Elevator passenger identification method, device, storage medium and equipment

Abstract

The application discloses an elevator passenger identification method, device, storage medium and equipment, and the method comprises the following steps: based on a character tracking algorithm and a first camera in the elevator, detecting first body image information of each person in the elevator and recording first time information; based on a character detection and access identification algorithm and a plurality of second cameras in all corridors, detecting second body image information of all people who go out or enter, corresponding house numbers and second time information; based on a character relocation algorithm, associating the first time information and the second time information with a difference within a threshold range, and one-to-one matching the associated first body image information and the second body image information to obtain the house number corresponding to the first body image information. Through the application, accurate identification of all people in the elevator is realized, thus solving the problems of 'free riding' and non-rideable visitors; meanwhile, the application makes the elevator riding process of people more unnoticeable, thus providing a better experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular to an elevator passenger identification method and device, a storage medium and equipment. BACKGROUND

[0002] With the implementation of national policy and the pursuit of more convenient life, installing elevators becomes the choice of many old communities. The old community usually installs elevators in a way that residents partly pay for it. Therefore, how to accurately charge the elevator becomes a problem to be solved.

[0003] The prerequisite for accurate elevator charging is accurate identification of elevator passengers. The existing identification methods mainly include the following two methods:

[0004] The first method is card swiping identification. A card swiping machine is installed in the elevator or at the door of each floor of the elevator. An electronic card is also issued to each resident who can ride the elevator. The card contains the information of the cardholder. When the resident rides the elevator, the resident needs to swipe the card first. The card swiping machine records the resident's riding information corresponding to the card. The elevator can be started only after swiping the card.

[0005] The second method is face recognition technology. A face recognition machine is installed in the elevator or at the door of each floor of the elevator. All the people who can ride the elevator need to upload their face images and corresponding personal information in advance. When the resident rides the elevator, the resident needs to align the face recognition machine first. The face recognition machine will set the elevator to a rideable state only after recognizing the face of the rideable resident.

[0006] The above two methods have two problems. The first problem is 'free riding'. That is, when the first person swipes the card or faces the face, other people who accompany the first person can also ride the elevator. This causes the omission of elevator passenger identification. The second problem is that visitors (including relatives of residents or on-site service personnel) cannot ride the elevator. Because the card or the pre-recorded face is required to ride the elevator, the visitors can only ride the elevator with the residents or climb the stairs, which causes great inconvenience to the visitors. SUMMARY

[0007] In view of the above technical problems, the present application provides an elevator passenger identification method, device, storage medium and equipment. By using the cooperation of multiple cameras in the elevator and the corridor, it is accurately tracked that each person who rides the elevator enters or leaves which household. The accurate identification of all elevator passengers is realized. The accurate charging of elevator passengers is realized. The problems of 'free riding' and visitor identification in the prior art are solved.

[0008] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0009] According to an aspect of the present application, an elevator passenger identification method is provided, the identification method comprising:

[0010] detecting first body image information of each person in the elevator and recording first time information based on a person tracking algorithm and a first camera in the elevator;

[0011] detecting second body image information, corresponding house number and second time information of all people who get out or get in based on a person detection and access identification algorithm and a plurality of second cameras in all corridors;

[0012] associating the first time information and the second time information whose difference is within a threshold range based on a person relocation algorithm, and one-to-one matching the associated first body image information and the second body image information to obtain the house number corresponding to the first body image information.

[0013] Further, the detecting first body image information of each person in the elevator and recording first time information comprises:

[0014] performing person identification on the captured video stream;

[0015] recording the full body image and the start time of taking the elevator when the person is identified;

[0016] recording the end time of taking the elevator when the identified person is detected to leave the elevator, and the time information includes the start time of taking the elevator and the end time of taking the elevator.

[0017] Further, the detecting second body image information, corresponding house number and second time information of all people who get out or get in comprises:

[0018] before detection, marking all door positions and corresponding house numbers in all second camera pictures;

[0019] when the detected person in any second camera picture is close to or in contact with the marked door, capturing an image block containing the person and the door in the second camera picture;

[0020] inputting the image block into a pre-trained image binary classification network to confirm whether the detected person is getting in or getting out;

[0021] recording the full body image, current time and house number of the detected person confirmed to be getting in or getting out.

[0022] Further, the one-to-one matching the associated first body image information and the second body image information to obtain the house number corresponding to the first body image information comprises:

[0023] According to the first time information corresponding to the first body image information, all the second body image information corresponding to the difference value within the threshold range of the second time information is screened out;

[0024] The similarity of the screened second body image information and the first body image information is calculated, and the second body image information with the highest similarity and greater than a set threshold is selected to confirm the household number of the personnel currently matched with the first body image information.

[0025] Further, the calculation of the similarity of the screened second body image information and the first body image information includes:

[0026] The first body image information and the second body image information containing a human body are respectively input into a pre-trained deep neural network to obtain corresponding two feature vectors;

[0027] The similarity of the two feature vectors is the similarity of the first body image information and the second body image information, and the similarity of the two feature vectors is represented by the included angle cosine value or the distance between vectors.

[0028] According to a second aspect of the present disclosure, an elevator passenger identification device is provided, comprising:

[0029] A first identification module is configured to detect first body image information of each personnel in the elevator based on a person tracking algorithm and a first camera in the elevator, and record first time information;

[0030] A second identification module is configured to detect second body image information, corresponding household number and second time information of all personnel who go out or enter based on a person detection and entrance identification algorithm and a plurality of second cameras in all corridors;

[0031] An analysis module is configured to associate the first time information and the second time information within the threshold range based on a person relocation algorithm, and one-to-one match the associated first body image information and the second body image information to obtain the household number corresponding to the first body image information.

[0032] According to a third aspect of the present disclosure, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the elevator passenger identification method as described above.

[0033] According to a fourth aspect of the present disclosure, an elevator passenger identification device is provided, comprising: a processor; and a memory arranged to store computer executable instructions, which when executed cause the processor to perform the elevator passenger identification method as described above.

[0034] The technical solution of the present disclosure has the following beneficial effects:

[0035] Compared with the existing elevator person identification system based on card swiping or face recognition, the present application realizes accurate identification of all personnel in the elevator, thus solving the problem of 'free riding' and the problem of visitors being unable to ride; at the same time, the present application does not require the riding personnel to perform any additional operation, and the riding process of people is more unobtrusive, thus providing a better experience. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 A flowchart of an elevator passenger identification method in an embodiment of the present disclosure;

[0037] Figure 2 A structural block diagram of an elevator passenger identification device in an embodiment of the present disclosure;

[0038] Figure 3 A terminal device for implementing an elevator passenger identification method in an embodiment of the present disclosure;

[0039] Figure 4 A computer readable storage medium storing an elevator passenger identification method in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the example implementations to those skilled in the art. The described features, structures, or characteristics can be combined in one or more implementations. In the following description, numerous specific details are provided to give a thorough understanding of implementations of the disclosure. One skilled in the relevant art will recognize, however, that the implementations of the disclosure can be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures have not been described in detail so as not to obscure the aspects of the disclosure.

[0041] Furthermore, the accompanying drawings are only schematic and are non-limiting. Like references signs denote like parts throughout the drawings and the specific illustrations in the drawings are to be understood as not limiting the possibilities open to the skilled person. Some of the blocks in the drawings can be functional blocks that do not necessarily have a direct relationship to one another or to physical or logical components illustrated in the drawings. Any known or future developed methods, components, devices, etc. can be used in the implementations of the present disclosure.

[0042] The application provides a product elevator passenger identification method. Referring to Figure 1 Fig. 1 is a flowchart of a product elevator passenger identification method provided by an embodiment of the application. The method can be applied to electronic devices such as personal computers and servers. The method can be executed by a device, which can be implemented by software and / or hardware. The method can specifically include the following steps S101-S103:

[0043] In step S101, based on a person tracking algorithm and a first camera in the elevator, first body image information of each person in the elevator is detected, and first time information is recorded.

[0044] The person tracking algorithm can be a computer vision technology for tracking and locating a person in a video stream or image sequence. For example, a color feature-based tracking algorithm: tracking based on color features of a person in an image, such as background modeling and color distribution modeling; a moving target detection and tracking algorithm: combining target detection and motion analysis techniques to track by detecting the motion of a target, such as Kalman filter, particle filter, etc.; an appearance feature-based tracking algorithm: tracking using appearance features (such as texture, shape) of a person, such as mean shift, CAMShift, etc.; a deep learning method: using a deep neural network for person tracking, such as a target tracking method based on a convolutional neural network.

[0045] Based on the above algorithm, the video stream is subjected to person identification. When a person is identified, the full-body image and the start time of taking the elevator are recorded. When the identified person is detected to leave the elevator, the end time of taking the elevator is recorded. The time information includes the start time of taking the elevator and the end time of taking the elevator.

[0046] In step S102, based on a person detection and entrance / exit detection algorithm and a plurality of second cameras in all corridors, second body image information of all people who go out or enter, corresponding door numbers and second time information are detected.

[0047] Specifically, the person detection and entrance / exit detection algorithm can be a pedestrian detection algorithm for detecting pedestrians in the corridor, such as a feature-based method (such as Haar features, HOG features) and a deep learning-based method (such as a convolutional neural network); or a face detection algorithm for detecting the faces of elevator passengers and combining a face recognition algorithm to identify their identities, such as a traditional feature extraction method (such as LBP, PCA) or a deep learning method (such as a convolutional neural network); or a behavior recognition algorithm can be used to determine whether they enter or leave the elevator by analyzing the motion trajectories and behaviors of elevator passengers, and associate them with the corresponding door.

[0048] Based on the above algorithm, the person who is entering or leaving is recorded, and the detected image and time are retained.

[0049] It is worth noting that the two steps S101 and S102 can be performed synchronously in real time, and there is no distinction between the order.

[0050] In step S103, based on the person repositioning algorithm, the first time information and the second time information within the threshold range are associated, the associated first body image information and the second body image information are matched one by one, and the corresponding house number of the first body image information is obtained.

[0051] Wherein, after detecting that there is a person in the elevator, i.e. obtaining the first body image information, it is necessary to determine the house number corresponding to the first body image information in order to determine the subject of payment. Here, in order to cover the two cases of getting on the elevator and getting off the elevator, the first time information and the second time information within the threshold range are associated, and all the people in the elevator are matched with the people entering and leaving the house in the nearby time corridor. Demonstratively, when getting on the elevator, the second body image information before the first time information occurs is matched with the first body image information that needs to be confirmed; when getting on the elevator, the second body image information after the first time information occurs is matched with the first body image information that needs to be confirmed.

[0052] In matching, the person repositioning algorithm used can be to use traditional feature extraction algorithms such as SIFT (Scale-Invariant Feature Transform) or SURF (Speeded-Up Robust Features) to extract feature descriptors of the person in the image. Then, the similarity of the person in the image is measured by calculating the feature similarity (such as Euclidean distance, cosine similarity, etc.) between the two images; or it can be to use convolutional neural networks (CNN) or pre-trained models (such as VGG, ResNet, etc.) to extract high-level semantic feature representations from images, and then compare the similarity of the feature vectors (such as cosine similarity, Euclidean distance, etc.) to compare the similarity of the person; or it can use a face recognition algorithm, which can include face detection, key point positioning, face alignment and feature extraction steps, and calculate the similarity by comparing the face feature vectors of two persons; or it can be based on metric learning methods, such as learning a suitable distance metric or similarity metric so that similar people have a closer distance.

[0053] In an embodiment, step S102 can specifically include: before detection, marking all door positions in all second camera pictures and corresponding house numbers; when a detected person in any second camera picture is close to or in contact with a marked door, intercepting an image block containing the person and the door in the second camera picture; inputting the image block into a pre-trained image binary classification network to determine whether the detected person is entering or leaving the door; recording the full-body image of the detected person confirmed to be entering or leaving the door, the current time, and the house number.

[0054] The marking is only set once when the second camera is installed. The marking can form a square box. In subsequent acquisition of second body image information, when a person appears in the square box, the second body image information of the person is calculated, that is, a video frame is recorded and input into a pre-trained image binary classification network for analysis of the action to determine whether the person is entering or leaving the door.

[0055] In an embodiment, step S103 includes: according to the first time information corresponding to the first body image information, screening all second body image information corresponding to a difference value of the second time information within a threshold range; calculating the similarity between the screened second body image information and the first body image information, selecting the second body image information with the highest similarity and greater than a set threshold, and confirming the house number of the person matched by the first body image information.

[0056] As an explanation of the above, all corridor personnel in a similar time are preliminarily screened according to the start and end times of the elevator ride of the person, the similarity between the elevator personnel and the corridor personnel is calculated according to the first body image information and the second body image information by using a person repositioning algorithm, and the corridor personnel corresponding to the highest similarity and greater than a certain artificial threshold are selected as the corridor personnel to be matched. Thus, the house numbers of all elevator ride personnel can be obtained, that is, the accurate identification of the elevator ride personnel is realized.

[0057] As a supplement, the calculation of the similarity between the screened second body image information and the first body image information includes: inputting the first body image information and the second body image information containing the full body of a person into a pre-trained deep neural network to obtain corresponding two feature vectors; the similarity between the two feature vectors is the similarity between the first body image information and the second body image information, and the similarity between the two feature vectors is represented by the included angle cosine value or the distance between the vectors.

[0058] Based on the above elevator passenger identification method, after the personnel riding the elevator is associated with the house number, accurate elevator charging can be implemented, such as one room account corresponding to one elevator payment account. As long as the personnel riding the elevator is associated with the house number, the charge can be deducted in the elevator payment account corresponding to the account.

[0059] Based on the same idea, as shown in Figure 2 An elevator passenger identification device is provided, comprising:

[0060] The first identification module 201 is configured to detect the first body image information of each personnel in the elevator based on a person tracking algorithm and a first camera in the elevator, and record the first time information;

[0061] The second identification module 202 is configured to detect the second body image information, the corresponding house number and the second time information of all personnel going out or going in based on a person detection and entrance / exit identification algorithm and a plurality of second cameras in all corridors;

[0062] The analysis module 203 is configured to associate the first time information and the second time information within a threshold range based on a person relocation algorithm, and match the associated first body image information and the second body image information one by one to obtain the house number corresponding to the first body image information.

[0063] Compared with the existing elevator person identification system based on card swiping or face recognition, the elevator passenger identification device realizes accurate identification of all personnel in the elevator, thereby solving the problems of 'free riding' and non-rideable visitors. At the same time, the invention does not require the rider to perform any additional operations, and the riding process is more unobtrusive, thus providing a better experience.

[0064] The specific details of each module / unit in the above device have been described in detail in the method part embodiments, and the undisclosed details can be referred to the method part embodiments. Therefore, further description is omitted.

[0065] Based on the same idea, the present specification also provides an elevator passenger identification device, as shown in Figure 3 .

[0066] The elevator passenger identification device can be a terminal device or a server provided in the above embodiments.

[0067] The elevator passenger identification device can vary greatly in configuration or performance and can include one or more processors 301, memory 302, and buses, and the memory 302 can store one or more applications or data. The memory 302 can include a readable medium in the form of a volatile memory unit, such as a random access memory (RAM) and / or a cache memory unit, for example, a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, and the like, which can be equipped on an electronic device, and can further include a read-only memory unit. The applications stored in the memory 302 can include one or more program modules (not shown in the figure), and such program modules include, but are not limited to, an operating system, one or more applications, other program modules, and program data, each of which or some combination thereof can include the implementation of a network environment. Further, the processor 301 can be configured to communicate with the memory 302 to execute a series of computer-executable instructions in the memory 302. The elevator passenger identification device can also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more I / O interfaces (input / output interfaces) 305, one or more external devices 306 (such as a keyboard) for communication, and can also communicate with one or more devices that enable a user to interact with the device, and / or any device that enables the device to communicate with one or more other computing devices (such as a router, network switch, etc.). Such communication can be through the I / O interface 305. Furthermore, the device can also communicate with one or more networks (such as a local area network (LAN)) through the wired or wireless interface 304.

[0068] The processor 301 can be composed of an integrated circuit in some embodiments, for example, can be composed of a single packaged integrated circuit, or can be composed of a plurality of packaged integrated circuits with the same function or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 301 is the control core (Control Unit) of the electronic device, which connects various components of the entire electronic device through various interfaces and lines, and executes various functions of the processing device and processes data by running or executing programs or modules (such as product elevator passenger identification programs, etc.) stored in the memory 11 and calling data stored in the memory 11.

[0069] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 302 and at least one processor 301, etc.

[0070] The power supply can be logically connected to the at least one processor 301 via a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power sources, a recharging device, a power fault detection circuit, a power converter or inverter, a power status indicator, and other components. The electronic device 1 may also include various sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated further here.

[0071] Optionally, the processing device may further include a user interface, which may be a display. Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0072] Figure 3 Only elevator passenger identification devices with components are shown; those skilled in the art will understand that... Figure 3 The structure shown does not constitute a limitation on elevator passenger identification devices and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0073] Specifically, in this embodiment, the elevator passenger identification device includes a memory and one or more programs, wherein one or more programs are stored in the memory, and one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the elevator passenger identification device, and is configured to be executed by one or more processors. The one or more programs include computer-executable instructions for performing the following:

[0074] Detect first body image information of each person in the elevator based on a person tracking algorithm and a first camera in the elevator, and record first time information;

[0075] Detect second body image information, corresponding house number and second time information of all people who go out or enter based on a person detection and entrance recognition algorithm and a plurality of second cameras in all corridors;

[0076] Correlate the first time information and the second time information whose difference is within a threshold range based on a person repositioning algorithm, and match the first body image information and the second body image information one by one to obtain the house number corresponding to the first body image information.

[0077] Based on the same idea, the exemplary embodiments of the present application also provide a computer readable storage medium having a program product stored thereon, which can implement the above-mentioned method of the present specification. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program codes for causing a terminal device to perform the steps described in the above-mentioned “Exemplary Methods” section according to various exemplary embodiments of the present disclosure when the program product is run on the terminal device.

[0078] Reference Figure 4 As shown, a program 400 for implementing the above-mentioned method according to the exemplary embodiments of the present disclosure is described, which can adopt a portable compact disc read-only memory (CD-ROM) and include program codes, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program, which can be used by or in conjunction with an instruction execution system, apparatus or device.

[0079] The program product can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium may, for example, be but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more wires, a portable disc, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0080] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting programs for use by or in conjunction with an instruction execution system, apparatus, or device.

[0081] Program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, CSS, and HTML, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0082] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the method according to the exemplary embodiments of this disclosure.

[0083] Furthermore, the above figures are merely illustrative representations of the processes included in the methods according to exemplary embodiments of this disclosure, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0084] It should be noted that, although several modules or units of the devices for action execution are mentioned in the above detailed description, such division into modules or units is not mandatory. Indeed, according to exemplary embodiments of the present disclosure, features and functionalities of two or more of the above-described modules or units can be embodied in one module or unit. Conversely, features and functionalities of one of the above-described modules or units can be further divided in multiple modules or units.

[0085] Other embodiments of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the features of the present disclosure disclosed herein. It is intended that the present disclosure cover any and all variations of the present disclosure that come within the scope of the following claims and their equivalents. It is intended that the specification and examples be considered exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0086] It is to be understood that the present disclosure is not limited to the precise structures described herein and shown in the drawings, and that various modifications and changes can be made without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the claims appended hereto.

Claims

1. An elevator passenger identification method, characterized by, The identification method comprises: Based on the person tracking algorithm and the first camera in the elevator, the first body image information of each person in the elevator is detected, and the first time information is recorded, specifically comprising: performing person identification on the video stream captured; when a person is identified, record the full body image and the start time of taking the elevator; when the identified person is detected to leave the elevator, record the end time of taking the elevator, and the first time information comprises the start time of taking the elevator and the end time of taking the elevator; Based on the person detection and access identification algorithm and a plurality of second cameras in all corridors, the second body image information of all people going out or going in, the corresponding house number and the second time information are detected; Based on the person repositioning algorithm, the first time information and the second time information with a difference within a threshold range are associated, the first body image information and the second body image information are matched one by one, and the house number corresponding to the first body image information is obtained, specifically comprising: according to the first time information corresponding to the first body image information, all second body image information corresponding to the second time information with a difference within a threshold range is screened out; the similarity of the screened second body image information and the first body image information is calculated, and the second body image information with the highest similarity and greater than a set threshold is selected, and the house number of the person currently matched with the first body image information is confirmed.

2. The elevator passenger identification method according to claim 1, characterized by, The detection of the second body image information of all people going out or going in, the corresponding house number and the second time information comprises: Before detection, mark all door positions in all second camera pictures and the corresponding house numbers; When the detected person in any second camera picture is close to or in contact with the marked door, the image block containing the person and the door in the second camera picture is intercepted; The image block is input into a pre-trained image binary classification network to confirm whether the detected person is going in or going out; The full body image, the current time and the house number of the detected person confirmed to be going in or going out are recorded.

3. The elevator passenger identification method according to claim 1, characterized by, The calculation of the similarity of the screened second body image information and the first body image information comprises: The first body image information and the second body image information containing the full body of a person are respectively input into a pre-trained deep neural network to obtain corresponding two feature vectors; The similarity of the two feature vectors is the similarity of the first body image information and the second body image information, and the similarity of the two feature vectors is represented by the included angle cosine value or the distance between vectors.

4. An elevator passenger identification device, characterized by It comprises: A first identification module is configured to detect the first body image information of each person in the elevator based on the person tracking algorithm and the first camera in the elevator, and record the first time information, specifically comprising: performing person identification on the video stream captured; when a person is identified, record the full body image and the start time of taking the elevator; when the identified person is detected to leave the elevator, record the end time of taking the elevator, and the first time information comprises the start time of taking the elevator and the end time of taking the elevator; The second identification module is configured to detect second body image information, corresponding house numbers and second time information of all people who go out or enter in, based on a person detection and entrance / exit identification algorithm and a plurality of second cameras in all corridors. The analysis module is configured to associate the first time information and the second time information with a difference within a threshold range, one-to-one match the first body image information and the second body image information, and obtain the house number corresponding to the first body image information, specifically including: filtering all the second body image information corresponding to the second time information with a difference within a threshold range according to the first time information corresponding to the first body image information; calculating the similarity of the filtered second body image information and the first body image information, selecting the second body image information with the highest similarity and greater than a set threshold, and confirming the house number of the person whose first body image information is currently matched.

5. An elevator passenger identification device, characterized by The device includes: a processor; and a memory arranged to store computer-executable instructions that, when executed, cause the processor to: detect first body image information of each person in the elevator and record first time information based on a person tracking algorithm and a first camera in the elevator, specifically including: performing person identification on the video stream captured; recording the full body image and the start time of taking the elevator when a person is identified; recording the end time of taking the elevator when the identified person is detected to leave the elevator, the first time information including the start time of taking the elevator and the end time of taking the elevator; detect second body image information, corresponding house numbers and second time information of all people who go out or enter in, based on a person detection and entrance / exit identification algorithm and a plurality of second cameras in all corridors; associate the first time information and the second time information with a difference within a threshold range, one-to-one match the first body image information and the second body image information, and obtain the house number corresponding to the first body image information based on a person relocation algorithm, specifically including: filtering all the second body image information corresponding to the second time information with a difference within a threshold range according to the first time information corresponding to the first body image information; calculating the similarity of the filtered second body image information and the first body image information, selecting the second body image information with the highest similarity and greater than a set threshold, and confirming the house number of the person whose first body image information is currently matched.

6. A computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the elevator passenger identification method according to any one of claims 1-3.

Images