Bridge crack sub-pixel recognition method and system based on moiré phase unwrapping

Abstract

The present invention discloses a sub-pixel bridge crack identification method and system based on moiré phase unwrapping, which relates to the field of structural health monitoring. The method includes: collecting a bridge image and uploading it to the system, processing the image to obtain a denoised image; performing sub-pixel registration on a reference grating image of the denoised image to obtain a registered reference grating image; using the registered reference grating image to generate a moiré image and calculate a global quality score, and calculating the wrapped phase for moiré images greater than a threshold, otherwise requiring the drone to re-collect; calculating the unwrapped phase based on pixel weights for moiré images greater than a threshold, and using the unwrapped phase to derive the displacement field of each pixel; combining the displacement gradient amplitude with the level set function to identify the crack area and quantify the crack information. Through automated moiré analysis and displacement field calculation, tiny cracks in the bridge can be accurately detected, and the size and depth of the cracks can be quantified, providing reliable support for bridge health monitoring and maintenance decision-making.

Description

Technical Field

[0001] The present invention relates to the field of structural health monitoring, and in particular to a method and system for sub-pixel identification of bridge cracks based on moiré phase unwrapping. Background Art

[0002] As vital infrastructure, bridges play a crucial role in transportation and logistics. Ensuring their safety and long-term stable operation is crucial for public safety management. With the advancement of urbanization and the continued increase in traffic volume, bridge aging and damage are becoming increasingly apparent, necessitating the increasing importance of bridge health monitoring and assessment. While traditional bridge inspection methods, such as manual inspection, acoustic testing, and structural sensor monitoring, offer some capabilities, they suffer from low efficiency, high costs, the inability to monitor in real time, and difficulty obtaining comprehensive information. Consequently, in recent years, bridge inspection methods based on image processing and drone technology have become a hot topic in research and application.

[0003] For decades, bridge health monitoring has typically relied on manual or ground-based inspections, methods often associated with time-consuming and labor-intensive processes and limited coverage of the entire bridge structure. The results of manual inspections not only rely on the inspector's experience and skills but are also susceptible to weather and environmental conditions, leading to significant uncertainty. Furthermore, bridge health monitoring also includes techniques such as acoustics, vibration, and strain monitoring. While these methods can provide information on the dynamic response of bridges, they often have limitations, as most cannot accurately capture the subtle deformations of surface cracks.

[0004] Although drone technology and image processing methods have been widely used in bridge inspection, they still face some challenges. First, the accuracy and robustness of image processing algorithms need to be further improved. In particular, in severe weather, complex lighting conditions, and low-contrast scenes, the image quality is poor, which affects the effectiveness of crack detection. Second, the moiré interferometry method has very high requirements for image registration, especially in the case of large-area, high-resolution images. The registration accuracy directly affects the final detection results. In addition, existing crack detection methods mostly rely on static images, while bridge surfaces may experience dynamic changes. This requires image processing technology to not only analyze static images but also adapt to changes in dynamic scenes. Therefore, new algorithms and technological innovations are urgently needed. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a bridge crack sub-pixel identification method and system based on moiré phase unwrapping to solve the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a bridge crack sub-pixel identification method based on moiré phase unwrapping, comprising:

[0007] S1. A high-resolution camera mounted on a drone captures images of the bridge surface and uploads them to a cloud system. The images are then geometrically corrected and denoised to produce a denoised image.

[0008] S2. Generate a reference grating image based on the optimal period for the denoised image, and use sub-pixel registration to obtain the registered reference grating image.

[0009] S3. Generate a moiré image using the registered reference grating image. Calculate the global quality score of the generated moiré image. Calculate the wrapped phase for moiré images with a score greater than a threshold. Otherwise, request the drone to resample.

[0010] S4. Calculate pixel weights for moiré images with a score greater than the scoring threshold, calculate the unwrapped phase based on the pixel weights, and derive the displacement field for each pixel using the interferometer principle.

[0011] S5. Analyze the displacement gradient amplitude in the displacement field and combine it with the level set function to identify the crack area and quantify the width, length, and depth of the crack, providing quantitative information for structural damage assessment.

[0012] The present invention is further configured such that S2 specifically includes:

[0013] The gradient amplitude of the denoised image is calculated, and the gradient amplitude is used to optimize the grating period to obtain the optimal period;

[0014] generating a reference grating image based on the optimal period;

[0015] The displacement is calculated using the phase correlation method, and sub-pixel registration is performed to obtain the registered reference grating image.

[0016] The present invention is further configured such that the gradient amplitude calculation logic is: ,in, is the gradient amplitude, For denoised images, is the differential vector continuous differential operator;

[0017] Optimal cycle calculation logic: ,in, is the optimal period, is the grating period, is a sine function, is the angular frequency of the sine function;

[0018] Reference raster calculation logic: ,in, is the reference raster image, and are the base value and amplitude of the reference image, respectively. is the grating waveform sinusoidal generating function;

[0019] Reference grating registration calculation logic: ,in, is the reference grating image after registration, and is the image translation displacement calculated by the phase correlation method, is the bicubic interpolation function.

[0020] The present invention is further configured such that S3 specifically includes:

[0021] The moiré pattern image is obtained by multiplying the registered reference grating image with the denoised image to synthesize the moiré pattern;

[0022] The global quality score of the generated moiré image is calculated using the local contrast and frequency domain energy ratio;

[0023] The wrapped phase is calculated for moiré images with a score greater than the threshold, otherwise the drone is required to re-sample.

[0024] The present invention is further configured such that the moiré image generation logic is: ,in, It is a moiré image;

[0025] Global quality score calculation logic: ,in, is the global quality score, and is the size of the image, is the local contrast, is the frequency domain energy ratio, is the noise compensation factor, is the noise level;

[0026] Local contrast calculation logic: ,in, is the local contrast, is the maximum grayscale value in the local area, is the minimum grayscale value in the local area;

[0027] Frequency domain energy ratio calculation logic: ,in, is the frequency domain energy ratio, is the frequency domain representation of the image, is the frequency domain representation of the desired image.

[0028] The present invention is further configured such that S4 specifically includes:

[0029] Pixel weights are obtained by combining global quality score, local contrast, and structural consistency;

[0030] Based on the pixel weight, the maximum spanning tree is used to calculate the path integral to obtain the unwrapped phase;

[0031] The displacement field for each pixel is derived using the interferometer principle on the unwrapped phase.

[0032] The present invention is further configured such that the pixel weight calculation logic is: ,in, is the pixel weight, is the global quality score, is the local contrast, For structural consistency, is the noise level;

[0033] Path integral calculation logic: ,in, To unwrap the phase, is the wrapping phase, To unwrap the phase gradient, is the wrapped phase gradient, is the rounding function, is the maximum spanning tree path, is the phase period;

[0034] Displacement field calculation logic: ,in, is the displacement field, The wavelength of the laser is used. Specifically, the displacement field calculation uses the interferometer principle to convert the phase difference into physical displacement. Phase change correspondence displacement.

[0035] The present invention is further configured such that S5 specifically includes:

[0036] The displacement field is used to calculate the sub-pixel displacement gradient amplitude;

[0037] The crack contour is described based on the displacement gradient amplitude and the level set function. The coordinate sets of all detected crack boundaries are added to the crack contour point set. The length and depth of the positions in the crack contour point set are calculated to obtain the crack information.

[0038] The present invention is further configured such that the displacement gradient amplitude calculation logic is: ,in, is the displacement gradient amplitude;

[0039] Crack profile description calculation logic: ,in, is the level set function, is the curvature, is the gradient magnitude of the level set function;

[0040] Crack contour point set construction logic: ;in, is the set of crack contour points;

[0041] Length calculation logic: ,in, is the crack length, is the total number of contour points, is the point where the crack contour is concentrated, is the next point in the crack contour set, is the Euclidean distance between crack contour points;

[0042] Depth calculation logic: ,in, is the crack depth, is a constant factor, is the maximum displacement field value of all points in the contour point set;

[0043] Width calculation logic: ,in, is the standard deviation, which is obtained by sampling each contour point along the direction of the normal vector to obtain the gray value profile of the image and performing Gaussian fitting.

[0044] The present invention also provides a bridge crack sub-pixel identification system based on moiré phase unwrapping, the system comprising:

[0045] Image acquisition and preprocessing module: The drone is equipped with a high-resolution camera to capture bridge surface images and upload them to the cloud system. The bridge surface images are then geometrically corrected and denoised to obtain denoised images.

[0046] Raster generation and registration module: Generates a reference grating image based on the optimal period for the denoised image, and uses sub-pixel registration to obtain the registered reference grating image;

[0047] Moiré synthesis and quality assessment module: Generates a moiré image using the registered reference grating image. Calculates the global quality score for the generated moiré image. If the moiré image exceeds the scoring threshold, the wrapped phase is calculated. Otherwise, the drone is required to resample.

[0048] Phase unwrapping and displacement calculation module: Calculates pixel weights for moiré images with a score greater than the scoring threshold, obtains the unwrapped phase based on the pixel weights, and uses the interferometer principle to derive the displacement field of each pixel from the unwrapped phase.

[0049] Crack Identification and Quantification Module: Analyzes the displacement gradient amplitude in the displacement field, combines the level set function to identify the crack area and quantify the width, length and depth of the crack, providing quantitative information for structural damage assessment.

[0050] The present invention provides a sub-pixel bridge crack identification method and system based on moiré phase unwrapping. The method uses a high-resolution camera mounted on an unmanned aerial vehicle (UAV) to capture a bridge surface image and upload it to a cloud system. The bridge surface image is then geometrically corrected and denoised to obtain a denoised image. A reference grating image is generated from the denoised image according to an optimal period, and a registered reference grating image is obtained using sub-pixel registration. A moiré image is generated using the registered reference grating image, and a global quality score is calculated for the generated moiré image. Moiré images with a score greater than a threshold are then parcelled, and those with a score less than the threshold are required to be re-sampled by the UAV. Pixel weights are calculated for the moiré images with a score greater than the threshold, and an unwrapped phase is calculated based on the pixel weights. The displacement field of each pixel is derived from the unwrapped phase using the interferometer principle. The displacement gradient amplitude in the displacement field is analyzed, and a level set function is used to identify crack areas and quantify the width, length, and depth of the cracks, providing quantitative information for structural damage assessment. The resulting beneficial effects include:

[0051] Efficient and accurate bridge surface damage detection: By using drones equipped with high-resolution cameras to capture bridge surface images and combining them with image denoising, image registration, moiré generation, and displacement field analysis, this method can accurately capture tiny cracks and other structural damage on the bridge surface. This method not only improves detection accuracy but also significantly increases detection efficiency, providing more comprehensive and accurate technical support for bridge health monitoring.

[0052] Automated Crack Detection and Quantification: The generated moiré image is automatically assessed for quality, combined with global quality scores such as local contrast and frequency-domain energy ratio. Cracks are then accurately detected and the wrapped phase is resolved. A gradient calculation method is used to derive the displacement field from the unwrapped phase, further quantifying the geometric characteristics of the cracks, such as width, length, and depth. This automated process significantly reduces the need for manual intervention and improves crack detection accuracy.

[0053] High-precision crack size and depth analysis: Sub-pixel displacement gradient amplitude analysis based on displacement field calculations, combined with level set functions for crack contour description, accurately identifies crack areas and quantifies crack width, length, and depth. By analyzing the collective crack contour points, the geometric characteristics of the cracks are accurately calculated, providing a quantitative basis for structural damage assessment. This technology is of great significance for long-term monitoring and maintenance decision-making of bridges.

[0054] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without inventive efforts. In the drawings:

[0056] Figure 1 This is a flow chart showing a method for sub-pixel identification of bridge cracks based on moiré phase unwrapping according to an exemplary embodiment of the present invention;

[0057] Figure 2 The figure is a schematic structural diagram of a bridge crack sub-pixel identification system based on moiré phase unwrapping according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0058] The following describes the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art will readily appreciate the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are intended only to illustrate the present invention and are not intended to limit the scope of protection of the present invention.

[0059] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0060] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention. Example 1

[0061] Sub-pixel recognition method for bridge cracks based on moiré phase unwrapping, such as Figure 1 As shown, including:

[0062] S1. A high-resolution camera mounted on a drone captures images of the bridge surface and uploads them to a cloud system. The images are then geometrically corrected and denoised to produce a denoised image.

[0063] S2. Generate a reference grating image based on the optimal period for the denoised image, and use sub-pixel registration to obtain the registered reference grating image.

[0064] S3. Generate a moiré image using the registered reference grating image. Calculate the global quality score of the generated moiré image. Calculate the wrapped phase for moiré images with a score greater than a threshold. Otherwise, request the drone to resample.

[0065] S4. Calculate pixel weights for moiré images with a score greater than the scoring threshold, calculate the unwrapped phase based on the pixel weights, and derive the displacement field for each pixel using the interferometer principle.

[0066] S5. Analyze the displacement gradient amplitude in the displacement field and combine it with the level set function to identify the crack area and quantify the width, length, and depth of the crack, providing quantitative information for structural damage assessment.

[0067] The present invention is further configured such that S2 specifically includes:

[0068] The gradient amplitude of the denoised image is calculated, and the gradient amplitude is used to optimize the grating period to obtain the optimal period;

[0069] generating a reference grating image based on the optimal period;

[0070] Phase correlation is used to calculate displacement and perform sub-pixel registration to obtain a registered reference grating image. Specifically, the gradient magnitude represents the edge strength or saliency of the image at that location. A larger gradient magnitude indicates a more dramatic image change at that location, typically indicating the presence of an edge or texture. These gradient magnitudes are used to optimize the grating period. The goal is to find a period that matches the image texture with the grating period. This optimization is achieved by maximizing the weighted sinusoidal response. A reference grating image is generated based on the optimal period. This reference image is an image with a periodic sinusoidal texture, so that the generated grating pattern matches the texture features of the original image. The registered reference grating is obtained by calculating the phase correlation between the images in the frequency domain using the phase correlation method. After the translation displacement is obtained, bicubic interpolation is used to achieve sub-pixel image translation. The basic idea of the phase correlation method is to calculate the correlation between images in the frequency domain through product multiplication. After Fourier transform, the phase component contains the spatial information of the image. The displacement between the images can be obtained using the complex conjugate product, which corresponds to the offset required for image registration.

[0071] The present invention is further configured such that the gradient amplitude calculation logic is: ,in, is the gradient amplitude, For denoised images, is the differential vector continuous differential operator;

[0072] Optimal cycle calculation logic: ,in, is the optimal period, is the grating period, is a sine function, is the angular frequency of the sine function;

[0073] Reference raster calculation logic: ,in, is the reference raster image, and are the base value and amplitude of the reference image, respectively. is the grating waveform sinusoidal generating function;

[0074] Reference grating registration calculation logic: ,in, is the reference grating image after registration, and is the image translation displacement calculated by the phase correlation method, is the bicubic interpolation function. Specifically, the gradient amplitude It is used to indicate the intensity of local changes in the image. The larger the gradient amplitude is, the more sensitive it is to the selection of the grating period. Indicates that the corrected image is at position The gradient at the position is a measure of the rate of change of image intensity, indicating the direction and rate of change of the image at that position. Refers to the Euclidean norm of the image gradient vector, which is used here to indicate the magnitude of the gradient. The larger the gradient, the more dramatic the image change, that is, the more obvious the edge. By adjusting the grating period To optimize the texture response of the image, the grating period The value range is This is a search interval, the purpose of which is to select the grating period that can best enhance the image features within this range. Indicates that the image position under the grating period The texture response of the sine function is to generate the same The corresponding fundamental frequency component expresses the texture characteristics of the grating in the form of fluctuations, matches the grating structure of different periods with the image content, and the angular frequency Indicates each position Grating frequency and grating period The relationship between Is the basic frequency of the grating, indicating the grating's fluctuation frequency within a period, which determines the periodicity of the sine wave. The frequency of a smaller grating will be very high and the fluctuations will be very dense, and if Larger fluctuations are sparser; refer to the raster image Based on the optimized grating period and sine wave function, Is the base brightness of the image, ensuring that the generated image value is within In the range, 127 is the amplitude of the raster image, ensuring that the generated image pixel values are from Changes between is the sinusoidal function that generates the grating pattern and defines the spatial periodicity of the pattern, where Indicates along Periodic variation of the axis, Indicates along The axis changes periodically, and its frequency is Half of the axis direction; reference raster image after registration The reference raster image is translated at the sub-pixel level by bicubic interpolation to align the target image. This translation adjusts the spatial position of the image so that the texture structure of the reference raster image and the target image are as consistent as possible. Perform translation operation on the reference raster image, adjust the image position to align with the target image, and use bicubic interpolation function

[0075] It is a high-precision interpolation method that calculates the new pixel value by considering the values of the 16 adjacent pixels in the image. Compared with nearest neighbor interpolation and bilinear interpolation, it can provide higher accuracy.

[0076] The present invention is further configured such that S3 specifically includes:

[0077] The moiré pattern image is obtained by multiplying the registered reference grating image with the denoised image to synthesize the moiré pattern;

[0078] The global quality score of the generated moiré image is calculated using the local contrast and frequency domain energy ratio;

[0079] For moiré images with a score greater than a threshold, a wrapping phase is calculated; otherwise, the drone must reacquire the image. Specifically, during the moiré synthesis process, the pixel values of the two images are multiplied together to create a new image that enhances the periodic structure in the image. This is equivalent to superimposing the texture information of the two images, creating a moiré effect. The global quality score combines the image's contrast, frequency domain purity, and noise level. A higher image quality score indicates better image detail, structure, and signal-to-noise ratio. A higher average contrast indicates rich image detail and clear structure; a higher frequency domain purity indicates significant image periodicity and less noise. Noise compensation is used to reduce the impact of noise and improve the image quality score. A scoring threshold is set. If the score is less than the threshold, the drone image corresponding to the moiré pattern is not clear enough for subsequent image processing and requires reacquiring the image at that location. If the score is greater than the threshold, the moiré image corresponding to the global quality score is high-quality and can be used for subsequent crack identification. The wrapping phase is calculated for this image and the next module is entered. Calculating phase wrapping is a prior art technique and will not be elaborated on here.

[0080] The present invention is further configured such that the moiré image generation logic is:

[0081] ,in, It is a moiré image;

[0082] Global quality score calculation logic: ,in, is the global quality score, and is the size of the image, is the local contrast, is the frequency domain energy ratio, is the noise compensation factor, is the noise level;

[0083] Local contrast calculation logic: ,in, is the local contrast, is the maximum grayscale value in the local area, is the minimum grayscale value in the local area;

[0084] Frequency domain energy ratio calculation logic: ,in, is the frequency domain energy ratio, is the frequency domain representation of the image, is the frequency domain representation of the desired image. Specifically, in the calculation of the moiré image, the features of the two images can be combined by multiplying the values of the two images at each pixel position. For example, if the two images contain different periodic structures, the product will highlight these structures. This operation can enhance the periodic interference in the image and can also be used to enhance certain specific texture features of the image. The global quality score quantifies the quality of the moiré image and evaluates the quality of the moiré image. The global quality score calculation combines the image contrast, frequency domain purity and noise level. The higher the image quality score, the better the image details, structure and signal-to-noise ratio. As input value, the global quality score of the moiré image is calculated by the formula, and the image size and Represents the width and height of the image respectively, The total number of pixels in the image, Represents the average contrast of the image, which is the average of the contrast values of all pixel positions; local contrast Location The contrast at the image is usually a measure of the local change in the grayscale value of the image. The contrast is used to measure the details and structural clarity of the image. A higher contrast usually means that there are more details in the image. and It is the maximum and minimum grayscale values in the local window, which reflect the degree of grayscale difference in the local area. Local contrast calculation is usually performed locally, and the contrast of each pixel value is calculated based on the grayscale values of the pixels around it; frequency domain energy ratio It is the energy ratio of the image in the frequency domain, indicating the purity of the image frequency domain or the concentration of the frequency domain information. It is used to measure the ratio of the image frequency components to noise or irrelevant components. The frequency domain representation of the image Obtained through Fourier transform of the image, it contains the information of the image at each frequency, and the frequency domain representation of the expected image It is also obtained through the Fourier transform of the image, which contains the information of the image at each frequency. The expected image is a preset template image, which provides a benchmark for the frequency domain energy ratio and is used to quantify the quality of the target image; the noise compensation factor The role of this factor is to compensate for the noise in the image and reduce the impact of noise on the quality score. is calculated through the variance of the moiré image, which indicates the noise intensity of the moiré image. A higher variance value indicates stronger noise in the image. The specific calculation formula is: ,in, Indicates that the moiré pattern image is at position The pixel variance at,variance is a measure of the change in pixel grayscale value in a local area of the image. A larger variance means that the pixel changes in the area are larger, which is usually related to noise or texture.

[0085] The present invention is further configured such that S4 specifically includes:

[0086] Pixel weights are obtained by combining global quality score, local contrast, and structural consistency;

[0087] Based on the pixel weight, the maximum spanning tree is used to calculate the path integral to obtain the unwrapped phase;

[0088] The interferometer principle is used to derive the displacement field of each pixel from the unwrapped phase. Specifically, the global quality score in the pixel weight is used to determine the overall signal-to-noise ratio of the tibial moiré image, the local contrast is used to reflect the fringe clarity of the moiré image, and the structural consistency is used to quantify the consistency of the fringe direction. The pixel weights calculated by coupling three physical features are more reliable than single weights. A maximum spanning tree is constructed so that the integral path preferentially passes through the high-weight path, reducing error propagation and making the obtained unwrapped phase more reliable. In fields such as image registration, object deformation, or motion analysis, the calculation of the displacement field is very important for understanding the spatial relationships and changes in the image. Using the interferometer principle to derive the displacement field of each pixel, especially in optical image processing, can help accurately capture local changes in the image, especially small deformations or displacements.

[0089] The present invention is further configured such that the pixel weight calculation logic is: ,in, is the pixel weight, is the global quality score, is the local contrast, For structural consistency, is the noise level;

[0090] Path integral calculation logic: ,in, To unwrap the phase, is the wrapping phase, To unwrap the phase gradient, is the wrapped phase gradient, is the rounding function, is the maximum spanning tree path, is the phase period;

[0091] Displacement field calculation logic: ,in, is the displacement field, is the laser wavelength. Specifically, the pixel weight By combining the global quality score, local contrast and structural consistency, a weight is assigned to each pixel according to the different characteristics of the image, thereby generating high-quality image synthesis results. The compensation of the noise level further ensures that the noise area will not affect the final image quality; structural consistency It is used to detect the stability of texture and shape of local areas. If the local structure of the image is relatively consistent, the weight of the area is higher, otherwise it is lower. Structural consistency can be calculated by edge detection, texture analysis and other methods. The value range of structural consistency is The larger the value, the more stable the structure; the noise level It is used to represent the noise level in a local area of the image. The area with large noise will affect the image quality, so it needs to be compensated. In the formula, the noise level is used as a factor to adjust the final weight to ensure that the weight of the noisy area is low. The formula is:

[0092] ,in Indicates that the moiré pattern image is at position The pixel variance at , the value range is greater than or equal to , the larger the value, the stronger the noise and the worse the image quality. In the calculation of pixel weights: high-quality areas will receive higher pixel weights, low-quality areas will receive lower pixel weights, and the noise area will be calculated by The introduction of is automatically suppressed, avoiding the interference of noise on image processing; path integral In the process, the edge weight of the maximum tree is defined according to the pixel weight, and the edges with larger pixel weights are prioritized to ensure that the high reliability area is integrated first. After the phase jump correction calculation, the correction value is accumulated along the path to gradually restore the true phase. The difference between the unwrapped phase gradient and the wrapped phase gradient is calculated and then used Integer multiples of To compensate the jump to the correction value, wrap the phase It is generated by calculating the global quality score of the moiré image in S3 and judging that it is greater than the threshold. It is an existing technology and will not be described in detail here. The maximum spanning tree path It is generated by the algorithm based on the integral path of pixel weights. The path sequence is directly determined by Determines the cumulative order that ultimately affects phase compensation, phase cycle It is the period length of the interference fringe and the basis for jump compensation. In displacement field calculation, the displacement field uses the interferometer principle to convert the phase difference into physical displacement. The conversion principle is that every Phase change correspondence The displacement of the laser wavelength is the laser physical constant, and its value is usually , is the interference optical path coefficient, which is usually .

[0093] The present invention is further configured such that S5 specifically includes:

[0094] The displacement field is used to calculate the sub-pixel displacement gradient amplitude;

[0095] The crack contour is described based on the displacement gradient amplitude and the level set function. The coordinate sets of all detected crack boundaries are added to the crack contour point set. The length and depth of the positions in the crack contour point set are calculated to obtain the crack information. Specifically, when calculating the displacement field, more subtle displacement changes can be captured with sub-pixel accuracy. The displacement gradient refers to the rate of change between adjacent pixels in the displacement field. It can reflect the local change trend in the image, especially at the edge of cracks or defects. The displacement gradient usually changes significantly. The sub-pixel displacement gradient amplitude can be accurate to a higher accuracy than the pixel scale, thereby identifying small cracks or damage in the image. The level set method is a mathematical tool for processing the evolution of curves or surfaces. Through the level set function, the crack boundaries in the image can be described and tracked. Its role is to effectively distinguish between crack areas and non-crack areas by setting a specific value range, and to more accurately model and describe cracks. Conclusion Combining the displacement gradient information and level set function of the crack in the displacement field can accurately describe the shape and contour of the crack. The crack contour point set consists of the detected crack boundary coordinates, and these point sets can be used to perform a more detailed geometric analysis of the crack. Based on the crack contour point set, the length and width of the crack can be obtained by calculating the Euclidean distance between the points in the crack contour point set. The Euclidean distance is used to measure the straight-line distance between two points. By analyzing the distribution of the crack boundary, the geometric dimensions of the crack can be accurately calculated. The depth of the crack is usually estimated by the maximum value of the displacement field or the grayscale profile obtained by vertical sampling from the crack boundary. Crack depth is an important parameter affecting structural safety. By quantifying the depth, the threat posed by cracks to bridge structures can be assessed.

[0096] The present invention is further configured such that the displacement gradient amplitude calculation logic is: ,in, is the displacement gradient amplitude;

[0097] Crack profile description calculation logic: ,in, is the level set function, is the curvature, is the gradient magnitude of the level set function;

[0098] Crack contour point set construction logic: ;in, is the set of crack contour points;

[0099] Length calculation logic: ,in, is the crack length, is the total number of contour points, is the point where the crack contour is concentrated, is the next point in the crack contour set, is the Euclidean distance between crack contour points;

[0100] Depth calculation logic: ,in, is the crack depth, is a constant factor, is the maximum displacement field value of all points in the contour point set;

[0101] Width calculation logic: ,in, is the standard deviation, which is obtained by Gaussian fitting the grayscale value profile of the image obtained by sampling each contour point along the direction of the normal vector. Specifically, the displacement gradient amplitude It measures the intensity of local deformation in the image, which varies greatly in the crack area. By calculating the displacement difference between adjacent pixels at each pixel in the image, the gradient amplitude of the displacement field is obtained. The larger the value, the closer the image corresponds to the edge of the crack or damage. The displacement field is The gradient in the direction, that is, the displacement change rate of each pixel adjacent to the pixel in the horizontal direction, The displacement field is The gradient in the direction, that is, the rate of change of the displacement of each pixel adjacent to the pixel in the vertical direction; the level set function represents the “implicit representation” of the crack boundary in the image, which is a function used to describe and track the dynamic changes of the crack boundary, where: , Indicates the edge of the crack represents the outer area of the crack, the curvature Indicates the degree of curvature of the crack boundary, and the calculation formula is: Higher curvature values correspond to more curved cracks. The gradient amplitude of the level set function represents the rate of change of the crack boundary. The gradient amplitude at the boundary is larger. This formula describes the evolution of the level set function over time, taking into account the curvature and displacement field of the crack. The first term is Make the crack boundary smooth, and the second The crack boundary is forced to expand along the displacement gradient direction, which helps to identify the crack area. By solving this equation, the crack boundary can be dynamically tracked and the crack contour representation can be gradually optimized. The crack contour point set Represents all points in the image that belong to the crack boundary, by filtering the level set function The point set is obtained; the crack length By calculating the distance between consecutive points in the crack contour point set, the length of the entire crack is accumulated. The Euclidean distance is used to measure the straight-line distance between two points; the crack depth It is obtained by finding the maximum displacement value in the set of crack contour points and then multiplying it by a constant factor. The constant factor is mainly used to adjust the quantitative range of depth, and its value usually depends on experiments or material properties. The maximum displacement value of all points in the crack contour point set; the crack width The standard deviation is obtained by performing Gaussian fitting on the grayscale profile of the crack boundary and then multiplying it by a constant factor. This method is a prior art and will not be described in detail here. Example 2

[0102] See also Figure 2 The exemplary bridge crack sub-pixel identification system based on moiré phase unwrapping includes:

[0103] Image acquisition and preprocessing module: The drone is equipped with a high-resolution camera to capture bridge surface images and upload them to the cloud system. The bridge surface images are then geometrically corrected and denoised to obtain denoised images.

[0104] Raster generation and registration module: Generates a reference grating image based on the optimal period for the denoised image, and uses sub-pixel registration to obtain the registered reference grating image;

[0105] Moiré synthesis and quality assessment module: Generates a moiré image using the registered reference grating image. Calculates the global quality score for the generated moiré image. If the moiré image exceeds the scoring threshold, the wrapped phase is calculated. Otherwise, the drone is required to resample.

[0106] Phase unwrapping and displacement calculation module: Calculates pixel weights for moiré images with a score greater than the scoring threshold, obtains the unwrapped phase based on the pixel weights, and uses the interferometer principle to derive the displacement field of each pixel from the unwrapped phase.

[0107] Crack Identification and Quantification Module: Analyzes the displacement gradient amplitude in the displacement field, combines the level set function to identify the crack area and quantify the width, length and depth of the crack, providing quantitative information for structural damage assessment.

[0108] It should be noted that the moiré phase unwrapping-based bridge crack sub-pixel identification system provided in the above-mentioned embodiment and the moiré phase unwrapping-based bridge crack sub-pixel identification method provided in the above-mentioned embodiment are based on the same concept. The specific manner in which each module and unit performs operations has been described in detail in the method embodiments and will not be repeated here. In actual applications, the moiré phase unwrapping-based bridge crack sub-pixel identification system provided in the above-mentioned embodiment can, as needed, allocate the above-mentioned functions to different functional modules, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0109] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0110] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0111] In this application, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0112] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0113] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0114] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0115] In the several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0116] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0117] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0118] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling 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 method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0119] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A sub-pixel bridge crack identification method based on moiré phase unwrapping, characterized by: include: S1. A high-resolution camera mounted on a drone captures images of the bridge surface and uploads them to a cloud system. The images are then geometrically corrected and denoised to produce a denoised image. S2. Generate a reference grating image based on the optimal period for the denoised image, and use sub-pixel registration to obtain a registered reference grating image. This includes: calculating the gradient amplitude of the denoised image, optimizing the grating period using the gradient amplitude to obtain the optimal period; generating a reference grating image based on the optimal period; calculating the displacement using the phase correlation method, and performing sub-pixel registration to obtain the registered reference grating image. Gradient amplitude calculation logic: ,in, is the gradient amplitude, For denoised images, is the continuous differential operator of the differential vector; the optimal period calculation logic is: ,in, is the optimal period, is the grating period, is a sine function, is the angular frequency of the sine function; refer to the grating calculation logic: ,in, is the reference raster image, and are the base value and amplitude of the reference image, respectively. is the grating waveform sine generating function; refer to the grating registration calculation logic: ,in, is the reference grating image after registration, and is the image translation displacement calculated by the phase correlation method, is the bicubic interpolation function; S3. Generate a moiré image using the registered reference grating image. Calculate the global quality score of the generated moiré image. Calculate the wrapped phase for moiré images with a score greater than a threshold. Otherwise, request the drone to resample. S4. Calculate pixel weights for moiré images with a score greater than the scoring threshold, calculate the unwrapped phase based on the pixel weights, and derive the displacement field for each pixel using the interferometer principle. S5. Analyze the displacement gradient amplitude in the displacement field and combine it with the level set function to identify the crack area and quantify the width, length, and depth of the crack, providing quantitative information for structural damage assessment.

2. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 1 is characterized in that: S3 specifically includes: The moiré pattern image is obtained by multiplying the registered reference grating image with the denoised image to synthesize the moiré pattern; The global quality score of the generated moiré image is calculated using the local contrast and frequency domain energy ratio; The wrapped phase is calculated for moiré images with a score greater than the threshold, otherwise the drone is required to re-sample.

3. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 2 is characterized in that: Moiré image generation logic: ,in, It is a moiré image; Global quality score calculation logic: ,in, is the global quality score, and is the size of the image, is the local contrast, is the frequency domain energy ratio, is the noise compensation factor, is the noise level; Local contrast calculation logic: ,in, is the local contrast, is the maximum grayscale value in the local area, is the minimum grayscale value in the local area; Frequency domain energy ratio calculation logic: ,in, is the frequency domain energy ratio, is the frequency domain representation of the image, is the frequency domain representation of the desired image.

4. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 1 is characterized in that: S4 specifically includes: Pixel weights are obtained by combining global quality score, local contrast, and structural consistency; Based on the pixel weight, the unwrapped phase is calculated by using the maximum spanning tree to find the path integral. The displacement field for each pixel is derived using the interferometer principle on the unwrapped phase.

5. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 4 is characterized in that: Pixel weight calculation logic: ,in, is the pixel weight, is the global quality score, is the local contrast, For structural consistency, is the noise level; Path integral calculation logic: ,in, To unwrap the phase, is the wrapping phase, To unwrap the phase gradient, is the wrapped phase gradient, is the rounding function, is the maximum spanning tree path, is the phase period; Displacement field calculation logic: ,in, is the displacement field, The wavelength of the laser is used. Specifically, the displacement field calculation uses the interferometer principle to convert the phase difference into physical displacement. Phase change correspondence displacement.

6. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 1 is characterized in that: S5 specifically includes: The displacement field is used to calculate the sub-pixel displacement gradient amplitude; The crack contour is described based on the displacement gradient amplitude and the level set function. The coordinate sets of all detected crack boundaries are added to the crack contour point set. The length and depth of the positions in the crack contour point set are calculated to obtain the crack information.

7. The bridge crack sub-pixel identification method based on moiré phase unwrapping according to claim 6 is characterized in that: Displacement gradient amplitude calculation logic: ,in, is the displacement gradient amplitude; Crack profile description calculation logic: ,in, is the level set function, is the curvature, is the gradient magnitude of the level set function; Crack contour point set construction logic: ;in, is the set of crack contour points; Length calculation logic: ,in, is the crack length, is the total number of contour points, is the point where the crack contour is concentrated, is the next point in the crack contour set, is the Euclidean distance between crack contour points; Depth calculation logic: ,in, is the crack depth, is a constant factor, is the maximum displacement field value of all points in the contour point set; Width calculation logic: ,in, is the standard deviation, which is obtained by sampling each contour point along the direction of the normal vector to obtain the gray value profile of the image and performing Gaussian fitting.

8. A bridge crack sub-pixel identification system based on moiré phase unwrapping, used to implement the bridge crack sub-pixel identification method based on moiré phase unwrapping according to any one of claims 1 to 7, characterized in that: include: Image acquisition and preprocessing module: The drone is equipped with a high-resolution camera to capture bridge surface images and upload them to the cloud system. The bridge surface images are then geometrically corrected and denoised to obtain denoised images. Raster generation and registration module: Generates a reference grating image based on the optimal period for the denoised image, and uses sub-pixel registration to obtain the registered reference grating image; Moiré synthesis and quality assessment module: Generates a moiré image using the registered reference grating image. Calculates the global quality score for the generated moiré image. If the moiré image exceeds the scoring threshold, the wrapped phase is calculated. Otherwise, the drone is required to resample. Phase unwrapping and displacement calculation module: Calculates pixel weights for moiré images with a score greater than the scoring threshold, obtains the unwrapped phase based on the pixel weights, and uses the interferometer principle to derive the displacement field of each pixel from the unwrapped phase. Crack Identification and Quantification Module: Analyzes the displacement gradient amplitude in the displacement field, combines the level set function to identify the crack area and quantify the width, length and depth of the crack, providing quantitative information for structural damage assessment.

Images