Bus passenger flow information acquisition method, device and equipment and storage medium

Abstract

The invention discloses a public transport passenger flow information acquisition method and device, equipment and a storage medium. The method comprises the following steps: acquiring an acquisition image at a door of a bus when the bus is parked at at least one station; for each station, identifying a passenger in at least one acquired image acquired by the station, obtaining image information of at least one passenger and a riding direction of each passenger, and determining the station as a riding station corresponding to each piece of image information; and matching the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, and determining the riding station of each passenger, the riding station including a get-on station and a get-off station. According to the embodiment of the invention, the accuracy and efficiency of obtaining the bus passenger flow information can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban public transportation, and in particular to a method, device, equipment and storage medium for acquiring public transportation passenger flow information. Background Art

[0002] With the development of urban public transportation, in order to optimize the level of public transportation operation and improve the quality of public transportation services, this can be achieved by obtaining public transportation passenger flow information.

[0003] At present, bus passenger flow information is mainly obtained by scanning the Wi-Fi and Bluetooth addresses of smart terminals carried by passengers.

[0004] However, the method of obtaining bus passenger flow information in the above manner has low collection accuracy, is prone to data distortion, and has insufficient collection efficiency. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for acquiring public transport passenger flow information. The embodiments of the present invention can improve the accuracy and efficiency of acquiring public transport passenger flow information.

[0006] In a first aspect, an embodiment of the present invention provides a method for obtaining public transport passenger flow information, the method comprising:

[0007] Acquire a captured image of a door of the public transport vehicle when the public transport vehicle stops at at least one station;

[0008] For each station, identify passengers in at least one captured image acquired at the station, obtain image information of at least one passenger and the boarding direction of each passenger, and determine the station as the boarding station corresponding to each image information;

[0009] According to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, the image information of each passenger is matched to determine the riding station of each passenger. The riding station includes the boarding station and the alighting station.

[0010] In a second aspect, an embodiment of the present invention further provides a device for acquiring public transportation passenger flow information, the device comprising:

[0011] An image acquisition module, used to acquire an image of a door of a public transport vehicle when the public transport vehicle stops at at least one station;

[0012] A passenger identification module is used to identify passengers in at least one collected image acquired at each station, obtain image information of at least one passenger and the boarding direction of each passenger, and determine the station as the boarding station corresponding to each image information;

[0013] The image information matching module is used to match the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, and determine the riding station of each passenger. The riding station includes the boarding station and the alighting station.

[0014] In a third aspect, an embodiment of the present invention further provides a bus passenger flow information acquisition device, the bus passenger flow information acquisition device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to at least one processor; wherein,

[0017] The memory stores a computer program that can be executed by at least one processor. The computer program is executed by at least one processor so that the at least one processor can execute the method for obtaining bus passenger flow information according to any embodiment of the present invention.

[0018] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for obtaining bus passenger flow information according to any embodiment of the present invention when executed.

[0019] The technical solution of the embodiment of the present invention can ensure that the position of the captured image is captured at the station by acquiring the captured image at the door of the bus when the bus stops at the station, and can objectively and real-time record the passengers' boarding and alighting behaviors; the way of collecting passenger information through images can be recorded according to actual behaviors, reducing the dependence of collected information on passenger operating behaviors and avoiding errors or omissions caused by dependence on passengers; by identifying the passenger information in the captured image and combining the passenger's riding direction, the specific behavior of the passenger at each station can be determined, and the analysis of the riding direction helps to distinguish the passenger's behavior at a certain station, ensuring the accurate matching of station information and passenger behavior; by combining the passenger's riding direction, riding direction and station data, the passenger's boarding station and alighting station can be accurately matched, thereby improving the integrity of the passenger's records, reducing the possibility of duplicate records or erroneous matching, ensuring the consistency and accuracy of the data, and providing more accurate data support for subsequent bus passenger flow analysis and operation optimization.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of a method for acquiring public transport passenger flow information provided according to Embodiment 1 of the present invention;

[0023] Figure 2 is a structural diagram of a device for acquiring public transportation passenger flow information provided according to Embodiment 2 of the present invention;

[0024] Figure 3 It is a structural schematic diagram of a bus passenger flow information acquisition device provided according to Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Embodiment 1

[0028] Figure 1 This is a flow chart of a method for obtaining public transport passenger flow information according to Embodiment 1 of the present invention. The embodiment of the present invention is applicable to the case of obtaining public transport passenger flow information. The method can be executed by a public transport passenger flow information obtaining device, which can be implemented in the form of hardware and / or software.

[0029] See also Figure 1 The method for obtaining bus passenger flow information shown includes:

[0030] S101. Acquire a captured image of a door of a public transportation vehicle when the public transportation vehicle stops at at least one station.

[0031] Among them, the captured image may refer to using a camera to capture images of the door when the bus stops at a bus stop to obtain passenger information. The purpose of the captured image is to provide initial data on the passenger situation in the area near the door when the bus stops at the station, and to provide basic information for subsequent passenger flow analysis. Cameras can be installed at each door of the bus. When the bus stops at a certain station, when passengers get on or off the bus at each door, the camera at the door will automatically capture the door and obtain the captured image at the door.

[0032] S102: For each station, identify passengers in at least one captured image acquired at the station, obtain image information of at least one passenger and the boarding direction of each passenger, and determine the station as the boarding station corresponding to each image information.

[0033] The image information may refer to data containing an image of a passenger. The image region containing the passenger may be used as the image information, or feature extraction may be performed on the image region containing the passenger to obtain a feature vector, which is used as the image information of the passenger.

[0034] The riding direction may refer to a boarding direction or an alighting direction. The riding station corresponding to the image information may refer to a boarding station and / or an alighting station of the passenger represented by the image information.

[0035] A large number of collected images can be obtained at each station, and at least one passenger can be identified in all collected images. Since a passenger can appear in multiple collected images, in order to reduce the amount of data processing and the complexity of data processing, the passengers can be counted in a hierarchical manner. The collected images corresponding to each station are processed in the same way. For each station, a passenger is identified in at least one collected image of a station, and the image information of the passenger can be extracted from the collected image in which the passenger is identified, and the passenger's boarding station can be determined based on the station where the bus stops when the collected image is taken. And the passenger's movement direction can be determined by identifying multiple continuous collected images of the passenger and the acquisition time of these multiple continuous collected images, and the passenger's boarding direction can be determined according to the passenger's movement direction and the relative position between the passenger and the bus. In this way, the image information, riding direction and boarding station of the same passenger can be obtained, and a corresponding relationship can be established between the image information, riding direction and boarding station.

[0036] The following methods may be used, and the following examples are only used as examples and are not intended to be limiting. In a specific example, there are multiple passengers in the acquired captured image, and the captured image is first segmented, and the segmented captured image is divided into multiple images, and then the feature vector of each passenger image is converted, and the image information may refer to the feature vectors corresponding to these images.

[0037] Among them, image segmentation can be performed by head-shoulder segmentation. Head-shoulder segmentation refers to segmenting the image position of each passenger in the collected image. The images taken by buses are taken from the top of the passengers' heads. The captured images can only observe the head and shoulder positions of each passenger. The image position of each passenger in the collected image is segmented.

[0038] Optionally, you can use the target detection algorithm to detect each passenger in the image, obtain the detection frame and coordinates of each passenger, segment each detected passenger area, calculate the feature vector for each segmented passenger area, and obtain the feature vector corresponding to each passenger area. Each passenger has an independent detection frame, and these passenger areas are processed in the order of the detection frames, and the index is kept unchanged during segmentation and feature calculation to ensure that the order of the output feature vectors is consistent with the order of the detection frames.

[0039] For example, when a bus stops at a station, image analysis can be used to determine if a passenger enters the bus from outside the bus, thereby determining that the passenger has boarded the bus; and when a passenger moves from inside the bus to outside the bus, thereby determining that the passenger has gotten off the bus.

[0040] In addition, the timestamps when passengers get on and off the bus can also be obtained. The timestamps when passengers get on and off the bus can be obtained by obtaining the acquisition time of the captured image, taking the timestamp when the passenger first appears in the appearance image as the boarding timestamp, and taking the timestamp when the passenger disappears from the appearance image and moves from inside the bus to outside the bus as the alighting timestamp.

[0041] Among them, the timestamp may refer to the moment when the passenger gets on and off the bus when the image is captured. The timestamp is used to subsequently arrange an accurate timetable for the public transportation system. For example, when the number of bus passengers increases significantly from a certain time to a certain time, more buses can be dispatched during this time period to solve the problem of crowded buses. On the other hand, the timestamp can also make a preliminary judgment on the abnormal behavior of passengers getting on and off the bus. For example, the boarding time of a passenger must be earlier than the disembarking time. When the boarding time of a passenger is later than the disembarking time, it means that the data is abnormal. This type of data can be analyzed separately to determine whether a malfunction of the on-board equipment has caused a passenger's boarding time to be later than the disembarking time.

[0042] Optionally, at least one passenger is identified in each collected image acquired at the site, and at least one appearance image of each passenger is screened out in each collected image; image information of each passenger is determined based on each appearance image of each passenger; and travel direction of each passenger is determined based on each appearance image of each passenger.

[0043] Among them, the appearance image may refer to a captured image containing a passenger. The image information of the passenger may be extracted from at least one appearance image of the same passenger. The direction of movement of the passenger relative to the shooting device may be determined based on the shooting time of the appearance image of the same passenger and the relative position of the passenger in each appearance image. According to the direction of movement and the relative position between the passenger and the bus, it is determined whether the passenger is getting on or off the bus, thereby determining the passenger's riding direction. Exemplarily, when a bus arrives at a certain station, the shooting device captures multiple captured images at the door, some of which show a passenger approaching the bus and stepping onto the bus. The appearance image may refer to the above series of images of passengers appearing.

[0044] By automatically identifying at least one passenger in each collected image obtained at the station, the need for manual identification is reduced, and images can be quickly processed and analyzed, thereby improving the efficiency of passenger identification; by screening the images that appear in the collected images and determining the image information and the riding direction based on the images that appear, the passenger's riding direction can be determined, so that the passenger's boarding image can be correctly matched in the subsequent disembarkation image range to generate complete passenger flow data; the image features of the passenger obtained from the image information are helpful for the subsequent calculation of similarity, avoiding errors and misjudgments caused by manual judgment of whether the passenger is the same.

[0045] Optionally, for each passenger, a target image is screened from the appearance images of each passenger; and the image features of the passenger are extracted from the target image of the passenger and used as the image information of the passenger.

[0046] The target image may refer to an image that contains the most representative feature information of the passenger. For example, an appearance image may be randomly selected from at least one appearance image as the target image; or an appearance image with the largest image area of ​​the passenger may be selected as the target image; or an image in which the passenger appears completely for the first time may be selected as the target image.

[0047] The role of the target image is to provide a clear and unobstructed perspective, from which the image features of the passenger can be fully extracted, reducing recognition errors caused by factors such as image blur or occlusion. Screening out the target image can improve recognition accuracy and avoid multiple calculations that waste computing resources and mismatches. For example, in the multiple frames of appearance images captured by the door camera, the appearance image of a certain passenger includes partially blurred, partially blocked, and completely clear images. According to preset rules, the appearance image that best meets the conditions is selected from the appearance images as the target image for subsequent extraction of the passenger's image features as the passenger's image information.

[0048] Image features may refer to feature vectors extracted from the target image. The image features are used to distinguish the visual features of different passengers, uniquely identify each passenger, distinguish them from other passengers, and provide a basis for subsequent image matching. For example, the target image of a passenger is a clear and complete image, and the features of the passenger's hairstyle, body shape, clothing, etc. can be extracted through a deep learning model or other methods.

[0049] By screening out the target image from the appearing images, the number of images that need to be processed is reduced, work efficiency is improved, and errors caused by repeatedly collecting images of the same passenger are avoided; extracting the image features of the passenger from the target image can ensure the consistency of the extracted information, which is helpful for subsequent data analysis and comparison, and avoids inconsistent features caused by different image sources or states; the extracted passenger image features serve as the passenger's image information, providing strong support for further data analysis.

[0050] Optionally, among the captured images acquired at the site, at least one captured image corresponding to each vehicle door is determined; the captured images corresponding to the vehicle door are multiple consecutive images taken of the area of ​​the vehicle door; for each vehicle door, at least one user is identified in each captured image corresponding to the vehicle door; based on the appearance status of each user in each captured image corresponding to the vehicle door, at least one passenger is screened for each user, and at least one appearance image of each passenger is determined.

[0051] The appearance state may refer to the state of the user at the vehicle door, and may include the user's stay time at the vehicle door.

[0052] Screening passengers from among the users may include determining users whose stay time is greater than or equal to a preset time threshold as passengers, and determining a captured image containing the user as at least one appearance image of the passenger.

[0053] At the bus stop, each captured image is obtained by continuously shooting the door area; the appearance image of each passenger is determined according to the appearance state of each user in the captured image corresponding to the door.

[0054] Determining at least one captured image corresponding to each door helps to concentrate image analysis on a specific area, thereby improving the quality of data collection. By focusing on the door area, detailed information related to the passengers' boarding and getting off the vehicle can be captured; each captured image corresponding to the door is a series of images taken of the door area, which can capture the dynamic behavior of passengers in real time, including states such as entering, leaving and staying. This continuity helps to more comprehensively understand the passenger's behavior patterns; identifying at least one user in each captured image corresponding to the door can reduce dependence on manual intervention, improve the efficiency of passenger identification, and quickly process a large number of images; screening at least one passenger based on the user's appearance status in each captured image corresponding to the door and determining the image in which he appears can effectively identify passengers who need attention and extract key information from complex data.

[0055] Optionally, when the user's behavior of two riding directions appears in each collected image corresponding to the door, it is determined that the user is not a passenger.

[0056] The behavior of the user having two boarding directions in each collected image corresponding to the door may refer to the behavior of the user having two different boarding directions in all collected images at the same door, for example, the user has a boarding direction in the first time period and a disembarking direction in the second time period, or the user has a disembarking direction in the first time period and a boarding direction in the second time period. The first time period is prior to the second time period. The time between the first time period and the second time period is less than the preset time threshold.

[0057] It is understandable that when a user gets off the bus again within a short time interval after getting on the bus through the same door, it can be determined that the user may have gotten off the bus temporarily due to crowding or other abnormal reasons, or the user got on the wrong bus, and the behavior can be determined to be abnormal public transportation riding behavior. For example, when the bus is crowded at a certain station, a passenger needs to get off the bus first because he needs to give way to other passengers getting off the bus. When other passengers have gotten off the bus, the passenger can get on the bus again. This behavior is not normal public transportation riding behavior, and the information cannot be used for the content of public transportation passenger flow information.

[0058] By analyzing the user's boarding direction behavior in each captured image corresponding to the door, it is possible to effectively determine whether the user is a passenger, which helps to distinguish between passengers and non-passengers and improve the accuracy of passenger identity confirmation, thereby improving the accuracy of passenger flow data; determining that the user is not a passenger can effectively reduce false alarms and misidentifications in passenger data records, improve overall reliability and trust, and focus resources and processing capabilities on real passengers.

[0059] S103, matching the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, and determining the riding station of each passenger, where the riding station includes a boarding station and an alighting station.

[0060] Among them, matching the image information of each passenger may refer to determining the boarding and alighting stations of the same passenger's ride, thereby matching and associating the boarding image information and alighting image information of a certain passenger to form the complete passenger flow information of the passenger. By matching the boarding and alighting image information of passengers, the complete riding information of each passenger can be integrated, including information such as the bus route taken, the location information of the boarding and alighting stations, and the time when the passenger boarded and alighted. For example, a passenger boards at the XX station, and his boarding image information is identified. Then the passenger gets off at the YY station, and his alighting image information is captured again. By matching these two pieces of information, the complete riding behavior of the passenger can be determined.

[0061] Among them, according to the riding direction and boarding station of each passenger, the boarding image information and the alighting image information of the same passenger are matched to form complete passenger riding information.

[0062] Optionally, for each image information of the same riding direction, the image information of the passenger in the riding direction is matched with each image information in the opposite direction to obtain the image information of the riding direction corresponding to the image information of the riding direction, where the opposite direction is the riding direction opposite to the riding direction; the passenger of the image information of the riding direction and the passenger of the corresponding image information of the opposite direction are determined to be the same passenger; the riding station corresponding to the riding direction of the same passenger and the riding station corresponding to the corresponding opposite direction are determined to be the riding station of the same passenger.

[0063] Among them, the riding direction includes the boarding direction and the alighting direction. The riding information of all passengers' boarding direction and the riding information of all passengers' alighting direction are matched, and the riding station at the boarding position and the riding station at the alighting position of each passenger are used as the riding station of the same passenger.

[0064] By matching the image information of the same riding direction with the image information of the opposite direction, the passenger's identity can be confirmed more accurately, and the image information of different directions is used to enhance the reliability of identity matching; matching the image information of passengers in the riding direction and the opposite direction helps to fully understand the passenger's behavior trajectory and flow pattern, and provide strong support for passenger flow management; determining the riding station corresponding to the riding direction and the opposite direction of the same passenger helps to optimize the information management of the riding station. It can effectively integrate the passenger's boarding and alighting information and improve the efficiency of data utilization.

[0065] Optionally, the similarity between the image features of the passenger in the riding direction and the image features in the opposite direction is calculated; based on the similarity between the image features of the riding direction and the image features in the opposite direction, similar image features in the opposite direction are determined and used as image information in the opposite direction corresponding to the image information in the riding direction.

[0066] The similarity may refer to the similarity between the image features of the passenger's boarding direction and the image features of the opposite alighting direction. The similarity may be obtained through a deep learning model, or the similarity may be calculated through the cosine distance of a vector. The similarity is used to find the same passenger in the alighting image as in the boarding image, and also to find the same passenger in the alighting image as in the boarding image.

[0067] By calculating the similarity between the image features of the passenger in the riding direction and the image features in the opposite direction, the matching image features can be found more accurately, ensuring that the feature matching is based on specific similarity analysis and improving the reliability of the matching; by utilizing similarity calculation, image features in different directions can be automatically identified and matched, reducing manual intervention.

[0068] The similarity Sim between a passenger's boarding target image and alighting target image can be calculated by the following formula:

[0069]

[0070] In the formula, v new It refers to the feature vector of a passenger’s boarding target image, v i It refers to the feature vector of the image with the i-th riding direction being the alighting direction. It is matched by finding the feature vector that is most similar to the feature vector of the target image among the feature vectors of all the images with the riding direction being the alighting direction.

[0071] In a specific example, for the captured boarding target image, combined with the current bus route and the location of the current bus stop, an initial boarding OD (Origin and Destination) record OD is generated. up as follows:

[0072] OD up ={(Line,ID new ,t1,null1,S org ,null2)}

[0073] In the formula, Line refers to the current bus route, ID new is the unique identifier of the passenger, t1 is the boarding timestamp of the image capture, null1 is the null value 1, S orgRefers to the boarding station information obtained by converting GPS coordinates. Null2 is null value 2. Since the initial OD record only records the passenger's boarding information, there is no alighting station information and alighting timestamp, and the record is null value 1 and null value 2.

[0074] Calculate v new With each v in the alight image feature library i and find the eigenvector v with the highest similarity. i , and the v with the highest similarity best The corresponding passenger ID is assigned to v new , and store the matching results in the following format:

[0075] OD down =OD final ={(Line,ID best ,t1,t2,S org ,S dst )}

[0076] In the formula, OD down To get off the OD record, OD final For the final storage format, get off OD record and the final storage format OD final Keep it consistent, Line refers to the current bus route, ID best Refers to the ID in the boarding OD record new The closest ID, t1 is the boarding timestamp captured by the image, t2 is the getting off timestamp captured by the image, S org Refers to the boarding station information obtained by converting GPS coordinates. dst Refers to the alighting station information converted from GPS coordinates.

[0077] The final OD dataset can be expressed as follows:

[0078] OD_Data = {OD final}

[0079] In a specific example, these OD data are transmitted to the background analysis platform for big data analysis. Through the passenger flow OD data analysis system, the bus operator can comprehensively analyze the passenger flow of routes, stations, and vehicles, optimize route planning and capacity allocation, and realize intelligent scheduling, accurate timetable arrangement, line network optimization, and multi-mode customized bus functions based on the comprehensive information such as the number of passengers in the riding interval, boarding time points, and concentrated boarding and alighting stations.

[0080] The technical solution of the embodiment of the present invention can ensure that the position of the captured image is captured at the station by acquiring the captured image at the door of the bus when the bus stops at the station, and can objectively and real-time record the passengers' boarding and alighting behaviors; the way of collecting passenger information through images can be recorded according to actual behaviors, reducing the dependence of collected information on passenger operating behaviors and avoiding errors or omissions caused by dependence on passengers; by identifying the passenger information in the captured image and combining the passenger's riding direction, the specific behavior of the passenger at each station can be determined, and the analysis of the riding direction helps to distinguish the passenger's behavior at a certain station, ensuring the accurate matching of station information and passenger behavior; by combining the passenger's riding direction, riding direction and station data, the passenger's boarding station and alighting station can be accurately matched, thereby improving the integrity of the passenger's records, reducing the possibility of duplicate records or erroneous matching, ensuring the consistency and accuracy of the data, and providing more accurate data support for subsequent bus passenger flow analysis and operation optimization.

[0081] Embodiment 2

[0082] Figure 2 1 is a structural diagram of a bus passenger flow information acquisition device provided according to Embodiment 2 of the present invention. The embodiment of the present invention is applicable to the case of bus passenger flow information acquisition, the device can execute the bus passenger flow information acquisition method, and the device can be implemented in the form of hardware and / or software.

[0083] See also Figure 2 The bus passenger flow information acquisition device shown includes: an image acquisition module 201, a passenger identification module 202 and an image information matching module 203, wherein:

[0084] An image acquisition module 201 is used to acquire an image of a door of a public transport vehicle when the public transport vehicle stops at at least one station;

[0085] The passenger identification module 202 is used to identify passengers in at least one collected image acquired at each station, obtain image information of at least one passenger and the riding direction of each passenger, and determine the station as the riding station corresponding to each image information;

[0086] The image information matching module 203 is used to match the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, and determine the riding station of each passenger. The riding station includes the boarding station and the alighting station.

[0087] The technical solution of the embodiment of the present invention can ensure that the position of the captured image is captured at the station by acquiring the captured image at the door of the bus when the bus stops at the station, and can objectively and real-time record the passengers' boarding and alighting behaviors; the way of collecting passenger information through images can be recorded according to actual behaviors, reducing the dependence of collected information on passenger operating behaviors and avoiding errors or omissions caused by dependence on passengers; by identifying the passenger information in the captured image and combining the passenger's riding direction, the specific behavior of the passenger at each station can be determined, and the analysis of the riding direction helps to distinguish the passenger's behavior at a certain station, ensuring the accurate matching of station information and passenger behavior; by combining the passenger's riding direction, riding direction and station data, the passenger's boarding station and alighting station can be accurately matched, thereby improving the integrity of the passenger's records, reducing the possibility of duplicate records or erroneous matching, ensuring the consistency and accuracy of the data, and providing more accurate data support for subsequent bus passenger flow analysis and operation optimization.

[0088] Optionally, the passenger identification module 202 is specifically used for:

[0089] Identifying at least one passenger in each collected image acquired at the station, and screening out at least one appearance image of each passenger in each collected image;

[0090] Determining image information of each passenger according to each appearance image of each passenger;

[0091] Based on each appearance image of each passenger, the boarding direction of each passenger is determined.

[0092] Optionally, the passenger identification module 202 is specifically used for:

[0093] For each passenger, a target image is selected from the appearance images of each passenger;

[0094] From the target image of the passenger, the image features of the passenger are extracted and used as the image information of the passenger.

[0095] Optionally, the image information matching module 203 is specifically used for:

[0096] For each image information of the same riding direction, the image information of the passenger in the riding direction is matched with each image information in the opposite direction to obtain the image information in the opposite direction corresponding to the image information in the riding direction, where the opposite direction is the riding direction opposite to the riding direction;

[0097] Determine that a passenger in the image information of the riding direction and a passenger in the corresponding image information of the opposite direction are the same passenger;

[0098] The boarding station corresponding to the boarding direction of the same passenger and the boarding station corresponding to the corresponding opposite direction are determined as the boarding stations of the same passenger.

[0099] Optionally, the image information matching module 203 is specifically used for:

[0100] Calculate the similarity between the image features of the passenger in the riding direction and the image features in the opposite direction;

[0101] According to the similarity between the image features of the riding direction and each image feature of the opposite direction, similar image features of the opposite direction are determined and used as the image information of the opposite direction corresponding to the image information of the riding direction.

[0102] Optionally, the passenger identification module 202 is specifically used for:

[0103] Determine at least one collected image corresponding to each vehicle door from among the collected images acquired at the site; each collected image corresponding to the vehicle door is a plurality of continuous images captured of the area of ​​the vehicle door;

[0104] For each vehicle door, identifying at least one user in each collected image corresponding to the vehicle door;

[0105] According to the appearance status of each user in each collected image corresponding to the vehicle door, at least one passenger is screened for each user, and at least one appearance image of each passenger is determined.

[0106] Optionally, the passenger identification module 202 is specifically used for:

[0107] If the user shows two riding directions in each collected image corresponding to the door, it is determined that the user is not a passenger.

[0108] The bus passenger flow information device provided in the embodiment of the present invention can execute the bus passenger flow information method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the bus passenger flow information method.

[0109] Embodiment 3

[0110] Figure 3 It is a structural schematic diagram of a bus passenger flow information acquisition device provided according to Embodiment 3 of the present invention.

[0111] like Figure 3As shown, the bus passenger flow information device 300 includes at least one processor 301, and a memory connected to the at least one processor 301, such as a read-only memory (ROM) 302, a random access memory (RAM) 303, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 301 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 302 or the computer program loaded from the storage unit 308 to the random access memory (RAM) 303. In RAM303, various programs and data required for the operation of the bus passenger flow information device 300 can also be stored. The processor 301, ROM302 and RAM303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0112] A number of components in the bus passenger flow information device 300 are connected to the I / O interface 305, including: an input unit 306, such as a keyboard, a mouse, etc.; an output unit 307, such as various types of displays, speakers, etc.; a storage unit 308, such as a disk, an optical disk, etc.; and a communication unit 309, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 309 allows the bus passenger flow information device 300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0113] Processor 301 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of processor 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. Processor 501 executes the various methods and processes described above, such as the bus passenger flow information method.

[0114] In some embodiments, the bus passenger flow information method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 308. In some embodiments, part or all of the computer program may be loaded and / or installed on the bus passenger flow information device 300 via the ROM 302 and / or the communication unit 309. When the computer program is loaded into the RAM 303 and executed by the processor 301, one or more steps of the bus passenger flow information method described above may be performed. Alternatively, in other embodiments, the processor 301 may be configured to execute the bus passenger flow information method in any other appropriate manner (e.g., by means of firmware).

[0115] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0117] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0118] To provide interaction with a user, the systems and techniques described herein may be implemented on an operation detection device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which a user can provide input to the bus passenger flow information device. Other types of devices may also be used to provide interaction with a user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0119] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0120] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS (Virtual Private Server) services.

[0121] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0122] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for obtaining public transport passenger flow information, characterized in that: The method comprises: Acquire a captured image of a door of the public transport vehicle when the public transport vehicle stops at at least one station; For each of the stations, a passenger is identified in at least one collected image acquired at the station, image information of at least one passenger and a riding direction of each of the passengers is obtained, and the station is determined as a riding station corresponding to each of the image information; According to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, the image information of each passenger is matched to determine the riding station of each passenger, and the riding station includes a boarding station and an alighting station.

2. The method according to claim 1, characterized in that The identifying of passengers in at least one collected image acquired at the station to obtain image information of at least one passenger and the riding direction of each passenger includes: Identifying at least one passenger in each of the collected images acquired at the site, and screening out at least one appearance image of each of the passengers in each of the collected images; Determining image information of each of the passengers according to each appearance image of each of the passengers; The boarding direction of each passenger is determined based on each appearance image of each passenger.

3. The method according to claim 2, characterized in that Determining the image information of each passenger according to each appearance image of each passenger includes: For each of the passengers, a target image is selected from the images in which the passengers appear; The image features of the passenger are extracted from the target image of the passenger and used as the image information of the passenger.

4. The method according to claim 3, characterized in that The step of matching the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger to determine the riding station of each passenger includes: For each image information of the same riding direction, the image information of the passenger in the riding direction is matched with each image information in the opposite direction to obtain image information in the opposite direction corresponding to the image information in the riding direction, wherein the opposite direction is the riding direction opposite to the riding direction; Determine that the passenger of the image information in the riding direction and the passenger of the corresponding image information in the opposite direction are the same passenger; The boarding station corresponding to the boarding direction of the same passenger and the boarding station corresponding to the corresponding opposite direction are determined as the boarding stations of the same passenger.

5. The method according to claim 4, characterized in that The matching of the image information of the passenger in the riding direction with the image information in the opposite direction to obtain the image information in the opposite direction corresponding to the image information in the riding direction includes: Calculating the similarity between the image features of the passenger in the riding direction and each image feature in the opposite direction; According to the similarity between the image features of the riding direction and each image feature of the opposite direction, similar image features of the opposite direction are determined and used as the image information of the opposite direction corresponding to the image information of the riding direction.

6. The method according to claim 2, characterized in that The step of identifying at least one passenger in each of the collected images acquired at the site, and screening out at least one appearance image of each of the passengers in each of the collected images, comprises: Determine at least one collected image corresponding to each vehicle door among the collected images acquired by the site; each collected image corresponding to the vehicle door is a plurality of continuous images taken of the area of ​​the vehicle door; For each of the vehicle doors, identifying at least one user in each collected image corresponding to the vehicle door; According to the appearance status of each of the users in each of the collected images corresponding to the vehicle door, at least one passenger is screened from each of the users, and at least one appearance image of each of the passengers is determined.

7. The method according to claim 6, characterized in that The step of screening at least one passenger for each of the users and determining at least one appearance image of each of the passengers comprises: If the user shows two riding directions in each collected image corresponding to the door, it is determined that the user is not a passenger.

8. A device for acquiring public transport passenger flow information, characterized in that: The device comprises: An image acquisition module, used to acquire an image of a door of a public transport vehicle when the public transport vehicle stops at at least one station; A passenger identification module, for each of the stations, identifying a passenger in at least one collected image acquired at the station, obtaining image information of at least one passenger and a riding direction of each of the passengers, and determining the station as a riding station corresponding to each of the image information; The image information matching module is used to match the image information of each passenger according to the riding direction of each passenger and the riding station corresponding to the image information of each passenger, and determine the riding station of each passenger, wherein the riding station includes the boarding station and the alighting station.

9. A device for acquiring public transport passenger flow information, characterized in that: The bus passenger flow information acquisition device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for acquiring bus passenger flow information according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for obtaining bus passenger flow information according to any one of claims 1 to 7 when executed.

Images