An intelligent monitoring method for the change amount of soil erosion in water and soil treatment

Abstract

The present invention relates to the field of image processing technology. More specifically, the present invention relates to an intelligent monitoring method for the change amount of soil erosion for soil and water treatment. The method includes collecting suspected landslide areas in remote sensing images of each period, obtaining the erosion risk factors of each suspected landslide area in each period, clustering the suspected landslide areas in each period, and obtaining the degree of erosion of each risk area in each period; designating the most recent period as the current period, obtaining the erosion growth of each risk area in the current period; obtaining the environmental stability of each risk area in the current period; based on the erosion growth and environmental stability, obtaining the overall attention of each risk area in the current period, and setting the acquisition frequency of each risk area in the current period according to the overall attention. The present invention takes into account the erosion conditions of different regions in the area to be detected, and improves the accuracy of erosion monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing. More specifically, the present invention relates to an intelligent monitoring method for the change amount of soil erosion for soil and water treatment. Background Art

[0002] Soil erosion, as one of the key issues in the global ecological environment field, seriously threatens the sustainable utilization of land resources, the balance and stability of the ecosystem, and the development of human society. In areas with severe soil and water loss, the soil fertility drops sharply, resulting in a significant reduction in crop yields and threatening regional food security.

[0003] The emergence of unmanned aerial vehicle (UAV) monitoring technology has brought a new perspective to soil erosion monitoring. It can obtain surface information over a large area and periodically, and indirectly estimate the change amount of soil erosion by analyzing the changes in landslide areas and other situations in remote sensing images of different periods. However, when using UAVs to monitor soil erosion, it often conducts fixed-frequency monitoring of the area to be detected, without considering the different erosion conditions of different regions in the area to be detected, and may miss sudden erosion events (such as the peak of soil and water loss caused by heavy rain) or dynamic evolution processes (such as the rapid expansion of landslides). Summary of the Invention

[0004] In order to solve the problem that when using UAVs to monitor soil erosion, it often conducts fixed-frequency monitoring of the area to be detected and may miss sudden erosion events, the present invention proposes an intelligent monitoring method for the change amount of soil erosion for soil and water treatment, and the method includes the following steps:

[0005] Collect remote sensing images of each period; based on the remote sensing images of each period, obtain each suspected landslide area and its slope in each period; based on the slope change of each suspected landslide area in each period, obtain the erosion risk factor of each suspected landslide area in each period; cluster the suspected landslide areas in each period according to the erosion risk factor to obtain each risk area in each period; obtain the erosion degree of each risk area in each period;

[0006] According to the erosion degree, obtain the erosion change rate of each risk area in each period; record the most recent period as the current period, and take the sum of the erosion change rates of the same risk area in all periods as the erosion growth of each risk area in the current period; according to the stability of the erosion degree of the same risk area in all periods, obtain the environmental stability of each risk area in the current period; based on the erosion growth and environmental stability, obtain the overall attention of each risk area in the current period.

[0007] Set the collection frequency of each risk area in the current period according to the overall attention of each risk area in the current period.

[0008] The innovation of the present invention lies in obtaining the overall attention of each risk area in the current period according to the erosion growth and environmental stability of the risk areas in the current period, and setting the acquisition frequency of each risk area in the current period according to the overall attention of each risk area in the current period, so as to realize different frequency monitoring for different erosion areas in the area to be detected.

[0009] Preferably, the obtaining of each suspected landslide area and its slope in each period includes:

[0010] Training a neural network based on the landslide areas manually marked in a large number of remote sensing images, and inputting the remote sensing images of each period into the trained neural network to obtain the suspected landslide areas in the remote sensing images of each period, denoted as the suspected landslide areas in each period;

[0011] Facilitating subsequent analysis of the suspected landslide areas in each period.

[0012] Extracting the slope of each suspected landslide area in each period according to the digital elevation model.

[0013] Preferably, the obtaining of the erosion risk factor of each suspected landslide area in each period includes:

[0014] Obtaining the maximum and minimum values of the slope change of the jth suspected landslide area in the ith period;

[0015] ;

[0016] In the formula, represents the erosion risk factor of the jth suspected landslide area in the ith period; represents the degree of slope change of the jth suspected landslide area in the ith period; represents the maximum value of the slope change of the jth suspected landslide area in the ith period; represents the minimum value of the slope change of the jth suspected landslide area in the ith period.

[0017] Facilitating subsequent clustering of the suspected landslide areas in each period according to the erosion factors, and analyzing each risk area in each period.

[0018] Preferably, the obtaining of the degree of slope change of the jth suspected landslide area in the ith period includes:

[0019] Preset the number of neighborhood periods \(T\). Denote the \(i\)-th period and the \(T\) periods before it as the neighborhood periods of the \(i\)-th period. Obtain the position coordinates of the center point of the \(j\)-th suspected landslide area in the \(i\)-th period, denoted as the target coordinates. If there is a suspected landslide area at the position of the target coordinates in any neighborhood period of the \(i\)-th period, denote the slope of this suspected landslide area as the target slope of this neighborhood period of the \(i\)-th period. If there is no suspected landslide area, the target slope of this neighborhood period of the \(i\)-th period is 0. Obtain the standard deviation of the target slope differences between all adjacent neighborhood periods at the \(i\)-th moment, denoted as the degree of slope change of the \(j\)-th suspected landslide area in the \(i\)-th period.

[0020] Preferably, the obtaining of the maximum and minimum values of the slope change of the \(j\)-th suspected landslide area in the \(i\)-th period includes:

[0021] Obtain the maximum value among the target slope differences between all adjacent neighborhood periods at the \(i\)-th moment, denoted as the maximum value of the slope change of the \(j\)-th suspected landslide area in the \(i\)-th period. Obtain the minimum value among the target slope differences between all adjacent neighborhood periods at the \(i\)-th moment, denoted as the minimum value of the slope change of the \(j\)-th suspected landslide area in the \(i\)-th period.

[0022] Preferably, the obtaining of each risk area in each period includes:

[0023] For any suspected landslide area in any period, take the erosion factor and position coordinates of the suspected landslide area as the feature vector of the suspected landslide area. Use the DBSCAN clustering algorithm to cluster the feature vectors of all suspected landslide areas in the period to obtain several clustering clusters. Take a region composed of all suspected landslide areas in each clustering cluster as a risk area in the period.

[0024] Preferably, the obtaining of the erosion change rate of each risk area in each period includes:

[0025] ;

[0026] In the formula, represents the erosion change rate of the \(k\)-th risk area in the \(i\)-th period; represents the degree of erosion of the \(k\)-th risk area in the \(i\)-th period; represents the degree of erosion of the \(k\)-th risk area in the \((i - 1)\)-th period.

[0027] Preferably, the obtaining of the environmental stability of each risk area in the current period includes:

[0028] Use the least squares method to perform a linear fit on the degree of erosion of the k-th risk area in all periods to obtain the erosion fitting line of the k-th risk area;

[0029] ;

[0030] wherein, represents the environmental stability of the k-th risk area in the current period; represents the number of all periods; represents the slope of the erosion fitting line of the k-th risk area; represents the change rate of the degree of erosion of the k-th risk area between the i-th period and the i+1-th period; exp() represents the exponential function with the natural constant as the base.

[0031] It is convenient to adaptively obtain the overall attention of the risk area according to the environmental stability of the risk area subsequently.

[0032] Preferably, the obtaining of the overall attention of each risk area in the current period includes:

[0033] Taking the product of the reciprocal of the environmental stability of the k-th risk area in the current period and the erosion growth of the k-th risk area in the current period as the overall attention of the k-th risk area in the current period.

[0034] Preferably, the setting of the acquisition frequency of each risk area in the current period according to the overall attention of each risk area in the current period includes:

[0035] Obtain the mean value of the difference in overall attention between each risk area in the current period and all other risk areas, denoted as the importance degree of each risk area in the current period, perform linear normalization on the importance degree to obtain the attention degree of each risk area in the current period, set a high threshold T1 and a low threshold T2. If the attention degree is greater than or equal to T1, set the monitoring frequency of the corresponding risk area to once a day; if the attention degree is less than or equal to T2, set the monitoring frequency of the corresponding risk area to once a month; if the attention degree is greater than T2 and less than T1, set the monitoring frequency of the corresponding risk area to once a week.

[0036] It is possible to set different acquisition frequencies for areas with different erosion degrees and not miss sudden erosion events.

[0037] The present invention has the following beneficial effects: The object of the present invention is to obtain the overall attention of each risk area in the current period according to the erosion growth and environmental stability of the risk areas in the current period, and set the acquisition frequency of each risk area in the current period according to the overall attention of each risk area in the current period, so as to achieve different frequency monitoring for different erosion areas in the area to be detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] By referring to the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown by way of example and not limitation, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0039] Figure 1 is a flowchart of the steps of an intelligent monitoring method for soil erosion change amount for soil and water treatment according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention.

[0041] Please refer to Figure 1 , which shows a flowchart of the steps of an intelligent monitoring method for soil erosion change amount for soil and water treatment provided by an embodiment of the present invention. The method includes the following steps:

[0042] S001. Collect remote sensing images of each period.

[0043] In the embodiment of the present invention, a drone is used to shoot the area to be detected every month to obtain remote sensing images of each period.

[0044] S002. Obtain the slope of each suspected landslide area in each period, and based on the slope change of each suspected landslide area in each period, obtain the erosion risk factor of each suspected landslide area in each period, cluster the suspected landslide areas in each period, and obtain the erosion degree of each risk area in each period.

[0045] It should be noted that when using a drone to monitor soil erosion, the area to be detected is often monitored at a fixed frequency without considering the different erosion conditions of different areas in the area to be detected. Therefore, the present invention needs to dynamically adjust the monitoring frequency of the drone according to the change trend of the erosion degree of different areas in the area to be detected;

[0046] Moreover, when there are more landslide areas and larger areas in remote sensing images, it will exacerbate soil erosion and lead to more serious soil loss. Therefore, for remote sensing images of each period, a neural network is used to obtain all suspected landslide areas in the remote sensing images of each period, and all suspected landslide areas of each period are obtained. Also, it is known that the steeper the slope, the higher the risk of landslide and the degree of erosion. It is also necessary to obtain the slope of each suspected landslide area in each period for subsequent analysis.

[0047] In the embodiment of the present invention, the neural network used in this embodiment is YOLOv3, and the method for obtaining the dataset for training this neural network is as follows:

[0048] A large number of remote sensing images are obtained, and the landslide areas are manually marked in each remote sensing image using bounding boxes, and this marking result is recorded as the label of each remote sensing image; the remote sensing images and their corresponding labels are collected to form a dataset; this dataset is used to train this neural network, and the loss function used during the training process is the mean square error loss function; the specific training process is well-known content of the neural network, and the specific training process is not described in detail in this embodiment; the remote sensing images of each period are input into the trained neural network to obtain the suspected landslide areas in the remote sensing images of each period, which are recorded as the suspected landslide areas of each period.

[0049] According to the digital elevation model (DEM), the slope of each suspected landslide area in each period is extracted.

[0050] It should be noted that long-term crustal movements, fault activities, or other natural phenomena may also cause changes in the slope, but the slope change is a long-term and slow process, while the occurrence of landslides caused by soil erosion is often accompanied by a drastic change in the slope. Therefore, if the slope of a certain suspected landslide area changes drastically and rapidly at any period, it may indicate that the possibility of landslide caused by erosion in this suspected landslide area is greater.

[0051] In the embodiment of the present invention, the specific method for obtaining the erosion risk factor of the j-th suspected landslide area at the i-th moment is as follows:

[0052] Preset the number of neighboring periods T = 10, and the i-th period and the previous T periods are recorded as the neighboring periods of the i-th period; obtain the position coordinates of the center point of the j-th suspected landslide area in the i-th period, which are recorded as the target coordinates; if there is a suspected landslide area at the position of the target coordinates in any neighboring period of the i-th period, the slope of this suspected landslide area is recorded as the target slope of this neighboring period of the i-th period; if there is no suspected landslide area at the position of the target coordinates in any neighboring period of the i-th period, then the target slope of this neighboring period of the i-th period is 0;

[0053] It should be noted that if the number of periods before the $i$-th period does not meet $T$, continue to obtain data from earlier periods in the database until the number of periods before the $i$-th period meets $T$.

[0054] Obtain the standard deviation of the target slope differences between all adjacent neighborhood periods at the $i$-th moment, denoted as the degree of slope change of the $j$-th suspected landslide area in the $i$-th period;

[0055] Obtain the maximum value among the target slope differences between all adjacent neighborhood periods at the $i$-th moment, denoted as the maximum slope change of the $j$-th suspected landslide area in the $i$-th period; obtain the minimum value among the target slope differences between all adjacent neighborhood periods at the $i$-th moment, denoted as the minimum slope change of the $j$-th suspected landslide area in the $i$-th period;

[0056] Obtain the erosion risk factor for each suspected landslide area in each period:

[0057] ;

[0058] In the formula, represents the erosion risk factor of the $j$-th suspected landslide area in the $i$-th period; represents the degree of slope change of the $j$-th suspected landslide area in the $i$-th period; represents the maximum slope change of the $j$-th suspected landslide area in the $i$-th period; represents the minimum slope change of the $j$-th suspected landslide area in the $i$-th period;

[0059] The greater the degree of slope change of the $j$-th suspected landslide area in the $i$-th period, the greater the possibility that the $j$-th suspected landslide area in the $i$-th period is eroded; The larger the value of, the greater the slope change of the $j$-th suspected landslide area in the $i$-th period and the greater the erosion risk factor.

[0060] It should be noted that cluster the suspected landslide areas in each period to obtain each risk area at each moment. If a certain risk area contains a larger number of suspected landslide areas, it means that the overall landslide area in this risk area is more concentrated and the degree of erosion is greater; and the greater the average value of the erosion risk factors of all suspected landslide areas in the risk area, the greater the degree of erosion of this risk area.

[0061] In an embodiment of the present invention, for any suspected landslide area in any period, the erosion factor and location coordinates of the suspected landslide area are used as the feature vector of the suspected landslide area; the DBSCAN clustering algorithm is used to cluster the feature vectors of all suspected landslide areas in the period to obtain several clusters; a region composed of all suspected landslide areas in each cluster is used as a risk area in the period;

[0062] Obtain the erosion degree of each risk area in each period:

[0063] ;

[0064] wherein, represents the erosion degree of the k-th risk area in the i-th period; represents the number of suspected landslide areas in the k-th risk area in the i-th period; represents the number of all suspected landslide areas in the i-th period; represents the mean value of the erosion risk factors of all suspected landslide areas in the k-th risk area in the i-th period; represents the mean value of the erosion risk factors of all suspected landslide areas in all risk areas in the i-th period.

[0065] S003. Obtain the erosion change rate of each risk area in each period; record the most recent period as the current period, and obtain the erosion growth of each risk area in the current period according to the increase in the erosion change rate of the same risk area in each period; obtain the environmental stability of each risk area in the current period according to the stability of the erosion degree of the risk area in all periods; based on the erosion growth and environmental stability, obtain the overall attention of each risk area in the current period.

[0066] It should be noted that when the most recent period is recorded as the current period, the risk areas in the current period are evolved from the risk areas in historical periods through soil erosion. Therefore, any risk area in the current period must correspond to the risk areas in each historical period; thus, if the erosion degree of a certain risk area in the current period has been increasing in the previous periods, it means that the erosion trend of this risk area in the current period is increasing, so the attention to this risk area in the current period needs to be increased;

[0067] Therefore, first, according to the difference in the erosion degree of the same risk area in adjacent periods, the erosion change rate of each risk area in each period is obtained; if the erosion change rate of the same risk area in each period is larger, it indicates that the risk area in the current period may face more serious erosion risk and needs to be focused on, then the erosion growth of the risk area at the current moment is larger; if the erosion change rate of the same risk area in each period tends to be negative or gradually decreases, it indicates that the soil erosion problem of the risk area in the current period may be alleviated, which may be related to natural recovery or effective prevention and control measures, and then the attention can be reduced, and at this time, the erosion growth of the risk area in the current period is smaller.

[0068] It should be further noted that by the spatial distance between the centers of the risk areas in two adjacent periods, the two risk areas with the closest spatial distance are selected, and the same risk area is corresponding in adjacent periods, so as to ensure that the same risk area can be accurately corresponding in different periods;

[0069] In the embodiment of the present invention, the erosion change rate of each risk area in each period is obtained:

[0070] ;

[0071] In the formula, represents the erosion change rate of the k-th risk area in the i-th period; represents the erosion degree of the k-th risk area in the i-th period; represents the erosion degree of the k-th risk area in the (i - 1)-th period;

[0072] Taking the most recent period as the current period, the erosion growth of each risk area in the current period is obtained:

[0073] ;

[0074] In the formula, represents the erosion growth of the k-th risk area in the current period; represents the total number of all periods; represents the erosion change rate of the k-th risk area in the i-th period; norm() represents the normalization function;

[0075] It should be noted that by comparing the erosion degrees of the same risk area in all periods, it is possible to identify whether the erosion degree of the risk area is stable in each period. For a risk area with a stable erosion change trend, it indicates that the natural environment of the risk area is relatively stable in each period, and the existing management measures can be continued; for a risk area with an unstable erosion change trend, it indicates that the soil erosion of the risk area may be affected by multiple natural environmental factors, and more interventions and measures are needed to reduce its volatility and avoid the intensification of sudden erosion.

[0076] Therefore, first use the least squares method to perform a linear fit on the erosion degrees of the same risk area in all periods to obtain the erosion fitting line of the risk area; if the slope of the erosion fitting line of any risk area is close to the change rate of the erosion degree of the risk area in adjacent periods, it means that the erosion degree of the risk area is very stable in all periods, and the greater the environmental stability of the risk area in the current period. At this time, it indicates that the natural environment of the risk area is relatively stable, and the existing management measures can be continued; on the contrary, when the difference is greater, it indicates that the natural environment of the risk area is unstable at this time, indicating that the environmental stability of the risk area in the current period is smaller, and the soil erosion may be affected by multiple factors, and more interventions and measures are needed to reduce its volatility and avoid the intensification of sudden erosion.

[0077] In the embodiment of the present invention, use the least squares method to perform a linear fit on the erosion degrees of the k-th risk area in all periods to obtain the erosion fitting line of the k-th risk area;

[0078] Obtain the environmental stability of each risk area in the current period:

[0079] ;

[0080] In the formula, represents the environmental stability of the k-th risk area in the current period; represents the number of all periods; represents the slope of the erosion fitting line of the k-th risk area; represents the change rate of the erosion degree of the k-th risk area in the i-th period and the i+1-th period; exp() represents the exponential function with the natural constant as the base;

[0081] It should be noted that if the environmental stability of any risk area in the current period is lower and the erosion trend is higher, then the overall attention to the risk area in the current period is greater.

[0082] Multiply the reciprocal of the environmental stability of the k-th risk area in the current period by the erosion growth of the k-th risk area in the current period, and use the product as the overall attention of the k-th risk area in the current period; similarly, obtain the overall attention of each risk area in the current period.

[0083] S004. Set the collection frequency of each risk area in the current period according to the overall attention of each risk area in the current period.

[0084] It should be noted that in order to efficiently collect data on different risk areas under limited resources, it is necessary to sort the high-efficiency collection frequencies of different risk areas in different current periods. Therefore, the attention of each risk area in the current period can be obtained by comparing the overall attention of different risk areas.

[0085] In the embodiment of the present invention, obtain the mean value of the difference in overall attention between each risk area in the current period and all other risk areas, denote it as the importance degree of each risk area in the current period, and perform linear normalization processing on the importance degree to obtain the attention of each risk area in the current period.

[0086] Preset a high threshold T1 = 0.7 and a low threshold T2 = 0.3. If the attention is greater than or equal to T1, set the monitoring frequency of the corresponding risk area to once a day; if the attention is less than or equal to T2, set the monitoring frequency of the corresponding risk area to once a month; if the attention is greater than T2 and less than T1, set the monitoring frequency of the corresponding risk area to once a week.

[0087] The above are only the preferred embodiments of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent monitoring method for the change amount of soil erosion in soil and water treatment, characterized in that, Including: Collect remote sensing images for each period; Based on the remote sensing images for each period, obtain each suspected landslide area and its slope for each period; Based on the slope change of each suspected landslide area for each period, obtain the erosion risk factor of each suspected landslide area for each period; Cluster the suspected landslide areas for each period according to the erosion risk factor, and obtain each risk area for each period; Obtain the erosion degree of each risk area for each period; According to the erosion degree, obtain the erosion change rate of each risk area for each period; Record the most recent period as the current period, and take the sum of the erosion change rates of the same risk area in all periods as the erosion growth of each risk area in the current period; According to the stability of the erosion degree of the same risk area in all periods, obtain the environmental stability of each risk area in the current period; Based on the erosion growth and environmental stability, obtain the overall attention of each risk area in the current period; Set the acquisition frequency of each risk area in the current period according to the overall attention of each risk area in the current period.

2. The intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, wherein The obtaining of each suspected landslide area and its slope for each period includes: Train a neural network based on the landslide areas manually marked in a large number of remote sensing images, and input the remote sensing images for each period into the trained neural network to obtain the suspected landslide areas in the remote sensing images for each period, denoted as the suspected landslide areas for each period; Extract the slope of each suspected landslide area for each period according to the digital elevation model.

3. The intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, wherein The obtaining of the erosion risk factor of each suspected landslide area for each period includes: Obtain the maximum and minimum values of the slope change of the j-th suspected landslide area in the i-th period; ； In the formula, represents the erosion risk factor of the j-th suspected landslide area in the i-th period; represents the degree of slope change of the j-th suspected landslide area in the i-th period; represents the maximum value of slope change of the j-th suspected landslide area in the i-th period; represents the minimum value of slope change of the j-th suspected landslide area in the i-th period.

4. An intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 3, characterized in that, The obtaining of the slope change degree of the j-th suspected landslide area in the i-th period includes: Preset the number of neighboring periods T, and denote the i-th period and the previous T periods as the neighboring periods of the i-th period; obtain the position coordinates of the center point of the j-th suspected landslide area in the i-th period, denoted as the target coordinates; if there is a suspected landslide area at the position of the target coordinates in any neighboring period of the i-th period, record the slope of this suspected landslide area as the target slope of this neighboring period of the i-th period; if there is no suspected landslide area, the target slope of this neighboring period of the i-th period is 0; obtain the standard deviation of the target slope differences between all adjacent neighboring periods at the i-th moment, denoted as the slope change degree of the j-th suspected landslide area in the i-th period.

5. An intelligent monitoring method for the amount of soil erosion change for soil and water treatment according to claim 3 or 4, characterized in that, The obtaining of the maximum and minimum values of the slope change of the j-th suspected landslide area in the i-th period includes: Obtain the maximum value among the target slope differences between all adjacent neighboring periods at the i-th moment, denoted as the maximum value of the slope change of the j-th suspected landslide area in the i-th period; obtain the minimum value among the target slope differences between all adjacent neighboring periods at the i-th moment, denoted as the minimum value of the slope change of the j-th suspected landslide area in the i-th period.

6. The intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, wherein The obtaining of each risk area for each period includes: For any suspected landslide area in any period, the erosion factor and location coordinates of the suspected landslide area are used as the feature vector of the suspected landslide area; the DBSCAN clustering algorithm is used to cluster the feature vectors of all suspected landslide areas in the period to obtain several clusters; a region composed of all suspected landslide areas in each cluster is used as a risk area in the period.

7. An intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, characterized in that, The obtaining of the erosion change rate of each risk area in each period includes: ； In the formula, represents the erosion change rate of the k-th risk area in the i-th period; represents the degree of erosion of the k-th risk area in the i-th period; represents the degree of erosion of the k-th risk area in the (i - 1)-th period.

8. The intelligent monitoring method for the amount of soil erosion change for soil and water treatment according to claim 1, characterized in that, The obtaining of the environmental stability of each risk area in the current period includes: Using the least squares method to perform a linear fit on the erosion degree of the k-th risk area in all periods to obtain the erosion fit line of the k-th risk area; ； In the formula, represents the environmental stability of the k-th risk area in the current period; represents the number of all periods; represents the slope of the erosion fitting line of the k-th risk area; represents the change rate of the erosion degree of the k-th risk area between the i-th period and the (i + 1)-th period; exp() represents the exponential function with the natural constant as the base.

9. The intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, characterized in that, The obtaining of the overall attention of each risk area in the current period includes: The product of the reciprocal of the environmental stability of the k-th risk area in the current period and the erosion growth of the k-th risk area in the current period is used as the overall attention of the k-th risk area in the current period.

10. The intelligent monitoring method for the change amount of soil erosion for soil and water treatment according to claim 1, characterized in that The setting of the acquisition frequency of each risk area in the current period according to the overall attention of each risk area in the current period includes: Obtain the mean value of the difference in overall attention between each risk area in the current period and all other risk areas, which is denoted as the importance degree of each risk area in the current period. Perform linear normalization on the importance degree to obtain the attention degree of each risk area in the current period. Set a high threshold T1 and a low threshold T2. If the attention degree is greater than or equal to T1, set the monitoring frequency of the corresponding risk area to once a day; if the attention degree is less than or equal to T2, set the monitoring frequency of the corresponding risk area to once a month; if the attention degree is greater than T2 and less than T1, set the monitoring frequency of the corresponding risk area to once a week.

Images