Elevator access control security method and system based on image recognition model

Abstract

The present invention provides an elevator access control security method and system based on an image recognition model, the method comprising: when a user is detected entering the interior of the elevator, a front full-body image corresponding to the current user is collected in real time by a camera device, and a face image in the front full-body image is extracted by a PP-ShiTu image recognition model; a number of target facial features contained in the face image are detected, and a corresponding target image is found in a preset image database according to the target facial features; it is determined whether the target image is consistent with the face image; if so, the current user is determined to be a resident, and the corresponding target user information is found according to the target image; a corresponding control code is generated according to the target user information, and the control code is sent to the mobile terminal of the current user, so that the user can activate the elevator through the control code. The present invention can avoid the occurrence of recognition errors, and correspondingly improves the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an elevator access control security method and system based on an image recognition model. Background Art

[0002] With the progress of the times and the rapid development of productivity, elevators are installed in existing office buildings, shopping malls, high-end hotels and other buildings to enable people to quickly reach the required floor and provide convenience for people.

[0003] Among them, existing high-end hotels, high-end office buildings and high-end communities usually install face recognition systems inside existing elevators in order to strengthen the management of people entering the elevators and improve security effects. Only after the face recognition is passed can the user enable the floor button and reach the corresponding floor.

[0004] However, the operating logic of the face recognition system installed in the elevators of existing high-end hotels, high-end office buildings or high-end residential areas is relatively simple, that is, the image collected by the camera is directly compared with the image stored in the image database, and after finding the consistent image, the user will be prompted to activate the elevator. However, in this way, during the recognition process, when the user directly uses the photo of the user that has been entered into the face recognition system, the face recognition can also be passed and the elevator can be activated. Therefore, the existing technology has the problem of low recognition accuracy, which leads to certain safety hazards. Summary of the invention

[0005] Based on this, the purpose of the present invention is to provide an elevator access control security method and system based on an image recognition model to solve the problem that the elevator security system in the prior art has certain safety hazards.

[0006] The first aspect of the embodiment of the present invention proposes:

[0007] An elevator access control security method based on an image recognition model, wherein the method comprises:

[0008] When a user is detected entering the elevator, a front full-body image corresponding to the current user is collected in real time by a camera device, and a face image in the front full-body image is extracted by a PP-ShiTu image recognition model;

[0009] Detecting a number of target facial features contained in the face image, and searching for a corresponding target image in a preset image database according to the target facial features, wherein the target image is unique;

[0010] Determining whether the target image is consistent with the face image;

[0011] If it is determined in real time that the target image is consistent with the face image, the current user is determined to be a resident, and the corresponding target user information is searched out according to the target image;

[0012] A corresponding control code is generated according to the target user information, and the control code is sent to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time.

[0013] The beneficial effect of the present invention is that by collecting the front full-body image of the user in real time, it is possible to judge whether it is the user himself. Furthermore, based on the front full-body image, the corresponding face image is detected, and the corresponding target image is found in a preset image database according to the target facial features contained in the face image, so that it is possible to judge whether the current user is a resident user. Based on this, the corresponding control code is sent to the mobile terminal of the current user in real time according to the target user information of the current user, so that the current user can input the current control code to start the current elevator accordingly, thereby effectively avoiding the occurrence of recognition errors. At the same time, the recognition accuracy is high, which eliminates safety hazards and improves the user experience.

[0014] Preferably, the step of extracting the face image in the frontal full-body image by using the PP-ShiTu image recognition model comprises:

[0015] Calling out the PP-PicoDet algorithm in the PP-ShiTu image recognition model, and identifying the human body area and the background area in the frontal full-body image by the PP-PicoDet algorithm;

[0016] Removing the background area and performing image enhancement processing on the human body area to extract a corresponding human body image;

[0017] Performing a full scan on the human body image to obtain pixel values ​​corresponding to a plurality of pixel points in the human body image, and detecting whether each pixel value is within a preset threshold value one by one;

[0018] If it is detected that some of the pixel values ​​are within the preset threshold, it is determined that the area where the pixel points corresponding to the some of the pixel values ​​are located is the face image.

[0019] Preferably, the step of determining that the region where the pixel points corresponding to the plurality of pixel values ​​are located is the face image comprises:

[0020] Screening out a number of target pixel points whose pixel values ​​are within the preset threshold, and highlighting the edges of the target pixel points to identify the target areas covered by the target pixel points;

[0021] The edge of the target area is identified by a preset edge detection algorithm, and corresponding contour lines are added at the edge to generate the face image.

[0022] Preferably, the step of generating a corresponding control code according to the target user information includes:

[0023] When the target image is acquired, the target user information is searched in the user registration database according to the target image, and the user identity information and check-in time information contained in the target user information are extracted;

[0024] Extracting the letter information contained in the user identity information, and simultaneously extracting the numeric information contained in the check-in time information;

[0025] The control code is generated according to the letter information and the number information.

[0026] Preferably, the step of extracting letter information contained in the user identity information and simultaneously extracting digital information contained in the check-in time information comprises:

[0027] Extracting the name column corresponding to the user contained in the user identity information, and extracting the full pinyin name corresponding to the user's name in the name column;

[0028] Performing forward maximum step length word segmentation processing on the full pinyin name to generate a number of corresponding letter units, wherein the letter units constitute the letter information;

[0029] The check-in time included in the check-in time information is subjected to reverse maximum step length word segmentation processing to generate a plurality of corresponding digital units, and the plurality of digital units constitute the digital information.

[0030] Preferably, the step of generating the control code according to the letter information and the digital information includes:

[0031] Extracting a corresponding target letter unit from the letter information according to a first preset weight, and extracting a corresponding target number unit from the number information according to a second preset weight;

[0032] The target letter units and the target number units are randomly arranged to generate a number of corresponding initial serial numbers, and a target serial number is selected from the initial serial numbers to use the target serial number as the start password of the elevator, and the control code is generated based on the start password.

[0033] Preferably, the step of generating the control code based on the startup password includes:

[0034] When the start password is obtained, the floor information corresponding to the current user is extracted from the target user information, and the corresponding floor value is extracted from the floor information;

[0035] A mapping relationship between the start password and the floor value is established to generate the control code.

[0036] When it is detected that the elevator does not enable the corresponding floor through the control code, the target letter unit and the target number unit are randomly arranged again to generate a number of corresponding secondary sequence numbers;

[0037] A target sequence number is reselected from the plurality of secondary sequence numbers, and the reselected target sequence number is sent to the elevator, so that the elevator can re-activate the corresponding floor.

[0038] A second aspect of an embodiment of the present invention provides an elevator access control security system based on an image recognition model, wherein the system includes:

[0039] The acquisition module is used to acquire a front full-body image corresponding to the current user in real time through a camera device when a user is detected entering the elevator, and extract a face image from the front full-body image through a PP-ShiTu image recognition model;

[0040] A search module, used to detect a number of target facial features contained in the face image, and search for a corresponding target image in a preset image database according to the target facial features, wherein the target image is unique;

[0041] A judgment module, used to judge whether the target image is consistent with the face image;

[0042] A processing module, for determining that the current user is a resident if it is determined in real time that the target image is consistent with the face image, and finding the corresponding target user information according to the target image;

[0043] The sending module is used to generate a corresponding control code according to the target user information, and send the control code to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time.

[0044] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the acquisition module is specifically used for:

[0045] Calling out the PP-PicoDet algorithm in the PP-ShiTu image recognition model, and identifying the human body area and the background area in the frontal full-body image by the PP-PicoDet algorithm;

[0046] Removing the background area and performing image enhancement processing on the human body area to extract a corresponding human body image;

[0047] Performing a full scan on the human body image to obtain pixel values ​​corresponding to a plurality of pixel points in the human body image, and detecting whether each pixel value is within a preset threshold value one by one;

[0048] If it is detected that some of the pixel values ​​are within the preset threshold, it is determined that the area where the pixel points corresponding to the some of the pixel values ​​are located is the face image.

[0049] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the search module is specifically used for:

[0050] Screening out a number of target pixel points whose pixel values ​​are within the preset threshold, and highlighting the edges of the target pixel points to identify the target areas covered by the target pixel points;

[0051] The edge of the target area is identified by a preset edge detection algorithm, and corresponding contour lines are added at the edge to generate the face image.

[0052] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is specifically used for:

[0053] When the target image is acquired, the target user information is searched in the user registration database according to the target image, and the user identity information and check-in time information contained in the target user information are extracted;

[0054] Extracting the letter information contained in the user identity information, and simultaneously extracting the numeric information contained in the check-in time information;

[0055] The control code is generated according to the letter information and the number information.

[0056] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is also specifically used for:

[0057] Extracting the name column corresponding to the user contained in the user identity information, and extracting the full pinyin name corresponding to the user's name in the name column;

[0058] Performing forward maximum step length word segmentation processing on the full pinyin name to generate a number of corresponding letter units, wherein the letter units constitute the letter information;

[0059] The check-in time included in the check-in time information is subjected to reverse maximum step length word segmentation processing to generate a plurality of corresponding digital units, and the plurality of digital units constitute the digital information.

[0060] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is also specifically used for:

[0061] Extracting a corresponding target letter unit from the letter information according to a first preset weight, and extracting a corresponding target number unit from the number information according to a second preset weight;

[0062] The target letter units and the target number units are randomly arranged to generate a number of corresponding initial serial numbers, and a target serial number is selected from the initial serial numbers to use the target serial number as the start password of the elevator, and the control code is generated based on the start password.

[0063] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the elevator access control security system based on the image recognition model also includes a detection module, and the detection module is specifically used for:

[0064] When it is detected that the elevator does not enable the corresponding floor through the control code, the target letter unit and the target number unit are randomly arranged again to generate a number of corresponding secondary sequence numbers;

[0065] A target sequence number is reselected from the plurality of secondary sequence numbers, and the reselected target sequence number is sent to the elevator, so that the elevator can re-activate the corresponding floor.

[0066] The third aspect of the embodiment of the present invention proposes:

[0067] A computer comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the elevator access control and security method based on the image recognition model as described above is implemented.

[0068] The fourth aspect of the embodiments of the present invention proposes:

[0069] A readable storage medium stores a computer program, wherein when the program is executed by a processor, the elevator access control security method based on the image recognition model as described above is implemented.

[0070] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A flowchart of an elevator access control and security method based on an image recognition model provided in the first embodiment of the present invention;

[0072] Figure 2 A schematic diagram of a simulated user in an elevator access control and security method based on an image recognition model provided in a second embodiment of the present invention;

[0073] Figure 3 This is a structural block diagram of an elevator access control security system based on an image recognition model provided in the third embodiment of the present invention.

[0074] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0075] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0076] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0077] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0078] See also Figure 1 , shown is the elevator access control security method based on the image recognition model provided by the first embodiment of the present invention. The elevator access control security method based on the image recognition model provided by this embodiment can avoid the occurrence of recognition errors, and at the same time, the recognition accuracy is high, which eliminates safety hazards and improves the user experience.

[0079] Specifically, the elevator access control security method based on the image recognition model provided in this embodiment includes the following steps:

[0080] Step S10, when it is detected that a user enters the elevator, a front full-body image corresponding to the current user is collected in real time by a camera device, and a face image in the front full-body image is extracted by a PP-ShiTu image recognition model;

[0081] Specifically, in this embodiment, it should be noted that the intelligent elevator access control security method provided in this embodiment is implemented based on the PP-ShiTu image recognition model, wherein the PP-ShiTi image recognition model mainly implements two functions, one is subject detection, and the other is feature extraction, so that the corresponding target can be effectively detected in the acquired image to complete the subsequent judgment process. Specifically, the above-mentioned subject detection can effectively distinguish the target object from the background, and can correspondingly remove interference features. More specifically, the target detection model used in the process of subject detection is the PP-PicoDet algorithm, so that the detection of the target object can be effectively completed. In addition, the above-mentioned feature extraction adopts the CPU-level lightweight backbone network PP-LCNet, and combines the metric learning arcmargin algorithm, so that the required features can be effectively extracted.

[0082] Based on this, in this step, it should be noted that when the pressure sensor detects that a user has entered the elevator, the camera device pre-installed in the current elevator is immediately used to capture the frontal full-body image of the user entering the elevator in real time. On this basis, the face image in the current frontal full-body image is extracted through the above-mentioned PP-ShiTu image recognition model. Specifically, the face image corresponds to the face of the current user.

[0083] Step S20, detecting a plurality of target facial features contained in the face image, and searching for a corresponding target image in a preset image database according to the plurality of target facial features, wherein the target image is unique;

[0084] Furthermore, in this step, it should be noted that after the required facial image is detected through the above steps, in order to accurately identify which registered user the current facial image corresponds to, this step will further detect in real time several target facial features contained in the current facial image. Specifically, the several target facial features may include facial organs such as eyes, nose, mouth, eyebrows and ears in the current facial image.

[0085] Based on this, the corresponding target image is immediately found in the preset image database according to the current target facial features. Specifically, the target image is unique. More specifically, the target image is the image taken when the current user checks in.

[0086] Step S30, determining whether the target image is consistent with the face image;

[0087] Furthermore, in this step, it should be noted that after the required target image is acquired through the above steps, this step will determine in real time whether the current target image is consistent with the above real-time acquired face image.

[0088] Specifically, this step will specifically determine whether the facial features in the current target image are consistent with the facial features in the above face image.

[0089] Step S40: if it is determined in real time that the target image is consistent with the face image, the current user is determined to be a resident, and the corresponding target user information is searched out according to the target image;

[0090] Furthermore, in this step, it should be noted that if this step determines in real time that the target image is consistent with the facial image, it can be correspondingly determined that the current user is a check-in person who has checked in. It should be pointed out that the check-in personnel will register the user information at the front desk in advance for subsequent management.

[0091] More specifically, the user information of each user will include the facial image corresponding to each user, based on which the corresponding target user information can be reversely searched according to the target image found in real time, so as to facilitate subsequent processing.

[0092] Step S50, generating a corresponding control code according to the target user information, and sending the control code to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time.

[0093] Finally, in this step, it should be noted that after obtaining the required target user information through the above steps, this step can further generate a corresponding control code based on the current target user information, and at the same time, send the target serial number to the user's mobile terminal in real time.

[0094] Furthermore, after receiving the target serial number, the user only needs to input the target serial number into a pre-set input box in the current elevator to enable the current elevator. Specifically, the control code is one-time, thereby effectively eliminating safety hazards.

[0095] When in use, by collecting the user's frontal full-body image in real time, it can be judged whether it is the user himself. Furthermore, the corresponding facial image is detected based on the frontal full-body image, and the corresponding target image is found in the preset image database according to the target facial features contained in the facial image, so that it can be judged whether the current user is a checked-in user. Based on this, the corresponding control code is sent to the current user's mobile terminal in real time according to the current user's target user information, so that the current user can input the current control code to start the current elevator, thereby effectively avoiding the occurrence of recognition errors. At the same time, the recognition accuracy is high, which eliminates safety hazards and improves the user experience.

[0096] It should be noted that the above implementation process is only to illustrate the feasibility of the present application, but this does not mean that the smart elevator access control and security method of the present application has only the above-mentioned implementation process. On the contrary, as long as the smart elevator access control and security method of the present application can be implemented, it can be included in the feasible implementation plan of the present application.

[0097] In summary, the intelligent elevator access control security method provided by the above embodiments of the present invention can avoid the occurrence of recognition errors, and at the same time has high recognition accuracy, which eliminates safety hazards and improves the user experience.

[0098] See also Figure 2 The second embodiment of the present invention also provides an intelligent elevator access control and security method. The difference between the intelligent elevator access control and security method provided by this embodiment and the intelligent elevator access control and security method provided by the first embodiment is that:

[0099] Specifically, in this embodiment, it should be noted that the step of extracting the face image in the frontal full-body image by using the PP-ShiTu image recognition model includes:

[0100] Calling out the PP-PicoDet algorithm in the PP-ShiTu image recognition model, and identifying the human body area and the background area in the frontal full-body image by the PP-PicoDet algorithm;

[0101] Removing the background area and performing image enhancement processing on the human body area to extract a corresponding human body image;

[0102] Performing a full scan on the human body image to obtain pixel values ​​corresponding to a plurality of pixel points in the human body image, and detecting whether each pixel value is within a preset threshold value one by one;

[0103] If it is detected that some of the pixel values ​​are within the preset threshold, it is determined that the area where the pixel points corresponding to the some of the pixel values ​​are located is the face image.

[0104] Specifically, in this embodiment, it should be noted that when the above-mentioned full-body frontal image is obtained, the above-mentioned PP-ShiTu image recognition model will be called out in the algorithm database, and further, the PP-PicoDet algorithm in the current PP-ShiTu image recognition model will be called out. At the same time, the human body area and background area in the current full-body frontal image will be further identified by the current PP-PicoDet algorithm. Based on this, the background area that will cause image recognition interference can be effectively removed. Furthermore, in order to improve the accuracy of subsequent recognition, this embodiment will also perform image enhancement processing such as wave reduction and noise filtering on the currently screened human body area, so that the required human body image can be extracted accordingly.

[0105] After obtaining the human body image through the above steps, this embodiment will also perform a full scan of the current human body image to scan out all the pixel points contained in the current human body image. At the same time, the pixel values ​​corresponding to the current pixel points are synchronously obtained. Based on this, each current pixel value is detected one by one to see if it is within the preset threshold. Specifically, if it is detected that the current pixel values ​​are within the preset threshold, it can be accurately determined that the area where the pixel points corresponding to the current pixel values ​​are located is a face image. Correspondingly, if it is detected that the current pixel values ​​are not within the preset threshold, it can be accurately determined that the area where the current pixel values ​​exist is an interference area.

[0106] Specifically, in this embodiment, it is also necessary to explain:

[0107] The step of determining that the area where the pixel points corresponding to the plurality of pixel values ​​are located is the face image comprises:

[0108] Screening out a number of target pixel points whose pixel values ​​are within the preset threshold, and highlighting the edges of the target pixel points to identify the target areas covered by the target pixel points;

[0109] The edge of the target area is identified by a preset edge detection algorithm, and corresponding contour lines are added at the edge to generate the face image.

[0110] The above method can accurately screen out the required target pixel points, and at the same time, can perform corresponding highlighting, so that the target area covered by the current target pixel points can be clearly displayed in the current image.

[0111] Furthermore, the edge of the current target area is identified by using an existing edge detection algorithm, and contour lines are added at the edge accordingly, so that the required face image can be finally displayed.

[0112] In addition, in this embodiment, it should be noted that the above step of generating a corresponding control code according to the target user information includes:

[0113] When the target image is acquired, the target user information is searched in the user registration database according to the target image, and the user identity information and check-in time information contained in the target user information are extracted;

[0114] Extracting the letter information contained in the user identity information, and simultaneously extracting the numeric information contained in the check-in time information;

[0115] The control code is generated according to the letter information and the number information.

[0116] In addition, in this embodiment, it should be noted that in order to allow only users who have checked in to use the elevator, outsiders are excluded to eliminate the safety risks of the users who have checked in. This embodiment also generates a corresponding target sequence number in real time and allows users to use the elevator through the target sequence number.

[0117] Specifically, after obtaining the required target image through the above steps, this step will further search for the target user information containing the current target image in the pre-set user registration library according to the current target image, and at the same time, extract the user identity information and check-in time information contained in the current target user information. Specifically, for example, the user identity information can be identity information such as "name Zhang San (zhang san), age 25, gender male and ID number", and correspondingly, for example, the check-in time information can be "2023-05-01 10:36", based on this, this embodiment can further extract the letter information contained in the current user identity information, and the digital information contained in the current check-in time information. On this basis, the above control code can be randomly generated according to the current letter information and the current digital information, and invented to the user for use.

[0118] In addition, in this embodiment, it should be noted that the steps of extracting the letter information contained in the user identity information and simultaneously extracting the numeric information contained in the check-in time information include:

[0119] Extracting the name column corresponding to the user contained in the user identity information, and extracting the full pinyin name corresponding to the user's name in the name column;

[0120] Performing forward maximum step length word segmentation processing on the full pinyin name to generate a number of corresponding letter units, wherein the letter units constitute the letter information;

[0121] The check-in time included in the check-in time information is subjected to reverse maximum step length word segmentation processing to generate a plurality of corresponding digital units, and the plurality of digital units constitute the digital information.

[0122] In addition, in the present embodiment, it is also necessary to explain that in order to facilitate the extraction of information and to correspondingly improve the data processing efficiency, the present embodiment will specifically extract the name column corresponding to the current user contained in the above-mentioned user identity information, such as the above-mentioned "Zhang San", and further extract the pinyin full name corresponding to the name of the current user in the current name column, such as the above-mentioned "zhang san", wherein it can be understood that the current pinyin full name is composed of several English letters, based on this, the present embodiment will further perform forward maximum step size word segmentation processing on the current pinyin full name to generate several corresponding letter units, such as "z / h / a / n / g / s / a / n", and the current several letter units constitute the above-mentioned letter information.

[0123] Furthermore, this embodiment also performs reverse maximum step segmentation processing on the check-in time contained in the above check-in time information to generate a number of corresponding digital units, such as "2 / 0 / 1 / 3 / year / 0 / 5 / month / 0 / 1 day 1 / 0 o'clock / 3 / 0 / minute", and the current number of digital units constitute the above digital information.

[0124] In this embodiment, it should be noted that the step of generating the control code according to the letter information and the digital information includes:

[0125] Extracting a corresponding target letter unit from the letter information according to a first preset weight, and extracting a corresponding target number unit from the number information according to a second preset weight;

[0126] The target letter units and the target number units are randomly arranged to generate a number of corresponding initial serial numbers, and a target serial number is selected from the initial serial numbers to use the target serial number as the start password of the elevator, and the control code is generated based on the start password.

[0127] Among them, in this embodiment, it should be pointed out that after the required letter information and digital information are respectively obtained through the above steps, in order to randomly generate the above control code, the corresponding target letter unit will be extracted from the above letter information according to the first preset weight, and at the same time, the corresponding target digital unit will be extracted from the above digital information according to the second preset weight. Preferably, the above first preset weight and the above second preset weight are both 50%, and the control code provided in this embodiment consists of six digits, of which letters occupy three digits and digits occupy three digits, so as to perform corresponding permutations and combinations.

[0128] Based on this, the current target letter unit and the target number unit are randomly arranged through the existing random arrangement algorithm, so that several initial sequence numbers can be generated accordingly. Further, only one target sequence number needs to be selected from the current several initial sequence numbers, and the current target sequence number can be used as the above-mentioned start password. Specifically, the start password contains three numbers and three letters, so that the current user can activate the current elevator through the start password. For example, the selected target letter units are "z, h and s", and the selected target number units are "5, 3 and 6", so the selected start password can be "z5s3s6".

[0129] In this embodiment, it should be noted that the method further includes:

[0130] When it is detected that the elevator does not enable the corresponding floor through the control code, the target letter unit and the target number unit are randomly arranged again to generate a number of corresponding secondary sequence numbers;

[0131] A target sequence number is reselected from the plurality of secondary sequence numbers, and the reselected target sequence number is sent to the elevator, so that the elevator can re-activate the corresponding floor.

[0132] In this embodiment, it should be noted that in actual application, the user may fail to input the received target sequence number within the preset time due to negligence, so it is necessary to randomly arrange the target letter unit and the target number unit again through the above steps to generate a number of corresponding secondary sequence numbers. Further, a target sequence number is re-selected from the current secondary sequence number and sent to the elevator again to finally deliver the user to the required floor.

[0133] Among them, in this embodiment, it is also necessary to point out that:

[0134] The step of generating the control code based on the startup password comprises:

[0135] When the start password is obtained, the floor information corresponding to the current user is extracted from the target user information, and the corresponding floor value is extracted from the floor information;

[0136] A mapping relationship between the start password and the floor value is established to generate the control code.

[0137] The above method can be used to extract the floor information in the target user information in real time, that is, the check-in floor that the user has checked in beforehand. Based on this, the specific floor value is extracted from the current floor information. At the same time, a mapping relationship between the two and the above-mentioned start password is established. On this basis, when the elevator is enabled, the current elevator can enable the required floor button, so that the user does not need to perform manual operation, which correspondingly improves the user experience.

[0138] In this embodiment, it should also be pointed out that the method further includes:

[0139] When a plurality of users are detected entering the interior of the elevator, the front panoramic images corresponding to the plurality of users are collected in real time by the camera device, and the front panoramic images are grayed to generate corresponding gray images;

[0140] Detecting the grayscale values ​​corresponding to each pixel in the grayscale image one by one, and determining whether each grayscale value is within a preset grayscale threshold;

[0141] If it is determined that there is a grayscale value within the preset grayscale threshold, the pixel corresponding to the current grayscale value is determined to be a target pixel, and a target area where several target pixels exist is identified, wherein the target area includes face areas corresponding to the multiple users respectively;

[0142] Searching for a corresponding target face image in the face area according to the target user information, and determining that the target face image is consistent with the target image in the database;

[0143] If it is determined that the target face image is consistent with the target image in the database, the control code is sent to the elevator, so that the elevator activates the corresponding floor through the control code.

[0144] Among them, in this embodiment, it is also necessary to point out that when multiple users enter the elevator at the same time, in order to accurately find the user who has checked in among the current users, it is necessary to first collect the panoramic image inside the current elevator and perform the corresponding grayscale processing. Based on this, the corresponding grayscale value is detected, and each grayscale value is judged one by one whether it is within the preset grayscale threshold. Specifically, if yes, it means that the pixel point corresponding to the current grayscale value is the required target pixel point, and it can be further determined that the area where the current target pixel points exist is the face area of ​​the current users. On this basis, according to the user face image that has been stored in the above target user information, the required target face image is correspondingly found in the current face area, and it is judged in real time whether the current target face image is consistent with the target image stored in the database. Specifically, if yes, it means that the current target face image is the face image of the user who has checked in. Based on this, the control code generated in real time can be sent to the current elevator, so that the current elevator can enable the corresponding floor through the control code.

[0145] It should be pointed out that the implementation principle and some technical effects of the method provided in the second embodiment of the present invention are the same as those of the first embodiment. For the sake of brief description, for matters not mentioned in this embodiment, reference may be made to the corresponding contents provided in the first embodiment.

[0146] In summary, the intelligent elevator access control security system provided by the above embodiments of the present invention can avoid the occurrence of recognition errors, and at the same time has high recognition accuracy, which eliminates safety hazards and improves the user experience.

[0147] See also Figure 3 , which shows an elevator access control security system based on an image recognition model provided by a third embodiment of the present invention, wherein the system comprises:

[0148] The acquisition module is used to acquire a front full-body image corresponding to the current user in real time through a camera device when a user is detected entering the elevator, and extract a face image from the front full-body image through a PP-ShiTu image recognition model;

[0149] A search module, used to detect a number of target facial features contained in the face image, and search for a corresponding target image in a preset image database according to the target facial features, wherein the target image is unique;

[0150] A judgment module, used to judge whether the target image is consistent with the face image;

[0151] A processing module, for determining that the current user is a resident if it is determined in real time that the target image is consistent with the face image, and finding the corresponding target user information according to the target image;

[0152] The sending module is used to generate a corresponding control code according to the target user information, and send the control code to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time.

[0153] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the acquisition module is specifically used for:

[0154] Calling out the PP-PicoDet algorithm in the PP-ShiTu image recognition model, and identifying the human body area and the background area in the frontal full-body image by the PP-PicoDet algorithm;

[0155] Removing the background area and performing image enhancement processing on the human body area to extract a corresponding human body image;

[0156] Performing a full scan on the human body image to obtain pixel values ​​corresponding to a plurality of pixel points in the human body image, and detecting whether each pixel value is within a preset threshold value one by one;

[0157] If it is detected that some of the pixel values ​​are within the preset threshold, it is determined that the area where the pixel points corresponding to the some of the pixel values ​​are located is the face image.

[0158] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the search module is specifically used for:

[0159] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is specifically used for:

[0160] When the target image is acquired, the target user information is searched in the user registration database according to the target image, and the user identity information and check-in time information contained in the target user information are extracted;

[0161] Extracting the letter information contained in the user identity information, and simultaneously extracting the numeric information contained in the check-in time information;

[0162] The control code is generated according to the letter information and the number information.

[0163] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is also specifically used for:

[0164] Extracting the name column corresponding to the user contained in the user identity information, and extracting the full pinyin name corresponding to the user's name in the name column;

[0165] Performing forward maximum step length word segmentation processing on the full pinyin name to generate a number of corresponding letter units, wherein the letter units constitute the letter information;

[0166] The check-in time included in the check-in time information is subjected to reverse maximum step length word segmentation processing to generate a plurality of corresponding digital units, and the plurality of digital units constitute the digital information.

[0167] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the sending module is also specifically used for:

[0168] Extracting a corresponding target letter unit from the letter information according to a first preset weight, and extracting a corresponding target number unit from the number information according to a second preset weight;

[0169] The target letter units and the target number units are randomly arranged to generate a number of corresponding initial serial numbers, and a target serial number is selected from the initial serial numbers to use the target serial number as the start password of the elevator, and the control code is generated based on the start password.

[0170] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the elevator access control security system based on the image recognition model also includes a detection module, and the detection module is specifically used for:

[0171] When it is detected that the elevator does not enable the corresponding floor through the control code, the target letter unit and the target number unit are randomly arranged again to generate a number of corresponding secondary sequence numbers;

[0172] A target sequence number is reselected from the plurality of secondary sequence numbers, and the reselected target sequence number is sent to the elevator, so that the elevator can re-activate the corresponding floor.

[0173] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the elevator access control security system based on the image recognition model also includes a prompt module, and the prompt module is specifically used for:

[0174] Among them, in the above-mentioned elevator access control security system based on the image recognition model, the elevator access control security system based on the image recognition model also includes a detection module, and the detection module is specifically used for:

[0175] When a plurality of users are detected entering the interior of the elevator, the front panoramic images corresponding to the plurality of users are collected in real time by the camera device, and the front panoramic images are grayed to generate corresponding gray images;

[0176] Detecting the grayscale values ​​corresponding to each pixel in the grayscale image one by one, and determining whether each grayscale value is within a preset grayscale threshold;

[0177] If it is determined that there is a grayscale value within the preset grayscale threshold, the pixel corresponding to the current grayscale value is determined to be a target pixel, and a target area where several target pixels exist is identified, wherein the target area includes face areas corresponding to the multiple users respectively;

[0178] Searching for a corresponding target face image in the face area according to the target user information, and determining that the target face image is consistent with the target image in the database;

[0179] If it is determined that the target face image is consistent with the target image in the database, the control code is sent to the elevator, so that the elevator activates the corresponding floor through the control code.

[0180] A fourth embodiment of the present invention provides a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the elevator access control and security method based on the image recognition model provided in the above embodiment is implemented.

[0181] The fifth embodiment of the present invention provides a readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the elevator access control and security method based on the image recognition model provided in the above embodiment is implemented.

[0182] The present invention also has the following beneficial effects:

[0183] 1. By collecting the user's full-body frontal image, it can accurately determine whether the user is using the elevator, thus preventing others from using the elevator;

[0184] 2. By sending dynamic control codes to the elevator in real time, it can prevent the elevator from being used at will.

[0185] In summary, the intelligent elevator access control security method and system provided by the above embodiments of the present invention can avoid the occurrence of recognition errors, and at the same time, the recognition accuracy is high, which eliminates safety hazards and improves the user experience.

[0186] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0187] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0188] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0189] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0190] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0191] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An elevator access control security method based on an image recognition model, characterized in that: The method comprises: When a user is detected entering the elevator, a front full-body image corresponding to the current user is collected in real time by a camera device, and a face image in the front full-body image is extracted by a PP-ShiTu image recognition model; Detecting a number of target facial features contained in the face image, and searching for a corresponding target image in a preset image database according to the target facial features, wherein the target image is unique; Determining whether the target image is consistent with the face image; If it is determined in real time that the target image is consistent with the face image, the current user is determined to be a resident, and the corresponding target user information is searched out according to the target image; Generate a corresponding control code according to the target user information, and send the control code to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time; The step of generating a corresponding control code according to the target user information comprises: When the target image is acquired, the target user information is searched in the user registration database according to the target image, and the user identity information and check-in time information contained in the target user information are extracted; Extracting the letter information contained in the user identity information, and simultaneously extracting the numeric information contained in the check-in time information; Generate the control code according to the letter information and the digital information; The step of extracting the letter information contained in the user identity information and simultaneously extracting the digital information contained in the check-in time information includes: Extracting the name column corresponding to the user contained in the user identity information, and extracting the full pinyin name corresponding to the user's name in the name column; Performing forward maximum step length word segmentation processing on the full pinyin name to generate a number of corresponding letter units, wherein the letter units constitute the letter information; Performing reverse maximum step length word segmentation processing on the check-in time contained in the check-in time information to generate a number of corresponding digital units, wherein the number of digital units constitute the digital information; The step of generating the control code according to the letter information and the digital information comprises: Extracting a corresponding target letter unit from the letter information according to a first preset weight, and extracting a corresponding target number unit from the number information according to a second preset weight; The target letter unit and the target number unit are randomly arranged to generate a number of corresponding initial sequence numbers, and a target sequence number is selected from the initial sequence numbers to use the target sequence number as the start password of the elevator; When the start password is obtained, the floor information corresponding to the current user is extracted from the target user information, and the corresponding floor value is extracted from the floor information; Establishing a mapping relationship between the start password and the floor value to generate the control code; The method further comprises: When it is detected that the elevator does not enable the corresponding floor through the control code, the target letter unit and the target number unit are randomly arranged again to generate a number of corresponding secondary sequence numbers; A target sequence number is reselected from the plurality of secondary sequence numbers, and the reselected target sequence number is sent to the elevator, so that the elevator can re-activate the corresponding floor.

2. The elevator access control and security method based on the image recognition model according to claim 1 is characterized in that: The step of extracting the face image in the frontal full-body image by the PP-ShiTu image recognition model comprises: Calling out the PP-PicoDet algorithm in the PP-ShiTu image recognition model, and identifying the human body area and the background area in the frontal full-body image by the PP-PicoDet algorithm; Removing the background area and performing image enhancement processing on the human body area to extract a corresponding human body image; Performing a full scan on the human body image to obtain pixel values ​​corresponding to a plurality of pixel points in the human body image, and detecting whether each pixel value is within a preset threshold value one by one; If it is detected that some of the pixel values ​​are within the preset threshold, it is determined that the area where the pixel points corresponding to the some of the pixel values ​​are located is the face image.

3. The elevator access control and security method based on the image recognition model according to claim 2 is characterized in that: The step of determining that the area where the pixel points corresponding to the plurality of pixel values ​​are located is the face image comprises: Screening out a number of target pixel points whose pixel values ​​are within the preset threshold, and highlighting the edges of the target pixel points to identify the target areas covered by the target pixel points; The edge of the target area is identified by a preset edge detection algorithm, and corresponding contour lines are added at the edge to generate the face image.

4. The elevator access control and security method based on the image recognition model according to claim 1 is characterized in that: The method further comprises: When a plurality of users are detected entering the interior of the elevator, the front panoramic images corresponding to the plurality of users are collected in real time by the camera device, and the front panoramic images are grayed to generate corresponding gray images; Detecting the grayscale values ​​corresponding to each pixel in the grayscale image one by one, and determining whether each grayscale value is within a preset grayscale threshold; If it is determined that there is a grayscale value within the preset grayscale threshold, a pixel corresponding to the current grayscale value is determined to be a target pixel, and a target area where several target pixels exist is identified, wherein the target area includes face areas corresponding to the multiple users respectively; Searching for a corresponding target face image in the face area according to the target user information, and determining that the target face image is consistent with the target image in the database; If it is determined that the target face image is consistent with the target image in the database, the control code is sent to the elevator, so that the elevator activates the corresponding floor through the control code.

5. An elevator access control security system based on an image recognition model, characterized in that: The elevator access control and security method based on the image recognition model according to any one of claims 1 to 4 is implemented, and the system comprises: The acquisition module is used to acquire a front full-body image corresponding to the current user in real time through a camera device when a user is detected entering the elevator, and extract a face image from the front full-body image through a PP-ShiTu image recognition model; A search module, used to detect a number of target facial features contained in the face image, and search for a corresponding target image in a preset image database according to the target facial features, wherein the target image is unique; A judgment module, used to judge whether the target image is consistent with the face image; A processing module, for determining that the current user is a resident if it is determined in real time that the target image is consistent with the face image, and finding the corresponding target user information according to the target image; The sending module is used to generate a corresponding control code according to the target user information, and send the control code to the elevator, so that the elevator activates the corresponding floor through the control code, and the control code is one-time.

Images