Detection Method, Device, Medium and Equipment for Snow and Ice Melting Infiltration

Abstract

The present application discloses a detection method, device, medium and equipment for snow and ice ablation and infiltration. The method includes: obtaining a remote sensing image of a target area, inputting the remote sensing image into multiple different trained snow cover classification models respectively to obtain the classification results of each snow cover classification model; performing fusion processing on the classification results of each snow cover classification model, and calculating the snow cover area of the target area according to the fusion result; constructing a relationship model between the infiltration amount of meltwater and the ablation amount of snow and ice, and calculating the infiltration amount of meltwater in the target area based on the snow cover area of the target area and the relationship model between the infiltration amount of meltwater and the ablation amount of snow and ice. Since the classification results of multiple snow cover classification models are fused in the present application, the accuracy of the fused classification results is relatively high, and the obtained snow cover area is also relatively accurate. At the same time, the relationship model between snow and ice ablation and the infiltration amount of water takes into account the complex situation of snow and ice ablation in open-pit mines in cold regions. Therefore, the accuracy of the calculated infiltration amount of meltwater is high.

Description

Technical Field

[0001] The present invention relates to the technical field of ice and snow melting, and particularly relates to a detection method, device, medium and equipment for ice and snow melting and infiltration. Background Art

[0002] In alpine and high-altitude regions, mine exploitation faces extremely severe natural condition challenges. Especially during the winter snow accumulation and spring ice and snow melting stages, significant changes will occur in the seepage field of the mine pit. With the continuous increase in the mining depth of the mine pit and the change of seasons, coupled with the influence of mining disturbances, the original groundwater seepage field may redistribute, greatly increasing the risks of slope water seepage and water inrush. In addition, due to the complexity of the slope rock and soil properties and the diversity of influencing factors, along with dangerous factors such as snowmelt water seepage, slope landslides may be caused during the exploitation activities of open-pit mines in alpine and high-altitude regions. Especially as the mining depth of the mine increases, the safety problems of high and steep slopes become more prominent. Therefore, it is extremely important to detect the infiltration of ice and snow melt in the mining area.

[0003] Currently, for the detection of the infiltration of ice and snow melt in open-pit mines in alpine and high-altitude regions, it mainly relies on satellite remote sensing technology. By obtaining images from satellites, the reflectance data of ice and snow in the visible light band can be obtained. As the ice and snow melt, its surface will become rough, the impurities will increase, and the reflectance will change. By comparing the reflectance data of different periods, the process of ice and snow melting and infiltration can be monitored.

[0004] However, there are some interference factors on the surface of the mining area, and the obtained reflectance cannot accurately reflect the state of snow cover. Therefore, the existing detection methods for ice and snow melting in the mining area are difficult to accurately detect the situation of ice and snow melting and infiltration. Summary of the Invention

[0005] In view of this, the present invention provides a detection method, device, medium and equipment for ice and snow melting and infiltration, mainly aiming to solve the problem that the existing detection methods for ice and snow melting in the mining area are difficult to accurately detect the ice and snow melting and infiltration.

[0006] According to one aspect of the present application, a detection method for ice and snow melting and infiltration is provided, and the method includes:

[0007] Obtain a remote sensing image of the target area, and input the remote sensing image into a plurality of different trained snow cover classification models respectively to obtain the classification results of each snow cover classification model, where the classification results include the confidence levels of multiple category labels corresponding to each pixel point in the remote sensing image;

[0008] Fuse the classification results of each snow cover classification model, and calculate the snow cover area of the target area according to the fusion result;

[0009] Construct a relationship model between the meltwater infiltration volume and the ice and snow ablation volume, and calculate the meltwater infiltration volume of the target area based on the snow cover area of the target area and the relationship model between the meltwater infiltration volume and the ice and snow ablation volume;

[0010] Among them, the process of fusing the classification results of each snow cover classification model and calculating the snow cover area of the target area according to the fusion result includes:

[0011] For each snow cover classification model, obtain the confidence level corresponding to each category, multiply the confidence level of the category label corresponding to each pixel by the confidence level corresponding to the category label to obtain the confidence level allocation data corresponding to each pixel;

[0012] Based on the confidence level allocation data corresponding to each pixel of each snow cover classification model and a preset conflict factor calculation formula, calculate the conflict factor;

[0013] Based on the conflict factor, the confidence level allocation data corresponding to each pixel of each snow cover classification model, and a preset confidence level fusion calculation formula, calculate the fused confidence level allocation data corresponding to each pixel;

[0014] Based on the fused confidence level allocation data of each pixel, determine the comprehensive category of the pixel, and multiply the number of pixels with the comprehensive category of snow cover by the preset image spatial resolution as the snow cover area of the target area.

[0015] Optionally, before obtaining the confidence level corresponding to each category for each snow cover classification model, the detection method for ice and snow ablation infiltration further includes:

[0016] Obtain the verification data set corresponding to each snow cover classification model, input the verification data set into each trained snow cover classification model respectively, and obtain the model classification result corresponding to each snow cover classification model;

[0017] Based on the actual classification result and the model classification result of the verification data set, conduct positive and negative statistics on the classification results, and generate a confusion matrix corresponding to each snow cover classification model according to the statistical results;

[0018] Based on the confusion matrix corresponding to each snow cover classification model and a preset confidence level calculation formula, calculate the confidence level corresponding to each category of each snow cover classification model.

[0019] Optionally, the construction of the relationship model between the meltwater infiltration volume and the ice and snow ablation volume includes:

[0020] Construct a snow sublimation model using the surface temperature method;

[0021] Based on the balance relationship of the total snow and ice volume, the snow and ice ablation volume, and the surface runoff, and the snow sublimation model, a relationship model between the meltwater infiltration volume and the snow and ice ablation volume is constructed.

[0022] Optionally, the relationship model between the meltwater infiltration volume and the snow and ice ablation volume is:

[0023] ,

[0024] Wherein, R is the meltwater infiltration volume; is the snow density; S n is the snow cover area; h is the average snow depth; I is the surface runoff, is the evaporation, R n is the net radiation, S is the energy stored in the snow and soil, is the air density, C p is the specific heat at constant pressure of air, f(u x ) is the ground energy transfer equation, ( T s -T a ) is the temperature difference between the snow surface and the air.

[0025] Optionally, before inputting the remote sensing image into multiple trained different snow cover classification models respectively to obtain the classification results of each snow cover classification model, the detection method for snow and ice ablation and infiltration further includes:

[0026] Obtain multiple original remote sensing images of the training area, and determine the snow cover area in each original remote sensing image;

[0027] Intercept each of the original remote sensing images according to a preset image size to obtain multiple training images, and determine the snow cover classification category and the snow cover area of each training image;

[0028] Based on the multiple training images and their corresponding snow cover classification categories and snow cover areas, train multiple different initial snow cover classification models respectively to obtain multiple trained snow cover classification models.

[0029] Optionally, the multiple snow cover classification models include snow cover classification machine learning models constructed by using the K-nearest neighbor algorithm, the random forest algorithm, the support vector machine algorithm, and the adaptive boosting algorithm respectively.

[0030] According to another aspect of the present application, there is provided a detection device for snow and ice ablation and infiltration, including:

[0031] A classification module for obtaining a remote sensing image of a target area, inputting the remote sensing image into multiple different trained snow cover classification models respectively, and obtaining the classification results of each snow cover classification model, where the classification results include the confidence levels of multiple category labels corresponding to each pixel point in the remote sensing image;

[0032] A fusion module for fusing the classification results of each snow cover classification model and calculating the snow cover area of the target area according to the fusion result;

[0033] A calculation module for constructing a relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow, and calculating the infiltration amount of meltwater in the target area based on the snow cover area of the target area and the relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow;

[0034] Wherein, the fusion module is further configured to:

[0035] For each of the snow cover classification models, obtain the confidence level corresponding to each category, multiply the confidence level of the category label corresponding to each pixel point by the confidence level corresponding to the category label, and obtain the confidence level assignment data corresponding to each pixel point;

[0036] Calculate the conflict factor based on the confidence level assignment data corresponding to each pixel point of each snow cover classification model and a preset conflict factor calculation formula;

[0037] Calculate the fused confidence level assignment data corresponding to each pixel point based on the conflict factor, the confidence level assignment data corresponding to each pixel point of each snow cover classification model, and a preset confidence level fusion calculation formula;

[0038] Based on the fused confidence level assignment data corresponding to each pixel point, determine the comprehensive category of the pixel point, and use the product of the number of pixel points with the comprehensive category of snow cover and the preset image spatial resolution as the snow cover area of the target area.

[0039] Optionally, the detection device for ice and snow ablation infiltration further includes:

[0040] A category confidence level acquisition module for obtaining the verification data set corresponding to each snow cover classification model, inputting the verification data set into each trained snow cover classification model respectively, and obtaining the model classification results corresponding to each snow cover classification model; performing positive and negative statistics on the classification results based on the actual classification results and the model classification results of the verification data set, and generating a confusion matrix corresponding to each snow cover classification model according to the statistical results; calculating based on the confusion matrix corresponding to each snow cover classification model and a preset confidence level calculation formula to obtain the confidence level corresponding to each category of each snow cover classification model.

[0041] Optionally, the calculation module is further configured to:

[0042] Construct a snow sublimation model using the surface temperature method;

[0043] Based on the balance relationship of the total amount of ice and snow, the amount of ice and snow ablation, and the surface runoff and the snow sublimation model, construct a relationship model between the amount of meltwater infiltration and the amount of ice and snow ablation.

[0044] Optionally, the relationship model between the amount of meltwater infiltration and the amount of ice and snow ablation is:

[0045] ,

[0046] where R is the amount of meltwater infiltration; is the snow density; S n is the snow-covered area; h is the average snow depth; I is the surface runoff, is the evaporation, R n is the net radiation, S is the energy stored in the snow and soil, is the air density, C p is the specific heat at constant pressure of air, f(u x ) is the ground energy transfer equation, ( T s -T a ) is the temperature difference between the snow surface and the air.

[0047] Optionally, the detection device for ice and snow ablation infiltration further includes:

[0048] A training module, configured to obtain a plurality of original remote sensing images of a training area, and determine the snow-covered areas in each original remote sensing image; intercept each of the original remote sensing images according to a preset image size to obtain a plurality of training images, and determine the snow classification category and the snow-covered area of each training image; based on the plurality of training images and their corresponding snow classification categories and snow-covered areas, train a plurality of different initial snow classification models respectively to obtain a plurality of trained snow classification models.

[0049] Optionally, the plurality of snow classification models include snow classification machine learning models constructed using the K-nearest neighbor algorithm, the random forest algorithm, the support vector machine algorithm, and the adaptive boosting algorithm respectively.

[0050] According to another aspect of the present application, there is provided a storage medium storing at least one executable instruction, and the executable instruction causes a processor to perform operations corresponding to the above-mentioned detection method for ice and snow ablation and infiltration.

[0051] According to another aspect of the present application, there is provided a computer device, including: a processor, a memory, a communication interface, and a communication bus. The processor, the memory, and the communication interface complete communication with each other through the communication bus;

[0052] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations corresponding to the above-mentioned detection method for ice and snow ablation and infiltration.

[0053] By means of the above technical solutions, the technical solutions provided by the embodiments of the present invention have at least the following advantages:

[0054] A detection method, device, equipment and medium for ice and snow ablation and infiltration provided by the present application perform snow cover classification detection on a remote sensing image of a target area through multiple different snow cover classification models to obtain multiple classification results, fuse the multiple classification results, calculate the snow cover area according to the classification results after fusion processing, input the calculated snow cover area into a relationship model between meltwater infiltration and ice and snow ablation, and calculate the meltwater infiltration amount of the target area. Since the results of multiple snow cover classification models are fused, the accuracy of the classification results after fusion is relatively high, and the obtained snow cover area is relatively accurate, improving the accuracy of the meltwater infiltration amount. At the same time, the relationship model between ice and snow ablation and infiltration water volume takes into account the complex conditions of ice and snow ablation in open-pit mines in cold regions, which also improves the accuracy of the meltwater infiltration amount.

[0055] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention are specifically exemplified below. Description of the Drawings

[0056] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. And throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0057] Figure 1 A flowchart of a detection method for ice and snow ablation and infiltration provided by an embodiment of the present application is shown;

[0058] Figure 2Shows a flowchart of another method for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0059] Figure 3 Shows the snow classification result images output by multiple different snow classification models of a method for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0060] Figure 4 Shows a schematic diagram of fusing the snow classification results output by multiple different snow classification models of a method for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0061] Figure 5 Shows the image after fusing the snow classification results output by multiple different snow classification models of a method for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0062] Figure 6 Shows a comparison diagram between the classification result of the snow cover map obtained by shooting the actual mining area by an unmanned aerial vehicle and the snow classification result after fusion of a method for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0063] Figure 7 Shows a block diagram of the composition of a device for detecting ice and snow ablation and infiltration provided by an embodiment of the present application;

[0064] Figure 8 Shows a schematic structural diagram of a computer device provided by an embodiment of the present invention.

[0065] Among them,

[0066] Figure 7 In: 702 - Classification module; 704 - Fusion module; 706 - Calculation module;

[0067] Figure 8 In: 802 - Processor; 804 - Communication interface; 806 - Memory; 808 - Communication bus; 810 - Program. Specific embodiments

[0068] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other.

[0069] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes the specific implementation manner, structure, features, and their effects of the application according to the present invention in detail in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "embodiments" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0070] In view of the problem that the current detection methods for ice and snow ablation in mining areas are difficult to accurately detect the infiltration of ice and snow ablation, the embodiment of the present application provides a detection method for the infiltration of ice and snow ablation, as Figure 1 shown, the method includes:

[0071] 102: Obtain the remote sensing image of the target area, and input the remote sensing image into multiple different trained snow cover classification models respectively to obtain the classification results of each snow cover classification model;

[0072] 104: Perform fusion processing on the classification results of each snow cover classification model, and calculate the snow cover area of the target area according to the fusion result;

[0073] 106: Construct a relationship model between the meltwater infiltration amount and the ice and snow ablation amount, and calculate the meltwater infiltration amount of the target area based on the snow cover area of the target area and the relationship model between the meltwater infiltration amount and the ice and snow ablation amount.

[0074] Specifically, the target area is the open-pit mine slope. Obtain the high-resolution optical remote sensing image of the snow-covered area of the open-pit mine slope, perform operations such as preprocessing, resampling, and panchromatic sharpening on the remote sensing image, and then input the preprocessed remote sensing image into multiple different trained snow cover classification models, such as the snow cover classification models established by the K-nearest neighbor algorithm, random forest algorithm, support vector machine algorithm, and adaptive boosting algorithm. Through these different snow cover classification models, classify the preprocessed remote sensing image into snow and non-snow.

[0075] Use the Dempster-Shafer framework to fuse the classification results output by multiple different snow cover classification models to obtain the fused classification result, calculate the snow cover area according to the fused classification result; based on parameters such as environmental conditions and topography, construct a relationship model between the meltwater infiltration amount and the ice and snow ablation amount on the basis of the heat conduction equation, and substitute the calculated snow cover area into the relationship model between the meltwater infiltration amount and the ice and snow ablation amount to calculate the meltwater infiltration amount of the target area.

[0076] The present application provides a method for detecting snowmelt infiltration. Compared with the prior art, multiple different snow cover classification models are used to perform snow cover classification detection on the remote sensing image of the target area, obtaining multiple classification results. The multiple classification results are fused, and the snow cover area is calculated according to the classification results after fusion processing. The calculated snow cover area is input into the relationship model between the infiltration amount of meltwater and snowmelt, and the infiltration amount of meltwater in the target area is calculated. Since the results of multiple snow cover classification models are fused, the accuracy of the classification results after fusion is relatively high, and the obtained snow cover area is relatively accurate, improving the accuracy of the infiltration amount of meltwater. At the same time, the relationship model between snowmelt and infiltration water volume considers the complex situation of snowmelt in open-pit mines in cold regions, which also improves the accuracy of the infiltration amount of meltwater.

[0077] In an embodiment of the present invention, as Figure 2 shown, the classification result includes the confidence levels of multiple category labels corresponding to each pixel point in the remote sensing image; fusing the classification results of each snow cover classification model and calculating the snow cover area of the target area according to the fusion result includes:

[0078] 202: For each snow cover classification model, obtain the confidence level corresponding to each category, multiply the confidence level of the category label corresponding to each pixel point by the confidence level corresponding to the category label to obtain the confidence level allocation data corresponding to each pixel point;

[0079] 204: Based on the confidence level allocation data corresponding to each pixel point of each snow cover classification model and the preset conflict factor calculation formula, calculate the conflict factor;

[0080] 206: Based on the conflict factor, the confidence level allocation data corresponding to each pixel point of each snow cover classification model, and the preset confidence level fusion calculation formula, calculate the fused confidence level allocation data corresponding to each pixel point;

[0081] 208: Based on the fused confidence level allocation data of each pixel point, determine the comprehensive category of the pixel point, and multiply the number of pixel points with the comprehensive category of snow cover by the preset image spatial resolution as the snow cover area of the target area.

[0082] In this embodiment, the combination rule of the DS (Dempster-Shafer) framework is used to fuse these different classification results, so as to obtain a more reliable classification result. Obtain the classification results of different snow cover classification models for the same remote sensing image, and calculate the trust assignment data corresponding to each pixel point in the classification results of each snow cover classification model. Taking a pixel point in the classification result of a snow cover classification model as an example, obtain the trust of this snow cover classification model based on a preset verification dataset, multiply the confidence of the class label corresponding to this pixel point in the classification result by the trust of the category corresponding to this class label, and use the product as the trust assignment data corresponding to this pixel point. Using the same method, calculate the trust assignment data corresponding to each pixel point in the classification results of each snow cover classification model.

[0083] Then use the DS fusion rule to combine the trust assignment data corresponding to each pixel point in the classification results of different snow cover classification models to obtain comprehensive fused trust assignment data. Specifically, first calculate the conflict factor corresponding to each pixel point. After obtaining the conflict factor, substitute the conflict factor corresponding to each pixel point and the trust assignment data corresponding to each pixel point of each snow cover classification model into a preset trust fusion calculation formula to calculate the fused trust assignment data corresponding to each pixel point.

[0084] Taking two different snow cover classification models as an example, the BPA (trust assignment data) of each snow cover classification model for pixel point P is:

[0085] ,

[0086] where, and are the trusts of snow cover classification model i for "snow cover" and "non-snow cover" respectively, is the trust of snow cover classification model i for uncertain classification results.

[0087] Based on the trust assignment data of each pixel point in the classification results of each snow cover classification model, calculate the conflict factor corresponding to each pixel point. The conflict factor K represents the inconsistency between classifiers, and the calculation formula is:

[0088] ,

[0089] where, is the trust of the first snow cover classification model for "snow cover", is the trust of the second snow cover classification model for "non-snow cover". For two snow cover classification models M1 and M2, the conflict factor is:

[0090] ,

[0091] Among them, is the confidence that the pixel is snow cover in the first snow cover classification model, is the confidence that the pixel is non - snow cover in the first snow cover classification model, is the confidence that the pixel is snow cover in the second snow cover classification model, is the confidence that the pixel is non - snow cover in the second snow cover classification model.

[0092] Based on the Dempster - Shafer framework, fuse the classification results of different classification models, and use the Dempster combination rule to calculate the fused BPA:

[0093] ,

[0094] For two snow cover classification models M1 and M2, the fused BPA:

[0095] ,

[0096] Among them, is the confidence of the first snow cover classification model for uncertain classification results, is the confidence of the second snow cover classification model for uncertain classification results.

[0097] If there are multiple classification models, use the Dempster combination rule for fusion in turn. For example:

[0098] First, fuse M1 and M2 to get M 12 , then fuse M 12 and M3 to get M 123 , and so on, until the BPA of all classification models is fused to obtain the fused confidence assignment data of this pixel. According to the fused confidence assignment data of this pixel, determine the comprehensive category of this pixel. Usually, the category with the highest confidence is selected as the final comprehensive category. Use the above method to determine the comprehensive category of each pixel.

[0099] In an embodiment of the present invention, for each snow cover classification model, before obtaining the confidence corresponding to each category, the detection method for ice and snow ablation infiltration further includes:

[0100] Obtain the verification dataset corresponding to each snow cover classification model, input the verification dataset into each trained snow cover classification model respectively, and obtain the model classification results corresponding to each snow cover classification model;

[0101] Based on the actual classification results and model classification results of the verification dataset, conduct positive and negative statistics on the classification results, and generate the confusion matrix corresponding to each snow cover classification model according to the statistical results;

[0102] Based on the confusion matrix corresponding to each snow cover classification model and a preset confidence calculation formula, calculate to obtain the confidence corresponding to each category of each snow cover classification model.

[0103] Specifically, while obtaining the training data set, also obtain the validation data set. Taking a snow cover classification model as an example, after the snow cover classification model is trained, input the validation data set into the trained snow cover classification model to obtain the model classification result corresponding to the validation data set. Compare the model classification result corresponding to the validation data set with the actual classification result to determine the true and false of each validation sample. For example, (TP) = true positive, a sample correctly predicted as a positive class, (FN) = false negative, a sample that is actually a positive class but predicted as a negative class, (FP) = false positive, a sample that is actually a negative class but predicted as a positive class, (TN) = true negative, a sample correctly predicted as a negative class. Determine the number of true and false samples to generate the confusion matrix corresponding to this snow cover classification model.

[0104] For each snow cover classification model, calculate its confidence in different categories according to the confusion matrix corresponding to the snow cover classification model. For the category C i , the confidence of the classification model can be calculated by the following formula:

[0105] ,

[0106] where TP i and TN i are respectively the number of true positive examples and the number of true negative examples of the category C i , FP i and FN i are respectively the number of false positive examples and the number of false negative examples of the category C i .

[0107] The uncertainty of the classification model can be estimated by the error rate:

[0108] ,

[0109] Calculate the confidence corresponding to each category of each snow cover classification model respectively by using the above method.

[0110] In an embodiment of the present invention, construct a relationship model between meltwater infiltration and ice and snow ablation amount, including:

[0111] Construct a snow sublimation model by using the surface temperature method;

[0112] Based on the balance relationship of the total amount of ice and snow, the ablation amount of ice and snow, and the surface runoff, and the snow sublimation model, a relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow is constructed.

[0113] Specifically, the infiltration amount of meltwater is expressed as the product of snow density, snow area, and average snow thickness, and is associated with the snow sublimation amount and surface runoff. Combining environmental parameters such as snow density, snow area, average snow thickness, surface runoff, net radiation, energy storage, air density, latent heat of sublimation, specific heat of air, surface wind speed, and temperature difference between the snow surface and air, and fully considering the complex situation of ice and snow ablation in open-pit mines in cold regions, a quantitative modeling between ice and snow ablation and infiltration water volume is studied by adopting an energy balance model.

[0114] An ice and snow evaporation model is constructed by using the surface temperature method through collecting information such as the climate in the study area. The surface temperature method is based on the temperature difference between the air and the snow surface. Combining with the ground energy transfer equation, the sensible heat flux of the snow surface is calculated. Bringing the sensible heat flux into the surface energy balance equation can indirectly estimate the latent heat of evaporation of the snow cover, thereby calculating the snow sublimation amount. The snow sublimation model describes the process of snow cover turning into water vapor, where the snow sublimation amount is related to net radiation, energy storage, air density, latent heat of sublimation, specific heat of air, surface wind speed, and temperature difference between the snow surface and air.

[0115] The snow sublimation model is preliminarily constructed by using the surface temperature method:

[0116] ,

[0117] In the formula, is the latent heat of evaporation (W·kg -1 ); E is the snow sublimation amount (kg·m -2 ); R n is the net radiation (W·m -2 ); S is the energy stored in the snow and soil (W·m -2 ); is the air density (kg·m -3 ); C p is the specific heat at constant pressure of air (J / kg·°C); f(u x ) is the ground energy transfer equation; ( T s -T a ) is the temperature difference between the snow surface and the air (°C).

[0118] The relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow is:

[0119] ,

[0120] Among them, R is the infiltration amount of meltwater; is the snow density; S n is the snow cover area; h is the average snow depth; I is the surface runoff, is the evaporation, R n is the net radiation, S is the energy stored in snow and soil, is the air density, C p is the specific heat at constant pressure of air, f(u x ) is the ground energy transfer equation, ( T s -T a ) is the temperature difference between the snow surface and the air.

[0121] In an embodiment of the present invention, before inputting the remote sensing image into multiple different trained snow classification models respectively to obtain the classification results of each snow classification model, the detection method for snow and ice ablation infiltration further includes:

[0122] Obtain multiple original remote sensing images of the training area, and determine the snow-covered areas in each original remote sensing image;

[0123] Intercept each original remote sensing image according to a preset image size to obtain multiple training images, and determine the snow classification category and snow-covered area of each training image;

[0124] Based on multiple training images and their corresponding snow classification categories and snow-covered areas, train multiple different initial snow classification models respectively to obtain multiple trained snow classification models.

[0125] Specifically, obtain the high-resolution optical remote sensing image of the snow-covered open-pit mine slope as the original remote sensing image, perform operations such as preprocessing, resampling, and panchromatic sharpening on the original remote sensing image, and then use the normalized difference snow index method combined with the manual confirmation method to determine the snow-covered range in the original remote sensing image as the true value of the training set for subsequent model training.

[0126] The preprocessing operations of remote sensing spectral images include atmospheric correction to eliminate the influence of the atmosphere on the optical signals received by the sensor, geometric correction to eliminate the influence of terrain, and radiometric correction to convert the values received by the sensor into radiance values. In addition, in multi-spectral images, the spectral resolution of the spectral bands is higher than that of the spatial bands, while the spatial resolution of the panchromatic band is higher than that of the spectral bands. The panchromatic sharpening method is used, and the multi-spectral bands and spatial bands are merged based on the panchromatic band. The images processed by panchromatic sharpening can identify land features with high precision. In the panchromatic sharpening method, in order to maintain the quality of spatial information and spectral information, the local mean and variance matching (LMVM) algorithm is used. The local mean and variance matching algorithm is a technique for image processing, which can adjust the local contrast and brightness distribution of the image. This method measures the local mean and variance of the image, and then adjusts the image according to the required style or features. Through local mean and variance matching, the visual quality of the image can be improved, the details of the image can be enhanced, and the image can be made clearer and easier to observe. Since the study area is a mining area, the terrain of the mining area is relatively complex, resulting in a large difference in the illumination received by the remote sensing images on the shady slope and the sunny slope. Terrain correction refers to using different conversion methods to map the radiance of all pixels to a certain standard plane to reduce the influence of terrain undulation on the radiance of the image and make the spectral characteristics of ground objects more accurately presented.

[0127] Select an area with a suitable size in the original remote sensing image as the training area, and perform cropping. After cropping, it is divided into a training data set and a validation data set. For example, an image of 1000×1000 pixels is selected. Select snow and non-snow samples in the selected image area. Among the two categories (snow and non-snow samples), 70% of the samples are selected for training, and the remaining 30% are used for validating the trained model. After generating the classification image, the overall accuracy and global kappa coefficient are used to evaluate the validation metrics. The global kappa coefficient is calculated through the confusion matrix.

[0128] The normalized difference snow index (NDSI) is the most critical parameter in snow detection. This index has the ability to distinguish ice / snow from clouds. In the visible light range, the reflectivity of ice and snow remains above 50%, showing white on the single-band grayscale image. In the near-infrared band, the reflectivity of ice and snow continuously decreases to about 20%. At the blue light band of 0.49 µm, ice and snow show a reflectivity peak of more than 80%. As the wavelength increases, the reflectivity decreases sharply. This law can be used to construct the normalized snow index (NDSI). The calculation formula of NDSI:

[0129] ,

[0130] B 3, B 11They are the 3rd green band and the 11th short-wave infrared band of the spectral data corresponding to the remote sensing image respectively. The threshold used in this application is 0.4, and the threshold of NDSI>0.4 is used to distinguish snow pixels (including snow / ice, snow under shadow, water bodies) and non-snow pixels.

[0131] Since the determination of the threshold is not accurate enough, manual correction is carried out on the basis of the normalized difference snow index method to determine the snow-covered area in the training image set.

[0132] Use the K-Nearest Neighbors (KNN), Random Forest (RF), Support Vector Machine (SVM), and Adaptive Boosting (Adaboost) algorithms to establish a machine learning classification model. The training data set is input into these different initial snow classification models respectively to train the initial snow classification models and obtain the trained snow classification models.

[0133] The implementation principle of the K-Nearest Neighbors classification algorithm is as follows:

[0134] Take all samples with known classes as references.

[0135] By calculating the distances between the unknown sample and all known samples, select the K known samples closest to the unknown sample.

[0136] According to the majority-voting rule, classify the unknown sample into the class with the largest proportion among the K nearest neighbor samples. K represents the number of nearest neighbor sample instances to be selected. Since when classifying using the K-Nearest Neighbors classification algorithm, only the selected K value is used to determine the class of the sample to be classified, rather than relying on the method of discriminating the class domain to determine the class, the K-Nearest Neighbors algorithm is more applicable to sample sets with overlapping or poor class domains compared to other methods.

[0137] The implementation principle of the Support Vector Machine algorithm is as follows:

[0138] The difference between the Support Vector Machine algorithm and traditional machine learning methods is that the Support Vector Machine relies on the principle of minimizing structural risk for training. To achieve superior generalization performance, the goal of the Support Vector Machine is to balance the complexity of the model and the learning ability in order to achieve the best processing effect. The solution process of the Support Vector Machine is expressed as a convex quadratic programming problem with linear constraint conditions, and the global optimal solution is calculated based on optimization theory.

[0139] The implementation principle of the Random Forest algorithm is as follows:

[0140] The ensemble learning algorithm of random forest is constructed based on multiple decision trees. Compared with requiring too high classification accuracy for each decision tree, random forest only needs a part of the trees to have high accuracy. In this algorithm, each decision tree independently trains and classifies the data, and finally all decision trees vote on the data to determine its category by a majority decision. During the training process, each decision tree corresponds to a subset of the training data, and this technique is called the bagging (bootstrap aggregating or bagging) sampling method. This method increases the diversity and generalization ability of the model by randomly and repeatedly extracting subsets from the entire training set. When generating the random forest algorithm, each decision tree is trained based on a different subset of data, and finally the integrated result obtains the final classification result.

[0141] The implementation principle of the AdaBoost algorithm is as follows:

[0142] The core idea of Adaptive Boosting (AdaBoost) is to adaptively adjust the sample weights, so that new weak classifiers are generated in each iteration, focusing more on the samples misclassified in the previous round, thereby continuously improving the overall performance of the model. This process is repeated iteratively until a preset extremely low error rate is reached or the preset maximum number of iterations is reached, and finally a strong classifier with high accuracy is formed. One of the advantages of AdaBoost is that it can handle high-dimensional data and complex classification problems well, and it is not easy to overfit on this basis. In practical applications, AdaBoost is widely used in binary classification and multi-classification problems.

[0143] The AdaBoost iterative algorithm process is as follows:

[0144] Initialize the weight distribution of the training data: Suppose there are N samples, and initially the weights of each sample are the same, that is, 1 / N.

[0145] Train the weak classifier: In the AdaBoost algorithm, the process of training the weak classifier is very important. In each round, a weak classifier is trained based on the training data with the current weight distribution. During this process, if a sample is correctly classified, its weight will decrease when constructing the next round of training set; on the contrary, the weight of the sample that is not correctly classified will increase. Such a dynamic adjustment strategy ensures that the subsequent weak classifiers can focus more on the misclassified samples in the previous round, thereby improving the overall efficiency of the entire model. This iterative process will continue until the preset error rate is reached or the maximum number of iterations is reached, and finally a strong classifier is constructed. This adaptive weight adjustment training method endows AdaBoost with excellent performance in solving complex classification problems, while enhancing the stability and generalization ability of the algorithm.

[0146] Construction of the strong classifier: In the process of constructing the strong classifier, the combination of weak classifiers is not treated equally, but is combined according to their classification error rates. In the final classification decision, the weak classifier with a low error rate will obtain a higher weight, thus more significantly affecting the result; on the contrary, the weak classifier with a high error rate will be given a lower weight and have a smaller impact on the final classification. This weight adjustment strategy reflects its effectiveness in practical applications.

[0147] Taking the high-resolution optical satellite images of certain regions as the original remote sensing images, and taking the remote sensing images of another target region as an example, the detection of snow and ice ablation infiltration is carried out. The highest spatial resolution of the original remote sensing images is 10 meters, and it has various band types including visible light, infrared, and SWIR, etc. Through these bands, land phenomena such as clouds, ice, water, and snow can be distinguished and detected, and these changes can be monitored. Preprocess the original remote sensing images. Determine the snow-covered area in the original remote sensing graphics, select images of 1000×1000 pixels, crop the original remote sensing area to obtain multiple samples, and divide the multiple samples into a training data set and a validation data set. Based on the training data set, the initial snow classification models using the K-Nearest Neighbors (KNN) algorithm, Random Forest (RF) algorithm, Support Vector Machine (SVM) algorithm, and Adaptive Boosting (Adaboost) algorithm are trained respectively to obtain multiple different trained snow classification models.

[0148] Obtain the remote sensing images of the target region, input the remote sensing images into the above-mentioned different trained snow classification models, and obtain multiple classification results. The snow classification results obtained using the above snow classification models are as Figure 3 shown. Figures a, b, c, and d are the snow classification results output by the snow classification models using the K-Nearest Neighbors algorithm, Random Forest algorithm, Support Vector Machine, and Adaptive Boosting algorithm in sequence. The confusion matrices, overall accuracy (OA), and global kappa coefficient (K) corresponding to these snow classification models are shown in Tables 1 - 5. Based on the DS framework, the snow classification results output by the above four snow classification models are fused. The schematic diagram of the fusion process is as Figure 4 shown, and the fusion result is as Figure 5 shown. Compare the classification result of the snow-covered map obtained by shooting the actual mining area by drone with the snow classification result after fusion. The comparison diagram is as Figure 6 shown. Figures a and b are the classification result of the snow-covered map obtained by shooting the actual mining area by drone and the snow classification result after fusion in sequence. It can be seen from the comparison that the accuracy of the snow classification after fusion is relatively high.

[0149] Table 1 Confusion matrix of KNN classification

[0150]

[0151] Table 2 RF Classification Confusion Matrix

[0152]

[0153] Table 3 Adaboost Classification Confusion Matrix

[0154]

[0155] Table 4 SVM Classification Confusion Matrix

[0156]

[0157] Table 5 KAPPA Index and Overall Accuracy

[0158]

[0159] Collect information such as the climate and soil of the target area to obtain the parameters of the relationship model between meltwater infiltration and ice and snow ablation. The obtained model parameters are shown in Table 6. Based on these parameters and the calculated snow cover area, calculate the meltwater infiltration of the target area.

[0160] Table 6 Model Construction Parameters

[0161]

[0162] Furthermore, as an implementation of the method described above Figure 1 shown, an embodiment of the present invention provides a detection device for ice and snow ablation infiltration, as Figure 7 shown, the device includes:

[0163] A classification module 702, configured to obtain a remote sensing image of the target area, input the remote sensing image into a plurality of different trained snow classification models respectively, and obtain the classification results of each snow classification model;

[0164] A fusion module 704, configured to perform fusion processing on the classification results of each snow classification model, and calculate the snow cover area of the target area according to the fusion results;

[0165] A calculation module 706, configured to construct a relationship model between meltwater infiltration and ice and snow ablation, and calculate the meltwater infiltration of the target area based on the snow cover area of the target area and the relationship model between meltwater infiltration and ice and snow ablation.

[0166] The present application provides a detection device for snow and ice melting and infiltration. Compared with the prior art, snow cover classification detection is performed on a remote sensing image of a target area through multiple different snow cover classification models to obtain multiple classification results. The multiple classification results are fused, and the snow cover area is calculated according to the result after fusion processing. The calculated snow cover area is input into a relationship model between the infiltration amount of melting water and snow and ice melting, and the infiltration amount of melting water in the target area is calculated. Since the results of multiple snow cover classification models are fused, the accuracy of the classification results is relatively high, the obtained snow cover area is relatively accurate, the accuracy of the infiltration amount of melting water is improved, and at the same time, the relationship model between snow and ice melting and the infiltration water volume takes into account the complex situation of snow and ice melting in open-pit mines in cold regions, which also improves the accuracy of the infiltration amount of melting water.

[0167] In one embodiment, the classification result includes the confidence of multiple category labels corresponding to each pixel point in the remote sensing image; the fusion module 704 is further configured to:

[0168] For each snow cover classification model, obtain the trust degree corresponding to each category, multiply the confidence of the category label corresponding to each pixel point by the trust degree corresponding to the category label to obtain the trust degree allocation data corresponding to each pixel point;

[0169] Based on the trust degree allocation data corresponding to each pixel point of each snow cover classification model and a preset conflict factor calculation formula, calculate the conflict factor;

[0170] Based on the conflict factor, the trust degree allocation data corresponding to each pixel point of each snow cover classification model, and a preset trust degree fusion calculation formula, calculate the fusion trust degree allocation data corresponding to each pixel point;

[0171] Based on the fusion trust degree allocation data of each pixel point, determine the comprehensive category of the pixel point, and multiply the number of pixel points with the comprehensive category of snow cover by the preset image spatial resolution as the snow cover area of the target area.

[0172] In one embodiment, the detection device for snow and ice melting and infiltration further includes:

[0173] A category trust degree acquisition module, configured to obtain a verification data set corresponding to each snow cover classification model, input the verification data set into each trained snow cover classification model respectively to obtain a model classification result corresponding to each snow cover classification model; perform positive and negative statistics on the classification results based on the actual classification results and model classification results of the verification data set, and generate a confusion matrix corresponding to each snow cover classification model according to the statistical results; perform calculations based on the confusion matrix corresponding to each snow cover classification model and a preset trust degree calculation formula to obtain the trust degree corresponding to each category of each snow cover classification model.

[0174] In one embodiment, the calculation module 706 is further configured to:

[0175] Construct a snow sublimation model using the surface temperature method;

[0176] Based on the balance relationship of the total amount of ice and snow, the ablation amount of ice and snow, the surface runoff, and the snow sublimation model, construct a relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow.

[0177] In one embodiment, the relationship model between the infiltration amount of meltwater and the ablation amount of ice and snow is:

[0178] ,

[0179] wherein, R is the infiltration amount of meltwater; is the snow density; S n is the snow-covered area; h is the average thickness of the snow cover; I is the surface runoff, is the evaporation, R n is the net radiation, S is the energy stored in the snow and soil, is the air density, C p is the specific heat at constant pressure of air, f(u x ) is the ground energy transfer equation,( T s -T a ) is the temperature difference between the snow surface and the air.

[0180] In one embodiment, the detection device for ice and snow ablation infiltration further includes:

[0181] A training module, configured to obtain a plurality of original remote sensing images of a training area, and determine the snow-covered areas in each original remote sensing image; intercept each original remote sensing image according to a preset image size to obtain a plurality of training images, and determine the snow classification category and the snow-covered area of each training image; respectively train a plurality of different initial snow classification models based on the plurality of training images and their corresponding snow classification categories and snow-covered areas to obtain a plurality of trained snow classification models.

[0182] In one embodiment, the plurality of snow classification models include snow classification machine learning models constructed by using the K-nearest neighbor algorithm, the random forest algorithm, the support vector machine algorithm, and the adaptive boosting algorithm respectively.

[0183] According to an embodiment of the present invention, there is provided a storage medium storing at least one executable instruction, and the computer executable instruction can execute the detection method for ice and snow ablation infiltration in any of the above method embodiments.

[0184] Figure 8 The schematic structural diagram of a computer device provided according to an embodiment of the present invention is shown. The specific embodiments of the present invention do not limit the specific implementation of the computer device.

[0185] As Figure 8 shown, the computer device may include: a processor 802, a communications interface 804, a memory 806, and a communication bus 808.

[0186] Among them: the processor 802, the communications interface 804, and the memory 806 communicate with each other through the communication bus 808.

[0187] The communications interface 804 is used to communicate with network elements of other devices such as clients or other servers.

[0188] The processor 802 is used to execute the program 810, and specifically can execute the relevant steps in the above-described embodiment of the method for detecting snowmelt infiltration.

[0189] Specifically, the program 810 may include program code, and the program code includes computer operation instructions.

[0190] The processor 802 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention. One or more processors included in the computer device may be of the same type of processor, such as one or more CPUs; or may be of different types of processors, such as one or more CPUs and one or more ASICs.

[0191] The memory 806 is used to store the program 810. The memory 806 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0192] The program 810 is specifically used to cause the processor 802 to perform the following operations:

[0193] Obtain a remote sensing image of the target area, input the remote sensing image into a plurality of different trained snow cover classification models respectively, and obtain the classification results of each snow cover classification model;

[0194] Fuse the classification results of each snow cover classification model, and calculate the snow cover area of the target area according to the fusion result;

[0195] Construct a relationship model between the meltwater infiltration volume and the ice and snow ablation volume, and calculate the meltwater infiltration volume of the target area based on the snow cover area of the target area and the relationship model between the meltwater infiltration volume and the ice and snow ablation volume.

[0196] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. In one embodiment, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0197] The above embodiments are only exemplary embodiments of the present application and are not used to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art can make various modifications or equivalent replacements within the essence and protection scope of the present application, and such modifications or equivalent replacements should also be regarded as falling within the protection scope of the present application.

Claims

1. A method for detecting ice and snow melting and infiltration, characterized in that: include: Acquire a remote sensing image of the target area, and input the remote sensing image into a plurality of different snow classification models that have been trained to obtain a classification result of each snow classification model, wherein the classification result includes the confidence of a plurality of category labels corresponding to each pixel in the remote sensing image; The classification results of each snow classification model are fused, and the snow cover area of ​​the target area is calculated according to the fusion results; Constructing a relationship model between meltwater infiltration and ice and snow melting, and calculating the meltwater infiltration in the target area based on the snow cover area of ​​the target area and the relationship model between meltwater infiltration and ice and snow melting; The step of fusing the classification results of each snow classification model and calculating the snow cover area of ​​the target area according to the fusion results includes: For each of the snow classification models, the confidence corresponding to each category is obtained, and the confidence of the category label corresponding to each pixel is multiplied by the confidence corresponding to the category label to obtain the confidence distribution data corresponding to each pixel; Calculate the conflict factor based on the trust distribution data corresponding to each pixel point of each snow classification model and a preset conflict factor calculation formula; Based on the conflict factor, the trust distribution data corresponding to each pixel of each snow classification model and a preset trust fusion calculation formula, the fused trust distribution data of each pixel is calculated; Based on the fused trust distribution data of each pixel, the comprehensive category of the pixel is determined, and the product of the number of pixels with the comprehensive category of snow and the preset image spatial resolution is used as the snow coverage area of ​​the target area.

2. The method for detecting ice and snow melting and infiltration as claimed in claim 1, characterized in that: Before obtaining the confidence level corresponding to each category for each snow classification model, the method for detecting snow and ice melting infiltration further includes: Obtain a validation data set corresponding to each snow classification model, and input the validation data set into each trained snow classification model to obtain a model classification result corresponding to each snow classification model; Perform positive and negative statistics of the classification results based on the actual classification results of the validation data set and the model classification results, and generate a confusion matrix corresponding to each of the snow classification models according to the statistical results; The confidence corresponding to each category of each snow classification model is obtained by performing calculation based on the confusion matrix corresponding to each snow classification model and a preset confidence calculation formula.

3. The method for detecting ice and snow melting and infiltration according to claim 1, characterized in that: The construction of the relationship model between meltwater infiltration and ice and snow melting includes: The snow sublimation model was constructed using the surface temperature method; Based on the balance relationship between the total amount of ice and snow, the amount of ice and snow melt, and the surface runoff and the snow sublimation model, a relationship model between meltwater infiltration and ice and snow melt is constructed.

4. The method for detecting ice and snow melting and infiltration as claimed in claim 3, characterized in that: The relationship model between the amount of meltwater infiltration and the amount of ice and snow melting is: ， in, R is the meltwater infiltration; is the snow density; S n is the snow area; h is the average thickness of snow; I is the surface runoff, For evaporation, R n is the net radiation, S Energy stored in snow and soil, is the air density, C p is the specific heat of air at constant pressure, f (u x ) is the ground energy transfer equation, ( T s -T a ) is the temperature difference between the snow surface and the air.

5. The method for detecting ice and snow melting and infiltration according to claim 1, characterized in that: Before inputting the remote sensing images into a plurality of trained snow classification models to obtain classification results of each snow classification model, the snow melting and infiltration detection method further includes: Acquire multiple original remote sensing images of the training area and determine the snow covered area in each original remote sensing image; Intercepting each of the original remote sensing images according to a preset image size to obtain a plurality of training images, and determining the snow classification category and snow coverage area of ​​each training image; Based on the multiple training images and their corresponding snow classification categories and snow-covered areas, multiple different initial snow classification models are trained respectively to obtain multiple trained snow classification models.

6. The method for detecting ice and snow melting and infiltration as claimed in claim 5, characterized in that: The multiple snow classification models include snow classification machine learning models constructed using K nearest neighbor algorithm, random forest algorithm, support vector machine algorithm and adaptive boosting algorithm respectively.

7. A device for detecting ice and snow melting and infiltration, characterized in that: include: A classification module, used to obtain a remote sensing image of a target area, and input the remote sensing image into a plurality of trained snow classification models to obtain a classification result of each snow classification model, wherein the classification result includes the confidence of a plurality of category labels corresponding to each pixel in the remote sensing image; A fusion module is used to fuse the classification results of each snow classification model and calculate the snow cover area of ​​the target area according to the fusion results; A calculation module is used to construct a relationship model between meltwater infiltration and ice and snow melting, and calculate the meltwater infiltration of the target area based on the snow cover area of ​​the target area and the relationship model between meltwater infiltration and ice and snow melting; Wherein, the fusion module is also used for: For each of the snow classification models, the confidence corresponding to each category is obtained, and the confidence of the category label corresponding to each pixel is multiplied by the confidence corresponding to the category label to obtain the confidence distribution data corresponding to each pixel; Calculate the conflict factor based on the trust distribution data corresponding to each pixel point of each snow classification model and a preset conflict factor calculation formula; Based on the conflict factor, the trust distribution data corresponding to each pixel of each snow classification model and a preset trust fusion calculation formula, the fused trust distribution data of each pixel is calculated; Based on the fused trust distribution data of each pixel, the comprehensive category of the pixel is determined, and the product of the number of pixels with the comprehensive category of snow and the preset image spatial resolution is used as the snow coverage area of ​​the target area.

8. A storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the method for detecting ice and snow melting and infiltration as described in any one of claims 1-6.

9. A computer device comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the method for detecting ice and snow melting and infiltration as described in any one of claims 1-6.

Images