Building defect identification method and device, equipment and storage medium

Abstract

The invention discloses a building defect recognition method and device, equipment and a storage medium, and relates to the technical field of defect recognition, and the method comprises the steps: obtaining a to-be-recognized building image, and carrying out the target detection of the to-be-recognized building image, and obtaining a target object in the to-be-recognized building image; performing defect detection on the target object, and determining a defect problem of the target object and a defect problem category corresponding to the defect problem; and generating a defect identification result of the target object based on the defect problem and the corresponding defect problem category. According to the method, the defect problems are classified, and the targeted identification strategy is adopted according to different defect problem categories, so that the problem of low identification precision of a traditional method in a complex scene is solved, and the accuracy and efficiency of building defect identification are improved.

Description

Technical Field

[0001] The present application relates to the technical field of defect identification, and in particular to a method, device, equipment and storage medium for identifying building defects. Background Art

[0002] Image recognition technology has shown great potential in the field of building defect detection, especially in conventional scenarios, it can effectively identify potential problems in buildings. However, when faced with specific complex scenarios, the limitations of this technology become apparent.

[0003] For example, if there is no door stopper installed on the entrance door, since the door itself is not damaged, traditional image recognition models often have difficulty effectively identifying such defects. In addition, image recognition models also face huge challenges in complex scenarios such as fully furnished houses. Fully furnished houses often contain a large number of decorations and furniture, which dramatically increases the amount of information that the model needs to process, resulting in an overload on the detection model. In such cases, the model may be unable to effectively distinguish between building defects that need to be detected and normal decorative elements, which greatly reduces the detection effect and reduces the accuracy of identifying building defects.

[0004] Therefore, how to improve the accuracy of defect recognition of image recognition technology when facing complex building scenes is a problem that needs to be solved urgently. Summary of the invention

[0005] The main purpose of this application is to provide a building defect recognition method, device, equipment and storage medium, aiming to solve the technical problem of how to improve the defect recognition accuracy of image recognition technology when facing complex building scenes.

[0006] To achieve the above objectives, the present application proposes a building defect recognition method, which comprises: Acquire a building image to be identified, and perform target detection on the building image to be identified to obtain a target object in the building image to be identified; Performing defect detection on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem; Based on the defect problem and the corresponding defect problem category, a defect identification result of the target object is generated.

[0007] In one embodiment, the defect problem category includes at least one of a single defect problem, a compound defect problem, and a cascade defect problem, wherein the single defect problem is a defect problem that exists independently of any sub-target object, the compound defect problem is a defect problem formed by a combination of multiple sub-target objects, and the cascade defect problem is a defect problem of a target defect sub-problem to be determined among multiple defect sub-problems; The step of generating a defect identification result of the target object based on the defect problem and the corresponding defect problem category comprises: In a case where the defect problem category is a single defect problem, generating a defect identification result of a first sub-target object associated with the defect problem based on the defect problem; In the case where the defect problem category is a composite defect problem, generating defect identification results for each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem; In the case where the defect problem category is a cascade defect problem, performing secondary defect detection on a third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object; The existing defect recognition results of the first sub-target object and / or the second sub-target object and / or the third sub-target object are integrated to obtain the defect recognition result of the target object.

[0008] In one embodiment, the step of generating defect identification results for each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem includes: Extracting features of each second sub-target object associated with the composite defect problem to obtain defect features of each second sub-target object; Comparing each defect feature with a reference defect feature to obtain composite defect information corresponding to the composite defect problem and a region where the composite defect information is located; A defect recognition result of each second sub-target object is generated based on the composite defect information and the corresponding area position.

[0009] In one embodiment, the step of performing secondary defect detection on the third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object includes: Calling a corresponding sub-goal detection model according to a third sub-goal object associated with the cascade defect problem; Extracting defect features of the third sub-target object through the sub-target detection model, and comparing the defect features with reference defect features to obtain refined defect information corresponding to the cascade defect problem and the regional location of the refined defect information; A defect recognition result of the third sub-target object is generated based on the refined defect information and the corresponding area position.

[0010] In one embodiment, the defect recognition results of the first sub-target object, the second sub-target object, and the third sub-target object include each defect information and the area position corresponding to any defect information, and the step of integrating the defect recognition results of the first sub-target object, the second sub-target object, and the third sub-target object to obtain the defect recognition result of the target object includes: Annotating the image of the building to be identified based on the position of each area to obtain an annotated defect detection image; Performing region spacing detection on each marked region in the defect detection image to obtain a region spacing detection result of the defect detection image; The marked areas are merged according to the area spacing detection result to obtain an adjusted defect detection image, and the adjusted defect detection image and the corresponding defect information are used as the defect recognition result of the target object.

[0011] In one embodiment, the step of performing defect detection on the target object and determining the defect problem of the target object and the defect problem category corresponding to the defect problem includes: Acquire the location data of the building image to be identified, and perform a relevance search on the target object according to the location data; When the search result of the association search is that the target object has an associated object, feature comparison is performed on the building image to be identified based on feature information of the associated object to obtain defect problems of each second sub-target object, and the defect problem category of the defect problem is determined to be a composite defect problem, wherein each second sub-target object includes a target sub-object associated with the target object and a corresponding associated object; When the retrieval result of the association retrieval is that the target object has no associated object, defect detection is performed on the target object to obtain the defect problem of the target object, and the defect problem category of the defect problem is determined as a single defect problem or a cascade defect problem.

[0012] In one embodiment, the step of performing defect detection on the target object, obtaining the defect problem of the target object, and determining the defect problem category of the defect problem as a single defect problem or a cascade defect problem includes: Performing defect detection on the target object to obtain defect problems of the target object, and performing detailed analysis on the defect problems; When the detailed analysis result shows that the defect problem has multiple defect sub-problems, the defect problem category of the defect problem is determined to be a cascade defect problem; When the detailed analysis result shows that the defect problem does not have any defect sub-problems, the defect problem category of the defect problem is determined to be a single defect problem.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a building defect recognition device, the building defect recognition device comprising: An object detection module is used to obtain a building image to be identified, and perform target detection on the building image to be identified to obtain a target object in the building image to be identified; A problem classification module, used to perform defect detection on the target object, determine the defect problem of the target object and the defect problem category corresponding to the defect problem; The result generation module is used to generate a defect identification result of the target object based on the defect problem and the corresponding defect problem category.

[0014] In addition, to achieve the above-mentioned purpose, the present application also proposes an electronic device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the building defect identification method as described above.

[0015] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the building defect identification method described above are implemented.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the building defect identification method described above are implemented.

[0017] One or more technical solutions proposed in this application have at least the following technical effects: The present application first acquires a building image to be identified, and performs target detection on the building image to be identified to obtain a target object in the building image to be identified, so as to accurately identify the target object to be detected from the building image; then, defect detection is performed on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem, so as to accurately identify and classify the defect problem of the target object, and provide an effective basis for the subsequent execution of different defect identification processes for different defect problem categories; finally, based on the defect problem and the corresponding defect problem category, a defect identification result of the target object is generated, so as to perform targeted defect identification based on the defect problem category, thereby supporting the effective identification of building defects in various complex scenarios.

[0018] In summary, the present application effectively distinguishes the categories of defect problems, thereby performing targeted defect identification for different categories of defect problems, avoiding the problem that traditional defect identification methods are unable to effectively identify defects in different complex scenes due to the complexity of building scenes, and achieving accurate and efficient identification of building defects in various complex scenes, thereby improving the efficiency and accuracy of building quality inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 A schematic diagram of a process flow provided for the first embodiment of the building defect identification method of the present application; Figure 2 A schematic diagram of a flow chart provided for the second embodiment of the building defect identification method of the present application; Figure 3 A schematic diagram of a brief process of a building defect identification method provided in Example 2 of the present application; Figure 4 This is a schematic diagram of the module structure of the building defect identification device according to an embodiment of the present application; Figure 5 Schematic diagram of the device structure of the hardware operating environment involved in the building defect identification method in the embodiment of the present application. DETAILED DESCRIPTION

[0022] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0023] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of the embodiment of the present application is: obtaining a building image to be identified, and performing target detection on the building image to be identified to obtain a target object in the building image to be identified; performing defect detection on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem; based on the defect problem and the corresponding defect problem category, generating a defect identification result of the target object.

[0025] Since existing image recognition technology has certain limitations when facing certain complex scenarios, for example, when there is no door stopper installed on the entrance door, since the door itself is not damaged, traditional image recognition models often find it difficult to effectively identify such defects. In addition, image recognition models also face huge challenges in complex scenarios such as fine-decorated houses. Fine-decorated houses often contain a large number of decorations and furniture, which dramatically increases the amount of information that the model needs to process, resulting in an overload on the detection model. In such a case, the model may be unable to effectively distinguish which are the building defects that need to be detected and which are normal decorative elements, which greatly reduces the detection effect and reduces the accuracy of identifying building defects. Therefore, how to improve the defect recognition accuracy of image recognition technology in the face of complex building scenes is a problem that needs to be solved urgently.

[0026] The present application provides a solution by effectively distinguishing the categories of defect problems, thereby performing targeted defect identification for different categories of defect problems, avoiding the problem that traditional defect identification methods are unable to effectively identify defects in different complex scenes due to the complexity of building scenes, and achieving accurate and efficient identification of building defects in various complex scenes, thereby improving the efficiency and accuracy of building quality inspection.

[0027] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a building defect recognition device, etc. The following takes the building defect recognition device as an example to illustrate this embodiment and the following embodiments.

[0028] Based on this, the present application embodiment provides a method for identifying building defects. Figure 1 , Figure 1This is a flow chart of the first embodiment of the building defect identification method of the present application.

[0029] In this embodiment, the building defect identification method includes steps S10 to S30: Step S10, acquiring a building image to be identified, and performing target detection on the building image to be identified to obtain a target object in the building image to be identified; It should be noted that the target object refers to the specific building components or structures extracted from the building image through the target detection algorithm, such as doors, windows, walls, floors, pipes, etc. These objects are the basic units of defect identification.

[0030] It is understandable that, since there are often many objects (such as doors, walls, decorative components, etc.) in complex building scenes, step S10 is performed to accurately separate the target objects to be analyzed from the complex building scenes through target detection technology, which can avoid the problems of false detection or missed detection caused by environmental interference in traditional methods, and solve the limitation that the target objects are difficult to distinguish from the background (such as furniture and decorations) in complex scenes. At the same time, it also prevents the problem of the model being unable to focus on key detection areas due to information overload, thereby providing high-confidence candidate areas for target defect detection, reducing redundant data processing, and laying the foundation for subsequent classification and detailed analysis.

[0031] Exemplarily, a pre-trained YOLO model is used to perform corresponding target detection on the input building image to be identified, or an adaptive threshold segmentation algorithm can be used to preprocess the input image to enhance the target edge features. The detection accuracy of building components such as doors and window frames in complex decorative scenes is improved by introducing an attention mechanism module. The non-maximum suppression algorithm is used to remove duplicate detection frames and then output the pixel-level located target object boundary box to ensure the effective separation of the target object and background interference objects (such as furniture and decorations).

[0032] Step S20, performing defect detection on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem; It should be noted that the defect problem of the target object refers to the abnormal state of the target object in terms of form, function or installation specifications, such as missing door stops (functional defects), cracks in the wall (morphological defects), and excessive installation gaps between the door frame and the wall of the door and window project (installation defects), etc.; the defect problem category refers to the type divided according to the cause of the defect and the scope of impact, such as a single defect problem, that is, an independent problem that only affects a single object, such as a damaged door handle; a compound defect problem, that is, a defect caused by the coordinated action of multiple objects, such as missing door stops, tilted door frames and deformed door leaves that together lead to poor closure, etc.; a cascade defect problem, that is, a complex defect that requires multi-level detection to locate the main cause, such as water leakage may be caused by pipe damage or sealing failure, which requires secondary verification.

[0033] It is understandable that since traditional image recognition technology often confuses complex defect problem types and leads to model misjudgment, step S20 is performed to introduce a dynamic defect category discrimination mechanism and design differentiated detection strategies for different defect types. This overcomes the model misjudgment problem caused by failure to distinguish defect levels, prevents a single detection logic from being unable to handle multi-factor coupled defects or excessive data redundancy leading to limitations on recognition accuracy, thereby significantly improving the matching degree between the detection logic and the actual scenario.

[0034] For example, this can be accomplished through the collaborative work of multiple models: for single defect problems, a pre-trained ResNet-50 model is used to perform recognition tasks such as surface damage recognition; for complex defect problems, multiple target objects are extracted, such as the geometric features of door frames and door leaves (such as gap width and edge alignment) for multi-dimensional feature fusion analysis; for cascade defect problems, a motion trajectory simulation unit based on a physics engine is called to verify the correlation between, for example, the missing door stopper and the deformation of the door frame, and the defect type and category labels are synchronously output through a multi-task learning framework to improve detection efficiency.

[0035] In a feasible implementation, step S20 may include steps S21 to S23: Step S21, obtaining the location data of the building image to be identified, and performing a relevance search on the target object according to the location data; It should be noted that location data refers to the specific shooting location of the building image to be identified, such as the entrance, kitchen, etc.; association retrieval is based on the spatial topological relationship (such as adjacent, contact, nested) or functional dependency (such as the existence of door stops depends on the door body) of the location data, and retrieves other components associated with the target object from the building database or image (such as the door at the entrance and the door stop are related objects).

[0036] It is understandable that since some defect types (especially missing components) are usually difficult to effectively identify based on traditional identification methods, step S21 is performed to obtain the position data of the target object to determine whether it has a physical or functional relationship with the surrounding building components, thereby avoiding the problem of missed structural defects caused by isolated detection, and solving the problem that traditional methods cannot effectively identify complex defects due to ignoring the spatial relationship between the target object and the surrounding components, thereby improving the systematic nature of defect detection and ensuring that the synergy of related objects is included in the analysis scope.

[0037] Exemplarily, the EXIF ​​information of the building image (such as GPS coordinates, camera focal length) is parsed to obtain the spatial coordinates of the target object in the three-dimensional model, and a topological relationship index is established based on a spatial database (such as PostGIS). The R-tree algorithm is used to retrieve related objects (such as door frames and door leaves, doors and door stoppers) that are less than a certain value away from the target object or have a physical connection relationship. At the same time, a logical check is performed in combination with the functional dependency rule library (such as "door stoppers must be installed on the side of the door body") to exclude decorative non-structural components (such as decorative patterns on door handles). Finally, a list of related objects and a spatial relationship map are output to ensure that the correlation of physical coupling defects such as the lack of door stoppers on the entrance doors is accurately captured.

[0038] Step S22, when the search result of the association search is that the target object has an associated object, feature comparison is performed on the building image to be identified based on feature information of the associated object to obtain defect problems of each second sub-target object, and the defect problem category of the defect problem is determined to be a composite defect problem, wherein each second sub-target object includes a target sub-object associated with the target object and a corresponding associated object; It should be noted that associated objects refer to other building components that have physical connections, functional dependencies or spatial constraints with the target object; the second sub-target object refers to the detection unit composed of the target object and its associated objects (such as the "door + door stopper" combination), which is used for collaborative analysis of complex defects.

[0039] It can be understood that when there are associated objects with the target object, it is necessary to analyze the matching degree between the two through feature comparison in order to identify the composite defects caused by synergy. Therefore, performing step S22 can avoid the problem of only detecting a single object and ignoring the combined defects of multiple objects, thereby accurately determining the defect status of the composite defect.

[0040] Exemplarily, high-precision modeling is performed on the target object (such as a door leaf) and its associated objects (such as a door frame), the geometric features of the door leaf edge curvature and the door frame installation flatness are extracted, and the feature vector is input into a pre-trained neural network for similarity matching. If the gap width between the door leaf and the door frame deviates too much from the theoretical value or the edge misalignment angle is too large, the composite defect judgment logic is triggered.

[0041] Step S23, when the retrieval result of the association retrieval is that the target object has no associated object, defect detection is performed on the target object to obtain the defect problem of the target object, and the defect problem category of the defect problem is determined as a single defect problem or a cascade defect problem.

[0042] It is understandable that if the target object has no associated objects, its independent defects (single defects) are directly detected or the potential main causes (cascade defects) are verified through multi-level detection, so step S23 is performed to avoid misjudgment caused by excessive association, thereby achieving efficient and accurate defect classification for isolated objects, reducing redundant calculations and improving classification efficiency.

[0043] Exemplarily, if the detection result is that there is no associated object for the target door handle, a corresponding defect detection is performed on the door handle to obtain the defect problem corresponding to the door handle, and the single defect / cascade defect judgment logic is triggered.

[0044] In this implementation, the physical / functional association between the target object and the associated objects is constructed through multi-source data fusion, and combined with dynamic judgment logic, the problems of missed detection of complex defects, misjudgment of cascade defects and redundant calculation caused by isolated detection in traditional methods are avoided, and systematic defect cause analysis and accurate classification are achieved, which significantly improves the accuracy of defect detection in complex building scenarios and the value of engineering repair guidance.

[0045] Step S30: generating a defect recognition result of the target object based on the defect problem and the corresponding defect problem category.

[0046] It should be noted that the defect identification results may include structured data such as defect type, location, category, and repair suggestions, such as an image report and text description that annotates the deformed area of ​​the door frame: "The door frame is tilted 3° to the left (compound defect, coordinates [X,Y]-[X',Y']), it is recommended to adjust the hinge."

[0047] It is understandable that, since traditional defect recognition methods lack the ability to characterize complex defects, step S30 is performed to generate the final recognition result using a divide-and-conquer strategy based on the classification results. This can solve the problem of inaccurate positioning (such as failure to merge associated areas) or misdescription (such as failure to reflect the causal relationship between multiple defects) caused by ignoring the differences in defect types when integrating results. Based on the type-adaptive integration results, it can support the output of a test report that has both spatial positioning accuracy and semantic integrity (such as marking the synergistic impact areas of compound defects), greatly improving the interpretability of the test results and the engineering guidance value.

[0048] For example, a rectangular annotation box and text description with confidence score are directly output for a single defect, a polygon annotation containing related areas and a cause analysis report are generated for a composite defect, and a hierarchical diagnosis suggestion is formed by superimposing secondary inspection results on cascade defects, and finally a standardized defect distribution map and repair priority list are generated.

[0049] In a feasible implementation manner, the defect problem category includes at least one of a single defect problem, a compound defect problem, and a cascade defect problem, wherein the single defect problem is a defect problem that exists independently of any sub-target object, the compound defect problem is a defect problem formed by a combination of multiple sub-target objects, and the cascade defect problem is a defect problem of a target defect sub-problem to be determined among multiple defect sub-problems; Step S30 may include steps S31 to S34: Step S31, when the defect problem category is a single defect problem, generating a defect identification result of a first sub-target object associated with the defect problem based on the defect problem; It should be noted that the first sub-target object refers to an independent target object directly associated with a single defect problem (such as the surface layer of the wall where the wall crack is located), and the defect analysis can be completed without relying on other components.

[0050] It can be understood that since a single defect exists independently, a defect identification result can be directly generated for an independent object, so performing step S31 can avoid the problem of incorrectly associating a simple defect with other non-existent factors, thereby quickly locating and outputting independent defects, shortening the detection time and computing resource consumption of simple defects.

[0051] Exemplarily, in a single defect scenario, the corresponding defect recognition result is generated directly based on the defect problem, such as using the U-Net segmentation model to extract the morphological features of the wall crack (i.e., the first sub-target object). For example, when the length is >10cm and the width is >2mm, the judgment is triggered, and the defect recognition result including the crack judgment result and the crack position (marked with pixel coordinates) is directly output.

[0052] Step S32, when the defect problem category is a composite defect problem, generating defect identification results for each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem; It is understandable that composite defects are caused by the coordinated action of multiple related objects, and the defect characteristics of all related objects need to be analyzed simultaneously. Therefore, performing step S32 can avoid the problem of collaborative defects (the door still cannot be closed) left behind by analyzing only a single component (such as adjusting the door leaf but ignoring the tilt of the door frame), thereby ensuring that composite defects can be accurately identified and handled.

[0053] In a feasible implementation manner, the step of generating defect identification results of each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem in step S32 may include steps S321 to S323: Step S321, extracting features of each second sub-target object associated with the composite defect problem to obtain defect features of each second sub-target object; It is understandable that since the coupling of multiple defects often leads to feature confusion, performing step S321 can avoid the problem of failure to distinguish the independent features of each sub-object in the composite defect, resulting in a decrease in feature extraction accuracy under cross-interference. By separating and extracting the key features of each sub-object, the separation and resolvability of independent features in the composite defect can be improved.

[0054] Step S322, comparing each defect feature with a reference defect feature to obtain composite defect information corresponding to the composite defect problem and a region where the composite defect information is located; It should be noted that the reference defect feature refers to a pre-established standardized defect feature database, which contains historical detection data or typical defect parameters derived from theoretical models, and serves as a comparison benchmark to achieve defect classification and quantitative analysis of correlation; composite defect information refers to the structured data of the interaction relationship between multi-sub-object defects obtained through feature comparison; the regional location of the composite defect information refers to the spatial range of the synergistic effect of multiple sub-objects in the composite defect, which can be digitally calibrated through a coordinate system or topological relationship.

[0055] It is understandable that in order to establish an objective comparison mechanism based on the standardized feature library to eliminate the risk of subjective misjudgment in the composite defect correlation analysis, step S322 is performed. Through algorithmic comparison, composite defect information containing information such as defect correlation strength and action direction can be generated, which can avoid misjudgment of associated defects and accurately locate the synergistic effect area of ​​composite defects.

[0056] Step S323: generating defect recognition results for each second sub-target object based on the composite defect information and the corresponding area position.

[0057] It can be understood that in order to map the global correlation analysis results of the composite defect to the independent identification strategy of each sub-object, step S323 is performed, which can avoid the problem of ignoring the interaction relationship between the sub-objects in the composite defect, thereby generating an associated defect identification result based on the composite defect information and the regional position, ensuring that the result can accurately characterize the corresponding composite defect.

[0058] In this embodiment, by constructing a hierarchical analysis mechanism for composite defects, feature decoupling extraction technology is used to separate the independent defect features of each second sub-target object, and an algorithmic comparison is performed based on a standardized reference defect feature library, thereby avoiding the cross-interference caused by undecoupled composite defect features, misjudgment of association relationships caused by reliance on subjective experience, and ignoring the synergistic effects of isolated objects in traditional methods. This achieves quantitative analysis of the correlation strength of composite defects and precise positioning of the defect action area of ​​multiple sub-objects, thereby ultimately improving the recognition accuracy of composite defects.

[0059] Step S33, when the defect problem category is a cascade defect problem, performing secondary defect detection on a third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object; It should be noted that the third sub-target object refers to the object in the cascade defect where shallow defects are detected (such as a faucet with appearance defects, but it is impossible to know whether the defect is caused by installation defects or surface rust). More detailed secondary inspections are required to determine the specific defect conditions.

[0060] It is understandable that, since there is a primary and secondary causal relationship between cascading defects, it is necessary to further trace the specific causes, so step S33 is performed to avoid misjudging the cascading defects as single defects, resulting in the lack of engineering guidance value in the obtained defect identification results, thereby achieving accurate identification of defects at a deeper level, and then supporting the output of a radical building repair plan.

[0061] In a feasible implementation manner, the step of performing secondary defect detection on the third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object in step S33 may include steps S331 to S333: Step S331, calling a corresponding sub-goal detection model according to the third sub-goal object associated with the cascade defect problem; It should be noted that the sub-target detection model refers to a special detection model that is pre-trained or dynamically constructed for the defect characteristics of a specific sub-object, and it achieves accurate identification and causal analysis of the defects of the sub-object by integrating multi-source data and domain knowledge.

[0062] It is understandable that since it is often difficult for one recognition model to support deeper and more detailed analysis and recognition of a large number of different objects, step S331 is performed to locate the source of the fundamental defect through a dedicated detection model. This can solve the problem of missed detection or misjudgment of deep defects caused by the use of a unified detection model in traditional methods, and can effectively reduce the performance configuration requirements of the preliminary detection model, thereby achieving the effect of improving both recognition accuracy and model application costs.

[0063] For example, after initially detecting the appearance defects on the surface of an object, the pre-trained sub-target detection model library based on the convolutional neural network is used to dynamically load the adapted model parameters according to the material type (such as metal, plastic) and surface characteristics (such as reflectivity and roughness) of the third sub-target object. For example, for metal surfaces with high reflectivity, the detection model trained by fusion multispectral imaging data is called, and the surface image is input into the model after illumination normalization preprocessing, and the local feature map for color uniformity and oxidation spots is output to adapt to the differences in light sensitivity of different materials and ensure the model's ability to capture subtle color differences.

[0064] Step S332, extracting defect features of the third sub-target object through the sub-target detection model, and comparing the defect features with reference defect features to obtain refined defect information corresponding to the cascade defect problem and the regional position where the refined defect information is located; It should be noted that refining defect information refers to extracting more specific defect information of the sub-target object through cascade defect analysis. For example, after initially identifying that there are defects in the appearance of an object, the sub-target detection model corresponding to the object is called to perform deeper color judgment, texture analysis and other processing on the surface of the object to obtain the data.

[0065] It can be understood that in order to obtain more detailed defect problem information of the target object, step S332 is performed. In-depth feature extraction and comparison are performed through the sub-target detection model adapted to the sub-target object. This can avoid the problem that deep features are difficult to capture through a general model, thereby effectively improving the recognition accuracy of complex defect problems.

[0066] Exemplarily, the loaded sub-target detection model is used to perform multi-scale feature extraction on the surface image, the color difference distribution is quantified through color space conversion, and the local texture variation coefficient is calculated in combination with the texture analysis algorithm. Then, the extracted color difference gradient matrix and texture variation heat map are compared pixel by pixel with the standard sample data of the same material in the reference defect feature library, and a dynamic threshold segmentation algorithm (such as the adaptive Otsu algorithm) is used to determine the abnormal area, and refined defect information including color difference deviation values ​​and texture fracture paths is generated. The defect area is located on the surface of the object through coordinate mapping technology, and specific labels are marked.

[0067] Step S333: generating a defect recognition result of the third sub-target object based on the refined defect information and the corresponding area position.

[0068] For example, according to the color difference deviation value and texture fracture length in the refined defect information, combined with the spatial distribution of the regional position, the preset defect pattern library is matched to output the defect type, severity level and repair priority. At the same time, a structured report containing defect quantification parameters and spatial positioning is formed based on the coordinates of the regional position.

[0069] In this embodiment, a dynamic hierarchical detection framework for cascading defects is constructed, an adaptive model calling mechanism based on sub-object attribute features is adopted, and multi-dimensional feature analysis technology is combined to decouple and quantitatively characterize the defect features of the target object. A dynamic feature comparison algorithm and spatial mapping technology are used to generate structured recognition results containing defect information and defect locations. This avoids the omission of deep defects due to insufficient generalization and adaptation of detection models, misjudgment of cascade relationships due to feature coupling, and repair strategy problems caused by positioning ambiguity in traditional methods, thereby achieving accurate analysis and identification of cascade defect problems.

[0070] Step S34, integrating the existing defect recognition results of the first sub-target object and / or the second sub-target object and / or the third sub-target object to obtain the defect recognition result of the target object.

[0071] It should be noted that if only one or two of the first sub-target object, the second sub-target object or the third sub-target object exist, the defect recognition results of these one or two sub-target objects will be integrated; if all three sub-target objects exist, the defect recognition results of these three sub-target objects will be integrated.

[0072] Exemplarily, the detection results of the first sub-target object (wall cracks), the second sub-target object (door frame-door leaf) and the third sub-target object (faucet) are sorted by priority (cascade>composite>single) through the defect association map, and high-priority detection results are output preferentially in the same area.

[0073] In this embodiment, by constructing a classification detection and hierarchical integration mechanism, an independent analysis module is used to directly generate the defect identification result of the first sub-target object for a single defect problem, thereby avoiding the redundant detection, logical confusion and insufficient pertinence of the defect identification scheme caused by the traditional method due to the failure to distinguish the defect type; a multi-object collaborative analysis mechanism is introduced for the compound defect problem, and the defect characteristics of each second sub-target object are synchronously decoupled and associated, thereby solving the omission of related defects caused by isolated identification; a deep-level traceability detection mechanism is introduced for the cascade defect problem, and the specific defects of the third sub-target object are located through secondary defect detection, thereby overcoming the problem that the model needs to process too much information, resulting in excessive load on the detection model and the inability to effectively identify the contents of each sub-item in detail, and effectively improving the high accuracy of defect identification in scenarios with large amounts of data; finally, the detection results of the three types of sub-objects are integrated to achieve global coordination of the defect governance sequence and resource allocation, thereby achieving the effect of improving the accuracy of defect identification and reducing the consumption of detection resources.

[0074] This embodiment provides a method for identifying building defects, which effectively distinguishes the categories of defect problems, thereby performing targeted defect identification for different categories of defect problems, avoiding the problem that traditional defect identification methods are unable to effectively identify defects in different complex scenes due to the complexity of building scenes, and achieving accurate and efficient identification of building defects in various complex scenes, thereby improving the efficiency and accuracy of building quality inspection.

[0075] In a feasible implementation manner, step S23 may further include steps S231 to S233: Step S231, performing defect detection on the target object, obtaining defect problems of the target object, and performing detailed analysis on the defect problems; It is understandable that since traditional defect detection methods usually only identify surface anomalies and cannot reveal defect problems at a deeper level, resulting in a lack of specificity in defect classification, performing step S231 can avoid defect feature coupling caused by insufficient detection granularity, such as multiple sub-problems being misjudged as a single problem, or omission of key sub-problems, such as failure to discover deeper defect problems, thereby providing high-precision data support for subsequent classification based on a finer-grained analysis of the defect problems.

[0076] Exemplarily, a multi-task detection model based on deep learning is used to identify surface defects of a target object, and an attempt is made to decompose the surface defects through a feature decoupling algorithm to determine whether the surface defect has multiple more detailed defect sub-problems.

[0077] Step S232, when the detailed analysis result shows that the defect problem has multiple defect sub-problems, the defect problem category of the defect problem is determined to be a cascade defect problem; It should be noted that the refined analysis results refer to the defect structured data generated by multi-dimensional detection and feature decoupling technology, which contains independent / cascade attribute criteria to guide defect classification.

[0078] It is understandable that since the determination of cascade defects requires that there are multiple independent and parallel sub-category defects under the main defect category, performing step S232 can avoid the problem that defect identification cannot be specific to sub-category defects due to failure to identify the independent attributes of parallel sub-category defects under the same category, thereby ensuring the accuracy of subsequent defect identification.

[0079] Step S233: when the detailed analysis result shows that the defect problem does not have a defect sub-problem, the defect problem category of the defect problem is determined to be a single defect problem.

[0080] It can be understood that since the independence and integrity of a single defect need to be confirmed by verifying that there are no divisible parallel sub-category defects, performing step S233 can avoid defect identification anomalies caused by incorrect segmentation of the main defect, thereby clarifying the attributes of a single defect and saving ineffective consumption of computing resources.

[0081] In this embodiment, by conducting a more detailed analysis of the defect problem to evaluate whether the defect is divisible, the omission or misjudgment of parallel sub-category defects under the main defect category due to insufficient detection granularity is effectively avoided, thereby achieving accurate defect classification.

[0082] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 In step S34, the defect recognition results of the first sub-target object, the second sub-target object and the third sub-target object include each defect information and the area position corresponding to any defect information. Step S34 may also include steps S341 to S343: Step S341, annotating the image of the building to be identified based on the position of each area to obtain an annotated defect detection image; Step S342, performing region spacing detection on each marked region in the defect detection image to obtain a region spacing detection result of the defect detection image; It should be noted that the region spacing detection result indicates the spacing between the labeled regions, including the minimum distance value between the labeled regions calculated on the two-dimensional image and its distribution characteristics, which are used to determine whether the defect annotations can be merged.

[0083] It is understandable that after the defect conditions are marked, there may be densely packed feature borders in the same area, such as concrete cracks, exposed reinforcement, concrete honeycombs, etc. Therefore, step S342 is performed to perform spacing detection on each marked area, thereby providing an effective merging basis for the subsequent fusion of the marked areas.

[0084] Step S343 , merging the marked areas according to the area spacing detection result to obtain an adjusted defect detection image, and using the adjusted defect detection image and the corresponding defect information as the defect recognition result of the target object.

[0085] It is understandable that when there are too densely labeled areas in the same area, it will directly affect the user's knowledge of the specific defect conditions, so step S343 is performed to effectively avoid the complexity of the labeled areas by merging the too densely labeled areas, while also ensuring that the labeled data is not lost, thereby providing users with comprehensive and efficient defect identification result feedback.

[0086] In this embodiment, by effectively integrating and merging densely defect-annotated areas, it is effectively avoided that a large number of defects overlap in the same area, which is not conducive to the display of defect identification results, thereby affecting the guiding effect of the results on practical engineering applications. The final data is reliably and effectively integrated, providing users with comprehensive and efficient defect identification result feedback.

[0087] For example, to help understand the implementation process of the building defect identification method obtained by combining this embodiment with the above-mentioned embodiment 1, please refer to Figure 3 , Figure 3 A brief flow chart of a building defect identification method is provided, specifically: The image of the building to be identified is input into the first-level model detection (which can be a pre-trained YOLO model) to detect the target of the input image through the model, and the defect problem category is determined by combining the business definition with the target detection result to determine whether the defect problem of the target object is a valid problem. If it is valid, it is a single problem, a compound problem, or a cascade problem. If it is a single problem, the annotation boundaries (i.e., the annotation area) are directly merged in the subsequent process; if it is a compound problem, the annotation boundaries are merged after further logical processing; if it is a cascade problem, the input image is cropped based on the corresponding target object (i.e., the third sub-target object), and the cropped small image is input into the second-level model (i.e., the sub-target detection model) for cascade detection (i.e., secondary defect detection), and then the annotation boundaries are merged based on the detection results. Finally, the merged defect detection image and the corresponding defect information are output as the defect recognition result of the target object.

[0088] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the building defect identification method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0089] This application also provides a building defect identification device, please refer to Figure 4 , the building defect identification device comprises: The object detection module 10 is used to obtain a building image to be identified, and perform target detection on the building image to be identified to obtain a target object in the building image to be identified; A problem classification module 20 is used to perform defect detection on the target object, determine the defect problem of the target object and the defect problem category corresponding to the defect problem; The result generating module 30 is used to generate the defect identification result of the target object based on the defect problem and the corresponding defect problem category.

[0090] Optionally, the defect problem category includes at least one of a single defect problem, a compound defect problem and a cascade defect problem, wherein the single defect problem is a defect problem that exists independently of any sub-target object, the compound defect problem is a defect problem formed by a combination of multiple sub-target objects, and the cascade defect problem is a defect problem of a target defect sub-problem to be determined among multiple defect sub-problems; The result generation module 30 is also used for: In a case where the defect problem category is a single defect problem, generating a defect identification result of a first sub-target object associated with the defect problem based on the defect problem; In the case where the defect problem category is a composite defect problem, generating defect identification results for each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem; In the case where the defect problem category is a cascade defect problem, performing secondary defect detection on a third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object; The existing defect recognition results of the first sub-target object and / or the second sub-target object and / or the third sub-target object are integrated to obtain the defect recognition result of the target object.

[0091] Optionally, the result generating module 30 is further used for: Extracting features of each second sub-target object associated with the composite defect problem to obtain defect features of each second sub-target object; Comparing each defect feature with a reference defect feature to obtain composite defect information corresponding to the composite defect problem and a region where the composite defect information is located; A defect recognition result of each second sub-target object is generated based on the composite defect information and the corresponding area position.

[0092] Optionally, the result generating module 30 is further used for: Calling a corresponding sub-goal detection model according to a third sub-goal object associated with the cascade defect problem; Extracting defect features of the third sub-target object through the sub-target detection model, and comparing the defect features with reference defect features to obtain refined defect information corresponding to the cascade defect problem and the regional location of the refined defect information; A defect recognition result of the third sub-target object is generated based on the refined defect information and the corresponding area position.

[0093] Optionally, the defect recognition results of the first sub-target object, the second sub-target object and the third sub-target object include each defect information and a region position corresponding to any defect information, and the result generation module 30 is further used to: Annotating the image of the building to be identified based on the position of each area to obtain an annotated defect detection image; Performing region spacing detection on each marked region in the defect detection image to obtain a region spacing detection result of the defect detection image; The marked areas are merged according to the area spacing detection result to obtain an adjusted defect detection image, and the adjusted defect detection image and the corresponding defect information are used as the defect recognition result of the target object.

[0094] Optionally, the question classification module 20 is further used for: Acquire the location data of the building image to be identified, and perform a relevance search on the target object according to the location data; When the search result of the association search is that the target object has an associated object, feature comparison is performed on the building image to be identified based on feature information of the associated object to obtain defect problems of each second sub-target object, and the defect problem category of the defect problem is determined to be a composite defect problem, wherein each second sub-target object includes a target sub-object associated with the target object and a corresponding associated object; When the retrieval result of the association retrieval is that the target object has no associated object, defect detection is performed on the target object to obtain the defect problem of the target object, and the defect problem category of the defect problem is determined as a single defect problem or a cascade defect problem.

[0095] Optionally, the question classification module 20 is further used for: Performing defect detection on the target object to obtain defect problems of the target object, and performing detailed analysis on the defect problems; When the detailed analysis result shows that the defect problem has multiple defect sub-problems, the defect problem category of the defect problem is determined to be a cascade defect problem; When the detailed analysis result shows that the defect problem does not have any defect sub-problems, the defect problem category of the defect problem is determined to be a single defect problem.

[0096] The building defect recognition device provided by the present application adopts the building defect recognition method in the above embodiment, which can solve the technical problem of how to improve the defect recognition accuracy of image recognition technology when facing complex building scenes. Compared with the prior art, the beneficial effects of the building defect recognition device provided by the present application are the same as the beneficial effects of the building defect recognition method provided by the above embodiment, and the other technical features in the building defect recognition device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0097] The present application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the building defect identification method in the above-mentioned embodiment 1.

[0098] Reference below Figure 5 , which shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present application. The electronic devices in the embodiments of the present application may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions: tablet computers), PMPs (Portable Media Players: portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0099] like Figure 5As shown, the electronic device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which may perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 to a random access memory 1004. In the random access memory 1004, various programs and data required for the operation of the electronic device are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other via a bus 1005. An input / output interface 1006 is also connected to the bus. Generally, the following systems may be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although the electronic device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have instead.

[0100] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0101] The electronic device provided by the present application adopts the building defect recognition method in the above embodiment, which can solve the technical problem of how to improve the defect recognition accuracy of image recognition technology when facing complex building scenes. Compared with the prior art, the beneficial effects of the electronic device provided by the present application are the same as the beneficial effects of the building defect recognition method provided by the above embodiment, and the other technical features in the electronic device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0102] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0103] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0104] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the building defect identification method in the above-mentioned embodiment.

[0105] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0106] The computer-readable storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0107] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by an electronic device, the electronic device: obtains a building image to be identified, and performs target detection on the building image to be identified to obtain a target object in the building image to be identified; performs defect detection on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem; and generates a defect identification result of the target object based on the defect problem and the corresponding defect problem category.

[0108] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0109] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0110] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0111] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned building defect recognition method, and can solve the technical problem of how to improve the defect recognition accuracy of image recognition technology when facing complex building scenes. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the building defect recognition method provided in the above-mentioned embodiment, and will not be repeated here.

[0112] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned building defect identification method when executed by a processor.

[0113] The computer program product provided by the present application can solve the technical problem of how to improve the accuracy of defect recognition of image recognition technology when facing complex building scenes. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the building defect recognition method provided by the above embodiment, which will not be repeated here.

[0114] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for identifying building defects, characterized in that: The building defect identification method comprises: Acquire a building image to be identified, and perform target detection on the building image to be identified to obtain a target object in the building image to be identified; Performing defect detection on the target object to determine the defect problem of the target object and the defect problem category corresponding to the defect problem; Based on the defect problem and the corresponding defect problem category, a defect identification result of the target object is generated.

2. The building defect identification method according to claim 1, characterized in that: The defect problem category includes at least one of a single defect problem, a compound defect problem and a cascade defect problem, wherein the single defect problem is a defect problem that exists independently of any sub-target object, the compound defect problem is a defect problem formed by a combination of multiple sub-target objects, and the cascade defect problem is a defect problem of a target defect sub-problem to be determined among multiple defect sub-problems; The step of generating a defect identification result of the target object based on the defect problem and the corresponding defect problem category comprises: In a case where the defect problem category is a single defect problem, generating a defect identification result of a first sub-target object associated with the defect problem based on the defect problem; In the case where the defect problem category is a composite defect problem, generating defect identification results for each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem; In the case where the defect problem category is a cascade defect problem, performing secondary defect detection on a third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object; The existing defect recognition results of the first sub-target object and / or the second sub-target object and / or the third sub-target object are integrated to obtain the defect recognition result of the target object.

3. The building defect identification method according to claim 2, characterized in that: The step of generating defect identification results of each second sub-target object based on the defect problems of each second sub-target object associated with the composite defect problem comprises: Extracting features of each second sub-target object associated with the composite defect problem to obtain defect features of each second sub-target object; Comparing each defect feature with a reference defect feature to obtain composite defect information corresponding to the composite defect problem and a region where the composite defect information is located; A defect recognition result of each second sub-target object is generated based on the composite defect information and the corresponding area position.

4. The building defect identification method according to claim 2, characterized in that: The step of performing secondary defect detection on the third sub-target object associated with the cascade defect problem to obtain a defect identification result of the third sub-target object includes: Calling a corresponding sub-goal detection model according to a third sub-goal object associated with the cascade defect problem; Extracting defect features of the third sub-target object through the sub-target detection model, and comparing the defect features with reference defect features to obtain refined defect information corresponding to the cascade defect problem and the regional location of the refined defect information; A defect recognition result of the third sub-target object is generated based on the refined defect information and the corresponding area position.

5. The building defect identification method according to claim 2, characterized in that: The defect recognition results of the first sub-target object, the second sub-target object and the third sub-target object include each defect information and the area position corresponding to any defect information. The step of integrating the defect recognition results of the first sub-target object, the second sub-target object and the third sub-target object to obtain the defect recognition result of the target object includes: Annotating the image of the building to be identified based on the position of each area to obtain an annotated defect detection image; Performing region spacing detection on each marked region in the defect detection image to obtain a region spacing detection result of the defect detection image; The marked areas are merged according to the area spacing detection result to obtain an adjusted defect detection image, and the adjusted defect detection image and the corresponding defect information are used as the defect recognition result of the target object.

6. The building defect identification method according to claim 2, characterized in that: The step of performing defect detection on the target object and determining the defect problem of the target object and the defect problem category corresponding to the defect problem comprises: Acquire the location data of the building image to be identified, and perform a relevance search on the target object according to the location data; When the search result of the association search is that the target object has an associated object, feature comparison is performed on the building image to be identified based on feature information of the associated object to obtain defect problems of each second sub-target object, and the defect problem category of the defect problem is determined to be a composite defect problem, wherein each second sub-target object includes a target sub-object associated with the target object and a corresponding associated object; When the retrieval result of the association retrieval is that the target object has no associated object, defect detection is performed on the target object to obtain the defect problem of the target object, and the defect problem category of the defect problem is determined as a single defect problem or a cascade defect problem.

7. The building defect identification method according to claim 6, characterized in that: The step of performing defect detection on the target object to obtain the defect problem of the target object, and determining the defect problem category of the defect problem as a single defect problem or a cascade defect problem comprises: Performing defect detection on the target object to obtain defect problems of the target object, and performing detailed analysis on the defect problems; When the detailed analysis result shows that the defect problem has multiple defect sub-problems, the defect problem category of the defect problem is determined to be a cascade defect problem; When the detailed analysis result shows that the defect problem does not have any defect sub-problems, the defect problem category of the defect problem is determined to be a single defect problem.

8. A building defect identification device, characterized in that: The building defect identification device comprises: An object detection module is used to obtain a building image to be identified, and perform target detection on the building image to be identified to obtain a target object in the building image to be identified; A problem classification module, used to perform defect detection on the target object, determine the defect problem of the target object and the defect problem category corresponding to the defect problem; The result generation module is used to generate a defect identification result of the target object based on the defect problem and the corresponding defect problem category.

9. An electronic device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the building defect identification method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the building defect identification method according to any one of claims 1 to 7 are implemented.

Images