AR scenic area guide method and system based on positioning

Abstract

The invention relates to the technical field of AR preview, in particular to a positioning-based AR scenic area guide method and system, and the method comprises the steps: obtaining the real-time three-dimensional position of a tourist through a Google ARCore fusion positioning engine based on a preset scenic area three-dimensional model and a space reference coordinate system; carrying out visual anchor point dynamic matching according to the real-time three-dimensional position, and identifying a current scene area corresponding to the current position of the tourist; performing virtual content scheduling and space rendering according to the current scene area, and generating space rendering content; and obtaining a visual attention behavior of a tourist on the space rendering content, screening out tourist-interested guide content according to the visual attention behavior, and generating a visual guide path according to the tourist-interested guide content. According to the method, by adjusting the loading and scheduling strategy of the virtual content, the requirements of personalized content display and smooth space guidance are met, and the experience feeling of tourists is enhanced.

Description

TECHNICAL FIELD

[0001] The application relates to the field of AR preview technology, in particular to an AR scenic spot guide method and system based on positioning. BACKGROUND

[0002] With the development of augmented reality (AR) and mobile positioning technology, AR scenic spot guide systems gradually replace traditional text or voice guide and become an important means to improve the experience of tourists. Some existing AR guide systems realize basic positioning through GPS and an inertial measurement unit (IMU) and trigger guide content through a preset recognition image or a 3D model.

[0003] There are many existing solutions for applying AR to scenic spot guide, such as the invention patent CN115187760A published on October 14, 2022, which discloses an AR-based scenic spot guide method and system, an electronic device, and a storage medium, which can be applied to the fields of artificial intelligence, cloud computing, mobile Internet, virtual reality, augmented reality, holographic technology, or finance. The method comprises the following steps: acquiring an image around a scenic spot collected by a tourist; extracting a scenic spot feature from the image around the scenic spot through a spatial point cloud algorithm; comparing the scenic spot feature with a pre-established spatial map to obtain an explanation video of the current scenic spot and update the current number of tourists visiting the current scenic spot; and acquiring a real-time navigation route, a dynamically planned route with fewer tourists at the scenic spot, and an explanation video of the scenic spot after the tourist arrives at the scenic spot through a camera on the mobile device of the tourist, thereby improving the user experience.

[0004] For example, the invention patent CN118484091A published on August 13, 2024 discloses a virtual scenic spot visiting guide system based on VR technology, which relates to the technical field of scenic spot visiting guide. The virtual scenic spot visiting guide system based on VR technology comprises the following steps: starting the system, entering the virtual scenic spot environment, selecting the guide function, selecting the scenic spot, planning the system guide route, starting the guide, guiding the scenic spot, interactive experience, voice guide, and ending the guide. The device used in the virtual scenic spot visiting guide system based on VR technology comprises a machine body, an adjusting mechanism is installed at the bottom end of the machine body, and symmetrical limiting mechanisms are installed on the side of the machine body away from the adjusting mechanism. The adjusting mechanism can adjust the light shield cover, so that the position of the light shield cover can be adjusted according to the head of the patient, and the light shield cover can be adjusted according to the size of the head of the user, so that it can be better worn, thereby improving the use experience.

[0005] While the technical solutions in the aforementioned patent documents all possess certain technical effects, they, like other existing technologies, still suffer from some problems. For example, visual anchor point matching is insufficient: current technologies primarily employ fixed feature templates or local descriptors (such as SIFT, ORB, etc.) to construct a static anchor point library. This leads to high mismatch rates or anchor point recognition failures when tourists rapidly switch perspectives, encounter image occlusion, or have complex backgrounds, affecting the scheduling and presentation of subsequent guided tour content. There is also insufficient acquisition of tourist interests and preferences: existing guided tour path construction largely relies on recommendations based on the shortest geographical distance or a small number of hotspot interest points, resulting in a mismatch between the guided tour path and tourists' actual preferences. This fails to meet the needs of personalized content display and smooth spatial guidance, leading to a poor tourist experience.

[0006] Therefore, there is an urgent need for a location-based AR scenic area navigation method and system. Summary of the Invention

[0007] Therefore, it is necessary to provide a location-based AR scenic area navigation method and system that can improve data processing efficiency to address the aforementioned technical problems.

[0008] The technical solution of this invention is as follows: A location-based AR scenic area navigation method, the method comprising: Based on the preset 3D model of the scenic area and spatial reference coordinate system, the real-time 3D position of tourists is obtained through the Google ARCore fusion positioning engine. Based on the real-time 3D position, visual anchor points are dynamically matched, and the current scene area corresponding to the tourist's current location is identified; Based on the current scene area, virtual content is scheduled and spatial rendering is performed, and spatial rendering content is generated; The system acquires visitors' visual attention behavior towards the rendered spatial content, filters out the tour content that visitors are interested in based on the visual attention behavior, and generates a visual guide path based on the tour content that visitors are interested in.

[0009] Specifically, based on the real-time three-dimensional position, dynamic matching of visual anchor points is performed, and the current scene area corresponding to the tourist's current location is identified, including: Obtain a set of candidate anchor points within a preset anchor point range for the real-time 3D position; the number of candidate anchor points in the set is multiple. Based on the tourist's current camera image, the number of anchor point pairs that successfully match the anchor points in the candidate anchor point set is selected. Calculate the visual anchor point matching score based on each candidate anchor point set, and locate the current scene area corresponding to the tourist's current location based on the calculated visual anchor point matching score.

[0010] Specifically, a visual anchor point matching score is calculated based on each candidate anchor point set, and the current scene area corresponding to the tourist's current location is located based on the calculated visual anchor point matching score, including: The highest visual anchor matching score is selected from the calculated visual anchor matching score values; Determine whether the maximum visual anchor point matching score value is greater than the preset matching score threshold; If the determination is yes, then the scene area corresponding to the highest visual anchor point matching score will be set as the current scene area corresponding to the tourist's current location.

[0011] Specifically, the system acquires visitors' visual attention behavior towards the rendered spatial content, and filters out tour content that visitors are interested in based on this visual attention behavior, including: Acquire tourists' visual attention behavior towards the rendered content of the space, and extract tourists' points of interest from the visual attention behavior; Based on the extracted points of interest of tourists, guide content that tourists are interested in is selected.

[0012] Specifically, based on the extracted points of interest of tourists, the guided tour content that tourists are interested in is selected, including: The points of interest of tourists are clustered based on a clustering algorithm, and clustered regions are generated. Each clustered region contains multiple points of interest. Calculate the interest point density of each interest point in the clustered region, and calculate the average density and region radius of the clustered region based on each interest point; The region attraction score is calculated based on the average density, the region radius, and a preset interest point distribution feature extraction model. Based on the area's attractiveness score, guided tour content that tourists are interested in is selected.

[0013] Specifically, the guided tour content that tourists are interested in includes multiple points of interest, and a visual guided path is generated based on the guided tour content that tourists are interested in, including: The guide path weights are generated based on the shortest distance between points of interest corresponding to the tour content that the tourists are interested in and the density difference of points of interest between them. A guidance path is generated based on the visitor's current location, the 3D coordinates of the target point of interest, and the weight of the guidance path. Based on the physics engine in Unity 3D and ARCore scene rendering technology, the guide path is dynamically presented on the visitor's electronic terminal.

[0014] Specifically, based on a pre-defined 3D model of the scenic area and a spatial reference coordinate system, the real-time 3D location of tourists is obtained through the Google ARCore fusion positioning engine, including: Construct a 3D model of the scenic area and establish a spatial reference coordinate system; Based on the Google ARCore fusion positioning engine, multi-source fusion positioning is performed according to the spatial reference coordinate system to obtain the real-time three-dimensional position of tourists.

[0015] Specifically, it also includes a location-based AR scenic area navigation system, the system comprising: The 3D location acquisition module is used to obtain the real-time 3D location of tourists based on a preset 3D model of the scenic area and a spatial reference coordinate system, using the Google ARCore fusion positioning engine. The current area recognition module is used to dynamically match visual anchor points based on the real-time three-dimensional position and identify the current scene area corresponding to the tourist's current location. The spatial rendering generation module is used to schedule virtual content and perform spatial rendering based on the current scene area, and generate spatial rendering content. The guided path generation module is used to obtain the visual attention behavior of tourists towards the spatial rendering content, filter out the guided content that tourists are interested in based on the visual attention behavior, and generate a visual guided path based on the guided content that tourists are interested in.

[0016] Specifically, the current area recognition module is further configured to: acquire a set of candidate anchor points within a preset anchor point range for the real-time 3D location, wherein the number of candidate anchor point sets is multiple; based on the tourist's current camera image, filter out the number of anchor point pairs that successfully match the anchor points in the candidate anchor point set; calculate a visual anchor point matching score value according to each candidate anchor point set, and locate the current scene area corresponding to the tourist's current location based on the calculated visual anchor point matching score value.

[0017] Specifically, the current region identification module is further configured to: filter out the largest visual anchor point matching score from the calculated visual anchor point matching score; determine whether the largest visual anchor point matching score is greater than a preset matching score threshold; if the determination is yes, then set the scene region corresponding to the largest visual anchor point matching score as the current scene region corresponding to the tourist's current location.

[0018] Specifically, the guide path generation module is also used to: obtain the visual attention behavior of tourists towards the spatial rendering content, extract the tourists' points of interest from the visual attention behavior, and filter out the tour content that tourists are interested in based on the extracted points of interest.

[0019] Specifically, the guided path generation module is further configured to: cluster tourists' points of interest based on a clustering algorithm and generate clustered regions, each of which contains multiple points of interest; calculate the regional attractiveness score of each point of interest in the clustered region based on the average density, the region radius, and a preset point of interest distribution feature extraction model; and select guided tour content that tourists are interested in based on the regional attractiveness score.

[0020] Specifically, the guided tour content that tourists are interested in includes multiple points of interest (POIs), and the guide path generation module is further used to: generate guide path weights based on the shortest distance between POIs corresponding to the guided tour content that tourists are interested in and the density difference of POIs between POIs; generate a guide path based on the tourist's current location, the three-dimensional coordinates of the target POI, and the guide path weights; and dynamically present the guide path on the tourist's electronic terminal based on the physics engine in the Unity 3D engine and the scene rendering technology of ARCore.

[0021] Specifically, the three-dimensional location acquisition module is also used to: construct a three-dimensional model of the scenic area and establish a spatial reference coordinate system; perform multi-source fusion positioning based on the spatial reference coordinate system using the Google ARCore fusion positioning engine, and acquire the real-time three-dimensional location of tourists.

[0022] Optionally, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps described in the positioning-based AR scenic spot guidance method.

[0023] Optionally, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps described in the positioning-based AR scenic spot navigation method.

[0024] The technical effects achieved by this invention are as follows: The aforementioned location-based AR scenic area navigation method and system sequentially obtains the real-time 3D location of tourists using the Google ARCore fusion positioning engine based on a preset 3D model of the scenic area and a spatial reference coordinate system. It then dynamically matches visual anchor points based on the real-time 3D location and identifies the current scene area corresponding to the tourist's current location. Based on the current scene area, it schedules virtual content and renders the space, generating rendered content. It acquires the tourist's visual attention behavior towards the rendered content, filters out navigation content of interest based on this behavior, and generates a visual guidance path based on this content. This application, through a visual anchor point matching model, employs a multi-source fusion positioning and dynamic visual anchor point scoring mechanism to achieve stable and high-precision scene recognition capabilities, effectively overcoming the problem of poor adaptability of static templates to environmental changes in existing methods. This improves the stability and accuracy of scene recognition and enhances adaptability to environmental changes. Furthermore, through an interest point distribution feature extraction model, combined with spatial clustering and density estimation, it can dynamically assess the tourist's interest preference for the area based on the actual spatial distribution and adjust the loading and scheduling strategies of virtual content accordingly. This meets the needs of personalized content display and smooth spatial guidance, enhancing the tourist experience. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating a location-based AR scenic area navigation method in one embodiment; Figure 2 This is a structural block diagram of a location-based AR scenic spot guide system in one embodiment. Detailed Implementation

[0026] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0027] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0028] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0029] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0030] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0032] In one embodiment, a terminal is provided, the terminal being used to: obtain the real-time 3D position of tourists using the Google ARCore fusion positioning engine based on a preset 3D model of the scenic area and a spatial reference coordinate system; dynamically match visual anchor points according to the real-time 3D position and identify the current scene area corresponding to the tourist's current location; perform virtual content scheduling and spatial rendering according to the current scene area and generate spatially rendered content; obtain the tourist's visual attention behavior towards the spatially rendered content, filter out the tour guide content that the tourist is interested in based on the visual attention behavior, and generate a visual guidance path based on the tour guide content that the tourist is interested in.

[0033] The terminal may be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices.

[0034] In one embodiment, such as Figure 1 As shown, a location-based AR scenic area navigation method is provided, the method including: Step S100: Based on the preset 3D model of the scenic area and the spatial reference coordinate system, obtain the real-time 3D position of the tourists through the Google ARCore fusion positioning engine; Step S200: Perform dynamic matching of visual anchor points based on the real-time three-dimensional position, and identify the current scene area corresponding to the tourist's current location; Step S300: Perform virtual content scheduling and spatial rendering based on the current scene area, and generate spatial rendering content; Step S400: Obtain the visual attention behavior of tourists towards the spatial rendering content, filter out the tour content that tourists are interested in based on the visual attention behavior, and generate a visual guidance path based on the tour content that tourists are interested in.

[0035] The location-based AR scenic area navigation method and system described in this application sequentially utilizes a preset 3D model of the scenic area and a spatial reference coordinate system, via Google... The ARCore fusion positioning engine acquires the real-time 3D location of tourists; it performs dynamic visual anchor point matching based on the real-time 3D location and identifies the current scene area corresponding to the tourist's current location; it performs virtual content scheduling and spatial rendering based on the current scene area and generates spatial rendering content; it acquires the tourist's visual attention behavior towards the spatial rendering content, filters out the tour content that the tourist is interested in based on the visual attention behavior, and generates a visual guidance path based on the tour content that the tourist is interested in. This application achieves stable and high-precision scene recognition capabilities through a visual anchor point matching model, adopting multi-source fusion positioning and a dynamic visual anchor point scoring mechanism, effectively overcoming the problem of poor adaptability of static templates to environmental changes in existing methods, thereby improving the stability and accuracy of scene recognition and improving adaptability to environmental changes; through an interest point distribution feature extraction model, combined with spatial clustering and density estimation, it can dynamically assess the tourist's interest preference for the area based on the actual spatial distribution, and adjust the loading and scheduling strategy of virtual content accordingly, meeting the needs of personalized content display and smooth spatial guidance, and enhancing the tourist experience.

[0036] In one embodiment, step S100: Based on a preset 3D model of the scenic area and a spatial reference coordinate system, the real-time 3D position of the tourist is obtained through the Google ARCore fusion positioning engine, including: Step S110: Construct a 3D model of the scenic area and establish a spatial reference coordinate system; Step S120: Based on the Google ARCore fusion positioning engine, perform multi-source fusion positioning according to the spatial reference coordinate system and obtain the real-time three-dimensional position of the tourist.

[0037] Specifically, in step S110, the goal of constructing the three-dimensional model of the scenic area is to represent the actual physical environment of the scenic area in digital form and transform it into three-dimensional data that is easy for the AR system to understand and operate.

[0038] The 3D model of the scenic area includes spatial elements such as buildings, terrain, and major attractions, ensuring a precise correspondence between virtual information and the real environment. Establishing the spatial reference coordinate system binds these 3D models to specific locations in the real world, ensuring the system can accurately locate every position within the scenic area. This step provides an accurate geographic basis for all subsequent location-based operations.

[0039] The construction of a three-dimensional environmental model of a scenic area relies on the digital capture of the physical environment of the scenic area. Common methods include lidar scanning, photogrammetry, satellite or drone photography, etc.

[0040] Among them, lidar obtains precise distance information of objects in a scenic area by emitting a laser beam and calculating the time difference of the reflected laser. It uses this reflection data to generate a very dense and accurate point cloud dataset, which is used to reconstruct the three-dimensional environment of the scenic area, including elements such as buildings, trees, and mountains. It has the advantages of high precision and high efficiency, and is especially suitable for complex terrain and high-rise buildings. However, the amount of data generated is huge, and effective data processing is required.

[0041] Photogrammetry generates 3D models using high-resolution aerial images or multi-angle images taken by ground cameras, employing image recognition and computer vision technologies. It has the advantages of low cost and suitability for surveying large-scale scenic areas, but its data density is lower and its accuracy is slightly inferior to that of LiDAR.

[0042] Satellite or drone photography quickly acquires high-resolution images or point cloud data by scanning scenic areas on-site. It has a wide coverage area and can collect panoramic data of large scenic areas, making it more flexible. However, its accuracy is limited by the stability of the aerial photography platform and the quality of the sensors.

[0043] After the 3D model of the scenic area is created, a spatial reference coordinate system needs to be established to map all 3D objects in the scenic area to real geographic coordinates. This allows the system to identify the specific location of tourists in the real world and accurately overlay virtual content onto the physical scene. Depending on the scale of the scenic area and actual needs, the coordinate system can be either a global coordinate system or a local coordinate system.

[0044] The global coordinate system is suitable for modeling large-scale scenic areas and can directly cooperate with the Global Positioning System (GPS). By mapping the position of each object in the scenic area to the global coordinate system, seamless cooperation with GPS is ensured, providing tourists with global positioning and navigation. The local coordinate system is used for smaller scenic areas, or when higher-precision spatial data is required. This method does not rely on the global coordinate system, is relatively simple to implement, and can adapt to the local environmental data of the scenic area.

[0045] When using LiDAR or UAV data, there is often a conversion between the sensor coordinate system and the actual geographic coordinate system. This requires calibration algorithms, combined with GPS data or known reference point locations, to perform the coordinate system conversion. These include external calibration methods (setting calibration points at known ground coordinates to match sensor data from LiDAR / camera / UAVs with the actual geographic coordinate system) and ICP (for point cloud data registration, suitable for multi-sensor fusion) algorithms. Furthermore, because data generated by different sensors (such as LiDAR, photogrammetry, and UAVs) may be inconsistent, point cloud registration algorithms (such as ICP and NDT) and filtering techniques (such as RANSAC) are needed to optimize this data and ensure the accuracy of the 3D model.

[0046] Furthermore, in real-world scenarios, due to the limitations of a single sensor (such as GPS), it is impossible to provide sufficiently accurate location data indoors or in complex terrain. In order to improve positioning accuracy, reduce errors, and ensure efficient navigation, multi-source fusion positioning is achieved by using Google ARCore's fusion positioning engine. This integrates data from multiple sensors (such as GPS, IMU, and visual sensors) and combines this data with the 3D environment model and spatial reference coordinate system established in step S1. This enables high-precision positioning and navigation. In other words, in step S120, multi-source fusion positioning is performed based on the spatial reference coordinate system using the Google ARCore fusion positioning engine, and the real-time 3D position of the visitor is obtained.

[0047] Specifically, Google ARCore's fusion positioning engine relies on the visual-inertial fusion positioning (VIO) method, which combines visual information (from cameras) and data from the inertial measurement unit (IMU) to achieve accurate estimation of the device's position and attitude. This can help compensate for potential interference with GPS signals in scenic areas (such as obstruction by buildings and trees). By combining this visual data with the 3D environment model, ARCore's positioning engine enables the system to perform more stable and accurate positioning, thereby obtaining a more accurate 3D position.

[0048] Furthermore, the visitor's real-time location changes continuously while walking or driving. Through Google ARCore's fusion positioning engine, the visitor's position is dynamically corrected each time new location information is updated, combining the spatial reference coordinate system and the environmental model to ensure that the virtual information during the guided tour is accurately superimposed onto the real-world scene.

[0049] In one embodiment, step S200: dynamically matching visual anchor points based on the real-time 3D position and identifying the current scene area corresponding to the tourist's current location, including: Step S210: Obtain a set of candidate anchor points within the preset anchor point range for the real-time 3D position, wherein the number of candidate anchor points in the set is multiple; Step S220: Based on the tourist's current camera image, filter out the number of anchor point pairs that successfully match the anchor points in the candidate anchor point set; Step S230: Calculate the visual anchor point matching score based on each candidate anchor point set, and locate the current scene area corresponding to the tourist's current location based on the calculated visual anchor point matching score.

[0050] In this embodiment, the preset anchor point range is pre-set, specifically a range formed with the real-time 3D position as the center and a preset distance R as the radius. Multiple candidate anchor point sets within the preset anchor point range corresponding to the real-time 3D position are selected from a pre-set anchor point database. Next, features are extracted from the tourist's current camera image, and nearest neighbor matching is performed with anchor points in each candidate scene anchor point set to form a set of anchor points that successfully match the anchor points in the candidate anchor point set. The number of anchor point pairs that successfully match the anchor points in the candidate anchor point set is then selected. Finally, a visual anchor point matching score is calculated based on a preset visual anchor point matching model, and the current scene area corresponding to the tourist's current location is located based on the calculated visual anchor point matching score.

[0051] The visual anchor point matching model is as follows: ,in, The visual anchor point matching score is used to determine the confidence level of the match between the current frame image and anchor points in a candidate anchor point set. This represents the number of anchor pairs that successfully match the current image with anchors in the candidate anchor set. This represents the total number of anchor points in the candidate anchor point set. The matching density metric measures the proportion of matching points to the total number of anchor points. The structure distribution factor indicates whether the matching anchor points are uniformly distributed on the structure of the reference image. The structural saliency factor represents the importance of the location of the matching point in the visual structure. , and For pre-set weight parameters, weight parameters , and The sum is 1. For example, the weight parameters... , and The values ​​are 0.2, 0.5 and 0.3 respectively.

[0052] Specifically, for the structural distribution factor This value measures the spatial uniformity of matching anchor points. A value closer to 1 indicates a more balanced distribution of matching points within the image structure, thus improving the robustness of matching decisions. (Structure Distribution Factor) The specific calculations are as follows: ,in, This represents the average Euclidean distance of the matched anchor points in the candidate anchor point set relative to the image center in the reference image. is the standard deviation of the mean Euclidean distance.

[0053] Specifically, The calculation formula is: ,in, Let be the Euclidean distance of the i-th matching anchor point relative to the center of the image. The calculation formula is: .

[0054] Specifically, for structural significance factors This metric is used to quantify the visual structural importance or discriminative power of the region where the matching point is located. A larger value indicates that the points matching the candidate anchor point in the current frame image are more concentrated in structurally important and highly discriminative locations, thus enhancing the matching reliability. The calculation formula is as follows: ,in, To assign semantic saliency weight to the i-th matching anchor point in the reference image, an image structure analysis network is used to hierarchically label structural elements in the image, such as building edges, mountain outlines, and sculpture edges, assigning different saliency weights to different locations. Feature points located on semantically critical edges are given greater weight, while those located in textured areas, such as grass or walls, are given less weight. The image structure analysis network can be a convolutional network (GCN) or an image segmentation algorithm, as shown in the figure. Semantically critical edges include the top lines of buildings and the outlines of stone monuments.

[0055] Existing scene recognition methods, including ORB / SIFT+RANSAC which rely solely on the number of local image matches for scene confirmation, those that depend on entity markers and static labels for scene recognition, and those that use panoramic recognition for scene confirmation, are difficult to stably identify sub-scenes under conditions of high repetition textures, partial occlusion, and changing lighting. They suffer from poor adaptability, high cost, and susceptibility to occlusion.

[0056] In this embodiment, multi-dimensional structural modeling is used to improve the robustness of recognition and judgment. Furthermore, the comparison area is dynamically filtered according to the tourist's current three-dimensional position, which has higher spatial matching efficiency and deployment flexibility. It also improves the adaptability to uncertain content such as occlusion and lighting. Moreover, the semantic structure-guided matching method is more in line with the complex background features of scenic area buildings and natural landscapes.

[0057] Therefore, based on the acquired real-time 3D location of tourists, a visual anchor point matching model is used to dynamically match visual anchor points to further identify the specific sub-scene area where the tourists are located, thereby achieving scene recognition and obtaining scene recognition results. After acquiring the real-time 3D location of tourists, it is necessary to further accurately identify the actual scene location where the tourists are currently located in order to achieve binding with the virtual tour content. This step achieves a dynamically adaptive scene recognition method by jointly judging image semantic features and spatial data, enhancing the system's robustness and accuracy, and significantly reducing deployment and maintenance costs.

[0058] In one embodiment, step S230: calculating a visual anchor point matching score based on each candidate anchor point set, and locating the current scene area corresponding to the tourist's current location based on the calculated visual anchor point matching score, includes: Step S231: Select the highest visual anchor point matching score from the calculated visual anchor point matching score values; Step S232: Determine whether the maximum visual anchor point matching score is greater than the preset matching score threshold; Step S233: If the determination is yes, then set the scene area corresponding to the largest visual anchor point matching score to the current scene area corresponding to the tourist's current location.

[0059] In this embodiment, the obtained maximum visual anchor point matching score is compared with a set threshold. If the score is greater than the threshold, it is determined that the tourist is located in the corresponding scene area, that is, the maximum visual anchor point matching score is selected from the calculated visual anchor point matching scores. It is then determined whether the maximum visual anchor point matching score is greater than the preset matching score threshold. If it is, the scene area corresponding to the maximum visual anchor point matching score is set as the current scene area corresponding to the tourist's current location. If it is not, the previous recognition state is maintained or it is marked as unknown.

[0060] In one embodiment, step S300: Perform virtual content scheduling and spatial rendering based on the current scene area, and generate spatial rendering content, as follows: In this embodiment, the goal of virtual content scheduling is to correctly load and render guide content, such as attraction descriptions, route directions, and historical background, within the visitor's visual field based on their current perspective and location. Spatial rendering, on the other hand, ensures the seamless integration of this virtual content with the real environment, achieving a natural and smooth augmented reality experience and presenting virtual information from the visitor's current perspective. This embodiment uses the Unity3D engine combined with ARCore's rendering module to implement virtual content scheduling and spatial rendering.

[0061] The Unity3D engine is a very popular game and augmented reality development engine, especially suitable for the creation and rendering of virtual content, including the creation and management of virtual content, the scheduling of virtual content, the optimization of virtual content animation and interaction, and rendering effects. Using the Unity3D engine, those skilled in the art can create rich virtual content, such as virtual characters, buildings, and text labels, and organize these virtual objects into manageable components and scenes, flexibly controlling which virtual objects need to be loaded and which need to be unloaded. When a visitor approaches a specific attraction or enters a scene area, the Unity3D engine's event management system triggers the loading and unloading of the corresponding virtual content. This process is based on the current scene area corresponding to the visitor's current location obtained in step S200, ensuring that each virtual object appears at the correct time and location. The Unity3D engine can not only render static virtual content but also supports dynamic animation of virtual objects and visitor interaction processing such as touch, drag, and rotation. For example, a visitor can click on a virtual building model to view its detailed information or control the rotation or scaling of virtual content through gestures. The Unity3D engine has a built-in rendering system that can handle rendering effects such as lighting, shadows, and reflections, ensuring a natural transition of virtual content in the real environment.

[0062] ARCore is an augmented reality SDK launched by Google, specifically designed to support AR experiences on Android devices for spatial rendering. ARCore uses technologies such as motion tracking, environment understanding, and lighting estimation to perceive the visitor's environment in real time, providing Unity3D with accurate spatial references to ensure virtual content displays correctly in the real-world environment. ARCore's ARAnchor function binds scene areas to virtual content, ensuring virtual objects always remain in the correct position. For example, when a visitor moves, the virtual content updates with their viewpoint, maintaining stability. ARCore's depth perception function can identify the relative positions of virtual objects and real-world objects in real time, thus handling occlusion. For example, when virtual guide information appears in front of a visitor, ARCore ensures that the information is not obscured by real-world objects, remaining clearly visible. ARCore can provide real-time lighting information for virtual content through ambient light estimation, ensuring that the lighting of the virtual content matches the real-world environment. This allows the brightness, shadows, and other effects of virtual objects to dynamically adjust with changes in ambient light, making the virtual content blend more seamlessly with the real-world environment.

[0063] In one embodiment, step S400: obtaining the visual attention behavior of tourists towards the spatial rendering content, and filtering out the tour content that tourists are interested in based on the visual attention behavior, including: Step S410: Obtain the visual attention behavior of tourists towards the rendered spatial content, and extract the tourists' points of interest from the visual attention behavior; Step S420: Filter out the tour content that tourists are interested in based on the extracted tourist interest points.

[0064] In this embodiment, by first acquiring the visual attention behavior of tourists towards the rendered spatial content, the tourist's points of interest are extracted from the visual attention behavior; then, based on the extracted tourist's points of interest, the tour content that the tourist is interested in is filtered out, thereby achieving content that can better match the user's interests and improve the experience.

[0065] In one embodiment, step S420: filtering out tour content that tourists are interested in based on the extracted tourist interest points includes: Step S421: Cluster the tourists' points of interest based on the clustering algorithm and generate cluster regions, each of which contains multiple points of interest; Step S422: Calculate the interest point density of each interest point in the clustered region, and calculate the average density and region radius of the clustered region based on each interest point; Step S423: Calculate the region attractiveness score based on the average density, the region radius, and the preset interest point distribution feature extraction model; Step S424: Based on the area attraction score, filter out the guided tour content that tourists are interested in.

[0066] In this embodiment, a preset spatial clustering algorithm is first used to spatially cluster the tourists' points of interest, resulting in different clustered regions. Each region contains a relatively dense set of points of interest, which is the clustered region. Next, the point of interest density of each point of interest in the clustered region is calculated, and the average density and region radius of the clustered region are calculated based on each point of interest.

[0067] The interest point density is calculated using a kernel density estimation method in the prior art, and the specific calculation formula is as follows: ,in, Interest point density, used to represent the degree of concentration of interest points around a given point. The number of points of interest in the clustered region. Let i be the three-dimensional spatial coordinates of the point of interest. Let j be the three-dimensional spatial coordinates of the point of interest. The bandwidth parameter of the Gaussian kernel function is... Let i be the three-dimensional spatial coordinates of the point of interest. The three-dimensional spatial coordinates of the point of interest j The Euclidean distance between them.

[0068] The average density of the clustered regions is calculated based on the following formula: ,in, The average density of the clustered regions.

[0069] The radius of the region is calculated based on the following formula: ,in, For the region radius, The centroid of the region is represented by the mean coordinates of the three-dimensional spatial coordinates of all points of interest. The location of the center of gravity of the region 3D spatial coordinates of point of interest i The Euclidean distance between them.

[0070] Then, based on the following formula, the regional attractiveness score is calculated according to the average density, the region radius, and the preset interest point distribution feature extraction model: ,in, The regional attractiveness score represents the degree of interest and preference tourists have for that region. average density The normalized density, after normalization, is obtained by comparing the average density of interest points within the region with the maximum average density of interest points across all regions. The radius of the region represents its spatial extent. This represents the maximum radius of all areas within the scenic area.

[0071] Finally, based on the area attractiveness score, guided tour content that tourists are interested in is selected. Areas with higher attractiveness scores indicate that tourists have a higher interest in those areas and that they better match tourists' interests and preferences. Based on the attractiveness scores of these areas of interest, subsequent virtual content scheduling is optimized, prioritizing the loading of guided tour content or augmented reality information for areas that tourists are interested in.

[0072] Therefore, in this embodiment, after spatial rendering is completed, the point of interest distribution features of tourists' visual attention behavior are extracted using an interest point distribution feature extraction model based on the spatial rendering content. This yields tourists' interest preferences, which are then used to optimize subsequent virtual content scheduling and guided tour route recommendations. Guided tour content or augmented reality information for areas that tourists are interested in is prioritized, thereby enhancing tourists' immersion and interactive experience.

[0073] Existing methods for calculating tourist preferences typically rely on distance or nearby points. These methods are ill-suited to complex geographical environments, lacking adaptability and flexibility. Furthermore, traditional methods fail to accurately reflect the distribution characteristics of tourist points of interest within scenic areas. In contrast, this model combines spatial clustering and density estimation, making it more flexible, adaptable to complex geographical environments, and capable of handling irregularly distributed point-of-interest data. It can also dynamically assess tourists' interest in a region based on its actual spatial distribution and adjust virtual content loading and scheduling strategies accordingly. This allows it to handle different types of attractions and points of interest, providing refined spatial analysis and improving scalability and accuracy.

[0074] In one embodiment, the guided tour content of interest to tourists includes multiple points of interest. Step S400, generating a visual guide path based on the guided tour content of interest to tourists, includes: Step S431: Generate guide path weights based on the shortest distance between points of interest corresponding to the tour content that the tourists are interested in and the density difference of points of interest between them; Step S432: Generate a guidance path based on the visitor's current location, the 3D coordinates of the target point of interest, and the guidance path weight; Step S433: Based on the physics engine in Unity 3D engine and ARCore scene rendering technology, dynamically present the guide path on the visitor's electronic terminal.

[0075] In this embodiment, step S431 specifically includes: generating guide path weights based on the following formula, according to the shortest distance between points of interest corresponding to the tour content of interest to the tourist, the density difference of points of interest between points of interest, and the tourist preference matching degree: ,in, Let i be the weight of the guiding path between interest point i and interest point j, where i and j belong to the set of all interest points on the path from the visitor's current location to the target interest point. Let i be the distance between points of interest i and j. This represents the maximum distance between two points of interest, used to normalize the distance between these points. The difference in interest point density between points of interest within the region, where, , Let be the average density of all interest points in regions i and j, respectively. This represents the maximum value of the density difference between interest points in two regions, used to normalize the density difference of interest points within the regions. To match tourist preferences, This is a vector of historical points of interest visited by tourists, where each dimension represents whether the tourist has visited or preferred a certain type of point of interest. Let i be the attribute vector of interest point i, with dimensions and same, The total number of points of interest preferred by tourists is used to normalize the tourist preference matching degree. As the first weight, As the second weight, As a third weight, the specific value is preset by those skilled in the art, and this application does not make specific limitations or examples.

[0076] Furthermore, the guiding path is generated based on the following formula: ,in, To guide the path, The three-dimensional spatial coordinates of the tourist's target point of interest. Let i be the weight of the guiding path between interest point i and interest point j, where i and j belong to the set of all interest points on the path from the visitor's current location to the target interest point.

[0077] Existing technologies typically focus only on the shortest distance or the least time for a route, failing to meet the diverse needs of tourists. This embodiment, however, comprehensively considers the distance of the route, the density of points of interest, and tourist preferences, enabling route generation to adapt to tourists' interests and needs in real time. This makes the guided route more personalized, enhances the tourist experience, and better meets tourist requirements. Finally, through the physics engine in the Unity 3D engine and ARCore's scene rendering technology, the guided route is dynamically presented, and information such as the route direction and distance is displayed on the tourist's screen, enhancing the interactivity and immersion of the guided tour.

[0078] In one embodiment, such as Figure 2 As shown, a location-based AR scenic area navigation system is also provided, the system comprising: The 3D location acquisition module is used to obtain the real-time 3D location of tourists based on a preset 3D model of the scenic area and a spatial reference coordinate system, using the Google ARCore fusion positioning engine. The current area recognition module is used to dynamically match visual anchor points based on the real-time three-dimensional position and identify the current scene area corresponding to the tourist's current location. The spatial rendering generation module is used to schedule virtual content and perform spatial rendering based on the current scene area, and generate spatial rendering content. The guided path generation module is used to obtain the visual attention behavior of tourists towards the spatial rendering content, filter out the guided content that tourists are interested in based on the visual attention behavior, and generate a visual guided path based on the guided content that tourists are interested in.

[0079] In another embodiment, the current area recognition module is further configured to: acquire a set of candidate anchor points within a preset anchor point range for the real-time three-dimensional location, wherein the number of candidate anchor point sets is multiple; based on the tourist's current camera image, filter out the number of anchor point pairs that successfully match the anchor points in the candidate anchor point set; calculate a visual anchor point matching score value according to each candidate anchor point set, and locate the current scene area corresponding to the tourist's current location based on the calculated visual anchor point matching score value.

[0080] In another embodiment, the current region identification module is further configured to: filter out the largest visual anchor matching score from the calculated visual anchor matching score values; determine whether the largest visual anchor matching score is greater than a preset matching score threshold; if the determination is yes, then set the scene region corresponding to the largest visual anchor matching score as the current scene region corresponding to the tourist's current location.

[0081] In another embodiment, the guide path generation module is further configured to: acquire the visual attention behavior of tourists towards the spatial rendering content, extract the tourists' points of interest from the visual attention behavior, and filter out the tour content that tourists are interested in based on the extracted points of interest.

[0082] In another embodiment, the guide path generation module is further configured to: cluster tourists' points of interest based on a clustering algorithm and generate clustered regions, each of the clustered regions containing multiple points of interest; calculate the regional attractiveness score of each point of interest in the clustered region based on the average density, the region radius, and a preset point of interest distribution feature extraction model; and filter out tour guide content that tourists are interested in based on the regional attractiveness score.

[0083] In another embodiment, the tour content of interest to tourists includes multiple points of interest, and the guide path generation module is further configured to: generate guide path weights based on the shortest distance between the points of interest corresponding to the tour content of interest to tourists and the density difference of the points of interest between the points of interest; generate a guide path based on the tourist's current location, the three-dimensional coordinates of the target point of interest and the guide path weights; and dynamically present the guide path on the tourist's electronic terminal based on the physics engine in the Unity 3D engine and the scene rendering technology of ARCore.

[0084] In another embodiment, the three-dimensional location acquisition module is further used to: construct a three-dimensional model of the scenic area and establish a spatial reference coordinate system; perform multi-source fusion positioning based on the spatial reference coordinate system using the Google ARCore fusion positioning engine, and acquire the real-time three-dimensional location of the tourists.

[0085] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps described in the location-based AR scenic spot navigation method above.

[0086] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps described in the location-based AR scenic area navigation method.

[0087] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0088] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0089] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0090] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0091] This application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above method embodiments.

[0092] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0093] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to implement the steps described in the above-described method embodiments.

[0094] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0095] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0096] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0097] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and 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 through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0098] 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.

[0099] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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. Such 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 this application, and should all be included within the protection scope of this application.

[0100] One embodiment of this application also provides a computer device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above-described methods.

[0101] The computer device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above description is an example of a computer device and does not constitute a limitation on the computer device. It may include more or fewer components than described above, or a combination of certain components, or different components, such as input / output devices, network access devices, etc.

[0102] The processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0103] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard drive or RAM. In other embodiments, the memory may be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory may include both internal and external storage units of the computer device. The memory is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory can also be used to temporarily store data that has been output or will be output.

[0104] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0105] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for AR scenic area tour based on positioning, characterized in that, The method comprises: Based on the preset three-dimensional model of the scenic spot and the spatial reference coordinate system, the real-time three-dimensional position of the tourist is obtained through the Google ARCore fusion positioning engine; According to the real-time three-dimensional position, the visual anchor point dynamic matching is carried out, and the current scene area corresponding to the current position of the tourist is identified; According to the current scene area, virtual content scheduling and space rendering are carried out, and space rendering content is generated; Obtain the visual attention behavior of the tourist to the space rendering content, and filter out the tourist interested tour content according to the visual attention behavior, and generate a visual guide path according to the tourist interested tour content. 2.The AR scenic spot tour guide method based on positioning according to claim 1, wherein, According to the real-time three-dimensional position, the visual anchor point dynamic matching is carried out, and the current scene area corresponding to the current position of the tourist is identified, which comprises: Obtain the candidate anchor point set of the real-time three-dimensional position in the preset anchor point range, and the number of the candidate anchor point set is multiple; Based on the current camera image of the tourist, the number of anchor points successfully matched with the anchor points in the candidate anchor point set is filtered out; According to each of the candidate anchor point set, the visual anchor point matching score value is calculated, and the current scene area corresponding to the current position of the tourist is located according to the calculated visual anchor point matching score value. 3.The AR scenic spot tour guide method based on positioning according to claim 2, characterized in that, According to each of the candidate anchor point set, the visual anchor point matching score value is calculated, and the current scene area corresponding to the current position of the tourist is located according to the calculated visual anchor point matching score value, which comprises: From the calculated visual anchor point matching score value, the maximum visual anchor point matching score value is filtered out; Determine whether the maximum visual anchor point matching score value is greater than the preset matching score threshold value; If it is determined to be yes, the scene area corresponding to the maximum visual anchor point matching score value is set as the current scene area corresponding to the current position of the tourist. 4.The AR scenic spot tour guide method based on positioning according to claim 1, wherein, Obtain the visual attention behavior of the tourist to the space rendering content, and filter out the tourist interested tour content according to the visual attention behavior, which comprises: Obtain the visual attention behavior of the tourist to the space rendering content, and extract the interest points of the tourist from the visual attention behavior; According to the extracted interest points of the tourist, the tourist interested tour content is filtered out.

5. The AR scenic spot tour guide method based on positioning according to claim 4, characterized in that, According to the extracted interest points of the tourist, the tourist interested tour content is filtered out, which comprises: Based on the clustering algorithm, the interest points of the tourist are clustered, and a clustering area is generated, each of the clustering area contains multiple interest points; The interest point density of each interest point in the clustering area is calculated, and the average density and the area radius of the clustering area are calculated according to each of the interest points; According to the average density, the area radius and the preset interest point distribution feature extraction model, the area attraction score is calculated; According to the area attraction score, the tourist interested tour content is filtered out. 6.The AR scenic spot tour guide method based on positioning according to claim 1, wherein, The tourist interested tour content comprises multiple interest points, and a visual guide path is generated according to the tourist interested tour content, which comprises: According to the shortest distance between the interest points corresponding to the tourist interested tour content, the interest point density difference between the interest points, a guide path weight is generated; According to the three-dimensional coordinates of the current position of the tourist, the target interest point and the guide path weight, a guide path is generated; Based on the physical engine in the Unity 3D engine and the scene rendering technology of ARCore, the guide path is dynamically presented on the electronic terminal of the visitor. 7.The AR scenic spot tour guide method based on positioning according to claim 1, wherein, Based on the preset three-dimensional model of the scenic spot and the spatial reference coordinate system, the real-time three-dimensional position of the visitor is obtained through the Google ARCore fusion positioning engine, including: constructing a three-dimensional model of the scenic spot and establishing a spatial reference coordinate system; Based on the spatial reference coordinate system, the Google ARCore fusion positioning engine performs multi-source fusion positioning and obtains the real-time three-dimensional position of the visitor.

8. A positioning-based AR scenic area guide system, characterized in that, The system comprises: a three-dimensional position acquisition module configured to obtain the real-time three-dimensional position of the visitor based on the preset three-dimensional model of the scenic spot and the spatial reference coordinate system through the Google ARCore fusion positioning engine; a current area identification module configured to dynamically match a visual anchor point according to the real-time three-dimensional position and identify a current scene area corresponding to the current position of the visitor; a spatial rendering generation module configured to schedule virtual content and perform spatial rendering according to the current scene area and generate spatial rendering content; a guide path generation module configured to obtain a visual attention behavior of the visitor to the spatial rendering content, filter out visitor interested tour content according to the visual attention behavior, and generate a visual guide path according to the visitor interested tour content. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method of any one of claims 1 to 7.

Images