Drought monitoring intelligent early warning system and method based on remote sensing image

Abstract

The invention discloses a drought monitoring intelligent early warning system and method based on remote sensing images, and belongs to the technical field of drought monitoring. Configuring a storage space according to the number of months, performing sample slice division on the storage space according to different vegetation areas, and recording the vegetation indexes of the vegetation areas randomly collected in different months into sample slices; updating the sample slices by taking a unit year as a cycle period to form periodic sample slices; configuring the step length of the random acquisition frequency, marking the vegetation index acquired each time by a step length node, and analyzing the time effectiveness of the step length based on the vegetation index; representing the periodical form of the aging degree through a mode of describing a slice matrix, and evaluating the drought influence correlation degree between vegetation regions; with the vegetation areas as clustering centers, based on drought influence relevancy, early warning collaboration degrees among different vegetation areas are analyzed, and an early warning sequence is formed; according to the invention, the drought monitoring accuracy can be improved, and the linkage early warning of the drought condition of the multi-vegetation area can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of drought monitoring, and in particular to an intelligent drought monitoring and early warning system and method based on remote sensing images. Background Art

[0002] Remote sensing technology is a technology that detects and identifies targets by sensing electromagnetic waves, visible light, and infrared radiation reflected or radiated from a distance. In the field of drought monitoring, remote sensing technology is mainly used to monitor vegetation indices. Drought can limit plant photosynthesis, leading to a decrease in vegetation indices.

[0003] In the prior art, vegetation index methods include, but are not limited to, the Modified Soil Adjusted Vegetation Index (MSAVI), which is generally applicable to the early growth stage of vegetation seedlings; the Red Edge Chlorophyll Vegetation Index (RECI), which is generally applicable to the active growth stage of vegetation; the Normalized Difference Vegetation Index (NDVI), which is generally applicable to the most active growth stage of vegetation in the middle growth stage; and the Normalized Difference Red Edge Vegetation Index (NDRE), which is generally applicable to the mature stage of vegetation growth.

[0004] However, the types of vegetation communities are different, and the plant growth characteristics in different vegetation areas are also different, resulting in the single vegetation index analysis method being inaccurate in monitoring drought conditions, and it is difficult to conduct a coordinated analysis of drought conditions in multiple vegetation areas. Summary of the Invention

[0005] The purpose of the present invention is to provide a drought monitoring intelligent early warning system and method based on remote sensing images to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A drought monitoring intelligent early warning system based on remote sensing images, the system includes: a storage module, a data acquisition module, a slice processing module and a regional drought collaborative early warning module;

[0008] The storage module configures storage space according to the number of months, divides the storage space into sample slices according to different vegetation areas, and records the vegetation index of the vegetation areas randomly collected in different months into the sample slices; updates the sample slices with the unit year as a cycle period, and forms periodic sample slices;

[0009] The data acquisition module is used to configure the step length of the random acquisition frequency in the periodic sample slices, mark the vegetation index collected each time with a step length node, and analyze the timeliness of the step length based on the vegetation index;

[0010] The slice processing module characterizes the periodic form of timeliness by depicting a slice matrix and evaluates the correlation of drought impact between vegetation areas;

[0011] The regional drought collaborative warning module is used to analyze the warning coordination between different vegetation areas based on the correlation of drought impact, taking the vegetation area as the cluster center, and form a warning sequence.

[0012] Furthermore, the storage module includes a database unit and a data slice unit;

[0013] The database unit is used to construct a storage space in the database to store sample cluster data. The storage space is marked with a monthly time range as the unit time span of sampling. The storage space corresponding to the mark of month a is recorded as SC a , a∈[1, 12] and a is an integer;

[0014] The data slicing unit is used to uniformly number the vegetation areas, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained by remote sensing imaging technology; with the annual cycle scale, the sample slice is marked with a periodic mark and recorded as a periodic sample slice Where t represents the cycle number.

[0015] Furthermore, the data acquisition module includes an acquisition frequency configuration unit and a frequency analysis unit;

[0016] The acquisition frequency configuration unit is used to periodically sample slices In the configuration of random acquisition frequency step, the random acquisition frequency step refers to the minimum time interval length of the acquisition behavior when the vegetation index is acquired through remote sensing imaging technology, and a vegetation index is obtained at a step node; the periodic sample slices The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (V i );

[0017] The frequency analysis unit is based on periodic sample slices Analyze and calculate the step size S ta (Vi )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices contained in .

[0018] Furthermore, the slice processing module includes a slice matrix unit and a matrix analysis unit;

[0019] The slice matrix unit constructs the vegetation area V based on the step size and the timeliness of the step size i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ; If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0;

[0020] The matrix analysis unit analyzes and evaluates the correlation of drought impact between vegetation areas based on the slice matrix. The formula is: Where V j represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

[0021] Furthermore, the regional drought collaborative early warning module includes a cluster analysis unit and an early warning sequence unit;

[0022] The cluster analysis unit is used to classify vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V i The drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i ); Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects;

[0023] The warning sequence unit is used to preset the warning coordination threshold. If the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i ).

[0024] A drought monitoring and intelligent early warning method based on remote sensing images, the method comprises the following steps:

[0025] Step S1: configuring storage space according to the number of months, dividing the storage space into sample slices according to different vegetation areas, recording the vegetation index of vegetation areas randomly collected in different months into the sample slices; updating the sample slices with the unit year as the cycle period, and forming periodic sample slices;

[0026] Step S2: In the periodic sample slice, configure the step length of the random sampling frequency, mark the vegetation index collected each time with a step length node, and analyze the timeliness of the step length based on the vegetation index;

[0027] Step S3: Characterize the periodic form of timeliness by depicting the slice matrix and evaluate the correlation of drought impact between vegetation areas;

[0028] Step S4: Taking the vegetation area as the cluster center, based on the drought impact correlation, analyze the warning coordination between different vegetation areas and form a warning sequence.

[0029] Furthermore, the specific implementation process of step S1 includes:

[0030] A storage space is constructed in the database to store sample cluster data. The storage space is marked with a monthly time range as the unit time span of sampling. The storage space corresponding to the month is recorded as SC a , a∈[1, 12] and a is an integer;

[0031] The vegetation areas are uniformly numbered, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained by remote sensing imaging technology;

[0032] Taking the year as the cycle scale, the sample slice is marked with a periodic mark and recorded as a periodic sample slice. Where t represents the cycle number;

[0033] According to the above method, vegetation indices are collected in different vegetation areas on a monthly basis, which makes the vegetation index collection method more flexible and applicable to the plant growth characteristics of different vegetation community types. For example, rice is mainly planted in a vegetation area, and the growth characteristics of rice in different months include development and maturity. In the development stage, the modified soil adjusted vegetation index method (MSAVI) can be used to collect vegetation indices, and in the maturity stage, the normalized difference red edge vegetation index method (NDRE) can be used to collect vegetation indices.

[0034] Furthermore, the specific implementation process of step S2 includes:

[0035] In periodic sample slicing In the configuration of random acquisition frequency step, the random acquisition frequency step refers to the minimum time interval length of the acquisition behavior when the vegetation index is acquired through remote sensing imaging technology, and a vegetation index is obtained at a step node; the periodic sample slices The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (Vi );

[0036] Based on periodic sample slicing Analyze and calculate the step size S ta (V i )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices included in ;

[0037] According to the above method, the flexibility of the present invention in collecting vegetation indices is also reflected in that data is collected by vegetation area and month to generate periodic sample slices, and the timeliness is only based on the same vegetation area and the same monthly cycle for analysis, avoiding the problem of data asymmetry caused by using different vegetation index analysis methods; at the same time, the size of the acquisition frequency step directly affects the flexibility and accuracy of the frequency adjustment. The smaller the step size, the more refined the frequency adjustment can be, which can better adapt to changes in monitoring needs. As the drought develops and monitoring needs change, the timeliness of the step size is used to indirectly reflect the drought situation in the vegetation area, so as to be applied to the changing characteristics of drought monitoring within a specific month.

[0038] Furthermore, the specific implementation process of step S3 includes:

[0039] Based on the step length and the timeliness of the step length, construct the vegetation area V i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ;

[0040] If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0;

[0041] Based on the slice matrix, the correlation of drought impact between vegetation areas is analyzed and evaluated using the formula: Where Vj represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

[0042] Furthermore, the specific implementation process of step S4 includes:

[0043] Vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V i The drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i );

[0044] Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects;

[0045] Preset warning coordination threshold, if the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i );

[0046] According to the above method, the impact range of drought often covers multiple vegetation areas. However, due to the differences in plant growth characteristics of different vegetation community types, the use of remote sensing technology has monitoring errors, and the feedback to plant growth characteristics also reflects the errors in the impact of drought. The present invention uses early warning collaborative analysis between vegetation areas to make the drought early warning effect more in line with plant growth characteristics, thereby making the early warning behavior more effective and the monitoring data more accurate.

[0047] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: in a drought monitoring intelligent early warning system and method based on remote sensing images provided by the present invention, storage space is configured according to the number of months, the storage space is divided into sample slices according to different vegetation areas, and the vegetation index of the vegetation areas randomly collected in different months is recorded in the sample slices; the sample slices are updated with a unit year as a cycle period to form periodic sample slices; the step length of the random collection frequency is configured, the vegetation index of each collection is marked with a step length node, and the timeliness of the step length is analyzed based on the vegetation index; the periodic form of the timeliness is characterized by depicting a slice matrix, and the correlation of drought impacts between vegetation areas is evaluated; with the vegetation area as the cluster center, the early warning coordination between different vegetation areas is analyzed based on the drought impact correlation to form an early warning sequence; the present invention can improve the accuracy of drought monitoring while realizing linked early warning of drought conditions in multiple vegetation areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0049] Figure 1 This is a structural diagram of an intelligent early warning system for drought monitoring based on remote sensing images according to the present invention;

[0050] Figure 2 This is a schematic diagram of the steps of an intelligent drought monitoring and early warning method based on remote sensing images of the present invention. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0052] See also Figure 1 In the first embodiment of the present invention, a drought monitoring intelligent early warning system based on remote sensing images is provided. The system includes a storage module, a data acquisition module, a slice processing module and a regional drought collaborative early warning module.

[0053] The storage module configures storage space according to the number of months, divides the storage space into sample slices according to different vegetation areas, and records the vegetation index of vegetation areas randomly collected in different months into the sample slices; updates the sample slices with the unit year as the cycle period, and forms periodic sample slices;

[0054] Preferably, the storage module includes a database unit and a data slicing unit;

[0055] The database unit is used to build a storage space in the database to store sample cluster data. The storage space is marked with a monthly time range as the unit time span of sampling. The storage space corresponding to the mark of month a is recorded as SC a , a∈[1, 12] and a is an integer;

[0056] Data slice unit is used to uniformly number vegetation areas, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained through remote sensing imaging technology; with the annual cycle scale, the sample slices are marked with a periodic mark and recorded as periodic sample slices Where t represents the cycle number.

[0057] The data acquisition module is used to configure the step length of random acquisition frequency in periodic sample slices, mark the vegetation index of each acquisition with a step length node, and analyze the timeliness of the step length based on the vegetation index;

[0058] Preferably, the data acquisition module includes an acquisition frequency configuration unit and a frequency analysis unit;

[0059] The acquisition frequency configuration unit is used to slice samples at periodic intervals. In the configuration of random acquisition frequency step, the random acquisition frequency step refers to the minimum time interval length of the acquisition behavior when the remote sensing image technology is used to collect and obtain the vegetation index, and a vegetation index is obtained at a step node; the periodic sample slices The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (V i );

[0060] Frequency analysis unit, based on periodic sample slices Analyze and calculate the step size S ta (V i )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices contained in .

[0061] The slice processing module characterizes the periodic form of timeliness by depicting the slice matrix and evaluates the correlation of drought impact between vegetation areas;

[0062] Preferably, the slice processing module includes a slice matrix unit and a matrix analysis unit;

[0063] Slice the matrix unit and construct the vegetation area V based on the step size and the timeliness of the step size i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ; If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0;

[0064] The matrix analysis unit analyzes and evaluates the correlation of drought impact between vegetation areas based on the slice matrix. The formula is: Where V j represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i(t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

[0065] The regional drought collaborative early warning module is used to analyze the early warning coordination between different vegetation areas based on the correlation of drought impact, taking the vegetation area as the cluster center, and form an early warning sequence;

[0066] Preferably, the regional drought collaborative early warning module includes a cluster analysis unit and an early warning sequence unit;

[0067] Cluster analysis unit, used to classify vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V i The drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i ); Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects;

[0068] The warning sequence unit is used to preset the warning coordination threshold. If the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i ).

[0069] See also Figure 2 In the second embodiment, a drought monitoring and intelligent early warning method based on remote sensing images is provided, which includes the following steps:

[0070] Step S1: configuring storage space according to the number of months, dividing the storage space into sample slices according to different vegetation areas, recording the vegetation index of vegetation areas randomly collected in different months into the sample slices; updating the sample slices with the unit year as the cycle period, and forming periodic sample slices;

[0071] For example, a storage space is constructed in the database to store sample cluster data. The storage space is marked with a time range of a month as the unit time span of sampling. The storage space corresponding to the mark of month a is recorded as SC a , a∈[1, 12] and a is an integer;

[0072] The vegetation areas are uniformly numbered, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained through remote sensing imaging technology;

[0073] With the year as the cycle scale, the sample slices are marked with a periodic mark and recorded as periodic sample slices. Where t represents the cycle number.

[0074] Step S2: In the periodic sample slice, configure the step length of the random sampling frequency, mark the vegetation index of each collection with a step length node, and analyze the timeliness of the step length based on the vegetation index;

[0075] For example, in periodic sample slices In the configuration of random acquisition frequency step, the random acquisition frequency step refers to the minimum time interval length of the acquisition behavior when the remote sensing image technology is used to collect and obtain the vegetation index, and a vegetation index is obtained at a step node; the periodic sample slices The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (V i );

[0076] Based on periodic sample slicing Analyze and calculate the step size S ta (V i )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices contained in .

[0077] Step S3: Characterize the periodic form of timeliness by depicting the slice matrix and evaluate the correlation of drought impact between vegetation areas;

[0078] For example, based on the step length and the timeliness of the step length, the vegetation area V is constructed. i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ;

[0079] If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0;

[0080] Based on the slice matrix, the correlation of drought impact between vegetation areas is analyzed and evaluated using the formula: Where V j represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

[0081] Step S4: Taking the vegetation area as the cluster center, based on the drought impact correlation, analyze the warning coordination between different vegetation areas and form a warning sequence;

[0082] For example, the vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V iThe drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i );

[0083] Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects;

[0084] Preset warning coordination threshold, if the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i ).

[0085] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0086] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A drought monitoring intelligent early warning method based on remote sensing images, characterized in that: The method comprises the following steps: Step S1: configuring storage space according to the number of months, dividing the storage space into sample slices according to different vegetation areas, recording the vegetation index of vegetation areas randomly collected in different months into the sample slices; updating the sample slices with the unit year as the cycle period, and forming periodic sample slices; Step S2: In the periodic sample slice, configure the step length of the random sampling frequency, mark the vegetation index collected each time with a step length node, and analyze the timeliness of the step length based on the vegetation index; Step S3: Characterize the periodic form of timeliness by depicting the slice matrix and evaluate the correlation of drought impact between vegetation areas; Step S4: Taking the vegetation area as the cluster center, based on the drought impact correlation, analyze the warning coordination between different vegetation areas and form a warning sequence.

2. The method for drought monitoring and intelligent early warning based on remote sensing images according to claim 1, characterized in that: The specific implementation process of step S1 includes: A storage space is constructed in the database to store sample cluster data. The storage space is marked with a monthly time range as the unit time span of sampling. The storage space corresponding to the month is recorded as SC a , a∈[1, 12] and a is an integer; The vegetation areas are uniformly numbered, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained by remote sensing imaging technology; Taking the year as the cycle scale, the sample slice is marked with a periodic mark and recorded as a periodic sample slice. Where t represents the cycle number.

3. The method for drought monitoring and intelligent early warning based on remote sensing images according to claim 2, characterized in that: The specific implementation process of step S2 includes: In periodic sample slicing In the configuration of random acquisition frequency step, the random acquisition frequency step refers to the minimum time interval length of the acquisition behavior when the vegetation index is acquired through remote sensing imaging technology, and a vegetation index is obtained at a step node; the periodic sample slices The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (V i ); Based on periodic sample slicing Analyze and calculate the step size S ta (V i )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices contained in .

4. The method for drought monitoring and intelligent early warning based on remote sensing images according to claim 3 is characterized in that: The specific implementation process of step S3 includes: Based on the step length and the timeliness of the step length, construct the vegetation area V i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ; If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0; Based on the slice matrix, the correlation of drought impact between vegetation areas is analyzed and evaluated using the formula: Where V j represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

5. The method for drought monitoring and intelligent early warning based on remote sensing images according to claim 4 is characterized in that: The specific implementation process of step S4 includes: Vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V i The drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i ); Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects; Preset warning coordination threshold, if the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i ).

6. A drought monitoring intelligent early warning system based on remote sensing images, characterized in that: The system includes: a storage module, a data acquisition module, a slice processing module and a regional drought collaborative early warning module; The storage module configures storage space according to the number of months, divides the storage space into sample slices according to different vegetation areas, and records the vegetation index of the vegetation areas randomly collected in different months into the sample slices; updates the sample slices with the unit year as a cycle period, and forms periodic sample slices; The data acquisition module is used to configure the step length of the random acquisition frequency in the periodic sample slices, mark the vegetation index collected each time with a step length node, and analyze the timeliness of the step length based on the vegetation index; The slice processing module characterizes the periodic form of timeliness by depicting a slice matrix and evaluates the correlation of drought impact between vegetation areas; The regional drought collaborative warning module is used to analyze the warning coordination between different vegetation areas based on the correlation of drought impact, taking the vegetation area as the cluster center, and form a warning sequence.

7. The drought monitoring intelligent early warning system based on remote sensing images according to claim 6, characterized in that: The storage module includes a database unit and a data slice unit; The database unit is used to construct a storage space in the database to store sample cluster data. The storage space is marked with a monthly time range as the unit time span of sampling. The storage space corresponding to the mark of month a is recorded as SC a , a∈[1, 12] and a is an integer; The data slicing unit is used to uniformly number the vegetation areas, and the i-th vegetation area is recorded as V i , based on the number of vegetation areas, the storage space SC a Divide the sample slices and divide the vegetation area V i The corresponding sample slice is recorded as SC a (V i ), randomly sample vegetation area V within a month i A set of several vegetation indices and recorded in the sample slice SC a (V i ), the vegetation index is collected and obtained by remote sensing imaging technology; with the annual cycle scale, the sample slice is marked with a periodic mark and recorded as a periodic sample slice Where t represents the cycle number.

8. The drought monitoring intelligent early warning system based on remote sensing images according to claim 7 is characterized by: The data acquisition module includes an acquisition frequency configuration unit and a frequency analysis unit; The acquisition frequency configuration unit is used to periodically sample slices In [1], configure the step length of the random acquisition frequency. The step length of the random acquisition frequency refers to the minimum time interval length of the acquisition behavior when acquiring vegetation indices through remote sensing imaging technology, and a vegetation index is obtained at each step length node. Slice the periodic sample The step length of the random sampling frequency corresponding to the configuration is recorded as S ta (V i ); The frequency analysis unit is based on periodic sample slices Analyze and calculate the step size S ta (V i )’s timeliness, the formula is Where, Indicates that the step size S ta The vegetation area V corresponding to the b-th step node under the acquisition frequency i The vegetation index, Represents periodic sample slices The total number of vegetation indices included in μ is the periodic sample slice. The average value of the vegetation indices contained in .

9. The drought monitoring intelligent early warning system based on remote sensing images according to claim 8, characterized in that: The slice processing module includes a slice matrix unit and a matrix analysis unit; The slice matrix unit constructs the vegetation area V based on the step size and the timeliness of the step size i The slice matrix is denoted as R i (t×a), the row number of the slice matrix corresponds to the period number, and the column number of the slice matrix corresponds to the month number, then the matrix element corresponding to the tth row and the ath column of the slice matrix is r ta ; If the step length S ta (V i ) is greater than or equal to the timeliness threshold, then let r ta =S ta (V i )=1, otherwise, let r ta =S ta (V i )=0; The matrix analysis unit analyzes and evaluates the correlation of drought impact between vegetation areas based on the slice matrix. The formula is: Where V j represents the jth vegetation area, R j (t×a) represents the vegetation area V j Slice matrix, NUM[R i (t×a)∩R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of 1s contained in the result of the Boolean matrix intersection operation between (t×a), NUM[R i (t×a)∪R j (t×a)] represents the slice matrix R i (t×a) and the slice matrix R j The number of ones contained in the result of the Boolean matrix union between (t×a).

10. The drought monitoring intelligent early warning system based on remote sensing images according to claim 9, characterized in that: The regional drought collaborative early warning module includes a cluster analysis unit and an early warning sequence unit; The cluster analysis unit is used to classify vegetation area V i As the cluster center, let j = j + 1, iteratively calculate the vegetation area V i Each vegetation area outside the vegetation area V i The drought impact correlation between them forms a set of drought-related regions, which is recorded as DR(V i ); Based on the drought-related regional set DR(V i ), analyze and calculate the vegetation area V i With vegetation area V j The early warning coordination degree between Where, represents the set of drought-related regions DR(V i ) contains the average value of the drought impact correlation, σ 2 represents the set of drought-related regions DR(V i ) contains the variance of the correlation of drought effects; The warning sequence unit is used to preset the warning coordination threshold. If the warning coordination degree EC(V i →V j ) is greater than or equal to the warning coordination threshold, the vegetation area V j Recorded in the collaborative warning sequence dr(V i ) in the drought-related area set DR(V i After each drought impact correlation degree in the ) participates in the analysis of the early warning coordination degree, the coordinated early warning sequence dr(V i ).

Images