Intelligent forestry monitoring processing system and method

Abstract

The invention provides an intelligent forestry monitoring processing system and method. The method comprises the following steps: extracting leaf texture features and leaf edge features of different color categories from leaf state monitoring images of vegetation in a forestry monitoring area; determining a difference feature vector between the leaf texture features of each color category, and further determining a vegetation degradation gradient in a vegetation leaf senescence and health transition region in the forestry monitoring region; determining an edge difference vector in the leaf edge feature of each color category, and further determining the vegetation influence degree of water and soil loss on vegetation in the forestry monitoring area; determining the vulnerability index of the vegetation in the forestry monitoring area according to the degradation gradient of the vegetation and the vegetation influence degree, and classifying the growth health state of the vegetation in the forestry monitoring area according to the vulnerability index. By adopting the scheme of the invention, the forestry monitoring image data processing method which can fuse multi-source features, dynamically associate environmental factors and quantify vegetation vulnerability can be realized.

Description

Technical Field

[0001] The present application relates to the field of monitoring image processing technology, and more specifically, to a smart forestry monitoring and processing system and method. Background Art

[0002] With the development of smart forestry monitoring technology, real-time monitoring and precise assessment of forestry ecosystems have become important areas of ecological protection. Traditional forestry monitoring relies heavily on manual inspections or single remote sensing data, making it difficult to dynamically capture subtle changes in vegetation growth. This is particularly evident in areas such as early identification of vegetation degradation and assessment of soil and water loss impacts. Existing technologies often limit the analysis of leaf status to static detection of single features (such as color or texture) and lack the ability to correlate multidimensional features (such as color category differences, edge variations, and spatial distribution). This makes it impossible to accurately quantify vegetation degradation gradients and the impact of environmental stress. Furthermore, existing forestry monitoring systems often use isolated indicators to assess vegetation health, making it difficult to comprehensively reflect the spatial heterogeneity of vegetation in terms of ecological vulnerability, limiting the accuracy of forestry disaster warnings and ecological restoration. Therefore, the industry faces the challenge of developing forestry monitoring image data processing methods that integrate multi-source features, dynamically correlate environmental factors, and quantify vegetation vulnerability. Summary of the Invention

[0003] The present application provides a smart forestry monitoring and processing system and method, which can realize a forestry monitoring image data processing method that integrates multi-source features, dynamically associates environmental factors and quantifies vegetation vulnerability.

[0004] In a first aspect, the present application provides a method for processing images for smart forestry monitoring, comprising the following steps: Collecting leaf status monitoring images of vegetation in forestry monitoring areas; Extracting leaf texture features and leaf edge features of different color categories from the leaf status monitoring image; Determine the difference feature vectors between the texture features of leaves of each color category, and determine the degradation gradient of vegetation in the transition zone between senescent and healthy leaves in the forestry monitoring area based on the difference feature vectors of the leaf textures and the spatial distribution of vegetation in the forestry monitoring area; Determine the edge difference vectors within the leaf edge features of each color category, correlate and fuse all edge difference vectors with the growth density of vegetation in the forestry monitoring area, and then obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area; The vulnerability index of vegetation in the forestry monitoring area is determined by the vegetation degradation gradient and the vegetation impact, and the growth health status of vegetation in the forestry monitoring area is classified according to the vulnerability index.

[0005] In some embodiments, extracting leaf texture features and leaf edge features of different color categories from the leaf status monitoring image specifically includes: Identify the various color categories of leaves; Extracting leaf texture features of different color categories from the leaf status monitoring image; Leaf edge features of different color categories are extracted from the leaf status monitoring image.

[0006] In some embodiments, determining the difference feature vectors between the leaf texture features of each color category specifically includes: Determine the texture difference between the texture features of leaves of two random color categories; The difference feature vectors between the leaf texture features of each color category are determined based on all the texture differences.

[0007] In some embodiments, determining the degradation gradient of vegetation in the transition zone between senescent and healthy leaves in the forestry monitoring area based on the difference feature vectors of the textures of the leaves and the spatial distribution of the vegetation in the forestry monitoring area specifically includes: Generate the growth density of vegetation in the forestry monitoring area; Performing a trend analysis on the degradation of leaves of vegetation in the forestry monitoring area based on the difference feature vectors of the textures of each leaf and the spatial distribution of vegetation in the forestry monitoring area to obtain a spreading trend of vegetation degradation; The degradation gradient of vegetation in the transition zone between senescent and healthy leaves of vegetation in the forestry monitoring area is determined according to the spreading trend of the vegetation degradation.

[0008] In some embodiments, determining the edge difference vectors within the leaf edge features of each color category specifically includes: Determine edge differences within leaf edge features for each color class; The edge difference vectors within the leaf edge features of each color category are determined based on all the edge differences.

[0009] In some embodiments, all edge difference vectors are correlated and fused with the growth density of vegetation in the forestry monitoring area to obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area. Specifically, the following is performed: Determine the growth density of vegetation in the forestry monitoring area; extracting a plurality of dominant edge difference vectors from all edge difference vectors; Correlating the growth density with all dominant edge difference vectors to obtain density-edge correlation information; The impact of soil erosion on vegetation in the forestry monitoring area is determined based on the density-edge association information.

[0010] In some embodiments, determining the vulnerability index of vegetation in a forestry monitoring area based on the vegetation degradation gradient and the vegetation impact specifically includes: Obtain topographical characteristics of forestry monitoring areas; Determining a spatial gradient of vegetation degradation in a forestry monitoring area based on the vegetation degradation gradient and the terrain characteristics; The vulnerability index of vegetation in the forestry monitoring area is determined by the spatial gradient and the vegetation impact.

[0011] In a second aspect, the present application provides a smart forestry monitoring and processing system, which includes a monitoring image processing unit, and the monitoring image processing unit includes: An acquisition module is used to collect leaf status monitoring images of vegetation in the forestry monitoring area; A processing module, configured to extract leaf texture features and leaf edge features of different color categories from the leaf status monitoring image; The processing module is further configured to determine difference feature vectors between leaf texture features of each color category, and determine a degradation gradient of vegetation in a transition zone between senescent and healthy leaves in the forestry monitoring area based on the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area; The processing module is further configured to determine edge difference vectors within the leaf edge features of each color category, correlate and fuse all edge difference vectors with the growth density of vegetation in the forestry monitoring area, and thereby obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area; An execution module is used to determine the vulnerability index of vegetation in the forestry monitoring area based on the vegetation degradation gradient and the vegetation impact, and classify the growth health status of vegetation in the forestry monitoring area according to the vulnerability index.

[0012] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned smart forestry monitoring image processing method.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned smart forestry monitoring image processing method.

[0014] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the smart forestry monitoring and processing system and method provided by the present application, leaf status monitoring images of vegetation in the forestry monitoring area are first collected; leaf texture features and leaf edge features of different color categories are extracted from the leaf status monitoring images; the difference feature vectors between the leaf texture features of each color category are determined, and the degradation gradient of vegetation in the transition area between aging and health of leaves in the forestry monitoring area is determined based on the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area; the edge difference vector within the leaf edge feature of each color category is determined, and all edge difference vectors are correlated and fused with the growth density of vegetation in the forestry monitoring area to obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area; the vulnerability index of vegetation in the forestry monitoring area is determined by the vegetation degradation gradient and the vegetation impact, and the vegetation growth health status in the forestry monitoring area is classified by the vulnerability index.

[0015] It can be seen that in the process of forestry monitoring data processing, this application can first capture the visual and physical changes of leaves by extracting texture features and edge features, providing multi-dimensional feature support for subsequent difference analysis, and through color category division, it can more finely identify leaves in different states, thus laying the foundation for subsequent degradation gradient and impact analysis; then, determining the difference feature vectors between the texture features of leaves in each color category can quantify the change trend between leaves in different color categories, and combined with spatial distribution, it can dynamically reflect the degradation process of vegetation, and can capture the transition state of vegetation from health to aging, providing a dynamic and quantitative indicator for evaluating the degree of vegetation degradation. Then, determining the edge difference vector within the leaf edge features of each color category can reflect the physical changes of the leaves. Combined with the growth density, it can quantify the impact of soil erosion on vegetation. It can also associate environmental factors (such as soil erosion) with the physical characteristics of vegetation, providing a more comprehensive environmental impact assessment. Finally, the vulnerability index of vegetation in the forestry monitoring area is determined based on the vegetation degradation gradient and the vegetation impact. The vulnerability index is a comprehensive indicator that can quantify the health status and degradation risk of vegetation. Through classification, the growth health status of vegetation can be managed in a hierarchical manner, providing a scientific basis for forestry protection and management, and helping to formulate targeted protection measures. The above scheme can realize a forestry monitoring image data processing method that integrates multi-source features, dynamically associates environmental factors, and quantifies vegetation vulnerability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is an exemplary flow chart of a method for processing smart forestry monitoring images according to some embodiments of the present application; Figure 2 is an exemplary flow chart for determining the degradation gradient of vegetation according to some embodiments of the present application; Figure 3 is a flowchart of data processing according to some embodiments of the present application; Figure 4 is a schematic structural diagram of a monitoring image processing unit according to some embodiments of the present application; Figure 5 It is a structural diagram of a computer device for implementing a smart forestry monitoring image processing method according to some embodiments of the present application. DETAILED DESCRIPTION

[0017] In order to better understand the technical solution of the present application, the technical solution of the present application will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0018] refer to Figure 1 , which is an exemplary flow chart of a method for processing smart forestry monitoring images according to some embodiments of the present application. The method 100 for processing smart forestry monitoring images mainly includes the following steps: In step 101, leaf status monitoring images of vegetation in a forestry monitoring area are collected.

[0019] In specific implementation, a multispectral camera is mounted on an unmanned aerial vehicle to capture multiple images of vegetation leaves in the forestry monitoring area, and all the images are used as leaf status monitoring images of vegetation in the forestry monitoring area, wherein the leaf status monitoring image represents a monitoring image of the growth status of vegetation leaves in the forestry monitoring area, which can be used to evaluate the growth health status of vegetation in the forestry monitoring area; in other embodiments, other methods can also be used for collection, which will not be elaborated here.

[0020] In step 102, leaf texture features and leaf edge features of different color categories are extracted from the leaf status monitoring image.

[0021] In some embodiments, extracting leaf texture features and leaf edge features of different color categories from the leaf status monitoring image can be achieved by using the following steps: Identify the various color categories of leaves; Extracting leaf texture features of different color categories from the leaf status monitoring image; Leaf edge features of different color categories are extracted from the leaf status monitoring image.

[0022] In specific implementation, the following method can be used to determine the various color categories of the leaves, namely: by converting the leaf image from the RGB color space into the hue, saturation, and brightness (HSV) color space, and using the HSV parameters to classify the colors, the various color categories of the leaves are obtained. Different colors have specific ranges in the HSV space. For example, the hue range of green is usually between 26 and 34, wherein the color category represents the color type of the leaves, such as green, yellow, red, etc.; in other embodiments, other methods can also be used for determination, which are not limited here.

[0023] In addition, the texture of the leaves can be described by image texture, and the characteristic values ​​representing the image texture include image contrast, correlation, energy and entropy. These characteristic values ​​can reflect the roughness and regularity of the leaf texture. In specific implementation, the extraction of leaf texture features of different color categories from the leaf status monitoring image can be achieved in the following manner, namely: using image processing technology to obtain the appearance characteristics of color, shape, and texture of multiple vegetation from the leaf status monitoring image, and each appearance feature is used as the leaf state of the vegetation, all leaf states are divided according to color categories to obtain leaf states of different color categories, and the corresponding areas of the leaf states of each color category in the leaf image are grayscaled, and the combined frequency of pixel grayscale values ​​in the corresponding areas of each color category in the image is counted according to the set distance and angle to construct a grayscale co-occurrence matrix, and the contrast, correlation, energy and entropy of the corresponding areas of different color categories are calculated based on the grayscale co-occurrence matrix, and each contrast, correlation, energy and entropy is used as the leaf texture feature corresponding to the different color categories; in other embodiments, other methods can also be used for extraction, which are not limited here.

[0024] In specific implementation, the following methods can be used to extract the edge features of leaves of different color categories from the leaf status monitoring image, namely: extracting the leaf states of different color categories from the leaf status monitoring image, graying, denoising, and enhancing the areas corresponding to the leaf states of different color categories in the image, using the threshold segmentation method to separate the leaves of different color categories from the background, and calculating the edges, areas, and perimeters of each leaf separated from the different color categories in the image through the edge detection algorithm, and using all the calculated geometric features of different color categories (i.e., the edges, areas, and perimeters of each leaf) as the leaf edge features of the corresponding color categories. In other embodiments, other methods can also be used for extraction, which are not limited here.

[0025] It should be noted that the leaf texture features described in this application represent the characteristics of the texture conditions of the corresponding colors in the collected images, and the leaf edge features represent the characteristics of the edge conditions of the leaves of vegetation in the forestry monitoring area, which can be used to identify the vegetation in the forestry monitoring area and take corresponding management measures.

[0026] In step 103, the difference feature vectors between the leaf texture features of each color category are determined, and the degradation gradient of vegetation in the transition area between aging and health of vegetation leaves in the forestry monitoring area is determined based on the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area.

[0027] In some embodiments, determining the difference feature vectors between the texture features of leaves of different color categories may be achieved by using the following steps: Determine the texture difference between the texture features of leaves of two random color categories; The difference feature vectors between the leaf texture features of each color category are determined based on all the texture differences.

[0028] In specific implementation, determining the texture difference between the leaf texture features of two random color categories can be achieved in the following manner, namely: the Euclidean distance can be used to calculate the texture difference between the leaf texture features of two random color categories, and the texture difference includes the difference in contrast, the difference in correlation, the difference in energy and the difference in entropy, wherein the texture difference represents the degree of difference in the texture of leaves of two random color categories, which can be used to predict changes in leaf texture; determining the difference feature vector between the leaf texture features of each color category based on all texture differences can be achieved in the following manner, namely: combining each texture difference into a vector, and using each vector as the difference feature vector between the leaf texture features of the corresponding two color categories, for example: difference feature vector = [d contrast, d correlation, d energy, d entropy]; in other embodiments, other methods can also be used for determination, which are not limited here.

[0029] It should be noted that the difference feature vector between the texture features of leaves of two color categories in this application represents the vector of the difference features of the texture between the two color categories in the forestry monitoring area, which can be used to classify the leaves in the forestry monitoring area, thereby judging the growth trend of the leaves in the forestry monitoring area.

[0030] In some embodiments, reference Figure 2 As shown in FIG. 1 , this figure is an exemplary flow chart for determining the degradation gradient of vegetation in some embodiments of the present application. In this embodiment, the degradation gradient of vegetation in the transition zone between aging and healthy leaves in a forestry monitoring area is determined based on the difference feature vectors of the textures of each leaf and the spatial distribution of vegetation in the forestry monitoring area. The following steps can be used to achieve this: First, in step 1031, the spatial distribution of vegetation in the forestry monitoring area is generated; Next, in step 1032, a trend analysis of the degradation of leaves in the forestry monitoring area is performed based on the difference feature vectors of the textures of the leaves and the spatial distribution of the vegetation in the forestry monitoring area to obtain a spreading trend of vegetation degradation. Finally, in step 1033, the degradation gradient of vegetation in the transition zone between senescent and healthy leaves of vegetation in the forestry monitoring area is determined according to the spreading trend of vegetation degradation.

[0031] In specific implementation, the spatial distribution of vegetation in the forestry monitoring area can be generated in the following manner, namely: an image of the forestry monitoring area is obtained by using a camera mounted on a drone in the existing technology, and the horizontal and vertical distributions of the positions of vegetation in the forestry monitoring area in the geographic space are analyzed from the image by combining remote sensing (RS) and geographic information system (GIS) technology, so as to use the horizontal and vertical distributions of the positions of vegetation in the geographic space as the spatial distribution of vegetation in the forestry monitoring area, wherein the spatial distribution represents the distribution of the positions of vegetation in the forestry monitoring area.

[0032] In addition, as the degree of leaf degradation increases, its texture characteristics will gradually change in a regular manner. Through long-term monitoring and analysis of the changes in the difference feature vectors of leaf textures, these regularities can be discovered, and then a corresponding relationship between the difference feature vectors and the degree of leaf degradation can be established. Moreover, the spatial distribution of vegetation is not isolated, and there are interactions and influences between different vegetations. Therefore, in specific implementation, the degradation trend of vegetation leaves in the forestry monitoring area is analyzed according to the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area. The spread trend of vegetation degradation can be obtained in the following way, namely: principal component analysis (PCA) is used to screen the key texture features of the difference feature vectors of all leaf textures, and the key texture features are calculated by support vector machine (SVM). The degradation probability value (0-1) of the leaves in the area corresponding to the color type of each difference eigenvector in the feature, wherein the degradation probability value represents the numerical value of the possibility of vegetation degradation in the area, and the attribute table of the vector data is associated with all the degradation probability values ​​and the spatial distribution of the vegetation in the forestry monitoring area to construct a dataset with spatial attributes of the vegetation leaf status in the forestry monitoring area, and linear regression is used in combination with the dataset with spatial attributes to analyze the spread direction and spread speed of vegetation degradation in the forestry monitoring area, and the spread direction and spread speed of vegetation degradation in the forestry monitoring area are used as the spread trend of vegetation degradation in the forestry monitoring area, wherein the spread trend represents the trend of the spread of vegetation degradation in the forestry monitoring area; in other embodiments, other methods can also be used to determine, which are not limited here.

[0033] In addition, the transitional area between the senescence and health of vegetation leaves is the zone where the vegetation changes from a healthy state to a senescent and degraded state. Within the transitional area, if the spread rate of vegetation degradation is relatively fast, it indicates that the degree of vegetation degradation changes significantly within a short distance, that is, the degradation gradient is relatively steep; conversely, if the spread rate is slow, the degradation gradient is relatively gentle. When specifically implemented, the degradation gradient of the vegetation in the transitional area between the senescence and health of vegetation leaves in the forestry monitoring area can be determined according to the spread trend of the vegetation degradation in the following manner: taking the center point of the starting area of the spread trend of the vegetation degradation in the forestry monitoring area as the center, measuring the distance d from each point (the center point of the spatial analysis unit in the forestry monitoring area (such as the center point of a 10m×10m grid)) in the direction of the spread trend of the vegetation degradation in the forestry monitoring area to this center, fitting a decay function (such as the exponential function P(d)=P0*e^{-kdv}, and the decay coefficient k can be solved through non-linear regression), where P(d) is the degradation value, P0 is the initial degradation value, and v is the spread rate in the spread trend of the vegetation degradation (the larger the absolute value, the faster the decay). Calculate the degradation value of each point in the direction of the spread trend of the vegetation degradation in the forestry monitoring area according to the above decay function, and divide the area according to the degradation value P (such as a high degradation area where P≥0.6, a medium degradation area where 0.3<P<0.6, and a healthy area where P≤0.3). The medium degradation area is the transitional area. Arrange all the degradation values that meet the medium degradation area in the order of the corresponding points to this center, and use the arranged sequence as the degradation gradient of the vegetation in the transitional area between the senescence and health of vegetation leaves in the forestry monitoring area. In other embodiments, other methods can also be used to determine it, which is not limited here.

[0034] It should be noted that the degradation value in this application represents the parameter value of the attenuation degree of the corresponding point in the forestry area, and the initial degradation value represents the parameter value of the degradation degree of the starting area of the spread trend of the vegetation degradation in the forestry monitoring area, which can be determined by the historical degradation degree of this starting area. For example, the average value of the historical degradation degree is used as the initial degradation value of this starting area.

[0035] It should be noted that the degradation gradient in this application represents the change gradient of the degradation degree of the vegetation leaves in the transitional area between the senescence and health in the forestry monitoring area, and can be used to analyze the degradation situation of the vegetation in the forestry monitoring area.

[0036] In step 104, determine the edge difference vectors within the edge features of the leaves of each color category, and associate and fuse all the edge difference vectors with the growth density of the vegetation in the forestry monitoring area, so as to obtain the vegetation impact degree of soil erosion on the vegetation in the forestry monitoring area.

[0037] In some embodiments, determining the edge difference vectors within the leaf edge features of each color category may be implemented using the following steps: Determine edge differences within leaf edge features for each color class; The edge difference vectors within the leaf edge features of each color category are determined based on all the edge differences.

[0038] In specific implementation, determining the edge differences within the leaf edge features of each color category can be achieved in the following manner, namely: using Euclidean distance to calculate the edge differences of the geometric features of all leaves within the leaf edge features of each color category, the edge differences include mean, variance, maximum and minimum, wherein the edge differences represent the degree of difference in the leaf edges of the color categories, which can be used to predict changes in the leaf edges; determining the edge difference vector within the leaf edge features of each color category based on all edge differences can be achieved in the following manner, namely: combining each edge difference into a vector, and using each vector as the edge difference vector within the leaf edge feature of the corresponding color category, for example: edge difference vector within the leaf edge feature = [d mean, d variance, d maximum, d minimum]; in other embodiments, other methods can also be used for determination, which are not limited here.

[0039] It should be noted that the edge difference vector within the leaf edge feature in this application represents the vector of the edge difference feature of a color category in the forestry monitoring area, which can be used to classify the leaves in the forestry monitoring area, thereby judging the growth trend of the leaves in the forestry monitoring area.

[0040] In some embodiments, associating and fusing all edge difference vectors with the growth density of vegetation in the forestry monitoring area to obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area can be achieved by the following steps: Determine the growth density of vegetation in the forestry monitoring area; extracting a plurality of dominant edge difference vectors from all edge difference vectors; Correlating the growth density with all dominant edge difference vectors to obtain density-edge correlation information; The impact of soil erosion on vegetation in the forestry monitoring area is determined based on the density-edge association information.

[0041] It should be noted that the vegetation coverage (i.e., vegetation growth density) is quantified by vegetation indices (such as NDVI and EVI) using the difference in vegetation reflectance in the near-infrared and red light bands. In specific implementation, the growth density of vegetation in the forestry monitoring area can be determined in the following way, namely: multispectral data of vegetation in the forestry monitoring area is collected through multispectral imaging of drones, and the vegetation index of each spatial analysis unit (such as a 10m×10m grid) in the forestry monitoring area is calculated in combination with the multispectral data using the normalized vegetation index method in the existing technology, and all vegetation indices are used as the growth density of vegetation in the forestry monitoring area.

[0042] In addition, the dimensionality of all edge difference vectors is reduced while retaining the main information (such as retaining the principal components of the first 80% of the edge difference vectors), thereby reducing the difficulty of analysis. In specific implementation, multiple dominant edge difference vectors can be extracted from all edge difference vectors in the following manner, namely: principal component analysis (PCA) is used to screen the key edge features of all edge difference vectors, and each of the screened edge difference vectors is used as the dominant edge difference vector, wherein the dominant edge difference feature represents the edge difference vector with the main effect; in other embodiments, other methods can also be used to obtain it, which is not limited here.

[0043] It should be noted that the growth density reflects the overall growth status and quantity distribution of vegetation, while the edge difference vector describes the local characteristics of the vegetation edge. There is an intrinsic connection between these characteristics. For example, the edges of areas with vigorous vegetation growth have more details and changes, which is related to the diverse vegetation morphology caused by the high growth density. Therefore, by associating the growth density with all dominant edge difference vectors, the growth status and structural characteristics of vegetation in the forestry monitoring area can be analyzed; in specific implementation, the growth density and all dominant edge difference vectors are associated to obtain density-edge correlation information, which can be achieved in the following way, namely: a density-edge correlation model is established based on a random forest, the growth density is used as the target variable of the density-edge correlation model, all dominant edge difference vectors are used as the characteristic variables of the density-edge correlation model, the correlation between different edge difference vectors and density is calculated through the density-edge correlation model, and all correlations are used as density-edge correlation information, wherein the density-edge correlation information represents information on the degree of correlation between density and edge; in other embodiments, other methods can also be used for determination, which is not limited here.

[0044] In addition, in forestry areas, where the density of vegetation is higher and the leaf edge is positively correlated with the density (i.e., the leaves in high-density areas are thicker and have stronger interception capacity), the higher the vegetation impact is, the lower the sensitivity of vegetation to soil erosion is, and vice versa. In specific implementation, the vegetation impact of soil erosion on vegetation in forestry monitoring areas can be determined based on the density-edge correlation information. This can be achieved in the following way, namely: normalizing all vegetation indices in the growth density to obtain the normalized coefficients of each vegetation index, and constructing an impact formula based on the machine learning algorithm. The framework of the impact formula is: The vegetation impact of a spatial analysis unit = α*D+β*(T−D), where α and β are weights, D is the normalization coefficient of the vegetation index corresponding to the spatial analysis unit, and (T−D) is the correlation between the density corresponding to the spatial analysis unit and the density-edge correlation information. α and β are determined using measured data on soil and water loss (e.g., giving priority to the effect of density on coverage, α=0.6, β=0.4). The average of the vegetation impacts of all spatial analysis units is used as the vegetation impact of soil and water loss on vegetation in the forestry monitoring area. In other embodiments, other methods may be used for determination, which are not limited here.

[0045] It should be noted that the vegetation impact degree in this application represents the degree of impact of soil erosion on vegetation in the forestry monitoring area, which can be used to judge the impact of vegetation under soil erosion conditions in the forestry monitoring area, thereby reducing the impact of soil erosion on vegetation assessment.

[0046] In step 105, the vulnerability index of the vegetation in the forestry monitoring area is determined by the vegetation degradation gradient and the vegetation impact, and the growth health status of the vegetation in the forestry monitoring area is classified according to the vulnerability index.

[0047] In some embodiments, determining the vulnerability index of vegetation in a forestry monitoring area based on the vegetation degradation gradient and the vegetation impact can be achieved by using the following steps: Obtaining topographical characteristics of forestry monitoring areas; Determining a spatial gradient of vegetation degradation in a forestry monitoring area based on the vegetation degradation gradient and the terrain characteristics; The vulnerability index of vegetation in the forestry monitoring area is determined by the spatial gradient and the vegetation impact.

[0048] In a specific implementation, obtaining the terrain characteristics of the forestry monitoring area can be achieved in the following manner, namely: obtaining the terrain characteristics of the forestry monitoring area from a database of a monitoring department corresponding to the forestry monitoring area, wherein the terrain characteristics represent the characteristics of the terrain in the forestry monitoring area, and the terrain characteristics include the shape, slope, and undulation of the terrain; In addition, it should be noted that by analyzing the relationship between terrain characteristics and vegetation degradation gradients, the spatial distribution pattern of vegetation degradation can be revealed; in specific implementation, the spatial gradient of vegetation degradation in the forestry monitoring area can be determined based on the vegetation degradation gradient and the terrain characteristics. This can be achieved in the following manner, namely: superimposing the vegetation degradation gradient and the terrain characteristics in a geographic information system (GIS), and using the superimposed result as the spatial gradient of vegetation degradation. After superimposing the vegetation degradation gradient and the terrain characteristics, each superimposed unit contains both vegetation degradation information and vegetation terrain information, wherein the spatial gradient of vegetation degradation represents the gradient of change after the degree of vegetation degradation in the forestry monitoring area is combined with the geographic space; in other embodiments, other methods can also be used for determination, which are not limited here.

[0049] In addition, it should be noted that combining the spatial gradient and the vegetation impact can more comprehensively evaluate the vulnerability of vegetation in the forestry monitoring area. The spatial gradient focuses on reflecting the spatial change trend and potential vulnerability distribution of vegetation, while the vegetation impact emphasizes the actual effect of external factors on vegetation. By combining the two through the weighted average method, an index that comprehensively reflects the vulnerability of vegetation can be obtained. In specific implementation, the vulnerability index of vegetation in the forestry monitoring area can be determined by the spatial gradient and the vegetation impact. That is, the spatial gradient and the vegetation impact can be combined according to certain weights using the weighted average method to calculate the vegetation vulnerability index of vegetation in the forestry monitoring area. For example, the spatial gradient is given a weight of w1 and the vegetation impact is given a weight of w2. Then the vegetation vulnerability index VI=w1×SG+w2×II, where SG is the spatial gradient of vegetation degradation and II is the vegetation impact index. The weight can be determined by a data statistical analysis method; in other embodiments, other methods can also be used for determination, which is not limited here.

[0050] It should be noted that the vegetation vulnerability index in this application represents the parameter value of the sensitivity and recovery capacity of vegetation in the forestry inspection area to vegetation degradation and land loss, which can be used to evaluate the performance of vegetation in the forestry monitoring area so that relevant departments can take corresponding measures.

[0051] In some embodiments, the classification of the vegetation growth health status in the forestry monitoring area using the vulnerability index can be achieved by the following steps: Determine vulnerability thresholds for forestry monitoring areas; comparing the vulnerability index to the vulnerability threshold; If the vulnerability index is greater than or equal to the vulnerability threshold, the vegetation growth health status in the forestry monitoring area is classified as a high vulnerability state; If the vulnerability index is less than the vulnerability threshold, the vegetation growth health status in the forestry monitoring area is classified as a low vulnerability state.

[0052] It should be noted that in this application, a high vulnerability state refers to a state in which vegetation is highly sensitive to external interference and has a low recovery ability under specific environmental conditions, and a low vulnerability state refers to a state in which vegetation is less sensitive to external interference and has a higher recovery ability under specific environmental conditions.

[0053] It should be noted that in this application, the vulnerability threshold can be set accordingly according to the specific needs of the vegetation in the forestry monitoring area. For example, if the vegetation in the forestry monitoring area is ordinary vegetation, the vulnerability threshold can be set in a high range. If the vegetation in the forestry monitoring area is scarce vegetation, the vulnerability threshold can be set in a low range. In other embodiments, for example, when the forestry monitoring area belongs to a forest in a natural ecology, the vulnerability threshold can be set in a high range because the forest in the natural ecology has strong recovery ability.

[0054] In some embodiments, reference Figure 3 As shown, this figure is a flow chart of data processing in some embodiments of the present application, such as Figure 3 As described above, data is collected from the forest area, and the collected data is pre-processed and then stored in a database, and then the data is analyzed and finally the analysis results are output.

[0055] In addition, in another aspect of the present application, in some embodiments, the present application provides a smart forestry monitoring and processing system, the smart forestry monitoring and processing system includes a monitoring image processing unit, reference Figure 4 , which is a schematic diagram of the structure of a monitoring image processing unit according to some embodiments of the present application. The monitoring image processing unit 400 includes: an acquisition module 401, a processing module 402 and an execution module 403, which are described as follows: Acquisition module 401, in this application, acquisition module 401 is mainly used to collect leaf status monitoring images of vegetation in the forestry monitoring area; Processing module 402, in this application, is used to extract leaf texture features and leaf edge features of different color categories from the leaf status monitoring image; It should be noted that the processing module 402 in the present application is also used to determine the difference feature vectors between the texture features of leaves of each color category, and determine the degradation gradient of vegetation in the transition zone between aging and healthy leaves in the forestry monitoring area based on the difference feature vectors of the leaf textures and the spatial distribution of vegetation in the forestry monitoring area; In addition, it should be noted that the processing module 402 in the present application is also used to determine the edge difference vectors within the leaf edge features of each color category, and to correlate and fuse all the edge difference vectors with the growth density of vegetation in the forestry monitoring area, thereby obtaining the vegetation impact of soil erosion on vegetation in the forestry monitoring area; Execution module 403. In this application, execution module 403 is mainly used to determine the vulnerability index of vegetation in the forestry monitoring area through the vegetation degradation gradient and the vegetation impact, and classify the vegetation growth health status in the forestry monitoring area according to the vulnerability index.

[0056] In addition, the present application also provides a computer device, which includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned smart forestry monitoring image processing method.

[0057] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device for implementing the smart forestry monitoring image processing method according to some embodiments of the present application. The smart forestry monitoring image processing method in the above embodiment can be Figure 5 The computer device 500 shown in FIG. 5 is implemented as shown in FIG. 5 . The computer device 500 includes at least one processor 501 , a communication bus 502 , a memory 503 , and at least one communication interface 504 .

[0058] The processor 501 may be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0059] The communication bus 502 may be used to transmit information between the aforementioned components.

[0060] The memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CDROM) or other optical disc storage, an optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory 503 may be independent and connected to the processor 501 via the communication bus 502. The memory 503 may also be integrated with the processor 501.

[0061] The memory 503 is used to store program code for executing the solution of the present application, and is controlled by the processor 501. The processor 501 is used to execute the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiment can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0062] The communication interface 504 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0063] In a specific implementation, as an example, a computer device may include multiple processors, each of which may be a single-core (singleCPU) processor or a multi-core (multiCPU) processor. The processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0064] The aforementioned computer device can be a general-purpose computer device or a dedicated computer device. In a specific implementation, the computer device can be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of this application do not limit the type of computer device.

[0065] In addition, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned smart forestry monitoring image processing method.

[0066] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0067] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A smart forestry monitoring image processing method, characterized in that: The steps include: Collecting leaf status monitoring images of vegetation in forestry monitoring areas; Extracting leaf texture features and leaf edge features of different color categories from the leaf status monitoring image; Determine the difference feature vectors between the texture features of leaves of each color category, and determine the degradation gradient of vegetation in the transition zone between senescent and healthy leaves in the forestry monitoring area based on the difference feature vectors of the textures of each leaf and the spatial distribution of vegetation in the forestry monitoring area; Determine the edge difference vectors within the leaf edge features of each color category, correlate and fuse all edge difference vectors with the growth density of vegetation in the forestry monitoring area, and then obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area; The vulnerability index of vegetation in the forestry monitoring area is determined by the vegetation degradation gradient and the vegetation impact, and the growth health status of vegetation in the forestry monitoring area is classified according to the vulnerability index.

2. The method according to claim 1, wherein Extracting leaf texture features and leaf edge features of different color categories from the leaf status monitoring image specifically includes: Identify the various color categories of leaves; Extracting leaf texture features of different color categories from the leaf status monitoring image; Leaf edge features of different color categories are extracted from the leaf status monitoring image.

3. The method according to claim 1, wherein Determining the difference feature vectors between the texture features of leaves of each color category specifically includes: Determine the texture difference between the texture features of leaves of two random color categories; The difference feature vectors between the leaf texture features of each color category are determined based on all the texture differences.

4. The method according to claim 1, wherein Based on the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area, the degradation gradient of vegetation in the transition area between aging and health of vegetation leaves in the forestry monitoring area is determined. Specifically, Generate the growth density of vegetation in the forestry monitoring area; Performing a trend analysis on the degradation of leaves of vegetation in the forestry monitoring area based on the difference feature vectors of the textures of each leaf and the spatial distribution of vegetation in the forestry monitoring area to obtain a spreading trend of vegetation degradation; The degradation gradient of vegetation in the transition zone between senescent and healthy leaves of vegetation in the forestry monitoring area is determined according to the spreading trend of the vegetation degradation.

5. The method according to claim 1, wherein Determining the edge difference vector within the leaf edge features of each color category specifically includes: Determine edge differences within leaf edge features for each color class; The edge difference vectors within the leaf edge features of each color category are determined based on all the edge differences.

6. The method according to claim 1, wherein All edge difference vectors and vegetation growth density in the forestry monitoring area are correlated and fused, and the vegetation impact of soil erosion on vegetation in the forestry monitoring area is obtained, which specifically includes: Determine the growth density of vegetation in the forestry monitoring area; extracting a plurality of dominant edge difference vectors from all edge difference vectors; Correlating the growth density with all dominant edge difference vectors to obtain density-edge correlation information; The impact of soil erosion on vegetation in the forestry monitoring area is determined based on the density-edge association information.

7. The method according to claim 1, wherein Determining the vulnerability index of vegetation in the forestry monitoring area by the vegetation degradation gradient and the vegetation impact specifically includes: Obtain topographical characteristics of forestry monitoring areas; Determining a spatial gradient of vegetation degradation in a forestry monitoring area based on the vegetation degradation gradient and the terrain characteristics; The vulnerability index of vegetation in the forestry monitoring area is determined by the spatial gradient and the vegetation impact.

8. A smart forestry monitoring and processing system, comprising a monitoring image processing unit, characterized in that: The monitoring image processing unit includes: An acquisition module is used to collect leaf status monitoring images of vegetation in the forestry monitoring area; A processing module, configured to extract leaf texture features and leaf edge features of different color categories from the leaf status monitoring image; The processing module is further configured to determine difference feature vectors between leaf texture features of each color category, and determine a degradation gradient of vegetation in a transition zone between senescent and healthy leaves in the forestry monitoring area based on the difference feature vectors of each leaf texture and the spatial distribution of vegetation in the forestry monitoring area; The processing module is further configured to determine edge difference vectors within the leaf edge features of each color category, correlate and fuse all edge difference vectors with the growth density of vegetation in the forestry monitoring area, and thereby obtain the vegetation impact of soil erosion on vegetation in the forestry monitoring area; An execution module is used to determine the vulnerability index of vegetation in the forestry monitoring area based on the vegetation degradation gradient and the vegetation impact, and classify the growth health status of vegetation in the forestry monitoring area according to the vulnerability index.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores code, and the processor is configured to obtain the code and execute the smart forestry monitoring image processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the smart forestry monitoring image processing method according to any one of claims 1 to 7 is implemented.

Images