Recognition and early warning method for hidden water body in front of construction tunnel

Abstract

The invention provides a recognition and early warning method for a hidden water body in front of a construction tunnel, and the method comprises the steps: S100, collecting an infrared thermogram of a tunnel face, and carrying out the filtering processing of the infrared thermogram; s200, inputting the filtered infrared thermogram into a mask-rcnn neural network subjected to depth separable convolution optimization for processing, judging whether water leakage exists on the tunnel face or not, and if the water leakage exists, calculating the real area of a water seepage region; s300, if it is judged that no leakage water exists, the temperature and temperature difference rule on the tunnel face is analyzed, whether a hidden water body exists in front of the tunnel face or not is judged based on the tunnel face temperature distribution rule, and if yes, the position size of the hidden water body is calculated; and S400, outputting analysis and judgment results of the leakage water and the hidden water body in front of the tunnel face. According to the method, the calculation efficiency can be improved, and the purpose of quickly detecting whether water leakage exists in front of the tunnel face in real time is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of tunnel water leakage detection, and in particular to a method for identifying and warning a hidden water body in front of a construction tunnel. Background Art

[0002] At present, tunnel leakage identification is mostly for lining and concrete leakage identification, and its purpose is to monitor the safety of tunnel operation. However, during tunnel construction, the water content of rock is highly correlated with rock strength and failure form. The use of non-destructive methods to monitor the water content in tunnel surrounding rock is extremely important for evaluating construction safety. At the same time, since the construction interval for non-destructive testing during tunnel construction is short, in order to achieve the construction optimization work in a limited time, it is necessary to ensure the computational efficiency of the optimization algorithm. Summary of the invention

[0003] The purpose of the present invention is to provide a method for identifying and warning hidden water bodies in front of a construction tunnel, thereby improving the computational efficiency of tunnel water leakage detection and achieving the purpose of real-time detection.

[0004] To achieve the above-mentioned purpose, the present invention provides a method for identifying and warning a hidden water body in front of a construction tunnel, the method comprising:

[0005] S100, collecting infrared thermal images of the tunnel face, and performing filtering processing on the infrared thermal images;

[0006] S200, inputting the infrared thermal image after filtering into the mask-rcnn neural network optimized by deep separable convolution for processing, determining whether there is water leakage on the tunnel face, and if there is water leakage, calculating the real area of ​​the water leakage area;

[0007] S300, if it is determined that there is no water leakage, then analyzing the temperature and temperature difference law on the tunnel face, and judging whether there is a hidden water body in front of the tunnel face based on the temperature distribution law of the tunnel face, and if there is a hidden water body, calculating the position and size of the hidden water body;

[0008] S400: Output the analysis and judgment results of water leakage and hidden water bodies in front of the tunnel face.

[0009] Furthermore, the infrared thermal image is filtered, specifically including:

[0010] S101, select a pixel point q from the infrared thermal image pixel Q i , query all neighboring points N through the neighbor index k (q i );

[0011] S102, calculate pixel point q i The normal vector n i;

[0012] S103, set the initial value of weight coefficient ∑ φ =0, and the initial value of weight factor λ=0;

[0013] S104, calculate pixel point q i and its neighboring point p i The distance ‖p i -q i ‖, and the normal distance <n i ,p i -q i >

[0014] S105: Update parameters λ=λ+φ <n i ,p i -q i >,∑ φ =∑ φ +φ;

[0015] σ d Indicates q i To its nearest neighbor point p i Distance influence weight factor, σ n Indicates q i With p i The distance vector is the normal vector at this point n i The projection on q i The influence weight factor of , φ represents the Gaussian kernel function;

[0016] S106, update pixel point q i location;

[0017] S107, repeat S101 to S106 to traverse all pixel points in the infrared thermal image atlas to complete the filtering process.

[0018] Furthermore, step S200 specifically includes:

[0019] S201, inputting the infrared thermal image after filtering into the backbone network, wherein the backbone network is composed of Depthwise convolution and Pointwise convolution to generate a feature map containing temperature field and rock texture information;

[0020] S202, based on the feature map generated in S202, the region proposal network generates a series of pre-designed bounding boxes that may contain water seepage areas, and arranges the proposed bounding boxes on the feature map to a more accurate position through the RoI Align operation;

[0021] S203, passing the fixed feature map to the normalized exponential function and the bounding box regression layer in parallel through the fully connected layer of the network to determine whether there is water seepage in the area, and giving different moisture content levels and the outer frame positions of the water seepage position respectively. At the same time, the fully convolutional network is used to generate a binary mask to segment the precise position of the water seepage from the background;

[0022] S204, after the binary mask is generated, the real area of ​​the tunnel surrounding rock corresponding to the infrared thermal image is calculated based on the shooting distance, and then the real area of ​​the seepage area is further calculated according to the ratio of the number of pixels of the binary mask to the number of pixels of the entire infrared thermal image.

[0023] Furthermore, step S300 specifically includes:

[0024] S301, establishing a rectangular coordinate system with the center of the face as the origin, the face height direction as the y-axis, and the face width direction as the x-axis;

[0025] S302, determining the location of the hidden water body by analyzing the temperature of the tunnel face and the temperature difference distribution law;

[0026] S303, calculating the location and size of the hidden water body.

[0027] Furthermore, the location of hidden water bodies can be determined by analyzing the temperature and temperature difference distribution of the tunnel face, including:

[0028] S401. When the distance between the tunnel face and the hidden water body is r, assuming that the hidden water body is directly in front of the tunnel face at y=m, the temperature T of any point on the tunnel face is calculated as follows:

[0029] T=0.5C1ln[(|ym|-r0) 2 +r 2 ]+C2

[0030] In the above formula, r0 represents the size of the hidden water body, C1 and C2 are constants related to the properties of the surrounding rock and water, and m represents the mapping position parameter of the hidden water body on the tunnel face;

[0031] S402, judging the location of the hidden water body according to the temperature at different locations on the tunnel face, specifically:

[0032] When there is a hidden water body in front of the tunnel, if the temperature of the hidden water body is higher than the temperature of the surrounding rock, the minimum temperature on the tunnel face appears at the boundary of the tunnel face, and the position of the maximum temperature corresponds to the position of the hidden water body.

[0033] When the temperature of the hidden water body is lower than the temperature of the tunnel face, the maximum temperature on the tunnel face appears at the boundary of the tunnel face, and the position of the minimum temperature corresponds to the position of the hidden water body.

[0034] If both the maximum and minimum values of the temperature on the heading face appear at the boundary of the heading face, further analyze the temperature gradient of the heading face.

[0035] Further, analyze the temperature gradient of the heading face, specifically including:

[0036] S501. Calculate the temperature gradient of it in the y direction based on the calculation formula of temperature T. The corresponding calculation formula is:

[0037]

[0038] Based on the above formula, it can be seen that when y > m, the gradient of the temperature field increases with the increase of y; when y < m, the gradient of the temperature field increases with the decrease of y.

[0039] S502. Judge the existence and orientation of the hidden water body according to the temperature gradient distribution laws I and II of the heading face. Law I is that the temperature gradient values on the heading face are symmetrically distributed with the position where the hidden water body exists (y = m) as the axis of symmetry; Law II is that on the heading face, the closer to the orientation where the hidden water body exists, the smaller the temperature gradient value, and the farther away from the orientation where the hidden water body exists, the larger the temperature gradient value; since the temperature difference at the same interval along the y-axis direction on the heading face is the accumulation of the temperature gradient, its temperature difference also follows this law.

[0040] After determining the orientation of the hidden water body, further analyze the relationship between the maximum temperature difference on the heading face and the position and size of the hidden water body. The maximum temperature difference ΔT on the heading face max The calculation formula is:

[0041]

[0042] In the above formula, Ts represents the normal temperature of the surrounding rock, T w represents the temperature of the hidden water body, k represents the thermal conductivity, h1 represents the heat convection coefficient of water, H represents the size of the heading face, r max represents the distance from the center of the water body when the temperature field of the surrounding rock returns to normal;

[0043] S504. Judge whether there is a hidden water body based on the distribution law of the maximum temperature difference on the heading face.

[0044] Further, the distribution laws of the maximum temperature difference on the heading face include:

[0045] When the size of the hidden water body in front of the heading face exceeds the size of the heading face, the temperature fields superimposed at different positions on the same heading face by the hidden water body are the same, and at this time, the temperature difference on the heading face is 0;

[0046] Since when there is a hidden water body in front of the heading face, there is always at least one temperature extreme value on the boundary of the heading face, the maximum temperature difference ΔT on the heading face maxIt is positively correlated with the tunnel face size H. The larger H is, the greater the maximum temperature difference on the tunnel face;

[0047] When there is a hidden water body in front of the tunnel face, as the distance r from the hidden water body decreases, the maximum temperature difference ΔT on the tunnel face increases. max gradually increase;

[0048] When other conditions are constant, the size of the hidden water body r0 in front of the tunnel face and the maximum temperature difference ΔT on the tunnel face max Inversely proportional.

[0049] Furthermore, the calculation of the location and size of the hidden water body specifically includes:

[0050] S601. Analyze the relative positions of the tunnel face and the detection points distributed along the axial direction and establish a mathematical model as shown in the following formula:

[0051]

[0052] In the above formula, Δr represents the axial spacing, ΔT 轴向 Indicates the maximum temperature difference under Δr spacing;

[0053] S602, solving the mathematical model to obtain the distance r between the tunnel face and the hidden water body and its size r0.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] The present invention provides a method for identifying and warning hidden water bodies in front of a construction tunnel. To ensure accuracy, a normal-based bilateral filtering method is used to process infrared thermal images to enhance the intensity of the boundary between the water-containing area and the dry area on the image. In terms of detection efficiency, a mask rcnn neural network structure is integrated with a deep separable convolution structure, and innovatively applied to rock water leakage detection to improve computing efficiency and achieve real-time detection, so that the moisture content of the rock can be quickly detected during a short construction interval. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0057] Figure 1 It is a schematic diagram of the overall process of a method for identifying and warning of hidden water bodies in front of a construction tunnel provided by an embodiment of the present invention.

[0058] Figure 2It is a schematic diagram of the backbone network structure provided by an embodiment of the present invention.

[0059] Figure 3 It is a schematic diagram of a tunnel face coordinate system provided by an embodiment of the present invention.

[0060] Figure 4 It is a schematic diagram of the relative positions of the tunnel face and the detection points distributed along the axial direction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The enumerated embodiments are only used to explain the present invention and are not used to limit the scope of the present invention.

[0062] Reference Figure 1 This embodiment provides a method for identifying and warning a hidden water body in front of a construction tunnel, the method comprising the steps of:

[0063] S100, collecting infrared thermal images of the tunnel face, and performing filtering processing on the infrared thermal images.

[0064] S200, inputting the infrared thermal image after filtering into the mask-rcnn neural network optimized by deep separable convolution for processing, judging whether there is water leakage in the tunnel face, and if there is water leakage, calculating the actual area of ​​the water leakage area.

[0065] S300: If it is determined that there is no water leakage, the temperature and temperature difference law on the tunnel face are analyzed, and based on the temperature distribution law of the tunnel face, it is determined whether there is a hidden water body in front of the tunnel face; if there is a hidden water body, the position and size of the hidden water body are calculated.

[0066] S400: Output the analysis and judgment results of water leakage and hidden water bodies in front of the tunnel face.

[0067] In step S100, since the rock moisture content in the seepage area and the dry area is gradually attenuated, the boundary of the two areas is blurred under the angle of the infrared thermal image, which brings difficulties to the water leakage identification and the calculation of the leakage and dry areas. To solve this problem, this embodiment adds a bilateral filtering algorithm based on normal vectors before the image is classified and segmented to enhance the distinction between the boundaries of the seepage area and the dry area. The bilateral filtering process based on normal is as follows: S101, select a pixel point q from the infrared thermal image pixel Q i , query all neighboring points N through the neighbor index k (q i );

[0068] S102, calculate pixel point q i The normal vector n i ;

[0069] S103, set the initial value of weight coefficient ∑ φ =0, and the initial value of weight factor λ=0;

[0070] S104, calculate pixel point q i and its neighboring point p i The distance ‖p i -q i ‖, and the normal distance <n i ,p i -q i >

[0071] S105: Update parameters λ=λ+φ <n i ,p i -q i >,∑ φ =∑ φ +φ;

[0072] σ d Indicates q i To its nearest neighbor point p i Distance influence weight factor, σ n Indicates q i With p i The distance vector is the normal vector at this point n i The projection on q i The influence weight factor of , φ represents the Gaussian kernel function;

[0073] S106, update pixel point q i location;

[0074] S107, repeat S101 to S106 to traverse all pixel points in the infrared thermal image atlas to complete the filtering process.

[0075] As an optional implementation, step S200 specifically includes:

[0076] S201, input the filtered infrared thermal image into the backbone network, refer to Figure 2 ,The backbone network is composed of Depthwise convolution and Pointwise convolution to generate a feature map containing ,temperature field and rock texture information.

[0077] S202. Based on the feature map generated by S202, the region proposal network generates a series of pre-designed bounding boxes that may contain water seepage areas, and arranges the proposed bounding boxes on the feature map to a more accurate position through the RoI Align operation, thereby effectively avoiding the two errors caused by convolution and deconvolution caused by the pooling operation.

[0078] S203. The fixed feature map is passed to the normalized exponential function (SoftMax) and the bounding box regression layer in parallel through the fully connected layer of the network to determine whether there is leakage water in the area, and different moisture content levels and external border positions of the leakage position are given respectively. At the same time, the fully convolutional network is used to generate a binary mask to segment the precise position of the leakage from the background. The precision of this segmentation is higher than that of the external frame of the border regression.

[0079] S204, after the binary mask is generated, the real area of ​​the tunnel surrounding rock corresponding to the infrared thermal image is calculated based on the shooting distance, and then the real area of ​​the seepage area is further calculated according to the ratio of the number of pixels of the binary mask to the number of pixels of the entire infrared thermal image.

[0080] Whether there is water seepage in the rock and the evaluation of the water seepage level are related to the mechanical properties of the rock. At the same time, the water seepage situation can also predict whether there is a risk of sudden water inrush in further construction. Therefore, the detection of whether there is seepage in the rock is related to the adjustment of the next construction method and the control of construction safety. It is very necessary to identify water seepage in the rock during the construction period. This optional implementation incorporates a deep separable convolutional neural network on the basis of the traditional mask rcnn model, thereby achieving the purpose of further improving computational efficiency.

[0081] After determining that there is no water leakage, in order to further determine whether there is a hidden water body ahead of the tunnel face excavation and predict potential risks in advance, as another optional implementation, step S300 includes a method for predicting hidden water bodies according to the temperature distribution law of the tunnel face:

[0082] S301. A rectangular coordinate system is established with the center of the tunnel face as the origin, the height direction of the tunnel face as the y-axis, and the width direction of the tunnel face as the x-axis, and the distribution law of temperature and temperature difference on the tunnel face is analyzed based on this. Figure 3 A schematic diagram of the rectangular coordinate system is given.

[0083] S302. Determine the location of hidden water bodies by analyzing the temperature of the face and the temperature difference distribution pattern.

[0084] S303, calculating the location and size of the hidden water body.

[0085] The position of hidden water bodies can be determined by analyzing the temperature and temperature difference distribution of the molecular face, including:

[0086] S401. When the distance between the tunnel face and the hidden water body is r, assuming that the hidden water body is directly in front of the tunnel face at y=m, the temperature T of any point on the tunnel face is calculated as follows:

[0087] T=0.5C1ln[(|ym|-r0) 2 +r2 +C2

[0088] In the above formula, r0 represents the size of the hidden water body, C1 and C2 are constants related to the properties of the surrounding rock and water, and m represents the mapping position parameter of the hidden water body on the heading face;

[0089] Based on the above formula, it can be seen that the temperature at different positions on the heading face is related to |y - m|.

[0090] S402. Determine the position of the hidden water body according to the temperature at different positions on the heading face, specifically:

[0091] When there is a hidden water body in front of the tunnel, if the temperature of the hidden water body is higher than the temperature of the surrounding rock, the minimum temperature on the heading face appears at the boundary of the heading face. At this time, the position of the maximum temperature corresponds to the orientation of the hidden water body;

[0092] When the temperature of the hidden water body is lower than the temperature of the heading face, the maximum temperature on the heading face appears at the boundary of the heading face. At this time, the position of the minimum temperature corresponds to the orientation of the hidden water body;

[0093] If both the maximum and minimum temperatures on the heading face appear at the boundary of the heading face, it is temporarily impossible to judge the temperature of the water, and it is necessary to further analyze the temperature gradient of the heading face.

[0094] Analyze the temperature gradient of the heading face, specifically including:

[0095] S501. Calculate the temperature gradient of the temperature T along the y direction based on the calculation formula, and the corresponding calculation formula is:

[0096]

[0097] Based on the above formula, it can be seen that when y > m, the gradient of the temperature field increases with the increase of y, and when y < m, the gradient of the temperature field increases with the decrease of y. Since y = m is the corresponding position of the hidden water body, it can be not considered when calculating the gradient.

[0098] S502. Judge the existence and orientation of the hidden water body according to the temperature gradient distribution rules I and II on the heading face. Rule I is: the temperature gradient values on the heading face are symmetrically distributed with the position where the hidden water body exists (y = m) as the axis of symmetry; Rule II is: on the heading face, the closer to the orientation where the hidden water body exists, the smaller the temperature gradient value, and the farther away from the orientation where the hidden water body exists, the larger the temperature gradient value; Since the temperature difference at the same interval along the y-axis direction on the heading face is the accumulation of the temperature gradient, its temperature difference also follows this rule.

[0099] The above temperature gradient distribution rules can also provide a theoretical basis for judging the existence and orientation of the hidden water body in front of the tunnel.

[0100] S503. After determining the position of the hidden water body, further analyze the relationship between the maximum temperature difference on the tunnel face and the position and size of the hidden water body. The maximum temperature difference on the tunnel face ΔT max The calculation formula is:

[0101]

[0102] In the above formula, Ts represents the normal temperature of the surrounding rock, T w represents the temperature of the hidden water body, k represents the thermal conductivity, h1 represents the thermal convection coefficient of water, H represents the size of the tunnel face, r max It indicates the distance from the center of the water body when the surrounding rock temperature field returns to normal. A large amount of engineering data shows that this distance is generally 20m to 50m from the boundary of the hidden water body.

[0103] S504. Determine whether there is a hidden water body based on the distribution pattern of the maximum temperature difference on the tunnel face.

[0104] From the above formula, we can see that when When the hidden water body is large enough, there will be no temperature difference on the tunnel face. At this time, the hidden water body cannot be predicted by the temperature field on the tunnel face. The maximum temperature difference on the tunnel face has the following rules:

[0105] When the size of the hidden water body in front of the tunnel face exceeds the size of the tunnel face, the temperature field of the hidden water body superimposed at different positions on the same tunnel face is the same, and the temperature difference on the tunnel face is 0 at this time;

[0106] When there is a hidden water body in front of the tunnel face, there is always at least one temperature maximum on the tunnel face boundary. Therefore, the maximum temperature difference ΔT on the tunnel face is max It is positively correlated with the tunnel face size H. The larger H is, the greater the maximum temperature difference on the tunnel face;

[0107] When there is a hidden water body in front of the tunnel face, as the distance r from the hidden water body decreases, the maximum temperature difference ΔT on the tunnel face increases. max gradually increase;

[0108] When other conditions are constant, the size of the hidden water body in front of the face is r o Maximum temperature difference ΔT on the tunnel face max Inversely proportional.

[0109] Since the calculation formula of C1 is a function of the size of the hidden water body r0, when the maximum temperature difference on the tunnel face, the size of the tunnel face, and the direction of the hidden water body m are known, the relationship between the distance r between the tunnel face and the hidden water body and its size r0 can be calculated by this formula. The position and size of the hidden water body can be further calculated by the maximum temperature difference of the tunnel face and the axial temperature difference of the tunnel near the tunnel face. The specific calculation method is as follows:

[0110] The relative position of the tunnel face and the detection points distributed along the axial direction is as follows: Figure 4 shown. Figure 4 The middle face is located on section 1-1', and the maximum temperature difference is ΔT max , the two detection points with an axial spacing of ΔR are points a and b in the figure, from which the formula can be obtained:

[0111]

[0112] In the above formula, Δr represents the axial spacing, ΔT 轴向 Indicates the maximum temperature difference under Δr spacing.

[0113] Depend on Figure 4 It can be seen that the distance r between the hidden water body and the tunnel face (detection point) in the above equations is the same. Based on this, the relationship between r and R in this case can be obtained. O The value of .

[0114] In summary, when detecting hidden water bodies, we can first determine qualitatively whether there is hidden water in front of the tunnel face based on the temperature distribution on the tunnel face and the temperature gradient distribution, and at the same time determine the existence of hidden water bodies. Then, we can establish the relationship between the temperature on the tunnel face and the position and size of the hidden water body through the maximum temperature difference on the tunnel face. Through a single tunnel face, we can only determine the relative relationship between the radius of the hidden water body and the distance between the tunnel face and the hidden water body. If we want to determine the true distance of the hidden water body, we need to consider the temperature field data of the tunnel face and the axial direction at the same time. The comprehensive calculation of the two can finally obtain the position and size of the hidden water body in front of the tunnel.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying and warning of hidden water bodies in front of a construction tunnel, characterized in that: The method comprises: S100, collecting infrared thermal images of the tunnel face, and performing filtering processing on the infrared thermal images; S200, inputting the infrared thermal image after filtering into the mask-rcnn neural network optimized by deep separable convolution for processing, determining whether there is water leakage on the tunnel face, and if there is water leakage, calculating the real area of ​​the water leakage area; S300, if it is determined that there is no water leakage, then analyzing the temperature and temperature difference law on the tunnel face, and judging whether there is a hidden water body in front of the tunnel face based on the temperature distribution law of the tunnel face, and if there is a hidden water body, calculating the position and size of the hidden water body; S400: Output the analysis and judgment results of water leakage and hidden water bodies in front of the tunnel face.

2. The method for identifying and warning of hidden water bodies in front of a construction tunnel according to claim 1 is characterized in that: The infrared thermal image is filtered, including: S101, select a pixel point q from the infrared thermal image pixel Q i , query all neighboring points N through the neighbor index k (q i ); S102, calculate pixel point q i The normal vector n i ; S103, set the initial value of weight coefficient ∑ φ =0, and the initial value of weight factor λ=0; S104, calculate pixel point q i and its neighboring point p i The distance ‖p i -q i ‖, and the normal distance <n i ,p i -q i > S105: Update parameters λ=λ+φ <n i ,p i -q i >,∑ φ =∑ φ +φ; σ d Indicates q i To its nearest neighbor point p i Distance influence weight factor, σ n Indicates q i With p i The distance vector is the normal vector at this point n i The projection on q i The influence weight factor of , φ represents the Gaussian kernel function; S106, update pixel point q i location; S107, repeat S101 to S106 to traverse all pixel points in the infrared thermal image atlas to complete the filtering process.

3. The method for identifying and warning of hidden water bodies in front of a construction tunnel according to claim 1 is characterized in that: Step S200 specifically includes: S201, inputting the infrared thermal image after filtering into the backbone network, wherein the backbone network is composed of Depthwise convolution and Pointwise convolution to generate a feature map containing temperature field and rock texture information; S202, based on the feature map generated in S202, the region proposal network generates a series of pre-designed bounding boxes that may contain water seepage areas, and arranges the proposed bounding boxes on the feature map to a more accurate position through the RoI Align operation; S203, passing the fixed feature map to the normalized exponential function and the bounding box regression layer in parallel through the fully connected layer of the network to determine whether there is water seepage in the area, and giving different moisture content levels and the outer frame positions of the water seepage position respectively. At the same time, the fully convolutional network is used to generate a binary mask to segment the precise position of the water seepage from the background; S204, after the binary mask is generated, the real area of ​​the tunnel surrounding rock corresponding to the infrared thermal image is calculated based on the shooting distance, and then the real area of ​​the seepage area is further calculated according to the ratio of the number of pixels of the binary mask to the number of pixels of the entire infrared thermal image.

4. The method for identifying and warning of hidden water bodies in front of a construction tunnel according to claim 1 is characterized in that: Step S300 specifically includes: S301, establishing a rectangular coordinate system with the center of the face as the origin, the face height direction as the y-axis, and the face width direction as the x-axis; S302, determining the location of the hidden water body by analyzing the temperature of the tunnel face and the temperature difference distribution law; S303, calculating the location and size of the hidden water body.

5. The method for identifying and warning of hidden water bodies ahead of a construction tunnel according to claim 4, characterized in that: The location of hidden water bodies is determined by analyzing the temperature and temperature difference distribution of the tunnel face, including: S401. When the distance between the tunnel face and the hidden water body is r, assuming that the hidden water body is directly in front of the tunnel face at y=m, the temperature T of any point on the tunnel face is calculated as follows: T=0.5C1ln[(|y-m|-r0) 2 +r 2 ]+C2 In the above formula, r0 represents the size of the hidden water body, C1 and C2 are constants related to the properties of the surrounding rock and water, and m represents the mapping position parameter of the hidden water body on the tunnel face; S402, judging the location of the hidden water body according to the temperature at different locations on the tunnel face, specifically: When there is a hidden water body in front of the tunnel, if the temperature of the hidden water body is higher than that of the surrounding rock, the minimum temperature value on the heading face appears at the boundary of the heading face. At this time, the position of the maximum temperature value corresponds to the orientation of the hidden water body; When the temperature of the hidden water body is lower than the temperature of the heading face, the maximum temperature value on the heading face appears at the boundary of the heading face. At this time, the position of the minimum temperature value corresponds to the orientation of the hidden water body; If both the maximum and minimum temperature values on the heading face appear at the boundary of the heading face, further analyze the temperature gradient of the heading face.

6. The method for identifying and warning of hidden water bodies ahead of a construction tunnel according to claim 5, characterized in that: Analyze the temperature gradient of the heading face, specifically including: S501. Calculate the temperature gradient of it along the y-direction based on the calculation formula of temperature T. The corresponding calculation formula is: Based on the above formula, it can be seen that when y > m, the gradient of the temperature field increases with the increase of y; when y < m, the gradient of the temperature field increases with the decrease of y; S502. Judge the existence and orientation of the hidden water body according to the distribution laws I and II of the temperature gradient on the heading face. Law I is: the temperature gradient values on the heading face are symmetrically distributed with the position where the hidden water body exists (y = m) as the axis of symmetry; Law II is: on the heading face, the closer to the orientation where the hidden water body exists, the smaller the temperature gradient value, and the farther away from the orientation where the hidden water body exists, the larger the temperature gradient value; since the temperature difference at the same interval along the y-axis direction on the heading face is the accumulation of the temperature gradient, its temperature difference also follows this law; S503. After determining the position of the hidden water body, further analyze the relationship between the maximum temperature difference on the tunnel face and the position and size of the hidden water body. The maximum temperature difference on the tunnel face ΔT max The calculation formula is: In the above formula, Ts represents the normal temperature of the surrounding rock, T w represents the temperature of the hidden water body, k represents the thermal conductivity, h1 represents the thermal convection coefficient of water, H represents the size of the tunnel face, r max Indicates the distance from the center of the water body when the surrounding rock temperature field returns to normal; S504. Judge whether there is a hidden water body based on the distribution law of the maximum temperature difference on the heading face.

7. The method for identifying and warning of hidden water bodies ahead of a construction tunnel according to claim 6, characterized in that: The distribution law of the maximum temperature difference on the heading face includes: When the size of the hidden water body in front of the heading face exceeds the size of the heading face, the temperature fields superimposed at different positions on the same heading face by the hidden water body are the same. At this time, the temperature difference on the heading face is 0; When there is a hidden water body in front of the tunnel face, there is always at least one temperature maximum on the tunnel face boundary. Therefore, the maximum temperature difference ΔT on the tunnel face is max It is positively correlated with the tunnel face size H. The larger H is, the greater the maximum temperature difference on the tunnel face; When there is a hidden water body in front of the tunnel face, as the distance r from the hidden water body decreases, the maximum temperature difference ΔT on the tunnel face increases. max gradually increase; When other conditions are constant, the size of the hidden water body r0 in front of the tunnel face and the maximum temperature difference ΔT on the tunnel face max Inversely proportional.

8. The method for identifying and warning of hidden water bodies ahead of a construction tunnel according to claim 7, characterized in that: The calculation of the position and size of the hidden water body specifically includes: S601. Analyze the relative positions of the heading face and the detection points distributed along the axial direction, and establish a mathematical model as shown in the following formula: In the above formula, Δr represents the axial spacing, ΔT 轴向 Indicates the maximum temperature difference under Δr spacing; S602. Solve the mathematical model to obtain the values of the distance r between the heading face and the hidden water body and its size r0.

Images