Vegetation detection method and system for soil environment restoration

Abstract

The invention relates to the technical field of soil image processing, in particular to a vegetation detection method and system for soil environment restoration. According to the method, color features and position distribution features of pixel points are analyzed, a vegetation coverage factor of each superpixel block is obtained, and vegetation coverage blocks are screened out; according to the position distribution of the pixel points in each vegetation coverage block and the chromaticity fluctuation characteristics, the artificial cultivation characteristic degree of each vegetation coverage block is obtained; according to the centroid position and the vegetation coverage factor of each superpixel block, obtaining a vegetation coverage class cluster and a non-vegetation coverage class cluster; and in combination with the relative position between each vegetation coverage block in each non-vegetation coverage class cluster and each vegetation coverage block in the adjacent vegetation coverage class cluster, obtaining the soil remediation vegetation significance of each super-pixel block, and further obtaining the soil remediation weighted significance of each super-pixel block. According to the method, the vegetation characteristics in the soil environment restoration process are accurately analyzed, so that the accuracy of soil environment restoration assessment is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of soil image processing, and in particular to a vegetation detection method and system for soil environmental remediation. Background Art

[0002] Soil remediation is a key measure for improving the quality of contaminated soil, and vegetation restoration is a key indicator for evaluating soil remediation effectiveness. Vegetation plays a crucial role in soil remediation, serving not only as an indicator of contamination but also as a tool for reducing pollutant levels through phytoremediation. Therefore, vegetation monitoring is a crucial tool for evaluating soil remediation effectiveness.

[0003] In the existing technology, the results of soil environmental remediation are evaluated by relying on a single vegetation index or spectral feature. However, the evaluation process is negatively affected by the inconsistent scale differences of different types of natural vegetation areas. The spatial distribution characteristics of the vegetation coverage area are not taken into account, resulting in poor accuracy of local soil environmental remediation assessment. Summary of the Invention

[0004] In order to solve the technical problem of poor accuracy in soil environmental remediation assessment due to failure to consider the spatial distribution characteristics of vegetation coverage areas, the present invention aims to provide a vegetation detection method and system for soil environmental remediation. The technical solutions adopted are as follows:

[0005] The present invention proposes a vegetation detection method for soil environmental remediation, the method comprising:

[0006] Obtain remote sensing images of soil areas;

[0007] Multiple superpixel blocks of the remote sensing image are obtained. Based on the color characteristics and position distribution characteristics of the pixels in each superpixel block, the vegetation cover factor of each superpixel block is obtained, and the vegetation cover blocks are screened out. Based on the position distribution and chromaticity fluctuation characteristics of the pixels in each vegetation cover block, the artificial cultivation characteristic degree of each vegetation cover block is obtained.

[0008] According to the centroid position and vegetation coverage factor of each superpixel block, all superpixel blocks are clustered to obtain block clustering clusters; according to the distribution of vegetation coverage factors of all superpixel blocks in different block clustering clusters, vegetation coverage clusters and non-vegetation coverage clusters are obtained;

[0009] Based on the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, the artificial cultivation characteristic degree and vegetation cover factor of the corresponding vegetation cover block, the soil restoration vegetation significance of each superpixel block is obtained; based on the soil restoration vegetation significance of each superpixel block, the significant distance between each superpixel block and different superpixel blocks, and the vegetation cover factors of other superpixel blocks, the soil restoration weighted significance of each superpixel block is obtained;

[0010] Vegetation is detected based on the weighted significance of soil restoration across all superpixel patches.

[0011] Furthermore, the method for obtaining the vegetation cover factor includes:

[0012] For any superpixel block, the over-green minus over-red index of each pixel is obtained and normalized and mapped as the vegetation color feature;

[0013] The CANNY algorithm is used to perform edge detection on superpixel blocks to obtain multiple edge lines. The slope variance of all edge pixels on the edge line where each edge pixel lies is obtained as the slope fluctuation feature of each edge pixel. The edge pixel with the smallest relative distance to each pixel is selected, and the slope fluctuation feature of the corresponding edge pixel is used as the local texture roughness coefficient of each pixel.

[0014] The vegetation coverage factor of each superpixel block is obtained according to the saturation channel value, vegetation color characteristics and local texture roughness coefficient of different pixel points in the HSV space. The saturation channel value is negatively correlated with the vegetation coverage factor, while the vegetation color characteristics and local texture roughness coefficient are both positively correlated with the vegetation coverage factor.

[0015] Furthermore, the method for obtaining the vegetation coverage blocks includes:

[0016] If the vegetation coverage factor of a superpixel block is greater than or equal to a preset coverage threshold, the corresponding superpixel block is used as a vegetation coverage block.

[0017] Furthermore, the method for obtaining the artificial cultivation characteristic degree includes:

[0018] For each vegetation cover block, the number of edge pixels on each edge line is counted as the length of the corresponding edge line; the mean difference in slope between each boundary pixel and the adjacent boundary pixel is obtained as the slope change rate of each boundary pixel;

[0019] The length of the edge line or the slope change rate of the boundary pixel point is used as the analysis data. The difference between adjacent analysis data is calculated in descending order. The maximum value of the corresponding element when the difference is the largest is selected as the data reference value. If the analysis data is greater than the analysis reference value, the edge line corresponding to the analysis data is used as the long edge line, or the boundary pixel point corresponding to the analysis data is used as the boundary segmentation point to obtain multiple block boundaries.

[0020] The PCA algorithm was used to obtain the principal component direction of each long edge line and block boundary, and the angle between the principal component direction and the horizontal direction. The artificial cultivation characteristic of each vegetation coverage block was obtained based on the minimum difference in the corresponding angle between each long edge line and different block boundaries in each vegetation coverage block, as well as the fluctuation degree of the chromaticity channel values of all pixel points in the HSV space. The minimum difference and fluctuation degree of the angle were positively correlated with the degree of artificial cultivation.

[0021] Furthermore, the method for obtaining the block clustering clusters includes:

[0022] The centroid coordinates and vegetation cover factor of each superpixel block are used to construct multidimensional data, and the Euclidean distance of the multidimensional data between superpixel blocks is obtained. K-means clustering is performed on all superpixel blocks to obtain block clusters.

[0023] Furthermore, the method for obtaining the vegetation cover cluster and the non-vegetation cover cluster includes:

[0024] Obtain the mean vegetation coverage factor of all superpixel blocks in each block cluster as the overall vegetation coverage level;

[0025] The differences in adjacent overall vegetation coverage levels are calculated in order from large to small, and the maximum value in the corresponding element when the difference is the largest is selected as the coverage level reference value; if the overall vegetation coverage level of a block cluster is greater than the coverage level reference value, the corresponding block cluster will be regarded as the vegetation coverage cluster, and the remaining block clusters will be regarded as non-vegetation coverage clusters.

[0026] Furthermore, the method for obtaining the soil remediation vegetation significance includes:

[0027] The soil environmental restoration degree of each non-vegetation cover cluster is obtained based on the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, as well as the vegetation cover factor of the corresponding vegetation cover block.

[0028] Obtain the relative distance between the centroid of each vegetation-covered block and each non-vegetation-covered cluster within the corresponding block area, select the minimum value of the relative distance between each vegetation-covered block and different non-vegetation-covered clusters, and use the corresponding non-vegetation-covered cluster as the reference cluster for each vegetation-covered block;

[0029] The artificial cultivation characteristic degree of each vegetation coverage block and the product of the soil environmental restoration degree of the corresponding reference cluster are obtained as the soil restoration significance coefficient; the ratio of the soil restoration significance coefficient of each vegetation coverage block and the minimum relative distance is obtained as the soil restoration vegetation significance of each vegetation coverage block; if the superpixel block is not a vegetation coverage block, the soil restoration vegetation significance of the corresponding superpixel block is set to a positive integer 1.

[0030] Furthermore, the method for obtaining the soil environmental remediation degree includes:

[0031] Obtain the relative distance distribution between the centroid of each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster as the restoration edge distance; construct a restoration edge distance sequence by sorting the corresponding modified edge distances between each non-vegetation cover cluster and all vegetation cover blocks in the adjacent vegetation cover cluster in ascending order;

[0032] According to the distribution order of vegetation coverage blocks corresponding to each element in the restoration edge distance sequence, a vegetation coverage factor sequence corresponding to all vegetation coverage blocks is constructed;

[0033] The correlation coefficient between the restoration edge distance series and the vegetation cover factor series was obtained and normalized as the soil environmental restoration degree of each non-vegetation covered area.

[0034] Furthermore, the method for obtaining the weighted significance of soil remediation includes:

[0035] Obtain the significant distance between each superpixel block and different other superpixel blocks, calculate the product of the significant distance and the vegetation cover factor of other superpixel blocks, obtain the product of the product of the product and the soil restoration vegetation significance value of each superpixel block, and perform positive correlation normalization as the weighted significance of soil restoration of each superpixel block.

[0036] The present invention also proposes a vegetation detection system for soil environment remediation, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements any one of the steps of the vegetation detection method for soil environment remediation.

[0037] The present invention has the following beneficial effects:

[0038] The present invention obtains multiple super-pixel blocks of remote sensing images, obtains the vegetation coverage factor of each super-pixel block according to the color characteristics and position distribution characteristics of the pixels in each super-pixel block, and screens out the vegetation coverage blocks, which can preliminarily screen out vegetation-related areas and reduce the interference of non-vegetation areas such as ground shadows, bare soil and buildings; obtains the artificial cultivation characteristic degree of each vegetation coverage block according to the position distribution of the pixels in each vegetation coverage block and the chromaticity fluctuation characteristics, and more accurately evaluates the contribution of artificial vegetation in the soil remediation area; and according to the centroid position and vegetation coverage factor of each super-pixel block, the vegetation coverage factor of each super-pixel block is screened out. All superpixel blocks are clustered to obtain block clusters. Through cluster analysis, superpixel blocks with similar centroid positions and vegetation cover factors are divided into the same cluster, which can better reflect the spatial distribution characteristics of vegetation. According to the distribution of vegetation cover factors of all superpixel blocks in different block clusters, vegetation cover clusters and non-vegetation cover clusters are obtained. According to the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, the artificial cultivation characteristic degree and vegetation cover factor of the corresponding vegetation cover block are used to obtain the soil remediation vegetation significance of each superpixel block. By quantifying the significance, vegetation areas with important significance for soil remediation can be identified. According to the soil remediation vegetation significance of each superpixel block, the significant distance between each superpixel block and different other superpixel blocks, and the vegetation cover factors of other superpixel blocks, the weighted soil remediation significance of each superpixel block is obtained, which helps to accurately reflect the overall effect of soil remediation. The present invention improves the accuracy of soil environment remediation assessment by accurately analyzing vegetation characteristics during the soil environment remediation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 A flow chart of a vegetation detection method for soil environmental remediation provided by one embodiment of the present invention;

[0041] Figure 2 A flow chart of a method for obtaining a vegetation cover factor provided by one embodiment of the present invention;

[0042] Figure 3 A flow chart of a method for obtaining artificial cultivation characteristics provided in one embodiment of the invention;

[0043] Figure 4A flow chart of a method for obtaining soil remediation vegetation significance provided in one embodiment of the invention. DETAILED DESCRIPTION

[0044] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a vegetation detection method and system for soil environmental remediation proposed in accordance with the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] The following describes in detail a vegetation detection method and system for soil environmental remediation provided by the present invention with reference to the accompanying drawings.

[0047] See also Figure 1 , which shows a method flow chart of a vegetation detection method for soil environmental remediation provided by an embodiment of the present invention, specifically comprising:

[0048] Step S1: Acquire a remote sensing image of a soil area.

[0049] In an embodiment of the present invention, remote sensing images can quickly cover a large area of soil, provide a global perspective, facilitate identification of the distribution of vegetation-covered and non-vegetation-covered areas, and help evaluate the growth status of vegetation in soil environmental restoration areas; first, remote sensing images of the soil area environment are collected using aerial photography equipment when the weather is clear.

[0050] It should be noted that, in order to facilitate subsequent image processing, the image is filtered and denoised through non-local mean filtering, which helps to improve the quality of the image, highlight the outline and details of the image, and make the image clearer; the non-local mean filtering algorithm is a technical means well known to those skilled in the art and will not be described in detail here.

[0051] Step S2: Obtain multiple superpixel blocks of the remote sensing image, obtain the vegetation coverage factor of each superpixel block based on the color characteristics and position distribution characteristics of the pixels in each superpixel block, and screen out the vegetation coverage blocks; obtain the artificial cultivation characteristic degree of each vegetation coverage block based on the position distribution of the pixels in each vegetation coverage block and the chromaticity fluctuation characteristics.

[0052] Remote sensing images contain a large number of pixels. Directly processing the pixels is not only computationally intensive but also prone to introducing noise. In order to reduce computational complexity, dividing the image into multiple superpixel blocks helps simplify the image analysis and understanding process. Multiple superpixel blocks of remote sensing images are obtained.

[0053] It should be noted that, in one embodiment of the present invention, a remote sensing image is segmented using a superpixel segmentation algorithm to obtain multiple superpixel blocks of the remote sensing image. Superpixel segmentation can aggregate pixels in an image into several larger pixel blocks, i.e., superpixel blocks, based on similarity criteria such as color, texture, or shape, thereby significantly reducing the dimensionality of image processing while retaining important image information. The specific superpixel segmentation algorithm is a technical means well known to those skilled in the art and will not be described in detail here.

[0054] The soil itself is darker in color and has a higher saturation than vegetation. Artificial objects often have regular geometric shapes and a more uniform texture distribution compared to vegetation areas. By analyzing color features and position distribution features, vegetation usually has specific spectral characteristics in remote sensing images. Color features can be used to effectively distinguish vegetation from non-vegetated areas. Position distribution features help understand the possibility of vegetation growth and texture distribution. Based on the color features and position distribution features of the pixels in each superpixel block, the vegetation cover factor of each superpixel block is obtained.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining the vegetation coverage factor is as follows: Figure 2 , which shows a flow chart of a method for obtaining a vegetation cover factor, including:

[0056] Step S201: For any superpixel block, obtain the over-green minus over-red index of each pixel point, and perform normalized mapping as the vegetation color feature.

[0057] The excessive green minus excessive red index helps to further highlight the green characteristics of vegetation, while suppressing the interference of non-vegetation areas such as soil and bare land, and can effectively enhance the contrast between vegetation and non-vegetation areas.

[0058] It should be noted that the specific over-green minus over-red index is a technical means well known to those skilled in the art and will not be described in detail here.

[0059] Step S202: Use the CANNY algorithm to perform edge detection processing on the superpixel blocks to obtain multiple edge lines; obtain the slope variance of all edge pixels on the edge line where each edge pixel is located as the slope fluctuation feature of each edge pixel; select the edge pixel with the smallest relative distance to each pixel point, and use the slope fluctuation feature of the corresponding edge pixel point as the local texture roughness coefficient of each pixel point.

[0060] The CANNY algorithm is a classic edge detection algorithm that effectively extracts edge information from images. By performing edge detection on superpixel blocks, clear edge lines can be obtained. The slope variation of pixels along the edge line reflects the smoothness or roughness of the edge. By assigning the slope fluctuation characteristics of the nearest edge pixel to each pixel, the roughness information of the edge can be extended to the entire area, thereby describing the roughness of the local texture.

[0061] It should be noted that, in some embodiments of the present invention, the relative distance can be obtained by existing distance calculation methods such as Euclidean distance or Manhattan distance. The specific means are technical means well known to those skilled in the art and will not be described in detail here.

[0062] It should be noted that, in one embodiment of the present invention, the method for obtaining the slope is to obtain the ratio of the vertical coordinate difference and the horizontal coordinate difference between each edge pixel point and the previous adjacent edge pixel point as the slope of each edge pixel point; the specific CANNY algorithm is a technical means well known to those skilled in the art and will not be elaborated here.

[0063] Step S203: According to the saturation channel value, vegetation color characteristics and local texture roughness coefficient of different pixel points in the HSV space, the vegetation coverage factor of each superpixel block is obtained. The saturation channel value is negatively correlated with the vegetation coverage factor, and the vegetation color characteristics and local texture roughness coefficient are both positively correlated with the vegetation coverage factor.

[0064] In one embodiment of the present invention, for each superpixel block, the vegetation cover factor is expressed as follows:

[0065]

[0066] Where C represents the vegetation coverage factor of each superpixel block; represents the local texture roughness coefficient of the i-th pixel in each superpixel block; S i represents the saturation channel value of the i-th pixel in the HSV space; n represents the number of pixels in the superpixel block; E i Represents the i-th vegetation color feature; norm() represents the normalization function.

[0067] In the formula of the vegetation cover factor, the smaller the local texture roughness coefficient of the i-th pixel in each superpixel block, the neater the texture distribution, the larger the saturation channel value of the i-th pixel in the HSV space, the greater the possibility that it is soil, the more object interference there is, the more the vegetation color characteristics are interfered with by other objects, the vegetation color characteristics are relatively reduced, and the smaller the vegetation cover factor.

[0068] The vegetation coverage factor can reflect the degree of vegetation coverage in a certain area. The larger the vegetation coverage factor, the more likely it is a vegetation covered block. By analyzing the vegetation coverage factor, vegetation covered blocks can be screened out.

[0069] Advantageously, in one embodiment of the present invention, the method for obtaining vegetation coverage blocks includes:

[0070] If the vegetation coverage factor of a superpixel block is greater than or equal to a preset coverage threshold, the corresponding superpixel block is used as a vegetation coverage block.

[0071] It should be noted that, in one embodiment of the present invention, the preset coverage threshold is set to 0.65; in other embodiments of the present invention, the preset coverage threshold may be set according to specific circumstances, which is not limited or elaborated herein.

[0072] In the process of soil environmental remediation, artificial planting of vegetation is a common method. Compared with naturally grown vegetation areas, the plant species in artificial vegetation areas are relatively single, the colors are relatively consistent, the distribution is more even, and they often appear in neat and uniform rows. Therefore, the artificial cultivation characteristic degree of each vegetation coverage block is obtained based on the position distribution of pixel points in each vegetation coverage block and the chromaticity fluctuation characteristics.

[0073] Preferably, in one embodiment of the present invention, the method for obtaining the artificial cultivation characteristic degree can be found in Figure 3 , which shows a flow chart of a method for obtaining artificial cultivation characteristics, including:

[0074] Step S301: For each vegetation cover block, count the number of edge pixels on each edge line as the length of the corresponding edge line; obtain the mean difference in slope between each boundary pixel and the adjacent boundary pixel points as the slope change rate of each boundary pixel point.

[0075] The length of the edge line reflects the continuity and significance of the edge. In the vegetation cover block, the long edge line may correspond to the boundary or internal structure of the vegetation area, which is used to quantify the significance of the edge line. The boundary refers to the edge line that intersects between blocks. The slope change rate of the boundary pixel point reflects the smoothness or tortuosity of the boundary.

[0076] Step S302: Use the length of the edge line or the slope change rate of the boundary pixel point as the analysis data, calculate the difference between adjacent analysis data in order from large to small, and select the maximum value in the corresponding element when the difference is the largest as the data reference value. If the analysis data is greater than the analysis reference value, use the edge line corresponding to the analysis data as the long edge line, or use the boundary pixel point corresponding to the analysis data as the boundary segmentation point to obtain multiple block boundaries.

[0077] By calculating the differences between adjacent analysis data, we can find the mutation points of the data distribution, which usually correspond to significant feature changes; selecting the maximum value when the difference is the largest as the reference value can ensure that the screened edge lines or boundary points are significant.

[0078] Step S303: Use the PCA algorithm to obtain the principal component direction of each long edge line and block boundary, and obtain the angle between the principal component direction and the horizontal direction; according to the minimum difference in the corresponding angles of the principal component directions between each long edge line and different block boundaries in each vegetation coverage block, and the degree of fluctuation of the chromaticity channel values of all pixel points in the HSV space, obtain the artificial cultivation characteristic of each vegetation coverage block. The minimum difference and the degree of fluctuation of the angle are positively correlated with the degree of artificial cultivation.

[0079] PCA can extract the main change direction of the data and is used to quantify the directional characteristics of edge lines or boundaries. In vegetation coverage blocks, the direction of the principal component can reflect the overall direction of vegetation arrangement. The angle reflects the degree of deviation of the direction of the edge line or boundary from the horizontal direction. The fluctuation degree of the chromaticity channel value reflects the uniformity of vegetation color. The more consistent the color distribution, the greater the degree of artificial cultivation.

[0080] It should be noted that, in one embodiment of the present invention, the degree of fluctuation can be expressed by calculating the variance. The larger the variance, the greater the degree of fluctuation, and the smaller the variance, the smaller the degree of fluctuation. In other embodiments of the present invention, the degree of fluctuation can also be expressed by calculating the range or standard deviation. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0081] In one embodiment of the present invention, for each vegetation coverage block, the formula for the artificial cultivation characteristic degree is expressed as:

[0082]

[0083] Among them, P represents the artificial cultivation characteristic degree of each vegetation cover block; Represents the variance of the chromaticity channel values of all pixels in each vegetation coverage block in the HSV space, that is, the degree of fluctuation; θ j represents the angle between the principal component direction of the jth long edge line and the horizontal direction; α represents the angle between the principal component direction of the block boundary and the horizontal direction; min|θ j -α| represents the minimum difference in the angle between the j-th long edge line and the principal component directions of different block boundaries; m represents the number of long edge lines.

[0084] In the formula of artificial cultivation characteristic degree, Add 0.01 to avoid the denominator of the formula being 0, which would make the formula meaningless; The larger the value, the greater the chromaticity fluctuation characteristics, the more uneven the distribution, and the less likely it is artificial cultivation; the smaller the minimum difference in the angle between the j-th long edge line and the principal component direction corresponding to the boundary of different blocks, the more consistent the principal component direction, the more likely it is cultivated in the same direction, and the greater the possibility of artificial vegetation.

[0085] Step S3: Cluster all superpixel blocks according to the centroid position and vegetation coverage factor of each superpixel block to obtain block clustering clusters; obtain vegetation coverage clusters and non-vegetation coverage clusters according to the distribution of vegetation coverage factors of all superpixel blocks in different block clustering clusters.

[0086] The centroid is the geometric center of each superpixel block, and the vegetation cover factor indicates the vegetation density or coverage ratio within the superpixel block. Combining these two types of data to form multidimensional data helps to simultaneously consider spatial location and vegetation cover. Through clustering, areas with similar vegetation cover patterns can be identified. All superpixel blocks are clustered based on the centroid position and vegetation cover factor of each superpixel block to obtain block clusters.

[0087] Preferably, in one embodiment of the present invention, the method for obtaining block clustering includes:

[0088] The centroid coordinates and vegetation cover factor of each superpixel block are used to construct multidimensional data, and the Euclidean distance of the multidimensional data between superpixel blocks is obtained. K-means clustering is performed on all superpixel blocks to obtain block clusters.

[0089] It should be noted that K-means clustering can divide superpixel blocks into K clusters, so that the points in the same cluster have high similarity and the differences between different clusters are large, which helps to identify areas with similar vegetation coverage and spatial distribution patterns. Among them, the K value is determined by the elbow rule. The specific K-means clustering is a technical means well known to those skilled in the art and will not be elaborated here.

[0090] It should be noted that, in other embodiments of the present invention, existing distance calculation methods such as the Manhattan distance between multidimensional data may also be used. The specific Euclidean distance and Manhattan distance are technical means well known to those skilled in the art and will not be described in detail here.

[0091] The vegetation coverage factor reflects the degree of vegetation coverage in the superpixel block and quantifies the vegetation coverage of the superpixel block. The higher the vegetation coverage factor, the higher the vegetation coverage in the block. According to the distribution of vegetation coverage factors of all superpixel blocks in different block clusters, vegetation coverage clusters and non-vegetation coverage clusters are obtained.

[0092] Preferably, in one embodiment of the present invention, the method for obtaining vegetation cover clusters and non-vegetation cover clusters includes:

[0093] Obtain the mean vegetation coverage factor of all superpixel blocks in each block cluster as the overall vegetation coverage level;

[0094] The differences in adjacent overall vegetation coverage levels are calculated in order from large to small, and the maximum value in the corresponding element when the difference is the largest is selected as the coverage level reference value; if the overall vegetation coverage level of a block cluster is greater than the coverage level reference value, the corresponding block cluster will be regarded as the vegetation coverage cluster, and the remaining block clusters will be regarded as non-vegetation coverage clusters.

[0095] Step S4: According to the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, the artificial cultivation characteristic degree and vegetation cover factor of the corresponding vegetation cover block, the soil restoration vegetation significance of each superpixel block is obtained; according to the soil restoration vegetation significance of each superpixel block, the significant distance between each superpixel block and different other superpixel blocks and the vegetation cover factors of other superpixel blocks, the soil restoration weighted significance of each superpixel block is obtained.

[0096] The relative position distribution is the spatial relationship between the non-vegetation cover cluster and the vegetation cover blocks in the adjacent vegetation cover cluster, reflecting the spatial impact of the vegetation cover blocks on the non-vegetation cover cluster. The closer the vegetation cover blocks are, the greater their contribution to the soil remediation of the non-vegetation cover cluster may be. The artificial cultivation characteristic and vegetation cover factor both reflect the possibility of artificial planting of plants in the vegetation cover blocks. According to the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, the soil remediation vegetation significance of each superpixel block is obtained.

[0097] Preferably, in one embodiment of the present invention, the method for obtaining soil remediation vegetation significance is as follows: Figure 4 , which shows a flow chart of a method for obtaining soil vegetation significance, including:

[0098] Step S401: Obtain the soil environmental restoration degree of each non-vegetation cover cluster based on the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, and the vegetation cover factor of the corresponding vegetation cover block.

[0099] Preferably, in one embodiment of the present invention, the method for obtaining the soil environmental remediation degree includes:

[0100] Obtain the relative distance distribution between the centroid of each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster as the restoration edge distance; construct a restoration edge distance sequence by sorting the corresponding modified edge distances between each non-vegetation cover cluster and all vegetation cover blocks in the adjacent vegetation cover cluster in ascending order;

[0101] According to the distribution order of vegetation coverage blocks corresponding to each element in the restoration edge distance sequence, a vegetation coverage factor sequence corresponding to all vegetation coverage blocks is constructed;

[0102] The correlation coefficient between the restoration edge distance series and the vegetation cover factor series was obtained and normalized as the soil environmental restoration degree of each non-vegetation covered area.

[0103] It should be noted that, in one embodiment of the present invention, the correlation coefficient between sequences can be obtained by calculating the Pearson correlation coefficient. The larger the correlation coefficient, the closer the sequences are, and the more it can reflect the progress of soil environmental remediation. The correlation coefficient is normalized using the sigmoid algorithm to obtain the soil environmental remediation degree in non-vegetation covered areas. In other embodiments of the present invention, the correlation coefficient can also be calculated using cosine similarity, and the correlation coefficient can be normalized to achieve normalization. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0104] Step S402: Obtain the relative distance between the centroid of each vegetation-covered block and each non-vegetation-covered cluster within the corresponding block area, select the minimum value of the relative distance between each vegetation-covered block and different non-vegetation-covered clusters, and use the corresponding non-vegetation-covered cluster as the reference cluster for each vegetation-covered block.

[0105] The spatial relationship between vegetation-covered blocks and non-vegetation-covered blocks was quantified by the relative distance between the centroids. The smaller the relative distance, the closer the blocks were, and the more the soil restoration characteristics of the vegetation-covered blocks could be demonstrated.

[0106] Step S403: Obtain the product of the artificial cultivation characteristic degree of each vegetation coverage block and the soil environmental restoration degree of the corresponding reference cluster as the soil restoration significance coefficient; obtain the ratio of the soil restoration significance coefficient of each vegetation coverage block and the minimum relative distance as the soil restoration vegetation significance of each vegetation coverage block; if the superpixel block is not a vegetation coverage block, set the soil restoration vegetation significance of the corresponding superpixel block to a positive integer 1.

[0107] The artificial cultivation characteristic index reflects the degree of artificial cultivation in the vegetation coverage block. The larger the artificial cultivation characteristic index is, the more likely the coverage block is a population planting area and the greater the soil vegetation significance is.

[0108] In one embodiment of the present invention, for each vegetation coverage block, the formula for soil remediation vegetation significance is:

[0109]

[0110] Where X represents the soil restoration vegetation significance of each vegetation cover block; P represents the artificial cultivation characteristic of each vegetation cover block; min(d) represents the minimum relative distance between each vegetation cover block and the centroid of the corresponding block area of different non-vegetation cover clusters; F represents the soil environmental restoration degree of the corresponding non-vegetation cover cluster when the relative distance between each vegetation cover block and different non-vegetation cover clusters is the minimum.

[0111] In the formula for soil remediation vegetation significance, the smaller the minimum relative distance, the closer the vegetation coverage block is to the centroid of the non-vegetation coverage area. The closer the distance, the more likely it is that the soil remediation characteristics are similar, and the vegetation growth situation can more effectively reflect the soil remediation progress of the vegetation coverage block. Therefore, the greater the soil environment remediation degree and the greater the artificial cultivation characteristic degree, the greater the soil remediation characteristics.

[0112] The weighted soil restoration significance of each superpixel block is obtained based on the soil restoration vegetation significance of each superpixel block, the significant distance between each superpixel block and different other superpixel blocks, and the vegetation coverage factor of other superpixel blocks.

[0113] Preferably, in one embodiment of the present invention, the method for obtaining the weighted significance of soil remediation includes:

[0114] Obtain the significant distance between each superpixel block and different other superpixel blocks, calculate the product of the significant distance and the vegetation cover factor of other superpixel blocks, obtain the product of the product of the product and the soil restoration vegetation significance value of each superpixel block, and perform positive correlation normalization as the weighted significance of soil restoration of each superpixel block.

[0115] It should be noted that, in an embodiment of the present invention, the method for obtaining the saliency distance includes: obtaining the saliency information of the superpixel block through a saliency detection method, and weighting the Euclidean distance of the center coordinates between the superpixel blocks according to the difference in saliency information as the saliency distance; the specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0116] In one embodiment of the present invention, for each superpixel block, the formula for weighted significance of soil remediation is expressed as:

[0117]

[0118] Where S represents the weighted significance of soil restoration in each superpixel block; X represents the significance of soil restoration vegetation in each superpixel block; C k represents the vegetation coverage factor of the kth other superpixel block; d k represents the significant distance between each superpixel block and the kth other superpixel block; K represents the number of other superpixel blocks; exp() represents an exponential function with a natural constant as the base.

[0119] In the formula of weighted significance of soil remediation, the exponential function with natural constant as the base is used to convert Perform negative correlation mapping; It represents the product of the significant distance between each superpixel block and other superpixel blocks and the vegetation coverage factor of the corresponding superpixel blocks. The larger the product accumulation value, the larger the significant distance between blocks, the greater the difference in performance characteristics between blocks, the larger the vegetation coverage factor of other superpixel blocks, the greater the credibility of the significant distance between blocks, and the greater the significance of soil remediation reflected by the corresponding superpixel block.

[0120] Step S5: Detect vegetation based on the weighted significance of soil restoration in each superpixel block.

[0121] It should be noted that, in another embodiment of the present invention, after obtaining the weighted significance of soil remediation of all superpixel blocks, the vegetation is detected, including: at different preset scales, each pixel in all superpixel blocks is analyzed based on the existing CA significance detection algorithm to obtain the soil remediation significance update of each pixel at each preset scale; obtaining the average of the soil remediation significance update of each pixel at all preset scales as the overall vegetation significance value of each pixel; the preset scale set is {100%, 80%, 50%, 30%}. In other embodiments of the present invention, the preset scale can be set according to the specific situation. The specific CA significance detection algorithm is a technical means well known to those skilled in the art and will not be elaborated here.

[0122] The overall vegetation significance value of each pixel is used as analysis data. The corresponding pixel when the analysis data is greater than the analysis reference value is obtained as the vegetation pixel of interest. Morphological closing operations are performed on all vegetation pixels of interest to obtain the vegetation region of interest. The real-time vegetation condition index and vegetation health index of the vegetation region of interest are obtained through the normalized vegetation index. The product of the vegetation condition index and vegetation health index is obtained as the real-time soil environmental remediation completion degree of the vegetation region of interest. The specific morphological closing operations and normalized vegetation index are well known to those skilled in the art and are not detailed here.

[0123] Based on this, by processing the accurate weighted significance of soil remediation in vegetation areas, identifying vegetation areas of interest, and analyzing the health status of corresponding vegetation, it is helpful to more accurately evaluate the soil environmental remediation situation. The higher the degree of completion of soil environmental remediation, the better the effect of soil environmental remediation.

[0124] In summary, the present invention obtains the vegetation coverage factor of each superpixel block by analyzing the color characteristics and position distribution characteristics of pixel points, and screens out vegetation coverage blocks; obtains the artificial cultivation characteristic degree of each vegetation coverage block based on the position distribution of pixels in each vegetation coverage block and the chromaticity fluctuation characteristics; obtains vegetation coverage clusters and non-vegetation coverage clusters based on the centroid position and vegetation coverage factor of each superpixel block; obtains the soil remediation vegetation significance of each superpixel block by combining the relative position distribution between each non-vegetation coverage cluster and each vegetation coverage block in the adjacent vegetation coverage cluster, and then obtains the soil remediation weighted significance of each superpixel block. The present invention improves the accuracy of soil environmental remediation assessment by accurately analyzing the vegetation characteristics in the soil environmental remediation process.

[0125] The present invention also proposes a vegetation detection system for soil environment remediation, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements any one of the steps of a vegetation detection method for soil environment remediation.

[0126] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0127] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A vegetation detection method for soil environmental remediation, characterized in that: The method comprises: Obtain remote sensing images of soil areas; Multiple superpixel blocks of the remote sensing image are obtained. Based on the color characteristics and position distribution characteristics of the pixels in each superpixel block, the vegetation cover factor of each superpixel block is obtained, and the vegetation cover blocks are screened out. Based on the position distribution and chromaticity fluctuation characteristics of the pixels in each vegetation cover block, the artificial cultivation characteristic degree of each vegetation cover block is obtained. According to the centroid position and vegetation coverage factor of each superpixel block, all superpixel blocks are clustered to obtain block clustering clusters; according to the distribution of vegetation coverage factors of all superpixel blocks in different block clustering clusters, vegetation coverage clusters and non-vegetation coverage clusters are obtained; Based on the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, the artificial cultivation characteristic degree and vegetation cover factor of the corresponding vegetation cover block, the soil restoration vegetation significance of each superpixel block is obtained; based on the soil restoration vegetation significance of each superpixel block, the significant distance between each superpixel block and different superpixel blocks, and the vegetation cover factors of other superpixel blocks, the soil restoration weighted significance of each superpixel block is obtained; Vegetation is detected based on the weighted significance of soil restoration across all superpixel patches.

2. A vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the vegetation cover factor includes: For any superpixel block, the over-green minus over-red index of each pixel is obtained and normalized and mapped as the vegetation color feature; The CANNY algorithm is used to perform edge detection on superpixel blocks to obtain multiple edge lines. The slope variance of all edge pixels on the edge line where each edge pixel lies is obtained as the slope fluctuation feature of each edge pixel. The edge pixel with the smallest relative distance to each pixel is selected, and the slope fluctuation feature of the corresponding edge pixel is used as the local texture roughness coefficient of each pixel. The vegetation coverage factor of each superpixel block is obtained according to the saturation channel value, vegetation color characteristics and local texture roughness coefficient of different pixel points in the HSV space. The saturation channel value is negatively correlated with the vegetation coverage factor, while the vegetation color characteristics and local texture roughness coefficient are both positively correlated with the vegetation coverage factor.

3. The vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the vegetation coverage blocks includes: If the vegetation coverage factor of a superpixel block is greater than or equal to a preset coverage threshold, the corresponding superpixel block is used as a vegetation coverage block.

4. The vegetation detection method for soil environmental remediation according to claim 2, characterized in that: The method for obtaining the artificial cultivation characteristic degree includes: For each vegetation cover block, the number of edge pixels on each edge line is counted as the length of the corresponding edge line; the mean difference in slope between each boundary pixel and the adjacent boundary pixel is obtained as the slope change rate of each boundary pixel; The length of the edge line or the slope change rate of the boundary pixel point is used as the analysis data. The difference between adjacent analysis data is calculated in descending order. The maximum value of the corresponding element when the difference is the largest is selected as the data reference value. If the analysis data is greater than the analysis reference value, the edge line corresponding to the analysis data is used as the long edge line, or the boundary pixel point corresponding to the analysis data is used as the boundary segmentation point to obtain multiple block boundaries. The PCA algorithm was used to obtain the principal component direction of each long edge line and block boundary, and the angle between the principal component direction and the horizontal direction. The artificial cultivation characteristic of each vegetation coverage block was obtained based on the minimum difference in the corresponding angle between each long edge line and different block boundaries in each vegetation coverage block, as well as the fluctuation degree of the chromaticity channel values of all pixel points in the HSV space. The minimum difference and fluctuation degree of the angle were positively correlated with the degree of artificial cultivation.

5. The vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the block clustering clusters includes: The centroid coordinates and vegetation cover factor of each superpixel block are used to construct multidimensional data, and the Euclidean distance of the multidimensional data between superpixel blocks is obtained. K-means clustering is performed on all superpixel blocks to obtain block clusters.

6. The vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the vegetation cover cluster and the non-vegetation cover cluster includes: Obtain the mean vegetation coverage factor of all superpixel blocks in each block cluster as the overall vegetation coverage level; The differences in adjacent overall vegetation coverage levels are calculated in order from large to small, and the maximum value in the corresponding element when the difference is the largest is selected as the coverage level reference value; if the overall vegetation coverage level of a block cluster is greater than the coverage level reference value, the corresponding block cluster will be regarded as the vegetation coverage cluster, and the remaining block clusters will be regarded as non-vegetation coverage clusters.

7. The vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the soil remediation vegetation significance includes: The soil environmental restoration degree of each non-vegetation cover cluster is obtained based on the relative position distribution between each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster, as well as the vegetation cover factor of the corresponding vegetation cover block. Obtain the relative distance between the centroid of each vegetation-covered block and each non-vegetation-covered cluster within the corresponding block area, select the minimum value of the relative distance between each vegetation-covered block and different non-vegetation-covered clusters, and use the corresponding non-vegetation-covered cluster as the reference cluster for each vegetation-covered block; The artificial cultivation characteristic degree of each vegetation coverage block and the product of the soil environmental restoration degree of the corresponding reference cluster are obtained as the soil restoration significance coefficient; the ratio of the soil restoration significance coefficient of each vegetation coverage block and the minimum relative distance is obtained as the soil restoration vegetation significance of each vegetation coverage block; if the superpixel block is not a vegetation coverage block, the soil restoration vegetation significance of the corresponding superpixel block is set to a positive integer 1.

8. The vegetation detection method for soil environmental remediation according to claim 7, characterized in that: The method for obtaining the soil environmental remediation degree includes: Obtain the relative distance distribution between the centroid of each non-vegetation cover cluster and each vegetation cover block in the adjacent vegetation cover cluster as the restoration edge distance; construct a restoration edge distance sequence by sorting the corresponding modified edge distances between each non-vegetation cover cluster and all vegetation cover blocks in the adjacent vegetation cover cluster in ascending order; According to the distribution order of vegetation coverage blocks corresponding to each element in the restoration edge distance sequence, a vegetation coverage factor sequence corresponding to all vegetation coverage blocks is constructed; The correlation coefficient between the restoration edge distance series and the vegetation cover factor series was obtained and normalized as the soil environmental restoration degree of each non-vegetation covered area.

9. The vegetation detection method for soil environmental remediation according to claim 1, characterized in that: The method for obtaining the weighted significance of soil remediation includes: Obtain the significant distance between each superpixel block and different other superpixel blocks, calculate the product of the significant distance and the vegetation cover factor of other superpixel blocks, obtain the product of the product of the product and the soil restoration vegetation significance value of each superpixel block, and perform positive correlation normalization as the weighted significance of soil restoration of each superpixel block.

10. A vegetation detection system for soil environmental remediation, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the vegetation detection method for soil environment remediation as claimed in any one of claims 1 to 9 are implemented.

Images