Building engineering crack detection method and system based on image recognition

Abstract

The invention discloses a building engineering crack detection method and system based on image recognition, and relates to the technical field of image processing. The method comprises the following steps: acquiring a constructional engineering surface image by using an image monitoring device; a building engineering drawing is introduced, and a basic illumination background is generated in combination with ambient light conditions and illumination light conditions; performing brightness balance optimization on a shadow area of the building engineering surface image, and filtering and cutting illumination dark spot partitions to obtain a primary processing image; performing morphological enhancement processing on the primary processing image to obtain a secondary processing image; recognizing a structural crack feature set and a non-structural crack feature set in combination with concrete surface features; and through the structural crack feature set and the non-structural crack feature set, the influence on the safety of the building structure is evaluated, and safety early warning of the building structure is carried out. The technical problems of low crack detection efficiency and insufficient accuracy in the prior art are solved, and the technical effect of efficiently and accurately detecting cracks is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for detecting cracks in construction projects based on image recognition. Background Art

[0002] With the continuous development of modern construction engineering technology, the requirements for the safety and durability of building structures are increasing. However, during the use of buildings, various cracks often appear on the surface of building projects due to various reasons such as material aging, construction defects, and environmental factors. These cracks not only affect the aesthetics of the building, but more importantly, they may indicate potential risks to the safety of the structure. Traditional crack detection methods mainly rely on manual inspections, which is not only inefficient, but also easily affected by human factors, making it difficult to ensure the accuracy of the detection results. Summary of the invention

[0003] The present application provides a method and system for detecting cracks in construction projects based on image recognition, which solves the technical problems of low efficiency and insufficient accuracy of crack detection in the prior art.

[0004] In view of the above problems, the present application provides a method and system for detecting cracks in construction projects based on image recognition.

[0005] In a first aspect of the present application, a method for detecting cracks in a construction project based on image recognition is provided, the method comprising: An image monitoring device is used to collect a surface image of a construction project, wherein the surface image of the construction project has an image shadow area mark; a construction project drawing is introduced, and a basic lighting background is generated in combination with ambient light conditions and lighting conditions; based on the basic lighting background, the shadow area of ​​the surface image of the construction project is optimized for brightness balance, and the lighting dark spot partitions are filtered and cropped to obtain a primary processed image; a set of structural elements is defined, and the primary processed image is morphologically enhanced to obtain a secondary processed image, and the morphological operation associated with the morphological enhancement includes an opening operation and a closing operation; through the secondary processed image, in combination with the concrete surface features, a structural crack feature set and a non-structural crack feature set are identified; through the structural crack feature set and the non-structural crack feature set, the impact on the safety of the building structure is evaluated, and a safety warning of the building structure is performed.

[0006] The second aspect of the present application provides a construction engineering crack detection system based on image recognition, the system comprising: An image acquisition module, the image acquisition module is used to use an image monitoring device to acquire a surface image of a construction project, the surface image of the construction project having an image shadow area mark; a basic illumination background generation module, the basic illumination background generation module is used to introduce a construction engineering drawing, and generate a basic illumination background in combination with ambient light conditions and illumination light conditions; a shadow area processing module, the shadow area processing module is used to optimize the brightness balance of the shadow area of ​​the surface image of the construction project based on the basic illumination background, and filter and crop the illumination dark spot partition to obtain a primary processed image; a morphological enhancement processing module, the morphological enhancement processing module is used to define a set of structural elements, perform morphological enhancement processing on the primary processed image, and obtain a secondary processed image, and the morphological operation mode associated with the morphological enhancement processing includes an opening operation and a closing operation; a crack feature recognition module, the crack feature recognition module is used to identify a structural crack feature set and a non-structural crack feature set through the secondary processed image in combination with concrete surface features; a safety assessment module, the safety assessment module is used to assess the impact on the safety of the building structure through the structural crack feature set and the non-structural crack feature set, and to provide a safety warning for the building structure.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, an image monitoring device is used to collect the surface image of the construction project, and the surface image of the construction project has the image shadow area mark. Then, the construction engineering drawings are introduced, and the basic lighting background is generated by combining the ambient light conditions and the lighting conditions. Then, based on the basic lighting background, the shadow area of ​​the surface image of the construction project is optimized for brightness balance, and the dark spot partition of the lighting is filtered and cropped to obtain a primary processed image. Further, a set of structural elements is defined, and the primary processed image is subjected to morphological enhancement processing to obtain a secondary processed image. The morphological operation associated with the morphological enhancement processing includes an opening operation and a closing operation. Next, through the secondary processed image, combined with the concrete surface features, the structural crack feature set and the non-structural crack feature set are identified. Finally, through the structural crack feature set and the non-structural crack feature set, the impact on the safety of the building structure is evaluated, and a safety warning of the building structure is performed. The technical problems of low efficiency and insufficient accuracy of crack detection in the prior art are solved, and the technical effect of efficient and accurate crack detection is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of a construction engineering crack detection method based on image recognition provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a construction engineering crack detection system based on image recognition provided in an embodiment of the present application.

[0010] Explanation of reference numerals: image acquisition module 11 , basic lighting background generation module 12 , shadow area processing module 13 , morphology enhancement processing module 14 , crack feature recognition module 15 , safety assessment module 16 . DETAILED DESCRIPTION

[0011] The present application solves the technical problems of low efficiency and insufficient accuracy of crack detection in the prior art by providing a construction engineering crack detection method and system based on image recognition.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a construction engineering crack detection method based on image recognition, wherein the method comprises: An image monitoring device is used to collect a surface image of a construction project, and the surface image of the construction project is provided with an image shadow area mark.

[0015] The surface of the construction project is monitored in real time through image monitoring equipment (high-definition camera) to obtain the surface image of the building. The surface image of the construction project not only contains the overall characteristics of the building surface, but also includes the shadow area caused by uneven lighting or obstructions. The shadow coverage area in the surface image of the construction project is annotated to generate an image shadow area mark.

[0016] Architectural engineering drawings are introduced, and the basic lighting background is generated by combining the ambient light conditions and the lighting conditions.

[0017] In the process of construction project monitoring and evaluation, construction engineering drawings are introduced and used as the reference basis for building structures. The accuracy of image processing is ensured by comparing the design details in the drawings with the actual collected building surface images. On this basis, the basic lighting background is generated by combining the actual site ambient light conditions (such as natural light intensity, light source position, reflection, etc.) and lighting conditions (such as the angle and brightness of artificial lighting, etc.). The basic lighting background is used to simulate the brightness and shadow distribution of the building surface under the current lighting environment as a reference for subsequent image processing. By generating a basic lighting background, the impact of lighting changes on the building surface image can be more accurately identified, so that shadow areas and light dark spots can be corrected to avoid misjudging the existence or severity of cracks due to lighting problems.

[0018] Based on the basic lighting background, the shadow area of ​​the building engineering surface image is optimized for brightness balance, and the lighting dark spot partitions are filtered and cropped to obtain a primary processed image.

[0019] By comparing the shadow area with the basic lighting background, the brightness deviation caused by uneven lighting or shadows is identified and adjusted to restore the brightness of the shadow area to the same level as the surrounding area. At the same time, the dark spot partitions formed by weak light in the image are filtered and cropped to eliminate the interference of these dark areas on subsequent image processing. After brightness optimization and smoothing, a primary processed image is generated, making the overall brightness of the image more uniform and the details clearer, providing a good foundation for the next step of morphological processing and crack identification.

[0020] Furthermore, based on the basic illumination background, the shadow area of ​​the building engineering surface image is optimized for brightness balance, and the illumination dark spot partitions are filtered and cropped to obtain a primary processed image, and the method includes: Based on the basic lighting background, the brightness balance parameters of the shadow area of ​​the building engineering surface image are determined; the brightness balance is optimized according to the brightness balance parameters of the shadow area of ​​the building engineering surface image to obtain a preprocessed image; based on the preprocessed image, the illumination dark spot partitions are filtered and cropped to obtain a primary processed image.

[0021] Specifically, by comparing the brightness difference between the basic lighting background and the shadow area in the actual image, the brightness balance parameter is determined. The brightness balance parameter is used to measure the amount of brightness adjustment required for the shadow area to achieve a visual effect similar to that of the normal lighting area; according to the determined brightness balance parameter, the shadow area is optimized for brightness, and the brightness of the shadow area is adjusted to be consistent with the surrounding non-shadow area, thereby avoiding crack detection errors caused by brightness differences; after completing the brightness balance optimization, a preprocessed image is generated, and the brightness of the shadow area in the preprocessed image has been balanced, so that the brightness distribution of the entire image is more uniform, which is convenient for subsequent processing steps; in the preprocessed image, dark spot partitions formed due to insufficient lighting are identified, these areas are cropped, and invalid information is eliminated to ensure that the retained image portion has sufficient light intensity and clarity; after processing, a primary processed image is obtained, which has been balanced and optimized in brightness, and no longer contains interfering dark spot areas, has clearer details, and is suitable for subsequent morphological processing and crack recognition.

[0022] Furthermore, based on the pre-processed image, filtering and cropping the illuminated dark spot partitions to obtain a primary processed image, the method includes: After the brightness balance is optimized, the illumination dark spot partitions are determined; based on the pre-processed image, the illumination dark spot partitions are locally filtered and cropped, and the cropped image is smoothed to obtain a primary processed image.

[0023] After completing the brightness balance optimization, the pre-processed image is analyzed to determine the illumination dark spot partitions. These dark spot partitions are usually local areas caused by insufficient lighting, shadows or poor reflections. Their brightness is lower than the average value of the overall image, which can easily affect the accuracy of subsequent crack detection. Based on the determined illumination dark spot partitions, these areas are locally filtered and cropped. By setting a brightness threshold, the areas below the threshold can be identified as dark spot areas and removed or replaced. After completing the cropping of the illumination dark spot partitions, the cropped image is smoothed to obtain a primary processed image. The primary processed image is more uniform in brightness and eliminates the interference caused by illumination dark spots.

[0024] A set of structural elements is defined, and morphological enhancement processing is performed on the primary processed image to obtain a secondary processed image. The morphological operation mode associated with the morphological enhancement processing includes an opening operation and a closing operation.

[0025] According to the morphological characteristics of building cracks, a suitable set of structural elements is defined. The structural elements can be matrices of different shapes and sizes (such as square, circular or linear structures). Their sizes and shapes should be adjusted according to the actual morphology of the cracks so as to more effectively extract the crack edge features. The morphological opening operation consists of corrosion and expansion operations, which are mainly used to remove small noise and artifacts in the image. When performing the opening operation on the primary processed image, the structural element is first used to corrode the image to eliminate the small noise in the image, and then the expansion operation is performed to restore the original structure. The morphological closing operation consists of expansion and corrosion operations, which are mainly used to close small holes or crack breakpoints in the image. When performing the closing operation, the image is first dilated to expand the target area (such as cracks) in the image, and then the corrosion operation is performed to restore the original shape of the crack. After the morphological opening and closing operations, the secondary processed image is obtained. After the crack features in the secondary processed image are morphologically enhanced, the edges are clearer, the noise and interference are effectively removed, and the morphological structure of the crack is more obvious, which is suitable for subsequent crack feature extraction and classification.

[0026] Furthermore, the morphological operation associated with the morphological enhancement process includes an opening operation, and the method includes: Evaluate local structural performance characteristics of a construction project, wherein the local structural performance characteristics include unit structural load and unit structural settlement; define a first structural element subset in a structural element set according to the local structural performance characteristics, wherein the first structural element subset is a circular structure, an elliptical structure, and a linear structure; based on the first structural element subset, disconnect a plurality of small cracks using an opening operation in a morphological operation method associated with the morphological enhancement processing.

[0027] Specifically, the local structural performance characteristics of the construction project are evaluated, including unit structural load and unit structural settlement, to identify stress concentration areas and settlement areas of the building, that is, high-risk locations for crack formation; based on the local structural performance characteristics obtained by the evaluation, a first structural element subset in the structural element set in the morphological operation is defined, and the first structural element subset includes circular structures, elliptical structures and linear structures. The circular and elliptical structural elements are used to process smooth cracks, while the linear structural elements are used for thinner or continuous cracks; based on the defined first structural element subset, an opening operation is performed on the primary processed image. Through the corrosion operation, the structural elements are used to disconnect multiple small cracks, and the discontinuous crack edges in the image are separated, and then the expansion operation is performed to restore the main structure of the crack.

[0028] Furthermore, the morphological operation associated with the morphological enhancement process includes a closing operation, and the method includes: Evaluate overall structural performance characteristics of a construction project, wherein the overall structural performance characteristics include global structural material shrinkage and global structural material expansion; define a second structural element subset in the structural element set based on the overall structural performance characteristics; and fill multiple small cracks based on the second structural element subset using a closing operation in a morphological operation method associated with the morphological enhancement processing.

[0029] Specifically, the overall structural performance characteristics of the construction project are evaluated, including global structural material shrinkage (such as shrinkage caused by water loss during the curing process of concrete) and global structural material expansion (such as thermal expansion and contraction caused by temperature changes) to identify areas where cracks may appear and their expansion trends; based on the overall structural performance characteristics obtained by the evaluation, a second structural element subset for closing operations is defined, and the second structural element subset includes larger or more complex shapes, such as circular, elliptical or rectangular structures, which can better close and fill gaps or discontinuous areas in the cracks; using the defined second structural element subset, a closing operation is performed on the primary processed image, and the edges of the cracks are expanded through the expansion operation to make the broken or discontinuous parts of the cracks easier to connect, and then the corrosion operation is used to restore the shape of the original crack and fill those small cracks or gaps.

[0030] Furthermore, a set of structural elements is defined, and morphological enhancement processing is performed on the primary processed image to obtain a secondary processed image. The method further includes: The crack candidate areas in the primary processed image are classified using a deep learning model to distinguish between structural cracks and non-structural cracks; based on examples of cracks in construction projects, crack morphological features are obtained, and the crack morphological features include branches, intersections, and endpoints; according to the crack morphological features and crack distribution features, the structural cracks and non-structural cracks in the crack candidate areas are matched with the morphological operation mode.

[0031] After completing the preliminary morphological processing, a deep learning model (such as a convolutional neural network (CNN)) is used to classify the crack candidate areas in the first-level processed image; the model can accurately distinguish between structural cracks and non-structural cracks by learning the morphological features, texture features, and distribution patterns of cracks. Structural cracks usually have a greater impact on building safety, while non-structural cracks may only be surface or decorative cracks with less impact. Based on crack instances, morphological features of cracks are extracted, including but not limited to geometric features such as branches, intersections, and endpoints. These features can help identify the complexity of cracks and their possible expansion paths. The branches and intersections of cracks reflect the complexity of cracks, while the endpoints show the starting or ending position of cracks. Combined with the distribution characteristics of cracks in building structures, the direction, length, density, etc. of cracks are analyzed. These distribution characteristics are related to building performance parameters such as structural load and settlement, which can further help identify the severity and potential risks of cracks. According to the morphological and distribution characteristics of cracks, appropriate morphological operations are matched to the structural cracks and non-structural cracks in the crack candidate area respectively; for structural cracks, more detailed closing operations and enhancement processing may be required to ensure accurate identification and tracking of cracks; for non-structural cracks, opening operations or simpler morphological processing methods can be used to eliminate surface noise or unimportant crack morphology.

[0032] The structural crack feature set and the non-structural crack feature set are identified through the secondary processed image in combination with the concrete surface features.

[0033] Image processing technology is used to extract key features of the concrete surface from the secondary processed image, including texture, surface roughness, pore distribution, etc. Based on the features of the concrete surface, cracks related to the stress and load distribution of the building structure are identified. These cracks usually show large crack widths, through cracks, or complex forms such as multiple branches and intersections, which may affect the overall stability of the building structure. At the same time, cracks that are limited to the surface, have no obvious depth, or have no serious impact on the structure are identified. These cracks are mostly cracks in the decorative layer, surface cracks caused by thermal expansion and contraction, etc., and their characteristics are usually small, shallow, and irregular cracks.

[0034] By analyzing the width, depth, shape, direction and other characteristics of the cracks in the secondary processed image and combining them with the concrete surface characteristics, the identified cracks are classified. Cracks with obvious structural risks are classified into the structural crack feature set; while cracks with no serious impact, such as surface cracks and decorative cracks, are classified into the non-structural crack feature set.

[0035] The impact on the safety of the building structure is evaluated through the structural crack feature set and the non-structural crack feature set, and a safety warning of the building structure is carried out.

[0036] By analyzing the feature sets of structural cracks and non-structural cracks, the impact of these cracks on the safety of building structures is evaluated. For structural cracks, the focus is on evaluating whether the cracks will affect the building's bearing capacity, durability and seismic resistance. The evaluation of non-structural cracks focuses on whether they affect the appearance or the normal use of other non-bearing components. If cracks that may endanger structural safety are identified, a safety warning will be automatically triggered.

[0037] Furthermore, the impact on the safety of the building structure is evaluated by using the structural crack feature set and the non-structural crack feature set, and a safety warning of the building structure is performed. The method includes: Based on the crack examples of the construction project, basic crack characteristics are extracted, and the basic crack characteristics include crack width characteristics, crack length characteristics, and crack depth characteristics; based on the basic crack characteristics, the impact on the safety of the building structure is evaluated to obtain a building structure safety index; a building structure safety threshold is set, the building structure safety index is compared with the building structure safety threshold, and it is determined whether to issue a safety warning instruction.

[0038] Specifically, based on examples of cracks in construction projects, basic crack characteristics are extracted, including crack width characteristics, crack length characteristics, and crack depth characteristics; based on the basic crack characteristics, an evaluation model for the impact of cracks on the safety of building structures is established using professional knowledge such as structural engineering principles and material mechanics, and the crack characteristic data is input into the evaluation model to calculate the building structure safety index, which is a quantitative indicator used to indicate the safety of the structure under the current crack state; a reasonable building structure safety threshold is set based on factors such as the design specifications of the building structure, historical data, and expert experience, and the threshold is used to determine whether the structure is in a safe state; the calculated building structure safety index is compared with the safety threshold; if the safety index is lower than the safety threshold, it indicates that there are safety hazards in the structure; when the safety index is lower than the safety threshold, a safety warning instruction is issued, which includes notifying relevant personnel, activating emergency plans, and conducting further inspections or repairs.

[0039] In summary, the embodiments of the present application have at least the following technical effects: First, an image monitoring device is used to collect the surface image of the construction project, and the surface image of the construction project has the image shadow area mark. Then, the construction engineering drawings are introduced, and the basic lighting background is generated by combining the ambient light conditions and the lighting conditions. Then, based on the basic lighting background, the shadow area of ​​the surface image of the construction project is optimized for brightness balance, and the dark spot partition of the lighting is filtered and cropped to obtain a primary processed image. Further, a set of structural elements is defined, and the primary processed image is subjected to morphological enhancement processing to obtain a secondary processed image. The morphological operation associated with the morphological enhancement processing includes an opening operation and a closing operation. Next, through the secondary processed image, combined with the concrete surface features, the structural crack feature set and the non-structural crack feature set are identified. Finally, through the structural crack feature set and the non-structural crack feature set, the impact on the safety of the building structure is evaluated, and a safety warning of the building structure is performed. The technical problems of low efficiency and insufficient accuracy of crack detection in the prior art are solved, and the technical effect of efficient and accurate crack detection is achieved.

[0040] Embodiment 2, based on the same inventive concept as the construction engineering crack detection method based on image recognition in the above embodiment, Figure 2 As shown, the present application provides a construction engineering crack detection system based on image recognition, wherein the system includes: An image acquisition module 11 is used to acquire a surface image of a construction project by using an image monitoring device, and the surface image of the construction project has an image shadow area mark; a basic illumination background generation module 12 is used to introduce a construction project drawing, and generate a basic illumination background by combining ambient light conditions and illumination conditions; a shadow area processing module 13 is used to optimize the brightness balance of the shadow area of ​​the surface image of the construction project based on the basic illumination background, and filter and crop the illumination dark spot partition to obtain a primary processed image; a morphological enhancement processing module 14 is used to generate a basic illumination background by introducing a construction project drawing and combining ambient light conditions and illumination conditions; a shadow area processing module 13 is used to optimize the brightness balance of the shadow area of ​​the surface image of the construction project based on the basic illumination background, and filter and crop the illumination dark spot partition to obtain a primary processed image; a morphological enhancement processing module 14 is used to generate a basic illumination background by combining ambient light conditions and illumination conditions; a shadow area processing module 13 ... The morphological enhancement processing module 14 is used to define a set of structural elements, perform morphological enhancement processing on the primary processed image, and obtain a secondary processed image. The morphological operation mode associated with the morphological enhancement processing includes an opening operation and a closing operation; the crack feature recognition module 15 is used to identify a structural crack feature set and a non-structural crack feature set through the secondary processed image in combination with concrete surface features; the safety assessment module 16 is used to evaluate the impact on the safety of the building structure through the structural crack feature set and the non-structural crack feature set, and to provide a safety warning for the building structure.

[0041] Furthermore, the shadow area processing module 13 is used to perform the following method: Based on the basic lighting background, the brightness balance parameters of the shadow area of ​​the building engineering surface image are determined; the brightness balance is optimized according to the brightness balance parameters of the shadow area of ​​the building engineering surface image to obtain a preprocessed image; based on the preprocessed image, the illumination dark spot partitions are filtered and cropped to obtain a primary processed image.

[0042] Furthermore, the shadow area processing module 13 is used to perform the following method: After the brightness balance is optimized, the illumination dark spot partitions are determined; based on the pre-processed image, the illumination dark spot partitions are locally filtered and cropped, and the cropped image is smoothed to obtain a primary processed image.

[0043] Furthermore, the morphology enhancement processing module 14 is used to perform the following method: Evaluate local structural performance characteristics of a construction project, wherein the local structural performance characteristics include unit structural load and unit structural settlement; define a first structural element subset in a structural element set according to the local structural performance characteristics, wherein the first structural element subset is a circular structure, an elliptical structure, and a linear structure; based on the first structural element subset, disconnect a plurality of small cracks using an opening operation in a morphological operation method associated with the morphological enhancement processing.

[0044] Furthermore, the morphology enhancement processing module 14 is used to perform the following method: Evaluate overall structural performance characteristics of a construction project, wherein the overall structural performance characteristics include global structural material shrinkage and global structural material expansion; define a second structural element subset in the structural element set based on the overall structural performance characteristics; and fill multiple small cracks based on the second structural element subset using a closing operation in a morphological operation method associated with the morphological enhancement processing.

[0045] Furthermore, the morphology enhancement processing module 14 is used to perform the following method: The crack candidate areas in the primary processed image are classified using a deep learning model to distinguish between structural cracks and non-structural cracks; based on examples of cracks in construction projects, crack morphological features are obtained, and the crack morphological features include branches, intersections, and endpoints; according to the crack morphological features and crack distribution features, the structural cracks and non-structural cracks in the crack candidate areas are matched with the morphological operation mode.

[0046] Furthermore, the security assessment module 16 is used to perform the following method: Based on the crack examples of the construction project, basic crack characteristics are extracted, and the basic crack characteristics include crack width characteristics, crack length characteristics, and crack depth characteristics; based on the basic crack characteristics, the impact on the safety of the building structure is evaluated to obtain a building structure safety index; a building structure safety threshold is set, the building structure safety index is compared with the building structure safety threshold, and it is determined whether to issue a safety warning instruction.

[0047] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0049] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A construction engineering crack detection method based on image recognition, characterized in that: The method comprises: An image monitoring device is used to collect a surface image of a construction project, wherein the surface image of the construction project has an image shadow area mark; Introduce architectural engineering drawings, combine ambient light conditions with lighting conditions, and generate basic lighting background; Based on the basic lighting background, the shadow area of ​​the building engineering surface image is optimized for brightness balance, and the lighting dark spot partitions are filtered and cropped to obtain a primary processed image; A set of structural elements is defined, and a morphological enhancement process is performed on the primary processed image to obtain a secondary processed image, wherein the morphological operation associated with the morphological enhancement process includes an opening operation and a closing operation; Using the secondary processed image and combining it with the concrete surface features, a structural crack feature set and a non-structural crack feature set are identified; The impact on the safety of the building structure is evaluated through the structural crack feature set and the non-structural crack feature set, and a safety warning of the building structure is carried out.

2. The construction engineering crack detection method based on image recognition as claimed in claim 1, characterized in that: Based on the basic illumination background, the shadow area of ​​the construction engineering surface image is optimized for brightness balance, and the illumination dark spot partitions are filtered and cropped to obtain a primary processed image. The method includes: Determining brightness balance parameters of shadow areas of the building engineering surface image based on the basic lighting background; Perform brightness balance optimization according to the brightness balance parameters of the shadow area of ​​the building engineering surface image to obtain a preprocessed image; Based on the pre-processed image, the illuminated dark spot partitions are filtered and cropped to obtain a primary processed image.

3. The construction engineering crack detection method based on image recognition as claimed in claim 2, characterized in that: Based on the pre-processed image, filtering and cropping the illuminated dark spot partitions to obtain a primary processed image, the method comprising: After the brightness balance is optimized, the light-dark spot partitioning is determined; Based on the pre-processed image, the illuminated dark spot partition is locally filtered and cropped, and the cropped image is smoothed to obtain a primary processed image.

4. The construction engineering crack detection method based on image recognition as claimed in claim 3, characterized in that: The morphological operation mode associated with the morphological enhancement process includes an opening operation, and the method includes: Evaluate local structural performance characteristics of a construction project, wherein the local structural performance characteristics include unit structural load and unit structural settlement; According to the local structural performance characteristics, a first structural element subset in the structural element set is defined, wherein the first structural element subset is a circular structure, an elliptical structure, and a linear structure; Based on the first structure element subset, a plurality of small cracks are disconnected by an opening operation in a morphological operation mode associated with the morphological enhancement process.

5. The construction engineering crack detection method based on image recognition as claimed in claim 4, characterized in that: The morphological operation mode associated with the morphological enhancement process includes a closing operation, and the method includes: Assessing overall structural performance characteristics of a construction project, the overall structural performance characteristics including global structural material shrinkage and global structural material expansion; Defining a second subset of structural elements in the set of structural elements according to the overall structural performance characteristics; Based on the second structure element subset, a plurality of small cracks are filled by a closing operation in a morphological operation mode associated with the morphological enhancement process.

6. The construction engineering crack detection method based on image recognition as claimed in claim 5, characterized in that: A set of structural elements is defined, and morphological enhancement processing is performed on the primary processed image to obtain a secondary processed image. The method further includes: Classifying the crack candidate regions in the primary processed image using a deep learning model to distinguish between structural cracks and non-structural cracks; Based on the example of cracks in construction projects, the crack morphological characteristics are obtained, wherein the crack morphological characteristics include branches, intersections, and endpoints; According to the crack morphological characteristics and crack distribution characteristics, the morphological operation mode is matched to the structural cracks and non-structural cracks in the crack candidate area.

7. The construction engineering crack detection method based on image recognition as claimed in claim 6, characterized in that: By using the structural crack feature set and the non-structural crack feature set, the impact on the safety of the building structure is evaluated and a safety warning of the building structure is performed. The method includes: Based on the construction engineering crack example, extracting crack basic features, the crack basic features including crack width features, crack length features, and crack depth features; Based on the basic characteristics of the cracks, the impact on the safety of the building structure is evaluated to obtain a building structure safety index; A building structure safety threshold is set, the building structure safety index is compared with the building structure safety threshold, and it is determined whether to issue a safety warning instruction.

8. The construction engineering crack detection system based on image recognition is characterized by: For implementing the construction engineering crack detection method based on image recognition according to any one of claims 1 to 7, the system comprises: An image acquisition module, the image acquisition module is used to acquire a surface image of a construction project using an image monitoring device, the surface image of the construction project having an image shadow area mark; A basic lighting background generation module, which is used to introduce architectural engineering drawings and generate a basic lighting background by combining ambient light conditions and lighting conditions; A shadow area processing module, the shadow area processing module is used to optimize the brightness balance of the shadow area of ​​the building engineering surface image based on the basic lighting background, and filter and crop the lighting dark spot partitions to obtain a primary processed image; A morphological enhancement processing module, wherein the morphological enhancement processing module is used to define a set of structural elements, perform morphological enhancement processing on the primary processed image, and obtain a secondary processed image. The morphological operation mode associated with the morphological enhancement processing includes an opening operation and a closing operation. A crack feature recognition module, the crack feature recognition module is used to recognize a structural crack feature set and a non-structural crack feature set through the secondary processed image in combination with concrete surface features; A safety assessment module is used to assess the impact on the safety of the building structure through the structural crack feature set and the non-structural crack feature set, and to provide a safety warning for the building structure.

Images