BIM Model Texture Dynamic Display Method, System, Device and Storage Medium

Abstract

The present invention discloses a method, system, device and storage medium for dynamically displaying BIM model textures, which relates to the technical field of BIM image dynamic processing. It includes constructing a BIM model containing component texture information, and real-time collecting key frame images of the video stream through a high-resolution camera; establishing a virtual camera in the BIM model, matching the virtual camera with the real camera, and automatically extracting and dynamically replacing the component textures in the BIM model; and dynamically displaying the BIM model textures in real time through the Web end. The method of the present invention realizes the data docking between components and real images by constructing a BIM model containing component texture information and combining with the collection of key frame images by a high-resolution camera. By matching the parameters of the virtual camera and the real camera in the BIM model, the real-time performance and automation degree of texture update are improved. Based on the Web end visualization display, the component texture information is displayed in real time, enhancing the user's interactive experience.

Description

Technical Field

[0001] The present invention relates to the technical field of BIM image dynamic processing, and specifically to a method, system, device and storage medium for dynamically displaying BIM model textures. Background Technique

[0002] With the wide application of building information modeling technology, the concept of digital management throughout the building life cycle has become increasingly popular. As the core supporting technology for the digital transformation of the construction industry, BIM is gradually extending from the design stage to the construction, operation and maintenance, and even the decommissioning stage. BIM is not only a three-dimensional modeling tool, but also an integrated platform containing multi-dimensional data such as spatial geometry, component attributes, engineering logic, and operation and maintenance information. In recent years, with the development of cloud computing, the Internet of Things (IoT), artificial intelligence (AI), and computer graphics and image processing technologies, the integrated application of BIM and external perception data has become a research hotspot. Especially in visualization, intelligent operation and maintenance, and construction site feedback, more and more scholars and enterprises are committed to synchronizing data and visually coupling BIM with the real environment. The texture display in traditional BIM models mostly relies on static texture mapping. At the initial stage of modeling, materials and textures are manually attached to enhance the visual realism of the model. However, this static binding method cannot reflect the real-time state changes of the building during use, nor can it respond to the actual changing needs of the construction environment.

[0003] There are several technical bottlenecks in the texture display of existing BIM systems. First, the component textures of current BIM models are mostly manually set once during the modeling stage, lacking dynamic linkage with the real building scene, and unable to reflect the state changes such as pollution, aging, damage, or repair of the building surface material over time, which limits the authenticity and practicality of BIM in the operation and maintenance visualization scenario. Second, the high-frequency updated camera video stream in the real environment has not been effectively connected to the BIM texture system, lacking a complete automated link from image extraction, component matching to texture replacement. Currently, most systems still rely on manual processing of key-frame images, which is inefficient, subjective, error-prone, and not suitable for the automated operation and maintenance of large-scale building complexes. Third, traditional image matching methods are mostly based on feature point alignment or full-image similarity calculation, and cannot accurately extract the texture area of specific BIM components in the image. Especially in the case of camera angle transformation, light interference, or occlusion, the recognition accuracy of traditional methods drops significantly.

[0004] In addition, in terms of visualization terminals, most existing systems are based on local desktop applications and cannot achieve real-time synchronous display of dynamic textures. Even if some platforms provide Web browsing interfaces, they lack a lightweight texture refresh mechanism and cannot achieve smooth component texture switching and rendering updates during user operations. The present invention constructs a virtual-real camera parameter matching mechanism, automatically extracts the texture regions corresponding to components in key frames, completes automatic recognition and replacement in combination with component IDs, and realizes dynamic loading and periodic refresh display of textures through the Web side, thereby establishing a dynamic coupling path at the texture level between the model and reality, improving the automation level, component-level real reflection ability, and user interaction experience of the system. Compared with traditional BIM systems, the present invention has achieved comprehensive optimization in terms of automatic extraction of component textures, accuracy of replacement logic, consistency of virtual-real perspective mapping, and lightweight rendering strategy on the Web side, and has extremely high practical popularization and engineering application value. Summary of the Invention

[0005] In view of the problems existing in the above-mentioned prior art, the present invention is proposed.

[0006] Therefore, the technical problems solved by the present invention are: the existing BIM model texture display methods have problems such as lagging updates, high manual dependence, inaccurate texture mapping, low system real-time performance, and inability to reflect the actual state changes of building components, and how to realize automatic extraction, accurate matching, and dynamic replacement of component textures based on real-time video images collected by cameras, and efficient visualization display through the Web side.

[0007] To solve the above technical problems, the present invention provides the following technical solutions. A method for dynamically displaying BIM model textures includes: constructing a BIM model containing component texture information, and real-time collecting key frame images of a video stream through a high-resolution camera; establishing a virtual camera in the BIM model, matching the virtual camera with the real camera, and automatically extracting and dynamically replacing the textures of components in the BIM model; dynamically displaying the BIM model textures in real time through a Web end. Real-time collecting key frame images of a video stream includes presetting high-resolution camera nodes in a building space, configuring camera collection parameters and image clarity thresholds, automatically extracting key frame images from the video stream at set time intervals, and renaming and classifying the qualified key frame images according to timestamps and storing them in a preset image file path according to preset conditions. Automatically extracting and dynamically replacing the textures of components in the BIM model includes constructing a corresponding virtual camera in the BIM model based on the perspective of the real camera, obtaining the visible area and component ID of the component in the current perspective, segmenting the collected key frame images, extracting the image areas corresponding to the BIM components, performing pixel-level mapping between the image coordinates and the model component coordinates, locating the original texture in the model texture file set according to the component ID, automatically adjusting the extracted image to the map size and format, synchronously writing the updated image into the component material attributes, automatically completing texture replacement, and recording the update timestamp.

[0008] As a preferred solution of the method for dynamically displaying BIM model textures according to the present invention, wherein: constructing the BIM model containing component texture information includes using Revit software to build a BIM model, setting texture categories and material parameters in the component property interface according to the building design drawings, attaching the texture information to the components in the BIM model, performing coordinate mapping processing on the texture images, and converting the BIM model into the IFC data format.

[0009] As a preferred solution of the method for dynamically displaying BIM model textures according to the present invention, wherein: real-time collecting key frame images of a video stream through a high-resolution camera includes arranging a high-resolution camera in a building space, recording the current three-dimensional coordinate information and shooting angle parameters, collecting real-time video stream data, and storing it in a preset database.

[0010] As a preferred solution of the method for dynamically displaying BIM model textures according to the present invention, wherein: collecting real-time video stream data includes extracting a preset number of pictures from the video stream data at a preset interval time to obtain a picture sequence that meets the preset conditions, extracting the texture information in the picture sequence, and storing the texture information in a texture file set to obtain a texture storage file set A.

[0011] As a preferred solution of the BIM model texture dynamic display method described in the present invention, wherein: the matching of the virtual camera and the real camera includes establishing a virtual camera node in the virtual environment based on the BIM model coordinate system, placing the virtual camera at the position of the real camera according to the three-dimensional coordinate information of the real camera, and setting the shooting angle parameters consistent with those of the real camera.

[0012] As a preferred solution of the BIM model texture dynamic display method described in the present invention, wherein: the automatic extraction and dynamic replacement of the component textures in the BIM model includes rendering the BIM model image from the perspective of the real camera by the virtual camera, identifying the BIM components included in the rendered image, obtaining the unique identification ID of the components, establishing the relationship between the component identification ID and the BIM component texture picture, indexing the component texture storage file set B, positioning the target component texture file to be replaced, segmenting the stored image in the texture storage file set A into parts only containing BIM components based on the pixel positions of the segmented image, and replacing the texture in the texture storage file set B with the cropped picture texture.

[0013] As a preferred solution of the BIM model texture dynamic display method described in the present invention, wherein: the real-time dynamic display of the BIM model texture through the Web side includes loading the BIM model IFC data on the Web side, identifying the corresponding components in the BIM model according to the component ID, mounting the latest texture image in the texture storage file set B on the surface of the BIM components on the Web side to replace the original texture, rendering and refreshing the component area with the updated texture, and updating the texture in real time through the browser front end based on the preset texture refresh frequency.

[0014] Another object of the present invention is to provide a BIM model texture dynamic display system, which can solve the problems of delayed texture synchronization update and large deviation between the display effect and the reality in the current BIM visualization technology through an automated process integrating high-resolution image acquisition, intelligent identification of component textures, and Web-side visual rendering.

[0015] As a preferred solution of the BIM model texture dynamic display system described in the present invention, it includes: a component texture acquisition module, a component texture matching and dynamic replacement module, and a Web-side dynamic display module; the component texture acquisition module includes a BIM model construction sub-module and a video image acquisition sub-module. The BIM model construction sub-module is used to establish a BIM model with component texture information, and the video image acquisition sub-module is used to collect and classify and store key frame images in the camera video stream; the component texture matching and dynamic replacement module includes a virtual camera modeling sub-module, an image segmentation sub-module, and a texture update sub-module. The virtual camera modeling sub-module is used to build a virtual camera with the same parameters as the real camera in the BIM model. The image segmentation sub-module is used to extract the image area of the corresponding component in the key frame image and crop the image texture corresponding to the component. The texture update sub-module is used to replace the cropped image texture onto the component; the Web-side dynamic display module includes an IFC data loading sub-module and a browser rendering sub-module. The IFC data loading sub-module is used to load the IFC data of the BIM model, and the browser rendering sub-module is used to control the update frequency of the component texture and display the updated texture on the Web side.

[0016] A computer device includes a memory and a processor. The memory stores a computer program. It is characterized in that when the processor executes the computer program, it implements the steps of any one of the methods in the BIM model texture dynamic display method.

[0017] A computer-readable storage medium stores a computer program. It is characterized in that when the computer program is executed by a processor, it implements the steps of any one of the methods in the BIM model texture dynamic display method.

[0018] The beneficial effects of the present invention: By constructing a BIM model containing component texture information and combining with a high-resolution camera to collect on-site key frame images, the present invention realizes the data docking between BIM components and real images, providing an accurate data basis for texture extraction and dynamic replacement. By establishing a virtual camera in the BIM model and matching it with the parameters of the real camera, the perspective consistency is ensured, so that the texture area extracted by image segmentation can be accurately mapped to the corresponding component and automatically replaced, significantly improving the real-time and automation of texture update. Finally, based on the visualization display mechanism on the Web side, the updated component texture is loaded and presented in a periodic refresh manner, realizing remote, efficient, and lightweight real-time visualization, enhancing the user's interactive experience and on-site restoration feeling. The overall technical solution constructs a closed-loop path from real perception to model response and then to remote display, with high real-time, high precision, and good scalability, and has important application value in the fields of intelligent construction and BIM operation and maintenance. Description of the Drawings

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a schematic flowchart of the BIM model texture dynamic display method provided by the first embodiment of the present invention.

[0021] Figure 2 It is the overall flowchart of the BIM model texture dynamic display system provided by the third embodiment of the present invention. Specific Embodiments

[0022] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0023] Embodiment 1, referring to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a BIM model texture dynamic display method, including:

[0024] S1: Construct a BIM model containing component texture information, and collect key frame images of the video stream in real time through a high-resolution camera.

[0025] It should be noted that constructing a BIM model containing component texture information includes using Revit software to build a BIM model, setting texture categories and material parameters in the component property interface according to the building design drawings, attaching texture information to the components in the BIM model, performing coordinate mapping processing on the texture images, and converting the BIM model into the IFC data format.

[0026] It should also be noted that the construction of the BIM model is operated using Revit software. According to the building design drawings, a three-dimensional geometric model including building components such as walls, floors, doors, windows, columns, and beams is created. During the modeling process, basic attributes such as component IDs, spatial positions, and geometric dimensions are set for each type of component. At the same time, in the material or surface texture property interface of the component, texture information is attached, including the texture picture file path, texture mapping method (such as the tiling, stretching, or UV mapping method for the texture), and texture parameters (such as transparency, reflectivity, and glossiness).

[0027] During the texture attachment process, the present invention preferably uses high-definition texture images (such as JPG or PNG formats) to ensure high visual fidelity during subsequent dynamic display. To ensure the geometric consistency between the image and the component surface, the system performs coordinate mapping processing on the texture image. The coordinate mapping processing includes converting the two-dimensional texture image coordinates into UV coordinates corresponding to the geometry of the three-dimensional component surface, so that the texture can be accurately presented on the component surface in the BIM model, avoiding phenomena such as stretching, offset, or deformation. This mapping can be automatically completed through the built-in material editor in Revit or by importing an external UV mapping plugin for auxiliary calibration. After completion, the BIM model is exported in the IFC standard format.

[0028] The coordinate mapping processing also includes calculating the two-dimensional image coordinates The intersection point of the spatial ray and the three-dimensional component surface is expressed as:

[0029] ;

[0030] Among them, The two-dimensional image coordinates are used to describe the position of the component texture in the image, represents the pixel coordinates of the image in the horizontal direction, represents the pixel coordinates of the image in the vertical direction, is the intersection point of the ray and the triangular element in the BIM model, which is the position corresponding to the pixel point in the two-dimensional image in the three-dimensional world, is the camera optical center (i.e., the origin coordinates of the real camera in the three-dimensional space), which can be simulated by the virtual camera in the BIM model, is the ray length parameter, represents the ray direction vector under the camera view calculated from the two-dimensional image coordinates. Starting from the two-dimensional image coordinates and combining the internal and external parameters of the camera, a spatial ray is generated, and the intersection point of the spatial ray and the three-dimensional component surface is calculated .

[0031] The calculation of the barycentric coordinate interpolation is expressed as:

[0032] ;

[0033] ;

[0034] Among them, , , are the three vertex coordinates of the component triangular element in the BIM model, represents the point The barycentric coordinate interpolation of the point within the triangular element is calculated in reverse based on the intersection point of the three-dimensional component surface Barycentric coordinate interpolation within a triangular element.

[0035] The UV coordinates corresponding to the geometry on the surface of a 3D component for a 2D image calculated according to barycentric coordinate interpolation are expressed as:

[0036] ;

[0037] Where is the final texture mapping coordinates of the point, that is, the coordinates of the corresponding point on the surface of the 3D component where the texture in the image is pasted, represents the horizontal coordinate component of the texture mapping corresponding to the point on the surface of the 3D component, represents the vertical coordinate component of the texture mapping corresponding to the point on the surface of the 3D component, , , are the UV mapping coordinates corresponding to the three vertex coordinates , , of the triangular element.

[0038] The BIM model constructed by the present invention not only has a clear component structure, complete material and texture parameters, but also realizes the coordinate mapping between the texture and the geometric model, providing a structured and callable basic data source for subsequent dynamic replacement and visual update of the texture, and supporting the technical implementation of the method for dynamic extraction and replacement of the texture.

[0039] Furthermore, key frame images of the video stream are collected in real time through a high-resolution camera, including arranging high-resolution cameras in the building space, recording the current 3D coordinate information and shooting angle parameters, collecting real-time video stream data, and storing it in a preset database.

[0040] It should also be noted that by using a high-resolution camera to take real-time pictures of the building space and extracting key frame images from the video stream as the data source for component texture update, high-resolution cameras are arranged inside or outside the building space according to a predetermined logic. The installation positions of the cameras are determined based on the distribution density of building components, spatial occlusion conditions, and key monitoring areas in the model. The 3D spatial coordinate information of each camera's installation position in the global coordinate system of the building needs to be recorded, and its shooting parameters, including pitch angle, yaw angle, roll angle, field of view angle, focal length, and imaging depth, are marked. The position information is used for subsequent synchronous mapping of virtual cameras and component matching calculations.

[0041] The present invention selects the lowest value of the preferred resolution of 1920×1080, and selects camera resolutions of 1920×1080 and above, which have good image clarity and low-light environment adaptability to ensure that the surface texture details of the components are completely restored in the captured images. All cameras are connected to the server through a local area network or a wireless network to continuously upload real-time video stream data.

[0042] Furthermore, real-time video stream data is collected, including extracting preset frames of pictures at preset intervals in the video stream data to obtain a picture sequence that meets the preset conditions, extracting the texture information in the picture sequence, and storing the texture information in a texture file set to obtain a texture storage file set A.

[0043] It should also be noted that a preselected scheme for the preset interval of the present invention is 1 second, and a preferred scheme for the preset number of frames is 1 frame, that is, 1 frame is selected as the key frame image every 1 second (the frame rate is 1 FPS) to form a picture sequence. Without significantly increasing the image processing burden, the present invention can cover the dynamic changes in the camera's field of view and meet the time resolution requirements for texture updates in most building operation and maintenance scenarios. Compared with high-frequency frame extraction (such as 30 FPS), the preset interval and preset number of frames of the present invention can greatly reduce data redundancy, reduce image storage and computing resource consumption in practical engineering applications, and are suitable for deployment in large-scale building complexes or scenarios where multiple cameras run synchronously. Compared with low-frequency frame extraction (such as 1 frame every 5 seconds), the strategy of 1 frame per second can ensure that in a building environment with frequent dynamic changes such as construction, renovation, or real-time monitoring, the subtle changes in the surface texture of the components and the changes in ambient light can be captured, effectively improving the response speed and real-time perception of component texture updates.

[0044] It should also be noted that the variance of the image grayscale is calculated as:

[0045] ;

[0046] where, is the variance of the image grayscale, is the width of the current key frame image, is the height of the current key frame image, is the original image after being processed by the Laplacian operator, represents the pixel coordinate index in the horizontal direction in the current key frame image, represents the pixel coordinate index in the vertical direction in the current key frame image, is the average grayscale response value of the entire Laplacian image.

[0047] A preferred solution for the preset conditions specifically includes: calculating the variance of the image grayscale through the Laplace operator, the grayscale gradient variance value is not less than 120, the average grayscale value of all pixels in the image is between 50 and 200, the pixel area of the BIM component accounts for no less than 30% of the total number of pixels in the entire image, and the occluded pixel area of the component in the image does not exceed 20% of the total pixel area of the component.

[0048] Setting the grayscale gradient variance value of 120 as the preferred threshold of the present invention can exclude poor-quality images caused by jitter, motion blur or focal length offset, ensuring that the extracted texture can be used for accurate replacement of the BIM component material. Using the average grayscale value of all pixels in the image being between 50 and 200 as the preferred threshold of our invention, if the image is too dark (average grayscale value less than 50) or too bright (average grayscale value greater than 200), it may cause loss of details or local overexposure of the BIM component. An average grayscale within this range can ensure that the image has appropriate brightness and darkness, which is beneficial for component boundary recognition and texture alignment. The average grayscale value threshold can make the image suitable for subsequent standardization processing and multi-camera synchronous mapping. To avoid too small a proportion of components in the image and too high a proportion of scene-irrelevant backgrounds, it is set that the proportion of component pixels is greater than or equal to 30%, ensuring that at least 30% of the area in the extracted image directly corresponds to the BIM component and has sufficient texture content. If the component is overly occluded in the image, even if the area is large enough, it is still not suitable as a texture source. Setting the occluded pixel proportion to be less than or equal to 20% as the preferred threshold of our invention can maximize the restoration of the true appearance of the component surface, while improving the visual consistency and credibility of the replaced model.

[0049] When the preset conditions are met, the image is determined to be a valid image frame and included in the image sequence. After the image sequence is formed, the texture area corresponding to the BIM model component in each frame of the image is extracted, and texture recognition, cropping and standardization processing are performed. The processed component texture images will be uniformly stored in the texture file set, denoted as texture storage file set A. The file set is archived according to the component ID, shooting timestamp and camera number, supporting the subsequent texture update module to quickly index the corresponding texture image according to the component identity, realizing accurate matching and replacement of the texture.

[0050] S2: Establish a virtual camera in the BIM model, match the virtual camera with the real camera, and automatically extract and dynamically replace the component texture in the BIM model.

[0051] It should be noted that matching the virtual camera with the real camera includes establishing a virtual camera node in the virtual environment based on the BIM model coordinate system, placing the virtual camera at the position of the real camera according to the three-dimensional coordinate information of the real camera, and setting the shooting angle parameters consistent with the real camera.

[0052] It should also be noted that based on the global three-dimensional coordinate system of the BIM model, the spatial position coordinates corresponding to the real camera are selected. After each real camera is installed, its spatial position is three-dimensionally located by a laser rangefinder, a total station, a SLAM positioning device, or the CAD coordinate drawing of the building structure, and an absolute position value including the X, Y, and Z axis coordinates is generated. In the BIM model environment, the camera or view point function interface provided by the three-dimensional modeling platform is called to initialize the virtual camera node. The origin coordinates of this node are set to the three-dimensional coordinates corresponding to the real camera to ensure the consistency of the spatial position. On the basis of the position matching being completed, the system further synchronizes the shooting parameters of the real camera, including: pitch angle, yaw angle, roll angle, focal length, field of view angle, imaging sensor ratio, shooting depth, and all the shooting parameters are synchronously set in the virtual camera node to ensure that when rendering, the range of components, the angle of components, and the boundary positions seen by the virtual camera are maximally consistent with the images captured by the real camera.

[0053] Furthermore, for the automatic extraction and dynamic replacement of component textures in the BIM model, it includes rendering the BIM model image from the perspective of the real camera by the virtual camera, identifying the BIM components included in the rendered image, obtaining the unique identification ID of the components, establishing the relationship between the component identification ID and the BIM component texture picture, indexing the component texture storage file set B, positioning the target component texture file to be replaced, segmenting the stored image in the texture storage file set A into parts that only contain BIM components based on the pixel positions of the segmented image, and replacing the texture in the texture storage file set B with the cropped picture texture.

[0054] It should also be noted that after the parameter matching between the virtual camera and the real camera is completed, the system renders an image frame in the BIM model environment that is consistent with the perspective of the real camera through the virtual camera. The rendered image starts from the virtual view point, presents the appearance and spatial layout of the model components in the current perspective, and has a definite frustum range. Identify the BIM components included in the rendered image, and the method used is the component triangle face element matching based on the three-dimensional scene projection algorithm, so as to accurately determine the set of visible components in the current perspective.

[0055] Specifically, by traversing all the component triangular facets in the BIM model, the three-dimensional vertex coordinate information of each facet is obtained, and the corresponding normal vector direction is calculated. For each facet, the system takes the virtual camera position as a reference, calculates the direction vector from the center point of the facet to the camera, and determines whether the facet faces the camera view by means of vector dot product. When the dot product of the facet normal vector and this direction vector is less than zero, it is determined that the facet is a forward-facing facet facing the camera and participates in the subsequent visibility judgment. According to the projection matrix of the virtual camera, the three vertex coordinates of the forward-facing facet are projected onto the image plane to obtain two-dimensional image coordinate points, and a corresponding projected triangular region is constructed on the image plane. For each pixel point in this region, the system calculates its corresponding depth value in three-dimensional space through the barycentric coordinate interpolation method and compares it with the depth buffer value at the same position. If the depth of the facet corresponding to the current pixel point is less than the minimum depth value recorded in the depth buffer, it means that the facet is in the foreground and can be directly observed by the camera. The system marks this facet as a "visible facet" and records the ID of the component it belongs to. Conversely, if there are other facets with shallower depths at this pixel position, it is determined that the current facet is occluded. After completing the pixel-level depth comparison of all facets, the system counts the number of facets marked as visible for each component ID in the image plane. If the number of pixel points where a certain component appears in the foreground area exceeds a set threshold (for example, more than 5 pixels or more than 1% of the facet area), then this component is included in the set of visible components under the current view. Generate the set of visible components corresponding to the current camera view, and combine the corresponding projection boundary coordinates to provide an accurate spatial reference basis for subsequent pixel clipping, UV mapping, and component texture replacement of the texture image.

[0056] For each identified component, the system extracts its unique identifier (component ID) defined in the BIM model and establishes a correspondence between the component and the texture image according to the component ID. The texture images of all BIM components are stored in the component texture storage file set B, and each texture image uses the component ID as the naming or indexing key value. The system can quickly locate the target texture file to be replaced based on this key value.

[0057] The present invention utilizes key frame images collected from a real - world camera, segments the images, and extracts the texture regions corresponding to the target components in the images. The images are sourced from the high - resolution picture sequences saved in the texture storage file set A. The image segmentation process includes methods such as component contour recognition, edge detection, semantic segmentation, or object detection. The system can map the pixel coordinates in the real - world image to the component regions in the BIM model in combination with the three - dimensional projection positions of the components in the BIM model, obtain the image cropping boundaries corresponding to the component textures, and based on the pixel boundary information, segment out the texture region images containing only the target components from the file set A, and perform necessary image size normalization, brightness correction, and format conversion to adapt to the requirements of the BIM model for texture images. The processed cropped texture images will be used as new textures to replace the original texture files of the corresponding components in the component texture file set B.

[0058] S3: Dynamically display the BIM model texture in real - time through the Web end.

[0059] It should be noted that dynamically displaying the BIM model texture in real - time through the Web end includes loading the IFC data of the BIM model at the Web end, identifying the corresponding components in the BIM model according to the component ID, mounting the latest texture images in the texture storage file set B to the surface of the BIM components at the Web end to replace the original textures, rendering and refreshing the component regions with updated textures, and updating the textures in real - time through the browser front - end based on the preset texture refresh frequency.

[0060] It should also be noted that first, initialize the three - dimensional visualization environment at the Web end and load the structured data of the BIM model. The BIM model data is exported in the IFC standard format. By parsing the IFC file, the system extracts the geometric shape information, spatial position relationship, unique identification ID, and original material texture parameters of each component in the model. The extracted information is mapped to the component object instances in the three - dimensional rendering engine at the Web end to achieve corresponding modeling with the BIM data. After the model is loaded, the system locates the components to be updated according to the component ID, and in combination with the texture storage file set B maintained by the server, extracts the latest texture image file corresponding to the component from it. The texture image file comes from the front - end image acquisition and processing module and has been processed by pixel segmentation, size standardization, and brightness correction, and can be directly used as the component texture for binding.

[0061] The Web - end system associates model components with corresponding texture images through component IDs, and calls the material replacement interface provided by the rendering engine to bind the new texture image to the surface material node of the component, replacing the original texture image. After the replacement is completed, the system renders and refreshes the local spatial area where the target component is located to ensure that the texture replacement takes effect immediately in the user - visible interface. Preferably, to avoid the performance burden caused by the overall model redrawing, only the components with updated textures or their local spatial units are subjected to local rendering operations, improving the system response speed and interaction performance.

[0062] To achieve continuous and real - time texture updates, the system supports setting the texture refresh time - frequency parameter T, such as refreshing once every 30 seconds or refreshing at the minute level. This parameter can be flexibly configured according to the usage scenario. For example, it can be set to high - frequency refresh in construction monitoring or equipment inspection scenarios, and low - frequency refresh in exhibition scenarios. During each refresh cycle, the front - end browser automatically requests the component texture status update record from the server. The system compares whether the currently displayed texture is consistent with the latest texture file path. If an update is detected, the texture is immediately replaced and the display area is refreshed, finally realizing the real - time dynamic display of BIM model component textures based on the Web platform.

[0063] Example 2 is an embodiment of the present invention, which provides a method for dynamically displaying BIM model textures. To verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0064] The test object is the BIM model of a six - story teaching building, which includes 462 standard component nodes, and the total number of texture material samples reaches 1180. A total of 6 system control groups are set up in the experiment: 2 groups are the current mainstream traditional BIM systems, 3 groups are prototype systems constructed based on the system architecture proposed in the present invention (using different resolutions and tuning parameters), and the other 1 group is an industry control system combined with augmented reality.

[0065] First, use Revit to build an architectural BIM model, and set unique identifiers and texture materials for each component. Six high - resolution cameras (mixing 4K and 1080P) are deployed in the building scene. Through a unified video processing module, key - frame images are extracted from the video stream at a frequency of 1 frame / second.

[0066] Subsequently, the experimental group system constructs corresponding virtual camera nodes for each camera in the BIM environment, and strictly completes the matching according to the real - camera coordinates and viewing - angle parameters. The component texture area is extracted through an image - segmentation algorithm, and the mapping binding between the component ID and the new texture image is completed, and the replacement operation is executed.

[0067] Finally, on the Web side, the rendering engine is called to load the model and refresh the component texture images in real time. By simulating the actual usage scenarios through user behaviors, key performance indicators such as the component recognition accuracy, average texture update latency, Web-side loading time, and visual consistency score of each group of systems are measured. The experimental data is shown in Table 1.

[0068] Table 1 Experimental data table of dynamic texture display of BIM models

[0069] Test subject Camera resolution (pixels) Key frame extraction frequency (frames per second) Texture recognition accuracy rate (%) Average texture update delay (seconds) Web-side loading time (seconds) Component visual consistency score (out of 100) Traditional BIM system A 1280x720 0.2 65.2 8.3 7.5 58 Traditional BIM system B 1920x1080 0.5 72.4 6.1 6.9 64 System No. 1 of the present invention 3840x2160 1 91.6 2.7 3.2 88 System No. 2 of the present invention 3840x2160 1 93.2 2.3 3 91 System No. 3 of the present invention 4096x2160 1 95.1 2 2.8 93 Augmented reality control group 1920x1080 0.5 78.9 4.8 4.5 72

[0070] As can be seen from the table data, traditional BIM systems generally have problems such as low texture recognition accuracy, large update latency, and long loading time in texture processing. For example, under the condition of relatively low image resolution (1280×720), the texture recognition accuracy of "traditional BIM system A" is only 65.2%, and the average texture update latency reaches 8.3 seconds, which cannot meet the requirements of real-time display.

[0071] In contrast, the system architecture proposed in the present invention ("No. 1 - 3 of the present invention system") shows significant advantages in key performance indicators. Under the condition of using a camera with a resolution of 3840×2160 and above, combined with the texture replacement mechanism driven by component ID and the local Web rendering update mechanism, the texture recognition accuracy is increased to over 93%, the average texture replacement latency is controlled within 2.0 - 2.7 seconds, and the Web-side loading time is reduced to 2.8 - 3.2 seconds, which is significantly better than other systems. The system of the present invention realizes the on-demand update and refresh strategy of the component surface texture, making the visual consistency score of the components generally exceed 90 points (full score 100), greatly improving the real restoration effect of the building scene, which is particularly crucial in interactive experience and remote visual management. Compared with the augmented reality control system, although the system of the present invention does not adopt somatosensory interaction technology, it has obvious advantages in information real-time and processing stability. The augmented reality system has problems of strong device dependence and large component recognition errors, with a texture recognition accuracy of 78.9% and an average latency of still 4.8 seconds, which is lower than the average level of the system of the present invention.

[0072] This embodiment fully demonstrates that the present invention has significant technological innovation in aspects such as component-level texture extraction and replacement automation, lightweight Web-side rendering, and system response speed. It is particularly suitable for application scenarios such as building digital twins, remote operation and maintenance display, and intelligent inspection assistance, and can effectively overcome the deficiencies of heavy manual intervention, response lag, and texture distortion in the prior art, having important practical value and promotion potential.

[0073] Example 3, refer to Figure 2, which is an embodiment of the present invention, provides a BIM model texture dynamic display system, including a component texture acquisition module 100, a component texture matching and dynamic replacement module 200, and a Web-side dynamic display module 300.

[0074] Among them, S4: The component texture acquisition module 100 includes a BIM model construction sub-module 101 and a video image acquisition sub-module 102. The BIM model construction sub-module 101 is used to establish a BIM model with component texture information, and the video image acquisition sub-module 102 is used to collect key frame images in the camera video stream and classify and store them.

[0075] It should also be noted that the video images and structural information extracted by the video image acquisition sub-module 102 will be synchronously transmitted to the component texture matching and dynamic replacement module 200 as processing inputs. At the same time, the component ID will be used to establish a texture file index relationship.

[0076] S5: The component texture matching and dynamic replacement module 200 includes a virtual camera modeling sub-module 201, an image segmentation sub-module 202, and a texture update sub-module 203. The virtual camera modeling sub-module 201 is used to build a virtual camera in the BIM model with the same parameters as the real camera. The image segmentation sub-module 202 is used to extract the image area of the corresponding component in the key frame image and crop the image texture corresponding to the component. The texture update sub-module 203 is used to replace the cropped image texture onto the component.

[0077] It should also be noted that after the component texture matching and dynamic replacement module 200 completes the cropping of the texture image and the replacement of the component, it will package the updated component texture path and the corresponding component ID to form binding information, which is written into the shared data structure or database. This binding information will be used as the data input source of the IFC data loading sub-module 301 to ensure that the Web side can directly call the latest texture image and locate it to the correct component object.

[0078] S6: The Web-side dynamic display module 300 includes an IFC data loading sub-module 301 and a browser rendering sub-module 302. The IFC data loading sub-module 301 is used to load the IFC data of the BIM model, and the browser rendering sub-module 302 is used to control the update frequency of the component texture and display the updated texture on the Web side.

[0079] It should also be noted that after the browser rendering sub-module 302 receives the component and texture binding path from the IFC data loading sub-module 301, it periodically accesses the texture storage file set B in combination with the texture refresh frequency set in the system, extracts the corresponding image resources from the database, and completes the replacement and refresh of the local component texture.

[0080] If a function is implemented in the form of 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, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods according to the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., all kinds of media that can store program codes.

[0081] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a predefined sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0082] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), fiber optic devices, and portable compact disc read-only memories (CDROMs). Additionally, a computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0083] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc. 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 preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for dynamically displaying BIM model textures, characterized in that: include: Build a BIM model containing component texture information and collect key frame images of video streams in real time through high-resolution cameras; Establish a virtual camera in the BIM model, match the virtual camera with the real camera, and automatically extract and dynamically replace the texture of components in the BIM model; Real-time dynamic display of BIM model textures through the Web terminal; Real-time acquisition of key frame images of video streams includes: presetting high-resolution camera nodes in the building space, configuring camera acquisition parameters and image clarity thresholds, automatically extracting key frame images from the video stream at set time intervals, and renaming and classifying key frame images that meet the requirements according to timestamps and storing them in a preset image file path according to preset conditions; The automatic extraction and dynamic replacement of component textures in the BIM model includes building a corresponding virtual camera in the BIM model based on the actual camera perspective, obtaining the visible area and component ID of the component under the current perspective, segmenting the acquired key frame images, extracting the image area corresponding to the BIM component, mapping the image coordinates with the model component coordinates at the pixel level, locating the original texture in the model texture file set according to the component ID, automatically adjusting the extracted image to the map size and format, synchronously writing the updated image into the component material properties, automatically completing the texture replacement, and recording the update timestamp.

2. The BIM model texture dynamic display method according to claim 1, characterized in that: The construction of the BIM model containing component texture information includes building the BIM model using Revit software, setting texture categories and material parameters in a component property interface according to building design drawings, attaching texture information to components in the BIM model, performing coordinate mapping processing on texture images, and converting the BIM model into an IFC data format.

3. The BIM model texture dynamic display method according to claim 1 or 2, characterized in that: The real-time acquisition of key frame images of video streams by high-resolution cameras includes arranging high-resolution cameras in the building space, recording current three-dimensional coordinate information and shooting angle parameters, acquiring real-time video stream data, and storing them in a preset database.

4. The BIM model texture dynamic display method according to claim 3, wherein: The real-time video stream data acquisition includes extracting a preset number of frames of pictures from the video stream data according to a preset interval time, obtaining a picture sequence that meets a preset condition, extracting texture information from the picture sequence, storing the texture information in a texture file set, and obtaining a texture storage file set A.

5. The BIM model texture dynamic display method according to any one of claims 1, 2 or 4, characterized in that: The matching of the virtual camera with the real camera includes establishing a virtual camera node in the virtual environment based on the BIM model coordinate system, placing the virtual camera at the real camera position according to the three-dimensional coordinate information of the real camera, and setting shooting angle parameters consistent with the real camera.

6. The BIM model texture dynamic display method according to claim 5, characterized in that: The automatic extraction and dynamic replacement of component textures in the BIM model includes rendering the BIM model image under the perspective of a real camera according to a virtual camera, identifying the BIM component contained in the rendered image, obtaining a unique identification ID of the component, establishing a relationship between the component identification ID and the BIM component texture image, indexing a component texture storage file set B, locating a target component texture file to be replaced, segmenting the image screen stored in the texture storage file set A into a portion containing only the BIM component based on the pixel position of the segmented image, and replacing the texture in the texture storage file set B with the cropped image texture.

7. The BIM model texture dynamic display method according to any one of claims 1, 2, 4 or 6, characterized in that: The real-time dynamic display of the BIM model texture through the Web side includes loading the IFC data of the BIM model on the Web side, identifying the corresponding components in the BIM model according to the component ID, mounting the latest texture image in the texture storage file set B on the surface of the BIM components on the Web side to replace the original texture, rendering and refreshing the component area with the updated texture, and updating the texture in real time through the browser front end based on the preset texture refresh frequency.

8. BIM model texture dynamic display system, characterized in that: It includes a component texture acquisition module (100), a component texture matching and dynamic replacement module (200), and a Web-side dynamic display module (300); The component texture acquisition module (100) includes a BIM model construction sub-module (101) and a video image acquisition sub-module (102). The BIM model construction sub-module (101) is used to establish a BIM model with component texture information, and the video image acquisition sub-module (102) is used to collect and classify and store the key frame images in the camera video stream. The component texture matching and dynamic replacement module (200) includes a virtual camera modeling sub-module (201), an image segmentation sub-module (202), and a texture update sub-module (203). The virtual camera modeling sub-module (201) is used to build a virtual camera with the same parameters as the real camera in the BIM model. The image segmentation sub-module (202) is used to extract the image area of the corresponding component in the key frame image and crop the image texture corresponding to the component. The texture update sub-module (203) is used to replace the cropped image texture onto the component. The Web-side dynamic display module (300) includes an IFC data loading sub-module (301) and a browser rendering sub-module (302). The IFC data loading sub-module (301) is used to load the IFC data of the BIM model, and the browser rendering sub-module (302) is used to control the update frequency of the component texture and display the updated texture on the Web side.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the BIM model texture dynamic display method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the BIM model texture dynamic display method described in any one of claims 1 to 7.

Images