Passenger Flow Counting Method, Equipment, and Flight Attendant Work Lights

Abstract

This invention discloses a passenger flow counting method, device, and attendant work light based on attendant work lights, relating to the field of lighting technology. The method includes: driving the work lights in a detection area to illuminate; acquiring first image information of the detection area captured by a first camera and second image information of the detection area captured by a second camera; calculating the disparity information of a target object based on the first and second image information; generating a depth map using triangulation based on the disparity information; identifying human feature information based on the depth map to extract target personnel; tracking the target personnel to identify target passengers from among the target personnel; and counting the number of target passengers. Using this invention can improve the accuracy of passenger flow calculation.

Description

Technical Field

[0001] This invention relates to the field of lighting technology, and in particular to a passenger flow counting method, device, and passenger service light based on a flight attendant work light. Background Technology

[0002] Flight attendant work lights are typically installed in the flight attendant work area of ​​an aircraft, primarily providing overhead ambient lighting for the flight attendants' workstations, and are an important component of the cabin lighting system. Currently, the function of flight attendant work lights is still mainly single-function lighting, and their control logic mainly revolves around basic lighting needs such as brightness adjustment and color temperature switching.

[0003] In recent years, with the continuous growth of civil aviation passenger traffic and airlines' increasing emphasis on operational efficiency and service quality, more and more airlines have begun to focus on and promote the application of accurate passenger counting within aircraft. By monitoring boarding, disembarking, and passenger distribution in real time, cabin crew can be effectively assisted in tasks such as headcount verification, seat guidance, and emergency management, thereby improving the level of intelligence in cabin management.

[0004] However, despite the progress made in open environments such as airport terminals, the widespread adoption of passenger flow counting technology in the enclosed, confined, and highly dynamic environment of aircraft cabins still faces many unique challenges, including: (1) The cabin environment is highly complex: the cabin space is cramped, the passageways are narrow, the people are densely packed and move frequently, and there is serious mutual obstruction between flight attendants and passengers, and between passengers themselves; at the same time, the lighting conditions in the cabin are highly variable, and are significantly affected by multiple light sources such as natural light from the windows, night flight mode, and reading lights. This places extremely high demands on the environmental adaptability, anti-interference ability and stability of the sensors and algorithms on which passenger flow counting depends.

[0005] (2) Strict data privacy and security compliance requirements: If visual sensors such as cameras are used for people counting, privacy regulations must be strictly followed to ensure that the collected data is fully anonymized and that no personal identity is identified or unnecessary biometric information is stored. This increases the complexity of system design and data processing and compliance costs to some extent.

[0006] (3) High cost of modification: Modifying or adding cabin equipment on existing aircraft that are already in operation involves multiple steps such as airworthiness certification, structural modification, wiring and system integration. Not only is the cost of modification significant, but it also requires aircraft parking time, affecting flight operation plans and constituting an important economic obstacle to the implementation of new technologies.

[0007] In summary, existing flight attendant work lights have limited functionality, and passenger counting methods have not yet been widely adopted in the cabin. Therefore, there is an urgent need to propose a new solution that can organically combine flight attendant work lights with passenger counting technology without significantly increasing modification costs and compliance risks. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a passenger flow counting method, device and passenger flow counting light based on flight attendant work lights, which can improve the accuracy of passenger flow calculation.

[0009] To address the aforementioned technical problems, this invention provides a passenger flow counting method based on flight attendant work lights, comprising: illuminating the work lights in a detection area; acquiring first image information of the detection area captured by a first camera and second image information of the detection area captured by a second camera; calculating the disparity information of a target object based on the first and second image information; generating a depth map using triangulation based on the disparity information; identifying human feature information based on the depth map to extract target personnel; tracking the target personnel to identify target passengers from among the target personnel; and counting the number of target passengers.

[0010] As an improvement to the above scheme, the step of tracking the target personnel to identify target passengers includes: assigning a temporary tracking identifier to the target personnel; continuously collecting the behavioral sequence and dwell time of the temporary tracking identifier, wherein the behavioral sequence is used to record the coordinate information of key points of the target personnel, the key points including head feature points and torso feature points, the torso feature points including shoulder feature points and hip feature points; determining whether the behavioral characteristics of the temporary tracking identifier match the preset flight attendant behavior based on the behavioral sequence and dwell time, if the determination is yes, then marking the temporary tracking identifier as "flight attendant", if the determination is no, then marking the temporary tracking identifier as "target passenger".

[0011] As an improvement to the above scheme, the flight attendant behaviors include nodding, bending over, and prolonged stays.

[0012] As an improvement to the above scheme, the step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior based on the behavioral sequence and dwell time includes: within a preset first duration, if the head feature point fluctuates at least a preset number of times, then the behavioral characteristics of the temporary tracking identifier match the nodding behavior; the fluctuation refers to the situation where the change amplitude of the Y coordinate of the head feature point is greater than or equal to a preset baseline change amplitude and the change frequency is within a preset nodding frequency range.

[0013] As an improvement to the above scheme, the step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset flight attendant behavior based on the behavioral sequence and dwell time includes: if the torso feature point moves and the duration of the movement is greater than or equal to a preset second duration, then the behavioral characteristics of the temporary tracking identifier match the bending behavior; the movement refers to the situation where the angle between the line connecting the shoulder feature point and the hip feature point and the vertical direction is greater than or equal to a preset reference angle or the descent of the Y coordinate of the hip feature point is greater than or equal to a preset reference descent.

[0014] As an improvement to the above scheme, the step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset flight attendant behavior based on the behavioral sequence and dwell time includes: if the key point changes and the dwell time is greater than or equal to the preset third duration, then the behavioral characteristics of the temporary tracking identifier match the long-term dwell behavior; the change includes nodding behavior, bending behavior and specific displacement of key points.

[0015] As an improvement to the above scheme, the passenger flow counting method based on the flight attendant work lights further includes: if the key point remains unchanged and the dwell time is greater than or equal to a preset fourth duration, then the temporary tracking identifier is marked as "suspicious dwelling" and an alarm signal is generated.

[0016] As an improvement to the above scheme, the passenger flow counting method based on flight attendant work lights further includes: sending the coordinate information of the key points corresponding to the temporary tracking identifier marked "flight attendant" to an external device, and / or counting the number of flight attendants.

[0017] Accordingly, the present invention also provides a passenger flow counting device based on a flight attendant work light, including a memory and a processor. The memory stores a computer program, wherein the processor executes the computer program to implement the steps of the above-mentioned passenger flow counting method based on a flight attendant work light.

[0018] Accordingly, the present invention also provides a flight attendant work light, comprising: a work light for illuminating a detection area; a first camera for acquiring first image information of the detection area; a second camera for acquiring second image information of the detection area; the aforementioned passenger flow counting device based on the flight attendant work light; and a housing for encapsulating the first camera, the second camera, the work light, and the passenger flow counting device.

[0019] The flight attendant work light of this invention can be applied to aviation, railway or other transportation fields. Implementing this invention has the following beneficial effects: This invention uses binocular parallax technology to construct depth maps without requiring personal information, thus effectively protecting passenger privacy. Simultaneously, it identifies human features through depth maps to extract target individuals, then tracks these individuals to identify target passengers, thereby achieving accurate passenger counts and effectively preventing the artificial increase in passenger flow caused by flight attendants moving around at the cabin doors.

[0020] Furthermore, this invention introduces temporary tracking identifiers and constructs behavioral sequences based on these identifiers, which can effectively protect the privacy of passengers and flight attendants. At the same time, this invention also achieves accurate identification of flight attendants and passengers by tracking behavioral sequences and dwell time, so as to separately collect data on target passengers and / or flight attendants. Attached Figure Description

[0021] Figure 1 This is a flowchart of the first embodiment of the passenger flow counting method based on flight attendant work lights of the present invention; Figure 2 This is a flowchart of the second embodiment of the passenger flow counting method based on flight attendant work lights of the present invention; Figure 3 This is a schematic diagram of an embodiment of the flight attendant work light of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. It is hereby declared that the directional terms such as up, down, left, right, front, back, inside, and outside used in this text are based solely on the accompanying drawings and are not intended to specifically limit the invention.

[0023] See Figure 1 , Figure 1 The flowchart of a first embodiment of the passenger flow counting method based on flight attendant work lights of the present invention is shown, which includes: S101, the working light in the drive detection area is turned on; When working, the work light must be turned on first to provide a light source for the inspection area.

[0024] S102, acquire the first image information of the detection area captured by the first camera and the second image information of the detection area captured by the second camera; S103, calculate the disparity information of the target object based on the first image information and the second image information; S104, Generate a depth map using triangulation based on parallax information; This invention is based on left and right binocular parallax and uses triangulation to obtain depth information. That is, a triangle is formed between the image planes of the two cameras and the target object. Therefore, if the positional relationship between the two cameras is known, the depth information of the object in the common field of view of the two cameras can be obtained.

[0025] S105, Identify human feature information based on depth map to extract target personnel; Before identification, human features can be modeled based on depth information to obtain a human detection geometric mathematical model; during the identification process, this human detection geometric mathematical model can be used to identify human feature information in the depth map in order to extract the target person.

[0026] Human body feature information includes three-dimensional features of the head and shoulders, but is not limited to this.

[0027] S106, Tracking target personnel to identify target passengers from among them; In other words, once a target person enters the detection area, the present invention will continuously detect the real-time location of the target person and track the trajectory of the target person's position changes, thereby identifying the type of the target person.

[0028] Accordingly, the steps for tracking target individuals to identify target passengers include: (1) Assign temporary tracking tags to the target personnel; When a target person is identified from the depth map, different temporary tracking identifiers are assigned to different target persons, such as ID_001 and ID_002.

[0029] (2) Continuously collect the behavioral sequences and dwell time of temporary tracking tags; It should be noted that the behavior sequence is used to record the coordinate information of key points of the target person. Key points include head feature points (such as nose tip, ears, and neck) and torso feature points. Torso feature points include shoulder feature points and hip feature points. (3) Determine whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior based on the behavioral sequence and dwell time; (4) If the determination is yes, then mark the temporary tracking identifier as "flight attendant"; (5) If the determination is no, the temporary tracking identifier will be marked as "target passenger".

[0030] It should be noted that when flight attendants are working, they often use "nodding behavior" to express states such as "confirmation and response" and "ticket check indication", and they also use "bending behavior" to express states such as "bending down to greet", "checking under the seat", "picking up items", and "adjusting equipment". They also use "prolonged standing behavior" to express states such as "standing in the inspection area for a long time". Therefore, this invention divides flight attendant behavior into: nodding behavior, bending behavior and prolonged standing behavior.

[0031] Accordingly, if the behavior sequence and dwell time can match the flight attendant's behavior, the temporary tracking identifier will be marked as "flight attendant", and the subsequent passenger flow counting results marked by the temporary tracking identifier will be specially marked or exempted.

[0032] The matching rules for nodding, bending over, and prolonged standing are explained in detail below: I. Nodding Behavior Within a preset first time period, if the head feature point fluctuates at least a preset number of times, the behavior characteristics of the temporary tracking marker match the nodding behavior; where fluctuation refers to the situation where the change amplitude of the Y coordinate of the head feature point is greater than or equal to the preset baseline change amplitude and the change frequency is within the preset nodding frequency range.

[0033] In this embodiment, the first duration is 2 seconds, the preset number of times is 2, the baseline change range is 15 pixels, and the nodding frequency range is 0.5~2Hz. Therefore, if within 2 seconds, the Y coordinate change range of the head feature point is greater than or equal to 15 pixels and the change frequency is between 0.5 and 2Hz more than twice, it can be determined that the target person is nodding.

[0034] In other embodiments, parameters such as the first duration, preset number of times, baseline change range, and nodding frequency range can be flexibly set according to requirements.

[0035] It should be noted that since the gait frequency of the target person is usually greater than 2Hz when walking normally, the above matching rules can effectively eliminate the natural bumps during walking.

[0036] II. Bending over If the torso feature point moves and the duration of the movement is greater than or equal to the preset second duration, the behavioral characteristics of the temporary tracking marker match the bending behavior; where movement refers to the situation where the angle between the line connecting the shoulder feature point and the hip feature point and the vertical direction is greater than or equal to the preset reference angle or the descent of the Y coordinate of the hip feature point is greater than or equal to the preset reference descent.

[0037] In this embodiment, the second duration is 1.5 seconds, the reference angle is 45°, and the reference descent is 30 pixels. Therefore, if the angle between the line connecting the shoulder feature point and the hip feature point and the vertical direction is greater than or equal to 45° and the duration of the movement is greater than or equal to 1.5 seconds, it can be determined that the target person is bending over. In addition, if the descent of the Y coordinate of the hip feature point is greater than or equal to 30 pixels and the duration of the movement is greater than or equal to 1.5 seconds, it can also be determined that the target person is bending over.

[0038] In other embodiments, parameters such as the second duration, reference angle, and reference descent amplitude can be flexibly set according to requirements.

[0039] It should be noted that the present invention introduces a duration, which can effectively exclude situations where the duration is short, such as squatting down to tie shoelaces.

[0040] III. Prolonged Stay Behavior If the key point changes and the dwell time is greater than or equal to the preset third duration, the behavioral characteristics of the temporary tracking marker match the long-term dwell behavior; among which, the changes include nodding behavior, bending behavior, and specific displacement of the key point (such as moving left or right or rotating).

[0041] In this embodiment, the third duration is 60 seconds; therefore, if the dwell time is greater than or equal to 60 seconds, and nodding, bending over, or specific displacement of key points occur during this period, it can be determined that the target person has engaged in long-term dwelling behavior.

[0042] In other embodiments, the third duration can be flexibly set as needed.

[0043] To prevent passengers from being mistakenly identified as sleeping or using their phones, this invention requires monitoring specific displacements at key points to ensure that the target person is not asleep or in a static state.

[0044] Furthermore, if the key points remain unchanged and the dwell time is greater than or equal to the preset fourth duration, the temporary tracking identifier will be marked as "suspicious dwelling" and an alarm signal will be generated.

[0045] In this embodiment, the third duration is 5 minutes. Therefore, if the dwell time is greater than or equal to 5 minutes and no nodding, bending or specific displacement of key points occurs during the period, the target person can be determined to be "suspiciously staying" and an alarm signal can be generated to notify the management personnel.

[0046] S107, Count the number of target passengers.

[0047] Therefore, the passenger flow counting method based on flight attendant work lights of the present invention can improve the identification accuracy of target passengers, and count the data of target passengers and / or flight attendants in real time, effectively avoiding the false increase in passenger flow caused by flight attendants walking at the cabin door.

[0048] See Figure 2 , Figure 2 The flowchart of a second embodiment of the passenger flow counting method based on flight attendant work lights of the present invention is shown, which includes: S201, the indicator light in the drive detection area is turned on; S202, acquire the first image information of the detection area captured by the first camera and the second image information of the detection area captured by the second camera; S203, calculate the disparity information of the target object based on the first image information and the second image information; S204, Generate a depth map using triangulation based on parallax information; S205, Identify human feature information based on depth map to extract target personnel; S206, Tracking target personnel to identify target passengers and flight attendants from among them; S207, sends the flight attendant's coordinate information to an external device; The coordinates of key points corresponding to the temporary tracking identifier marked "flight attendant" will be sent to external devices to facilitate safety supervision and personnel qualification verification by management personnel.

[0049] S208, count the number of target passengers and flight attendants respectively.

[0050] Accordingly, the present invention also discloses a passenger flow counting device based on a flight attendant work light, including a memory and a processor. The memory stores a computer program, wherein the processor executes the computer program to implement the steps of the above-mentioned passenger flow counting method based on a flight attendant work light.

[0051] See Figure 3 , Figure 3 An embodiment of the flight attendant work light of the present invention is shown, which includes a work light 1, a first camera 2, a second camera 3, a passenger flow counting device based on the flight attendant work light, and a housing 4. The work light 1, the first camera 2, and the second camera 3 are respectively connected to the passenger flow counting device based on the flight attendant work light. Specifically: The work light 1 is used to illuminate the inspection area; The first camera 2 is used to acquire first image information of the detection area; The second camera 3 is used to acquire second image information of the detection area; The passenger flow counting device based on the flight attendant work light includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the above-mentioned passenger flow counting method based on the flight attendant work light. Housing 4 is used to enclose the first camera, the second camera, the work light, and the passenger flow counting device.

[0052] During operation, the working lights need to be turned on first to provide a light source for the detection area. Then, the first image information and the second image information of the detection area are collected simultaneously by the first camera 2 and the second camera 3 respectively. Then, the passenger flow counting device realizes the number of target passengers and crew members based on the passenger flow counting method.

[0053] Therefore, the flight attendant work light of the present invention can be applied to aviation, railway or other transportation fields, which can improve the identification accuracy of target passengers, and collect data of target passengers and / or flight attendants in real time, effectively avoiding the false increase in passenger flow caused by flight attendants walking at the cabin door.

[0054] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A passenger flow counting method based on flight attendant work lights, characterized in that, include: The indicator light in the drive detection area is turned on; Acquire first image information of the detection area captured by the first camera and second image information of the detection area captured by the second camera; Calculate the disparity information of the target object based on the first image information and the second image information; A depth map is generated using triangulation based on the parallax information. Human feature information is identified based on the depth map to extract the target person; The target individuals are tracked to identify target passengers from among them; Count the number of the target passengers.

2. The passenger flow counting method based on flight attendant work lights as described in claim 1, characterized in that, The step of tracking the target individuals to identify target passengers includes: Assign temporary tracking identifiers to the target personnel; The behavior sequence and dwell time of the temporary tracking identifier are continuously collected. The behavior sequence is used to record the coordinate information of key points of the target person. The key points include head feature points and torso feature points. The torso feature points include shoulder feature points and hip feature points. Based on the behavioral sequence and dwell time, it is determined whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior. If the determination is yes, then the temporary tracking identifier will be marked as "flight attendant". If the determination is negative, the temporary tracking identifier will be marked as "target passenger".

3. The passenger flow counting method based on flight attendant work lights as described in claim 2, characterized in that, The flight attendant's behavior includes nodding, bending over, and prolonged stays.

4. The passenger flow counting method based on flight attendant work lights as described in claim 2 or 3, characterized in that, The step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior based on the behavioral sequence and dwell time includes: If the head feature point fluctuates at least a preset number of times within a preset first time period, then the behavioral characteristics of the temporary tracking identifier match the nodding behavior. The fluctuation refers to the situation where the change in the Y coordinate of the head feature point is greater than or equal to a preset baseline change range and the change frequency is within a preset nodding frequency range.

5. The passenger flow counting method based on flight attendant work lights as described in claim 2 or 3, characterized in that, The step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior based on the behavioral sequence and dwell time includes: If the torso feature point moves and the duration of the movement is greater than or equal to a preset second duration, then the behavioral characteristics of the temporary tracking marker match the bending behavior. The movement refers to the situation where the angle between the line connecting the shoulder feature point and the hip feature point and the vertical direction is greater than or equal to a preset reference angle, or the descent of the Y coordinate of the hip feature point is greater than or equal to a preset reference descent.

6. The passenger flow counting method based on flight attendant work lights as described in claim 2 or 3, characterized in that, The step of determining whether the behavioral characteristics of the temporary tracking identifier match the preset crew behavior based on the behavioral sequence and dwell time includes: If the key point changes and the dwell time is greater than or equal to the preset third duration, then the behavioral characteristics of the temporary tracking identifier match the long-term dwell behavior. The changes include nodding behavior, bending over behavior, and specific displacements of key points.

7. The passenger flow counting method based on flight attendant work lights as described in claim 6, characterized in that, Also includes: If the key points remain unchanged and the dwell time is greater than or equal to the preset fourth duration, the temporary tracking identifier will be marked as "suspicious dwelling" and an alarm signal will be generated.

8. The passenger flow counting method based on flight attendant work lights as described in claim 2, characterized in that, Also includes: Send the coordinates of key points corresponding to temporary tracking identifiers marked "crew members" to external devices, and / or count the number of crew members.

9. A passenger flow counting device based on a flight attendant work light, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the passenger flow counting method based on the crew member work light as described in any one of claims 1 to 8.

10. A flight attendant work light, characterized in that, include: Work lights are used to illuminate the inspection area; A first camera is used to acquire first image information of the detection area; A second camera is used to acquire second image information of the detection area; The passenger flow counting device based on the flight attendant work light as described in claim 9; The housing is used to enclose the first camera, the second camera, the work light, and the passenger flow counting device.

Images