A real-time remote sensing method, detection system and early warning system for surface anomalies

By constructing a prior knowledge base of surface remote sensing features and combining long-term and real-time remote sensing data, the system identifies and calculates surface anomaly areas, solving the problems of timeliness and applicability of satellite remote sensing detection and enabling timely and accurate detection and early warning of surface anomalies.

CN119091308BActive Publication Date: 2026-11-17BEIJING NORMAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310663062.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-06
Publication Date
2026-11-17
Estimated Expiration
2043-06-06

AI Technical Summary

Technical Problem

Existing satellite remote sensing methods lack timeliness, cannot be applied to all types of surface anomalies, and have low levels of automation, making it difficult to meet the needs of real-time remote sensing services.

Method used

By constructing a prior knowledge base of surface remote sensing features, and utilizing long-term time-series surface remote sensing feature information and real-time remote sensing data, potential areas of surface anomalies can be identified and anomaly intensity can be calculated, enabling real-time on-board remote sensing detection.

Benefits of technology

It enables real-time and accurate detection and early warning of surface anomalies, is highly adaptable, and can realize real-time detection of surface anomalies in orbit, shortening the detection cycle and improving the identification accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119091308B_ABST
    Figure CN119091308B_ABST
Patent Text Reader

Abstract

A method, system, and early warning system for real-time remote sensing detection of surface anomalies include: acquiring first long-term time-series surface remote sensing feature information to form a first long-term time-series surface remote sensing feature data set; constructing a prior knowledge base of surface remote sensing features; acquiring real-time on-board remote sensing data, calculating surface remote sensing feature information, and identifying potential areas of surface anomalies; identifying surface anomaly areas within the potential areas of surface anomalies based on the prior knowledge base of surface remote sensing features; and calculating the intensity of surface anomalies in the surface anomaly areas. This invention achieves effective capture of normal surface conditions by constructing a prior knowledge base of surface remote sensing features; identifies surface anomalies based on real-time acquired remote sensing data using the prior knowledge base, achieving high accuracy and fast computation efficiency, and can serve as a real-time on-orbit remote sensing detection technology system for surface anomalies; finally, calculating the intensity of surface anomaly areas facilitates corresponding emergency response based on intensity, demonstrating strong adaptability and ease of promotion.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of real-time remote sensing services, specifically to a real-time remote sensing detection method, detection system, and early warning system for surface anomalies. Background Technology

[0002] In recent years, with the intensification of global climate change and the increasing human activities, various sudden surface anomalies (such as natural disasters, environmental pollution, ecological damage, safety accidents, and illegal development) have occurred frequently, exhibiting characteristics such as wide distribution, high frequency of occurrence, rapid evolution, and heavy economic losses. These events seriously affect the normal evolution of the natural environment and human well-being, and restrict the healthy and sustainable development of human society and the economy. Therefore, real-time monitoring and early warning of various surface anomalies have become paramount in ensuring ecological security, human health, and high-quality and stable socio-economic development. There is an urgent need to construct a fully automated, real-time detection method system capable of detecting all surface anomalies.

[0003] Studies have shown that satellite remote sensing is the most effective method for real-time detection of large-scale surface anomalies. However, due to the suddenness, diversity, randomness, concealment, and complexity of surface anomalies, many anomalies often go undetected during actual detection processes, especially failing to be detected and warned of in the first instance. The timeliness and accuracy of satellite detection are far from meeting the real-time remote sensing service needs characterized by minute-level response, such as disaster relief, accident handling, pollution control, and law enforcement. The main reasons for this phenomenon are as follows:

[0004] First, current satellite remote sensing early warning systems for surface anomalies typically involve a lengthy process, from mission planning, satellite operation and control, and data transmission between satellite and ground, to the acquisition of monitoring information and the extraction of early warning knowledge. For example, for emergency remote sensing monitoring by natural resources departments, the process first requires developing a multi-satellite collaborative imaging plan, then acquiring satellite images through a ground-based satellite data center, and finally processing these images through an application system to extract relevant remote sensing information on surface anomalies. On average, it takes 1-3 days to extract early warning knowledge such as the location, extent, and intensity of the target event from remote sensing images. Therefore, the process from satellite imagery to real-time detection and early warning of surface anomalies suffers from a long service chain, numerous links, complex procedures, low levels of intelligence, and slow response, which falls far short of meeting the actual needs of emergency response.

[0005] Secondly, almost all current research on surface anomaly detection focuses only on specific surface anomaly events, such as surface thermal anomalies (Rembold et al., 2019), floods (DeVries et al., 2020), wildfires (Zhang et al., 2021), earthquakes (Liu et al., 2020), and forest disturbances (Tang et al., 2019). There is a lack of remote sensing systems applicable to all types of surface anomalies, making it impossible to effectively detect all surface anomalies simultaneously, resulting in a large number of surface anomalies being overlooked. Furthermore, simply combining and superimposing existing detection methods for different surface anomalies would significantly increase the complexity of the entire surface anomaly detection method, severely reducing computational efficiency. At the same time, logical conflicts between different surface anomaly detection methods and limitations of application scenarios could lead to false positives and false negatives in some areas, reducing the accuracy of surface anomaly identification and failing to meet the needs of real-time remote sensing services.

[0006] Third, most current remote sensing methods for detecting surface anomalies generally suffer from low levels of automation, complex model structures, and low accuracy in some scenarios, making it difficult to meet the high timeliness requirements of real-time remote sensing services. Currently, surface anomaly remote sensing methods can be broadly categorized into manual visual interpretation, threshold segmentation, overlay and multi-model ensemble methods, and machine learning methods. Manual visual interpretation is relatively simple, but often requires significant manual intervention, has low automation, and long computation cycles. Threshold segmentation primarily identifies surface anomaly areas by segmenting remote sensing indices suitable for surface anomaly assessment. However, due to the complexity and diversity of surface anomaly types, some surface anomalies are quite similar to some normal landforms, easily leading to low accuracy in some scenarios. While overlay and multi-model ensemble methods yield higher accuracy results, their model structures are complex and often require substantial ground-based measurements or other auxiliary data. Compared to the aforementioned methods, machine learning methods have a higher level of intelligence. However, they not only face the same problems as threshold segmentation methods but also challenges in selecting massive samples and model generalization, all of which increase the difficulty of timely and accurate surface anomaly detection. Summary of the Invention

[0007] (I) Purpose of the Invention

[0008] The purpose of this invention is to provide a real-time remote sensing method, detection system, and early warning system for surface anomalies that addresses the problem that the timeliness of traditional satellite remote sensing detection is seriously lagging behind the actual needs of surface anomaly monitoring and emergency response, as well as the problem that it cannot be applied to all different types of surface anomalies.

[0009] (II) Technical Solution

[0010] To address the above problems, this invention provides a method for real-time remote sensing detection of surface anomalies, comprising:

[0011] The first long-term time series surface remote sensing feature information was obtained to form the first long-term time series surface remote sensing feature data set;

[0012] Based on the first long-term time series of surface remote sensing feature data, a prior knowledge base of surface remote sensing features is constructed.

[0013] Acquire real-time remote sensing data from satellites, calculate surface remote sensing feature information, and identify potential areas of surface anomalies based on the current surface remote sensing feature information;

[0014] Based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features, identify areas of surface anomalies.

[0015] Calculate the intensity of surface anomalies in the identified surface anomaly areas.

[0016] Preferably, the construction of the prior knowledge base for surface remote sensing features includes: calculating and obtaining threshold values ​​for various types of surface anomaly remote sensing feature indicators, and determining whether there are outliers or discrete values ​​in the first long-term time series surface remote sensing feature data set.

[0017] max = Q3 + 1.5 * (Q3 - Q1)

[0018] min = Q1 - 1.5 * (Q3 - Q1)

[0019] Where max represents the maximum value of the threshold of the remote sensing feature index of surface anomalies, and min represents the minimum value of the threshold of the remote sensing feature index of surface anomalies. All values ​​in the first long-term remote sensing feature data set of surface anomalies are arranged from smallest to largest and divided into the first range, the second range, the third range and the fourth range. Q1 represents the largest value in the first range; Q3 represents the largest value in the third range.

[0020] Based on the threshold values ​​of various surface anomaly remote sensing feature indicators and the first long-term time series surface remote sensing feature data set, outliers are removed to obtain the second long-term time series surface remote sensing feature data set.

[0021] In the second long-term remote sensing feature data set, all values ​​are greater than or equal to the minimum value of the remote sensing feature threshold for surface anomalies, and less than or equal to the maximum value of the remote sensing feature threshold for surface anomalies.

[0022] The various remote sensing features of surface anomalies include vegetation features, water features, ice and snow features, and bare area features.

[0023] Preferably, the construction of the prior knowledge base for surface remote sensing features further includes: calculating the mean and standard deviation of surface anomaly remote sensing feature indicators based on the second long-term time series surface remote sensing feature data set.

[0024]

[0025]

[0026] Among them, f mean (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The average value of each remote sensing feature, v(x) i ,y i ,d k ,t n ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The t-th remote sensing feature n There are several values, where n represents the number of values ​​in the prior knowledge base of surface remote sensing features located at (x). i ,y i The pixel at position d is at position d. k The number of numerical values ​​in a remote sensing feature, f sd (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The standard deviation of each remote sensing feature.

[0027] Preferably, the identification of potential areas of surface anomalies includes:

[0028] Determine whether the vegetation feature indicators in the current surface remote sensing feature information meet the first criterion. If they do not meet the criterion, they are identified as potential areas of surface anomalies.

[0029] Preferably, identifying surface anomaly areas includes determining whether a potential surface anomaly area meets the second criterion; if it does, it is determined to be a surface anomaly area.

[0030] The second standard includes:

[0031] |F(x i y i d k )-f menn(x i y i d k )|>15*f sd (x i y i d k )

[0032] (F(x i y i d NDVI )>f mean (x i y i d NDVI ))&(f mean (x i y i d NDWI )<0)

[0033] (F(x i ,y i ,d NDWI )>f mean (x i y i d NDWI ))&(f mean (x i y i d NDWI )>=0)

[0034] Where, F(x) i ,y i ,d k ) indicates that the location in the remote sensing image acquired in real time is located at (x i ,y i The pixel at position d is at position d. k The numerical value of each remote sensing feature, F(x) i ,y i ,d NDVI ) indicates that the vegetation feature index data obtained from real-time acquisition of remote sensing images is located in (x i ,y i The pixel value at position F(x) i ,y i ,d NDWI ) indicates that the water body characteristic index data obtained from the inversion of real-time acquired remote sensing images is located in (x i ,y i The pixel value at position f mean (x i ,y i ,d NDVI ) represents the location (x) of the vegetation feature index in the prior knowledge base of surface remote sensing features. i ,y iThe average value of the pixels at position f mean (x i ,y i ,d NDWI ) represents the (x) in the prior knowledge base of surface remote sensing features located at the water body feature index. i ,y i The average value of the pixels at ().

[0035] Preferably, the calculation of the surface anomaly intensity of the surface anomaly region includes:

[0036] s1(x i ,y i )=(|F(x i ,y i d max )-f meann (x i y i d max )|) / f sd (x i y i d max )

[0037] Wherein, s1(x i ,y i ) represents located in (x i ,y i The surface anomaly intensity of the pixel at location ) F(x i ,y i ,d max ) represents the target image located in (x i ,y i The value of the remote sensing feature at point f that changed most significantly after the occurrence of the surface anomaly. mean (x i ,y i ,d max ) represents the prior knowledge base located at (x i ,y i f is the average value of the remote sensing features that changed most significantly after the occurrence of the surface anomaly at the pixel at location ) sd (x i ,y i ,d max ) represents the prior knowledge base located at (x i ,y i The standard deviation of the remote sensing feature at point () is the one that changes the most after the occurrence of a surface anomaly.

[0038] Preferably, calculating the surface anomaly intensity of the surface anomaly region further includes:

[0039]

[0040] Where S2 represents the intensity of surface anomalies at the patch scale, aera plaque This represents the area of ​​an isolated patch within an anomaly region on the Earth's surface, expressed in square meters.

[0041] Preferably, the calculation of the surface anomaly intensity of the surface anomaly region further includes:

[0042] S = a*s1 + b*s2

[0043] Where S is the surface anomaly intensity value, S1 represents the pixel-scale surface anomaly intensity, S2 represents the patch-scale surface anomaly intensity, and a and b are the weighting coefficients of the pixel-scale and patch-scale surface anomaly intensity, respectively.

[0044] Preferably, a real-time remote sensing system for detecting surface anomalies includes:

[0045] Acquisition module: Acquires the first long-term time series surface remote sensing feature information to form the first long-term time series surface remote sensing feature data set;

[0046] Construction module: Based on the first long-term time series of land surface remote sensing feature data set, construct a prior knowledge base of land surface remote sensing features;

[0047] Identification module: acquires real-time remote sensing data from the satellite, calculates surface remote sensing feature information, identifies potential areas of surface anomalies based on the current surface remote sensing feature information, and identifies surface anomaly areas based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features.

[0048] Calculation module: Calculates the intensity of surface anomalies in the identified surface anomaly areas.

[0049] Preferably, a real-time remote sensing early warning system for surface anomalies includes a satellite terminal and a local terminal, wherein the satellite terminal is installed on an in-orbit satellite and the local terminal is installed on the ground surface;

[0050] The satellite end uses the detection method described in any of the above items to calculate the surface anomaly detection results and surface anomaly intensity data, and transmits the surface anomaly detection results and surface anomaly intensity to the local end;

[0051] The local terminal receives the surface anomaly detection results and surface anomaly intensity data, and issues a surface anomaly warning based on the surface anomaly detection results and surface anomaly intensity data.

[0052] (III) Beneficial Effects

[0053] The above-described technical solution of the present invention has the following beneficial technical effects:

[0054] This invention utilizes long-term time-series remote sensing features to construct a prior knowledge base of surface remote sensing features, enabling effective capture of normal surface conditions. It performs two rounds of surface anomaly region identification based on surface remote sensing features calculated from real-time acquired remote sensing data and the prior knowledge base, achieving high identification accuracy. The intensity of surface anomaly regions is calculated, facilitating appropriate emergency response based on intensity, demonstrating strong adaptability and ease of promotion. This invention can serve as an on-orbit, real-time remote sensing detection technology system for surface anomalies, providing crucial support for real-time detection and early warning of surface anomalies. The entire method enables on-board storage and on-orbit computation, achieving real-time on-orbit detection of surface anomalies. Attached Figure Description

[0055] Figure 1 This is an overall flowchart of one embodiment of the present invention;

[0056] Figure 2 This is an overall concept diagram of one embodiment of the present invention;

[0057] Figure 3 This is a flowchart of the identification process according to another embodiment of the present invention;

[0058] Figure 4 This is a spatial distribution map of the surface anomaly area and the intensity of surface anomalies in an experimental area according to one embodiment of the present invention. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and the accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0060] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0061] In the description of this invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0062] Example 1

[0063] A method for real-time remote sensing detection of surface anomalies. Figure 1 The overall flow of one embodiment of the present invention is shown, as follows: Figure 1 As shown, it includes:

[0064] The first long-term time series remote sensing feature information of the land surface is acquired to form a first long-term time series remote sensing feature data set. The specific method and content of acquiring the first long-term time series remote sensing feature information are not limited here. Optionally, in this embodiment, the acquisition method includes information obtained by comprehensively observing the target area using one or more satellites such as hyperspectral, microwave, infrared, or high-resolution. The specific time for acquiring the first long-term time series remote sensing feature information is also not limited here. Optionally, remote sensing images from several years can be used as the first long-term time series remote sensing feature information. The imaging time can be selected to be as close as possible to the specific number of imaging days of the target area. Further, in this embodiment, remote sensing images of the target area from the previous four years are selected as the first long-term time series remote sensing feature information. Optionally, in this embodiment, the first long-term time series remote sensing feature information is preprocessed. The specific content of the preprocessing is not limited here; it can be cloud removal, correction, or alignment processing of the remote sensing images. Optionally, in this embodiment, the preprocessing includes screening the first long-term time series remote sensing feature information, removing the first long-term time series remote sensing feature information with a cloud content exceeding 60%.

[0065] This constitutes the first long-term time-series remote sensing feature data set of the Earth's surface. The specific content of this set is not limited; optionally, it can be classified according to remote sensing features. From the perspective of land cover type, the global Earth's surface can be roughly divided into four main land cover types: vegetation cover areas, water bodies, snow and ice, and bare areas without vegetation cover (such as bare land, impermeable surfaces, etc.). In this embodiment, the first long-term time-series remote sensing feature data set may optionally include: vegetation remote sensing features, water body remote sensing features, snow and ice remote sensing features, and bare area remote sensing features. Each land cover type has remote sensing features that can describe its characteristics in detail. When surface anomalies occur, they often cause varying degrees of damage to the original land cover type. This damage is reflected in the remote sensing features that describe the characteristics of different land covers as significant changes in the values ​​of the remote sensing features. To describe the first long-term time-series land surface remote sensing feature data set, a land surface anomaly remote sensing feature index system is constructed. The specific content of the index system is not limited here; optionally, in this embodiment, the construction of the land surface anomaly remote sensing feature index system includes:

[0066] Vegetation characteristic indicators:

[0067] Water body characteristic indicators:

[0068] Ice and snow characteristic indicators:

[0069] Characteristic indicators of exposed areas:

[0070] Wherein: NIR represents the near-infrared band, RED represents red wave, GREEN represents green wave, BLUE represents blue wave, and SWIR1 represents the short-wave infrared band.

[0071] The first long-term time series surface remote sensing feature data set is obtained by calculating the surface anomaly remote sensing feature index system based on the acquired first long-term time series surface remote sensing feature information. Here, the first long-term time series surface remote sensing feature data set is the four index sets mentioned above.

[0072] Based on the first long-term time-series land surface remote sensing feature dataset, a prior knowledge base for land surface remote sensing features is constructed. Within the land surface anomaly remote sensing feature index system, identifying areas where remote sensing features show significant changes is sufficient to determine land surface anomaly areas. Prior to this, the first long-term time-series land surface remote sensing feature dataset undergoes screening processing, and threshold values ​​for various land surface anomaly remote sensing feature indices are calculated and obtained separately.

[0073] max = Q3 + 1.5 * (Q3 - Q1)

[0074] min = Q1 - 1.5 * (Q3 - Q1)

[0075] Where max represents the maximum value of the threshold of the remote sensing feature index of surface anomalies, and min represents the minimum value of the threshold of the remote sensing feature index of surface anomalies. All values ​​in the first long-term remote sensing feature data set of surface anomalies are arranged from smallest to largest and divided into the first range, the second range, the third range and the fourth range. Q1 represents the largest value in the first range; Q3 represents the largest value in the third range.

[0076] Based on the threshold values ​​of various surface anomaly remote sensing feature indicators and the first long-term surface remote sensing feature data set, a second long-term surface remote sensing feature data set is obtained. The values ​​in the second long-term surface remote sensing feature data set satisfy the condition that they are greater than or equal to the minimum value of the threshold values ​​of the surface anomaly remote sensing feature indicators and less than or equal to the maximum value of the threshold values ​​of the surface anomaly remote sensing feature indicators. Various surface anomaly remote sensing feature indicators include vegetation feature indicators, water body feature indicators, snow and ice feature indicators, and bare area feature indicators.

[0077] Here, all values ​​in the first long-term time-series surface remote sensing feature data set are arranged from smallest to largest and divided into a first range, a second range, a third range, and a fourth range. It can be divided into four ranges equally or unequally. Optionally, in this embodiment, it is divided into four ranges equally. Q1 represents the 25th percentile number after all values ​​in the set are arranged from smallest to largest; Q3 refers to the 75th percentile number after all values ​​in the set are arranged from smallest to largest.

[0078] After outlier removal, the mean and standard deviation are calculated for each feature and each pixel. Based on the second long-term time-series surface remote sensing feature dataset, the mean and standard deviation of the surface anomaly remote sensing feature indicators are calculated respectively:

[0079]

[0080]

[0081] Among them, f mean (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The average value of each remote sensing feature, v(x) i ,y i ,d k ,t n ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The t-th remote sensing feature n There are several values, where n represents the number of values ​​in the prior knowledge base of surface remote sensing features located at (x). i ,y i The pixel at position d is at position d. k The number of numerical values ​​in a remote sensing feature, f sd (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The standard deviation of each remote sensing feature.

[0082] By integrating and storing the multi-year average and standard deviation data of these four main remote sensing features, a priori knowledge base of surface remote sensing features can be obtained, representing the normal state and reasonable fluctuation range of each remote sensing feature for each pixel. The spatial scope of the construction is not limited here; optionally, a comprehensive priori knowledge base that spatially covers the entire globe and is temporally continuous can be constructed and stored on satellite. In this way, any location requiring surface anomaly detection can achieve real-time detection of surface anomalies simply by calling this priori knowledge base.

[0083] The system acquires real-time remote sensing data from satellites, calculates surface remote sensing feature information, and identifies potential areas of surface anomalies. The spatial location of the calculation subject is not limited here; optionally, the calculation can be performed by an in-orbit satellite or a local satellite on the surface. In this embodiment, the calculation can optionally be performed by an in-orbit satellite. The acquisition method is not limited here; optional acquisition methods include information obtained through comprehensive observations using one or more satellites, such as hyperspectral, microwave, infrared, or high-resolution sensors.

[0084] Identify potential areas of surface anomalies; the specific content of these potential anomalies is not limited here. Figure 3 A flowchart of the identification process is shown, such as Figure 3 As shown, optionally, in this embodiment, according to the meaning of surface anomalies, surface anomalies often lead to varying degrees of surface damage, disrupting the relatively stable and naturally evolving equilibrium state of the surface. That is, surface anomalies represent a trend towards "deterioration." Based on real-time acquired remote sensing feature information, it is determined whether the vegetation feature indicators in the current surface remote sensing feature information meet the first criterion. If not, it is identified as a potential surface anomaly area. The specific content of the first criterion is not limited here. Optionally, in this embodiment, the first criterion is whether the NDVI value is greater than 0.4. If it is greater than 0.4, it is classified as a surface anomaly-free area; if it is less than or equal to 0.4, it is initially classified as a potential surface anomaly area. When the vegetation in a certain area is growing well, lush, and has high coverage, it completely fails to meet any of the characteristics of surface anomalies, and therefore must be a clearly surface anomaly-free area. It is set that when the NDVI is greater than 0.4, the area often has good vegetation growth, lush vegetation, and high coverage. Therefore, NDVI is obtained by inverting remote sensing images in real time, and then it is determined whether the NDVI value is greater than 0.4. If it is greater than 0.4, it is classified as an area without surface anomalies; if it is less than or equal to 0.4, it is identified as a potential area with surface anomalies.

[0085] Based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features, identifying surface anomaly areas includes determining whether the potential areas of surface anomalies meet the second criterion; if they do, they are identified as surface anomaly areas. The prior knowledge base of surface remote sensing features stores the average values ​​and standard deviations of four remote sensing features: NDVI, NDWI, NDSI, and BSI. The NDVI, NDWI, NDSI, and BSI data of the real-time acquired remote sensing images are obtained through inversion (using F(x)). i ,y i ,d k(This is represented by a symbol). Then, the pixel-by-pixel difference between the feature and the average value in the prior knowledge base of surface remote sensing features is compared. If the absolute value of the difference between one or more remote sensing features is greater than the set fixed threshold, it can be preliminarily identified as a surface anomaly. In this embodiment, the fixed threshold is set to 15 times the standard deviation, which is the optimal setting.

[0086] When there is a significant difference between the remote sensing feature values ​​of the real-time acquired remote sensing images and the remote sensing feature values ​​in the prior knowledge base of surface remote sensing features, it may be a surface anomaly or a phenomenon of land restoration or improved surface ecological quality, and it does not belong to the phenomenon of surface anomaly. Therefore, some restrictive conditions are added to eliminate the phenomenon of land restoration or improved surface ecological quality. The phenomenon of land restoration or improved surface ecological quality specifically includes: (1) the vegetation growth in non-water areas is better and the vegetation coverage is increased. That is, the NDVI value of the area with relatively small NDWI in the prior knowledge base of surface remote sensing features increases; (2) the water purity in water areas increases and the water level rises. That is, the NDWI value of the area with relatively high NDWI in the prior knowledge base of surface remote sensing features continues to rise. The second standard is expressed by the formula as follows:

[0087] |F(x i y i d k )-f mean (x i y i d k )|>15*f sd (x i y i d k )

[0088] (F(x i y i d NDVI )>f mean (x i y i d NDVI ))&(f mean (x i y i d NDWI )<0)

[0089] (F*x i y i d NDWI )>f mean (x i y i d NDWI ))&(f mean (x i y i d NDWI)>=0)

[0090] Where, F(x) i ,y i ,d k ) indicates that the location in the remote sensing image acquired in real time is located at (x i ,y i The pixel at position d is at position d. k The numerical value of each remote sensing feature, F(x) i ,y i ,d NDVI ) indicates that the NDVI data obtained from the inversion of real-time acquired remote sensing images is located in (x i ,y i The pixel value at position F(x) i ,y i ,d NDWI ) indicates that the NDWI data obtained by inversion from real-time acquired remote sensing images is located in (x i ,y i The pixel value at position f mean (x i ,y i ,d NDVI ) represents the (x) in the prior knowledge base of surface remote sensing features located in the NDVI remote sensing features. i ,y i The average value of the pixels at position f mean (x i ,y i ,d NDWI ) represents the (x) in the prior knowledge base of surface remote sensing features located in the NDWI remote sensing features. i ,y i The average value of the pixels at (). In this embodiment, setting the NDWI threshold to 0 can better distinguish between water bodies and non-water bodies, thereby promoting the effective identification of land restoration or surface ecological quality improvement phenomena.

[0091] Based on the identified surface anomaly regions, the intensity of the surface anomaly in these regions is calculated. After surface anomaly identification, the intensity of surface anomalies in different regions is further quantitatively described. At the pixel scale, surface anomaly intensity reflects the degree of difference in remote sensing features before and after the anomaly occurs. The greater the difference, the stronger the surface anomaly. Therefore, the surface anomaly intensity of a single pixel is determined by comparing the change in remote sensing features to the standard deviation of remote sensing features in the prior knowledge base of surface remote sensing features. The greater the ratio of change to standard deviation, the stronger the surface anomaly. Furthermore, the larger the distribution range or impact area of ​​a surface anomaly event, the higher its intensity is considered. Therefore, at the patch scale, surface anomaly intensity is closely related to patch area; the larger the patch area, the stronger the surface anomaly.

[0092] Pixel-scale intensity calculation includes:

[0093] s1(x i ,y i )=(|F(x i ,y i d max )-f mean (x i y i d max )|) / f sd (x i y i d max )

[0094] Wherein, s1(x i ,y i ) represents located in (x i ,y i The surface anomaly intensity of the pixel at location ) F(x i ,y i ,d max ) represents the location in the real-time acquired remote sensing image located at (x i ,y i The value of the remote sensing feature at point f that changed most significantly after the occurrence of the surface anomaly. mean (x i ,y i ,d max ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i f is the average value of the remote sensing features that changed most significantly after the occurrence of the surface anomaly at the pixel at location ) sd (x i ,y i ,d max ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The standard deviation of the remote sensing feature at point () is the one that changes the most after the occurrence of a surface anomaly.

[0095] Patch-scale intensity calculations include:

[0096]

[0097] Where S2 represents the intensity of surface anomalies at the patch scale, aera plaque This represents the area of ​​an isolated patch within an anomaly region on the Earth's surface, expressed in square meters.

[0098] After evaluating the surface anomaly intensity at both the pixel and patch scales, a weighted average is calculated to obtain the final estimated surface anomaly intensity. However, since the surface anomaly intensity values ​​at the pixel and patch scales differ significantly and are not on the same order of magnitude, a unified standardization process is applied. In this embodiment, the surface anomaly intensity values ​​at the pixel scale are mostly below 100, so these values ​​are retained. Furthermore, a complete surface anomaly event is necessarily composed of individual surface anomaly patches. Although the areas of these individual patches vary, in this embodiment, to match the patch-scale surface anomaly intensity values ​​with the pixel-scale intensity values, the patch area is divided by 250 square kilometers and then multiplied by 100. However, it is undeniable that a small portion of surface anomaly events involve extremely large areas, such as severe snowstorms, severe sandstorms, severe floods, and severe fires, where the area of ​​a single patch is very likely to exceed the set threshold (250 square kilometers). But in general, these situations are mostly low-probability events. Furthermore, in real life, people often classify such events as high-intensity surface anomaly events. Similarly, at the pixel scale, some pixels may have surface anomaly intensity values ​​exceeding 100. However, this does not affect the subsequent assessment of surface anomaly intensity; it is merely used as one piece of evidence that it is an ultra-high-intensity surface anomaly event. After unifying the surface anomaly intensities at the pixel and patch scales to the same order of magnitude as much as possible, weights are assigned to the surface anomaly intensities at both scales. The specific content of the weights is not limited here; optionally, in this embodiment, the same weights are assigned, and then the final estimated value of the surface anomaly intensity is obtained through comprehensive calculation.

[0099] Calculating the intensity of the surface anomaly region also includes:

[0100] S = a*s1 + b*s2

[0101] Wherein, S1 represents the surface anomaly intensity at the pixel scale, S2 represents the surface anomaly intensity at the patch scale, S is the surface anomaly intensity value, and a and b are the weighting coefficients for the surface anomaly intensity at the pixel and patch scales, respectively. The weighting coefficients for the surface anomaly intensity at the pixel and patch scales are not limited here; optionally, in this embodiment, the weighting coefficients for both are the same.

[0102] In this embodiment, Figure 4 The following diagram shows the spatial distribution of surface anomaly areas and surface anomaly intensities in the experimental area: Figure 4As shown, four remote sensing images were randomly selected for surface anomaly identification and accuracy verification, including: constructing a prior knowledge base for the experimental area images to obtain the average value and standard deviation of four remote sensing features: NDVI, NDWI, NDSI, and BSI; performing data preprocessing such as cloud removal on the real-time acquired remote sensing images and initially identifying potential surface anomaly areas; completing surface anomaly identification by setting reasonable thresholds; and calculating the estimated value of surface anomaly intensity by comprehensively considering pixel scale and patch scale.

[0103] like Figure 4 As shown, the real-time acquired remote sensing images and surface anomaly detection results for experimental area 1 are as follows: Figure 4 As shown in (a) and (b), it can be observed that the region experienced severe flooding. In particular, the river in the western part of the image breached its banks, causing massive floodwaters to spread to both sides and inundate large areas of land. Simultaneously, significant rainfall accumulated in the low-lying farmland areas in the eastern part of the image. Figure 4 (b) It can be observed that the surface anomaly areas identified in this study are mainly concentrated in a blocky pattern in the eastern part of the experimental area. Meanwhile, a small number of surface anomaly areas are also discretely distributed in a strip-like or scattered pattern in the western part of the experimental area. This is basically consistent with the spatial distribution pattern of the visual interpretation results, showing a high degree of similarity. Furthermore, 1000 verification points were randomly and evenly distributed throughout the entire experimental area, and accuracy evaluation was completed using relevant historical data and higher-resolution remote sensing imagery. The overall classification accuracy of the surface anomaly detection results in Experimental Area 1 is 92.20%.

[0104] like Figure 4 As shown, the real-time acquisition of remote sensing images and surface anomaly detection results for experimental area 2 are as follows: Figure 4 As shown in (d) and (e), it can be observed that forest fires occurred in the eastern and southeastern parts of the area, leaving behind large areas of burned land. Figure 4 (e) It can be observed that the surface anomaly areas are mainly concentrated in blocks in the eastern and southeastern parts of the experimental area. This is basically consistent with the spatial distribution pattern of the visual interpretation results. To conduct a quantitative evaluation, 900 verification points were randomly and evenly distributed throughout the experimental area. The overall classification accuracy of the surface anomaly detection results in Experimental Area 2 is 94.33%.

[0105] like Figure 4 As shown, the real-time acquired remote sensing images and surface anomaly detection results of experimental area 3 are as follows: Figure 4 As shown in (g) and (h), it can be observed that deforestation and destruction of ecological land have occurred in the northern and southern parts of this area. Figure 4 (h) It can be observed that the surface anomaly areas are mainly distributed in patches in scattered areas in the north and south. This is basically consistent with the spatial distribution pattern of the visual interpretation results. The overall classification accuracy of the surface anomaly detection results in Experimental Area 3 is 89.75%.

[0106] like Figure 4 As shown, the real-time acquisition of remote sensing images and surface anomaly detection results for experimental area 4 are as follows: Figure 4 As shown in (j) and (k), it can be observed that there are no surface anomalies in this area. Figure 4 (k) It can be observed that there are basically no surface anomalies in the results obtained by this invention, with only a few very small and sporadic surface anomaly areas detected in the northwest of the experimental area. The overall classification accuracy of the surface anomaly detection results in Experimental Area 4 is 97.38%.

[0107] The surface anomaly intensity results for experimental areas 1-4 are as follows: Figure 4 As shown in (c), (f), (i), and (l), the color represents the intensity of the surface anomaly. It can be observed that the high-intensity surface anomaly areas in Experimental Area 1 are mainly distributed along the original river channels in the western part of the experimental area, while the intensity of the surface anomaly areas in the eastern part is relatively lower. This is because heavy rainfall caused the river to breach its banks and rapidly rush towards both banks; the closer the area is to the original river channel, the more severe the damage, and the stronger the surface anomaly intensity. The eastern area, due to its relatively low elevation, accumulated some rainfall. However, its damage level was far lower than that of the river breach, and its surface anomaly intensity was also relatively low. In Experimental Area 2, the intensity of the surface anomaly areas in the central and eastern parts of the area is higher than that in the northern and southern parts. Analysis reveals that the key area of ​​the forest fire is in the central and eastern parts of the experimental area. This area was most affected by the fire, with almost all vegetation burned. The relatively smaller surface anomaly areas in the southern and northern parts of the area still have a certain number of surviving trees and vegetation, and were less affected by the fire. In Experimental Area 3, high-intensity surface anomalies are mainly concentrated in the largest surface anomaly patch in the central and northern parts. It can be observed that this area was previously a forest area with high vegetation cover. However, due to human deforestation, its land cover type has directly changed from forest to bare land. Because of the drastic change and relatively large area, the intensity of its surface anomalies is relatively high. Since the surface anomaly areas detected in Experimental Area 4 are very few and inconspicuous, the intensity of its surface anomalies is also relatively low.

[0108] This embodiment analyzes the main land cover types on the land surface, identifies the main remote sensing features that can effectively capture the varying degrees of damage caused by land surface anomalies, and forms a land surface anomaly remote sensing feature index system that can be applied to all types of land surface anomalies. Traditional remote sensing methods for detecting surface anomalies are mostly based on the premise that the type of surface anomaly event is known. The remote sensing feature indicators they construct tend to reflect the surface characteristics after the anomaly occurs, which means that these remote sensing features are only applicable to the selected type of surface anomaly event. This embodiment takes the surface anomaly receptor as the starting point and determines the surface anomaly by capturing the change information of the surface anomaly receptor pixel by pixel. This breaks the limitation of previous surface anomaly remote sensing detection being limited to a single type of surface anomaly and forms a more universal real-time detection method for surface anomalies. After constructing the surface anomaly remote sensing feature indicator system, this invention obtains the corresponding remote sensing features based on the inversion of multi-year surface images and calculates their average value and standard deviation to construct a priori knowledge base of the surface stable state. Since the priori knowledge base is constructed using multi-year remote sensing images and outliers are excluded, the priori knowledge base reflects the relatively stable and fluctuating equilibrium state of the surface that conforms to the laws of natural evolution. It provides an important basis for the real-time and accurate determination of subsequent surface anomalies.

[0109] Due to the complexity and diversity of surface anomalies, some surface anomalies share similar spectral, textural, and radiometric characteristics with some normal landforms. For example, landslides and debris flows are similar to bare soil areas, and barrier lakes are similar to normal water bodies. This makes it difficult for traditional threshold segmentation methods and the increasingly popular machine learning and deep learning methods to effectively distinguish between normal landforms and surface anomalies during surface anomaly detection. However, the comparative evaluation based on a prior knowledge base proposed in this embodiment can map the target image to prior knowledge, focusing more on the analysis of changes in the same area over time, thus avoiding misclassification caused by similar remote sensing characteristics to a certain extent.

[0110] Traditional change monitoring often relies on knowing that surface anomalies have occurred before searching for new and historical remote sensing images to identify the anomalous areas. This embodiment, however, can detect surface anomalies based on real-time acquired remote sensing images after establishing a prior knowledge base and decision rules. It possesses autonomous on-orbit diagnostic capabilities for surface anomalies, effectively distinguishing between anomaly and non-anomaly areas without human intervention. Furthermore, it eliminates the need to search for historical remote sensing images of the acquired data area and download the latest images and sub-satellite anomaly detection data. The surface anomaly detection algorithm is simply executed on the real-time acquired remote sensing images, directly transmitting the detection results to the local terminal, significantly reducing the surface anomaly detection cycle.

[0111] Example 2

[0112] A real-time remote sensing system for detecting surface anomalies, such as Figure 1 As shown, it includes:

[0113] Acquisition module: Acquires the first long-term time series surface remote sensing feature information to form the first long-term time series surface remote sensing feature data set;

[0114] Construction module: Based on the first long-term time series of surface remote sensing feature data set, construct a prior knowledge base of surface remote sensing features. Here, the storage location of the prior knowledge base of surface remote sensing features is not limited. Optionally, in this embodiment, the prior knowledge base of surface remote sensing features is stored on the satellite.

[0115] Identification module: acquires real-time remote sensing data from the satellite, calculates surface remote sensing feature information, identifies potential areas of surface anomalies based on the current surface remote sensing feature information, and identifies surface anomaly areas based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features.

[0116] Calculation module: Based on the identified surface anomaly areas, calculate the intensity of surface anomalies in the surface anomaly areas. There is no restriction on the spatial location of the calculation subject. Optionally, the calculation can be performed by an in-orbit satellite or by a local end on the surface. In this embodiment, the calculation can be performed by an in-orbit satellite.

[0117] In this embodiment, the prior knowledge base of surface remote sensing features is stored on-board, enabling on-orbit processing of real-time remote sensing detection of surface anomalies. It also directly calculates the intensity of surface anomalies via satellite, which is highly efficient and provides important support for real-time detection and early warning of surface anomalies.

[0118] Example 3

[0119] A real-time remote sensing early warning system for surface anomalies. Figure 2 An overall conceptual diagram of one embodiment of the present invention is shown, such as Figure 2 As shown, it includes a satellite terminal and a local terminal. The satellite terminal is located on an in-orbit satellite, and the local terminal is located on the ground. The specific content of the local terminal is not limited here. It can be a handheld terminal, a computer terminal, or a television terminal, etc. In this embodiment, a handheld terminal can be selected.

[0120] The satellite end uses the detection method described in any of the above items to calculate the surface anomaly detection results and surface anomaly intensity data, and transmits the surface anomaly detection results and surface anomaly intensity to the local end;

[0121] The local terminal receives the surface anomaly detection results and surface anomaly intensity data, and issues a surface anomaly early warning based on the surface anomaly detection results and surface anomaly intensity data. The surface anomaly detection results include at least one of the following: calculated surface remote sensing feature information, identified potential surface anomaly areas, and identified surface anomaly areas.

[0122] This embodiment takes the surface anomaly receptor as the starting point and constructs a priori knowledge base that can effectively describe the normal surface state based on the representative remote sensing characteristics of different land covers reflecting surface anomalies. By constructing the priori knowledge base and storing it on the satellite, it provides the possibility for on-orbit processing of real-time remote sensing detection of surface anomalies. The intensity of surface anomalies is directly calculated by the satellite and transmitted to the local end, which can facilitate the local end to directly issue early warnings based on the intensity of surface anomalies.

[0123] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of the invention should be included within the protection scope of the invention. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A method for real-time remote sensing detection of surface anomalies, characterized in that, include: The first long-term time series surface remote sensing feature information was obtained to form the first long-term time series surface remote sensing feature data set; Based on the first long-term time series of surface remote sensing feature data, a prior knowledge base of surface remote sensing features is constructed. Acquire real-time remote sensing data from satellites, calculate surface remote sensing feature information, and identify potential areas of surface anomalies; Based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features, identify areas of surface anomalies. Calculate the intensity of surface anomalies in the identified surface anomaly areas. The construction of the prior knowledge base for surface remote sensing features includes: calculating and obtaining threshold values ​​for various surface anomaly remote sensing feature indicators, and determining whether there are outliers or discrete values ​​in the first long-term time series surface remote sensing feature data set. Where max represents the maximum value of the threshold of the remote sensing feature index of surface anomalies, and min represents the minimum value of the threshold of the remote sensing feature index of surface anomalies. All values ​​in the first long-term remote sensing feature data set of surface anomalies are arranged from smallest to largest and divided into the first range, the second range, the third range and the fourth range. Q1 represents the largest value in the first range; Q3 represents the largest value in the third range. Based on the threshold values ​​of various surface anomaly remote sensing feature indicators and the first long-term time series surface remote sensing feature data set, remote sensing feature anomalies are removed to obtain the second long-term time series surface remote sensing feature data set. In the second long-term remote sensing feature data set, all values ​​are greater than or equal to the minimum value of the remote sensing feature threshold for surface anomalies, and less than or equal to the maximum value of the remote sensing feature threshold for surface anomalies. The various remote sensing features of surface anomalies include vegetation features, water features, ice and snow features, and bare surface features. The construction of the prior knowledge base for land surface remote sensing features also includes: calculating the mean and standard deviation of land surface anomaly remote sensing feature indicators based on the second long-term time series land surface remote sensing feature data set. Among them, f mean (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The average value of each remote sensing feature, v(x) i ,y i ,d k ,t n ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The t-th remote sensing feature n There are several values, where n represents the number of values ​​in the prior knowledge base of surface remote sensing features located at (x). i ,y i The pixel at position d is at position d. k The number of numerical values ​​in a remote sensing feature, f sd (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The standard deviation of each remote sensing feature.

2. The real-time remote sensing detection method for surface anomalies according to claim 1, characterized in that, The identification of potential areas of surface anomalies includes: The system determines whether the vegetation feature indicators in the calculated surface remote sensing feature information based on the acquired real-time satellite remote sensing data meet the first criterion. If they do not meet the criterion, the area is identified as a potential surface anomaly.

3. The real-time remote sensing detection method for surface anomalies according to claim 2, characterized in that, Identifying surface anomaly areas includes determining whether a potential surface anomaly area meets the second criterion; if it does, it is determined to be a surface anomaly area. The second standard includes: Where, F(x) i ,y i ,d k ) indicates that the location in the remote sensing image acquired in real time is located at (x i ,y i The pixel at position d is at position d. k The numerical value of each remote sensing feature, F(x) i ,y i ,d NDVI ) indicates that the vegetation feature index data obtained from real-time acquisition of remote sensing images is located in (x i ,y i The pixel value at position F(x) i ,y i ,d NDWI ) indicates that the water body feature data obtained from the inversion of real-time acquired remote sensing images is located in (x i ,y i The pixel value at position f mean (x i ,y i ,d NDVI ) represents the location (x) of the vegetation feature index in the prior knowledge base of surface remote sensing features. i ,y i The average value of the pixels at position f mean (x i ,y i ,d NDWI ) represents the (x) in the prior knowledge base of surface remote sensing features located at the water body feature index. i ,y i The average value of the pixels at ().

4. The real-time remote sensing detection method for surface anomalies according to claim 3, characterized in that: The calculation of the surface anomaly intensity in the surface anomaly region includes: Wherein, s1(x i ,y i ) represents located in (x i ,y i The surface anomaly intensity of the pixel at location ) F(x i ,y i ,d max ) represents the target image located in (x i ,y i The value of the remote sensing feature at point f that changed the most after the surface anomaly occurred. mean (x i ,y i ,d max ) represents the prior knowledge base located at (x i ,y i f is the average value of the remote sensing features that changed most significantly after the occurrence of the surface anomaly at the pixel at location ) sd (x i ,y i ,d max ) represents the prior knowledge base located at (x i ,y i The standard deviation of the remote sensing feature at point () is the one that changes the most after the occurrence of a surface anomaly.

5. The real-time remote sensing detection method for surface anomalies according to claim 4, characterized in that: The calculation of the surface anomaly intensity of the surface anomaly region further includes: Where S2 represents the intensity of surface anomalies at the patch scale, aera plaque This represents the area of ​​an isolated patch within an anomaly region on the Earth's surface, expressed in square meters.

6. The real-time remote sensing detection method for surface anomalies according to claim 5, characterized in that: The calculation of the surface anomaly intensity of the surface anomaly region further includes: Where S is the surface anomaly intensity value, S1 represents the pixel-scale surface anomaly intensity, S2 represents the patch-scale surface anomaly intensity, and a and b are the weighting coefficients of the pixel-scale and patch-scale surface anomaly intensity, respectively.

7. A real-time remote sensing system for detecting surface anomalies, characterized in that, include: Acquisition module: Acquires the first long-term time series surface remote sensing feature information to form the first long-term time series surface remote sensing feature data set; Construction module: Based on the first long-term time series of land surface remote sensing feature data set, construct a prior knowledge base of land surface remote sensing features; Identification module: acquires real-time remote sensing data from the satellite, calculates surface remote sensing feature information, and identifies potential areas of surface anomalies; based on the potential areas of surface anomalies and the prior knowledge base of surface remote sensing features, it identifies areas of surface anomalies. Calculation module: Calculates the intensity of surface anomalies in the identified surface anomaly areas; The construction of the prior knowledge base for surface remote sensing features includes: calculating and obtaining threshold values ​​for various surface anomaly remote sensing feature indicators, and determining whether there are outliers or discrete values ​​in the first long-term time series surface remote sensing feature data set. Where max represents the maximum value of the threshold of the remote sensing feature index of surface anomalies, and min represents the minimum value of the threshold of the remote sensing feature index of surface anomalies. All values ​​in the first long-term remote sensing feature data set of surface anomalies are arranged from smallest to largest and divided into the first range, the second range, the third range and the fourth range. Q1 represents the largest value in the first range; Q3 represents the largest value in the third range. Based on the threshold values ​​of various surface anomaly remote sensing feature indicators and the first long-term time series surface remote sensing feature data set, remote sensing feature anomalies are removed to obtain the second long-term time series surface remote sensing feature data set. In the second long-term remote sensing feature data set, all values ​​are greater than or equal to the minimum value of the remote sensing feature threshold for surface anomalies, and less than or equal to the maximum value of the remote sensing feature threshold for surface anomalies. The various remote sensing features of surface anomalies include vegetation features, water features, ice and snow features, and bare surface features. The construction of the prior knowledge base for land surface remote sensing features also includes: calculating the mean and standard deviation of land surface anomaly remote sensing feature indicators based on the second long-term time series land surface remote sensing feature data set. Among them, f mean (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The average value of each remote sensing feature, v(x) i ,y i ,d k ,t n ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The t-th remote sensing feature n There are several values, where n represents the number of values ​​in the prior knowledge base of surface remote sensing features located at (x). i ,y i The pixel at position d is at position d. k The number of numerical values ​​in a remote sensing feature, f sd (x i ,y i ,d k ) represents the prior knowledge base of surface remote sensing features located in (x i ,y i The pixel at position d is at position d. k The standard deviation of each remote sensing feature.

8. A real-time remote sensing early warning system for surface anomalies, characterized in that: It includes a satellite end and a local end, wherein the satellite end is located on the satellite in orbit and the local end is located on the ground. The satellite end uses the detection method described in any one of claims 1-6 to calculate the surface anomaly detection results and surface anomaly intensity data, and transmits the surface anomaly detection results and surface anomaly intensity to the local end; The local terminal receives the surface anomaly detection results and surface anomaly intensity data, and issues a surface anomaly warning based on the surface anomaly detection results and surface anomaly intensity data.