inside and outside the elevator car, wherein the number of images of each scene is more than or equal to 100 frames, and each frame is an image at different moments.
Creating a marking box (rectangular frame) in the image through an image marking tool LabelImg, marking the human head, wherein marking information is thatWherein->Represents the relative abscissa starting point, +.>Representing the start point of the relative ordinate,represents the relative width>Relative height, ++>Representing a category. Since only the head is labeled, the category is only "head" and therefore will be +.>Collectively labeled 0. Thereby establishing a detection model.
After the image is marked, a marking file with the format of xml is generated, marking information is stored, the original image is stored, and a detection data set is generated.
A120, dividing the detection data set into a training set for training and a verification set for model parameter adjustment, wherein the ratio of the training set to the verification set is 7:3.
A130, model training is carried out on the training set through a YOLOv3 method, and when the iteration times of the detection model reach a preset value or the accuracy of the detection model on the verification set reaches a preset threshold value, model training of the detection model is stopped, and the detection model with the highest accuracy on the verification set is selected.
S200, counting the first customer flow in each elevator carAnd the number of people with lost passenger flow outside the car, namely second passenger flow +.>。
B210, acquiring a first image in the car, extracting all frame images in the time period from the opening to the closing of the elevator door in the first image, marking the head of a person in each frame image, and generating according to the marking conditionFirst person number set of moments，/>Wherein->Representation->Middle->Labeling information of the person, i.e. +.>Rectangular frame of the individual's head.
B220, calculating the first center point of each headWherein->，；/>Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the individual's head rectangular box in the image.
Obtaining a pre-entered coordinate range of an elevator doorWherein->For the starting point of the abscissa of the elevator door in the image, < >>For the starting point of the ordinate of the elevator door in the image, -, for the elevator door>For the relative width of the elevator door in the image, < >>Is the relative height of the elevator door in the image.
Based on the centre of the elevator doorAnd the following formula:
calculating the Euclidean distance between the first center point and the center of the elevator door。
Recording the Euclidean distance between each first center point and the center of the elevator door in each frame of image，/>，For the total number of frames of all images in each elevator door open to closed time period. Calculating difference value of European distance between first center point and elevator door center in front and back two frames of images +.>If->Negative and in 60 consecutive frames or 2 consecutive seconds +.>Not decrease, i.e. ->，/>Indicating that the distance between the passenger and the elevator door is gradually reduced, and that the passenger has a tendency to leave the elevator; otherwise, it is indicated that the passenger does not have a tendency to leave the elevator.
For ease of understanding, the following is exemplified by way of frame count: assume that the Euclidean distance between the first center point of the first head and the center of the elevator door in the first frame isPixel, european distance between the first center point and the center of the elevator door in the second frame +.>Pixel, european distance between first center point and elevator door center in third frameA pixel. Euclidean distance between the first center point and the center of the elevator door in the second frame and the first frameThe difference is:the difference between the Euclidean distance between the first center point and the center of the elevator door in the third frame and the second frame is: />Pixels, and->If the following 58 frames of images are, the difference value of the Euclidean distance between the first center point and the center of the elevator door in every 2 frames of images is +.>Are all negative numbers and +.>The distance between the passenger and the elevator door is considered to be gradually decreasing and there is a tendency for the passenger to leave the elevator.
It can be understood that the trend of passengers leaving the elevator is judged by calculating the difference value of the Euclidean distance between the first center point and the center of the elevator door in every 2 frames of images within 2 seconds in the same way as the time counting mode.
Further, if the identified head of the person finally disappears from the current frame image and no more appears in the subsequent 30-60 frames or 1-2 seconds, indicating that the person leaves the elevator, recording the number of people leaving the elevator. At this time, from the first person count set +.>Corresponding annotation information is deleted.
It is considered that there are some passengers who initially fail to recognize the reason for getting off the elevator or the like, so that during the closing of the elevator door, the passenger suddenly rushes out of the elevator and causes the sensor in the elevator door to fail to recognize the passenger passing without stopping the door closing action. In this case, the number of images of 30 to 60 frames or 2 seconds is not necessarily satisfied from the disappearance of the first center point of the passenger from the image to the end of the extracted image in the images extracted from the first image, and at this time, only the last frame of the extracted image is calculated. That is, if the first center point is identified as disappearing from the image of 30-60 frames or 1-2 seconds before the elevator door closes, and until the last frame of image that the elevator door closes, the image no longer appears, indicating that the passenger left the elevator.
Since the passenger has a situation of adjusting the position in the elevator, when the passenger moves in the elevator, the passenger approaches to the elevator door, but the passenger does not want to go out of the elevator. It can also be understood that when the number of passengers in the elevator is large, there are cases where passengers at the elevator entrance walk out of the elevator actively for other passengers to conveniently exit the elevator, and after passengers needing to get off the elevator walk out, the passengers can get back into the elevator again, and at this time, the number of passengers cannot be counted as the newly increased passenger flow in the elevator car. Therefore, the Euclidean distance between the first center point and the center of the elevator door is calculated, the trend of passengers leaving the elevator is primarily deduced according to the Euclidean distance change condition, and then the judgment of passengers disappearing from the first image is combined to be used as further confirmation to judge that the passengers leave the elevator, so that the accuracy of detection is improved, and the accuracy of detecting the passenger flow is improved.
When it is detected in a certain frame of image that the current first person number set does not belong toIs present in the subsequent 150 frames or 5 seconds of images, indicating that someone is entering the elevator, writing the labeling information into the first person count set>In the elevator, the number of people entering the elevator is recorded>。
It is easy to understand that passengers also exist to quickly wash into the elevator during the process of closing the door of the elevator, or to quickly and repeatedly press the door closing button of the elevator after entering the elevator to quickly close the door of the elevatorA case of closing or the like, so that there is a case where the number of images of the passenger is less than 150 frames or 5 seconds, at this time, only the last frame of the extracted image is calculated. That is, only the image from which the elevator door is opened to the stage of closing the elevator door is extracted from the first image, and in the extracted image, it is detected that the image does not belong to the current first person number set in the first 149 frames or 5 seconds of the last frame image in which the elevator door is closedAnd the label remains in the last frame of image until the elevator door is closed, indicating that someone is entering the elevator.
Once the elevator opens and closes the door, the number of people newly increased in the car；
B230, acquiring door opening and closing information of the elevator from the main control board of the elevator, and counting door opening and closing times of the elevator in one dayBy the formula:
calculating a first customer flow in a car during a dayWherein->For elevator +.>The number of people newly added in the car with the door opened and closed again.
The first customer flow is cleared at zero per day, i.e. at zero per day,facilitating rapid statistics of the first customer traffic per day.
C210, obtaining a second image outside the car, marking the head of the person in the second image, and generating according to the marking conditionSecond population set of time of day->，/>Wherein->Representation->Middle->Labeling information of the person, i.e. +.>Rectangular frame of the individual's head.
C220 obtaining a second center point of each headWherein->，；/>Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the individual's head rectangular box in the image.
Extracting each time the elevator door is closed in the second imageAll frame images within the second to reopen time period, or +.>All frame images within the time period from the frame image to the reopening of the frame image are identified, the head of each frame image is identified, and the labeling information is written into the second head of people set +.>Is a kind of medium. Wherein (1)>In this embodiment, <' > a->Preferably 5;in this embodiment, 150 is preferable. If the marking information in the second image disappears from the image of the current frame and no longer appears in the image of the subsequent 5 seconds or in the image of 150 frames, indicating that someone leaves and no longer waits for an elevator, then the person is taken from the second people number set->Corresponding annotation information is deleted.
It will be readily appreciated that when the number of frames is used as the timing target, the images of all frames in the time period from when the elevator door is closed to when the elevator door is opened again can be acquired first, and then the images of all frames after the elevator door is closed can be removedAnd acquiring an image in the form of a frame image.
Since after the elevator door is open there is a passenger leaving the elevator, this part of the passenger has been counted as a first customer flow after entering the elevatorHowever, statistics should not be repeated. By waiting forSecond or +.>The form of the frame image is such that these passengers disappear from the second image and are not recognized. When the waiting time is too short, i.e. +.>Or->When the passengers are difficult to completely disappear from the second image, the passengers going out of the elevator are repeatedly counted; when the waiting time is too long, i.eOr->When the passengers disappear from the second image completely, a longer blank time exists in the period from the time when the second image is acquired to the time when the people head is marked, and passengers with statistical losses are easy to miss.
Calculate a second population setMarking information quantity, obtaining maximum value +.>And minimum valueBy the formula: />Calculating the number of lost people outside the lift car in the stage of closing the lift door to re-opening>。
C230 slave elevator masterThe control board acquires door opening and closing information of the elevator, and counts door opening and closing times of the elevator in one dayBy the formula:
wherein,for elevator +.>The number of people lost outside the elevator car when the door is opened for the second time,
and calculating the number of lost people outside the car in one day, namely the second passenger flow.
The second passenger flow volume is cleared at the zero point of each day, i.e. at the zero point of each day,the second passenger flow volume per day is convenient to rapidly count.
It will be appreciated that the statistics of the first customer flow and the statistics of the second customer flow are performed simultaneously during the operation of the elevator, and that B210-B230 and C210-C230 are only used to distinguish the statistics steps of the first customer flow and the second customer flow, and are not limited in order.
Further, compared with the method that the images of all frames in the first image and the second image are marked and the passenger flow volume is detected and calculated, the method omits repeated statistics and calculation in the process that the number of people in and out of an elevator car is unchanged in the ascending and descending processes of the elevator and effectively reduces the calculated amount by extracting part of the images in the first image and the second image to carry out passenger marking and passenger flow volume detection and calculation.
S300, through the formula:the total passenger flow per day of the elevator is calculated.
The embodiment of the invention also provides an elevator passenger flow volume data statistics device, which comprises
A first image acquisition module 2 for acquiring an image of the inside of the car;
a second image acquisition module 4 for acquiring an image outside the car;
one or more processors; a memory;
and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions that, when executed by the processor, cause the elevator traffic data statistics device to perform the elevator traffic data statistics method described above.
Referring to fig. 2 and 3, the first image acquisition module 2 and the second image acquisition module 4 may be devices capable of acquiring images, such as a camera and a high-definition camera, the first image acquisition module 2 may be disposed at a top end of a surface of the elevator car 1 facing the elevator door 3, specifically, opposite ends of an inner wall of the elevator car 1, the second image acquisition module 4 may be disposed above the elevator door 3, and the image acquisition range may cover a semicircular area (a dotted line area in the figure) with a radius of 3 meters, at least using a center of a projection of the elevator door on a horizontal plane as a center of a circle.
The embodiment of the invention also provides electronic equipment, which comprises: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the elevator traffic data statistics method described above.
The embodiment of the invention also provides a non-transitory computer readable storage medium storing computer instructions for causing a computer to execute the elevator passenger flow volume data statistics method.
The above embodiments are not intended to limit the scope of the present invention, so: all equivalent changes in structure, shape and principle of the invention should be covered in the scope of protection of the invention.

Claims (10)

1. A statistical method for elevator passenger flow data is characterized in that: comprises the following steps of
Acquiring a first image in a car and a second image of the car door opening area;
establishing a detection model, and detecting the number of heads in the first image;
generating based on the detection result of the detection modelFirst person count set of moments->，/>Wherein, the method comprises the steps of, wherein,representation->Middle->Labeling information of the individuals;
acquiring elevator door opening and closing information, calculating the opening and closing times of the elevator door every day, calculating the number of people newly entering the elevator car after opening and closing the elevator door every time, and calculating the first customer flow of each day in the elevator car by summing the number of people newly entering the elevator car after opening and closing the elevator door every time;
the detection model detects the number of heads in the second image;
generating based on the detection result of the detection modelSecond population set of time of day->，/>Wherein, the method comprises the steps of, wherein,representation->Middle->Labeling information of the individuals;
according to the elevator door opening and closing information, obtaining the maximum value of the number of people outside the lift car in the time period of opening and closing the elevator doorAnd minimum->By the formula: />Calculating the number of lost people out of the lift car from the stage of closing the lift door to re-opening>；
Acquiring elevator door opening and closing information, calculating the opening and closing times of the elevator door every day, and counting the number of lost people outside the elevator car after each elevator door is closedSumming, calculating a second passenger flow per day outside the car;
and obtaining the sum of the first customer flow and the second customer flow, namely the total customer flow of each day.
2. The elevator traffic data statistics method as recited in claim 1, wherein: the method for establishing the detection model comprises the following steps:
acquiring annotation information in the first image and the second image through an image annotation tool, and establishing a detection data set to store the annotation information, the first image and the second image;
dividing the data in the detection data set, and performing model training.
3. The elevator traffic data statistics method as recited in claim 2, wherein: the labeling information comprises relative abscissa starting pointsRelative ordinate starting point->Relative width->Relative height->Category->，
Wherein the category is uniformly marked as 0.
4. A method of statistical elevator traffic data according to claim 3, characterized in that: the method for calculating the number of people newly entering the car after each time the elevator door is opened and closed comprises the following steps:
extracting all frame images in the stage from the opening to the closing of the elevator door in a first image, extracting the labeling information of each person in the first image, and writing the labeling information into the first person number setIn (a) and (b);
acquiring a first center point of a human headWherein->，/>；
Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the personal head rectangular frame in the image;
acquiring the center point of the elevator door，
Calculating a euclidean distance between the first center point and the elevator door center point,
if the Euclidean distance between the first central point and the elevator door central point is gradually reduced, indicating that passengers tend to leave from the elevator car; otherwise, indicating that the passenger does not have a tendency to exit from the car;
if the first center point has a tendency to leave from the car and disappears from one frame of the first image and no more appears in the subsequent images of the first time threshold, indicating that a person leaves the elevator, and the number of people leaving the elevatorFor the first number of center points to disappear and from the first set of people +.>Removing the corresponding labeling information;
if a frame of the first image does not belong to the current first person number setIs continuously present in the image before the elevator door is completely closed, indicating that someone enters the elevator, the number of people entering the elevator +.>Is the number of people newly entering the car.
5. A method of statistical elevator traffic data according to claim 3, characterized in that: the method for judging the loss of the people outside the lift car comprises the following steps:
extracting all frame images from a second time threshold value after the elevator door is closed to a stage of reopening in the second image, extracting the labeling information of each person in the second image, and writing the labeling information into the second people number setIn (a) and (b);
acquiring a second center point of the human headWherein->，/>；
Is->Relative width of personal head rectangle frame in image, < ->Is->The relative height of the personal head rectangular frame in the image;
if the second center point disappears from one frame of the second image and no more appears in the subsequent images, indicating that people are lost and from the second people number setAnd removing the corresponding labeling information.
6. The elevator traffic data statistics method as recited in claim 1, wherein: zero points per day zero point the first customer flow, the second customer flow, and the total customer flow statistics.
7. An elevator passenger flow volume data statistics device which is characterized in that: comprising
The first image acquisition module acquires images in the car;
a second image acquisition module for acquiring an image outside the car;
one or more processors; a memory;
and one or more computer programs, wherein the one or more computer programs are stored in the memory, the one or more computer programs comprising instructions that, when executed by the processor, cause the elevator traffic data statistics apparatus to perform the elevator traffic data statistics method of any of claims 1-6.
8. The elevator traffic data statistics device of claim 7, wherein: the first image acquisition module is arranged at the top in the car.
9. An electronic device, characterized in that: comprising the following steps: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the elevator traffic data statistics method of any of claims 1-6.
10. A non-transitory computer readable storage medium storing computer instructions, characterized by: computer instructions for causing a computer to perform the elevator traffic data statistics method according to any one of claims 1-6.
CN202410139043.1A 2024-02-01 2024-02-01 Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium Active CN117690094B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410139043.1A CN117690094B (en) 2024-02-01 2024-02-01 Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410139043.1A CN117690094B (en) 2024-02-01 2024-02-01 Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN117690094A CN117690094A (en) 2024-03-12
CN117690094B true CN117690094B (en) 2024-04-09

Family

ID=90133723

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410139043.1A Active CN117690094B (en) 2024-02-01 2024-02-01 Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN117690094B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205930710U (en) * 2016-07-25 2017-02-08 南京信息工程大学 Subway stream of people dredges device
CN109598257A (en) * 2018-12-28 2019-04-09 福建工程学院 A kind of bootstrap technique and system of equilibrium subway carriage passenger capacity
CN110334569A (en) * 2019-03-30 2019-10-15 深圳市晓舟科技有限公司 The volume of the flow of passengers passes in and out recognition methods, device, equipment and storage medium
CN110443100A (en) * 2018-05-04 2019-11-12 郑州宇通客车股份有限公司 A kind of passenger flow statistical method, passenger flow statistical system and school bus
CN115303901A (en) * 2022-08-05 2022-11-08 北京航空航天大学 Elevator traffic flow identification method based on computer vision

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205930710U (en) * 2016-07-25 2017-02-08 南京信息工程大学 Subway stream of people dredges device
CN110443100A (en) * 2018-05-04 2019-11-12 郑州宇通客车股份有限公司 A kind of passenger flow statistical method, passenger flow statistical system and school bus
CN109598257A (en) * 2018-12-28 2019-04-09 福建工程学院 A kind of bootstrap technique and system of equilibrium subway carriage passenger capacity
CN110334569A (en) * 2019-03-30 2019-10-15 深圳市晓舟科技有限公司 The volume of the flow of passengers passes in and out recognition methods, device, equipment and storage medium
CN115303901A (en) * 2022-08-05 2022-11-08 北京航空航天大学 Elevator traffic flow identification method based on computer vision

Also Published As

Publication number Publication date
CN117690094A (en) 2024-03-12

Similar Documents

Publication Publication Date Title
CN108229509B (en) Method and device for identifying object class and electronic equipment
WO2017156772A1 (en) Method of computing passenger crowdedness and system applying same
CN106241584A (en) A kind of intelligent video monitoring system based on staircase safety and method
CN105844229B (en) A kind of calculation method and its system of passenger&#39;s crowding
CN107886761A (en) A kind of parking lot monitoring method based on unmanned plane
CN105303191A (en) Method and apparatus for counting pedestrians in foresight monitoring scene
CN112287827A (en) Complex environment pedestrian mask wearing detection method and system based on intelligent lamp pole
CN108198159A (en) A kind of image processing method, mobile terminal and computer readable storage medium
CN111209781B (en) Method and device for counting indoor people
CN107833328B (en) Access control verification method and device based on face recognition and computing equipment
CN115303901B (en) Elevator traffic flow identification method based on computer vision
CN110481606A (en) The metro passenger flow of view-based access control model identification technology guides system and method
KR102305038B1 (en) Server for Tracking Missing Child Tracking and Method for Tracking Moving Path of Missing Child based on Face Recognition based on Deep-Learning Therein
WO2022078134A1 (en) People traffic analysis method and system, electronic device, and readable storage medium
CN203165067U (en) Entrance population counting device oriented to open scene
CN117690094B (en) Elevator passenger flow volume data statistics method and device, electronic equipment and storage medium
CN103198327A (en) Open type scene-oriented entrance and exit population counting method and device
CN110809137A (en) Campus safety trampling prevention monitoring system and method
CN113869115A (en) Method and system for processing face image
CN109583397A (en) A kind of elevator examines the implementation method for artificial intelligent appraisement system of taking an examination
CN109345427A (en) The classroom video point of a kind of combination recognition of face and pedestrian&#39;s identification technology is to method
WO2023202346A1 (en) Smoking behavior detection method and apparatus, and related device
CN112686180A (en) Method for calculating number of personnel in closed space
CN105118073A (en) Human body head target identification method based on Xtion camera
TWI777689B (en) Method of object identification and temperature measurement

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