Artificial intelligence-based concealed engineering quality inspection method and device

Abstract

The application discloses a concealed engineering quality inspection method and device based on artificial intelligence, and the method comprises the following steps: receiving input concealed engineering images and concealed initial detection information, then obtaining matched target model features from a building virtual model, extracting image features and combining the image features with the concealed initial detection information and the target model features, inputting an intelligent prediction model, obtaining a structure deformation prediction result, receiving concealed change images and concealed new detection information, obtaining corresponding image comparison features and target detection information, and performing consistency comparison and verification on the structure deformation prediction information and the image features to obtain a quality verification result. In the building construction process, the method can obtain the structure deformation prediction result based on the intelligent prediction model, and perform consistency comparison and verification on the subsequently obtained concealed change images and concealed new detection information, so that the quality of the concealed engineering can be quickly and reliably inspected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent analysis, in particular to a concealed engineering quality inspection method and device based on artificial intelligence. BACKGROUND

[0002] The quality of building engineering is detected by experienced professional engineers through appearance observation, measurement and other methods. Appearance observation is the first step of quality detection and an important detection, especially for concealed engineering such as steel reinforcement engineering, the spacing of steel bars and other parameters have a great influence on the stress of the formed structure. However, after the concrete is covered, it is not possible to detect the steel bar discharge situation again, and only destructive sampling detection can be used. However, destructive re-inspection in the prior art will inevitably affect the construction period and cause economic losses. At the same time, destructive re-inspection may also damage the good quality and need to be re-constructed, and it is also possible that the defective area is not found, which has a large randomness. Therefore, the prior art method has the problem of being unable to quickly detect the quality of concealed engineering. SUMMARY

[0003] In order to overcome the deficiencies of the prior art scheme, the embodiments of the present application provide a concealed engineering quality inspection method and device based on artificial intelligence, which aims to solve the problem that the prior art method cannot quickly detect the quality of concealed engineering.

[0004] In order to solve the technical problems existing in the prior art, in a first aspect, the embodiments of the present application provide a concealed engineering quality inspection method based on artificial intelligence, which comprises:

[0005] If the input concealed engineering image and concealed initial detection information are received, the matching target model features are obtained from the pre-set building virtual model;

[0006] According to the pre-set structure feature extraction rule, the image features corresponding to the concealed engineering image are extracted;

[0007] The image features, the concealed initial detection information and the target model features are input into the pre-set intelligent prediction model to predict the corresponding structure deformation prediction information;

[0008] If the input concealed change image and concealed new detection information are received, the target image and target detection information matching the structure deformation prediction information are obtained;

[0009] According to the structure feature extraction rule, the image comparison features corresponding to the target image are extracted;

[0010] The target detection information and the image contrast feature are compared with the structure deformation prediction information according to preset comparison rules to obtain a corresponding quality check result.

[0011] In a second aspect, the embodiment of the present application further provides a concealed engineering quality inspection device based on artificial intelligence, which comprises:

[0012] A target model feature acquisition unit is configured to acquire a target model feature matched from a preset building virtual model if the input concealed engineering image and concealed initial detection information are received.

[0013] An image feature acquisition unit is configured to extract an image feature corresponding to the concealed engineering image according to preset structure feature extraction rules.

[0014] A prediction information acquisition unit is configured to input the image feature, the concealed initial detection information and the target model feature into a preset intelligent prediction model to predict corresponding structure deformation prediction information.

[0015] A matching unit is configured to acquire a target image and target detection information matched with the structure deformation prediction information if the input concealed change image and concealed new detection information are received.

[0016] A feature extraction unit is configured to extract an image contrast feature corresponding to the target image according to the structure feature extraction rules.

[0017] A quality check result acquisition unit is configured to compare the target detection information and the image contrast feature with the structure deformation prediction information according to preset comparison rules to obtain a corresponding quality check result.

[0018] In a third aspect, the embodiment of the present application further provides a computer device, which comprises a processor, a network interface, a memory and a communication bus, wherein the processor, the network interface and the memory complete mutual communication through the communication bus.

[0019] The memory is configured to store a computer program.

[0020] The processor is configured to execute the program stored on the memory to realize the steps of the concealed engineering quality inspection method based on artificial intelligence.

[0021] In a fourth aspect, the embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by the processor to realize the steps of the concealed engineering quality inspection method based on artificial intelligence.

[0022] The embodiment of the present application discloses a concealed engineering quality inspection method and device based on artificial intelligence, which comprises the following steps: receiving input concealed engineering images and concealed initial detection information, obtaining matching target model features from a building virtual model, extracting image features and combining them with the concealed initial detection information and the target model features, inputting an intelligent prediction model, obtaining a structure deformation prediction result, receiving concealed change images and concealed new detection information, obtaining corresponding image comparison features and target detection information, and performing consistency comparison and verification with the structure deformation prediction information and the image features to obtain a quality verification result. The above method can obtain a structure deformation prediction result based on an intelligent prediction model during the construction of a building, and perform consistency comparison and verification with the subsequently obtained concealed change images and concealed new detection information, so as to realize rapid and reliable inspection of the quality of concealed engineering, without the need for destructive sampling detection, thereby greatly improving the inspection efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0024] Figure 1 is a method flowchart of the concealed engineering quality inspection method based on artificial intelligence provided by the embodiment of the present application;

[0025] Figure 2 is a schematic block diagram of the concealed engineering quality inspection device based on artificial intelligence provided by the embodiment of the present application;

[0026] Figure 3 is a schematic block diagram of the computer device of the embodiment of the present application. DETAILED DESCRIPTION

[0027] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0028] It should be understood that when used in the specification and the appended claims, the terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or sets thereof.

[0029] It should also be understood that the terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and the appended claims of the present application, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0030] It should be further understood that the term "and / or" used in the specification and the appended claims of the present application means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0031] The embodiments of the present application provide an artificial intelligence-based concealed engineering quality inspection method, which is applied to a terminal device or a server. The terminal device or the server executes a stored software program to realize the artificial intelligence-based concealed engineering quality inspection method.

[0032] As shown in Figure 1 The method includes steps S110-S160.

[0033] S110, if the input concealed engineering image and concealed initial detection information are received, a matching target model feature is obtained from a pre-set building virtual model.

[0034] The user (engineering management personnel) can collect and input the concealed engineering image at the building construction site, and can also input the concealed initial detection information. The concealed initial detection information is the numerical information detected by the sensor arranged on the physical building structure. The sensor can be used for stress detection of the concealment in the building, so as to obtain corresponding numerical information. The stress detection includes stress generated by concealed gravity extrusion, tensile force generated by concealed stretching, etc. Therefore, the obtained concealed initial detection information contains detection values corresponding to multiple stress detection types.

[0035] Among them, concealed engineering includes steel bar concealed engineering, pre-embedded part concealed engineering, pipeline concealed engineering, etc. Steel bar concealed engineering refers to the part of the steel bar engineering that is covered or wrapped by subsequent processes such as concrete pouring and wall building, and cannot be directly observed and inspected. Concealed engineering mainly involves the configuration, connection and installation of steel bars in concrete structures; its quality directly affects the safety, durability and stability of the building structure, so strict acceptance procedures are needed to ensure compliance with design and specification requirements. Concealed engineering is the engineering that embeds building materials, components or equipment inside the structure (such as foundation, wall, floor) during construction, which is covered by the outer decoration or structure. For example: structural type: foundation steel, concrete components, internal structure of masonry; pipeline type: water and electricity pipeline, heating and ventilation pipeline, network wiring; waterproof and thermal insulation type: roof waterproof layer, thermal insulation layer, basement damp-proof layer. Its technical features include: irreversibility: difficult to inspect directly after covering, need to complete acceptance before concealment; high risk: quality problems can easily cause safety hazards (such as pipeline leakage, structure cracking).

[0036] According to industry standards, concealed engineering includes the following seven categories: foundation engineering, foundation soil treatment, foundation steel bar binding, concrete pouring, pile foundation construction, etc.; structural engineering, steel bar specifications, spacing, anchoring length in beams, columns and plates; steel structure welding joints; pipeline engineering, water supply and drainage pipeline materials and connections, electrical cable laying (strong / weak current), heating and ventilation pipeline insulation; waterproof engineering, roof waterproof material lap joint, basement damp-proof layer, bathroom closed water test; base engineering, suspended ceiling furring, floor base, wall plaster layer, door and window sleeve base; equipment installation, equipment foundation pre-embedded part, pipeline wall penetration sealing treatment, ventilation pipeline concealed part; special engineering, tunnel anchor rod, underwater riprap, geological drilling (requires special monitoring).

[0037] The key requirements for quality control of concealed engineering include: acceptance procedures must involve multiple parties: after self-inspection by the construction unit, the supervision, construction and design units jointly accept and sign to confirm; time limit requirement: written acceptance application must be submitted 48 hours before concealment, and if the supervisor does not arrive on time, it can be considered as default qualified (evidence must be kept). Technical verification method: on-site inspection: measure size, check material certificate, sample test (such as steel bar tension test). Image recording: record important parts (such as pile foundation, waterproof layer) throughout the process and archive. Sampling and retesting: more than 30% of the self-inspection results are retested, especially for easy shortcuts (such as insulation layer thickness). Common problems and risks: using inferior materials: such as replacing low-grade steel bars, reducing waterproof layer thickness; process violation: direct burial of electrical wires without sleeve, unsealed pipeline joints; procedure omission: covering without acceptance, leading to disputes with no evidence.

[0038] Hidden works are the physical parts that need to be covered and buried in building engineering, including foundation, structure, pipeline and other engineering contents that directly affect safety and function. Its core feature is invisible after being hidden, so it must pass through strict acceptance procedures, multi-party participation on-site verification and image recording to ensure quality, and avoid quality problems caused by cutting corners or program missing. In actual engineering, hidden works management is the "life and death line" of quality control.

[0039] According to the input hidden engineering image and hidden initial detection information, the target model feature matching the hidden engineering image and the hidden initial detection information is obtained from the virtual building model. Specifically, the virtual building model is a virtual model (Building Information Modeling, BIM) generated according to the design scheme of the building engineering. The virtual building model includes internal steel structure and external concrete structure. Since the quality inspection of hidden engineering only involves part of the area of building engineering, it is not necessary to analyze the complete virtual building model as a whole, and only the features of the local building area corresponding to the hidden engineering image and the hidden initial detection information need to be obtained.

[0040] In specific embodiments, step S110 includes the following sub-steps: obtaining a component in the building virtual model corresponding to the shooting position of the hidden engineering image as a first component according to the shooting position of the hidden engineering image; obtaining a component in the building virtual model associated with the detection point in the hidden initial detection information as a second component according to the detection point in the hidden initial detection information; combining the first component and the second component as a target component; and obtaining a model parameter in the building virtual model matching the target component as a corresponding target model feature.

[0041] Specifically, the hidden engineering image can be mapped to the virtual building model according to the shooting position of the hidden engineering image. Then, the virtual building model is viewed based on the shooting position, so that the part of the component corresponding to the shooting position can be determined. According to the mapping method, the component in the virtual building model corresponding to the shooting position can be obtained as the first component. If multiple hidden engineering images are included, the hidden engineering images are mapped respectively to obtain the components corresponding to the hidden engineering images, and after deduplication, the components are obtained as the corresponding first components.

[0042] Further, in addition to distinguishing the stress detection types in the concealed initial detection information, the detection points corresponding to the detection values are also included, that is, the specific position information of the sensors, and each detection value corresponds to a detection point and a stress detection type. The sensor placed at each detection point is in direct contact with at least two components, and the component in contact with the sensor corresponding to the detection point is the component associated with the detection point, that is, each detection point is associated with at least two components in the virtual building model; according to the detection points in the concealed initial detection information, the components associated with each detection point in the virtual building model are obtained as the second components.

[0043] The first component and the second component obtained by the above steps are combined as the target component, and the target component includes a complete building structure corresponding to the area requiring engineering quality analysis. In addition to including steel bars, the target component also includes components such as concrete. Further, the model parameters in the virtual building model that match the target component are obtained as the target model parameters corresponding to the concealed engineering image and the concealed initial detection information. Each component in the virtual building model corresponds to the corresponding model parameters, including component size parameters, component position parameters, density parameters, material type parameters, etc.

[0044] S120, according to the pre-configured structure feature extraction rule, the image features corresponding to the concealed engineering image are extracted.

[0045] The corresponding image features can be extracted from the concealed engineering image by the pre-configured structure feature extraction rule. The concealed engineering image is an image of the building engineering obtained after laying steel structure materials such as steel bars and completing the pouring of concrete, and the obtained image features can be used to reflect the external building component features of the concealed engineering.

[0046] In specific embodiments, step S120 includes a sub-step: processing the concealed engineering image according to the image processing parameters in the structure feature extraction rule to obtain a corresponding contour image; and extracting the corresponding image parameters from the contour image according to the parameter extraction item in the structure feature extraction rule as the image features corresponding to the concealed engineering image.

[0047] Specifically, the hidden engineering image can be processed according to image processing parameters in the structural feature extraction rule, and the image processing parameters include a dissolution ratio and a contour limitation threshold. The pixel contrast of each pixel point in the hidden engineering image can be calculated first, and the pixel contrast of the pixel point is the difference information of the pixel point compared with surrounding pixel points. The greater the difference between the pixel value of the pixel point and the pixel value of the surrounding pixel points, the greater the pixel contrast. After the pixel contrast of each pixel point in the hidden engineering image is calculated, the pixel points are sorted in descending order of the pixel contrast, and a part of the pixel points at the top of the sorting are further extracted from the sorted pixel points according to the dissolution ratio parameter. The pixel points are filled according to the positions of the extracted pixel points to obtain an initial contour image, and the contours in the initial contour image are further screened according to the contour limitation threshold. Specifically, adjacent pixel points in the initial contour image can be combined to form a corresponding contour, and non-adjacent pixel points cannot be combined. Whether the number of pixels contained in the contour and the contour length exceed the contour limitation threshold is determined, and the contour whose number of pixels and contour length exceed the contour limitation threshold is screened from the contours and combined into a corresponding contour image.

[0048] The parameter extraction item configured in the structural feature extraction rule can extract corresponding image parameters from the contour image through the parameter extraction item, and the obtained image parameters are used as image features corresponding to the hidden engineering image. The gap between two adjacent parallel contours with a large distance in the contour image is the building external contour (such as the roof beam contour); and the parameter extraction item can be set to include the mean value of the distance between the two adjacent parallel contours with a large distance, the mean value of the straightness tolerance value of the contour, and other feature information.

[0049] S130, input the image features, the hidden initial detection information and the target model features into a preset intelligent prediction model to obtain corresponding structural deformation prediction information.

[0050] The obtained image features, hidden initial detection information, and target model features are input into a pre-configured intelligent prediction model, and the input information is associated and analyzed by the intelligent prediction model to obtain structure deformation prediction information after a certain time. The intelligent prediction model is a neural network model constructed based on artificial intelligence. The intelligent prediction model includes an input layer, an associated analysis layer, and an output layer. The input layer includes multiple input nodes. The associated analysis layer includes one or more groups of associated nodes. One group of associated nodes is arranged vertically, and multiple groups of associated nodes are arranged in a multi-column distribution form. The associated nodes establish corresponding associated formulas with the input nodes, the output nodes, or other associated nodes in an adjacent group of associated nodes. The associated formulas are configured with corresponding coefficient values. The associated formulas can establish an association between two nodes. The output nodes in the output layer are dynamically configured according to the hidden initial detection information. The output layer includes multiple output nodes. Some of the output nodes correspond to output deformation prediction values, and the other output nodes correspond to output detection prediction values. Each detection prediction value corresponds to an output node, and the number of output nodes outputting detection prediction values is equal to the number of detection values in the hidden initial detection information. The structure deformation prediction information can be obtained by obtaining the prediction values output by the output nodes. The structure deformation prediction information corresponds to one or more prediction time parameters.

[0051] Before using the intelligent prediction model, the intelligent prediction model can also be trained by multiple groups of training data. Each group of training data includes input features and output features. The input features are input into the intelligent prediction model. The input features also include time parameters, which are the time difference between the time when the output features are obtained and the time when the input features are obtained. The prediction information is obtained by comparing the corresponding prediction information and the output features. The prediction information is the predicted feature obtained based on the input features corresponding to the changes predicted according to the time parameters. The prediction information reflects the predicted features of the input features after a certain time. The loss value between the prediction information and the output features is used to train the intelligent prediction model to adjust the parameters in the model. The intelligent prediction model is iteratively trained by multiple groups of training parameters, and the trained intelligent prediction model is finally used.

[0052] In specific embodiments, before step S130, the method further includes the step of configuring corresponding prediction time parameters in the intelligent prediction model according to pre-set prediction time information.

[0053] Further, before prediction by the intelligent prediction model, the prediction time parameter in the model can also be configured according to the preset prediction time information. One or more time values can be set in the prediction time information, such as setting the time value in the prediction time information as 7 days, 14 days, one month, three months, etc. The number of input nodes configured in the intelligent prediction model for inputting the prediction time parameter is equal to the number of time values set in the prediction time information.

[0054] S140, if the input hidden change image and hidden new detection information are received, the target image and target detection information matching the structure deformation prediction information are obtained.

[0055] The user (engineering management personnel) can perform subsequent monitoring on the construction project, take local images of the construction project to obtain the input hidden change image, and continuously monitor the construction project by the sensor to obtain the input hidden new detection information. Then, after receiving the input information, the target image and target detection information matching the structure deformation prediction information can be obtained.

[0056] In specific embodiments, step S140 includes the following sub-steps: obtaining the image time difference between the collection time of each image in the hidden change image and the input time of the hidden engineering image; obtaining the detection time difference between the detection time of each detection information in the hidden new detection information and the input time of the hidden detection information; obtaining the image matching the prediction time parameter as the target image according to the prediction time parameter of the structure deformation prediction information; and obtaining the detection information matching the prediction time parameter as the target detection information.

[0057] The images in the hidden change image may not all meet the corresponding detection requirements. Therefore, the image time difference between the collection time of each image in the hidden change image and the input time of the hidden engineering image can be obtained, and each image in the hidden change image corresponds to an obtainable image time difference. Further, the detection time difference between the detection time of each detection information in the hidden new detection information and the input time of the hidden detection information is obtained.

[0058] The images in the hidden change image are screened by the image time difference. Specifically, the image matching the prediction time parameter as the target image can be obtained according to the image time difference; and the detection information matching the prediction time parameter as the target detection information can also be obtained according to the detection time difference.

[0059] For example, the image time differences of three images in the hidden change image are 3 days, 7 days and 9 days respectively, and the prediction time parameter is 7 days. Then, the image with the image time difference of 7 days can be obtained as the target image.

[0060] S150, extracting image contrast features corresponding to the target image according to the structural feature extraction rule.

[0061] The image contrast features corresponding to the target image are extracted according to the structural feature extraction rule. The parameter extraction item for extracting the image contrast features is different from the parameter extraction item for extracting the image features. The parameter extraction item can be set as the average value of the flatness, the external convex deformation variable, the external contour straightness tolerance value, and the like. The image contrast features can be used to inspect the deformation of the construction project.

[0062] S160, performing consistency comparison and verification of the target detection information, the image contrast features, and the structural deformation prediction information according to a preset comparison and verification rule to obtain a corresponding quality verification result.

[0063] The consistency comparison and verification of the target detection information, the image contrast features, and the structural deformation prediction information are performed according to the comparison and verification rule, that is, the consistency of the target detection information, the image contrast features, and the structural deformation prediction information is verified, so as to obtain the quality verification result. If the consistency of the target detection information, the image contrast features, and the structural deformation prediction information satisfies the comparison and verification rule, the quality verification result of passing the verification is obtained. If the consistency does not satisfy the comparison and verification rule, the quality verification result of failing the verification is obtained. The structural deformation prediction information includes prediction information corresponding to each prediction time parameter, so the deformation of the construction project at different prediction time parameters can be inspected. The specific process of performing the comparison and verification of the prediction information of multiple prediction time parameters is similar to the process of performing the comparison and verification of the prediction information of one prediction time parameter. The subsequent steps of the present application focus on the comparison and verification of the prediction information of one prediction time parameter.

[0064] In specific embodiments, the step S160 includes a sub-step of judging whether each prediction value in the structural deformation prediction information satisfies a value detection condition in the comparison and verification rule. If the structural deformation prediction information satisfies the value detection condition, the value difference between each prediction value in the structural deformation prediction information and the value corresponding to the target detection information and the image contrast features is obtained. It is judged whether the value difference of each prediction value is located in a corresponding value verification interval in the comparison and verification rule. If the value difference of each prediction value is located in the corresponding value verification interval, the quality verification result of passing the verification is obtained. If the image feature difference information does not satisfy the image difference detection condition, or the value difference of any prediction value is not located in the corresponding value verification interval, the quality verification result of failing the verification is obtained.

[0065] Specifically, it can be judged whether each prediction value in the structure deformation prediction information satisfies a value detection condition in the comparison verification rule. The value detection condition can be set to correspond to each prediction value. That is, it can be judged whether each prediction value is located in the corresponding detection interval. If each prediction value is located in the corresponding detection interval, it is determined that the value detection condition is satisfied. Otherwise, it is determined that the value detection condition is not satisfied.

[0066] If it is determined that the value detection condition is satisfied, a value difference corresponding to each prediction value in the structure deformation prediction information and the target detection information and the image contrast feature is further obtained. Specifically, a value difference between the deformation prediction value in the structure deformation prediction information and the image contrast feature can be obtained, and a value difference between the detection prediction value in the structure deformation prediction information and the target detection information can be obtained. Each prediction value can correspond to a value difference.

[0067] The comparison verification rule can be set to correspond to each prediction value. It is further judged whether each value difference is located in the corresponding value verification interval. If the value difference of each prediction value is located in the corresponding value verification interval, a quality verification result of passing the verification is obtained. If the image feature difference information does not satisfy the image difference detection condition, or the value difference of any prediction value is not located in the corresponding value verification interval, a quality verification result of failing the verification is obtained. The quality verification result can correspond to one or more prediction results corresponding to the prediction time parameter.

[0068] In specific embodiments, after step S160, the method further includes the step of: judging whether the target detection information, the image contrast feature and the structure deformation prediction information satisfy a preset recheck condition; and if the recheck condition is satisfied, generating recheck prompt information corresponding to the structure deformation prediction information.

[0069] Further, after obtaining the quality verification result, it can be judged whether the target detection information, the image contrast feature and the structure deformation prediction information meet the re-inspection condition. The re-inspection condition is set with the building specification parameter range corresponding to each prediction value. The value difference between each prediction value in the structure deformation prediction information and the value corresponding to the target detection information and the image contrast feature can be obtained, and it is judged whether the value difference of each prediction value is located in the corresponding building specification parameter range. If the value difference is located in the corresponding building specification parameter range, it is determined that the re-inspection condition is not met. At this time, although the construction project may have quality defects, it still meets the basic building specification. If any value difference is not located in the corresponding building specification parameter range, it is determined that the re-inspection condition is met. At this time, the construction project does not meet the basic building specification, and needs to be re-inspected by destructive sampling. At this time, the re-inspection prompt information corresponding to the structure deformation prediction information can be generated. The re-inspection prompt information includes the deformation position corresponding to the structure deformation prediction information. Therefore, the re-inspection prompt information can prompt the user (engineering management personnel) to re-inspect the corresponding deformation position in the construction project.

[0070] The above-mentioned embodiment discloses a concealed engineering quality inspection method and device based on artificial intelligence. The method comprises: receiving an input concealed engineering image and concealed initial detection information, obtaining a matching target model feature from a building virtual model, extracting image features and combining the image features with the concealed initial detection information and the target model feature and inputting an intelligent prediction model to obtain a structure deformation prediction result, receiving a concealed change image and concealed new detection information and obtaining corresponding image contrast features and target detection information, and performing consistency comparison and verification with the structure deformation prediction information and the image features to obtain a quality verification result. The above-mentioned method can obtain a structure deformation prediction result based on an intelligent prediction model and perform consistency comparison and verification with a subsequent concealed change image and concealed new detection information, so as to realize rapid and reliable inspection of the quality of the concealed engineering, without the need for destructive sampling detection, thereby greatly improving the inspection efficiency.

[0071] The embodiment of the present application also provides a concealed engineering quality inspection device based on artificial intelligence. The concealed engineering quality inspection device based on artificial intelligence can be configured in a terminal device or a server. The concealed engineering quality inspection device based on artificial intelligence is used to execute any embodiment of the above-mentioned concealed engineering quality inspection method based on artificial intelligence. Specifically, please refer to Figure 2 , Figure 2 The schematic block diagram of the concealed engineering quality inspection device based on artificial intelligence provided by the embodiment of the present application is shown in the figure.

[0072] As Figure 2As shown, the artificial intelligence-based concealed engineering quality inspection device 100 comprises a target model feature acquisition unit 110, an image feature acquisition unit 120, a prediction information acquisition unit 130, a matching unit 140, a feature extraction unit 150, and a quality check result acquisition unit 160.

[0073] The target model feature acquisition unit 110 is configured to acquire a matching target model feature from a preset building virtual model if the input concealed engineering image and concealed initial detection information are received.

[0074] The image feature acquisition unit 120 is configured to extract an image feature corresponding to the concealed engineering image according to a preset structure feature extraction rule.

[0075] The prediction information acquisition unit 130 is configured to input the image feature, the concealed initial detection information, and the target model feature into a preset intelligent prediction model to predict a corresponding structure deformation prediction information.

[0076] The matching unit 140 is configured to acquire a target image and target detection information matching the structure deformation prediction information if the input concealed change image and concealed new detection information are received.

[0077] The feature extraction unit 150 is configured to extract an image comparison feature corresponding to the target image according to the structure feature extraction rule.

[0078] The quality check result acquisition unit 160 is configured to perform consistency comparison and verification on the target detection information, the image comparison feature, and the structure deformation prediction information according to a preset comparison and verification rule to obtain a corresponding quality check result.

[0079] The artificial intelligence-based concealed engineering quality inspection device provided in the embodiment of the present application applies the above-described artificial intelligence-based concealed engineering quality inspection method. The input concealed engineering image and concealed initial detection information are received, and then a matching target model feature is acquired from a building virtual model. The image feature is extracted and combined with the concealed initial detection information and the target model feature and input into an intelligent prediction model to obtain a structure deformation prediction result. The concealed change image and concealed new detection information are received, and corresponding image comparison features and target detection information are acquired and compared with the structure deformation prediction information and the image feature to obtain a quality check result. The above-described method can acquire a structure deformation prediction result based on an intelligent prediction model and compare it with the subsequently acquired concealed change image and concealed new detection information for consistency comparison and verification during the building construction process, thereby realizing rapid and reliable inspection of the quality of the concealed engineering without the need for destructive sampling detection, and thus greatly improving the inspection efficiency.

[0080] The above-mentioned artificial intelligence-based concealed engineering quality inspection device can be realized in the form of a computer program, which can run on a computer device as shown in the drawings. Figure 3

[0081] Please refer to Figure 3 , Figure 3 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device can be a terminal device or a server for executing an artificial intelligence-based concealed engineering quality inspection method to inspect the quality of concealed engineering.

[0082] Referring to Figure 3 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a communication bus 501, wherein the memory can include a storage medium 503 and an internal memory 504.

[0083] The storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032, when executed, can cause the processor 502 to execute an artificial intelligence-based concealed engineering quality inspection method, wherein the storage medium 503 can be a volatile storage medium or a non-volatile storage medium.

[0084] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0085] The internal memory 504 provides an environment for the execution of the computer program 5032 in the storage medium 503, and the computer program 5032, when executed by the processor 502, can cause the processor 502 to execute an artificial intelligence-based concealed engineering quality inspection method.

[0086] The network interface 505 is configured to perform network communication, such as providing transmission of data information, etc. Those skilled in the art can understand that Figure 3 the structure shown in the drawings is only a block diagram of part of the structure related to the present application scheme, and does not constitute a limitation on the computer device 500 to which the present application scheme is applied. The specific computer device 500 can include more or fewer components than those shown in the drawings, or combine certain components, or have a different component arrangement.

[0087] The processor 502 is configured to run the computer program 5032 stored in the memory to realize the corresponding functions in the above-mentioned artificial intelligence-based concealed engineering quality inspection method.

[0088] Those skilled in the art can understand Figure 3 ​The embodiments of the computer device shown in the figures do not constitute a limitation on the specific structure of the computer device, and in other embodiments, the computer device can include more or fewer components than shown, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device can only include the memory and the processor, and in such embodiments, the structure and function of the memory and the processor are consistent with the embodiments shown. Figure 3 The embodiments shown are consistent with the structure and function of the memory and the processor, and are not described again here.

[0089] It should be understood that in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and the processor 502 can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0090] In another embodiment of the present application, a computer readable storage medium is provided. The computer readable storage medium can be a volatile or non-volatile computer readable storage medium. The computer readable storage medium stores a computer program, wherein the computer program is executed by a processor to implement the steps contained in the above-mentioned artificial intelligence-based concealed engineering quality inspection method.

[0091] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, apparatuses and units can refer to the corresponding processes in the foregoing method embodiments, which are not described again here. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in a general manner. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0092] In several embodiments provided by the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, or a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, devices or units, and can be electric, mechanical or in other forms.

[0093] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0094] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0095] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part of the prior art that contributes to the technical solutions, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a computer readable storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned computer readable storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various media that can store program codes.

[0096] The above description 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 can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for inspecting the quality of concealed engineering works based on artificial intelligence, characterized in that, The method includes: If the input hidden project image and initial hidden detection information are received, the matching target model features are obtained from the preset building virtual model; Image features corresponding to the hidden engineering image are extracted according to preset structural feature extraction rules; The image features, the concealed initial detection information, and the target model features are input into a preset intelligent prediction model to predict the corresponding structural deformation prediction information. If the input hidden change image and hidden new detection information are received, the target image and target detection information that match the structural deformation prediction information are obtained; Based on the structural feature extraction rules, image contrast features corresponding to the target image are extracted; The target detection information and the image contrast features are compared and verified with the structural deformation prediction information according to the preset comparison and verification rules to obtain the corresponding quality verification results. The acquisition of the target image and target detection information that match the structural deformation prediction information includes: Obtain the image time difference between the acquisition time of each image in the concealed change image and the input time of the concealed engineering image; Obtain the detection time difference between the detection time of each detection information in the newly added hidden detection information and the input time of the hidden engineering image; Based on the prediction time parameters of the structural deformation prediction information, an image whose image time difference matches the prediction time parameters is obtained as the target image; The detection information that matches the detection time difference with the predicted time parameter is used as the target detection information.

2. The method for inspecting the quality of concealed works based on artificial intelligence according to claim 1, characterized in that, The step of obtaining matching target model features from a pre-set virtual building model includes: Based on the shooting location of the concealed engineering image, the component that maps to the shooting location in the building virtual model is obtained as the first component; Based on the detection points in the concealed initial detection information, the components associated with the detection points in the building virtual model are obtained as the second components; Combine the first component and the second component to form the target component; The model parameters that match the target component in the building virtual model are obtained as the corresponding target model features.

3. The method for inspecting the quality of concealed works based on artificial intelligence according to claim 1, characterized in that, The step of extracting image features corresponding to the hidden engineering image according to preset structural feature extraction rules includes: The hidden engineering image is processed according to the image processing parameters in the structural feature extraction rules to obtain the corresponding contour image; According to the parameter extraction item in the structural feature extraction rule, the corresponding image parameters are extracted from the contour image as image features corresponding to the hidden engineering image.

4. The method for inspecting the quality of concealed works based on artificial intelligence according to claim 1, characterized in that, Before inputting the image features, the initial concealment detection information, and the target model features into a preset intelligent prediction model to predict the corresponding structural deformation prediction information, the method further includes: Configure the corresponding prediction time parameters in the intelligent prediction model based on the preset prediction time information.

5. The method for inspecting the quality of concealed works based on artificial intelligence according to claim 1, characterized in that, The step of performing a consistency comparison and verification between the target detection information and the image contrast features and the structural deformation prediction information according to preset comparison and verification rules, and obtaining the corresponding quality verification result, includes: Determine whether each predicted value in the structural deformation prediction information meets the numerical detection conditions in the comparison and verification rules; If the structural deformation prediction information satisfies the numerical detection condition, obtain the numerical difference between each predicted value in the structural deformation prediction information and the numerical difference between the target detection information and the image comparison features; Determine whether the numerical difference between each predicted value is within the corresponding numerical verification interval in the comparison test rule; If the numerical difference of each predicted value is within the corresponding numerical verification interval, a quality verification result that passes the verification is obtained. If the image feature difference information does not meet the image difference detection conditions, or if the numerical difference of any of the predicted values ​​is not within the corresponding numerical verification interval, a quality verification result of failure is obtained.

6. The method for inspecting the quality of concealed works based on artificial intelligence according to claim 1 or 5, characterized in that, After performing a consistency comparison and verification between the target detection information and the image contrast features and the structural deformation prediction information according to preset comparison and verification rules, and obtaining the corresponding quality verification result, the method further includes: Determine whether the target detection information, the image contrast features, and the structural deformation prediction information meet the preset verification conditions; If the re-verification conditions are met, a re-verification prompt message corresponding to the structural deformation prediction information is generated.

7. A concealed engineering quality inspection device based on artificial intelligence, characterized in that, The device includes: The target model feature acquisition unit is used to acquire matching target model features from a preset building virtual model if it receives the input hidden project image and initial hidden detection information. The image feature acquisition unit is used to extract image features corresponding to the hidden project image according to preset structural feature extraction rules; The prediction information acquisition unit is used to input the image features, the hidden initial detection information and the target model features into a preset intelligent prediction model to predict the corresponding structural deformation prediction information. The matching unit is used to obtain a target image and target detection information that match the structural deformation prediction information if it receives the input hidden change image and hidden new detection information. The feature extraction unit is used to extract image comparison features corresponding to the target image according to the structural feature extraction rules; The quality verification result acquisition unit is used to perform consistency comparison verification between the target detection information and the image comparison features and the structural deformation prediction information according to the preset comparison verification rules, and obtain the corresponding quality verification result. The acquisition of the target image and target detection information that match the structural deformation prediction information includes: Obtain the image time difference between the acquisition time of each image in the concealed change image and the input time of the concealed engineering image; Obtain the detection time difference between the detection time of each detection information in the newly added hidden detection information and the input time of the hidden engineering image; Based on the prediction time parameters of the structural deformation prediction information, an image whose image time difference matches the prediction time parameters is obtained as the target image; The detection information that matches the detection time difference with the predicted time parameter is used as the target detection information.

8. A computer device, characterized in that, The device includes a processor, a network interface, a memory, and a communication bus, wherein the processor, network interface, and memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in a memory, it implements the steps of the hidden engineering quality inspection method based on artificial intelligence as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the concealed engineering quality inspection method based on artificial intelligence as described in any one of claims 1-6.

Images