Remote sensing identification method for complex mining area ground fissures

Abstract

The invention provides a complex mining area ground fissure remote sensing identification method, which comprises the following steps: carrying out non-shadow area filtering based on a sliding window, carrying out mining area shadow brightening based on HSV color compensation, and eliminating interference of sun shadow on ground fissure identification; filtering ground fracture noise based on ground fracture evaluation indexes of a linear index and a shape index, and identifying an initial ground fracture; a comprehensive evaluation and judgment mechanism is established by combining morphological processing and regional connectivity analysis, and the real form of the fissure is precisely reduced in a hierarchical and multi-scale mode; and based on multi-spatial scale threshold segmentation iteration, considering the length, the tail end trend, the inter-domain distance and the angle of a connected domain, performing layered threshold growth, and removing fine noise in combination with an automatic extraction mechanism to finish accurate identification of the ground fracture. According to the method, the accuracy of complex mining area crack extraction is improved, and a brand new technical support is provided for complex mining area ground crack identification.

Description

Technical Field

[0001] The invention relates to the technical field of ground fissure extraction in complex mining areas, and in particular to a remote sensing identification method for ground fissures in complex mining areas. Background Art

[0002] The continuous exploitation of underground coal resources will lead to a series of geological environmental problems such as surface subsidence in mining areas and ground fissures under the action of external forces. The above problems seriously affect the safety of ground buildings, land production capacity and the lives and property of surrounding people. Therefore, effective identification of ground fissures in mining areas has important theoretical and practical significance for the protection of national life and property safety, the management and restoration of green mines, etc.

[0003] However, traditional on-site detection is relatively costly, inefficient, and difficult to obtain the fine features of ground fissures. The use of satellite remote sensing data for ground fissure detection is affected by resolution and timeliness, and can only identify large-width ground fissures. In recent years, with the rapid development of drones and airborne laser radar (LiDAR) technology, it has been widely used to carry out mine ecological restoration. UAV remote sensing technology can obtain a variety of remote sensing data that show clear ground fissure characteristics, such as three-dimensional point clouds, digital surface models (DSMs), and digital orthophotos (DOMs). It has low cost, high accuracy, and high efficiency, reflecting the superiority of drone remote sensing technology in convenient ground fissure identification. However, since the extraction of ground fissures in mining areas is easily affected by complex terrain and noise background, manual visual interpretation alone is time-consuming and laborious and prone to omissions. It is often necessary to use digital image processing technology, machine learning algorithms, etc. to achieve automatic extraction of ground fissures in complex mining areas.

[0004] At present, automatic crack extraction based on digital image processing technology is one of the most commonly used methods. It has achieved a lot of results in extracting cracks in single backgrounds such as concrete pavements, buried pipelines, and house walls. However, for mining areas with complex surface environments and large terrain undulations, automatic extraction of ground cracks based on two-dimensional drone images is relatively limited. The main background noises such as shrubs, snow, and light shadows affect the accuracy of ground crack extraction. With the continuous development of computer vision technology, automatic crack identification based on machine learning algorithms has become the research focus of many scholars. For example, convolutional neural networks (CNNs) can effectively capture the grid-like topological structure of images, and better identify small cracks without being affected by noise. However, automatic crack identification requires a large amount of labeled data, and the accuracy of deep learning results is greatly affected by label accuracy, which has certain limitations. Summary of the invention

[0005] The purpose of the present invention is to provide a solution that can further improve the accuracy of ground fissure identification in complex mining areas in response to the deficiencies in the above-mentioned background technology.

[0006] In order to achieve the above object, the present invention provides a method for remote sensing identification of ground fissures in complex mining areas, comprising the following steps: S1, filtering the non-shadow area based on the sliding window, brightening the shadow of the mining area based on HSV color compensation, and eliminating the interference of the sun's shadow on the identification of ground fissures; S2, based on the ground fissure evaluation index of linear index and shape index, the ground fissure noise is filtered out and the initial ground fissure is identified; S3, combining morphological processing with regional connectivity analysis, establishes a comprehensive evaluation and judgment mechanism to accurately restore the true morphology of ground fissures at different levels and scales; S4, based on multi-spatial scale threshold segmentation iteration, considers the length of the connected domain, the end direction, the distance and angle between domains, performs hierarchical threshold growth, and combines the automatic extraction mechanism to eliminate small noise to complete the accurate identification of ground fissures.

[0007] Furthermore, S1 includes the following sub-steps: S11, setting the sliding window size, counting the number of pixels in each window when sliding the window step by step, suppressing small-scale noise according to the morphological characteristics of the non-shadow area to retain the large shadow area; S12, for the shadow areas that meet the conditions, the mine shadows are brightened based on HSV color compensation, the V channel in the HSV space is used to brighten the shadows, and then the H and S channels are locally filtered to make the color effect of the shadow area closer to the real color.

[0008] Furthermore, S11 is expressed by the following formula: ; in, is the pixel value to be retained, is the pixel value of the input image, for The number of black pixels inside, is the empirical threshold. When the above conditions are met, The pixel coverage is white, the black pixel value is 0, and the white pixel value is 255. The specific definition is: ; in, Indicates Centered at The sliding window Represents a sliding window Each pixel point, is the indicator function.

[0009] Furthermore, the linear index in S2 is expressed by express: ; in, and They represent the length and width of the minimum circumscribed rectangle of the connected domain where the ground fissure is located. The connected domain is the image area composed of points identified as ground fissure pixels in the image; Shape index by express: ; in, It represents the area of ​​the circumscribed circle of the minimum circumscribed rectangle of the connected domain where the ground fissure is located.

[0010] Furthermore, the continuity assumption is set in S3: is the gray level of the background pixel, is the average gray value of the initial ground fissure pixels. If , the pixel is determined as a ground fissure object; If the above conditions are not met, the following linear assumption is performed: a ground fissure compensation template that satisfies the ground fissure relationship is set, and if the pixel is determined to match one of the defined templates, the condition is further checked. ,in , is the threshold, the threshold >Threshold , if satisfied, the background pixel is classified as an initial ground crack pixel.

[0011] Furthermore, in S4, starting from the initial pixel of each ground fissure that meets the linearity and length conditions, the ground fissure pixels are tracked along the main direction of the connected domain, and the bright pixel endpoint finally tracked is defined as the seed point; Then, the search starts from the seed point and the Bresenham algorithm is used to grow the ground fissure pixels to achieve gap connection. The search conditions include the radius of the search area. ,angle and the length of the connected domain where the ground fissure is located , the ground fissure extraction conditions include the length of the ground fissure after the gap is connected .

[0012] Furthermore, for the search conditions, ; ; ; ; in, , , Respectively represent the radius, length and angle initial constraint values ​​of the search area, The length condition for extracting ground fissures is: , , are iterative thresholds, is the iteration termination condition, Indicates the number of iterations.

[0013] The above scheme of the present invention has the following beneficial effects: The remote sensing identification method of ground fissures in complex mining areas provided by the present invention, on the one hand, brightens the shadow of the mining area, performs tone compensation based on the HSV tone difference between the shadow area and the real image, and eliminates the interference of the sun shadow on the extraction of ground fissures; on the other hand, two ground fissure evaluation indicators are proposed and combined with the connected area analysis to filter out noise, and the crack morphology processing based on the continuity and linear assumptions is used to restore the real shape of the cracks. Based on the multi-spatial scale threshold segmentation iteration, the length of the connected domain, the terminal direction and the distance and angle between the connected domains are considered to perform threshold layered growth, and automatic extraction is set to achieve noise removal and complete identification of ground fissures, providing a new technical support for the identification of ground fissures in complex mining areas; Other beneficial effects of the present invention will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a flow chart of the method steps of the present invention; Figure 2 It is a schematic diagram of the accuracy results of ground fissure identification in a complex mining area in a specific case of the present invention. DETAILED DESCRIPTION

[0015] The following describes the embodiments of the present disclosure through specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0016] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0017] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure, and the illustrations only show the components related to the present disclosure rather than the number, shape and size of the components in actual implementation. The type, quantity and proportion of each component in actual implementation may be changed at will, and the component layout type may also be more complicated. In addition, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the aspects described can be practiced without these specific details.

[0018] like Figure 1 As shown, an embodiment of the present invention provides a remote sensing identification method for ground fissures in a complex mining area, which specifically includes the following steps: S1, filtering of non-shadow areas based on sliding windows, brightening of mining area shadows based on HSV color compensation, and eliminating the interference of sun shadows on ground fissure identification.

[0019] This step specifically includes the following sub-steps: S11, set a certain sliding window size, count the number of pixels in each window while sliding the window step by step, and effectively suppress small-scale noise according to the morphological characteristics of the non-shadow area to retain the large shadow area. The specific expression is as follows: ; in, is the pixel value to be retained, is the pixel value of the input image, for The number of black pixels inside, is the empirical threshold. When the above conditions are met, The pixels are covered with white, where the black pixel value is 0 and the white pixel value is 255. The specific definition is: ; in, Indicates Centered at The sliding window Represents a sliding window Each pixel point, is the indicator function.

[0020] For example, in one specific implementation of this embodiment, the sliding window size may be set to 15, and the number of black pixels may be set to 65.

[0021] S12, after completing the non-shadow area filtering, the shadow of the mine area is brightened based on HSV color compensation for the shadow area that meets the conditions. Specifically, the V channel in the HSV space is used to brighten the shadow, and then the H and S channels are locally filtered, so that the color effect of the shadow area is closer to the real color.

[0022] S2, based on the ground fissure evaluation index of linear index and shape index, the ground fissure noise is filtered out to identify the initial ground fissures.

[0023] In this step, the linear index is calculated by The specific expression is as follows: ; in, and They respectively represent the length and width of the minimum circumscribed rectangle of the connected domain where the ground fissure is located. The connected domain refers to the image area composed of points identified as ground fissure pixels in the image.

[0024] Shape index by The specific expression is as follows: ; in, It represents the area of ​​the circumscribed circle of the minimum circumscribed rectangle of the connected domain where the ground fissure is located.

[0025] It should be noted that the larger the linear index is, the more linear it is, and it is judged to be a ground fissure; the smaller the linear index is, the more blocky it is, and it is judged to be noise; the smaller the shape index is, the closer the area is to a line, which means that the area is likely to be a longitudinal or transverse ground fissure.

[0026] The ground fissure area processed by the preliminary segmentation is larger than the small spot noise. By combining the area size with the ground fissure linear index and shape index and setting a certain threshold, it is possible to effectively distinguish linear ground fissures from blocky and small-area noise, which is convenient for subsequent more accurate ground fissure identification. Therefore, in this embodiment, the object pixels identified in this step are called initial ground fissures.

[0027] S3, combining morphological processing with regional connectivity analysis, establishes a comprehensive evaluation and judgment mechanism to accurately restore the true morphology of ground fissures at different levels and scales.

[0028] In this step, the continuity assumption is set: is the gray level of the background pixel, is the average gray value of the initial ground fissure pixels. If , and judge this pixel as a ground fissure object.

[0029] If the above conditions are not met, the following linear assumptions are implemented to achieve hierarchical and multi-scale accurate restoration of the true form of ground fissures: Set a ground fissure compensation template that satisfies the ground fissure relationship. If the pixel is determined to match one of the defined templates, further check the conditions. ,in , is the threshold, the threshold >Threshold If this condition is met, the background pixel is classified as an initial ground crack pixel.

[0030] S4, based on multi-spatial scale threshold segmentation iteration, considers the length of the connected domain, the end direction, and the distance and angle between domains, and performs hierarchical threshold growth. Combined with the automatic extraction mechanism, small noise is eliminated to complete the accurate identification of ground fissures.

[0031] It should be noted that the multi-spatial scale threshold segmentation iteration is mainly to solve the problem of incomplete filtering of small noise. It is necessary to find the endpoint of the ground fissure as the seed point. The specific iterative method is to start from the initial pixel of each ground fissure that meets the linearity and length conditions, and track the ground fissure pixels along the main direction of the connected domain. Finally, the tracked bright pixel endpoint is defined as the seed point.

[0032] Then, the search starts from the seed point and the Bresenham algorithm is used to grow the pixels of the ground fissures to achieve gap connection. The search condition is determined by the radius of the search area. ,angle and the length of the connected domain where the ground fissure is located These three factors jointly determine that the ground fissure extraction conditions are determined by the length of the ground fissure after the gap is connected. It should be noted that the Bresenham algorithm is a classic algorithm used to generate straight lines or arcs in digital image processing. It is widely used in graphics rendering and path tracking. The Bresenham algorithm grows pixels used to accurately track and connect ground fissures and solve the problem of gap connection of ground fissures. Its core advantage lies in the efficient pixel generation method. Under certain search conditions, it can accurately "grow" or "fill" the ground fissure path along the target direction, and finally complete the accurate extraction of ground fissures.

[0033] For search criteria, set ; ; ; ; in, , , Respectively represent the radius, length and angle initial constraint values ​​of the search area, The length condition for extracting ground fissures is: , , are iterative thresholds, is the iteration termination condition, Indicates the number of iterations.

[0034] It should be noted that, in practical applications, the above threshold settings should be made according to actual conditions, and the growth search range should be continuously expanded and the ground fissure extraction range should be refined until the extraction size does not meet the definition of ground fissures, and the growth and extraction iterations are stopped. This solves the problem of incomplete filtering of the above-mentioned small noise, and not only achieves the effective removal of small noise in complex mining areas, but also ensures the complete extraction of large-size ground fissures, realizing the automation of multi-spatial scale threshold segmentation iteration.

[0035] For example, in one specific implementation of this embodiment, the iteration threshold Set to 5, , , Set to 60, 220 and 30 respectively.

[0036] As described above, the remote sensing identification method for ground fissures in complex mining areas provided in this embodiment, on the one hand, brightens the shadow of the mining area, performs tone compensation based on the HSV tone difference between the shadow area and the real image, and eliminates the interference of the sun shadow on the extraction of ground fissures; on the other hand, two ground fissure evaluation indicators are proposed and combined with connected area analysis to filter out noise, and crack morphology processing based on continuity and linear assumptions is used to restore the true shape of the cracks. Based on multi-spatial scale threshold segmentation iteration, the length of the connected domain, the end direction and the distance and angle between the connected domains are considered to perform threshold layered growth, and automatic extraction is set to achieve noise removal and complete identification of ground fissures, providing new technical support for the identification of ground fissures in complex mining areas.

[0037] The feasibility of this method is further verified by a specific case. A mining area in a certain place is used as a demonstration area, and drone remote sensing image data covering the mining area in February 2023 is selected. In order to reflect the advantages of this method, the accuracy, completeness and kappa coefficient of the traditional mining area ground fissure extraction method based on digital image processing technology are compared with this method (where the kappa coefficient is an indicator of consistency test, which is used to measure whether the ground fissure identification result is consistent with the actual situation). The results are as follows Figure 2 As shown. Figure 2 It can be seen that compared with the traditional method, this method can greatly improve the accuracy of ground fissure identification. Its completeness is improved from 31.46% of the traditional method to 60.96%, its accuracy is improved from 28.06% to 83.79%, and its kappa coefficient is improved from 28.15% to 70.07%. Therefore, this method can be used to accurately identify ground fissures in mining areas under the influence of large deformation and complex surface environment.

[0038] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A remote sensing identification method for ground fissures in complex mining areas, characterized in that: The steps include: S1, filtering the non-shadow area based on the sliding window, brightening the shadow of the mining area based on HSV color compensation, and eliminating the interference of the sun's shadow on the identification of ground fissures; S2, based on the ground fissure evaluation index of linear index and shape index, the ground fissure noise is filtered out and the initial ground fissure is identified; S3, combining morphological processing with regional connectivity analysis, establishes a comprehensive evaluation and judgment mechanism to accurately restore the true morphology of ground fissures at different levels and scales; S4, based on multi-spatial scale threshold segmentation iteration, considers the length of the connected domain, the end direction, the distance and angle between domains, performs hierarchical threshold growth, and combines the automatic extraction mechanism to eliminate small noise to complete the accurate identification of ground fissures.

2. A remote sensing identification method for ground fissures in complex mining areas according to claim 1, characterized in that: S1 includes the following sub-steps: S11, setting the sliding window size, counting the number of pixels in each window when sliding the window step by step, suppressing small-scale noise according to the morphological characteristics of the non-shadow area to retain the large shadow area; S12, for the shadow areas that meet the conditions, the mine shadows are brightened based on HSV color compensation, the V channel in the HSV space is used to brighten the shadows, and then the H and S channels are locally filtered to make the color effect of the shadow area closer to the real color.

3. A remote sensing identification method for ground fissures in complex mining areas according to claim 2, characterized in that: S11 is expressed by the following formula: ； in, is the pixel value to be retained, is the pixel value of the input image, for The number of black pixels inside, is the empirical threshold. When the above conditions are met, The pixel coverage is white, the black pixel value is 0, and the white pixel value is 255. The specific definition is: ； in, Indicates Centered at The sliding window Represents a sliding window Each pixel point, is the indicator function.

4. A remote sensing identification method for ground fissures in complex mining areas according to claim 2, characterized in that: S2 linear index through express: ； in, and They represent the length and width of the minimum circumscribed rectangle of the connected domain where the ground fissure is located. The connected domain is the image area composed of points identified as ground fissure pixels in the image; Shape index by express: ； in, It represents the area of ​​the circumscribed circle of the minimum circumscribed rectangle of the connected domain where the ground fissure is located.

5. A remote sensing identification method for ground fissures in complex mining areas according to claim 4, characterized in that: The continuity assumption is set in S3: is the gray level of the background pixel, is the average gray value of the initial ground fissure pixels. If , the pixel is determined as a ground fissure object; If the above conditions are not met, the following linear assumption is performed: a ground fissure compensation template that satisfies the ground fissure relationship is set, and if the pixel is determined to match one of the defined templates, the condition is further checked. ,in , is the threshold, the threshold >Threshold , if satisfied, the background pixel is classified as an initial ground crack pixel.

6. A remote sensing identification method for ground fissures in complex mining areas according to claim 5, characterized in that: In S4, starting from the initial pixel of each ground fissure that meets the linearity and length conditions, the ground fissure pixels are tracked along the main direction of the connected domain, and the bright pixel endpoint finally tracked is defined as the seed point; Then, the search starts from the seed point and the Bresenham algorithm is used to grow the ground fissure pixels to achieve gap connection. The search conditions include the radius of the search area. ,angle and the length of the connected domain where the ground fissure is located , the ground fissure extraction conditions include the length of the ground fissure after the gap is connected .

7. A remote sensing identification method for ground fissures in complex mining areas according to claim 6, characterized in that: For search conditions, ； ； ； ； in, , , Respectively represent the radius, length and angle initial constraint values ​​of the search area, The length condition for extracting ground fissures is: , , are iterative thresholds, is the iteration termination condition, Indicates the number of iterations.

Images