Fabricated building outer wall stress detection method and device

Abstract

The invention discloses a fabricated building exterior wall stress detection method and device, and the method comprises the steps: obtaining a building wall exterior facade image collected by an unmanned plane in real time, and obtaining an exterior facade grayscale image; identifying a target image for the external facade grayscale image, and determining the space coordinates of a target mark; performing adaptive weighted median filtering processing on the target image to obtain a filtered noise reduction image; carrying out binarization processing on the noise reduction image by adopting a second derivative edge operator to obtain each edge line of the target image; analyzing the shape data of each edge line of the obtained single target image, and comparing and marking each edge line; comparing the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain crack defect types corresponding to the shape features consistent with the shape data; and generating an outer wall stress detection report according to the space coordinates of the target mark and the crack defect type.

Description

Technical Field

[0001] The present invention relates to the technical field of building wall stress detection, and in particular to a method and device for detecting stress of an assembled building exterior wall. Background Art Prefabricated building walls are mostly composed of prefabricated reinforced concrete shear walls, which are composed of prefabricated wall molds and cast-in-place concrete of different strength grades and ages. Due to concrete shrinkage and creep, complex local structure, loose vibration and construction technology, hidden damage such as interface peeling and internal voids are prone to occur. Therefore, traditional methods mostly use tapping method, impact sound transmission method and ultrasonic concrete tomography for measurement.

[0002] However, existing prefabricated walls are mostly characterized by high floors, large areas and wide ranges. Traditional methods require multiple and repeated manual operations, which have low accuracy and low measurement efficiency, and are time-consuming and labor-intensive. Summary of the invention

[0003] The object of the present invention is to provide a method and device for detecting stress on the exterior wall of an assembled building, so as to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solution: a method for detecting stress on the exterior wall of an assembled building, comprising the following contents: S1. Obtaining a building wall facade image collected by a drone in real time, and graying the building wall facade image to obtain a facade grayscale image; S2, identifying whether the facade grayscale image contains a target image, and when it is identified that the target image is contained, determining the spatial coordinates of a target mark, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; S3, performing adaptive weighted median filtering on the target image to obtain a filtered denoised image; S4, using a second-order derivative edge operator to perform binarization processing on the denoised image to obtain edge lines of the target image; S5, analyzing the acquired shape data of each edge line of the target image, and performing comparative marking on each edge line; S6, comparing the shape data with shape features of known crack defect types in a pre-stored crack defect library, and obtaining a crack defect type corresponding to the shape feature consistent with the shape data; S7. Generate an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

[0005] Preferably, the step S1 further includes the following steps: Acquire multiple crack images and crack defect types that match the crack images one-to-one entered by a user, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape features of the crack defect type include a pre-stored crack depth, a pre-stored crack length, a pre-stored crack width and a pre-stored crack edge inclination angle that are matched one-to-one with the crack image.

[0006] Preferably, the step of analyzing the acquired shape data of each edge line of the target image and comparing and marking each edge line includes: Divide the single target image into multiple sub-regions, and calculate binary LBP values ​​for pixels of each sub-region based on a template of a fixed size, the binary LBP values ​​being Local Binary Patterns; Calculating an LBP histogram of each sub-region according to the binary LBP value, and connecting the LBP histograms of the sub-regions to form a global LBP feature; The global LBP features are stored as the shape data, and are annotated through data display to achieve visualization.

[0007] Preferably, the step of comparing the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain the crack defect type corresponding to the shape features consistent with the shape data comprises: The shape data of the target image includes the real-time crack length, real-time crack width and real-time crack edge inclination angle of the target image; Compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle respectively; The shape feature with the greatest similarity is used as a specified shape feature consistent with the shape data, and the crack defect type corresponding to the specified shape feature is obtained.

[0008] Preferably, the step of generating an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type comprises: Marking the spatial coordinates of the target mark as the actual position of the crack marks on the scanned building wall facade; The crack defect type includes the predicted crack depth of the crack mark; An external wall stress detection report including the actual position and the crack depth is generated according to the actual position of the crack mark on the exterior wall of the building and the crack depth of the crack mark.

[0009] On the other hand, a device for detecting stress on the exterior wall of an assembled building is provided, comprising the method for detecting stress on the exterior wall of an assembled building as described above, wherein the detection device comprises: The first acquisition module is used to acquire the building wall facade image collected by the drone in real time, and grayscale the building wall facade image to obtain the facade grayscale image, wherein the drone is connected to a remote monitoring terminal via a controller; a first analysis module, connected to the first acquisition module, for identifying whether the facade grayscale image contains a target image, and determining the spatial coordinates of a target mark when the target image is identified to be contained, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; A denoising module, connected to the first analysis module, for performing an adaptive weighted median filtering process on the target image to obtain a filtered denoised image; A first calculation module, connected to the denoising module, is used to perform binarization processing on the denoised image using a second-order derivative edge operator to obtain edge lines of the target image; a second analysis module, connected to the first calculation module, for analyzing the shape data of each edge line of the target image obtained, and for comparing and marking each edge line, wherein the shape data of the target image includes a real-time crack length, a real-time crack width, and a real-time crack edge inclination angle of the target image; A second calculation module, connected to the second analysis module, is used to compare the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain a crack defect type corresponding to the shape feature consistent with the shape data; A generation module, connected to the first analysis module and the second calculation module, is used to generate an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

[0010] Preferably, the prefabricated building exterior wall stress detection device further includes a pre-storage module, which is connected to the second calculation module and is used to obtain multiple crack images entered by a user and crack defect types that match the crack images one-to-one, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape features of the crack defect type include a pre-stored crack depth, a pre-stored crack length, a pre-stored crack width and a pre-stored crack edge inclination angle that are matched one-to-one with the crack image.

[0011] Preferably, the second analysis module includes: A first calculation unit, connected to the first calculation module, is used to divide the single target image into a plurality of sub-regions, and calculate a binary LBP value for a pixel point of each sub-region based on a template of a fixed size; a second calculation unit, connected to the first calculation unit, for calculating an LBP histogram of each sub-region according to the binary LBP value, and connecting the LBP histograms of the sub-regions to form a global LBP feature; The labeling unit is connected to the second computing unit and the remote monitoring terminal, stores the global LBP features as the shape data, labels them through data display, and transmits them to the remote monitoring terminal to achieve visualization.

[0012] Preferably, the second calculation module includes: A comparison unit, used to compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle; The third calculation unit is used to take the shape feature with the greatest similarity as the specified shape feature consistent with the shape data, and obtain the crack defect type corresponding to the specified shape feature.

[0013] Preferably, the device for detecting stress on the exterior wall of an assembled building is also provided with an early warning module, which is connected to the generation module and the remote monitoring terminal module, and is used to analyze whether the crack depth corresponding to the actual position of the crack mark in the exterior wall stress detection report exceeds the early warning value. If it exceeds the early warning value, an early warning signal is generated and sent to the remote monitoring terminal.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: by providing a drone equipped with a calculator, the drone accurately obtains the image of the exterior facade of the building wall, thereby replacing the step of manually obtaining the condition of the exterior facade of the wall; further, by obtaining the exterior facade image of the building wall collected by the drone in real time, the crack marks are used as the target image, the spatial coordinates of the target image are obtained, and then the preliminary target image is filtered to confirm the edge features of the actual crack marks by calculation, and then the marks consistent with the currently detected target image are retrieved from a preset crack defect library according to the edge features, so as to confirm the specific depth of the crack marks currently photographed by the drone, and finally, combined with the specific coordinate position and specific depth of the crack marks, an external wall stress detection report matching the current wall facade is quickly generated, and the overall stress detection step is simple and accurate, which also effectively improves the accuracy of stress detection and ensures the efficiency of stress detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the process of the method for detecting stress on the exterior wall of an assembled building of the present invention; Figure 2 It is a schematic structural diagram of the prefabricated building exterior wall stress detection device of the present invention.

[0016] Figure 2 In: 201, first acquisition module; 202, first analysis module; 203, noise reduction module; 204, first calculation module; 205, second analysis module; 2051, first calculation unit; 2052, second calculation unit; 2053, labeling unit; 206, second calculation module; 2061, comparison unit; 2062, third calculation unit; 207, generation module; 208, pre-storage module; 209, early warning module. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Embodiment 1: See also Figure 1 The present invention provides a technical solution: a method for detecting stress on the exterior wall of an assembled building, comprising the following contents: Step S1, obtaining a building wall facade image collected by a drone in real time, and graying the building wall facade image to obtain a facade grayscale image; Before step S1, data preprocessing is required, which includes the following: Acquire multiple crack images and crack defect types that match the crack images one-to-one entered by a user, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape characteristics of the crack defect type include pre-stored crack depth, pre-stored crack length, pre-stored crack width and pre-stored crack edge inclination angle that match the crack image one-to-one; the above data is composed by storing historical crack opening data.

[0019] Step S2, identifying whether the facade grayscale image contains a target image, and when it is identified that the target image is contained, determining the spatial coordinates of a target mark, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; In the above step S2, the spatial coordinates of the target image can clearly indicate the specific position of the crack marks on the outer surface of the wall, so as to accurately analyze the cause of the stress change of the wall and the main position of the change.

[0020] Step S3, performing adaptive weighted median filtering on the target image to obtain a filtered denoised image; Step S4, using a second-order derivative edge operator to perform binarization processing on the denoised image to obtain edge lines of the target image; Step S5, analyzing the acquired shape data of each edge line of the target image, and performing comparative marking on each edge line; In the above step S5, the analysis step specifically includes: Dividing the single target image into a plurality of sub-regions, and calculating a binary LBP value for a pixel point in each sub-region based on a template of a fixed size; Calculating an LBP histogram of each sub-region according to the binary LBP value, and connecting the LBP histograms of the sub-regions to form a global LBP feature; The global LBP features are stored as the shape data, and are annotated through data display to achieve visualization.

[0021] The visualization method in the above steps includes: annotating the image data taken by the drone and transmitting it to the remote monitoring terminal for display, so that the query personnel can intuitively understand the shape data of the target image.

[0022] Step S6, comparing the shape data with shape features of known crack defect types in a pre-stored crack defect library, and obtaining a crack defect type corresponding to the shape feature consistent with the shape data; The above step S6 includes: The shape data of the target image includes the real-time crack length, real-time crack width and real-time crack edge inclination angle of the target image; Compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle respectively; The shape feature with the greatest similarity is used as a specified shape feature consistent with the shape data, and the crack defect type corresponding to the specified shape feature is obtained.

[0023] Step S7: Generate an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

[0024] The above step S7 includes: The step of generating an external wall stress detection report according to the spatial coordinates of the target mark and the crack defect type comprises: Marking the spatial coordinates of the target mark as the actual position of the crack marks on the scanned building wall facade; The crack defect type includes the predicted crack depth of the crack mark; An external wall stress detection report including the actual position and the crack depth is generated according to the actual position of the crack mark on the exterior wall of the building and the crack depth of the crack mark.

[0025] Embodiment 2: See also Figure 2 On the other hand, the present invention provides a device for detecting stress on the exterior wall of an assembled building, including the method for detecting stress on the exterior wall of an assembled building as described above, wherein the detection device includes a first acquisition module 201, a first analysis module 202, a noise reduction module 203, a first calculation module 204, a second analysis module 205, a second calculation module 206, a generation module 207 and an early warning module 209, specifically: The first acquisition module 201 is used to acquire the building wall facade image collected by the drone in real time, and grayscale the building wall facade image to obtain the facade grayscale image, wherein the drone is connected to a remote monitoring terminal via a controller; The first analysis module 202 is connected to the first acquisition module 201, and is used to identify whether the facade grayscale image contains a target image, and when it is identified that the target image is contained, determine the spatial coordinates of a target mark, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; A noise reduction module 203, connected to the first analysis module 202, is used to perform adaptive weighted median filtering on the target image to obtain a filtered noise-reduced image; A first calculation module 204, connected to the denoising module 203, is used to perform binarization processing on the denoised image using a second-order derivative edge operator to obtain edge lines of the target image; A second analysis module 205, connected to the first calculation module 204, is used to analyze the shape data of each edge line of the target image obtained, and compare and mark each edge line, wherein the shape data of the target image includes the real-time crack length, real-time crack width and real-time crack edge inclination angle of the target image; Wherein, the second analysis module 205 includes: A first calculation unit 2051, connected to the first calculation module 204, is used to divide the single target image into multiple sub-regions, and calculate a binary LBP value for a pixel point of each sub-region based on a template of a fixed size; A second calculation unit 2052, connected to the first calculation unit 2051, is used to calculate the LBP histogram of each sub-region according to the binary LBP value, and connect the LBP histograms of the sub-regions to form a global LBP feature; The labeling unit 2053 is connected to the second calculation unit 2052 and the remote monitoring terminal, stores the global LBP feature as the shape data, labels it through data display, and transmits it to the remote monitoring terminal to achieve visualization.

[0026] A second calculation module 206, connected to the second analysis module 205, is used to compare the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain a crack defect type corresponding to the shape feature consistent with the shape data; Wherein, the second calculation module 206 includes: A comparison unit 2061 is used to compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle respectively; The third calculation unit 2062 is used to take the shape feature with the greatest similarity as the specified shape feature consistent with the shape data, and obtain the crack defect type corresponding to the specified shape feature.

[0027] The generating module 207 is connected to the first analyzing module 202 and the second calculating module 206, and is used to generate an external wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

[0028] In addition, the prefabricated building exterior wall stress detection device further includes a pre-storage module 208, which is connected to the second calculation module 206 and is used to obtain multiple crack images entered by a user and crack defect types that match the crack images one-to-one, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape features of the crack defect type include a pre-stored crack depth, a pre-stored crack length, a pre-stored crack width and a pre-stored crack edge inclination angle that are matched one-to-one with the crack image.

[0029] In order to ensure that management personnel can check in time when an early warning appears in the detection report, the prefabricated building exterior wall stress detection device is also provided with an early warning module 209. The early warning module 209 is connected to the generation module 207 and the remote monitoring terminal module, and is used to analyze whether the crack depth corresponding to the actual position of the crack mark in the exterior wall stress detection report exceeds the early warning value. If it exceeds the early warning value, an early warning signal is generated and sent to the remote monitoring terminal.

[0030] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting stress on the exterior wall of an assembled building, characterized in that: The detection method includes the following steps: S1. Obtaining a building wall facade image collected by a drone in real time, and graying the building wall facade image to obtain a facade grayscale image; S2, identifying whether the facade grayscale image contains a target image, and when it is identified that the target image is contained, determining the spatial coordinates of a target mark, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; S3, performing adaptive weighted median filtering on the target image to obtain a filtered denoised image; S4, using a second-order derivative edge operator to perform binarization processing on the denoised image to obtain edge lines of the target image; S5, analyzing the acquired shape data of each edge line of the target image; S6, comparing the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain a crack defect type corresponding to a shape feature consistent with the shape data; S7. Generate an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

2. A method for detecting stress on the exterior wall of an assembled building according to claim 1, characterized in that: Before step S1, the method further includes: Acquire multiple crack images and crack defect types that match the crack images one-to-one entered by a user, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape features of the crack defect type include a pre-stored crack depth, a pre-stored crack length, a pre-stored crack width and a pre-stored crack edge inclination angle that are matched one-to-one with the crack image.

3. A method for detecting stress on the exterior wall of an assembled building according to claim 2, characterized in that: The step of analyzing the acquired shape data of each edge line of the target image comprises: Dividing the single target image into a plurality of sub-regions, and calculating a binary LBP value for a pixel point in each sub-region based on a template of a fixed size; Calculating an LBP histogram of each sub-region according to the binary LBP value, and connecting the LBP histograms of the sub-regions to form a global LBP feature; The global LBP features are stored as the shape data, and are annotated through data display to achieve visualization.

4. A method for detecting stress on the exterior wall of an assembled building according to claim 3, characterized in that: The shape data of the target image includes the real-time crack length, real-time crack width and real-time crack edge inclination angle of the target image; correspondingly, the shape data is compared with the shape features of known crack defect types in a pre-stored crack defect library to obtain the crack defect type corresponding to the shape features consistent with the shape data, including: Compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle, respectively, to obtain a shape feature with the greatest similarity to the shape data of the target image; The shape feature with the greatest similarity is used as a specified shape feature consistent with the shape data, and the crack defect type corresponding to the specified shape feature is obtained.

5. A method for detecting stress on the exterior wall of an assembled building according to claim 4, characterized in that: The step of generating an external wall stress detection report according to the spatial coordinates of the target mark and the crack defect type comprises: Marking the spatial coordinates of the target mark as the actual position of the crack marks on the scanned building wall facade; The crack defect type includes the predicted crack depth of the crack mark; An external wall stress detection report including the actual position and the crack depth is generated according to the actual position of the crack mark on the exterior wall of the building and the crack depth of the crack mark.

6. A prefabricated building exterior wall stress detection device, characterized in that: The method for detecting stress on the exterior wall of an assembled building according to any one of claims 1 to 5, wherein the detection device comprises: The first acquisition module is used to acquire the building wall facade image collected by the drone in real time, and grayscale the building wall facade image to obtain the facade grayscale image, wherein the drone is connected to a remote monitoring terminal via a controller; a first analysis module, connected to the first acquisition module, for identifying whether the facade grayscale image contains a target image, and determining the spatial coordinates of a target mark when the target image is identified to be contained, wherein the target image includes a defective image, and the target mark is a crack mark corresponding to the defective image; A denoising module, connected to the first analysis module, for performing an adaptive weighted median filtering process on the target image to obtain a filtered denoised image; A first calculation module, connected to the denoising module, is used to perform binarization processing on the denoised image using a second-order derivative edge operator to obtain edge lines of the target image; a second analysis module, connected to the first calculation module, for analyzing the shape data of each edge line of the target image obtained, and for comparing and marking each edge line, wherein the shape data of the target image includes a real-time crack length, a real-time crack width, and a real-time crack edge inclination angle of the target image; A second calculation module, connected to the second analysis module, is used to compare the shape data with shape features of known crack defect types in a pre-stored crack defect library to obtain a crack defect type corresponding to the shape feature consistent with the shape data; A generation module, connected to the first analysis module and the second calculation module, is used to generate an exterior wall stress detection report according to the spatial coordinates of the target mark and the crack defect type.

7. The device for detecting stress on the exterior wall of an assembled building according to claim 6, characterized in that: It also includes a pre-storage module, which is connected to the second calculation module and is used to obtain multiple crack images entered by a user and crack defect types that match the crack images one-to-one, and store the multiple crack images and crack defect types that match the crack images one-to-one in a preset crack defect library; The shape features of the crack defect type include a pre-stored crack depth, a pre-stored crack length, a pre-stored crack width and a pre-stored crack edge inclination angle that are matched one-to-one with the crack image.

8. The device for detecting stress on the exterior wall of an assembled building according to claim 7, characterized in that: The second analysis module comprises: A first calculation unit, connected to the first calculation module, is used to divide the single target image into a plurality of sub-regions, and calculate a binary LBP value for a pixel point of each sub-region based on a template of a fixed size; a second calculation unit, connected to the first calculation unit, for calculating an LBP histogram of each sub-region according to the binary LBP value, and connecting the LBP histograms of the sub-regions to form a global LBP feature; The labeling unit is connected to the second computing unit and the remote monitoring terminal, stores the global LBP features as the shape data, labels them through data display, and transmits them to the remote monitoring terminal to achieve visualization.

9. The device for detecting stress on the exterior wall of an assembled building according to claim 8, characterized in that: The second calculation module includes: A comparison unit, used to compare the real-time crack length with the pre-stored crack length, the real-time crack width with the pre-stored crack width, and the real-time crack edge inclination angle with the pre-stored crack edge inclination angle; The third calculation unit is used to take the shape feature with the greatest similarity as the specified shape feature consistent with the shape data, and obtain the crack defect type corresponding to the specified shape feature.

10. The device for detecting stress on the exterior wall of an assembled building according to claim 9, characterized in that: An early warning module is also provided, which is connected to the generation module and the remote monitoring terminal module and is used to analyze whether the crack depth corresponding to the actual position of the crack mark in the external wall stress detection report exceeds the early warning value. If it exceeds the early warning value, an early warning signal is generated and sent to the remote monitoring terminal.

Images