Geological radar forecast waveform chart index quantization analysis method based on pixel recognition

Abstract

The application discloses a kind of geological radar forecast waveform chart index quantization analysis methods based on pixel identification, comprising the following steps: S1, cutting geological radar advance geological forecast waveform chart;S2, point by point index each pixel point RGB value of cutting after waveform chart, calculate the RGB sum of each pixel point, segmented statistics each pixel point RGB sum distribution;S3, determine RGB gain threshold, and the pixel point of RGB sum less than or equal to RGB gain threshold is carried out target color gain;S4, determine the coordinate of each channel of waveform chart;S5, calculate the amplitude value of each wave crest, frequency;S6, according to preset in-phase axis continuity rate, calculate the continuity of each wave crest of each channel;S7, to waveform chart partition, calculate waveform chart energy attenuation rate.The present application carries out cutting, gain and other pretreatment to geological radar forecast waveform chart using pixel identification method, eliminates interference factor, and it is convenient for index quantization analysis.The quantization result of the index of radar forecast waveform chart is automatically calculated.

Description

Technical Field

[0001] This invention relates to the field of advanced geological prediction for tunnels. More specifically, this invention relates to a method for quantitative analysis of indicators in ground-penetrating radar prediction waveforms based on pixel recognition. Background Technology

[0002] Ground-penetrating radar (GPR), as a short-range advanced forecasting method in tunnel excavation, has the advantages of speed and convenience. However, in practice, GPR forecasts are often analyzed using commercial software to generate waveform diagrams. The quality of these waveform diagram interpretations determines the accuracy of the geological conditions of the surrounding rock ahead of the tunnel face, providing guidance for tunnel excavation. Currently, the interpretation of radar waveform diagrams generally relies on the experience of forecasters, qualitatively describing the representational degree of various indicators in the waveform diagrams to determine the geological conditions ahead of the tunnel face. Due to the varying experience of forecasters in interpreting waveform diagrams, the forecast conclusions often contain significant subjective errors, frequently resulting in discrepancies between the forecast conclusions and the exposed surrounding rock. This can seriously lead to geological disasters and affect construction progress.

[0003] To address the aforementioned issues, a pixel-based method for quantitative analysis of radar forecast waveform indicators is proposed. This method avoids human experience errors, improves the accuracy of advanced geological forecasts, and provides reasonable guidance for tunnel construction. Summary of the Invention

[0004] To achieve these objectives and other advantages according to the present invention, a preferred embodiment of the present invention provides a method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition, comprising the following steps: S1. Trim the waveform diagram of advanced geological prediction by ground-penetrating radar; S2. Index the RGB values ​​of each pixel in the cropped waveform image point by point, sum the R, G, and B values ​​of each pixel to obtain the RGB sum value, and statistically analyze the distribution of the RGB sum value of each pixel in the whole image segment by segment. S3. Based on the actual display effect of the waveform, determine the RGB gain threshold, and apply target color gain to pixels whose RGB sum values ​​are less than or equal to the RGB gain threshold in the waveform, and change pixels with values ​​greater than the RGB gain threshold to white pixels. S4. Perform a pixel traversal search on the gained waveform to determine the coordinates of each channel of the waveform; S5. Search for the starting point and ending point of each wave peak one by one, record the coordinates, and calculate the amplitude and frequency of each wave peak. S6. Calculate the continuity of each peak in each channel according to the preset phase axis continuity rate. S7. Divide the waveform into regions and calculate the energy decay rate of the waveform.

[0005] Preferably, step S1 specifically includes the following steps: Redundant pixels were removed from the original waveform image of the ground-penetrating radar forecast, retaining only all pixels of each track. A coordinate system was constructed with the upper left corner of the cropped waveform image as the origin and the direction of the survey line length as the horizontal direction. x The axis is perpendicular to the direction of the probe depth. y The coordinate system of the axes.

[0006] Preferably, step S2 specifically includes the following steps: performing pixel-by-pixel indexing on the cropped waveform image to obtain the pixel value of each point, i.e., the RGB value; calculating the sum of the R, G, and B values ​​of each pixel point to obtain the RGB sum value; dividing the RGB sum value of each pixel point into segments for statistical analysis, specifically the segment intervals 765-688, 687-612, 611-535, 534-459, 458-382, 381-306, 305-229, 228-153, 152-76, and 75-0; and calculating the proportion of the RGB sum value of each pixel point in each interval.

[0007] Preferably, step S3 specifically includes the following steps: based on the statistical results of step S2, select a target range in combination with the actual display effect of the waveform, and determine the upper limit of the target range as the RGB gain threshold; for pixels in the entire waveform whose RGB sum value is less than or equal to the RGB gain threshold, change the gain to the target color.

[0008] Preferably, step S4 specifically includes: using a pixel traversal search method in the coordinate system to search for the distribution of pixels on the x-axis along the y-axis direction; terminating the traversal search when 10-200 target color pixels are found in the x-axis direction, or when the waveform diagram requires no less than 10 channels and no more than 200 channels; recording the coordinates of each target color pixel as the x-axis coordinate value of each channel; if the width of a channel is not 1 pixel, then the position of the rightmost pixel in the positive x-axis direction of that channel is identified as the x-axis coordinate of that channel.

[0009] Preferably, step S5 specifically includes: based on the coordinates determined in step S4, performing a pixel traversal search along the y-axis for each channel; when the i-th pixel is white and the (i+1)-th pixel is the target color, the (i+1)-th pixel is identified as the starting point of the corresponding peak; when the i-th pixel is the target color and the (i+1)-th pixel is white, the i-th pixel is identified as the ending point of the corresponding peak; searching for the maximum width of the target color in the x-axis direction within the interval from the starting point to the ending point of each peak, and using this maximum width as the amplitude value of the peak; the frequency of each peak is calculated using the following formula: in, This is the y-coordinate value of the endpoint of the nth peak of the mth wave. ε is the y-coordinate of the starting point of the nth peak of the mth channel; ε is the dielectric constant of the rock mass.

[0010] Preferably, step S6 specifically includes: based on radar detection accuracy and engineering experience, pre-setting the allowable error Δy in depth between adjacent peaks; for each peak, determining whether there are other peaks within the vertical direction y±Δy of adjacent channels; if they exist, determining that the in-phase axis corresponding to the peak is continuous; if they do not exist, determining that the in-phase axis corresponding to the peak is discontinuous.

[0011] Preferably, step S7 specifically includes: dividing the waveform into several regions according to the characteristics of the waveform; calculating the cumulative amplitude of all peaks in each region; calculating the energy density of each region, i.e., the ratio of the cumulative amplitude to the area of ​​the region; comparing the energy density of different regions in the vertical direction, and obtaining the energy attenuation of the waveform based on the comparison results.

[0012] The present invention has at least the following beneficial effects: I. This invention utilizes pixel recognition methods to perform preprocessing such as cropping and gain adjustment on the waveform image of ground-penetrating radar forecasts, eliminating interference factors and facilitating quantitative analysis of indicators.

[0013] II. This invention identifies the degree of characterization of various indicators in radar forecast waveform diagrams.

[0014] III. The quantitative results of various indicators of the radar forecast waveform diagram automatically calculated by this invention.

[0015] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the steps of the pixel-based ground-penetrating radar forecast waveform index quantification analysis method in this invention.

[0017] Figure 2 This is a schematic diagram of the waveform in S1 of the present invention.

[0018] Figure 3 This is a schematic diagram of pixels in S5 of the present invention. Detailed Implementation

[0019] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0020] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0021] Those skilled in the art should understand that, in the disclosure of this invention, the terms "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the above terms should not be construed as limiting this invention.

[0022] It is understood that the term "a" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0023] like Figure 1-3 As shown, a preferred embodiment of the present invention provides a method for quantitative analysis of indicators in ground-penetrating radar forecast waveforms based on pixel recognition, comprising the following steps: S1. Trim the waveform diagram of advanced geological prediction by ground-penetrating radar; Specifically, the following steps are included: Redundant pixels were removed from the original waveform image of the ground-penetrating radar forecast, retaining only all pixels of each track. A coordinate system was constructed with the upper left corner of the cropped waveform image as the origin and the direction of the survey line length as the horizontal direction. x The axis is perpendicular to the direction of the probe depth. y For details on the coordinate system of the axes, please refer to [link / reference]. Figure 2 Remove redundant pixels such as borders, redundant annotations, and blank areas, and retain only all pixels corresponding to each waveform that can reflect the geological signal. This ensures that subsequent analysis is carried out only on effective geological information and avoids interference from irrelevant pixels.

[0024] S2. Index the RGB values ​​of each pixel in the cropped waveform image point by point, sum the R, G, and B values ​​of each pixel to obtain the RGB sum value, and statistically analyze the distribution of the RGB sum value of each pixel in the whole image segment by segment. Specifically, the following steps are included: The cropped waveform is indexed pixel by pixel to obtain the pixel value (RGB value) of each point. The sum of the R, G, and B values ​​of each pixel is calculated to obtain the RGB sum value. The RGB sum value of each pixel is then segmented and statistically analyzed in the following intervals: 765-688, 687-612, 611-535, 534-459, 458-382, 381-306, 305-229, 228-153, 152-76, and 75-0. The proportion of the RGB sum value of each pixel in each interval is then calculated.

[0025] S3. Based on the actual display effect of the waveform, determine the RGB gain threshold, and apply target color gain to pixels whose RGB sum values ​​are less than or equal to the RGB gain threshold in the waveform, and change pixels with values ​​greater than the RGB gain threshold to white pixels. Specifically, the following steps are included: Based on the statistical results of step S2, a target range is selected in conjunction with the actual display effect of the waveform. The upper limit of the target range is determined as the RGB gain threshold. For pixels in the entire waveform whose RGB sum value is less than or equal to the RGB gain threshold, the gain is changed to the target color.

[0026] S4. Perform a pixel traversal search on the gained waveform to determine the coordinates of each channel of the waveform; Specifically, it includes: A pixel-based traversal search method is used to search for the distribution of pixels along the x-axis along the y-axis. When 10-200 target color pixels are found along the x-axis, or when the waveform diagram has at least 10 and no more than 200 channels, the traversal search is terminated. The coordinates of each target color pixel are recorded as the x-axis coordinates of each channel. If the width of a channel is not 1 pixel, the position of the rightmost pixel along the positive x-axis of that channel is taken as the x-axis coordinate of that channel.

[0027] S5. Search for the starting point and ending point of each wave peak one by one, record the coordinates, and calculate the amplitude and frequency of each wave peak. Specifically, it includes: Based on the coordinates determined in step S4, a pixel traversal search is performed along the y-axis for each channel. When the i-th pixel is white and the (i+1)-th pixel is the target color pixel, the (i+1)-th pixel is identified as the starting point of the corresponding peak; when the i-th pixel is the target color pixel and the (i+1)-th pixel is white, the i-th pixel is identified as the ending point of the corresponding peak. See details. Figure 3 The maximum width of the target color along the x-axis is searched sequentially within the interval from the start to the end of each peak, and this maximum width is taken as the amplitude value of that peak; the frequency of each peak is calculated using the following formula: in, This is the y-coordinate value of the endpoint of the nth peak of the mth wave. ε is the y-coordinate of the starting point of the nth peak of the mth channel; ε is the dielectric constant of the rock mass.

[0028] S6. Calculate the continuity of each peak in each channel according to the preset phase axis continuity rate. Specifically, it includes: Based on radar detection accuracy and engineering experience, the allowable error Δy in depth between adjacent peaks is preset; for each peak, it is determined whether there are other peaks within the vertical direction y±Δy of the adjacent channel; if they exist, the in-phase axis corresponding to the peak is determined to be continuous; if they do not exist, the in-phase axis corresponding to the peak is determined to be discontinuous.

[0029] S7. Divide the waveform into regions and calculate the energy decay rate of the waveform.

[0030] Specifically, it includes: Based on the characteristics of the waveform, the waveform is divided into several regions; the cumulative amplitude of all peaks in each region is calculated; the energy density of each region is calculated, which is the ratio of the cumulative amplitude to the area of ​​the region; in the vertical direction, the relationship between the energy densities of different regions is compared, and the energy attenuation of the waveform is obtained based on the comparison results.

[0031] This invention, through the implementation of the above seven steps, achieves quantitative analysis of key indicators in ground-penetrating radar (GPR) forecast waveforms. This method breaks away from the traditional reliance on manual experience for qualitative judgment. This invention utilizes pixel recognition to perform preprocessing such as cropping and gain adjustment on the GPR forecast waveforms, eliminating interference factors and facilitating quantitative analysis of indicators. Furthermore, it automatically calculates quantitative indicators such as the continuity rate of the in-phase axis, frequency, amplitude, and energy attenuation, accurately reflecting key information such as the geological stratum continuity and medium properties of the surrounding rock ahead of the tunnel face. This helps reduce the risk of geological disasters and ensures the smooth progress of construction.

[0032] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A method for quantitative analysis of indicators in ground-penetrating radar forecast waveforms based on pixel recognition, characterized in that, Includes the following steps: S1. Trim the waveform diagram of advanced geological prediction by ground-penetrating radar; S2. Index the RGB values ​​of each pixel in the cropped waveform image point by point, sum the R, G, and B values ​​of each pixel to obtain the RGB sum value, and statistically analyze the distribution of the RGB sum value of each pixel in the whole image segment by segment. S3. Based on the actual display effect of the waveform, determine the RGB gain threshold, and apply target color gain to pixels whose RGB sum values ​​are less than or equal to the RGB gain threshold in the waveform, and change pixels with values ​​greater than the RGB gain threshold to white pixels. S4. Perform a pixel traversal search on the gained waveform to determine the coordinates of each channel of the waveform; S5. Search for the starting point and ending point of each wave peak one by one, record the coordinates, and calculate the amplitude and frequency of each wave peak. S6. Calculate the continuity of each peak in each channel according to the preset phase axis continuity rate. S7. Divide the waveform into regions and calculate the energy decay rate of the waveform.

2. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 1, characterized in that, Step S1 specifically includes the following steps: Redundant pixels were removed from the original waveform image of the ground-penetrating radar forecast, retaining only all pixels of each track. A coordinate system was constructed with the upper left corner of the cropped waveform image as the origin and the direction of the survey line length as the horizontal direction. x The axis is perpendicular to the direction of the probe depth. y The coordinate system of the axes.

3. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 1, characterized in that, Step S2 specifically includes the following steps: Indexing each pixel of the cropped waveform image to obtain the pixel value (RGB value) of each point; calculating the sum of the R, G, and B values ​​of each pixel to obtain the RGB sum value; segmenting the RGB sum value of each pixel into intervals: 765-688, 687-612, 611-535, 534-459, 458-382, 381-306, 305-229, 228-153, 152-76, 75-0; and calculating the proportion of the RGB sum value of each pixel in each interval.

4. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 2, characterized in that, Step S3 specifically includes the following steps: Based on the statistical results of step S2, select the target interval in combination with the actual display effect of the waveform, and determine the upper limit of the target interval as the RGB gain threshold; for pixels in the entire waveform whose RGB sum value is less than or equal to the RGB gain threshold, change the gain to the target color.

5. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 1, characterized in that, Step S4 specifically includes: using a pixel traversal search method in the coordinate system to search for the distribution of pixels on the x-axis along the y-axis direction; terminating the traversal search when 10-200 target color pixels are found in the x-axis direction, or when the waveform diagram requires no less than 10 channels and no more than 200 channels; recording the coordinates of each target color pixel as the x-axis coordinate value of each channel; if the width of a channel is not 1 pixel, then the position of the rightmost pixel along the positive x-axis direction of that channel is identified as the x-axis coordinate of that channel.

6. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 5, characterized in that, Step S5 specifically includes: Based on the coordinates determined in step S4, performing a pixel traversal search along the y-axis for each channel; when the i-th pixel is white and the (i+1)-th pixel is the target color, the (i+1)-th pixel is identified as the starting point of the corresponding peak; when the i-th pixel is the target color and the (i+1)-th pixel is white, the i-th pixel is identified as the ending point of the corresponding peak; searching for the maximum width of the target color along the x-axis within the interval from the starting point to the ending point of each peak, and using this maximum width as the amplitude value of the peak; the frequency of each peak is calculated using the following formula: in, This is the y-coordinate value of the endpoint of the nth peak of the mth wave. ε is the y-coordinate of the starting point of the nth peak of the mth channel; ε is the dielectric constant of the rock mass.

7. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 1, characterized in that, Step S6 specifically includes: based on radar detection accuracy and engineering experience, preset the allowable error Δy in depth between adjacent peaks; for each peak, determine whether there are other peaks within the vertical direction y±Δy of the adjacent channel; if they exist, determine that the in-phase axis corresponding to the peak is continuous; if they do not exist, determine that the in-phase axis corresponding to the peak is discontinuous.

8. The method for quantitative analysis of ground-penetrating radar forecast waveform indicators based on pixel recognition according to claim 1, characterized in that, Step S7 specifically includes: dividing the waveform into several regions based on its characteristics; calculating the cumulative amplitude of all peaks in each region; calculating the energy density of each region, i.e., the ratio of the cumulative amplitude to the area of ​​that region; comparing the energy density of different regions in the vertical direction, and determining the energy attenuation of the waveform based on the comparison results.

Images