A geological disaster detection method based on image processing

Abstract

The present invention discloses a geological disaster detection method based on image processing, which relates to the field of disaster monitoring technology and includes the following steps: S1, obtaining initial road surface image data of the area to be detected, and geometrically correcting the initial road surface image data to obtain road surface feature image data; S2, using the outer edge coefficients and pixel grayscale values ​​of the spots in the road surface feature image data to screen and eliminate all spots, and use the remaining spots as the re-detection area; S3, processing the re-detection area to obtain the crack area of ​​the road surface. The present invention accurately locates the crack area, helps to timely discover early signs of geological disasters, provides a key basis for subsequent disaster warning, assessment and prevention, and helps to take timely and effective measures to reduce the losses caused by geological disasters.

Description

Technical Field

[0001] The present invention relates to the technical field of disaster monitoring, and in particular to a geological disaster detection method based on image processing. Background Art

[0002] As natural disasters pose a serious threat to human life, property, and the stability of the natural environment, monitoring and early warning of geological hazards have long been a research priority in geology, disaster science, and related engineering fields. The occurrence of geological hazards is often accompanied by a series of complex geological environmental changes, with pavement damage being a common early warning sign. For example, before or during geological hazards such as earthquakes, landslides, and debris flows, pavement surfaces may exhibit abnormalities such as cracks, subsidence, and heaves due to the movement and deformation of geological bodies. Timely and accurate detection of these pavement anomalies is crucial for early warning and prevention of geological hazards.

[0003] Traditional methods for detecting pavement changes due to geological hazards rely primarily on manual inspections. Inspectors regularly conduct on-site inspections of areas where geological hazards may occur, visually inspecting or using simple measuring tools to detect abnormalities such as cracks in the pavement. However, this method has numerous limitations. For one thing, manual inspections are inefficient, making it difficult to conduct comprehensive, real-time monitoring of large areas, and prone to missing early and subtle pavement changes. Furthermore, manual inspection results are significantly influenced by subjective factors, and the experience and judgment criteria of different inspectors may vary, making it difficult to guarantee the accuracy and reliability of the results.

[0004] With the rapid development of image processing technology, its application in geological disaster detection has gradually attracted attention. Image processing technology can automatically extract feature information from images by analyzing and processing acquired image data, thereby enabling the detection and identification of target objects. In geological disaster pavement detection, image processing technology can be used to obtain road surface image data of the area to be inspected and conduct in-depth analysis of this data to identify any anomalies in the road surface.

[0005] In practical applications, acquired road image data often exhibits certain geometric distortions. This is due to a variety of factors, including lens distortion in imaging devices (such as cameras), shooting angle, and terrain undulations. Geometric distortion can affect the accuracy of subsequent image analysis. Furthermore, road image data often contains a significant amount of noise and irrelevant information, which can interfere with subsequent road anomaly detection, reducing accuracy and efficiency. Therefore, an effective method is needed to accurately detect cracks in the road surface. Summary of the Invention

[0006] In order to solve the above problems, the present invention proposes a geological disaster detection method based on image processing.

[0007] The technical solution of the present invention is: a geological disaster detection method based on image processing comprises the following steps:

[0008] S1. Acquire initial road surface image data of the area to be detected, and perform geometric correction on the initial road surface image data to obtain road surface feature image data;

[0009] S2. Using the outer edge coefficients and pixel grayscale values ​​of the spots in the road feature image data, all spots are screened and removed, and the remaining spots are used as re-detection areas;

[0010] S3. Process the re-detected area to obtain the crack area of ​​the road surface.

[0011] Furthermore, S2 includes the following sub-steps:

[0012] S21, extracting a plurality of spots from road surface feature image data;

[0013] S22, generating outer edge coefficients for each spot;

[0014] S23. Filter and remove the spots using the outer edge coefficients of the spots and the grayscale values ​​of the internal pixels to obtain a re-detection area.

[0015] Furthermore, S22 includes the following sub-steps:

[0016] S221, generating a minimum enclosing circle for each spot, and extracting four equally spaced edge points on the edge of the minimum enclosing circle;

[0017] S222, combining the grayscale values ​​of the four edge points into a spot edge feature matrix;

[0018] S224. Taking the maximum eigenvalue of the spot edge feature matrix as the spot edge coefficient.

[0019] The beneficial effect of the above further solution is that in the present invention, in S221, the four points extracted from the edge of the minimum enclosing circle should be symmetrical about the center of the circle. In S222, the expression of the spot outer edge feature matrix is: Where, Represents the grayscale value of the first edge point, Represents the grayscale value of the second edge point, Represents the grayscale value of the third edge point, The grayscale value of the fourth edge point represents the fourth edge point. In road feature image data, the presence of spots can be caused by a variety of factors, including road anomalies caused by geological disasters (such as blurred areas at crack edges, spot-like features in the image caused by localized bumps or depressions due to geological deformation, etc.). By extracting multiple spots, we can comprehensively cover areas in the image where anomalies may exist, avoiding missing potential information related to geological disasters and laying the foundation for subsequent accurate detection. Extracting spots from the image is equivalent to structuring the image data. The originally continuous image data is decomposed into discrete spot objects, each with its own characteristic information (such as position and shape). The outer edge coefficient is a quantitative indicator designed by the present invention to describe the shape characteristics of spots. It can express the shape characteristics of spots in numerical form, facilitating subsequent differentiation and screening using mathematical methods. Using both the outer edge coefficient and the grayscale value of the internal pixels of a spot for screening and elimination is a screening method that integrates multi-dimensional features. The outer edge coefficient reflects the shape characteristics of the spot, while the grayscale value of the internal pixels reflects the brightness information of the spot. For example, pavement cracks typically exhibit specific grayscale variations in images, while the grayscale values ​​of spots like noise and stains may differ from those of cracks. By combining these two characteristics, we can more accurately distinguish between spots related to geological hazards and other unrelated spots, improving screening accuracy.

[0020] Furthermore, S23 includes the following sub-steps:

[0021] S231, extracting the grayscale value of each pixel in the outer edge of the spot, and constructing a contour loss function for the spot using the outer edge coefficient;

[0022] S232: Determine whether the grayscale value of the pixel points inside the spot and the contour loss function meet the set spot screening conditions. If so, retain the spot; otherwise, remove the spot.

[0023] The beneficial effect of the above further scheme is that: in the present invention, the grayscale value of each pixel point in the outer edge of the spot is extracted to capture the brightness change information of the spot edge in detail. In road surface images related to geological disasters, the edges of abnormal areas such as cracks often have unique grayscale characteristics. For example, the edges of cracks may show a different grayscale from the surrounding road surface due to material differences or different light reflections. The contour loss function can quantify the degree of difference between the spot contour and a certain ideal shape (such as the abnormal shape caused by geological disasters). For example, if the cracks caused by geological disasters usually appear in a relatively regular or definite direction in the image, then the contour loss function can measure the degree of mismatch between the actual spot contour and this ideal shape, thereby more comprehensively describing the characteristics of the spots and improving the accuracy of screening.

[0024] Furthermore, in S231, the contour loss function of the spot The expression is:

[0025] ;

[0026] Where, represents the logarithmic function, represents the exponential function, represents the outer edge coefficient of the spot, Represents the mean grayscale value of all pixels on the outer edge of the spot.

[0027] Furthermore, in S232, the expression of the spot screening condition is set as follows: Where, Represents the mean grayscale value of the internal pixels on the outer edge of the spot, represents the contour loss function of the blob, represents the set compromise factor, represents the logarithmic function, Represents the exponential function.

[0028] The trade-off factor is between 0 and 1.

[0029] Furthermore, in S3, a dilation operation is performed on each spot in the re-detected area, and the spots after the dilation operation are used as crack areas of the road surface.

[0030] The beneficial effects of the present invention are as follows: the present invention performs geometric correction on the acquired image data, eliminating geometric distortion caused by factors such as imaging equipment, shooting angle and terrain; the spots in the road feature image data are screened and eliminated using the outer edge coefficient and pixel grayscale value of the spots. The outer edge coefficient can reflect the shape characteristics of the spots, and the grayscale value reflects the brightness information of the spots. By comprehensively considering these two factors, it is possible to effectively distinguish irrelevant spots such as noise on the road surface from areas where abnormalities may exist, avoiding interference of these irrelevant information on subsequent detection; the spots remaining after screening are used as re-detection areas, and crack areas are accurately located, which helps to timely discover early signs of geological disasters, provide key basis for subsequent disaster warning, assessment and prevention, and help to take timely and effective measures to reduce losses caused by geological disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of the geological disaster detection method based on image processing. DETAILED DESCRIPTION

[0032] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0033] like Figure 1As shown, the present invention provides a geological disaster detection method based on image processing, comprising the following steps:

[0034] S1. Acquire initial road surface image data of the area to be detected, and perform geometric correction on the initial road surface image data to obtain road surface feature image data;

[0035] S2. Using the outer edge coefficients and pixel grayscale values ​​of the spots in the road feature image data, all spots are screened and removed, and the remaining spots are used as re-detection areas;

[0036] S3. Process the re-detected area to obtain the crack area of ​​the road surface.

[0037] In this embodiment of the present invention, S2 includes the following sub-steps:

[0038] S21, extracting a plurality of spots from road surface feature image data;

[0039] S22, generating outer edge coefficients for each spot;

[0040] S23. Filter and remove the spots using the outer edge coefficients of the spots and the grayscale values ​​of the internal pixels to obtain a re-detection area.

[0041] In this embodiment of the present invention, S22 includes the following sub-steps:

[0042] S221, generating a minimum enclosing circle for each spot, and extracting four equally spaced edge points on the edge of the minimum enclosing circle;

[0043] S222, combining the grayscale values ​​of the four edge points into a spot edge feature matrix;

[0044] S224. Taking the maximum eigenvalue of the spot edge feature matrix as the spot edge coefficient.

[0045] In the present invention, in S221, the four points extracted from the edge of the minimum enclosing circle should be symmetrical about the center of the circle. In S222, the expression of the spot edge feature matrix is: Where, Represents the grayscale value of the first edge point, Represents the grayscale value of the second edge point, Represents the grayscale value of the third edge point, The grayscale value of the fourth edge point represents the fourth edge point. In road feature image data, the presence of spots can be caused by a variety of factors, including road anomalies caused by geological disasters (such as blurred areas at crack edges, spot-like features in the image caused by localized bumps or depressions due to geological deformation, etc.). By extracting multiple spots, we can comprehensively cover areas in the image where anomalies may exist, avoiding missing potential information related to geological disasters and laying the foundation for subsequent accurate detection. Extracting spots from the image is equivalent to structuring the image data. The originally continuous image data is decomposed into discrete spot objects, each with its own characteristic information (such as position and shape). The outer edge coefficient is a quantitative indicator designed by the present invention to describe the shape characteristics of spots. It can express the shape characteristics of spots in numerical form, facilitating subsequent differentiation and screening using mathematical methods. Using both the outer edge coefficient and the grayscale value of the internal pixels of a spot for screening and elimination is a screening method that integrates multi-dimensional features. The outer edge coefficient reflects the shape characteristics of the spot, while the grayscale value of the internal pixels reflects the brightness information of the spot. For example, pavement cracks typically exhibit specific grayscale variations in images, while the grayscale values ​​of spots like noise and stains may differ from those of cracks. By combining these two characteristics, we can more accurately distinguish between spots related to geological hazards and other unrelated spots, improving screening accuracy.

[0046] In this embodiment of the present invention, S23 includes the following sub-steps:

[0047] S231, extracting the grayscale value of each pixel in the outer edge of the spot, and constructing a contour loss function for the spot using the outer edge coefficient;

[0048] S232: Determine whether the grayscale value of the pixel points inside the spot and the contour loss function meet the set spot screening conditions. If so, retain the spot; otherwise, remove the spot.

[0049] In the present invention, the grayscale value of each pixel point in the outer edge of the spot is extracted to carefully capture the brightness change information of the spot edge. In road surface images related to geological disasters, the edges of abnormal areas such as cracks often have unique grayscale characteristics. For example, the edge of the crack may show a different grayscale from the surrounding road surface due to material differences or different light reflections. The contour loss function can quantify the degree of difference between the spot contour and a certain ideal shape (such as the abnormal shape caused by geological disasters). For example, if the cracks caused by geological disasters usually appear in a relatively regular or definite direction in the image, then the contour loss function can measure the degree of mismatch between the actual spot contour and this ideal shape, thereby more comprehensively describing the characteristics of the spot and improving the accuracy of screening.

[0050] In the embodiment of the present invention, in S231, the contour loss function of the spot is The expression is:

[0051] ;

[0052] Where, represents the logarithmic function, represents the exponential function, represents the outer edge coefficient of the spot, Represents the mean grayscale value of all pixels on the outer edge of the spot.

[0053] In the embodiment of the present invention, in S232, the expression of the spot screening condition is set as follows: Where, Represents the mean grayscale value of the internal pixels on the outer edge of the spot, represents the contour loss function of the blob, represents the set compromise factor, represents the logarithmic function, Represents an exponential function. The tradeoff factor is between 0 and 1.

[0054] In the embodiment of the present invention, in S3, a dilation operation is performed on each spot in the re-detected area, and the spots after the dilation operation are used as crack areas of the road surface.

[0055] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A geological disaster detection method based on image processing, characterized in that: The following steps are involved: S1. Acquire initial road surface image data of the area to be detected, and perform geometric correction on the initial road surface image data to obtain road surface feature image data; S2. Using the outer edge coefficients and pixel grayscale values ​​of the spots in the road feature image data, all spots are screened and removed, and the remaining spots are used as re-detection areas; S3, processing the re-detected area to obtain the crack area of ​​the road surface; The S2 includes the following sub-steps: S21, extracting a plurality of spots from road surface feature image data; S22, generating outer edge coefficients for each spot; S23, using the outer edge coefficient of the spot and the gray value of the internal pixel point to screen and remove the spot, and obtain a re-detection area; The S22 includes the following sub-steps: S221, generating a minimum enclosing circle for each spot, and extracting four equally spaced edge points on the edge of the minimum enclosing circle; S222, combining the grayscale values ​​of the four edge points into a spot edge feature matrix; S223, taking the maximum eigenvalue of the spot outer edge feature matrix as the spot outer edge coefficient; The S23 includes the following sub-steps: S231, extracting the grayscale value of each pixel in the outer edge of the spot, and constructing a contour loss function for the spot using the outer edge coefficient; S232, determining whether the grayscale value of the pixel points inside the spot and the contour loss function meet the set spot screening conditions, if so, retain the spot, otherwise remove the spot; In S231, the contour loss function of the spot The expression is: ； Where, represents the logarithmic function, represents the exponential function, represents the outer edge coefficient of the spot, Represents the mean grayscale value of all pixels on the outer edge of the spot; In the step S232, the expression of the spot screening condition is set as follows: Where, Represents the mean grayscale value of the internal pixels on the outer edge of the spot, represents the contour loss function of the blob, represents the set compromise factor, represents the logarithmic function, Represents the exponential function.

2. The geological disaster detection method based on image processing according to claim 1, characterized in that: In S3, a dilation operation is performed on each spot in the re-detected area, and the spots after the dilation operation are used as crack areas of the road surface.

Images