Intelligent inspection implementation method, system and device based on intelligent inspection equipment
Patent Information
- Application Number
- CN202611028292.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-10
- Publication Date
- 2026-08-11
AI Technical Summary
[0003]本发明实施例提供了基于智能巡检设备的智能巡检实现方法、系统及设备，旨在解决现有技术中对建筑的巡检路线的设定以人工基于巡检规范要求为主设定，依赖人员经验，不同的人员制定的路径差异大，易出现重点节点漏查、次要区域重复巡检的问题
[0008]本发明实施例提供了基于智能巡检设备的智能巡检实现方法、系统及设备，方法包括：用户终端根据所选定的待巡检建筑BIM模型对应产生智能巡检指令，并发送至服务器；服务器响应于用户终端发送的智能巡检指令，获取与智能巡检指令对应的待巡检建筑信息；服务器获取由智能安全头盔上的图像采集装置所采集的当前建筑实景图像，并与待巡检建筑信息对应的BIM模型相匹配，得到当前巡检起点定位数据；服务器根据待巡检建筑信息的当前施工阶段信息及当前巡检起点定位数据，生成当前巡检路径规划数据，并将当前巡检路径规划数据发送至AI眼镜；其中，当前巡检路径规划数据以当前巡检起点定位数据对应的定位位置为起点定位位置；AI眼镜显示与当前巡检路径规划数据对应的巡检路径，以引导用户完成对待巡检建筑信息对应的建筑进行巡检。本发明实施例能在用户进入待巡检建筑信息对应的建筑后，通过当前建筑实景图像与待巡检建筑信息对应的BIM模型相匹配智能确定了当前巡检起点定位数据，结合待巡检建筑信息的当前施工阶段信息进行自动路径规划，得到当前巡检路径规划数据并发送至AI眼镜上显示，以引导用户完成对所述待巡检建筑信息对应的建筑进行巡检。
Smart Images
Figure CN122549696A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent building technology, and in particular to intelligent inspection implementation methods, systems and equipment based on intelligent inspection devices. Background Technology
[0002] During construction, on-site inspections at different stages effectively assess the actual construction progress of each node. Currently, inspection routes are primarily determined manually based on inspection specifications. Specifically, this involves manually marking key inspection points such as structural beams and columns, mechanical and electrical pipelines, openings, fire protection facilities, material storage areas, and construction machinery operating areas, excluding undeveloped and closed-off areas, before initially drafting the inspection route. However, this manual route-setting method relies heavily on personnel experience, leading to significant differences in route plans among different personnel. This can result in missed inspections of key nodes and repeated inspections of less important areas. Summary of the Invention
[0003] This invention provides an intelligent inspection method, system, and device based on intelligent inspection equipment. It aims to solve the problem that in the prior art, the setting of building inspection routes is mainly based on manual inspection specifications, which relies on personnel experience. Different personnel may set different routes, which can easily lead to missed inspections of key nodes and repeated inspections of secondary areas.
[0004] In a first aspect, embodiments of the present invention provide an intelligent inspection method based on intelligent inspection equipment, applied to an intelligent inspection system. The intelligent inspection system includes intelligent inspection equipment, a server, and a user terminal. The intelligent inspection equipment includes at least AI glasses and an intelligent safety helmet. The intelligent safety helmet is equipped with an image acquisition device. The AI glasses, the image acquisition device, and the user terminal are all communicatively connected to the server. The method includes: The user terminal generates intelligent inspection instructions based on the selected BIM model of the building to be inspected and sends them to the server. The server responds to the intelligent inspection command sent by the user terminal and obtains the building information to be inspected corresponding to the intelligent inspection command. The server acquires the current real-world image of the building captured by the image acquisition device on the smart safety helmet, and matches it with the BIM model corresponding to the building information to be inspected to obtain the current inspection starting point positioning data. The server generates current inspection path planning data based on the current construction stage information of the building to be inspected and the current inspection starting point location data, and sends the current inspection path planning data to the AI glasses; wherein, the current inspection path planning data takes the location corresponding to the current inspection starting point location data as the starting point location; The AI glasses display the inspection path corresponding to the current inspection path planning data to guide the user to complete the inspection of the building corresponding to the building information to be inspected.
[0005] Secondly, embodiments of the present invention also provide an intelligent inspection implementation system based on intelligent inspection equipment, configured within an intelligent inspection system. The intelligent inspection system includes intelligent inspection equipment, a server, and a user terminal. The intelligent inspection equipment includes at least AI glasses and an intelligent safety helmet. The intelligent safety helmet is equipped with an image acquisition device. The AI glasses, the image acquisition device, and the user terminal are all communicatively connected to the server. The intelligent inspection implementation system based on intelligent inspection equipment is used to implement the method described in the first aspect above.
[0006] Thirdly, embodiments of the present invention also provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect above.
[0007] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, can implement the method described in the first aspect above.
[0008] This invention provides a method, system, and device for intelligent inspection based on intelligent inspection equipment. The method includes: a user terminal generating an intelligent inspection command based on the selected BIM model of the building to be inspected and sending it to a server; the server responding to the intelligent inspection command sent by the user terminal and obtaining the building information to be inspected corresponding to the intelligent inspection command; the server obtaining a real-world image of the current building captured by an image acquisition device on an intelligent safety helmet and matching it with the BIM model corresponding to the building information to be inspected to obtain current inspection starting point positioning data; the server generating current inspection path planning data based on the current construction stage information of the building information to be inspected and the current inspection starting point positioning data, and sending the current inspection path planning data to AI glasses; wherein, the current inspection path planning data uses the positioning position corresponding to the current inspection starting point positioning data as the starting point positioning position; the AI glasses displaying the inspection path corresponding to the current inspection path planning data to guide the user to complete the inspection of the building information to be inspected. This invention enables the intelligent determination of the current inspection starting point location data by matching the current real-scene image of the building with the BIM model corresponding to the building information after the user enters the building to be inspected. It also performs automatic path planning by combining the current construction stage information of the building information to obtain the current inspection path planning data and sends it to the AI glasses for display, so as to guide the user to complete the inspection of the building information to be inspected. Attached Figure Description
[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0010] Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent inspection method based on intelligent inspection equipment provided in an embodiment of the present invention. Figure 2 A flowchart illustrating the intelligent inspection implementation method based on intelligent inspection equipment provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the first sub-process of the intelligent inspection implementation method based on intelligent inspection equipment provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of the second sub-process of the intelligent inspection implementation method based on intelligent inspection equipment provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the third sub-process of the intelligent inspection implementation method based on intelligent inspection equipment provided in an embodiment of the present invention; Figure 6A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation
[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0012] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.
[0013] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.
[0014] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0015] Please also refer to Figure 1 and Figure 2 ,in Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent inspection implementation method based on intelligent inspection equipment according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the intelligent inspection implementation method based on intelligent inspection equipment provided in an embodiment of the present invention. Figure 1 As shown, the intelligent inspection method based on intelligent inspection equipment provided in this embodiment of the invention is applied to an intelligent inspection system. The intelligent inspection system includes an intelligent inspection device 10, a server 20, and a user terminal 30. The intelligent inspection device 10 includes at least AI glasses 11 and an intelligent safety helmet 12. The intelligent safety helmet 12 is equipped with an image acquisition device 121. The AI glasses 11, the image acquisition device 121, and the user terminal 30 are all communicatively connected to the server 20. The user terminal 30 is a smart terminal used by the user (such as a smartphone, tablet, etc.). Figure 2 As shown, the method includes the following steps S110-S150.
[0016] S110. The user terminal generates an intelligent inspection command based on the selected BIM model of the building to be inspected and sends it to the server.
[0017] In this embodiment, after a user successfully logs into the intelligent inspection platform deployed on the server by entering their user account and password on the user terminal, the platform will display several building BIM models for which the current user account has inspection permissions. The user then needs to select the building BIM model to be inspected based on the actual building site they are currently in. After the selection is completed, the resulting inspection command is sent to the server.
[0018] S120. The server responds to the intelligent inspection command sent by the user terminal and obtains the building information to be inspected corresponding to the intelligent inspection command.
[0019] In this embodiment, when the server receives the intelligent inspection instruction sent by the user terminal, it needs to parse and obtain the building information to be inspected corresponding to the intelligent inspection instruction, so as to use the building corresponding to the building information to be inspected as the target object that the user needs to inspect.
[0020] In one embodiment, step S120 includes: Obtain the building's unique number, current construction stage information, and current inspectable floor range corresponding to the BIM model of the building to be inspected, to form the building information to be inspected.
[0021] In this embodiment, since the building corresponding to the building information to be inspected has different inspection requirements at different construction stages when it is the target object of inspection, it is necessary to parse and obtain the unique building number, current construction stage information, and current inspectable floor range corresponding to the BIM model of the building to be inspected in the intelligent inspection instruction, so as to form the building information to be inspected.
[0022] S130. The server acquires the current real-world image of the building captured by the image acquisition device on the smart safety helmet, and matches it with the BIM model corresponding to the building information to be inspected, to obtain the current inspection starting point positioning data.
[0023] In this embodiment, before the user begins the on-site inspection of the building corresponding to the building information to be inspected, they have already put on a smart safety helmet (because wearing a safety helmet is required to enter a construction site to meet regulatory requirements) and AI glasses (which generally also have an image acquisition device, but since the smart safety helmet already has an image acquisition device and prioritizes it as the image acquisition end of the entire smart inspection system, there is no need for the AI glasses to repeatedly collect on-site images). The user has also selected the building to be inspected through their terminal, and the server has determined the building information. At this point, when the user is on the first floor (or another floor) of the building as the inspection starting point, the image acquisition device on the smart safety helmet can first collect at least one real-scene image of the current building and upload it to the server. Once the server obtains the real-scene image of the current building and matches it with the BIM model corresponding to the building information to be inspected, it can automatically obtain the location data of the current inspection starting point. This method of locating the user's inspection starting point eliminates the need for manual setting by the user; they only need to control the smart safety helmet to collect one real-scene image of the current building at the inspection starting point. It is important to note that, in order for the server to more accurately match the current inspection starting point location data with the current building scene image, after the user puts on the smart safety helmet, the image acquisition device should be kept horizontal and its viewpoint should be aimed at the main building structure in front of the user (such as ensuring that the current building scene image includes walls, columns, beams, etc.).
[0024] In one embodiment, such as Figure 3 As shown, step S130 includes: S131. Obtain the semantic features of the current building corresponding to the current building real-scene image through a pre-trained deep learning model; S132. Match the current building semantic features with the building semantic features corresponding to each BIM model in the BIM model corresponding to the building information to be inspected, filter out the target building semantic features with the highest feature similarity to the current building semantic features, and obtain the target BIM model corresponding to the target building semantic features. S133. Obtain the positioning data corresponding to the target BIM model as the current inspection starting point positioning data.
[0025] In this embodiment, in order to extract the semantic features of the current building corresponding to the current building real-scene image more quickly in the server, the pre-trained deep learning model used is the SegFormer model. The SegFormer model includes a hierarchical Transformer encoder MiT and a full MLP lightweight decoder. The hierarchical Transformer encoder includes an input layer (used to perform convolution and overlapping block processing on the image to extract the image feature map), 4 Transformer encoder layers and an encoder output layer (used to output 4 sets of multi-scale features, such as F1 (1 / 4), F2 (1 / 8), F3 (1 / 16), and F4 (1 / 32) to cover the semantic information of the building from details to the whole). Each Transformer encoder includes a convolutional downsampling layer, a multi-head self-attention network and a feedforward network connected in sequence.
[0026] After obtaining the semantic features of the current building corresponding to the current building's real-world image through a deep learning model, it can be matched with the semantic features of each BIM model corresponding to the building information to be inspected (the cosine similarity or Euclidean distance between the current building's semantic features and the semantic features of each BIM model can be calculated as feature similarity). The target building semantic feature with the highest feature similarity to the current building's semantic features is then selected, and the target BIM model corresponding to this target building semantic feature is obtained from the BIM model. Finally, the positioning data corresponding to the target BIM model is obtained as the positioning data for the current inspection starting point. It can be seen that this automatic matching of image semantic features can quickly achieve the user's current location within the building to be inspected.
[0027] In one embodiment, the method further includes the following step before step S133: Obtain the current shooting altitude corresponding to the previous building real scene image, and obtain the target model altitude corresponding to the target BIM model; If it is determined that the current shooting altitude is the same as the altitude of the target model, then a reliable result label is added to the target BIM model.
[0028] In this embodiment, to achieve more accurate positioning, it is necessary to further compare the current shooting altitude corresponding to the previous building scene image with the target model altitude corresponding to the target BIM model. If the current shooting altitude and the target model altitude are the same, it indicates that the target BIM model accurately matches the previous building scene image and that their floor positions are also the same. At this point, a reliable result label can be added to the target BIM model, making it more reliable and thus selected as the target BIM model.
[0029] S140. The server generates current inspection path planning data based on the current construction stage information of the building to be inspected and the current inspection starting point positioning data, and sends the current inspection path planning data to the AI glasses.
[0030] The current inspection path planning data uses the location corresponding to the current inspection starting point location data as the starting point location.
[0031] In this embodiment, the server has the current inspection starting point location data and further obtains the current construction stage information of the building to be inspected (early construction stage, mid-construction stage, late construction stage, etc., and in actual use, more refined divisions can be made according to the user's actual needs). At this time, the server can automatically generate the current inspection path planning data (which takes the location corresponding to the current inspection starting point location data as the starting point location) and send the current inspection path planning data to the AI glasses. Then, the navigation path corresponding to the current inspection path planning data is superimposed and displayed on the lenses of the AI glasses without affecting normal vision, thereby guiding the user to conduct intelligent inspections.
[0032] In one embodiment, such as Figure 4 As shown, step S140 includes: S141. Obtain the range of currently inspectable floors and the information of currently inspectable areas corresponding to the current construction stage information; S142. Generate the current inspection path planning data in the order from the lowest floor to the highest floor, taking the location of the current inspection starting point positioning data as the starting point positioning location, and within the limited location range corresponding to the current inspectable floor range and the current inspectable area information.
[0033] In this embodiment, if the current inspectable floor range and current inspectable area information can be extracted from the current construction stage information, and other information such as the current restricted area information can also be obtained, then the server can follow the principle of sequentially from the lower floors to the higher floors and taking the location of the current inspection starting point positioning data as the starting point positioning location. After traversing all inspection points of the current inspectable floor range and current inspectable area information, the current inspection path planning data can be automatically generated by combining the corresponding path planning algorithm.
[0034] In one embodiment, such as Figure 5 As shown, step S142 includes: S1421. Obtain a set of candidate BIM model key points corresponding to the current inspectable floor range and the current inspectable area information; wherein, the set of candidate BIM model key points includes multiple candidate BIM model key points, and each candidate BIM model key point has corresponding floor positioning information and inspection priority information. S1422. Obtain the current inspection path planning data corresponding to the key point set of the candidate BIM model according to the preset ant colony algorithm.
[0035] In this embodiment, when performing path planning on the server, the current inspection path planning data is first obtained. Since each candidate BIM model key point in the candidate BIM model key point set has corresponding floor positioning information and inspection priority information, and the distance of each walkable path between each candidate BIM model key point and its neighboring candidate BIM model key points is also known, the current inspection path planning data corresponding to the candidate BIM model key point set can then be obtained according to the ant colony algorithm. When using the ant colony algorithm for path planning, taking the planning of multiple walkable paths between a candidate BIM model key point and its neighboring candidate BIM model key points as an example, the shorter the path distance between two points, the higher the priority of selection. Moreover, among the multiple points adjacent to the starting point, the point with the higher inspection priority is selected first (of course, if two points adjacent to the starting point have the same inspection priority, one of them can be randomly selected). When the above principles are followed for ant colony algorithm path planning, and all candidate BIM model key points in the candidate BIM model key point set have been traversed, the current inspection path planning data can be obtained.
[0036] S150. The AI glasses display the inspection path corresponding to the current inspection path planning data to guide the user to complete the inspection of the building corresponding to the building information to be inspected.
[0037] In this embodiment, after the server sends the inspection path corresponding to the current inspection path planning data to the AI glasses, the lenses of the AI glasses display the inspection path corresponding to the current inspection path planning data. Furthermore, during this navigation process, the image acquisition device on top of the safety smart helmet can continue to acquire images of the current building at a preset image acquisition cycle to locate the user's current building position and update the location in real time to the corresponding path positioning point on the inspection path corresponding to the current inspection path planning data. Through this real-time display of the inspection path by the AI glasses, the user can be intelligently guided to complete the inspection of the building corresponding to the building information to be inspected.
[0038] In one embodiment, step S150 includes: The AI glasses obtain the key points of the current candidate BIM model corresponding to the user's current location in the current inspection path planning data, and obtain the key inspection prompts corresponding to the key points of the current candidate BIM model from the inspection knowledge base corresponding to the building information to be inspected in the server, and broadcast them by voice.
[0039] In this embodiment, since AI glasses generally have a voice broadcast function, when the AI glasses combine the real-time feedback from the server on the user's current location and determine the key points of the current candidate BIM model corresponding to the user's current location in the current inspection path planning data, they can also obtain the key inspection prompts corresponding to the key points of the current candidate BIM model from the inspection knowledge base corresponding to the building information to be inspected in the server, and broadcast them by voice, thereby prompting the user to focus on inspecting the key items at the current candidate BIM model key points, thus achieving a more comprehensive intelligent building inspection.
[0040] As can be seen, the implementation of this method can intelligently determine the current inspection starting point location data by matching the current real-scene image of the building with the BIM model corresponding to the building information to be inspected after the user enters the building to be inspected. It can also automatically plan the path by combining the current construction stage information of the building information to be inspected, obtain the current inspection path planning data, and send it to the AI glasses for display, so as to guide the user to complete the inspection of the building corresponding to the building information to be inspected.
[0041] Corresponding to the above-described intelligent inspection implementation method based on intelligent inspection equipment, this invention also provides an intelligent inspection implementation system based on intelligent inspection equipment. For example... Figure 1 As shown, the intelligent inspection system based on intelligent inspection equipment is configured in the intelligent inspection system. The intelligent inspection system includes intelligent inspection equipment 10, server 20 and user terminal 30. The intelligent inspection equipment 10 includes at least AI glasses 11 and intelligent safety helmet 12. The intelligent safety helmet 12 is equipped with an image acquisition device 121. The AI glasses 11, the image acquisition device 121 and the user terminal 30 are all communicatively connected to the server 20. The user terminal 30 is a smart terminal used by the user (such as a smartphone, tablet computer, etc.).
[0042] The user terminal 30 is used to generate intelligent inspection instructions based on the selected BIM model of the building to be inspected, and send them to the server.
[0043] In this embodiment, after a user successfully logs into the intelligent inspection platform deployed on the server by entering their user account and password on the user terminal, the platform will display several building BIM models for which the current user account has inspection permissions. The user then needs to select the building BIM model to be inspected based on the actual building site they are currently in. After the selection is completed, the resulting inspection command is sent to the server.
[0044] The server 20 is used to respond to the intelligent inspection command sent by the user terminal and obtain the building information to be inspected corresponding to the intelligent inspection command.
[0045] In this embodiment, when the server receives the intelligent inspection instruction sent by the user terminal, it needs to parse and obtain the building information to be inspected corresponding to the intelligent inspection instruction, so as to use the building corresponding to the building information to be inspected as the target object that the user needs to inspect.
[0046] The server 20 is specifically used for: Obtain the building's unique number, current construction stage information, and current inspectable floor range corresponding to the BIM model of the building to be inspected, to form the building information to be inspected.
[0047] In this embodiment, since the building corresponding to the building information to be inspected has different inspection requirements at different construction stages when it is the target object of inspection, it is necessary to parse and obtain the unique building number, current construction stage information, and current inspectable floor range corresponding to the BIM model of the building to be inspected in the intelligent inspection instruction, so as to form the building information to be inspected.
[0048] The server 20 is also used to acquire the current real-scene image of the building captured by the image acquisition device on the smart safety helmet, and match it with the BIM model corresponding to the building information to be inspected to obtain the current inspection starting point positioning data.
[0049] In this embodiment, before the user begins the on-site inspection of the building corresponding to the building information to be inspected, they have already put on a smart safety helmet (because wearing a safety helmet is required to enter a construction site to meet regulatory requirements) and AI glasses (which generally also have an image acquisition device, but since the smart safety helmet already has an image acquisition device and prioritizes it as the image acquisition end of the entire smart inspection system, there is no need for the AI glasses to repeatedly collect on-site images). The user has also selected the building to be inspected through their terminal, and the server has determined the building information. At this point, when the user is on the first floor (or another floor) of the building as the inspection starting point, the image acquisition device on the smart safety helmet can first collect at least one real-scene image of the current building and upload it to the server. Once the server obtains the real-scene image of the current building and matches it with the BIM model corresponding to the building information to be inspected, it can automatically obtain the location data of the current inspection starting point. This method of locating the user's inspection starting point eliminates the need for manual setting by the user; they only need to control the smart safety helmet to collect one real-scene image of the current building at the inspection starting point. It is important to note that, in order for the server to more accurately match the current inspection starting point location data with the current building scene image, after the user puts on the smart safety helmet, the image acquisition device should be kept horizontal and its viewpoint should be aimed at the main building structure in front of the user (such as ensuring that the current building scene image includes walls, columns, beams, etc.).
[0050] The server 20 is also specifically used for: The semantic features of the current building corresponding to the current building real-scene image are obtained through a pre-trained deep learning model. The current building semantic features are matched with the building semantic features corresponding to each BIM model in the BIM model corresponding to the building information to be inspected. The target building semantic features with the highest feature similarity to the current building semantic features are selected, and the target BIM model corresponding to the target building semantic features is obtained. Obtain the positioning data corresponding to the target BIM model as the positioning data of the current inspection starting point.
[0051] In this embodiment, in order to extract the semantic features of the current building corresponding to the current building real-scene image more quickly in the server, the pre-trained deep learning model used is the SegFormer model. The SegFormer model includes a hierarchical Transformer encoder MiT and a full MLP lightweight decoder. The hierarchical Transformer encoder includes an input layer (used to perform convolution and overlapping block processing on the image to extract the image feature map), 4 Transformer encoder layers and an encoder output layer (used to output 4 sets of multi-scale features, such as F1 (1 / 4), F2 (1 / 8), F3 (1 / 16), and F4 (1 / 32) to cover the semantic information of the building from details to the whole). Each Transformer encoder includes a convolutional downsampling layer, a multi-head self-attention network and a feedforward network connected in sequence.
[0052] After obtaining the semantic features of the current building corresponding to the current building's real-world image through a deep learning model, it can be matched with the semantic features of each BIM model corresponding to the building information to be inspected (the cosine similarity or Euclidean distance between the current building's semantic features and the semantic features of each BIM model can be calculated as feature similarity). The target building semantic feature with the highest feature similarity to the current building's semantic features is then selected, and the target BIM model corresponding to this target building semantic feature is obtained from the BIM model. Finally, the positioning data corresponding to the target BIM model is obtained as the positioning data for the current inspection starting point. It can be seen that this automatic matching of image semantic features can quickly achieve the user's current location within the building to be inspected.
[0053] In one embodiment, the server 20 is further specifically used for: Obtain the current shooting altitude corresponding to the previous building real scene image, and obtain the target model altitude corresponding to the target BIM model; If it is determined that the current shooting altitude is the same as the altitude of the target model, then a reliable result label is added to the target BIM model.
[0054] In this embodiment, to achieve more accurate positioning, it is necessary to further compare the current shooting altitude corresponding to the previous building scene image with the target model altitude corresponding to the target BIM model. If the current shooting altitude and the target model altitude are the same, it indicates that the target BIM model accurately matches the previous building scene image and that their floor positions are also the same. At this point, a reliable result label can be added to the target BIM model, making it more reliable and thus selected as the target BIM model.
[0055] The server 20 is also used to generate current inspection path planning data based on the current construction stage information of the building to be inspected and the current inspection starting point positioning data, and send the current inspection path planning data to the AI glasses.
[0056] The current inspection path planning data uses the location corresponding to the current inspection starting point location data as the starting point location.
[0057] In this embodiment, the server has the current inspection starting point location data and further obtains the current construction stage information of the building to be inspected (early construction stage, mid-construction stage, late construction stage, etc., and in actual use, more refined divisions can be made according to the user's actual needs). At this time, the server can automatically generate the current inspection path planning data (which takes the location corresponding to the current inspection starting point location data as the starting point location) and send the current inspection path planning data to the AI glasses. Then, the navigation path corresponding to the current inspection path planning data is superimposed and displayed on the lenses of the AI glasses without affecting normal vision, thereby guiding the user to conduct intelligent inspections.
[0058] In one embodiment, the server 20 is further specifically used for: Obtain the range of currently inspectable floors and the information of currently inspectable areas corresponding to the current construction stage information; The current inspection path planning data is generated in the order from the lowest floor to the highest floor, with the current inspection starting point location data as the starting point location, and within the limited location range corresponding to the current inspectable floor range and the current inspectable area information.
[0059] In this embodiment, if the current inspectable floor range and current inspectable area information can be extracted from the current construction stage information, and other information such as the current restricted area information can also be obtained, then the server can follow the principle of sequentially from the lower floors to the higher floors and taking the location of the current inspection starting point positioning data as the starting point positioning location. After traversing all inspection points of the current inspectable floor range and current inspectable area information, the current inspection path planning data can be automatically generated by combining the corresponding path planning algorithm.
[0060] In one embodiment, the step of generating the current inspection path planning data in order from lower to higher floors, using the location of the current inspection starting point positioning data as the starting point positioning location, and within the limited location range corresponding to the current inspectable floor range and the current inspectable area information, includes: Obtain a set of key points of candidate BIM models corresponding to the current inspectable floor range and the current inspectable area information; wherein, the set of key points of candidate BIM models includes multiple key points of candidate BIM models, and each key point of candidate BIM models has corresponding floor positioning information and inspection priority information. The current inspection path planning data corresponding to the key point set of the candidate BIM model is obtained according to the preset ant colony algorithm.
[0061] In this embodiment, when performing path planning on the server, the current inspection path planning data is first obtained. Since each candidate BIM model key point in the candidate BIM model key point set has corresponding floor positioning information and inspection priority information, and the distance of each walkable path between each candidate BIM model key point and its neighboring candidate BIM model key points is also known, the current inspection path planning data corresponding to the candidate BIM model key point set can then be obtained according to the ant colony algorithm. When using the ant colony algorithm for path planning, taking the planning of multiple walkable paths between a candidate BIM model key point and its neighboring candidate BIM model key points as an example, the shorter the path distance between two points, the higher the priority of selection. Moreover, among the multiple points adjacent to the starting point, the point with the higher inspection priority is selected first (of course, if two points adjacent to the starting point have the same inspection priority, one of them can be randomly selected). When the above principles are followed for ant colony algorithm path planning, and all candidate BIM model key points in the candidate BIM model key point set have been traversed, the current inspection path planning data can be obtained.
[0062] The AI glasses 11 are also used to display the inspection path corresponding to the current inspection path planning data, so as to guide the user to complete the inspection of the building corresponding to the building information to be inspected.
[0063] In this embodiment, after the server sends the inspection path corresponding to the current inspection path planning data to the AI glasses, the lenses of the AI glasses display the inspection path corresponding to the current inspection path planning data. Furthermore, during this navigation process, the image acquisition device on top of the safety smart helmet can continue to acquire images of the current building at a preset image acquisition cycle to locate the user's current building position and update the location in real time to the corresponding path positioning point on the inspection path corresponding to the current inspection path planning data. Through this real-time display of the inspection path by the AI glasses, the user can be intelligently guided to complete the inspection of the building corresponding to the building information to be inspected.
[0064] In one embodiment, the AI glasses 11 are specifically used for: The system obtains the key points of the current candidate BIM model corresponding to the user's current location in the current inspection path planning data, retrieves the key inspection prompts corresponding to the key points of the current candidate BIM model from the inspection knowledge base corresponding to the building information to be inspected in the server, and broadcasts them via voice.
[0065] In this embodiment, since AI glasses generally have a voice broadcast function, when the AI glasses combine the real-time feedback from the server on the user's current location and determine the key points of the current candidate BIM model corresponding to the user's current location in the current inspection path planning data, they can also obtain the key inspection prompts corresponding to the key points of the current candidate BIM model from the inspection knowledge base corresponding to the building information to be inspected in the server, and broadcast them by voice, thereby prompting the user to focus on inspecting the key items at the current candidate BIM model key points, thus achieving a more comprehensive intelligent building inspection.
[0066] As can be seen, the implementation of this system can intelligently determine the current inspection starting point location data by matching the current real-scene image of the building with the BIM model corresponding to the building information to be inspected after the user enters the building to be inspected. It can also automatically plan the path by combining the current construction stage information of the building information to be inspected, obtain the current inspection path planning data, and send it to the AI glasses for display, so as to guide the user to complete the inspection of the building information to be inspected.
[0067] The aforementioned intelligent inspection system based on intelligent inspection equipment can be implemented as a computer program, which can be used in, for example... Figure 6 It runs on the computer device shown.
[0068] Please see Figure 6 , Figure 6 This is a schematic block diagram of a computer device provided in an embodiment of the present invention. This computer device integrates any of the intelligent inspection implementation systems based on intelligent inspection equipment provided in the embodiments of the present invention.
[0069] See Figure 6 The computer device 400 includes a processor 402, a memory, and a network interface 405 connected via a system bus 401. The memory may include a storage medium 403 and internal memory 404.
[0070] The storage medium 403 may store an operating system 4031 and a computer program 4032. The computer program 4032 includes program instructions, which, when executed, cause the processor 402 to execute an intelligent inspection implementation method based on an intelligent inspection device.
[0071] The processor 402 provides computing and control capabilities to support the operation of the entire computer device.
[0072] The internal memory 404 provides an environment for the computer program 4032 in the storage medium 403 to run. When the computer program 4032 is executed by the processor 402, the processor 402 can execute the above-mentioned intelligent inspection implementation method based on the intelligent inspection device.
[0073] This network interface 405 is used for network communication with other devices. Those skilled in the art will understand that... Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the computer device to which the present invention is applied. A specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0074] The processor 402 is used to run the computer program 4032 stored in the memory to implement the intelligent inspection method based on the intelligent inspection device as described above.
[0075] It should be understood that, in this embodiment of the invention, the processor 402 may be a Central Processing Unit (CPU), or it may 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. The general-purpose processor may be a microprocessor or any conventional processor.
[0076] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.
[0077] Therefore, the present invention also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program, wherein the computer program includes program instructions. When executed by a processor, the program instructions cause the processor to perform the intelligent inspection implementation method based on an intelligent inspection device as described above.
[0078] The storage medium can be any computer-readable storage medium that can store program code, such as a USB flash drive, external hard drive, read-only memory (ROM), magnetic disk, or optical disk.
[0079] 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, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. 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 implementations should not be considered beyond the scope of this invention.
[0080] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.
[0081] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the device of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention 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.
[0082] 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 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 all or part of the 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 to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.
[0083] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.
Claims
1. A method for implementing intelligent inspection based on intelligent inspection equipment, applied to an intelligent inspection system, characterized in that, The intelligent inspection system includes intelligent inspection equipment, a server, and a user terminal. The intelligent inspection equipment includes at least AI glasses and a smart safety helmet. The smart safety helmet is equipped with an image acquisition device. The AI glasses, the image acquisition device, and the user terminal are all communicatively connected to the server. The method includes: The user terminal generates intelligent inspection instructions based on the selected BIM model of the building to be inspected and sends them to the server. The server responds to the intelligent inspection command sent by the user terminal and obtains the building information to be inspected corresponding to the intelligent inspection command. The server acquires the current real-world image of the building captured by the image acquisition device on the smart safety helmet, and matches it with the BIM model corresponding to the building information to be inspected to obtain the current inspection starting point positioning data. The server generates current inspection path planning data based on the current construction stage information of the building to be inspected and the current inspection starting point location data, and sends the current inspection path planning data to the AI glasses; wherein, the current inspection path planning data takes the location corresponding to the current inspection starting point location data as the starting point location; The AI glasses display the inspection path corresponding to the current inspection path planning data to guide the user to complete the inspection of the building corresponding to the building information to be inspected.
2. The method according to claim 1, characterized in that, The step of obtaining the building information to be inspected corresponding to the intelligent inspection command includes: Obtain the building's unique number, current construction stage information, and current inspectable floor range corresponding to the BIM model of the building to be inspected, to form the building information to be inspected.
3. The method according to claim 1, characterized in that, The process of acquiring the current building scene image captured by the image acquisition device on the smart safety helmet and matching it with the BIM model corresponding to the building information to be inspected to obtain the current inspection starting point positioning data includes: The semantic features of the current building corresponding to the current building real-scene image are obtained through a pre-trained deep learning model. The current building semantic features are matched with the building semantic features corresponding to each BIM model in the BIM model corresponding to the building information to be inspected. The target building semantic features with the highest feature similarity to the current building semantic features are selected, and the target BIM model corresponding to the target building semantic features is obtained. Obtain the positioning data corresponding to the target BIM model as the positioning data of the current inspection starting point.
4. The method according to claim 3, characterized in that, Before the step of obtaining the positioning data corresponding to the target BIM model as the current inspection starting point positioning data, the method further includes: Obtain the current shooting altitude corresponding to the previous building real scene image, and obtain the target model altitude corresponding to the target BIM model; If it is determined that the current shooting altitude is the same as the altitude of the target model, then a reliable result label is added to the target BIM model.
5. The method according to claim 1, characterized in that, The step of generating current inspection path planning data based on the current construction stage information of the building to be inspected and the current inspection starting point location data includes: Obtain the range of currently inspectable floors and the information of currently inspectable areas corresponding to the current construction stage information; The current inspection path planning data is generated in the order from the lowest floor to the highest floor, with the current inspection starting point location data as the starting point location, and within the limited location range corresponding to the current inspectable floor range and the current inspectable area information.
6. The method according to claim 5, characterized in that, The process of generating the current inspection path planning data in ascending order from lower to higher floors, using the current inspection starting point location data as the starting point location, and within the limited location range corresponding to the current inspectable floor range and the current inspectable area information, includes: Obtain a set of key points of candidate BIM models corresponding to the current inspectable floor range and the current inspectable area information; wherein, the set of key points of candidate BIM models includes multiple key points of candidate BIM models, and each key point of candidate BIM models has corresponding floor positioning information and inspection priority information. The current inspection path planning data corresponding to the key point set of the candidate BIM model is obtained according to the preset ant colony algorithm.
7. The method according to claim 1, characterized in that, The AI glasses display the inspection path corresponding to the current inspection path planning data, including: The AI glasses obtain the key points of the current candidate BIM model corresponding to the user's current location in the current inspection path planning data, and obtain the key inspection prompts corresponding to the key points of the current candidate BIM model from the inspection knowledge base corresponding to the building information to be inspected in the server, and broadcast them by voice.
8. An intelligent inspection system based on intelligent inspection equipment, configured in an intelligent inspection system, characterized in that, The intelligent inspection system includes intelligent inspection equipment, a server, and a user terminal. The intelligent inspection equipment includes at least AI glasses and an intelligent safety helmet. The intelligent safety helmet is equipped with an image acquisition device. The AI glasses, the image acquisition device, and the user terminal are all communicatively connected to the server. The intelligent inspection implementation system based on the intelligent inspection equipment is used to implement the intelligent inspection implementation method based on the intelligent inspection equipment as described in any one of claims 1-7.
9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the intelligent inspection method based on the intelligent inspection device as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which includes program instructions that, when executed by a processor, can implement the intelligent inspection method based on an intelligent inspection device as described in any one of claims 1-7.