Gate dual-mode directional voice guiding method based on region division and related equipment

Abstract

The invention relates to the field of space region division and voice guidance, in particular to a gate dual-mode directional voice guidance method based on region division and related equipment. The method comprises the following steps: collecting a gate peripheral space image; according to the space image and a first preset division proportion, an in-channel interaction area and at least one out-channel pre-guiding area are divided, and the in-channel interaction area is used for performing face verification on passengers in the in-channel interaction area; a verification signal of the face and the certificate of the passenger is obtained from the in-channel interaction area, a first voice instruction is generated according to the verification signal, and a first set of loudspeakers is controlled according to the first voice instruction to deliver narrow-band sound beam voice of a first preset sound beam angle to the in-channel interaction area; and acquiring image information conforming to the special passenger characteristics from the out-of-channel pre-guiding area, generating a corresponding second voice instruction according to the image information, and controlling a second group of loudspeakers to deliver coverage sound beam voice at a second preset sound beam angle to the out-of-channel pre-guiding area according to the second voice instruction.

Description

Technical Field

[0001] This application relates to the field of spatial area division and voice guidance, and in particular to a dual-mode directional voice guidance method and related equipment for turnstiles based on area division. Background Technology

[0002] In areas with high pedestrian traffic, such as transportation hubs and public places, turnstiles play a crucial role in improving pedestrian flow efficiency and management. As people's demands for service quality continue to rise, providing accurate and clear voice guidance at turnstiles is becoming increasingly important. A good voice guidance system can help passengers quickly understand the passage procedures and precautions, reducing their dwell time at the turnstiles, thereby improving overall passage efficiency and enhancing the passenger travel experience. This is of great significance for ensuring the orderly operation of public places.

[0003] Existing turnstile voice guidance technologies typically employ a single-mode voice broadcasting approach. One common method is to install ordinary loudspeakers near the turnstiles to broadcast uniform voice prompts throughout the surrounding area, ensuring passengers receive the same message regardless of whether they are inside or outside the turnstile. Another method involves on-site human guidance, where staff verbally instruct passengers on how to operate and pass through the turnstiles. Some systems also display text prompts on the turnstile screens to assist passengers in completing the passage process.

[0004] However, existing voice guidance methods have significant drawbacks. Single-mode voice broadcasts cannot accurately guide passengers in different areas. Passengers undergoing facial verification or other interactive operations within the channel may be disturbed by irrelevant surrounding voices, affecting verification efficiency. Meanwhile, special passengers waiting in line outside the channel, such as those carrying large luggage, using wheelchairs, or strollers, lack specific pre-guidance prompts, potentially causing obstruction when entering the turnstile due to insufficient preparation. Manual guidance is not only labor-intensive but also difficult to provide comprehensive and timely coverage. On-screen text prompts also have limitations for passengers with poor eyesight or unfamiliar with the operation.

[0005] Therefore, how to design a dual-mode directional voice guidance method that can provide accurate voice guidance for passengers in different regions is a technical problem that urgently needs to be solved. Summary of the Invention

[0006] The purpose of this application is to overcome the above-mentioned technical problems and provide a dual-mode directional voice guidance method and related equipment for turnstiles based on regional division. It can provide accurate voice guidance according to the needs of passengers in different regions, so as to improve the efficiency of face and document verification in the channel, while providing special pre-guidance prompts for special passengers and reducing passage obstacles.

[0007] In a first aspect, one embodiment of this application discloses a dual-mode directional voice guidance method for turnstiles based on region division, which adopts the following scheme: A dual-mode directional voice guidance method for turnstiles based on region division includes: acquiring spatial images surrounding the turnstile; dividing the turnstile into an in-channel interaction area and at least one out-of-channel pre-guidance area according to the spatial images and a first preset division ratio, wherein the in-channel interaction area is used for facial verification of passengers in the in-channel interaction area; acquiring verification signals of passenger faces and identification documents from the in-channel interaction area, generating a first voice command based on the verification signals, and controlling a first set of loudspeakers to project narrow-beam voice with a first preset beam angle to the in-channel interaction area according to the first voice command; acquiring image information conforming to special passenger characteristics from the out-of-channel pre-guidance area, generating a corresponding second voice command based on the image information, and controlling a second set of loudspeakers to project a covering beam voice with a second preset beam angle to the out-of-channel pre-guidance area according to the second voice command.

[0008] By adopting the above technical solution, acquiring spatial images around the turnstile allows for a comprehensive understanding of the surrounding environment. Based on the spatial images and a first preset division ratio, an interactive area within the channel and at least one pre-guidance area outside the channel are defined. This allows for the targeted setting of areas for facial verification and areas for pre-guiding passengers about to enter the turnstile. Verification signals of the passenger's face and identification are obtained from the interactive area within the channel, generating a first voice command. This command controls a first set of loudspeakers to project a narrow-beam voice signal with a first preset beam angle into the interactive area within the channel, accurately delivering voice information related to the verification process to passengers undergoing facial and identification verification. Image information matching the characteristics of special passengers is obtained from the pre-guidance area outside the channel, generating a corresponding second voice command. This command controls a second set of loudspeakers to project a comprehensive voice signal with a second preset beam angle into the pre-guidance area outside the channel, providing timely reminders and guidance to special passengers and preventing inconvenience caused by special circumstances.

[0009] Optionally, the step of dividing the channel into an interactive area and at least one pre-guided area outside the channel based on the spatial image and a first preset division ratio includes: identifying the gate channel boundary from the spatial image and determining the interactive area inside the channel based on the first preset division ratio; determining at least one pre-guided area outside the channel based on the number of interactive areas inside the channel, the preset guidance area, and the second preset division ratio, wherein the second preset division ratio is used to determine the spatial layout of at least one pre-guided area outside the channel relative to the gate entrance.

[0010] By adopting the above technical solution, the boundary of the gate channel can be identified from the spatial image and the interactive area within the channel can be determined according to the first preset division ratio. This can accurately locate the area where passengers can undergo facial verification, improving the targeting and accuracy of the verification process. Based on the number of interactive areas within the channel, the number of preset guidance areas, and the second preset division ratio, at least one pre-guidance area outside the channel can be determined, and its spatial layout relative to the gate entrance can be clearly defined. This can achieve a reasonable division of the queuing area outside the gate, facilitating targeted guidance for special passengers and improving overall guidance efficiency and passenger passage experience.

[0011] Optionally, the step of acquiring the verification signal of the passenger's face and ID card from the interaction area within the channel, and generating a first voice command based on the verification signal, to control a first group of loudspeakers to project a narrow-beam voice with a first preset beam angle into the interaction area within the channel according to the first voice command, includes: acquiring the verification signal generated by the face acquisition device and the ID card reader, wherein the verification signal includes at least one of a verification start signal, a verification success signal, or a verification failure signal; retrieving the corresponding broadcast text content and broadcast control parameters from a preset voice command mapping table based on the verification signal to generate the first voice command, wherein the broadcast control parameters include broadcast priority and / or broadcast repetition count; controlling the first group of loudspeakers to generate a narrow-beam voice with the first preset beam angle according to the first voice command, and modulating the broadcast text content onto the narrow-beam voice for voice projection.

[0012] By adopting the above technical solution, verification signals including verification start signal, verification success signal, or verification failure signal generated by the face acquisition device and document reader can be obtained in a timely and accurate manner to know the face and document verification status of passengers in the interaction area of ​​the channel. Based on the verification signal, the corresponding broadcast text content and broadcast control parameters are retrieved from the preset voice command mapping table to generate the first voice command. The voice command can be accurately generated according to different verification situations, and the broadcast control parameters can flexibly control the broadcast priority and / or the number of broadcast repetitions. Based on the first voice command, the first group of loudspeakers is controlled to generate a narrow beam with a first preset beam angle, and the broadcast text content is modulated into the narrow beam for voice delivery, which can realize directional voice guidance in the interaction area of ​​the channel, so that the voice accurately covers the target area.

[0013] Optionally, the step of acquiring image information matching the characteristics of special passengers from the pre-guidance area outside the passage, and generating a corresponding second voice command based on the image information, to control a second group of loudspeakers to project a covering sound beam with a second preset sound beam angle onto the pre-guidance area outside the passage according to the second voice command, includes: acquiring an image sequence in the pre-guidance area outside the passage; identifying the image sequence to determine whether there is image information matching the characteristics of special passengers; if so, retrieving the corresponding broadcast text content and broadcast control parameters from a preset voice command mapping table based on the identified image information to generate a corresponding second voice command; and controlling a second group of loudspeakers to project a covering sound beam with the second preset sound beam angle onto the pre-guidance area outside the passage according to the second voice command.

[0014] By adopting the above technical solution, image sequences in the pre-guidance area outside the passage can be obtained, providing a data foundation for subsequent identification of special passenger characteristics. The image sequences are identified to determine whether there is image information that matches the characteristics of special passengers. If so, the corresponding broadcast text content and broadcast control parameters are retrieved from the preset voice command mapping table based on the identified image information to generate a second voice command. This can accurately generate appropriate voice commands for special passengers. Based on the second voice command, the second set of loudspeakers is controlled to project a second preset sound beam angle of sound into the pre-guidance area outside the passage, which can effectively cover the pre-guidance area outside the passage and provide voice guidance for special passengers.

[0015] Optionally, the special passenger characteristics include at least one of the following image recognition features: carrying luggage exceeding a preset threshold size, being elderly, using assistive devices, or exhibiting behaviors such as holding an infant or leading a young child, wherein the assistive devices include wheelchairs, canes, and strollers.

[0016] By adopting the above technical solutions, passengers carrying luggage exceeding a preset threshold can be identified as special passenger characteristics. This allows for the identification of passengers carrying large luggage, enabling targeted voice guidance to help them pass through the turnstiles more smoothly. Similarly, identifying the elderly, those using wheelchairs, canes, or strollers as special passenger characteristics allows for the accurate identification of passengers with mobility impairments or those carrying infants, providing timely voice prompts and guidance. Furthermore, identifying passengers holding infants or leading young children as special passenger characteristics allows for attention to passengers with children, providing them with appropriate voice guidance and ensuring their safety during passage.

[0017] Optionally, the horizontal width of the first preset sound beam angle is less than or equal to 30 degrees, and the vertical width is less than or equal to 20 degrees; the horizontal width of the second preset sound beam angle is greater than 60 degrees.

[0018] By adopting the above technical solution, setting the horizontal width of the first preset sound beam angle to less than or equal to 30 degrees and the vertical width to less than or equal to 20 degrees allows the narrow-beam sound to accurately concentrate its coverage in the interactive area within the channel, preventing sound from spreading to other areas and ensuring that passengers within the channel can clearly and accurately receive verification-related voice information. Setting the horizontal width of the second preset sound beam angle to greater than 60 degrees allows the covering sound beam to have a wider coverage in the pre-guidance area outside the channel, effectively covering special passengers in the pre-guidance area outside the channel and ensuring that special passengers can receive the corresponding guidance voice in a timely manner.

[0019] Secondly, another embodiment of this application discloses a dual-mode directional voice guidance system for turnstiles based on region division, which adopts the following scheme: A dual-mode directional voice guidance system for turnstiles based on region division includes: an in-channel verification module for verifying the face and identification of passengers who have entered the turnstile passage area; an out-of-channel perception module for collecting image data of a pre-set length of queuing area at the turnstile entrance for identifying special passenger characteristics; a region division and recognition module for dividing the region inside and outside the channel based on the image data and identifying the behavioral characteristics of queuing passengers; a directional audio playback module for generating and projecting directional sound beams with different beamwidths; and a dual-mode voice control module, which is communicatively connected to the in-channel face verification module, the out-of-channel perception module, the region division and recognition module, and the directional audio playback module, respectively, for receiving two types of information to generate two sets of independent voice commands, and driving the directional audio playback module to project directional sound beams with different beamwidths to the in-channel and out-of-channel areas respectively; the two types of information include status information corresponding to the in-channel passenger verification results and identification information of special passengers outside the channel.

[0020] By adopting the above technical solutions, the internal verification module can verify the faces and documents of passengers who have entered the gate passage area, avoiding the problems of low efficiency and error-proneness of manual verification, thus improving the efficiency and accuracy of gate passage. The external sensing module collects image data of the pre-set length of the queuing area at the gate entrance to identify the characteristics of special passengers, allowing for proactive attention to their needs and providing more timely services compared to traditional methods. Furthermore, the area division and recognition module divides the area inside and outside the gate based on image data and identifies the behavioral characteristics of queuing passengers, enabling the rational planning of the area inside and outside the gate. This system facilitates the categorization and guidance of different passengers, improving upon the previous chaotic and unclear guidance situation. The directional audio playback module generates and delivers directional sound beams with different beamwidths, accurately transmitting voice information to designated areas and reducing sound propagation interference. The dual-mode voice control module receives status information corresponding to passenger verification results within the channel and identification information of special passengers outside the channel, generating two independent sets of voice commands. These commands drive the directional audio playback module to deliver directional sound beams of different beamwidths to areas inside and outside the channel, achieving precise voice guidance for passengers in different situations inside and outside the channel, thus improving the overall quality of guidance services.

[0021] Optionally, the channel verification module is integrated into the camera above the gate operation screen, used to collect faces and verify documents of passengers who have entered the gate passage area; the channel external sensing module is installed in the camera above the gate channel, and the field of view of the corresponding camera covers the queuing area of ​​the gate entrance with a preset length.

[0022] By adopting the above technical solution, the internal verification module is integrated into the camera above the gate operation screen, which can collect facial images and verify documents of passengers who have entered the gate passage area, thus realizing the identity verification of passengers in the channel; the external sensing module is installed in the camera above the gate channel and its field of view covers the queuing area of ​​the gate entrance with a preset length, which can collect image data of the area to identify special passenger characteristics.

[0023] Thirdly, another embodiment of this application discloses a computer-readable storage medium, which adopts the following scheme: A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0024] Fourthly, another embodiment of this application discloses a computer-readable storage medium, which adopts the following scheme: An electronic device includes: a memory and a processor, the memory being used to store a computer program; the processor being used to execute the computer program to implement the steps of the method described above.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. Acquiring spatial images around the turnstile allows for a comprehensive understanding of the surrounding environment. Based on the spatial images and a first preset division ratio, an interactive area within the channel and at least one pre-guidance area outside the channel are defined. This allows for the targeted setting of areas for facial verification and pre-guidance of passengers about to enter the turnstile. Verification signals of the passenger's face and identification are obtained from the interactive area within the channel, generating a first voice command. This command controls a first set of loudspeakers to project a narrow-beam voice signal with a first preset beam angle into the interactive area within the channel, accurately delivering voice information related to the verification process to passengers undergoing facial and identification verification. Image information matching the characteristics of special passengers is obtained from the pre-guidance area outside the channel, generating a corresponding second voice command. This command controls a second set of loudspeakers to project a comprehensive voice signal with a second preset beam angle into the pre-guidance area outside the channel, providing timely reminders and guidance to special passengers and preventing inconvenience caused by special circumstances. 2. Identifying the gate channel boundary from the spatial image and determining the interaction area within the channel according to the first preset division ratio can accurately locate the area for passengers to undergo facial verification, improving the targeting and accuracy of the verification process; determining at least one pre-guidance area outside the channel based on the number of interaction areas and preset guidance areas within the channel and the second preset division ratio, and clarifying its spatial layout relative to the gate entrance, can achieve a reasonable division of the queuing area outside the gate, facilitating targeted guidance for special passengers, improving overall guidance efficiency and passenger passage experience; 3. Acquire verification signals generated by the face capture device and document reader, including verification start signal, verification success signal, or verification failure signal, to promptly and accurately ascertain the face and document verification status of passengers in the interaction area within the passageway; generate a first voice command by retrieving the corresponding broadcast text content and broadcast control parameters from a preset voice command mapping table based on the verification signal, accurately generating voice commands according to different verification situations, and flexibly controlling the broadcast priority and / or the number of broadcast repetitions based on the broadcast control parameters; control the first group of loudspeakers to generate a narrow beam with a first preset beam angle based on the first voice command, and modulate the broadcast text content into the narrow beam for voice delivery, enabling directional voice guidance in the interaction area within the passageway, ensuring accurate voice coverage of the target area. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating a dual-mode directional voice guidance method for turnstiles based on region division, as disclosed in an embodiment of this application. Figure 2 for Figure 1 A flowchart illustrating step S20 in a region-division-based dual-mode directional voice guidance method for turnstiles; Figure 3 for Figure 1 A flowchart illustrating step S30 in a disclosed dual-mode directional voice guidance method for turnstiles based on region division; Figure 4 for Figure 1 A flowchart illustrating step S40 in a region-division-based dual-mode directional voice guidance method for turnstiles; Figure 5 This is a schematic diagram of the structure of a dual-mode directional voice guidance system for turnstiles based on region division, as disclosed in another embodiment of this application. Detailed Implementation

[0027] The present application will be further described in detail below with reference to the accompanying drawings.

[0028] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0029] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a” and “the” as used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0030] It should be understood that although the terms "first," "second," etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0031] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0032] [First Embodiment] See Figure 1 The first embodiment of this application discloses a dual-mode directional voice guidance method for turnstiles based on region division, which includes the following steps: S10. Acquire images of the space surrounding the turnstile; The spatial images can be captured by a wide-angle / fisheye camera deployed directly above or diagonally above the turnstile channel. The camera's field of view can completely cover the queuing area from the ticket verification end (inside) of the turnstile to a certain distance (e.g., 15 meters) in front, thus enabling the acquisition of spatial images around the turnstile.

[0033] S20. Based on the spatial image and the first preset division ratio, divide the channel into an interactive area and at least one pre-guided area outside the channel. The spatial image includes the turnstile itself, the physical boundaries of the passage (such as railings), ground markings, and passengers within that area. The first preset division ratio is the proportion of the in-passage interaction area and the out-of-passage pre-guidance area in the entire spatial image. For example, if the first preset division ratio is set to 60%, then the "in-passage interaction area" is defined as a rectangular or approximately trapezoidal spatial area extending 9 meters (15 meters * 60%) inward from the passage entrance. The in-passage interaction area is used for facial verification of passengers within the interaction area. The out-of-passage pre-guidance area is used for passengers outside the passage to queue and for voice guidance.

[0034] In this embodiment, the first preset division ratio is set based on actual needs and experience. For example, a certain area near the face verification device within the gate channel is designated as the interaction area within the channel. The determination and number of pre-guided areas outside the channel can be set according to actual conditions, such as setting one or more pre-guided areas. For example, in another embodiment, before step S20, the spatial image can be preprocessed, such as perspective correction and distortion removal, to obtain a projection image that better reflects the real physical space.

[0035] For details, see Figure 2 Step S20 includes: S21. Identify the gate channel boundary from the spatial image, and determine the interaction area within the channel according to the first preset division ratio; Among them, image recognition technology can be used to identify the boundary of the gate channel. By analyzing the collected spatial images, the boundary position of the gate channel can be found.

[0036] S22. Based on the number of interactive areas and preset guidance areas within the channel and the second preset division ratio, determine at least one pre-guidance area outside the channel, wherein the second preset division ratio is used to determine the spatial layout of at least one pre-guidance area outside the channel relative to the gate entrance.

[0037] Specifically, the number of pre-guided areas outside the passageway can be set according to actual conditions, such as setting one or more pre-guided areas. The second preset division ratio is used to determine the spatial layout of the pre-guided area outside the passageway relative to the gate entrance. For example, an area with a certain distance and range in front of the gate entrance can be divided into the pre-guided area outside the passageway.

[0038] As described above, step S20 can pre-divide areas, which facilitates the system to accurately determine the status of passengers located in different functional zones (inside the channel: verification in progress / success / failure; outside the channel: normal / special passenger), thereby triggering the most appropriate and timely voice guidance instructions, significantly shortening the overall passage time for passengers, and improving the service experience for special passengers such as the elderly, the infirm, the disabled, and pregnant women.

[0039] S30. Obtain the verification signal of the passenger's face and document from the interaction area in the channel, generate a first voice command based on the verification signal, and control the first group of loudspeakers to project a narrow beam voice with a first preset sound beam angle into the interaction area in the channel according to the first voice command. In the interactive area within the passageway, facial recognition devices (such as dedicated cameras) and document readers (such as RFID card readers and passport scanners) deployed on the turnstiles verify faces and documents to generate corresponding verification signals.

[0040] The verification signals include a verification start signal, a verification success signal, and a verification failure signal. The verification start signal is automatically triggered when a passenger enters the "interaction area within the passage" and the equipment detects a valid face or ID card approaching, indicating that a verification process has begun. The verification success signal is generated when the facial recognition algorithm matches the ID information and meets the release rules, indicating that the passenger has passed identity verification. The verification failure signal is generated when recognition fails, the comparison is inconsistent, the ID card is invalid, or the timeout occurs, indicating that the verification has failed and passenger intervention is required.

[0041] For details, see Figure 3 Step S30 specifically includes: S31. Obtain the verification signal generated by the face capture device and the document reader; The verification signal includes at least one of the following: verification start signal, verification success signal, or verification failure signal.

[0042] The verification start signal is triggered when the verification processing unit detects a passenger entering the "interaction area within the passageway" via an image sensor or proximity sensor and successfully initiates a verification. This signal indicates that the system is ready and is waiting for or collecting identity information. The successful verification signal is generated when the matching degree between the features extracted by the facial recognition algorithm and the features of the identity information read from the document exceeds a preset threshold, and all business rules (such as valid ticket and authorized permissions) are verified. This signal is the decisive indication to allow passage; Verification failure signals can be further divided into several subtypes, such as: face comparison failure signal, document reading failure or invalid signal, verification timeout signal (such as the passenger not completing the operation within the time limit), and business rule verification failure signal (such as expired ticket).

[0043] S32. Based on the verification signal, retrieve the corresponding broadcast text content and broadcast control parameters from the preset voice command mapping table to generate the first voice command; The voice command mapping table is pre-set and stores the broadcast text content and broadcast control parameters corresponding to different verification signals. The broadcast control parameters include broadcast priority and / or the number of broadcast repetitions. The voice command mapping table is as follows: Verification signal type Broadcast text content Broadcast control parameters Verification begins Verification in progress, please wait. Priority: Medium; Number of repetitions: 1; Emotional tone: Steady Verification successful Verification passed, please proceed quickly. Priority: High; Repetition Count: 1; Tone of Voice: Affirmative Verification failed - Face does not match If the verification fails, please use the manual verification process. Priority: High; Repetition Count: 2; Emotional Tone: Hint Verification failed - Invalid document If the verification fails, please use the manual verification process. Priority: High; Repetition Count: 2; Emotional Tone: Hint The broadcast priority is used to schedule multiple instructions when they may conflict. For example, "verification failed" usually has a higher priority than "verification started" to ensure that important prompts are broadcast in a timely manner. The number of broadcast repetitions can be used to reinforce the reminder effect for critical operation instructions (such as verification failure). This can be preset and modified according to actual circumstances, and is not limited here.

[0044] S33. Control the first group of loudspeakers to generate a narrow beam with a first preset beam angle according to the first voice command, and modulate the text content to be broadcast into the narrow beam for voice transmission.

[0045] The first group of loudspeakers is a digital beamforming loudspeaker array, consisting of multiple (e.g., 8 or 16) high-performance loudspeaker units arranged in a specific geometry (e.g., linear or arc-shaped). Each unit is driven by an independent power amplifier channel and is independently controlled by a digital signal processor (DSP). Of course, ordinary loudspeakers combined with a beam control device can also be used to achieve narrow beam delivery.

[0046] The first preset beam angle has a horizontal width of less than or equal to 30 degrees and a vertical width of less than or equal to 20 degrees. This narrow beam can accurately project speech into the interactive area within the channel, reducing interference to other areas.

[0047] S40. Obtain image information that matches the characteristics of special passengers from the pre-guidance area outside the passage, generate a corresponding second voice command based on the image information, and control the second set of loudspeakers to project a covering sound beam with a second preset sound beam angle to the pre-guidance area outside the passage according to the second voice command.

[0048] This step enables intelligent identification and regional directional voice guidance for special passengers waiting in the area outside the turnstile channel, transforming passive response into proactive care service. Through spatially targeted acoustic broadcasting, it significantly optimizes the humanization level of public services and the overall environmental order while improving passage efficiency.

[0049] For details, see Figure 4 Step S40 includes: S41. Obtain the image sequence in the pre-guided area outside the channel; The images are continuously captured by surveillance cameras deployed directly above the turnstile or in front of the entrance. These cameras acquire video image sequences of the pre-guided area outside the turnstile at a preset frame rate (e.g., 15-30 fps) to ensure that the dynamic behavior of passengers can be captured.

[0050] S42. Recognize the image sequence and determine whether there is image information that matches the characteristics of special passengers. If so, retrieve the corresponding broadcast text content and broadcast control parameters from the preset voice command mapping table according to the recognized image information to generate the corresponding second voice command. The acquired image sequences are fed into a visual analysis unit (such as an edge computing device with an integrated AI acceleration chip), which runs a pre-trained deep learning model for multi-level recognition. Object detection: First, the model identifies all “human” targets in the image and determines their bounding box locations.

[0051] Feature classification and behavior recognition: For each detected "person" target, the model further analyzes its visual features and behavioral patterns, and matches them with a predefined "special passenger feature" database.

[0052] This feature library defines and identifies the following typical scenarios: Large luggage carriers: This is determined by detecting whether the size of the luggage next to the passenger exceeds a preset threshold (such as the sum of length, width and height being greater than 1.5 meters).

[0053] Users of mobility aids: Identify the specific shape and structure of devices such as wheelchairs, walking aids (canes), or strollers.

[0054] Accompanying persons with young children: Through posture recognition, determine whether there are typical human joint point spatial relationships such as "holding an infant" or "leading a child".

[0055] The generation of the second voice command also relies on a preset voice command mapping table, as follows: Identified feature types Broadcast text content broadcast Control parameters Guiding purpose Large luggage Dear passengers, please prepare your tickets in advance, and please watch your step if you have large luggage. Priority: Medium; Beam Pattern: Area Coverage Advance reminders to avoid congestion Action assistive tools Please note that there are accessible / mother-and-baby lanes ahead; please follow the signs. Priority: High; Beam Pattern: Area Coverage Proactive triage and care guidance Traveling with children Passengers traveling with children, please take good care of your children and prepare your identification documents in advance. Priority: Medium; Beam Pattern: Area Coverage Safety reminders and efficiency improvements lingering and looking around Passengers who need assistance should pay attention to the signs ahead or consult staff. Priority: Medium; Beam Pattern: Area Coverage Proactively provide help options Based on the identified feature types, the aforementioned mapping table is consulted to generate a second voice command. This command also includes the broadcast text and control parameters. The control parameters focus more on adapting to area broadcasting; for example, they may include a "sound beam coverage" parameter to guide the speaker array in adjusting its coverage angle, ensuring that all relevant passengers within the pre-guidance area can hear it.

[0056] S43. Control the second group of loudspeakers to project a covering sound beam with a second preset sound beam angle to the pre-guided area outside the channel according to the second voice command.

[0057] The second set of speakers is a wide-angle coverage speaker array, used to generate a directional sound beam with a horizontal width of more than 60 degrees. This can attract the attention of target passengers without creating a strong sense of "private conversation" like a narrow beam, which is in line with the etiquette of semi-public signs in public areas.

[0058] The implementation principle of this embodiment is as follows: This method rationally divides the area surrounding the turnstile, collects corresponding signals and generates targeted voice commands based on the characteristics and needs of passengers in different areas, and controls the loudspeakers to broadcast different types of sound beams. This approach avoids the shortcomings of single-mode voice broadcasting in existing technologies, accurately guides passengers in different areas, reduces the time passengers spend at the turnstile, improves overall passage efficiency, reduces labor costs, and also solves the limitations of on-screen text prompts. It provides passengers with a higher quality passage guidance service and has made a significant improvement and contribution to existing turnstile voice guidance technology.

[0059] [Second Embodiment] See Figure 5 The second embodiment of this application discloses a dual-mode directional voice guidance system for turnstiles based on region division, which is used to execute the method described in the first embodiment above, including: a channel core verification module 210, a channel external perception module 220, a region division and recognition module 230, a directional audio playback module 240, and a dual-mode voice control module 250.

[0060] The gate verification module 210 verifies the faces and identification documents of passengers who have entered the gate area. Integrated above the gate's control panel, the camera comprises two key hardware units: a face capture unit and an identification document reading unit. The face capture unit captures high-quality facial images that meet algorithm requirements. The identification document reading unit is a contactless IC card reader (such as an ID card or transportation card), a passport / visa page scanner, or a QR code reader, used to quickly and accurately read the electronic or optical information from the passenger's identification document. After verification, based on the comparison results (whether the matching degree exceeds a threshold) and business rules (ticket validity, access permissions, etc.), it outputs a "verification successful" or "verification failed" signal.

[0061] The external sensing module 220 is used to collect image data of the queuing area of ​​a preset length at the gate entrance for the identification of special passenger characteristics. Specifically, the external sensing module 220 is installed above or in front of a wide-angle / fisheye camera on the top of the gate passage, with a field of view covering a queuing area of ​​approximately 15 meters in front of the gate entrance. For each individual, it analyzes their visual characteristics to determine if they match predefined "special passenger characteristics," such as carrying oversized luggage, using a wheelchair / stroller, holding an infant, or exhibiting hesitant or looking-around behavior patterns. Finally, the sensing results are sent in real-time to the dual-mode voice control module and the area segmentation and recognition module.

[0062] The region segmentation and recognition module 230 is used to segment the area inside and outside the channel based on the image data and to identify the behavioral characteristics of queuing passengers. Specifically, the region segmentation and recognition module 230 includes dynamic spatial segmentation and behavioral characteristic recognition. Dynamic spatial segmentation is used to delineate an interactive area inside the channel according to preset rules (such as a first preset segmentation ratio); outside the channel, it dynamically delineates virtual boundaries (such as fan-shaped areas) of one or more "external pre-guidance areas" in the image coordinate system according to preset parameters such as the "number of guidance areas" and the "second preset segmentation ratio." These areas are the acoustic projection target areas for subsequent comprehensive voice guidance.

[0063] Behavioral feature recognition is used to perform more refined behavioral analysis of passengers within the defined pre-guidance zone. For example, multi-target tracking algorithms can be used to identify behaviors such as holding an infant or leading a young child. This function complements and deepens the "feature classification" function of the channel-external sensing module 220, focusing on the analysis of dynamic behavioral patterns, together providing richer decision-making basis for precise guidance.

[0064] The directional audio playback module 240 is used to generate and deliver directional sound beams with different beamwidths. This module is responsible for converting electronic voice commands into a physical sound field that can be precisely controlled in three-dimensional space, forming the physical basis for "directional voice guidance." In this embodiment, the directional sound beams with different beamwidths are described above for narrow-beam and covering sound beams, and will not be repeated here.

[0065] The dual-mode voice control module 250 is connected to the in-channel face verification module 210, the out-of-channel perception module 220, the area division and recognition module 230, and the directional audio playback module 240, respectively. It is used to receive two types of information to generate two sets of independent voice commands, and drive the directional audio playback module 240 to project directional sound beams with different beamwidths to the areas inside and outside the channel, respectively. The two types of information include the status information corresponding to the passenger verification results inside the channel and the identification information of special passengers outside the channel.

[0066] Among them, the dual-mode voice control module 250 is the "brain" and "central nervous system" of the system, responsible for information aggregation, intelligent decision-making and collaborative control, and is the key to the entire system to achieve "dual-mode" guidance.

[0067] Used to receive "status information" from the channel kernel verification module: namely, event signals such as verification start, success, and failure.

[0068] Used to receive "identification information" from the channel external sensing and area division module: namely, the type, location and pre-guidance area information of special passengers.

[0069] Used to generate the first voice command: Based on the received verification status information, the internal "channel-wide voice command mapping table" is queried to generate the corresponding first voice command. This command includes the broadcast content (such as "Please proceed") and broadcast control parameters (such as priority).

[0070] Used to generate the second voice command: Based on the received special passenger identification information and their location, the "Outside-Channel Voice Command Mapping Table" is queried to generate the corresponding second voice command. This command includes different broadcast content (such as "Please be careful with large luggage") and control parameters (such as area broadcast mode flag, target area ID).

[0071] For collaborative driving: The two sets of voice commands generated are sent to the directional audio playback module separately and independently, and command it to switch to the corresponding working mode (narrow beam / overlay beam), driving it to deliver precise directional voice to the designated "interactive area within the channel" or "pre-guided area outside the channel".

[0072] This module achieves unified scheduling and efficient collaboration of two types of information flows, two sets of decision-making logic, and two acoustic modes, ensuring that guidance for different scenarios inside and outside the channel can be executed seamlessly, in parallel, and without interference, thus realizing the intelligent, humanized, and highly efficient full-process voice guidance service pursued by the patent.

[0073] The implementation principle of this embodiment is as follows: Through the collaborative work of its various modules, the system achieves precise voice guidance for passengers in different areas surrounding the turnstile. Each module has a clear division of labor, accurately collecting information, dividing areas, identifying passenger characteristics, and generating targeted voice commands for delivery. Compared to existing turnstile voice guidance systems, this system better meets the needs of passengers in different areas, reduces passage obstruction, improves passage efficiency, and provides passengers with higher-quality passage guidance services, representing a significant improvement and contribution to existing turnstile voice guidance systems.

[0074] [Third Embodiment] A computer-readable storage medium is disclosed in the third embodiment of this application. The computer-readable storage medium is, for example, a non-volatile memory, such as magnetic media (e.g., hard disks, floppy disks, and magnetic tapes), optical media (e.g., CD-ROMs and DVDs), magneto-optical media (e.g., optical discs), and hardware devices specifically configured to store and execute computer-executable instructions (e.g., read-only memory (ROM), random access memory (RAM), flash memory, etc.). A computer program is stored on the computer-readable storage medium. The computer-readable storage medium can be executed by one or more processors or processing devices to implement the region-division-based gate dual-mode directional voice guidance method described in the foregoing embodiments.

[0075] Furthermore, it is understood that the foregoing embodiments are merely illustrative examples of the present invention. Provided that the technical features do not conflict, the structure is not contradictory, and the purpose of the invention is not violated, the technical solutions of the various embodiments can be arbitrarily combined and used.

[0076] In the embodiments provided by this invention, it should be understood that the disclosed methods, systems, and measuring devices can be implemented in other ways. For example, the modules included in the systems described above are merely illustrative, and the division of modules is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0077] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0078] Furthermore, in the various embodiments of the present invention, the functional units / modules can be integrated into one processing unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated into one unit / module. The integrated unit / module described above can be implemented in hardware or in the form of hardware plus software functional units / modules.

[0079] The integrated units / modules implemented as software functional units / modules described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause one or more processors of a computer measurement device (which may be a personal computer, server, or network measurement device, etc.) to execute some steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A dual-mode directional voice guidance method for turnstiles based on region division, characterized in that, include: Acquire images of the space surrounding the turnstile; Based on the spatial image and the first preset division ratio, an interactive area inside the channel and at least one pre-guided area outside the channel are divided. The interactive area inside the channel is used to perform facial verification on passengers in the interactive area inside the channel. The system obtains the verification signals of the passenger's face and ID from the interaction area within the channel, generates a first voice command based on the verification signals, and controls a first group of loudspeakers to project a narrow beam of voice with a first preset beam angle into the interaction area within the channel according to the first voice command. Image information matching the characteristics of special passengers is obtained from the pre-guidance area outside the passage. A corresponding second voice command is generated based on the image information. The second voice command controls the second set of loudspeakers to project a covering sound beam with a second preset sound beam angle onto the pre-guidance area outside the passage.

2. The method according to claim 1, characterized in that, The step of dividing the space image and the first preset division ratio into an in-channel interaction area and at least one out-of-channel pre-guidance area includes: Identify the gate channel boundary from the spatial image, and determine the interaction area within the channel according to the first preset division ratio; Based on the number of interaction areas and preset guidance areas within the channel and the second preset division ratio, at least one pre-guidance area outside the channel is determined, wherein the second preset division ratio is used to determine the spatial layout of at least one pre-guidance area outside the channel relative to the gate entrance.

3. The method according to claim 1, characterized in that, The step of acquiring the verification signal of the passenger's face and ID from the interaction area within the passage, generating a first voice command based on the verification signal, and controlling a first group of loudspeakers to project a narrow-beam voice with a first preset beam angle into the interaction area within the passage according to the first voice command includes: The verification signal generated by the face capture device and the document reader is acquired, wherein the verification signal includes at least one of a verification start signal, a verification success signal, or a verification failure signal; The corresponding broadcast text content and broadcast control parameters are retrieved from the preset voice command mapping table according to the verification signal to generate the first voice command. The broadcast control parameters include broadcast priority and / or broadcast repetition number. The first voice command controls the first group of loudspeakers to generate a narrow beam with the first preset beam angle, and modulates the text content to be broadcast onto the narrow beam for voice transmission.

4. The method according to claim 1, characterized in that, The step of acquiring image information matching the characteristics of special passengers from the pre-guidance area outside the passage, generating a corresponding second voice command based on the image information, and controlling a second group of loudspeakers to project a covering sound beam with a second preset sound beam angle onto the pre-guidance area outside the passage according to the second voice command includes: Obtain the image sequence in the pre-guided area outside the channel; The image sequence is identified to determine whether there is image information that matches the characteristics of a special passenger. If so, the corresponding broadcast text content and broadcast control parameters are retrieved from the preset voice command mapping table based on the identified image information to generate the corresponding second voice command. According to the second voice command, the second group of loudspeakers are controlled to project a sound beam with the second preset sound beam angle to the pre-guided area outside the channel.

5. The method according to claim 1, characterized in that, The special passenger characteristics include at least one of the following image recognition features: The following are considered as criteria for eligibility: carrying luggage exceeding a preset threshold, being elderly, using assistive devices, or exhibiting behaviors such as holding an infant or leading a young child, where the assistive devices include wheelchairs, canes, and strollers.

6. The method according to claim 1, characterized in that, The horizontal width of the first preset sound beam angle is less than or equal to 30 degrees, and the vertical width is less than or equal to 20 degrees. The horizontal width of the second preset sound beam angle is greater than 60 degrees.

7. A dual-mode directional voice guidance system for turnstiles based on region division, characterized in that, include: The channel verification module is used to verify the face and documents of passengers who have entered the gate passage area; The external sensing module is used to collect image data of the queuing area of ​​a preset length at the gate entrance to identify special passenger characteristics; The area division and recognition module is used to divide the area inside and outside the channel based on the image data and to identify the behavioral characteristics of queuing passengers. A directional audio playback module is used to generate and deliver directional sound beams with different beamwidths; The dual-mode voice control module is communicatively connected to the face verification module inside the channel, the perception module outside the channel, the area division and recognition module, and the directional audio playback module, respectively. It is used to receive two types of information to generate two independent sets of voice commands, and drive the directional audio playback module to project directional sound beams with different beamwidths to the areas inside and outside the channel, respectively. The two types of information include the status information corresponding to the passenger verification results inside the channel and the identification information of special passengers outside the channel.

8. The system according to claim 7, characterized in that, The channel verification module is integrated into the camera above the gate operation screen, and is used to collect facial images and verify documents of passengers who have entered the gate passage area. The external sensing module is installed inside the camera above the gate channel, and the field of view of the corresponding camera covers the queuing area of ​​the gate entrance with a preset length.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. An electronic device, characterized in that, include: Memory and processor, wherein the memory is used to store computer programs; The processor is configured to implement the steps of the method as described in any one of claims 1 to 6 when executing the computer program.

Images