Intelligent identification method for crushing degree of tunnel surrounding rock

Abstract

The invention discloses a tunnel surrounding rock breaking degree intelligent identification method, and relates to the technical field of geological engineering and image processing, and the method comprises the steps: image preprocessing, including noise removal and image enhancement; feature extraction, including edge detection; the image segmentation comprises threshold segmentation; feature analysis, including crack analysis, contrast analysis and homogeneity analysis; the method has the beneficial effects that the identification efficiency and accuracy can be effectively improved, and an important guarantee is provided for safe construction of tunnel engineering.

Description

Technical Field

[0001] The invention relates to the technical field of geological engineering and image processing, and in particular to a method for intelligently identifying the degree of fragmentation of surrounding rocks of a tunnel. Background Art

[0002] During the construction of tunnel projects, the stability of surrounding rock has an important impact on the safety and quality of the project. The degree of surrounding rock fragmentation is directly related to the stability and construction safety of the tunnel. The traditional method of identifying the degree of surrounding rock fragmentation mainly relies on manual observation and experience judgment, and has the following problems: 1. High subjectivity: The manual identification method relies on the experience of geological engineers. The judgment results of different personnel may vary greatly and lack consistency.

[0003] 2. Inefficiency: Manual identification requires a lot of on-site investigation and manual analysis, which is time-consuming and labor-intensive and cannot meet the needs of rapid construction.

[0004] 3. Lack of accuracy: Due to the limited resolution ability of the human eye, manual identification is difficult to accurately identify tiny cracks and broken features, which may lead to misjudgment and missed judgment.

[0005] With the development of computer technology and image processing technology, the automatic identification method based on image processing has gradually received attention. Through image processing technology, the degree of surrounding rock fragmentation can be quickly and accurately identified, and the efficiency and accuracy of identification can be improved. However, when the existing image processing methods are applied to the identification of the degree of surrounding rock fragmentation, there are still some technical difficulties, such as low image quality and difficulty in feature extraction. Therefore, there is an urgent need for an effective image processing method that can accurately extract the fragmentation characteristics of tunnel surrounding rocks and realize intelligent identification. Summary of the invention

[0006] The purpose of the present invention is to provide a method for intelligently identifying the degree of fragmentation of tunnel surrounding rocks in order to solve the above-mentioned problems. Through image processing technology, the degree of fragmentation of tunnel surrounding rocks is automatically and intelligently identified, thereby improving the accuracy and efficiency of the identification.

[0007] The present invention achieves the above-mentioned purpose through the following technical solutions: A method for intelligently identifying the degree of crushing of surrounding rock of a tunnel comprises the following steps: Image preprocessing: including noise removal and image enhancement; Feature extraction: including edge detection; Image segmentation: including threshold segmentation; Feature analysis: including crack analysis, contrast analysis and homogeneity analysis.

[0008] Preferably, the noise removal uses Gaussian filtering in a discretized window sliding window convolution manner.

[0009] Preferably: the image enhancement uses histogram equalization, which specifically includes image segmentation, local histogram calculation, local histogram equalization and interpolation processing.

[0010] Preferably, the edge detection uses a Canny edge detection algorithm.

[0011] Preferably, the image segmentation uses a threshold segmentation method, and a threshold is set according to the gray value of the image to segment the image into a broken area and a non-broken area.

[0012] Preferably: the crack analysis includes calculating parameters such as length, width and density of the cracks to quantify the degree of fragmentation; the contrast analysis includes calculating the contrast of the processed image data, which reflects the difference in pixel values ​​in the image; and calculating the homogeneity of the processed image data, which reflects the consistency of the grayscale in the image.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The intelligent identification method of the degree of crushing of tunnel surrounding rocks provided by the present invention realizes the automatic and intelligent identification of the degree of crushing of tunnel surrounding rocks through image processing technology, improves the accuracy and efficiency of identification, and has the following advantages compared with the traditional manual identification method: Automation and intelligence: Through image processing algorithms and deep learning models, the degree of surrounding rock fragmentation can be automatically identified, reducing manual intervention and improving work efficiency; High accuracy: Using a variety of image processing technologies and deep learning models, it can accurately extract and analyze the crushing characteristics of the surrounding rock and provide reliable identification results; Strong consistency: The recognition results of the algorithm and model are highly consistent, avoiding the subjectivity and differences in manual recognition; Real-time: By optimizing algorithms and models, the degree of rock fragmentation of the tunnel surrounding rock can be identified in real time, meeting the real-time monitoring needs of engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative labor.

[0015] Figure 1 It is a processing diagram of a method for intelligently identifying the degree of crushing of tunnel surrounding rocks described in the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings: Example 1

[0017] A method for intelligently identifying the degree of crushing of surrounding rock of a tunnel comprises the following steps: Step S1: Image preprocessing: 1.1 Noise Removal: Use Gaussian filtering to remove noise from the image, ensure image quality, and improve image clarity and contrast; reduce random noise in the image, enhance image smoothness, and lay the foundation for subsequent processing; after using Gaussian filtering, the high-frequency information of the image (such as edges, noise, etc.) will be weakened, making the image look smoother, but if the kernel is too large, it may also make the image too blurred and reduce its clarity. This application uses a medium kernel size of 7×7; 1.2 Image Enhancement: Adaptive histogram equalization is used to improve the contrast of the image, making the details of the image more visible, highlighting the detailed features of the surrounding rock, facilitating subsequent processing, and enhancing the local contrast of the image; compared with traditional histogram equalization, adaptive histogram equalization can more effectively enhance image details, especially in some images with shadows or uneven brightness; its working principle is: in adaptive histogram equalization, the image is divided into multiple small blocks, the histogram of each small block is calculated and equalized, and finally the difference processing is performed; (1) Image segmentation First, adaptive histogram equalization divides the image into several small blocks (also called "context areas" or "windows"). These small blocks are local areas of the image and are usually of the same size. The granularity of the division (i.e., the size of the small blocks) can be adjusted as needed. Select block size: The size of the block directly affects the effect of adaptive histogram equalization. Smaller blocks can enhance local contrast more finely, but may introduce more noise. Larger blocks tend to smooth the process and the effect is close to global equalization. (2) Local histogram calculation For each small block, adaptive histogram equalization will calculate the histogram of the pixel values ​​in the area separately. The histogram represents the frequency distribution of pixel values ​​of various gray levels in the small block. Calculate the local histogram: count the pixels in each small block and generate the distribution of its gray level, i.e. histogram; (3) Local histogram equalization After obtaining the local histogram, adaptive histogram equalization performs an equalization operation on the histogram. The purpose of histogram equalization is to redistribute the grayscale values ​​of pixels so that the grayscale distribution of pixels in the small block is more uniform, thereby enhancing the contrast. Equalization: The local histogram is converted into a uniform distribution through the cumulative distribution function (CDF), and the grayscale values ​​of the pixels are redistributed accordingly; (4) Interpolation processing Since each small block is processed independently, directly splicing them may form obvious boundaries between small blocks, causing the image to look discontinuous; in order to eliminate these block effects, adaptive histogram equalization uses interpolation processing; Boundary smoothing: For each pixel, adaptive histogram equalization not only considers the equalization results of the small block where it is located, but also considers the equalization results of its neighboring small blocks. The final pixel value is obtained by weighted averaging these results. This interpolation method can smoothly transition the differences between different small blocks and reduce the blocking effect. Step S2: Feature extraction: In the edge detection of feature extraction, the Canny edge detection algorithm is used to extract the edges in the image and identify the boundaries of cracks and broken areas. The Canny algorithm locates the significant edges in the image by calculating the gradient change of the pixel gray value; Step S3: Image segmentation: Image segmentation uses the image threshold segmentation method: setting a threshold according to the gray value of the image to segment the image into broken areas and non-broken areas. Threshold segmentation is a simple and effective method suitable for situations where image features are obvious. There are many threshold segmentation methods. This application uses an adaptive threshold segmentation method. Step S4: Feature analysis: The crack parameters, contrast and homogeneity of the rock are calculated from the processed images, and combined with the parameter boundary criteria of intact rock and broken rock, an analysis method for intelligently identifying the degree of surrounding rock fragmentation is obtained; 4.1 Crack analysis Calculate the length, width and density of cracks to quantify the degree of fragmentation. Fracture analysis can provide more detailed fragmentation information. The more cracks there are, the longer and wider they are, the higher the degree of fragmentation of the rock. 4.2 Contrast Analysis Calculate the contrast of the processed image data. The contrast reflects the difference in pixel values ​​in the image; it describes the grayscale difference between adjacent pixels. The larger the value, the rougher the texture of the image and the greater the grayscale difference. If the contrast is high, it usually means that there are obvious edges or boundaries in the image, which means that the rock is more broken. 4.3 Homogeneity analysis The homogeneity of the image data after calculation and processing. Homogeneity reflects the consistency of the grayscale in the image. The larger its value is, the more uniform the texture of the image is and the smaller the grayscale variation is. High homogeneity usually means that the regional variation in the image is smaller and the texture is smoother, which means that the degree of rock fragmentation is lower.

[0018] The Gaussian filtering in step S1 uses a discretized window sliding window convolution method.

[0019] The Gaussian template and the Gaussian template parameter calculation formula are existing well-known technologies and will not be described in detail here.

[0020] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for illustrating the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An intelligent identification method for the degree of crushing of tunnel surrounding rock, characterized by: The following steps are involved: Image preprocessing: including noise removal and image enhancement; Feature extraction: including edge detection; Image segmentation: including threshold segmentation; Feature analysis: including crack analysis, contrast analysis and homogeneity analysis.

2. According to claim 1, a method for intelligently identifying the degree of crushing of surrounding rock of a tunnel is characterized by: The noise removal uses Gaussian filtering in a discretized window sliding window convolution manner.

3. The method for intelligently identifying the degree of crushing of surrounding rock of a tunnel according to claim 1 is characterized by: The image enhancement uses histogram equalization, which specifically includes image segmentation, local histogram calculation, local histogram equalization and interpolation processing.

4. The method for intelligently identifying the degree of crushing of surrounding rock of a tunnel according to claim 1 is characterized in that: The edge detection uses the Canny edge detection algorithm.

5. The method for intelligently identifying the degree of crushing of surrounding rock of a tunnel according to claim 1 is characterized by: The image segmentation uses a threshold segmentation method, sets a threshold according to the gray value of the image, and divides the image into a broken area and a non-broken area.

6. The method for intelligently identifying the degree of crushing of surrounding rock of a tunnel according to claim 1 is characterized by: The crack analysis includes calculating parameters such as the length, width and density of the cracks to quantify the degree of fragmentation; the contrast analysis includes calculating the contrast of the processed image data, which reflects the difference in pixel values ​​in the image; and calculating the homogeneity of the processed image data, which reflects the consistency of the grayscale in the image.

Images