Efficient monitoring and intelligent irrigation management platform for radar rainfall data

Abstract

The invention relates to the technical field of data processing, in particular to an efficient monitoring and intelligent irrigation management platform for radar rainfall data. Obtaining a first abnormal point location and a non-initial abnormal point location according to a data volume feature in an intensity data cluster of the echo intensity and a data difference feature of the intensity data cluster; obtaining a point location connected domain according to the intensity data cluster of the non-initial abnormal point location; obtaining a spatial noise feature value according to the position distribution feature between the point location connected domains and the data difference feature in the point location connected domains; obtaining a time noise characteristic value according to the historical echo intensity of the non-initial abnormal point location; obtaining a second abnormal point location according to the spatial noise characteristic value and the time noise characteristic value; correcting the corresponding echo intensity according to the position features of the first abnormal point location and the second abnormal point location; and the rainfall is determined according to the corrected echo intensity, and the irrigation rule is adjusted, so that the accuracy of rainfall judgment and irrigation regulation is improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a high-efficiency monitoring and intelligent irrigation management platform for radar precipitation data. Background Technology

[0002] A smart irrigation management platform refers to the intelligent management of irrigated areas through data analysis and water volume control, aiming to improve the automation and accuracy of irrigation. Existing smart irrigation technologies can combine radar precipitation data monitoring to track rainfall patterns in subsequent precipitation areas, thereby controlling irrigation volume and preventing over-flooding or under-irrigation. Radar precipitation measurement utilizes meteorological radar to detect the scattering effect of precipitation particles on electromagnetic waves, calculating precipitation intensity and distribution. However, in actual monitoring, radar precipitation monitoring is susceptible to interference from complex environments, leading to inaccurate precipitation data in some areas. This reduces the accuracy of the smart irrigation management platform in adjusting regional irrigation volumes, ultimately negatively impacting crop growth. Summary of the Invention

[0003] To address the aforementioned technical problems, the present invention aims to provide an efficient monitoring and intelligent irrigation management platform for radar precipitation data. The specific technical solution adopted is as follows: The data acquisition module is used to acquire the echo intensity of different locations monitored by the meteorological radar; The first rainfall analysis module is used to obtain different intensity data clusters based on the differences in echo intensity at all points; to obtain the initial anomaly degree based on the data volume characteristics in the intensity data clusters and the data difference characteristics between the intensity data clusters and other intensity data clusters; and to obtain the first anomaly points and non-initial anomaly points based on the initial anomaly degree. The second rainfall analysis module is used to obtain different connected regions of points based on the intensity data clusters of the non-initial anomaly points; to obtain spatial noise feature values ​​of the connected regions of points based on the positional distribution characteristics between the connected regions of points and the data difference characteristics within the connected regions of points; to obtain temporal noise feature values ​​based on the historical echo intensity variation characteristics of the non-initial anomaly points; to obtain the noise level based on the spatial noise feature values ​​and the temporal noise feature values; and to obtain the second anomaly point based on the noise level. The data adjustment module is used to correct the corresponding echo intensity based on the location characteristics of the first and second abnormal points; and to determine the precipitation and adjust the irrigation rules based on the corrected echo intensity.

[0004] Furthermore, the step of obtaining different intensity data clusters based on the differences in echo intensity at all locations includes: The echo intensity of all points is clustered using the K-means clustering algorithm to obtain different intensity data clusters.

[0005] Further, the step of obtaining the initial anomaly degree based on the data volume characteristics in the intensity data cluster and the data difference characteristics between the intensity data cluster and other intensity data clusters includes: Calculate the reciprocal of the echo intensity quantity value in the intensity data cluster and normalize it to obtain the sparsity; calculate the average of the absolute values ​​of the differences between the cluster centers of the intensity data cluster and all other intensity data clusters and normalize it to obtain the data difference; calculate the average of the sparsity and the data difference to obtain the initial anomaly of the intensity data cluster.

[0006] Further, the step of obtaining the first anomaly point and the non-initial anomaly point based on the initial anomaly degree includes: When the initial anomaly of the intensity data cluster exceeds a preset first threshold, the point corresponding to the echo intensity within the intensity data cluster is the first anomaly point; otherwise, it is a non-initial anomaly point.

[0007] Furthermore, the step of obtaining different point connectivity components based on the intensity data clusters of the non-initial anomaly points includes: Connect adjacent non-initial outlier points that belong to the same intensity data cluster to obtain a point connectivity region.

[0008] Further, the step of obtaining the spatial noise feature value of the connected regions based on the positional distribution characteristics between the connected regions and the data difference characteristics within the connected regions includes: The sum of the nearest distances between the connected region of the point and each other connected region of the same intensity data cluster is calculated and normalized to obtain the distribution interval; the average of the absolute values ​​of the differences between the echo intensity of all points in the connected region and the mean echo intensity is calculated and normalized to obtain the data dispersion; the average of the distribution interval and the data dispersion is calculated to obtain the spatial noise characteristic value of the connected region of the point.

[0009] Furthermore, the step of obtaining the temporal noise feature value based on the variation characteristics of the echo intensity in the history of the non-initial anomaly point includes: Calculate and normalize the standard deviation of the echo intensity of the non-initial anomaly point during a preset historical period to obtain the degree of dispersion; calculate and normalize the sum of the absolute values ​​of the differences in echo intensity between adjacent times during the preset historical period of the non-initial anomaly point to obtain the degree of fluctuation; calculate the average of the degree of dispersion and the degree of fluctuation to obtain the time noise characteristic value.

[0010] Further, the step of obtaining the noise level based on the spatial noise feature value and the temporal noise feature value includes: The noise level of the non-initial anomaly point is obtained by calculating the average of the temporal noise feature value and the spatial noise feature value of the connected domain of the non-initial anomaly point.

[0011] Further, the step of obtaining the second anomaly location based on the noise level includes: The non-initial abnormal points whose noise level exceeds a preset second threshold are designated as the second abnormal points.

[0012] Furthermore, the step of correcting the corresponding echo intensity based on the positional characteristics of the first and second anomaly points includes: The connected regions of points where there is no second abnormal point are treated as normal connected regions; the echo intensity of the first or second abnormal point is replaced with the average echo intensity of the nearest normal connected region.

[0013] The present invention has the following beneficial effects: In this invention, acquiring intensity data clusters can distinguish different echo intensities and initially identify significant noise data within the echo intensity. Acquiring the initial anomaly level can characterize the likelihood that the intensity data cluster is noise data, and acquiring the first anomaly point and non-initial anomaly points can distinguish significant noise data. Acquiring the point connectivity domain can further distinguish noise data that may exist in non-initial points. Since the characteristics of the point connectivity domain differ between normal rainfall areas and environmental interference areas, acquiring spatial noise feature values ​​can characterize the possibility of noise data in the point connectivity domain. Since the echo intensity variation characteristics differ between normal rainfall areas and environmental interference areas, acquiring temporal noise feature values ​​can characterize the possibility of noise data in non-initial anomaly points. Acquiring the noise level can further accurately characterize the possibility that non-initial anomaly points are noise data. Finally, by correcting the corresponding echo intensity based on the location characteristics of the first and second anomaly points, the accuracy of rainfall intensity judgment in irrigation areas can be improved, resulting in more accurate irrigation volume adjustment. Attached Figure Description

[0014] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a block diagram of an efficient radar precipitation data monitoring and intelligent irrigation management platform provided in one embodiment of the present invention. Detailed Implementation

[0016] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a high-efficiency monitoring and intelligent irrigation management platform for radar precipitation data proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0018] The following description, in conjunction with the accompanying drawings, details the specific solution of the high-efficiency monitoring and intelligent irrigation management platform for radar precipitation data provided by the present invention.

[0019] Please see Figure 1 The diagram illustrates a block diagram of an efficient monitoring and intelligent irrigation management platform for radar precipitation data according to an embodiment of the present invention. The platform includes the following modules: The data acquisition module S1 is used to acquire the echo intensity of different locations monitored by the meteorological radar.

[0020] In this embodiment of the invention, the implementation scenario involves an intelligent irrigation management platform adjusting the irrigation amount based on future precipitation data monitored by meteorological radar, thereby avoiding problems of excessive or insufficient irrigation. First, meteorological radar is used to monitor and collect precipitation data over a period of time in the irrigated area. The radar monitoring is multi-point, with each point monitoring and collecting the echo intensity of a portion of the area. In the two-dimensional radar mosaic, each point corresponds to an echo intensity, and the degree of precipitation can be calculated from the echo intensity.

[0021] The first rainfall analysis module S2 is used to obtain different intensity data clusters based on the differences in echo intensity of all points; to obtain the initial anomaly degree based on the data volume characteristics in the intensity data cluster and the data difference characteristics between the intensity data cluster and other intensity data clusters; and to obtain the first anomaly point and non-initial anomaly points based on the initial anomaly degree.

[0022] Radar monitoring of precipitation is susceptible to interference from complex environments, leading to noise data in the echo intensity. This noise typically includes ground clutter and thermal noise from the radar system. Ground clutter, such as buildings, towers, and dust and smoke generated by human activities, reflects radar waves, creating false strong echoes at certain locations. Thermal noise from the radar system, such as random signals from electronic devices, forms scattered points in weak echo areas. Noise in the echo intensity interferes with the assessment of precipitation intensity, affecting the accuracy of irrigation adjustments and negatively impacting crop growth. Therefore, it is necessary to correct the echo intensity at certain points to improve the accuracy of precipitation intensity analysis. Firstly, different intensity data clusters are obtained based on the differences in echo intensity across all points. Preferably, in this embodiment, the step of obtaining intensity data clusters includes: clustering the echo intensity of all points using a K-means clustering algorithm to obtain different intensity data clusters. It should be noted that the K-means clustering algorithm is existing technology, and its specific steps will not be elaborated further. The number of clusters is obtained using the elbow method. Intensity data clusters can group data with similar echo intensities into a single cluster. Because rainfall has a localized effect, when rainfall is about to occur in a certain area, it will cause a large area of ​​points to be similarly affected by the echo intensity, which has a larger impact range than ground clutter and thermal noise. Furthermore, the echo intensity affected by rainfall differs significantly from the echo intensity of ground clutter and thermal noise. Therefore, the initial anomaly degree can be obtained based on the data volume characteristics in the intensity data cluster and the data difference characteristics between the intensity data cluster and other intensity data clusters.

[0023] Preferably, in this embodiment of the invention, the step of obtaining the initial anomaly level includes: calculating the reciprocal of the number of echo intensity values ​​in the intensity data cluster and normalizing it to obtain the sparsity level; in this embodiment of the invention, the normalization method is linear normalization. The smaller the amount of data in the intensity data cluster, the greater the sparsity, meaning that the echo intensity in the cluster is more likely to be noise data generated by environmental interference. Calculate the average of the absolute values ​​of the differences between the cluster centers of the intensity data cluster and all other intensity data clusters and normalize it to obtain the data difference level; since noise data caused by ground clutter or thermal noise differs significantly from the echo intensity of normally monitored areas, while the echo intensity at different points in normal rainfall areas is relatively close, the greater the difference in echo intensity between this intensity data cluster and other intensity data clusters, the greater the data difference level, meaning that this intensity data cluster is more likely to be generated by environmental interference. Calculate the average of the sparsity level and the data difference level to obtain the initial anomaly level of the intensity data cluster; the greater the initial anomaly level, the more likely the echo intensity in the intensity data cluster is to be noise data rather than true echo intensity.

[0024] Furthermore, the first and non-initial anomaly points can be obtained based on the initial anomaly degree. Preferably, in this embodiment, when the initial anomaly degree of the intensity data cluster exceeds a preset first threshold, the point corresponding to the echo intensity within that intensity data cluster is the first anomaly point; otherwise, it is a non-initial anomaly point. Since the initial anomaly degree in this embodiment ranges from 0 to 1, the initial anomaly degree of noisy data tends to 1, while the initial anomaly degree of normal echo intensity tends to 0. These two are distributed at opposite ends of the value range. Therefore, in this embodiment, the preset first threshold is set to 0.5, which can be determined by the implementer according to the implementation scenario. The first anomaly point means that the echo intensity at that point is significantly abnormal. For non-initial anomaly points, the reliability of the echo intensity needs to be further determined.

[0025] The second rainfall analysis module S3 is used to obtain different point connectivity regions based on the intensity data clusters of non-initial anomaly points; to obtain spatial noise characteristic values ​​of point connectivity regions based on the location distribution characteristics between point connectivity regions and the data difference characteristics within point connectivity regions; to obtain temporal noise characteristic values ​​based on the historical echo intensity variation characteristics of non-initial anomaly points; to obtain the noise level based on the spatial noise characteristic values ​​and temporal noise characteristic values; and to obtain the second anomaly point based on the noise level.

[0026] Since the first anomaly points were selected based on echo intensity values, they could only distinguish relatively obvious noise data. However, some noise data had echo intensity values ​​similar to those of normal rainfall areas, leading to these points being mistakenly identified as rainfall areas. For example, dense insect swarms are common in irrigated farmland areas. The aggregation and distribution characteristics of insect swarms result in a large distribution range, and their weak echoes can easily be confused with light rain, forming noise data that is not easily distinguishable. Therefore, further analysis of non-initial anomaly points is necessary. First, different point connectivity regions are obtained based on the intensity data clusters of non-initial anomaly points. Specifically, this includes connecting adjacent non-initial anomaly points belonging to the same intensity data cluster to obtain a point connectivity region; and forming different point connectivity regions in the radar mosaic of the two-dimensional plane. Because the distribution of insect swarms varies in density and shape, the corresponding noise data in the two-dimensional plane will exhibit discrete and gap-like characteristics; while the point connectivity regions corresponding to rainfall areas are relatively dense. Furthermore, the density of insect swarms varies, resulting in significant differences in the intensity of reflected echoes. Meanwhile, rainfall intensities are similar within the same rainfall area, leading to similar echo intensities. Therefore, the spatial noise characteristic value of a point-connected domain can be obtained based on the location distribution characteristics between point-connected domains and the data difference characteristics within the point-connected domains.

[0027] Preferably, in this embodiment of the invention, the step of obtaining spatial noise feature values ​​includes: calculating and normalizing the sum of the nearest distances between the connected regions of a point and each other connected regions within the same intensity data cluster to obtain the distribution interval; the greater the distance between connected regions of points, the larger the distribution interval, meaning that the connected region of a point is more likely to be formed by the influence of insect swarms. Calculating and normalizing the average of the absolute values ​​of the differences between the echo intensity and the mean echo intensity of all points within the connected region of a point to obtain the data dispersion; the greater the difference between the echo intensity and the mean echo intensity of points within the connected region of a point, the larger the data dispersion, meaning that the echo intensity distribution is more inconsistent, and more likely to be caused by differences in the density of insect swarms; the smaller the data dispersion, the more similar the echo intensity, and more likely to be the echo intensity corresponding to a rainfall area. Calculating the average of the distribution interval and the data dispersion to obtain the spatial noise feature value of the connected region of a point; the larger the spatial noise feature value, the more likely the connected region of a point is caused by the distribution of insect swarms. The formula for obtaining spatial noise feature values ​​includes:

[0028] In the formula, R represents the spatial noise characteristic value of the connected domain of the point. This indicates normalization, and N represents the number of other connected components of the same intensity within the same data cluster for this point. This represents the shortest distance between the connected component of this point and the connected component of the nth other point. The interval of distribution is represented by M, where M represents the number of points within the connected domain of a point. This represents the echo intensity at the m-th point. This represents the mean echo intensity of the connected region at that point. This indicates the degree of data dispersion.

[0029] Furthermore, rainfall intensity changes relatively slowly or at the same location over a period of time, while insect swarms move quickly and are unlikely to remain stationary at the same location for an extended period. Therefore, a significant change in echo intensity across multiple historical monitoring sessions at the same location is more likely to be caused by insect swarms. Thus, temporal noise characteristic values ​​are obtained based on the historical echo intensity variation characteristics of non-initial anomaly locations. Preferably, in this embodiment, the step of obtaining temporal noise characteristic values ​​includes: calculating and normalizing the standard deviation of echo intensity over a preset historical period at non-initial anomaly locations to obtain the degree of dispersion; a larger degree of dispersion indicates a greater difference in echo intensity, making it less likely to be influenced by rainfall. In this embodiment, the preset historical period is the range of the ten most recent monitoring times, including the current monitoring time, which can be determined by the implementer according to the implementation scenario. The sum of the absolute values ​​of the differences in echo intensity between adjacent times within the preset historical period at non-initial anomaly locations is calculated and normalized to obtain the degree of fluctuation; a larger degree of fluctuation indicates a faster change in echo intensity, making it less likely to be influenced by rainfall. Calculate the average of the dispersion and fluctuation to obtain the temporal noise characteristic value of the non-initial anomaly point; the larger the temporal noise characteristic value, the more likely the echo intensity of the non-initial anomaly point is to be affected by the insect swarm.

[0030] After obtaining the spatial noise characteristic values ​​and temporal noise characteristic values, the noise level can be obtained based on these values. Preferably, in this embodiment of the invention, the step of obtaining the noise level includes: calculating the average of the temporal noise characteristic values ​​and the spatial noise characteristic values ​​of the connected domain of the non-initial anomaly point to obtain the noise level of the non-initial anomaly point; the higher the noise level, the more likely the echo intensity of the non-initial anomaly point is to be noise data, making it difficult to reflect the true precipitation intensity. Furthermore, a second anomaly point can be obtained based on the noise level, specifically by: identifying non-initial anomaly points with noise levels exceeding a preset second threshold as second anomaly points; since the noise level range is from 0 to 1, and the noise levels of noise data and normal echo data are distributed at both ends of the range, in this embodiment of the invention, the preset second threshold is 0.5, which can be determined by the implementer according to the implementation scenario. The second anomaly point represents a point whose echo intensity is similar to other points in the region but belongs to noise data.

[0031] The data adjustment module S4 is used to correct the corresponding echo intensity based on the location characteristics of the first and second abnormal points; and to determine the precipitation and adjust the irrigation rules based on the corrected echo intensity.

[0032] After identifying the first and second anomalous points across all locations, to improve the accuracy of rainfall intensity assessment, the echo intensity of these points needs to be adjusted. Therefore, the echo intensity is corrected based on the location characteristics of the first and second anomalous points. Preferably, in this embodiment, the step of correcting the echo intensity includes: treating the connected regions of points without second anomalous points as normal connected regions; since rainfall conditions are similar in adjacent areas, the echo intensities of adjacent points are similar; therefore, the echo intensity of the first or second anomalous point is replaced with the average echo intensity of the nearest normal connected region. After echo intensity correction, rainfall can be determined based on the corrected echo intensity, and irrigation rules can be adjusted. The radar echo intensity is converted into rainfall intensity using the existing QPE (Quantitative Estimation of Rainfall) method, and adjustments are made through an intelligent irrigation management platform based on future rainfall intensity. For example, if heavy rainfall occurs in the future, the regional irrigation amount is reduced; or if there is no rainfall in the future, the regional irrigation amount is increased. Implementers can determine the irrigation adjustment rules according to the implementation scenario, thereby ensuring the growth environment of water-sensitive crops. Thus, by adjusting the echo intensity of potentially noisy data in the weather radar, the accuracy of rainfall intensity judgment in the irrigation area can be improved, making irrigation volume adjustment more accurate.

[0033] In summary, the embodiments of the present invention provide an efficient monitoring and intelligent irrigation management platform for radar precipitation data; first anomaly points and non-initial anomaly points are obtained based on the data volume characteristics and data difference characteristics of the intensity data clusters of echo intensity; the point connectivity domains are obtained based on the intensity data clusters of the non-initial anomaly points; spatial noise characteristic values ​​are obtained based on the positional distribution characteristics between the point connectivity domains and the data difference characteristics within the point connectivity domains; temporal noise characteristic values ​​are obtained based on the historical echo intensity of the non-initial anomaly points; second anomaly points are obtained based on the spatial noise characteristic values ​​and temporal noise characteristic values; the corresponding echo intensity is corrected based on the positional characteristics of the first and second anomaly points; and precipitation is determined and irrigation rules are adjusted based on the corrected echo intensity, thereby improving the accuracy of rainfall judgment and irrigation regulation.

[0034] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0035] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A high-efficiency monitoring and intelligent irrigation management platform for radar precipitation data, characterized in that, The platform includes the following modules: The data acquisition module is used to acquire the echo intensity of different locations monitored by the meteorological radar; The first rainfall analysis module is used to obtain different intensity data clusters based on the differences in echo intensity at all points; and to obtain the initial anomaly degree based on the data volume characteristics in the intensity data clusters and the data difference characteristics between the intensity data clusters and other intensity data clusters. The first and non-initial anomaly points are obtained based on the initial anomaly level. The second rainfall analysis module is used to obtain different point connectivity regions based on the intensity data clusters of the non-initial abnormal points; and to obtain the spatial noise characteristic value of the point connectivity regions based on the positional distribution characteristics between the point connectivity regions and the data difference characteristics within the point connectivity regions. The temporal noise characteristic value is obtained based on the variation characteristics of the echo intensity in the history of the non-initial anomaly points; the noise level is obtained based on the spatial noise characteristic value and the temporal noise characteristic value. The second anomaly location is obtained based on the noise level. The data adjustment module is used to correct the corresponding echo intensity based on the location characteristics of the first and second abnormal points; and to determine the precipitation and adjust the irrigation rules based on the corrected echo intensity.

2. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining different intensity data clusters based on the differences in echo intensity at all points includes: The echo intensity of all points is clustered using the K-means clustering algorithm to obtain different intensity data clusters.

3. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the initial anomaly level based on the data volume characteristics in the intensity data cluster and the data difference characteristics between the intensity data cluster and other intensity data clusters includes: Calculate the reciprocal of the echo intensity quantity value in the intensity data cluster and normalize it to obtain the sparsity; calculate the average of the absolute values ​​of the differences between the cluster centers of the intensity data cluster and all other intensity data clusters and normalize it to obtain the data difference; calculate the average of the sparsity and the data difference to obtain the initial anomaly of the intensity data cluster.

4. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the first anomaly point and the non-initial anomaly point based on the initial anomaly degree includes: When the initial anomaly of the intensity data cluster exceeds a preset first threshold, the point corresponding to the echo intensity within the intensity data cluster is the first anomaly point; otherwise, it is a non-initial anomaly point.

5. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining different point connectivity components based on the intensity data clusters of the non-initial anomaly points includes: Connect adjacent non-initial outlier points that belong to the same intensity data cluster to obtain a point connectivity region.

6. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the spatial noise feature value of the connected regions of points based on the positional distribution characteristics between the connected regions of points and the data difference characteristics within the connected regions of points includes: The sum of the nearest distances between the connected region of the point and each other connected region of the same intensity data cluster is calculated and normalized to obtain the distribution interval; the average of the absolute values ​​of the differences between the echo intensity of all points in the connected region and the mean echo intensity is calculated and normalized to obtain the data dispersion; the average of the distribution interval and the data dispersion is calculated to obtain the spatial noise characteristic value of the connected region of the point.

7. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the time noise feature value based on the change characteristics of the echo intensity of the non-initial anomaly point in history includes: Calculate and normalize the standard deviation of the echo intensity of the non-initial anomaly point during a preset historical period to obtain the degree of dispersion; calculate and normalize the sum of the absolute values ​​of the differences in echo intensity between adjacent times during the preset historical period of the non-initial anomaly point to obtain the degree of fluctuation; calculate the average of the degree of dispersion and the degree of fluctuation to obtain the time noise characteristic value.

8. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the noise level based on the spatial noise characteristic value and the temporal noise characteristic value includes: The noise level of the non-initial anomaly point is obtained by calculating the average of the temporal noise feature value and the spatial noise feature value of the connected domain of the non-initial anomaly point.

9. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of obtaining the second abnormal location based on the noise level includes: The non-initial abnormal points whose noise level exceeds a preset second threshold are designated as the second abnormal points.

10. The efficient monitoring and intelligent irrigation management platform for radar precipitation data according to claim 1, characterized in that, The step of correcting the corresponding echo intensity based on the location characteristics of the first and second anomaly points includes: The connected regions of points where there is no second abnormal point are treated as normal connected regions; the echo intensity of the first or second abnormal point is replaced with the average echo intensity of the nearest normal connected region.

Images