Gate system and its forensics method

Abstract

The application discloses a gate system and a forensics method thereof, and the forensics method comprises the following steps: forensics starting, acquiring a face image collected by a face recognition camera and three-dimensional point cloud data collected by a three-dimensional point cloud sensor; analyzing the relative coordinate position of the face in the three-dimensional point cloud data; determining the face image for identity verification in the face image collected by the face recognition camera according to the relative coordinate position of the face; verifying the identity of the personnel by using the face image for identity verification; and controlling the opening and closing of the gate according to the verification result. The three-dimensional coordinate recognition technology can clearly distinguish the relative coordinate positions of different personnel, the face image for identity verification is determined by coordinate positioning, the accuracy and reliability of face recognition are improved, accurate identity verification is ensured, more intelligent and efficient forensics analysis and access control are realized, and the work burden of security personnel is also reduced.

Description

Technical Field

[0001] This invention relates to the field of turnstile technology, and in particular to a turnstile system and its evidence collection method. Background Technology

[0002] Traditional turnstiles are used to manage pedestrian flow and regulate entry and exit, primarily in subway turnstile systems, railway station ticket gate systems, access control systems, and tourist area ticket gate systems. Their most basic and core function is to allow only one person to pass through at a time. With advancements in turnstile functionality, features such as automatic ticket checking and ID recognition have been added. However, to ensure security, the requirements for ticket checking are becoming increasingly stringent, especially in railway ticket checking systems, which aim to achieve a unified identity for the person, ticket, and ID, eliminating discrepancies between the ticket and ID, or between the person and their ID.

[0003] In recent years, facial recognition turnstiles have become increasingly widely used. Many office buildings, train stations, and other high-traffic areas employ facial recognition turnstiles. Their identification principle is as follows: First, a camera captures facial images or videos of people passing through and identifies their facial information. Then, this information is compared with the personnel information in the system database. If the facial recognition comparison is successful, an opening signal is sent to the turnstile, and the turnstile opens, allowing the person to pass through. If the system database does not contain the person's facial information, they can pass through the turnstile by swiping a card.

[0004] However, traditional turnstiles and facial recognition turnstiles have a common problem: when there are many people queuing at the entrance / exit, the front-end camera may capture facial images or videos of both the person in front and the person behind them, or it may only capture the facial image or video of the person behind. When the turnstile opens and the person in front passes through, if someone behind tries to pass, the system will falsely report that their facial image or video has already been captured, preventing them from passing through. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address at least one defect of the related technologies mentioned in the background: in existing face recognition gates, when faced with dense crowds, the problem of misidentification of faces and accidental gate opening often occurs, so the present invention provides a gate system and a method for obtaining evidence therefrom.

[0006] The technical solution adopted by this invention to solve its technical problem is: to construct an evidence collection method for a turnstile system, wherein the turnstile system includes a turnstile, a face recognition camera, and a 3D point cloud sensor, and the evidence collection method includes the following steps:

[0007] Acquisition steps: The evidence collection begins by acquiring the face image captured by the face recognition camera and the 3D point cloud data captured by the 3D point cloud sensor;

[0008] Analysis steps: Analyze the relative coordinates of the face in the 3D point cloud data;

[0009] Location step: Determine the face image used for identity verification from the face image captured by the face recognition camera based on the relative coordinate position of the face;

[0010] Verification steps: Use the facial image used for identity verification to verify the identity of the person;

[0011] Control steps: Control the opening and closing of the gate based on the verification results.

[0012] Preferably, in the evidence collection method of the gate system described in the invention, the three-dimensional point cloud data includes three-dimensional shape and three-dimensional coordinates for characterizing a person.

[0013] Preferably, in the evidence collection method for the gate system described in the invention, the analysis step includes:

[0014] By processing the three-dimensional point cloud data, specific features of the human face can be identified.

[0015] The three-dimensional coordinates of the face are determined based on the specific features of the face, and the relative coordinate position of the face is obtained.

[0016] Preferably, in the evidence collection method for the gate system described in the invention, the identification of specific facial features by processing the three-dimensional point cloud data includes:

[0017] Specific facial features are identified based on the three-dimensional shape of the person in the three-dimensional point cloud data.

[0018] Preferably, in the evidence collection method for the gate system described in the invention, the positioning step includes:

[0019] The three-dimensional coordinates of the foreground face are determined based on the relative coordinate position of the faces.

[0020] Transform the three-dimensional coordinates of the foremost face and the pixel coordinates of the face image to the same reference coordinate system;

[0021] Determine whether the converted 3D coordinates of the foremost face match the pixel coordinates of one of the converted face images. If yes, use that face image for authentication; otherwise, return to the acquisition step.

[0022] Preferably, in the evidence collection method for the gate system described in the invention, the control step includes:

[0023] If the determination in the positioning step is negative, then the gate is controlled to close, and the process returns to the acquisition step.

[0024] Preferably, in the evidence collection method for the gate system described in the invention, the positioning step further includes:

[0025] After transforming the coordinate system, the coordinates are calibrated, including rotation, scaling, and translation.

[0026] Preferably, in the evidence collection method for the gate system described in the invention, the gate system further includes a database, and the verification step includes:

[0027] Face verification steps: Determine whether the face image used for identity verification matches the face images in the database;

[0028] And / or, ID card verification steps: extract the ID card information corresponding to the face image used for identity verification from the database, and determine whether it matches the ID card information entered by the person.

[0029] Preferably, in the evidence collection method for the gate system described in the invention, the control step includes:

[0030] If the facial recognition step and the ID card verification step are both determined to be yes, then the gate is controlled to open.

[0031] If the determination in the face verification step or the ID card verification step is negative, then the gate is controlled to close and the process returns to the acquisition step.

[0032] The present invention also constructs a turnstile system comprising:

[0033] The turnstile is used for management personnel to enter and exit;

[0034] A face recognition camera, used to capture face images;

[0035] A three-dimensional point cloud sensor, which is used to collect three-dimensional point cloud data;

[0036] A controller, wherein the controller is used to implement the evidence collection method of the gate system described in any of the above-mentioned embodiments.

[0037] By implementing this invention, the following beneficial effects are achieved:

[0038] This invention uses three-dimensional coordinate recognition technology to clearly distinguish the relative coordinate positions of different people. By using coordinate positioning to determine the facial image used for identity verification, it avoids the problem of accidentally opening doors due to misidentification of the face of the person behind, improves the accuracy and reliability of facial recognition, ensures accurate identity verification, and realizes more intelligent and efficient evidence collection analysis and access control, while also reducing the workload of security personnel. Attached Figure Description

[0039] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0040] Figure 1 This is a flowchart of an embodiment of the evidence collection method for the gate system of the present invention;

[0041] Figure 2 This is a flowchart of the analysis steps in the evidence collection method of the gate system of the present invention;

[0042] Figure 3 This is a flowchart of the positioning step in the evidence collection method of the gate system of the present invention;

[0043] Figure 4 This is a logic structure diagram of one embodiment of the gate system of the present invention. Detailed Implementation

[0044] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0045] It should be noted that the flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0046] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0047] It should be noted that the area closer to the turnstile is the front, and the area farther from the turnstile is the back.

[0048] like Figure 1 As shown, one embodiment of the present invention discloses an evidence collection method for a turnstile system. The turnstile system includes a turnstile for personnel to enter and exit, a face recognition camera for capturing facial images, and a three-dimensional point cloud sensor for capturing three-dimensional point cloud data. The evidence collection method includes acquisition steps, analysis steps, positioning steps, verification steps, and control steps, as detailed below:

[0049] Acquisition Steps: The evidence collection begins by acquiring facial images captured by a face recognition camera and 3D point cloud data collected by a 3D point cloud sensor.

[0050] Specifically, evidence collection begins when personnel are detected entering the turnstile area. In some embodiments, the turnstile area is a preset range in front of the turnstile, such as the area from the turnstile to the prohibition line in front of the turnstile. Personnel entering the turnstile area can be detected by infrared sensors, another face recognition camera, or other means, which is not limited here.

[0051] When a person is detected entering the turnstile area, the facial recognition camera and the 3D point cloud sensor work simultaneously. The facial recognition camera can capture at least one facial image in front of the turnstile or within the turnstile area. In some embodiments, at least one facial image in front of the turnstile or within the turnstile area can be captured within a first preset capture range. Specifically, when the queue of people is dispersed, the facial recognition camera can capture one facial image, while when the queue of people is dense, the facial recognition camera can capture facial images of people in front, behind, or behind, thus capturing at least one facial image. It can be understood that "at least one" includes one, two, three, or any number of facial images.

[0052] The 3D point cloud sensor can collect 3D point cloud data in front of or within the gate area. In some embodiments, the 3D point cloud data in front of or within the gate area can be collected within a second preset collection range. The second preset collection range can be greater than or equal to the first preset collection range to ensure the coverage of the 3D point cloud data. The 3D point cloud data includes the 3D shape and 3D coordinates used to characterize people, in order to identify and locate the relative coordinate positions of different people. In some embodiments, the 3D point cloud sensor is a LiDAR sensor, a depth camera, or other sensors capable of acquiring 3D point cloud data, which is not limited here.

[0053] Analysis steps: Analyze the relative coordinates of the face in the 3D point cloud data, specifically as follows... Figure 2 As shown, it includes:

[0054] S11: By processing 3D point cloud data, specific features of a human face are identified.

[0055] Specifically, specific facial features are identified based on the 3D shape of a person in 3D point cloud data. These specific facial features include the eyes, nose, and mouth. At least one specific facial feature can be identified; understandably, "at least one" includes one, two, three, or any number of features. In some embodiments, a face detection algorithm may be used to identify these features, such as a deep learning face detection model, etc., which is not limited here.

[0056] S12: Determine the three-dimensional coordinates of the face based on its specific features, and obtain the relative coordinate position of the face.

[0057] Specifically, since specific features of at least one face can be identified, the three-dimensional coordinates of at least one face can be determined. Understandably, "at least one" includes one, two, three, or any number of faces. When determining the three-dimensional coordinates of one face, in terms of the relative coordinate position of the faces, this unique three-dimensional coordinate is the coordinate of the foreground face. When determining the three-dimensional coordinates of at least two faces, in terms of the relative coordinate positions of the faces, it can be determined which face is in front and which face is behind, thus determining the three-dimensional coordinates of the foreground face.

[0058] Positioning steps: Based on the relative coordinates of the face, determine the facial image used for identity verification from the facial recognition camera image. This ensures that the correct facial image is used for identity verification and avoids accidental door opening due to incorrect facial recognition of people behind the camera. Specifically, as follows... Figure 3 As shown, it includes:

[0059] S21: Determine the three-dimensional coordinates of the foreground face based on its relative coordinate position.

[0060] S22: Transform the 3D coordinates of the foreground face and the pixel coordinates of the face image to the same reference coordinate system to facilitate coordinate positioning. In some embodiments, the transformation can be to the gate coordinate system or other absolute coordinate systems, which is not limited here. Furthermore, preferably, to ensure accurate correspondence between the coordinates before and after the transformation, the coordinates are calibrated after transformation to the same reference coordinate system, wherein the calibration includes rotation, scaling, and translation.

[0061] S23: Determine whether the 3D coordinates of the foremost face after conversion are consistent with the pixel coordinates of one of the converted face images. If yes, use that face image as the face image for identity verification. If no, prompt for secondary verification and return to the acquisition step, perform manual inspection, or swipe card.

[0062] Verification steps: Verify a person's identity using a facial image used for identity verification, specifically including:

[0063] The facial verification process involves determining whether the facial image used for identity verification matches the facial images in the database. The gate system also includes a database storing the facial images and identification information of registered individuals.

[0064] And / or, ID verification steps: Extract the ID information corresponding to the face image used for identity verification from the database, and determine whether it matches the ID information entered by the person.

[0065] Specifically, after facial recognition verification is successful, personnel can proceed with ID card verification, or vice versa, or only one of these verifications may be performed. Some embodiments of the turnstile include two gates, with the first gate opening after one verification is successful and the second gate opening after the other verification is successful. Understandably, other embodiments of the turnstile include one gate that can only open after both verifications are successful.

[0066] Control procedures: Based on the verification results, control the opening and closing of the gate to allow or deny personnel passage.

[0067] Specifically, if the location step fails to identify the correct answer, the gate will close, denying passage and returning to the acquisition step. If the face verification and ID card verification steps both succeed, the gate will open, allowing passage. If either the face verification or ID card verification step fails to identify the correct answer, the gate will close, denying passage and returning to the acquisition step.

[0068] This embodiment achieves multiple verifications of personnel by comparing coordinate positioning, facial recognition, and ID card information, thereby improving the accuracy and security of identity verification.

[0069] In some embodiments, the control steps further include: if multiple verification results fail, then activating the alarm system of the gate system.

[0070] In some embodiments, the method further includes a recording step, as follows:

[0071] Recording Steps: All verification results and access events are recorded, while the turnstile system's user interface provides operators with real-time feedback and historical data queries. For example, real-time feedback includes generating pedestrian flow statistics reports, access personnel information reports, and abnormal event reports, providing valuable references for security management in buildings and other locations. Management can better understand the building's pedestrian flow and make corresponding security and operational optimization decisions.

[0072] In some embodiments, the method further includes a maintenance step, as follows:

[0073] Maintenance steps: Receive configuration, maintenance, and update operations performed by operators through the user interface of the gate system, such as updating the face recognition algorithm and optimizing the 3D coordinate recognition parameters, to maintain the efficient operation and security of the gate system.

[0074] In some embodiments, the method further includes a monitoring step, as follows:

[0075] Monitoring steps: By analyzing the 3D point cloud data collected by the 3D point cloud sensor, illegal passage (person forcibly passing through when the gate is not open) or tailgating (two people inside one gate at the same time) events are identified. If an event is detected, the gate system's alarm system is activated. The 3D coordinate recognition technology implemented in this project provides real-time spatial monitoring capabilities, enabling timely detection and alarm of illegal passage and tailgating events, thus enhancing on-site safety management and emergency response capabilities.

[0076] like Figure 4 As shown, one embodiment of the present invention discloses a turnstile system, including a turnstile, a face recognition camera, a 3D point cloud sensor, a controller, a database, an alarm system, and a user interface, as detailed below:

[0077] A turnstile is used for personnel access and includes at least one gate, which can be understood to include one, two, or any number of gates. In some embodiments, a facial recognition camera and a 3D point cloud sensor are mounted at the front or rear of the turnstile.

[0078] A facial recognition camera is used to capture facial images. Specifically, it can capture at least one facial image in front of or within the gate area. In some embodiments, it can capture at least one facial image in front of or within the gate area according to a first preset capture range. When people are queuing dispersedly, the facial recognition camera can capture one facial image. When people are queuing densely, the facial recognition camera can capture facial images of people in front, behind, or behind, thus capturing at least one facial image. Understandably, "at least one" includes one, two, three, or any number of facial images.

[0079] A 3D point cloud sensor is used to collect 3D point cloud data, specifically 3D point cloud data in front of or within the gate area. In some embodiments, the 3D point cloud data in front of or within the gate area can be collected within a second preset collection range. This second preset collection range can be greater than or equal to a first preset collection range to ensure the coverage of the 3D point cloud data. The 3D point cloud data includes the 3D shape and 3D coordinates used to characterize people, in order to identify and locate the relative coordinate positions of different people. In some embodiments, the 3D point cloud sensor is a LiDAR sensor, a depth camera, or other sensors capable of acquiring 3D point cloud data; this is not limited to these specific sensors.

[0080] The database stores facial images and identification information of registered individuals. The controller implements the authentication method of the gate system described in the above embodiments, and will not be repeated here. In some embodiments, the controller is a central control server.

[0081] By implementing this invention, the following beneficial effects are achieved:

[0082] This invention uses three-dimensional coordinate recognition technology to clearly distinguish the relative coordinate positions of different people. By using coordinate positioning to determine the facial image used for identity verification, it avoids the problem of accidentally opening doors due to misidentification of the face of the person behind, improves the accuracy and reliability of facial recognition, ensures accurate identity verification, and realizes more intelligent and efficient evidence collection analysis and access control, while also reducing the workload of security personnel.

[0083] It is understood that the above embodiments only illustrate some implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that for those skilled in the art, without departing from the concept of the present invention, the above embodiments or technical features can be freely combined, and several modifications and improvements can be made. These all fall within the protection scope of the present invention. That is, the embodiments described "in some embodiments" can be freely combined with any of the embodiments above and below. Therefore, all equivalent transformations and modifications made within the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A method for obtaining evidence from a turnstile system, the turnstile system comprising a turnstile, a face recognition camera, and a 3D point cloud sensor, characterized in that, The evidence collection method includes the following steps: Acquisition steps: The evidence collection begins by acquiring the face image captured by the face recognition camera and the 3D point cloud data captured by the 3D point cloud sensor; Analysis steps: Analyze the relative coordinates of the face in the 3D point cloud data; Positioning steps: Determine the three-dimensional coordinates of the foreground face based on its relative coordinate position; transform the three-dimensional coordinates of the foreground face and the pixel coordinates of the face image to the same reference coordinate system; calibrate the coordinates after the coordinate system transformation, including rotation, scaling, and translation; determine whether the transformed three-dimensional coordinates of the foreground face are consistent with the pixel coordinates of one of the transformed face images; if so, use that face image as the face image for identity verification; otherwise, return to the acquisition step. Verification steps: Use the facial image used for identity verification to verify the identity of the person; Control steps: Control the opening and closing of the gate based on the verification results.

2. The method for obtaining evidence for a turnstile system according to claim 1, characterized in that, The three-dimensional point cloud data includes the three-dimensional shape and three-dimensional coordinates used to characterize the person.

3. The method for obtaining evidence for a turnstile system according to claim 1, characterized in that, The analysis steps include: By processing the three-dimensional point cloud data, specific features of the human face can be identified. The three-dimensional coordinates of the face are determined based on the specific features of the face, and the relative coordinate position of the face is obtained.

4. The method for obtaining evidence for a turnstile system according to claim 3, characterized in that, By processing the 3D point cloud data, specific features of a human face are identified, including: Specific facial features are identified based on the three-dimensional shape of the person in the three-dimensional point cloud data.

5. The method for obtaining evidence for a turnstile system according to claim 1, characterized in that, The control steps include: If the determination in the positioning step is negative, then the gate is controlled to close, and the process returns to the acquisition step.

6. The method for obtaining evidence for a turnstile system according to claim 1, characterized in that, The gate system also includes a database, and the verification steps include: Face verification steps: Determine whether the face image used for identity verification matches the face images in the database; And / or, ID card verification steps: extract the ID card information corresponding to the face image used for identity verification from the database, and determine whether it matches the ID card information entered by the person.

7. The method for obtaining evidence for a turnstile system according to claim 6, characterized in that, The control steps include: If the facial recognition step and the ID card verification step are both determined to be yes, then the gate is controlled to open. If the determination in the face verification step or the ID card verification step is negative, then the gate is controlled to close and the process returns to the acquisition step.

8. A turnstile system, characterized in that, include: The turnstile is used for management personnel to enter and exit; A face recognition camera, used to capture face images; A three-dimensional point cloud sensor, which is used to collect three-dimensional point cloud data; A controller for implementing the evidence collection method of the gate system according to any one of claims 1-7.

Images