Methods for Detecting Landfill Remediation Effectiveness Using Remote Sensing Image Analysis

Abstract

This application relates to the field of image processing technology and proposes a method for detecting the remediation effect of landfills using remote sensing image analysis. The method includes: acquiring a baseline infrared image, a registered visible light image, and an infrared image of the area to be detected for landfill remediation effect detection; identifying drainage ditch areas in the visible light image; determining the texture direction offset and corrected texture direction offset of pixels; determining the corrected gradient magnitude of pixels by combining the gradient magnitude of pixels in the infrared image; calculating the normal texture information entropy and texture mutual information entropy of a sliding window of a preset size at each position based on the projection values ​​of the corrected gradient magnitude of different pixels in the infrared image in the normal and tangential directions; determining the non-leakage confidence of each sliding window; and obtaining the monitoring result of the landfill remediation effect based on the pixels contained in the sliding window corresponding to the non-leakage confidence. This application aims to improve the accuracy of landfill remediation effect detection results.

Description

Technical Field

[0001] This application relates to the field of image processing technology, specifically to a method for detecting the remediation effect of landfills using remote sensing image analysis. Background Technology

[0002] To accurately assess the effectiveness of soil pollution remediation, monitor groundwater pollution, and gain a comprehensive understanding of ecosystem improvement, remote sensing images are needed to detect the remediation effects of landfills. However, when using remote sensing images to detect the remediation effects of drainage ditches in landfills, the localized heating caused by methane leaks can lead to varying degrees of expansion in different types of materials with different thermal inertia on the surface. This can result in pseudo-changes and shifts in the surface texture direction of the acquired images.

[0003] Generally, single-window and split-window algorithms can be used to perform linear temperature correction on pseudo-changes and offsets. However, linear correction does not conform to the characteristic that thermal inertia decreases with distance, which often leads to the corrected texture direction not matching the actual direction, thus making the detection results of landfill remediation effect inaccurate. Summary of the Invention

[0004] This application provides a method for detecting the remediation effect of landfills using remote sensing image analysis, in order to solve the problem that the surface texture direction of remote sensing images is subject to false changes and inaccurate offset correction, resulting in inaccurate landfill remediation effect detection results. The specific technical solution adopted is as follows:

[0005] One embodiment of this application provides a method for detecting the remediation effect of landfills using remote sensing image analysis. The method includes the following steps:

[0006] Collect baseline infrared images, registered visible light images, and infrared images of the area to be tested for landfill remediation effects, and identify drainage ditch areas in the visible light images;

[0007] Based on the distribution of pixel values ​​around the corresponding pixel in the visible light image, the texture direction offset of the pixel is determined. Combining the difference in pixel values ​​between the infrared image and the reference infrared image, and the positional distribution relationship between the corresponding pixel and the drainage ditch area in the infrared image and the visible light image, the corrected texture direction offset of the pixel in the infrared image is calculated. Combining the gradient magnitude of the pixel in the infrared image, the corrected gradient magnitude of the pixel is determined.

[0008] Based on the projection values ​​of the correction gradient magnitude of different pixels in the infrared image onto the normal and tangential directions, calculate the normal texture information entropy, tangential texture information entropy, and texture mutual information entropy of a sliding window of a preset size at each position.

[0009] Based on the differences in texture mutual information entropy between sliding windows at all locations and the normal texture information entropy of the sliding windows, the non-leaking confidence of each sliding window is determined. Based on the pixels contained in the sliding window corresponding to the non-leaking confidence, the monitoring results of landfill remediation effect are obtained.

[0010] Furthermore, the method for obtaining the texture direction offset of the pixel is as follows:

[0011] Let any pixel in the visible light image be the target visible light pixel. With the target visible light pixel as the center, establish a local window of the target visible light pixel. Obtain the contrast of the gray-level co-occurrence matrix of the local window of the target visible light pixel under different preset directions. The preset direction corresponding to the maximum value of the contrast of the gray-level co-occurrence matrix under all preset directions is recorded as the texture direction of the target visible light pixel.

[0012] The maximum difference between the texture direction of the target visible light pixel and all its adjacent pixels is recorded as the texture direction offset of the target visible light pixel.

[0013] The texture direction offset of the corresponding pixel in the visible light image is denoted as the texture direction offset of the pixel in the infrared image.

[0014] Furthermore, the method for calculating the corrected texture direction offset of pixels in the infrared image by combining the difference in pixel values ​​between corresponding pixels in the infrared image and the reference infrared image, and the positional distribution relationship between corresponding pixels in the infrared image and the drainage ditch area, includes the following specific methods:

[0015] The average difference between the pixel values ​​of all corresponding pixels in the infrared image and the reference infrared image is recorded as the local temperature rise of the infrared image.

[0016] Calculate the surface thermal inertia of each pixel in the infrared image;

[0017] Identify the area corresponding to the drainage ditch in the visible light image, and record all pixels in the area corresponding to the drainage ditch as drainage ditch pixels. Determine the straight-line distance of the drainage ditch between the corresponding pixels in the infrared image and the drainage ditch pixels based on the minimum Euclidean distance between the corresponding pixels in the infrared image and the drainage ditch pixels in the visible light image.

[0018] Based on the local temperature rise of the infrared image, the surface thermal inertia of the pixels in the infrared image, and the straight-line distance of the drainage ditch, the texture direction offset of the pixels in the infrared image is corrected, and the corrected texture direction offset of the pixels in the infrared image is calculated.

[0019] Furthermore, the method for obtaining the straight-line distance of the drainage ditch among pixels in the infrared image is as follows:

[0020] The minimum Euclidean distance between the corresponding pixel in the infrared image and all the pixels in the drainage ditch in the visible light image is denoted as the straight-line distance of the pixel in the infrared image to the drainage ditch.

[0021] Furthermore, the formula for calculating the corrected texture direction offset is: in, This represents the corrected texture direction offset of a pixel within an infrared image. This represents the texture direction offset of a pixel within an infrared image. This represents a preset empirical coefficient, which has a value greater than or equal to 0.8 and less than or equal to 1.2. This represents the surface thermal inertia of a pixel within an infrared image. Indicates the local temperature rise in an infrared image; Represents the natural constant; This represents the straight-line distance between pixels in an infrared image and the drainage ditch. This indicates the preset distance parameter, which has a value greater than or equal to 0.5 and less than or equal to 1.0.

[0022] Furthermore, the method for obtaining the correction gradient magnitude of the pixel is as follows:

[0023] Calculate the gradient magnitude of each pixel in the infrared image;

[0024] The sum of the gradient magnitude of a pixel in an infrared image and the offset of the corrected texture direction is denoted as the corrected gradient magnitude of the pixel in the infrared image.

[0025] Furthermore, the specific method for obtaining the normal texture information entropy is as follows:

[0026] The normal texture image is obtained by projecting the corrected gradient magnitude of a pixel in the infrared image onto the normal direction.

[0027] Set a sliding window with a side length of the second preset length, and record the normalized value of the information entropy of the pixel values ​​of all pixels in the sliding window in the normal texture image as the normal texture information entropy of the sliding window.

[0028] Furthermore, the method for obtaining the texture mutual information entropy is as follows:

[0029] The ratio of the normal texture information entropy to the tangential texture information entropy of the sliding window at the same position is denoted as the texture mutual information entropy of the sliding window at the same position.

[0030] Furthermore, the method for determining the non-disclosure credibility is as follows:

[0031] The mean of the texture mutual information entropy of the sliding window at all positions is denoted as the comprehensive texture mutual information entropy; the range of the texture mutual information entropy at all positions is denoted as the range texture mutual information entropy; the absolute value of the difference between the texture mutual information entropy of the sliding window and the comprehensive texture mutual information entropy is denoted as the texture difference mutual information entropy of the sliding window; the ratio of the texture difference mutual information entropy of the sliding window to the range texture mutual information entropy is denoted as the texture mutual normalization information entropy of the sliding window.

[0032] The undisclosed credibility of the sliding window is obtained by weighting and summing the difference between the value 1 and the normalized information entropy of the texture of the sliding window and the normal texture information entropy of the sliding window with preset weights.

[0033] Furthermore, the specific method for obtaining the monitoring results of landfill remediation effectiveness based on the pixels contained in the sliding window corresponding to the undisclosed confidence level includes:

[0034] Based on the value of the undisclosed confidence level, the pixels within the sliding window are marked as well-repaired pixels, qualified pixels, and high-risk pixels, respectively.

[0035] The priorities for repairing excellent pixels, qualified pixels, and high-risk pixels are marked as 1, 2, and 3, respectively. When the same pixel is marked with different priorities at the same time, the pixel with the lowest priority is used as the pixel's label.

[0036] The area consisting of well-repaired pixels and qualified pixels is designated as the qualified repair area, while the area consisting of high-risk pixels is designated as the unqualified repair area.

[0037] The beneficial effects of this application are:

[0038] This application considers that when monitoring the remediation effect of drainage ditches in landfills, the localized heating caused by methane leakage will cause the surface material to expand. Based on the heat absorption and dissipation characteristics of the material itself and the characteristic that the thermal inertia of the pixel position decreases with the distance to the joint of the drainage ditch, the texture direction offset of the pixel is corrected. The corrected texture direction offset of the pixel in the infrared image is obtained, and the corrected gradient amplitude of the pixel is determined by combining it with the gradient amplitude of the pixel in the infrared image. Then, based on the difference in the texture change characteristics of the tangential and normal directions of the linear shape of the drainage ditch, the remediation features of the landfill and the pseudo texture signal caused by the methane leakage are extracted from the normal and tangential directions, respectively. The normal texture information entropy and tangential texture information entropy of the sliding window at each position are obtained, and the texture mutual information entropy is calculated. Furthermore, based on the difference in texture mutual information entropy between the sliding windows at all positions and the normal texture information entropy of the sliding window, the real remediation features of the landfill are extracted in a directional manner, and the non-leakage confidence of each sliding window is determined. Finally, based on the pixels contained in the sliding window corresponding to the non-leakage confidence, the monitoring results of the landfill remediation effect are obtained. This study aims to address the problem of inaccurate detection results of landfill remediation effects due to spurious changes in surface texture direction and inaccurate offset correction in remote sensing images, thereby improving the accuracy of landfill remediation effect detection results. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic flowchart of a landfill remediation effect detection method using remote sensing image analysis provided in one embodiment of this application;

[0041] Figure 2 This is a flowchart illustrating the process of obtaining the corrected texture direction offset according to one embodiment of this application. Detailed Implementation

[0042] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0043] Please see Figure 1The diagram illustrates a flowchart of a landfill remediation effectiveness detection method based on remote sensing image analysis, according to an embodiment of this application. The method includes the following steps:

[0044] Step S001: Collect the baseline infrared image, registered visible light image and infrared image of the area to be tested for landfill remediation effect, and identify the drainage ditch area in the visible light image.

[0045] A drone equipped with a visible light camera, an infrared camera, and a lidar was used to collect visible light images while the drone flew at an altitude of 100-150m over the area where the landfill remediation effect was to be tested. The registered visible light and infrared images were then obtained.

[0046] Using the YOLO object detection model, the region corresponding to the drainage ditch in the visible light image is identified, and all pixels within the region corresponding to the drainage ditch are recorded as drainage ditch pixels.

[0047] The use of the YOLO object detection model to identify the area corresponding to the drainage ditch in a visible light image is a well-known technique and will not be elaborated further.

[0048] Before implementing drainage ditch repair projects in areas where the effectiveness of landfill remediation needs to be tested, baseline infrared images are collected using infrared cameras.

[0049] At this point, registered visible light images, infrared images, and reference infrared images of the area to be tested for landfill remediation effects have been obtained.

[0050] Step S002: Based on the distribution of pixel values ​​around the corresponding pixel in the visible light image, determine the texture direction offset of the pixel. Combine the difference in pixel values ​​between the corresponding pixel in the infrared image and the reference infrared image, as well as the positional distribution relationship between the corresponding pixel in the infrared image and the drainage ditch area, calculate the corrected texture direction offset of the pixel in the infrared image. Combine the gradient magnitude of the pixel in the infrared image to determine the corrected gradient magnitude of the pixel.

[0051] When assessing the remediation effectiveness of drainage ditches in landfills, the localized heating caused by methane leaks can expand surface materials, leading to spurious changes and shifts in the surface texture direction of the acquired images. Single-window and split-window algorithms are commonly used to perform temperature-based linear corrections for these spurious changes and shifts. However, linear corrections do not reflect the characteristic that thermal inertia decreases with distance. Furthermore, they neglect the heat absorption and dissipation properties of the materials themselves and the influence of the linear shape of the drainage ditch on texture analysis, often resulting in mismatches between the corrected texture direction and the actual direction, thus leading to inaccurate landfill remediation effectiveness assessment results.

[0052] First, thermal inertia reflects a material's ability to resist temperature changes. Temperature and texture offset can be correlated based on thermal inertia parameters to eliminate interference from thermal anomalies. Then, based on the differences in texture variation characteristics of the linear shape of the drainage ditch in the tangential and normal directions, the remediation features of landfill and the pseudo-texture signals caused by methane leakage are extracted from the normal and tangential directions respectively, suppressing the interference of methane thermal anomalies and directionally extracting the real remediation features of landfill.

[0053] The average difference between the pixel values ​​of all corresponding pixels in the infrared image and the reference infrared image is recorded as the local temperature rise of the infrared image.

[0054] Denote any pixel in the visible light image as the target visible light pixel. A local window with a side length of a first preset length is established centered on the target visible light pixel. The local window of the target visible light pixel is obtained, and its gray-level co-occurrence matrix is ​​acquired in 12 preset directions. The contrast of the gray-level co-occurrence matrix is ​​calculated. The preset direction corresponding to the maximum contrast of the gray-level co-occurrence matrix in all preset directions is denoted as the texture direction of the target visible light pixel. The maximum difference between the texture directions of the target visible light pixel and all its adjacent pixels is denoted as the texture direction offset of the target visible light pixel. The texture direction offset of the corresponding pixel in the visible light image is denoted as the texture direction offset of the pixel in the infrared image.

[0055] In this embodiment, the first preset length is set to 15; the 12 preset directions in this embodiment are 0°, 15°, 30°, 45°, 60°, 75°, 90°, 105°, 120°, 135°, 150°, and 165°; the calculation of the contrast of the gray-level co-occurrence matrix is ​​a well-known technique and will not be described in detail here.

[0056] Based on the infrared image, calculate the surface thermal inertia of each pixel within the infrared image. The calculation of the surface thermal inertia of pixels in the infrared image is a well-known technique and will not be elaborated further.

[0057] The minimum Euclidean distance between the corresponding pixel in the infrared image and all the pixels in the drainage ditch in the visible light image is denoted as the straight-line distance of the pixel in the infrared image to the drainage ditch.

[0058] Based on the local temperature rise of the infrared image, the surface thermal inertia of the pixels in the infrared image, and the straight-line distance of the drainage ditch, the texture direction offset of the pixels in the infrared image is corrected, and the corrected texture direction offset of the pixels in the infrared image is calculated.

[0059] in, This represents the corrected texture direction offset of a pixel within an infrared image. This represents the texture direction offset of a pixel within an infrared image. This represents a preset empirical coefficient. In this embodiment, the empirical coefficient is set to 1. The empirical coefficient is used to adjust the thermal response of the object corresponding to the pixel in the infrared image. The empirical coefficient should be greater than or equal to 0.8 and less than or equal to 1.2. This represents the surface thermal inertia of a pixel within an infrared image. Indicates the local temperature rise in an infrared image; Represents the natural constant; This represents the straight-line distance between pixels in an infrared image and the drainage ditch. This represents a preset distance parameter. In this embodiment, the distance parameter is set to 0.75. The distance parameter should be greater than or equal to 0.5 and less than or equal to 1.0. The distance parameter is used to adjust the attenuation rate of thermal anomalies of objects corresponding to pixels in the infrared image as they diffuse outward from the joints of the drainage ditch.

[0060] It is understandable that thermal inertia is a comprehensive physical index that measures a material’s ability to resist temperature changes. The smaller the thermal inertia, the weaker the object’s ability to resist temperature changes, and the more significant the texture changes caused by the same amount of temperature change in infrared images, the greater the correction required.

[0061] The acquisition of the texture orientation offset correction incorporates thermal inertia, thermal distortion coupling, and the distance from the pixel in the infrared image to the drainage ditch seam, thus achieving the correction of the texture orientation offset. The greater the local temperature rise in the infrared image, and the smaller the surface thermal inertia of the pixel in the infrared image and the straight-line distance between the pixel and the drainage ditch, the greater the difference between the corrected texture orientation offset and the actual texture orientation offset.

[0062] At this point, the corrected texture direction offset of each pixel in the infrared image has been obtained. The flowchart for obtaining the corrected texture direction offset is as follows: Figure 2 As stated above.

[0063] Calculate the gradient magnitude of each pixel in the infrared image, and denot the sum of the gradient magnitude of each pixel in the infrared image and the offset of the corrected texture direction as the corrected gradient magnitude of the pixel in the infrared image.

[0064] The calculation of the gradient magnitude of pixels in an infrared image is a well-known technique and will not be elaborated further.

[0065] At this point, the corrected gradient magnitude of each pixel in the infrared image is obtained.

[0066] Step S003: Based on the projection values ​​of the correction gradient magnitude of different pixels in the infrared image in the normal and tangential directions, calculate the normal texture information entropy, tangential texture information entropy, and texture mutual information entropy of a sliding window of a preset size at each position.

[0067] Since the drainage ditches are linearly distributed, the texture uniformity perpendicular to the joint direction and the texture abrupt change along the joint direction will affect each other. In order to avoid interference in the horizontal and vertical directions and accurately locate the leak location, the entropy change is analyzed from both the normal and tangential directions. Specifically, the entropy change in the normal direction is used to determine whether the vegetation growth is uniform, and the entropy change in the tangential direction is used to identify the methane leak signal.

[0068] Based on the projection values ​​of the correction gradient magnitudes of different pixels in the infrared image onto the normal and tangential directions, the texture mutual information entropy of a sliding window of a preset size at each position is calculated.

[0069] The normal texture image is obtained by projecting the corrected gradient magnitude of a pixel in the infrared image onto the normal direction. A sliding window with a side length of a second preset length is set, and the normalized entropy of the information entropy of the pixel values ​​of all pixels within the sliding window in the normal texture image is recorded as the normal texture information entropy of the sliding window. With a sliding step size of 1 and a sliding direction from left to right and from top to bottom, the normal texture information entropy of the sliding window at each position is obtained.

[0070] The tangential texture image is obtained by projecting the corrected gradient magnitude of a pixel in the infrared image onto the tangential direction. A sliding window with a side length of a second preset length is set, and the normalized value of the entropy of the pixel values ​​of all pixels within the sliding window in the tangential texture image is recorded as the tangential texture entropy of the sliding window. With a sliding step size of 1 and a sliding direction from left to right and from top to bottom, the tangential texture entropy of the sliding window at each position is obtained.

[0071] The ratio of the normal texture information entropy to the tangential texture information entropy of the sliding window at the same position is denoted as the texture mutual information entropy of the sliding window at the same position.

[0072] When the vegetation grows more uniformly and the terrain is flatter, the restoration effect is better. The normal texture information entropy and the tangential texture information entropy of the sliding window are smaller. At this time, the texture mutual information entropy of the sliding window is smaller.

[0073] It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In practical applications, implementers may use other methods of existing technology, such as the maximum-minimum normalization method or the sigmoid function, to calculate the normalized value, and no limitation is made here.

[0074] At this point, the normal texture information entropy, tangential texture information entropy, and texture mutual information entropy of the sliding window at each position are obtained.

[0075] Step S004: Based on the differences in texture mutual information entropy between sliding windows at all locations and the normal texture information entropy of the sliding window, determine the non-leaking confidence of each sliding window. Based on the pixels contained in the sliding window corresponding to the non-leaking confidence, obtain the monitoring results of the landfill remediation effect.

[0076] The mean of the texture mutual information entropy of the sliding window at all positions is denoted as the comprehensive texture mutual information entropy; the range of the texture mutual information entropy at all positions is denoted as the range texture mutual information entropy; the absolute value of the difference between the texture mutual information entropy of the sliding window and the comprehensive texture mutual information entropy is denoted as the texture difference mutual information entropy of the sliding window; the ratio of the texture difference mutual information entropy of the sliding window to the range texture mutual information entropy is denoted as the texture mutual normalization information entropy of the sliding window.

[0077] The first preset weight is used as the weight of the normal texture information entropy of the sliding window, and the second preset weight is used as the weight of the difference between the value 1 and the normalized texture information entropy of the sliding window. The difference between the value 1 and the normalized texture information entropy of the sliding window and the normal texture information entropy of the sliding window are weighted and summed to obtain the unleashed confidence of the sliding window.

[0078] In this embodiment, the first weight is set to 0.6 and the second weight is set to 0.4. In actual application, as other implementation methods, the implementer can decide the values ​​of the first weight and the second weight. The sum of the first weight and the second weight should be 1. This application does not impose any other special restrictions.

[0079] The non-leakage confidence of the sliding window enables a quantitative assessment of repair quality and spatial visualization of leakage risk. The higher the non-leakage confidence of the sliding window, the better the repair quality and the lower the leakage risk at the corresponding position of the sliding window.

[0080] At this point, the undisclosed credibility of the sliding window in all locations has been obtained.

[0081] Pixels within a sliding window with a non-disclosure confidence level greater than or equal to the first preset threshold are marked as excellent repair pixels; pixels within a sliding window with a non-disclosure confidence level greater than or equal to the second preset threshold and less than the first preset threshold are marked as qualified repair pixels; and pixels within a sliding window with a non-disclosure confidence level less than or equal to the second preset threshold are marked as high-risk pixels.

[0082] The priorities for repairing excellent pixels, qualified pixels, and high-risk pixels are assigned as 1, 2, and 3, respectively. When the same pixel is marked with different priorities, the pixel with the lowest priority is used as the pixel's label.

[0083] Understandably, "excellent repair" refers to the pixel location where the landfill restoration effect is optimal, "qualified repair" refers to the pixel location where the landfill restoration effect is qualified, and "high-risk" refers to the pixel location where the landfill restoration effect is unqualified.

[0084] The area consisting of well-repaired pixels and qualified pixels is designated as the qualified repair area, while the area consisting of high-risk pixels is designated as the unqualified repair area.

[0085] Thus, the monitoring results of the landfill remediation effect were obtained.

[0086] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for detecting the remediation effect of landfills using remote sensing image analysis, characterized in that, The method includes the following steps: Collect baseline infrared images, registered visible light images, and infrared images of the area to be tested for landfill remediation effects, and identify drainage ditch areas in the visible light images; Based on the distribution of pixel values ​​around the corresponding pixel in the visible light image, the texture direction offset of the pixel is determined. Combining the difference in pixel values ​​between the infrared image and the reference infrared image, and the positional distribution relationship between the corresponding pixel and the drainage ditch area in the infrared image and the visible light image, the corrected texture direction offset of the pixel in the infrared image is calculated. Combining the gradient magnitude of the pixel in the infrared image, the corrected gradient magnitude of the pixel is determined. Based on the projection values ​​of the correction gradient magnitude of different pixels in the infrared image onto the normal and tangential directions, calculate the normal texture information entropy, tangential texture information entropy, and texture mutual information entropy of a sliding window of a preset size at each position. Based on the differences in texture mutual information entropy between sliding windows at all locations and the normal texture information entropy of the sliding windows, the non-leaking confidence of each sliding window is determined. Based on the pixels contained in the sliding window corresponding to the non-leaking confidence, the monitoring results of landfill remediation effect are obtained.

2. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for obtaining the texture direction offset of the pixel is as follows: Let any pixel in the visible light image be the target visible light pixel. With the target visible light pixel as the center, establish a local window of the target visible light pixel. Obtain the contrast of the gray-level co-occurrence matrix of the local window of the target visible light pixel under different preset directions. The preset direction corresponding to the maximum value of the contrast of the gray-level co-occurrence matrix under all preset directions is recorded as the texture direction of the target visible light pixel. The maximum difference between the texture direction of the target visible light pixel and all its adjacent pixels is recorded as the texture direction offset of the target visible light pixel. The texture direction offset of the corresponding pixel in the visible light image is denoted as the texture direction offset of the pixel in the infrared image.

3. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for calculating the corrected texture direction offset of pixels in the infrared image by combining the differences in pixel values ​​between corresponding pixels in the infrared image and the reference infrared image, and the positional distribution relationship between corresponding pixels in the infrared image and the drainage ditch area, includes the following specific methods: The average difference between the pixel values ​​of all corresponding pixels in the infrared image and the reference infrared image is recorded as the local temperature rise of the infrared image. Calculate the surface thermal inertia of each pixel in the infrared image; Identify the area corresponding to the drainage ditch in the visible light image, and record all pixels in the area corresponding to the drainage ditch as drainage ditch pixels. Determine the straight-line distance of the drainage ditch between the corresponding pixels in the infrared image and the drainage ditch pixels based on the minimum Euclidean distance between the corresponding pixels in the infrared image and the drainage ditch pixels in the visible light image. Based on the local temperature rise of the infrared image, the surface thermal inertia of the pixels in the infrared image, and the straight-line distance of the drainage ditch, the texture direction offset of the pixels in the infrared image is corrected, and the corrected texture direction offset of the pixels in the infrared image is calculated.

4. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 3, characterized in that, The method for obtaining the straight-line distance of the drainage ditch between pixels in the infrared image is as follows: The minimum Euclidean distance between the corresponding pixel in the infrared image and all the pixels in the drainage ditch in the visible light image is denoted as the straight-line distance of the pixel in the infrared image to the drainage ditch.

5. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 3, characterized in that, The formula for calculating the corrected texture direction offset is: in, This represents the corrected texture direction offset of a pixel within an infrared image. This represents the texture direction offset of a pixel within an infrared image. This represents a preset empirical coefficient, which has a value greater than or equal to 0.8 and less than or equal to 1.

2. This represents the surface thermal inertia of a pixel within an infrared image. Indicates the local temperature rise in an infrared image; Represents the natural constant; This represents the straight-line distance between pixels in an infrared image and the drainage ditch. This indicates the preset distance parameter, which has a value greater than or equal to 0.5 and less than or equal to 1.

0.

6. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for obtaining the correction gradient magnitude of the pixel is as follows: Calculate the gradient magnitude of each pixel in the infrared image; The sum of the gradient magnitude of a pixel in an infrared image and the offset of the corrected texture direction is denoted as the corrected gradient magnitude of the pixel in the infrared image.

7. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The specific method for obtaining the entropy of the normal texture information is as follows: The normal texture image is obtained by projecting the corrected gradient magnitude of a pixel in the infrared image onto the normal direction. Set a sliding window with a side length of the second preset length, and record the normalized value of the information entropy of the pixel values ​​of all pixels in the sliding window in the normal texture image as the normal texture information entropy of the sliding window.

8. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for obtaining the texture mutual information entropy is as follows: The ratio of the normal texture information entropy to the tangential texture information entropy of the sliding window at the same position is denoted as the texture mutual information entropy of the sliding window at the same position.

9. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for determining the non-disclosure credibility is as follows: The mean of the texture mutual information entropy of the sliding window at all positions is denoted as the comprehensive texture mutual information entropy; the range of the texture mutual information entropy at all positions is denoted as the range texture mutual information entropy; the absolute value of the difference between the texture mutual information entropy of the sliding window and the comprehensive texture mutual information entropy is denoted as the texture difference mutual information entropy of the sliding window; the ratio of the texture difference mutual information entropy of the sliding window to the range texture mutual information entropy is denoted as the texture mutual normalization information entropy of the sliding window. The undisclosed credibility of the sliding window is obtained by weighting and summing the difference between the value 1 and the normalized information entropy of the texture of the sliding window and the normal texture information entropy of the sliding window with preset weights.

10. The method for detecting the remediation effect of landfills using remote sensing image analysis according to claim 1, characterized in that, The method for obtaining the monitoring results of landfill remediation effectiveness based on the pixels contained in the sliding window corresponding to the non-disclosed confidence level includes the following specific methods: Based on the value of the undisclosed confidence level, the pixels within the sliding window are marked as well-repaired pixels, qualified pixels, and high-risk pixels, respectively. The priorities for repairing excellent pixels, qualified pixels, and high-risk pixels are marked as 1, 2, and 3, respectively. When the same pixel is marked with different priorities at the same time, the pixel with the lowest priority is used as the pixel's label. The area consisting of well-repaired pixels and qualified pixels is designated as the qualified repair area, while the area consisting of high-risk pixels is designated as the unqualified repair area.

Images