Building water seepage marking method and device based on robot cruise

Abstract

The invention discloses a building water seepage marking method, device and equipment based on robot cruise and a medium, and the method comprises the steps: obtaining an infrared image of a robot for cruise shooting of a to-be-detected building, inputting the infrared image to a pre-trained water seepage detection model for water seepage detection, and obtaining a plurality of candidate regions and corresponding water seepage probabilities; screening the plurality of candidate areas based on the water seepage probability and temperature data in the infrared image to obtain a water seepage area; coordinate conversion is carried out based on a pre-stored conversion relation between an architectural drawing coordinate system and a detection coordinate system, and a second coordinate of the water seepage area is obtained; and determining a water seepage area, and mapping the water seepage area to an architectural drawing of the to-be-detected building based on the second coordinates and the water seepage area to obtain a building water seepage labeling result. Through the technical scheme, the detection efficiency of building water seepage and the accuracy of the detection result can be improved, and the timeliness of water seepage feedback can be improved.

Description

Technical Field

[0001] This application belongs to the technical field of building inspection, and particularly relates to a building water seepage annotation method, device, equipment and medium based on robot cruise. Background Art

[0002] With the development of social economy, the number of buildings is increasing continuously. To ensure the safety and stability of building use, the quality inspection of building floors, especially the detection of water seepage problems in building floors, becomes very important. The water seepage problem of building floors refers to the phenomenon that water penetrates into other areas through the internal structure of the floor due to various reasons during the use of the building floor, thus affecting the safety and humidity of the floor. Identifying and evaluating the penetration or moisture conditions of each position of the floor according to the physical characteristics of the building floor, and then determining whether there is a water seepage phenomenon can effectively avoid the problem of untimely detection of damage to the floor structure.

[0003] In the prior art, the detection method for water seepage in building floors is usually to attach or connect a humidity detection instrument to the building to be detected, send detection instructions to the humidity detection instrument regularly, obtain the humidity data of the building to be detected, and determine whether there is a water seepage phenomenon in the building to be detected at the current moment by comparing the humidity data with the preset humidity data. Then, the area with water seepage phenomenon is determined according to the setting position of the humidity detection instrument and marked on the corresponding CAD drawing of the building to be detected.

[0004] However, the prior art requires a large number of humidity detection instruments, which is not suitable for large-area water seepage detection. At the same time, due to weather reasons, the humidity in the air will affect the detection results of the humidity detection instrument, resulting in inaccurate detection results for the water seepage of the building to be detected. At the same time, the prior art needs to manually mark the water seepage in the CAD drawing, resulting in the problem of untimely feedback of the water seepage situation. Summary of the Invention

[0005] The purpose of the embodiments of this application is to provide a building water seepage annotation method, device, equipment and medium based on robot cruise, which solves the problems of limited application range, inaccurate detection results and untimely feedback of water seepage situation existing in the building water seepage detection using the prior art. By identifying the water seepage situation of the building floor according to the physical characteristics of the building floor materials using infrared images, determining the water seepage area according to the water seepage probability of the infrared image and the preset water seepage probability threshold, and converting the water seepage area to the building drawing coordinate system for building water seepage annotation based on robot cruise, the purpose of automatically detecting and annotating water seepage based on infrared images can be achieved, improving the detection efficiency and accuracy of building water seepage, and at the same time facilitating the timely feedback of the water seepage situation.

[0006] In a first aspect, an embodiment of the present application provides a method for annotating building water seepage based on robot cruising, and the method includes: Obtain an infrared image captured by a robot during cruising of a building to be detected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, obtain multiple candidate regions and corresponding water seepage probabilities in the infrared image, and the infrared image includes temperature data of multiple pixel points; Screen multiple candidate regions based on the water seepage probability and temperature data to obtain the water seepage region in the infrared image; Obtain the first coordinates of the water seepage region in the detection coordinate system, and based on the conversion relationship between the building drawing coordinate system and the detection coordinate system stored in advance, convert the first coordinates to the building drawing coordinate system to obtain the second coordinates of the water seepage region; Identify multiple boundary pixel coordinates of the water seepage region in the building drawing coordinate system, determine the water seepage area based on the multiple boundary pixel coordinates, and map the water seepage region to the building drawing of the building to be detected based on the second coordinates and the water seepage area to obtain the annotation result of building water seepage based on robot cruising.

[0007] Optionally, screening multiple candidate regions based on the water seepage probability and temperature data includes: Group multiple candidate regions based on the temperature data to obtain multiple groups of candidate regions; Determine the overall water seepage probability of the same group of candidate regions based on the water seepage probability, and screen multiple candidate regions based on the overall water seepage probability.

[0008] Optionally, grouping multiple candidate regions based on the temperature data includes: Determine the pixel point temperature distribution information corresponding to each candidate region based on the temperature data; Determine the water seepage similarity between multiple candidate regions based on the pixel point temperature distribution information, and group multiple candidate regions based on the water seepage similarity.

[0009] Optionally, screening multiple candidate regions based on the overall water seepage probability includes: Calculate the average group water seepage probability of multiple groups of candidate regions based on the overall water seepage probability; Identify the magnitude relationship between the water seepage probability corresponding to each candidate region and the average group water seepage probability, and filter out the candidate regions with a water seepage probability less than the average group.

[0010] Optionally, before screening multiple candidate regions based on the water seepage probability and temperature data, the method further includes: Calculate the temperature difference data of the candidate region based on the temperature data of multiple pixel points, and perform accuracy verification on the water seepage probability based on the temperature difference data; Update the seepage probability based on the accuracy verification result to obtain the final seepage probability.

[0011] Optionally, the accuracy verification of the seepage probability based on the temperature difference data includes: Identify the association relationship between the temperature difference data and the preset temperature difference range, where the preset temperature difference range includes the preset material temperature difference range and the preset seepage temperature difference range; When the temperature difference data is within the preset seepage temperature difference range, determine that the seepage probability is accurate; When the temperature difference data is within the preset material temperature difference range, determine that the seepage probability is inaccurate.

[0012] Optionally, before calculating the temperature difference data of the candidate area based on the temperature data of multiple pixel points, the method further includes: Obtain the ambient temperature data and the material temperature rise data corresponding to the candidate area; Predict the highest seepage temperature of the candidate area based on the ambient temperature data and the material temperature rise data.

[0013] Correspondingly, the accuracy verification of the seepage probability based on the temperature difference data includes: Determine the estimated lowest seepage temperature of the candidate area based on the temperature difference data and the highest seepage temperature, and identify whether the actual lowest temperature in the temperature data of the candidate area is greater than the estimated lowest seepage temperature; Based on the actual lowest temperature identification result, perform accuracy verification on the seepage probability.

[0014] In a second aspect, an embodiment of the present application provides a building seepage annotation device based on robot cruising, and the device includes: A seepage probability determination module, configured to obtain an infrared image captured by a robot during cruising of a building to be detected, input the infrared image into a pre-trained seepage detection model, perform seepage detection based on the seepage detection model to obtain multiple candidate areas and corresponding seepage probabilities in the infrared image, and the infrared image includes temperature data of multiple pixel points; A seepage area detection module, configured to screen multiple candidate areas based on the seepage probability and the temperature data to obtain the seepage area in the infrared image; A seepage coordinate conversion module, configured to obtain the first coordinate of the seepage area in the detection coordinate system, and based on the conversion relationship between the building drawing coordinate system and the detection coordinate system stored in advance, convert the first coordinate to the building drawing coordinate system to obtain the second coordinate of the seepage area; A seepage area annotation module, configured to identify multiple boundary pixel coordinates of the seepage area in the building drawing coordinate system, determine the seepage area based on the multiple boundary pixel coordinates, and map the seepage area to the building drawing of the building to be detected based on the second coordinate and the seepage area to obtain the building seepage annotation result based on robot cruising.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented.

[0016] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0017] In a fifth aspect, an embodiment of the present application further provides a computer program product, which includes a computer program. The computer program is stored in a computer-readable storage medium, and at least one processor of the device reads and executes the computer program, so that the device executes the method described in the first aspect.

[0018] In the embodiment of the present application, an infrared image captured by a robot during a cruise of a building to be detected is obtained, and the infrared image is input into a pre-trained water seepage detection model. Water seepage detection is performed based on the water seepage detection model to obtain multiple candidate regions and corresponding water seepage probabilities in the infrared image. The infrared image includes temperature data of multiple pixel points; multiple candidate regions are screened based on the water seepage probability and temperature data to obtain water seepage regions in the infrared image; a first coordinate of the water seepage region in the detection coordinate system is obtained, and based on a pre-stored conversion relationship between the building drawing coordinate system and the detection coordinate system, the first coordinate is converted to the building drawing coordinate system to obtain a second coordinate of the water seepage region; multiple boundary pixel coordinates of the water seepage region in the building drawing coordinate system are identified, the water seepage area is determined based on the multiple boundary pixel coordinates, and the water seepage region is mapped to the building drawing of the building to be detected based on the second coordinate and the water seepage area to obtain a building water seepage annotation result based on robot cruise. Through the above-mentioned building water seepage annotation method based on robot cruise, the problems of low detection efficiency, inaccurate detection results, and untimely feedback of water seepage conditions existing in building water seepage detection using the prior art are solved. By determining the water seepage region according to the water seepage probability of the infrared image and a preset water seepage probability threshold, and converting the water seepage region to the building drawing coordinate system for building water seepage annotation based on robot cruise, the purpose of automatically performing water seepage detection and water seepage annotation based on the infrared image can be achieved, improving the detection efficiency of building water seepage and the accuracy of detection results, and at the same time facilitating the timely feedback of water seepage conditions. Description of the Drawings

[0019] Figure 1 is a flowchart of a building water seepage annotation method based on robot cruise provided by an embodiment of the present application; Figure 2It is an infrared schematic diagram of building water seepage to be detected provided by this application; Figure 3 It is a flowchart for screening candidate areas provided by an embodiment of this application; Figure 4 It is a flowchart of another building water seepage annotation method based on robot cruise provided by an embodiment of this application; Figure 5 It is a structural block diagram of a building water seepage annotation device based on robot cruise provided by an embodiment of this application; Figure 6 It is a structural block diagram of an electronic device provided by an embodiment of this application. Detailed implementation manners

[0020] In order to make the objectives, technical solutions, and advantages of this application clearer, the following further describes the specific embodiments of this application in detail with reference to the accompanying drawings. It can be understood that the specific embodiments described herein are only used to explain this application, rather than limiting this application. Additionally, it should be noted that for the sake of description, only parts related to this application are shown in the drawings rather than all the content. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. When the operations are completed, the process can be terminated, but there can also be additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0021] The following will clearly describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application belong to the scope of protection of this application.

[0022] The terms "first", "second", etc. in the specification and claims of this application are used to distinguish similar objects, rather than to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first", "second", etc. generally belong to the same category, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and / or" in the specification and claims means at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the associated objects before and after.

[0023] First, the application scenario of this solution can be a scenario for detecting building water seepage, especially a scenario for automatically detecting the water seepage situation of building floors by using a robot to cruise and capture infrared images. By determining the water seepage area according to the water seepage probability of the infrared image and a preset water seepage probability threshold, and converting the water seepage area to the building drawing coordinate system for building water seepage annotation based on robot cruise, the purpose of automatically detecting and annotating water seepage based on infrared images can be achieved, improving the detection efficiency of building water seepage and the accuracy of the detection results. At the same time, it is beneficial to timely feedback the water seepage situation. Based on the above application scenario, it can be understood that the execution subject of this solution can be an electronic device, such as intelligent terminals like mobile phones, tablets, and desktop computers.

[0024] Next, in conjunction with the accompanying drawings, a method, device, equipment, and medium for building water seepage annotation based on robot cruise provided by an embodiment of the present application will be described in detail through specific embodiments and their application scenarios.

[0025] Figure 1 is a flowchart of a method for building water seepage annotation based on robot cruise provided by an embodiment of the present application. As Figure 1 shown, it specifically includes the following steps: S101, Obtain an infrared image captured by a robot during the cruise of the building to be detected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate regions and corresponding water seepage probabilities in the infrared image. The infrared image includes temperature data of multiple pixel points.

[0026] Among them, the building to be detected can be a building that needs to be detected for water seepage or the internal structure of a building, such as the ceiling of a floor, the floor of a floor, etc. The infrared image can be a distribution map of the surface temperature values of the floor generated by capturing the infrared radiation emitted or reflected by the building to be detected using an infrared camera. The infrared image reflects the temperature difference on the surface of the building to be detected, and different pixel values represent different temperature values. The pre-trained water seepage detection model can be a model used to perform image recognition on the infrared image to predict the water seepage situation of the corresponding building structure. The candidate region can be an area in the infrared image predicted by the water seepage detection model where water seepage may occur. The water seepage probability can be data used to describe the possibility of water seepage in each candidate region. The temperature data can be the temperature value of the position of the building structure corresponding to each pixel point in the infrared image.

[0027] In one embodiment, an infrared image captured by a robot during a cruise inspection of a building to be inspected can be received in real time. The infrared image is input into a pre-trained water seepage detection model, and the water seepage detection model performs image recognition and processing on the infrared image, outputting an infrared image with water seepage areas outlined. The infrared image output by the water seepage detection model includes multiple candidate areas where water seepage may occur outlined by the model and the corresponding water seepage probability for each candidate area. The infrared image includes temperature data for each pixel point.

[0028] Taking the water seepage detection of the internal floors of the building to be inspected as an example, a lidar device can be pre-configured on the cruise robot and the cruise robot can be placed inside the building to be inspected. The robot can use lidar in combination with SLAM (Simultaneous Localization and Mapping) technology for autonomous navigation. SLAM technology allows the robot to build a map and locate itself in real time in an unknown environment, thus ensuring that it can automatically plan a cruise path within the floors of the building to be inspected, avoid collisions, and ensure coverage of all water seepage inspection areas. The lidar can provide high-precision environmental modeling and provide input data for the SLAM algorithm, enabling the robot to complete accurate positioning and mapping in a complex environment. During the autonomous cruise process of the robot, the floors are scanned in real time through the mounted infrared camera to capture temperature differences and identify potential water seepage areas based on the temperature differences.

[0029] Figure 2 It is an infrared schematic diagram of the building to be inspected provided by this application.

[0030] As Figure 2 shown in the figure, the figure includes dark strip areas and large light-colored areas. Among them, the dark strip areas are the areas where water seepage occurs in the building to be inspected, that is, the candidate areas. The darker the color of the candidate area, the more serious the water seepage degree in this area, and the higher the water seepage probability predicted by the water seepage detection model. The large light-colored areas in the figure are the areas where there is no water seepage in the building to be inspected. The numbers in the figure represent the pixel point temperature data at that point.

[0031] By using an infrared camera to sense the temperature changes on the surface of the building floors, water seepage areas usually exhibit different temperature characteristics from the surrounding areas due to the influence of moisture. Preprocessing operations such as denoising, enhancing contrast, and temperature calibration can be performed on the obtained infrared image, and the infrared image can be presented in white-hot mode to improve the accuracy of water seepage detection. The water seepage areas of the building floors usually exhibit lower temperature differences due to the influence of moisture. In the preprocessed infrared image, these areas affected by moisture and the areas not affected by moisture will exhibit different color depths, and the areas with darker colors are regarded as areas with potential signs of water seepage.

[0032] The water seepage detection model in this solution is a ResNet (Residual Networks) deep learning model. ResNet has extremely strong expressiveness and stability in image recognition tasks and is suitable for water seepage detection problems in complex environments. By utilizing the residual learning mechanism of ResNet, the problem of gradient disappearance during the training of deep networks can be effectively avoided, and the detection accuracy can be improved. Before training the water seepage detection model, a large number of labeled infrared image data need to be collected in advance. The labels include information such as whether there is a water seepage area in the image, the specific location of the water seepage area, and the water seepage probability value of this area. Data augmentation is performed on the training set of infrared image data to improve the robustness and generalization ability of the model. The data augmentation methods include rotation, scaling, cropping, and simulating water seepage areas of different angles and sizes to improve the model's ability to adapt to different scenarios; it also includes simulating infrared images under different temperature conditions to improve the model's adaptability to different temperature environments. During the model training process, the unlabeled infrared images in the training set are used as the model input, and the infrared image data with the labels of the area where the water seepage phenomenon is framed and the water seepage probability value label of this area are used as the output to train the water seepage detection model. The cross-entropy loss function is used to minimize the error between the label and the prediction result, and GPU acceleration is used for training. The network parameters are optimized through the gradient descent optimizer Adam. After the model training is completed, the trained ResNet model is used to predict the water seepage area of new infrared images. During the process of the water seepage detection model predicting infrared images, the model analyzes each pixel point in the image and predicts whether the area composed of multiple pixel points is a water seepage area and what the probability of this area being a water seepage area is. The output of the water seepage detection model is the water seepage area probability value.

[0033] The technology of using robot automatic cruise to collect infrared images and combining deep learning to automatically analyze infrared images in this solution enables the robot to autonomously inspect the water seepage phenomenon on the building floors and achieve the purpose of automatically identifying the water seepage area.

[0034] S102, Screen multiple candidate areas based on the water seepage probability and temperature data to obtain the water seepage area in the infrared image.

[0035] Among them, the water seepage area can be the area where there must be a water seepage phenomenon in the building to be detected.

[0036] In one embodiment, the sum of the water seepage probability and the temperature data of each candidate region can be calculated, and the sums of multiple candidate regions can be sorted from largest to smallest. The candidate regions with the sums ranked among the top preset number are used as the water seepage regions in the infrared image. Additionally, the water seepage probability weight and the temperature data weight can be preset in advance. By calculating the weighted sum of the water seepage probability, the water seepage probability weight, the temperature data, and the temperature data weight for each candidate region respectively, multiple candidate regions can be screened, and the candidate region with the largest weighted sum is used as the water seepage region in the infrared image. Further, the water seepage probability threshold and the temperature data threshold can be preset in advance. By comparing whether the water seepage probability of each candidate region is higher than the preset water seepage probability threshold and whether the sum of the pixel temperature data of each candidate region is higher than the preset temperature data threshold respectively, when the water seepage probability is higher than the preset water seepage probability threshold and the sum of the pixel temperature data is higher than the preset temperature data threshold, the candidate region is determined as the water seepage region in the infrared image.

[0037] S103. Obtain the first coordinates of the water seepage region in the detection coordinate system, and based on the pre-stored conversion relationship between the building drawing coordinate system and the detection coordinate system, convert the first coordinates to the building drawing coordinate system to obtain the second coordinates of the water seepage region.

[0038] Among them, the detection coordinate system can be a two-dimensional coordinate system constructed for the overall internal structure of the entire building to be detected. The detection coordinate system in this solution can be constructed by controlling the robot to move on the internal ground of the building to be detected, scanning the overall internal structure of the entire building to be detected, and using lidar combined with SLAM technology to construct the detection coordinate system of the entire building to be detected. The detection coordinate system of lidar combined with SLAM technology is initially a three-dimensional coordinate system. In order to achieve subsequent coordinate conversion with the building coordinate system and map the water seepage region to the building drawing, the three-dimensional coordinate system needs to be mapped to a two-dimensional coordinate system. The two-dimensional detection coordinate system in this solution takes the initial position of the robot moving on the internal ground of the building to be detected as the origin of the coordinate system, the positive front direction of the robot as the positive direction of the x-axis of the coordinate system, and the left side of the robot as the positive direction of the y-axis of the coordinate system. After the detection coordinate system is constructed, it will not change with the movement of the robot, and the robot can real-time locate its position coordinates in the detection coordinate system during cruising in the building to be detected, and calculate the position coordinates of the infrared images scanned during the cruising process and each region inside the infrared images in the detection coordinate system. The first coordinates can be the position coordinates of the water seepage region in the detection coordinate system. The building drawing coordinate system can be the coordinate system corresponding to the plane drawing of the building to be detected. The conversion relationship between the building drawing coordinate system and the detection coordinate system can be the calculation formula for converting the coordinates in the building drawing coordinate system to the corresponding coordinates in the detection coordinate system. The second coordinates can be the position coordinates of the water seepage region in the building drawing coordinate system.

[0039] In one embodiment, the first coordinate of the water seepage area in the detection coordinate system can be obtained by reading the cruise data of the robot, and the coordinate obtained by converting the first coordinate to the building drawing coordinate system can be calculated through the conversion relationship between the pre-stored building drawing coordinate system and the detection coordinate system, so as to obtain the second coordinate of the water seepage area.

[0040] The detection coordinate system in this solution is a SLAM two-dimensional coordinate system. During the robot's cruise, the robot uses lidar to scan the internal environment of the building to be detected, and gradually constructs a map containing internal building features such as building structures, walls, and obstacles through SLAM technology, usually a two-dimensional grid map or a point cloud map. SLAM technology can not only help the robot to locate in real time, but also integrate the internal environment information of the building to be detected into the map for subsequent positioning of the water seepage area. The building drawing in this solution is a CAD drawing, which is a two-dimensional plan of the internal environment of the building to be detected. The building drawing coordinate system is drawn in advance based on the local coordinate system of the building to be detected. There is a coordinate system conversion relationship between the detection coordinate of the water seepage area in the detection coordinate system and the actual coordinate in the building drawing coordinate system. In the internal environment of the building to be detected, some reference points with known positions, such as the positions of fixed objects such as doorways and corners, can be selected as reference points for coordinate conversion. By obtaining the first coordinate of these reference points in the detection coordinate system and the second coordinate in the building drawing coordinate system respectively, an affine transformation is used to calculate the conversion matrix between the detection coordinate system and the building drawing coordinate system, align the detection coordinate system with the building drawing coordinate system, and then convert the position coordinates detected in the SLAM system into the position coordinates in the CAD drawing coordinate system. Through coordinate registration, the coordinates of the water seepage area detected by the robot in the SLAM system can be mapped to the corresponding position in the CAD drawing for marking the water seepage area. The marking of the water seepage area includes information such as the abnormal temperature degree, position, and area of the water seepage. The robot pushes the detection results of the water seepage area and its position annotation on the CAD drawing to the maintenance personnel of the building to be detected in real time through the 5G wireless network, so that the maintenance personnel can use intelligent terminals such as mobile devices, tablets, or computers to view the detailed information of the water seepage area in time.

[0041] This solution combines SLAM technology for environment mapping and robot positioning. Through coordinate registration, it ensures that the water seepage detection information is highly aligned with the CAD drawing of the building, and can achieve more accurate and comprehensive water seepage identification capabilities, as well as intelligent cruise and real-time detection of water seepage information. Through real-time data collection and water seepage area feedback during the robot's cruise, and automatically marking the water seepage position on the CAD drawing of the building based on the coordinate conversion relationship and pushing it to the maintenance personnel in real time through the wireless network, the timeliness of water seepage feedback and water seepage maintenance can be improved.

[0042] S104. Identify the coordinates of multiple boundary pixels of the water seepage area in the building drawing coordinate system, determine the water seepage area based on the coordinates of the multiple boundary pixels, and map the water seepage area to the building drawing of the building to be detected based on the second coordinate and the water seepage area, so as to obtain the building water seepage annotation result based on robot cruise.

[0043] Among them, the boundary pixel coordinates can be the coordinates of the edge pixel points of the water seepage area in the building drawing coordinate system. The building water seepage annotation result is the building drawing data after marking the coordinates of the water seepage position and the water seepage area data of the building to be detected on the building drawing.

[0044] In one embodiment, the edge pixels of the water seepage area can be identified, and the corresponding boundary pixel coordinates of each edge pixel in the building drawing coordinate system can be obtained respectively, and the water seepage area can be calculated according to the boundary pixel coordinates. Determine the position of the water seepage area on the building drawing according to the second coordinate, determine the range of the water seepage area on the building drawing according to the water seepage area, and map the water seepage area to the building drawing of the building to be detected according to the position and the range, so as to mark the water seepage area on the building drawing, and obtain the building water seepage annotation result based on robot cruise.

[0045] The technical solution provided by the embodiments of the present application is to obtain the infrared image captured by the robot during the cruise of the building to be detected, input the infrared image into the pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate areas and corresponding water seepage probabilities in the infrared image. The infrared image includes the temperature data of multiple pixel points; filter the multiple candidate areas based on the water seepage probability and the temperature data to obtain the water seepage area in the infrared image; obtain the first coordinate of the water seepage area in the detection coordinate system, and based on the conversion relationship between the pre-stored building drawing coordinate system and the detection coordinate system, convert the first coordinate to the building drawing coordinate system to obtain the second coordinate of the water seepage area; identify the coordinates of multiple boundary pixels of the water seepage area in the building drawing coordinate system, determine the water seepage area based on the coordinates of the multiple boundary pixels, and map the water seepage area to the building drawing of the building to be detected based on the second coordinate and the water seepage area, so as to obtain the building water seepage annotation result based on robot cruise. Through the above building water seepage annotation method based on robot cruise, the problems of low detection efficiency, inaccurate detection results and untimely feedback of water seepage conditions existing in building water seepage detection using the existing technology are solved. By determining the water seepage area according to the water seepage probability of the infrared image and the preset water seepage probability threshold, and converting the water seepage area to the building drawing coordinate system for building water seepage annotation based on robot cruise, the purpose of automatically detecting and annotating water seepage based on infrared images can be achieved, the detection efficiency of building water seepage and the accuracy of detection results are improved, and at the same time, it is beneficial to timely feedback the water seepage situation.

[0046] Figure 3This is a flowchart for screening candidate regions provided by an embodiment of the present application. As Figure 3 shown, it specifically includes the following steps: S301, group multiple candidate regions based on temperature data to obtain multiple groups of candidate regions.

[0047] Among them, the groups of candidate regions can be groups obtained by grouping candidate regions with the same or similar pixel temperatures.

[0048] In one embodiment, the average pixel temperature of each candidate region can be calculated based on the temperature data, and candidate regions with relatively small differences in average pixel temperature are divided into the same group based on a clustering algorithm to obtain multiple groups of candidate regions.

[0049] In one embodiment, optionally, grouping multiple candidate regions based on temperature data includes: Determine the pixel point temperature distribution information corresponding to each candidate region based on the temperature data; Determine the seepage similarity between multiple candidate regions based on the pixel point temperature distribution information, and group the multiple candidate regions based on the seepage similarity.

[0050] Among them, the pixel point temperature distribution information can be information used to describe in which temperature ranges the pixel point temperature data in the same candidate region is distributed. The seepage similarity can be used to describe whether the seepage degrees between different candidate regions are the same. The pixel point temperature distribution information can represent the seepage degree of each candidate region. The lower the temperature in the pixel point temperature distribution information set, the more serious the seepage degree. According to the similarity between the pixel point temperature distribution information, the similarity of the seepage degrees between multiple candidate regions can be determined.

[0051] In one embodiment, the data that is relatively concentrated in the pixel temperature data corresponding to each candidate region can be determined as the pixel point temperature distribution information corresponding to each candidate region, calculate the distance between the pixel point temperature distribution information of multiple different candidate regions, and determine the seepage similarity between multiple candidate regions. According to the seepage similarity, candidate regions with higher similarity are divided into the same group to obtain the grouping results of multiple candidate regions. A similarity threshold can be preset, and multiple candidate regions with similarity higher than the preset similarity threshold are divided into the same group; alternatively, the number of groups can be preset, and based on the number of candidate regions and the preset number of groups, the number of candidate regions in each group is determined, and multiple candidate regions are grouped according to the number of candidate regions in each group in descending order of similarity.

[0052] In this solution, by determining the pixel temperature distribution information corresponding to each candidate area based on temperature data, determining the seepage similarity between multiple candidate areas, and grouping the multiple candidate areas based on the seepage similarity, the purpose of dividing subsequent areas with similar seepage degrees into the same group can be achieved, improving the rationality of grouping.

[0053] S302. Determine the overall seepage probability of the same candidate area group based on the seepage probability, and screen multiple candidate areas based on the overall seepage probability.

[0054] Among them, the overall seepage probability can be data used to describe the seepage possibility of the same group.

[0055] In one embodiment, the sum of the seepage probabilities corresponding to multiple candidate areas in each group can be calculated to determine the overall seepage probability of the same candidate area group; alternatively, the seepage probabilities corresponding to multiple candidate areas in each group can be sorted according to the magnitude relationship, and the maximum value of the seepage probabilities in each group can be determined as the overall seepage probability of the same candidate area group. According to the overall seepage probability, all candidate areas corresponding to the group with the smallest overall seepage probability among multiple groups are filtered out.

[0056] In one embodiment, optionally, screening multiple candidate areas based on the overall seepage probability includes: Calculating the average group seepage probability of multiple candidate area groups based on the overall seepage probability; Identifying the magnitude relationship between the seepage probability corresponding to each candidate area and the average group seepage probability, and filtering out candidate areas with seepage probabilities less than the average group.

[0057] In one embodiment, the average group seepage probability of multiple candidate area groups can be calculated according to the overall seepage probability and the number of candidate area groups. Compare the magnitude relationship between the seepage probability corresponding to each candidate area and the average group seepage probability, and filter out candidate areas with seepage probabilities less than the average group.

[0058] In this solution, by calculating the average group seepage probability of multiple candidate area groups based on the overall seepage probability, and filtering out candidate areas with seepage probabilities less than the average group according to the magnitude relationship between the seepage probability corresponding to each candidate area and the average group seepage probability, the purpose of screening seepage areas based on the overall seepage situation of candidate areas can be achieved, avoiding the problem that the seepage detection model is not accurate enough in identifying the seepage probability of infrared images due to external environmental interference, and improving the flexibility and accuracy of seepage area identification.

[0059] The technical solution provided by the embodiments of the present application groups multiple candidate regions according to temperature data, calculates the overall water seepage probability of the same candidate region group according to the water seepage probability, and screens multiple candidate regions based on the overall water seepage probability, so as to achieve the purpose of judging the water seepage region by combining temperature data and water seepage probability, and improve the accuracy of the water seepage region recognition result.

[0060] Figure 4 is a flowchart of another building water seepage annotation method based on robot cruising provided by the embodiments of the present application. As Figure 4 shown, the specific steps are as follows: S401, obtain the infrared image taken by the robot during the cruise of the building to be detected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate regions and corresponding water seepage probabilities in the infrared image. The infrared image includes temperature data of multiple pixel points.

[0061] S402, calculate the temperature difference data of the candidate regions based on the temperature data of multiple pixel points, and perform accuracy verification on the water seepage probability based on the temperature difference data.

[0062] Among them, the temperature difference data can be the difference between the highest pixel point temperature and the lowest pixel point dimension in each candidate region. The accuracy verification can be an operation for identifying whether the water seepage probability of each candidate region is accurate. The water seepage detection model in this solution is based on the water seepage probability of each candidate region obtained by image recognition of the infrared image. When there is an object occlusion problem in the building structure to be detected corresponding to the infrared image, due to the difference in materials between the occlusion object and the building to be detected, there will be a situation where the image color corresponding to the occlusion object in the captured infrared image is the same as or similar to the image color of the water seepage region. At this time, it is necessary to identify the water seepage probability according to the temperature data of the pixel points.

[0063] In one embodiment, the temperature difference data of the corresponding candidate region can be calculated according to the maximum temperature value and the minimum temperature value of multiple pixel points in each candidate region, the consistency between the temperature difference data of each candidate region and the water seepage probability is compared, and the inconsistent water seepage probability is determined as an inaccurate probability. The water seepage possibility of each candidate region can be sorted according to the size of the temperature difference data. The larger the temperature difference data, the higher the water seepage possibility; at the same time, the water seepage possibility of each candidate region is sorted according to the size of the water seepage probability. The larger the water seepage probability, the higher the water seepage possibility. Compare whether the positions of the same candidate region in the two sorts are the same, and determine the candidate region at the same position as the one where the temperature difference data is consistent with the water seepage probability, that is, the water seepage probability of this candidate region is accurate.

[0064] In one embodiment, optionally, performing accuracy verification on the water seepage probability based on the temperature difference data includes: Identify the correlation between the temperature difference data and the preset temperature difference range, where the preset temperature difference range includes a preset material temperature difference range and a preset seepage water temperature difference range; When the temperature difference data is within the preset seepage water temperature difference range, the seepage probability can be accurately determined; When the temperature difference data is within the preset material temperature difference range, the seepage probability cannot be accurately determined.

[0065] Among them, the preset material temperature difference range can be a temperature data region composed of the maximum and minimum values of the pixel point temperature difference caused by material differences set in advance. The preset seepage water temperature difference range is a temperature data region composed of the maximum and minimum values of the pixel point temperature difference caused by seepage water set in advance.

[0066] In one embodiment, the temperature difference data of each candidate region can be compared with the preset material temperature difference range and the preset seepage water temperature difference range respectively to identify the correlation between the temperature difference data and the preset temperature difference range. When the temperature difference data is within the preset seepage water temperature difference range, it indicates that the temperature difference of the candidate region is caused by the seepage water problem. At this time, the seepage probability can be accurately determined; when the temperature difference data is within the preset material temperature difference range, it indicates that the temperature difference of the candidate region is caused by different materials. At this time, the seepage probability cannot be accurately determined.

[0067] In this solution, by identifying the correlation between the temperature difference data and the preset temperature difference range and determining whether the seepage probability is accurate, the efficiency of accurately identifying the seepage probability can be improved.

[0068] In one embodiment, optionally, before calculating the temperature difference data of the candidate region based on the temperature data of multiple pixel points, the method further includes: Obtain the ambient temperature data and the material temperature rise data corresponding to the candidate region; Predict the highest seepage water temperature of the candidate region based on the ambient temperature data and the material temperature rise data.

[0069] Correspondingly, the accuracy verification of the seepage probability based on the temperature difference data includes: Determine the estimated lowest seepage water temperature of the candidate region based on the temperature difference data and the highest seepage water temperature, and identify whether the actual lowest temperature in the temperature data of the candidate region is greater than the estimated lowest seepage water temperature; Based on the identification result of the actual lowest temperature, perform accuracy verification on the seepage probability.

[0070] Among them, the material temperature rise data can be the temperature variable of the material of the building to be detected corresponding to the candidate region that changes with the ambient temperature. The highest seepage water temperature can be the highest temperature of the building to be detected when there is seepage in the candidate region. The estimated lowest seepage water temperature of the candidate region can be the lowest temperature of the region determined according to its actual temperature difference data when there is seepage in the candidate region.

[0071] In one embodiment, environmental temperature data and material temperature rise data corresponding to a candidate area can be obtained, and the highest seepage temperature corresponding to the candidate area can be calculated based on the environmental temperature data and the material temperature rise data. The estimated lowest seepage temperature in the case of seepage in the candidate area can be calculated based on the temperature difference data and the highest seepage temperature. The minimum temperature data in the pixel point temperature data is identified to obtain the actual lowest temperature of the candidate area, and it is compared whether the actual lowest temperature is greater than the estimated lowest seepage temperature. In the case where the actual lowest temperature is greater than the estimated lowest seepage temperature, it is determined that there is no seepage in the candidate area, and it is identified whether the seepage probability at this time is 0. If it is not 0, it is determined that the seepage probability is inaccurate; in the case where the actual lowest temperature is less than or equal to the estimated lowest seepage temperature, it is determined that there is seepage in the candidate area, and it is identified whether the seepage probability at this time is 0. If it is not 0, it is determined that the seepage probability is accurate.

[0072] In this solution, by predicting the highest seepage temperature of the candidate area based on the environmental temperature data and the material temperature rise data, determining the estimated lowest seepage temperature of the candidate area based on the temperature difference data and the highest seepage temperature, and performing accuracy verification on the seepage probability based on the actual lowest temperature identification result, the purpose of taking into account the influencing factor of the temperature difference change caused by the material temperature rise in the process of seepage probability identification can be achieved, and the comprehensiveness of the accuracy evaluation of the seepage probability and the accuracy of the final identification result of the seepage probability are improved.

[0073] S403. Update the seepage probability based on the accuracy verification result to obtain the final seepage probability.

[0074] In one embodiment, it can be determined whether the seepage probability is accurate according to the accuracy verification result, and the inaccurate seepage probability is set to 0 to obtain the final seepage probability. Since the seepage probability prediction error is caused by object occlusion, when there is object occlusion between the building to be detected and the infrared camera and there is no seepage, the seepage probability is not 0. At this time, the seepage probability is inaccurate, and setting it to 0 can update the inaccurate seepage probability.

[0075] S404. Screen multiple candidate areas based on the seepage probability and the temperature data to obtain the seepage area in the infrared image.

[0076] S405. Obtain the first coordinate of the seepage area in the detection coordinate system, and based on the pre-stored conversion relationship between the building drawing coordinate system and the detection coordinate system, convert the first coordinate to the building drawing coordinate system to obtain the second coordinate of the seepage area.

[0077] S406. Identify the coordinates of multiple boundary pixels of the water seepage area in the building drawing coordinate system, determine the water seepage area based on the coordinates of multiple boundary pixels, and map the water seepage area to the building drawing of the building to be detected based on the second coordinate and the water seepage area, so as to obtain the building water seepage annotation result based on robot cruise.

[0078] The technical solution provided by the embodiments of the present application, before screening multiple candidate areas based on the water seepage probability and temperature data, calculates the temperature difference data of the candidate areas based on the temperature data of multiple pixel points, verifies the accuracy of the water seepage probability based on the temperature difference data, and updates the water seepage probability based on the accuracy verification result to obtain the final water seepage probability, which can improve the accuracy of the water seepage area identification result and is beneficial to the accurate judgment of building water seepage.

[0079] Figure 5 It is a structural block diagram of a building water seepage annotation device based on robot cruise provided by the embodiments of the present application. As Figure 5 shown, it specifically includes the following: The water seepage probability determination module 501 is used to obtain the infrared image captured by the robot during the cruise of the building to be detected, input the infrared image into the pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate areas and corresponding water seepage probabilities in the infrared image. The infrared image includes the temperature data of multiple pixel points; The water seepage area detection module 502 is used to screen multiple candidate areas based on the water seepage probability and temperature data to obtain the water seepage area in the infrared image; The water seepage coordinate conversion module 503 is used to obtain the first coordinate of the water seepage area in the detection coordinate system, and convert the first coordinate to the building drawing coordinate system based on the pre-stored conversion relationship between the building drawing coordinate system and the detection coordinate system to obtain the second coordinate of the water seepage area; The water seepage area annotation module 504 is used to identify the coordinates of multiple boundary pixels of the water seepage area in the building drawing coordinate system, determine the water seepage area based on the coordinates of multiple boundary pixels, and map the water seepage area to the building drawing of the building to be detected based on the second coordinate and the water seepage area, so as to obtain the building water seepage annotation result based on robot cruise.

[0080] Optionally, the water seepage area detection module 502 is specifically used for: Group multiple candidate areas based on the temperature data to obtain multiple candidate area groups; Determine the overall water seepage probability of the same candidate area group based on the water seepage probability, and screen multiple candidate areas based on the overall water seepage probability.

[0081] Optionally, the water seepage area detection module 502 is specifically used for: Determine the pixel temperature distribution information corresponding to each candidate area based on the temperature data; Determine the seepage similarity between multiple candidate areas based on the pixel temperature distribution information, and group the multiple candidate areas based on the seepage similarity.

[0082] Optionally, the seepage area detection module 502 is specifically configured to: Calculate the average group seepage probability of multiple candidate area groups based on the overall seepage probability; Identify the magnitude relationship between the seepage probability corresponding to each candidate area and the average group seepage probability, and filter out the candidate areas with seepage probability less than the average group.

[0083] Optionally, the device further includes: An accuracy verification module, configured to calculate the temperature difference data of the candidate area based on the temperature data of multiple pixels, and perform accuracy verification on the seepage probability based on the temperature difference data; A seepage probability update module, configured to update the seepage probability based on the accuracy verification result to obtain the final seepage probability.

[0084] Optionally, the accuracy verification module is specifically configured to: Identify the association relationship between the temperature difference data and the preset temperature difference range, where the preset temperature difference range includes a preset material temperature difference range and a preset seepage temperature difference range; Determine that the seepage probability is accurate when the temperature difference data is within the preset seepage temperature difference range; Determine that the seepage probability is inaccurate when the temperature difference data is within the preset material temperature difference range.

[0085] Optionally, the device further includes: A temperature data acquisition module, configured to acquire ambient temperature data and the material temperature rise data corresponding to the candidate area; A maximum temperature prediction module, configured to predict the maximum seepage temperature of the candidate area based on the ambient temperature data and the material temperature rise data.

[0086] Correspondingly, the accuracy verification module is specifically configured to: Determine the estimated minimum seepage temperature of the candidate area based on the temperature difference data and the maximum seepage temperature, and identify whether the actual minimum temperature of the candidate area in the temperature data is greater than the estimated minimum seepage temperature; Perform accuracy verification on the seepage probability based on the actual minimum temperature identification result.

[0087] In the technical solution provided by the embodiment of the present application, the water seepage probability determination module is configured to obtain an infrared image captured by a robot during the cruise of a building to be detected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate regions and corresponding water seepage probabilities in the infrared image. The infrared image includes temperature data of multiple pixels; the water seepage area detection module is configured to screen the multiple candidate regions based on the water seepage probability and temperature data to obtain the water seepage area in the infrared image; the water seepage coordinate conversion module is configured to obtain the first coordinate of the water seepage area in the detection coordinate system, and based on the conversion relationship between the building drawing coordinate system and the detection coordinate system stored in advance, convert the first coordinate to the building drawing coordinate system to obtain the second coordinate of the water seepage area; the water seepage area annotation module is configured to identify multiple boundary pixel coordinates of the water seepage area in the building drawing coordinate system, determine the water seepage area based on the multiple boundary pixel coordinates, and map the water seepage area to the building drawing of the building to be detected based on the second coordinate and the water seepage area to obtain the building water seepage annotation result based on the robot cruise. Through the above-mentioned building water seepage annotation device based on robot cruise, the problems of low detection efficiency, inaccurate detection results, and untimely feedback of water seepage conditions existing in building water seepage detection using the prior art are solved. By determining the water seepage area according to the water seepage probability of the infrared image and the preset water seepage probability threshold, and converting the water seepage area to the building drawing coordinate system for building water seepage annotation based on robot cruise, the purpose of automatically performing water seepage detection and water seepage annotation based on infrared images can be achieved, improving the detection efficiency of building water seepage and the accuracy of detection results, and at the same time facilitating the timely feedback of water seepage conditions.

[0088] A building water seepage annotation device based on robot cruise in the embodiment of the present application can be configured in a device, or can be configured in components, integrated circuits, or chips in a terminal. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a handheld computer, a vehicle-mounted electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc. The non-mobile electronic device can be a server, a Network Attached Storage (NAS), a personal computer (PC), a television (TV), a teller machine, or a self-service machine, etc. The embodiment of the present application does not make specific limitations.

[0089] An architectural seepage annotation device based on robot cruise in the embodiments of the present application can be an operating system. The operating system can be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiments of the present application.

[0090] An architectural seepage annotation device based on robot cruise provided by the embodiments of the present application can implement each process realized by the above method embodiments. To avoid repetition, it will not be elaborated here.

[0091] As Figure 6 shown, the embodiments of the present application also provide an electronic device 600, including a processor 601, a memory 602, a program or instruction stored on the memory 602 and executable on the processor 601. When the program or instruction is executed by the processor 601, it implements each process of the above embodiments of the architectural seepage annotation method based on robot cruise, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0092] It should be noted that the electronic device in the embodiments of the present application includes the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0093] The embodiments of the present application also provide a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, it implements each process of the above embodiments of the architectural seepage annotation method based on robot cruise, and can achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0094] Among them, the processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc.

[0095] The embodiments of the present application further provide a program product, which includes program code. When the program product runs on a computer device, the program code is used to cause the computer device to execute the steps in the methods according to various exemplary embodiments of the present application described above in this specification. For example, the computer device can execute an architectural seepage annotation method recorded in the embodiments of the present application. The program product can be implemented by any combination of one or more readable media.

[0096] It should be noted that in this text, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including a..." does not exclude the presence of additional identical elements in the process, method, article or device that includes such element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the reverse order according to the functions involved. For example, the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, the features described with reference to certain examples may be combined in other examples.

[0097] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0098] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them fall within the protection scope of the present application.

[0099] The above is only the preferred embodiment of the present application and the technical principles applied. The present application is not limited to the specific embodiments described here. Various obvious changes, re-adjustments and substitutions that can be made by those skilled in the art will not depart from the protection scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments. Without departing from the concept of the present application, more other equivalent embodiments may be included, and the scope of the present application is determined by the scope of the claims.

Claims

1. A building water seepage marking method based on robot cruising, characterized in that: The method comprises: Obtain an infrared image taken by the robot while cruising the building to be inspected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate areas and corresponding water seepage probabilities in the infrared image, wherein the infrared image includes temperature data of multiple pixel points; Screening the plurality of candidate areas based on the water seepage probability and the temperature data to obtain the water seepage area in the infrared image; Acquire a first coordinate of the water seepage area in a detection coordinate system, and based on a pre-stored conversion relationship between a building drawing coordinate system and the detection coordinate system, convert the first coordinate to the building drawing coordinate system to obtain a second coordinate of the water seepage area; Identify multiple boundary pixel coordinates of the water seepage area in the architectural drawing coordinate system, determine the water seepage area based on the multiple boundary pixel coordinates, map the water seepage area to the architectural drawing of the building to be inspected based on the second coordinates and the water seepage area, and obtain the building water seepage annotation result based on robot cruising.

2. The building water seepage marking method based on robot cruising according to claim 1 is characterized in that: The screening of the plurality of candidate areas based on the water seepage probability and the temperature data comprises: Grouping the multiple candidate areas based on the temperature data to obtain multiple candidate area groups; The overall water seepage probability of the same candidate area group is determined based on the water seepage probability, and the plurality of candidate areas are screened based on the overall water seepage probability.

3. The building water seepage marking method based on robot cruising according to claim 2 is characterized in that: The grouping the plurality of candidate areas based on the temperature data comprises: Determine the temperature distribution information of the pixels corresponding to each of the candidate areas based on the temperature data; The water seepage similarity between the multiple candidate areas is determined based on the pixel temperature distribution information, and the multiple candidate areas are grouped based on the water seepage similarity.

4. The building water seepage marking method based on robot cruising according to claim 2 is characterized in that: The screening of the plurality of candidate areas based on the overall water infiltration probability comprises: Calculate the average group water infiltration probability of the plurality of candidate area groups based on the overall water infiltration probability; The magnitude relationship between the water seepage probability corresponding to each candidate area and the average group water seepage probability is identified, and the candidate areas whose water seepage probability is smaller than the average group are filtered out.

5. The building water seepage marking method based on robot cruising according to claim 1 is characterized in that: Before screening the plurality of candidate areas based on the water infiltration probability and the temperature data, the method further includes: Calculating temperature difference data of the candidate area based on the temperature data of the plurality of pixel points, and performing accuracy verification on the water seepage probability based on the temperature difference data; The water seepage probability is updated based on the accuracy verification result to obtain a final water seepage probability.

6. The building water seepage marking method based on robot cruising according to claim 5 is characterized in that: The accuracy verification of the water seepage probability based on the temperature difference data includes: Identify the correlation between the temperature difference data and a preset temperature difference range, wherein the preset temperature difference range includes a preset material temperature difference range and a preset water seepage temperature difference range; When the temperature difference data is within the preset water seepage temperature difference range, determining that the water seepage probability is accurate; When the temperature difference data is within the preset material temperature difference range, it is determined that the water seepage probability is inaccurate.

7. The building water seepage marking method based on robot cruising according to claim 5 is characterized in that: Before calculating the temperature difference data of the candidate area based on the temperature data of the plurality of pixel points, the method further includes: Acquiring ambient temperature data and material temperature rise data corresponding to the candidate area; Predicting the maximum water seepage temperature of the candidate area based on the ambient temperature data and the material temperature rise data; Accordingly, the accuracy verification of the water seepage probability based on the temperature difference data includes: Determining an estimated minimum water seepage temperature of the candidate area based on the temperature difference data and the maximum water seepage temperature, and identifying whether an actual minimum temperature of the candidate area in the temperature data is greater than the estimated minimum water seepage temperature; Based on the actual minimum temperature identification result, the water seepage probability is checked for accuracy.

8. A building water seepage marking device based on robot cruising, characterized in that: The device comprises: A water seepage probability determination module is used to obtain an infrared image taken by the robot while cruising the building to be inspected, input the infrared image into a pre-trained water seepage detection model, perform water seepage detection based on the water seepage detection model, and obtain multiple candidate areas in the infrared image and corresponding water seepage probabilities, wherein the infrared image includes temperature data of multiple pixel points; A water seepage area detection module, used for screening the multiple candidate areas based on the water seepage probability and the temperature data to obtain the water seepage area in the infrared image; A water seepage coordinate conversion module, used for obtaining a first coordinate of the water seepage area in a detection coordinate system, and based on a conversion relationship between a pre-stored architectural drawing coordinate system and the detection coordinate system, converting the first coordinate into the architectural drawing coordinate system to obtain a second coordinate of the water seepage area; A water seepage area marking module is used to identify multiple boundary pixel coordinates of the water seepage area in the architectural drawing coordinate system, determine the water seepage area based on the multiple boundary pixel coordinates, and map the water seepage area to the architectural drawing of the building to be inspected based on the second coordinates and the water seepage area to obtain a building water seepage marking result based on robot cruising.

9. An electronic device, characterized in that: It includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of a building water seepage marking method based on robot cruising as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, the steps of a building water seepage marking method based on robot cruising as described in any one of claims 1 to 7 are implemented.

Images