Farmland fire and straw burning monitoring method and system based on multi-source remote sensing data

Abstract

The present invention provides a method and system for monitoring farmland fire and straw burning based on multi-source remote sensing data, and relates to the technical field of fire point monitoring. The method comprises the following steps: obtaining and preprocessing remote sensing data to obtain temperature anomaly data, screening the temperature anomaly data to obtain farmland fire points; using an image recognition algorithm to identify and obtain farmland fire points, obtaining fire point images according to the farmland fire points, identifying the contours of the fire point images and performing denoising processing to obtain initial fire trace contours; judging whether the initial fire trace contour is a regular contour; if not, obtaining edge data of the initial fire trace contour and screening to obtain a target fire trace contour to monitor the regional scope of farmland fire / straw burning; the present application obtains farmland fire points by screening the acquired temperature anomaly data and identifies the initial fire trace contour, and screening the initial fire trace contour to obtain a target fire trace contour that meets the calculation accuracy requirements to monitor the regional scope of farmland fire / straw burning, thereby improving the calculation accuracy of the burning area.

Description

Technical Field

[0001] The present invention relates to the technical field of fire point monitoring, and in particular to a method and system for monitoring farmland fire and straw burning based on multi-source remote sensing data. Background Art

[0002] In order to improve the accuracy of straw burning area monitoring and accurately understand the actual situation of straw burning, it is necessary to accurately identify the straw burning area. In the prior art, the original edge contour of the burning area is generally directly obtained to calculate the burning area, and the burning area can generally be obtained by pixel points and resolution calculation. Since the shape of the original edge contour is mostly irregular, the original edge contour is not conducive to the accurate picking of pixel points, thus affecting the calculation accuracy of the burning area. Summary of the invention

[0003] Based on this, the purpose of the present invention is to provide a method and system for monitoring farmland fires and straw burning based on multi-source remote sensing data, which is used to solve the technical problem in the prior art that irregular original edge contours affect the calculation accuracy of the burning area.

[0004] The present invention provides a method for monitoring farmland fire and straw burning based on multi-source remote sensing data, comprising:

[0005] Acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data, wherein the temperature anomaly data includes real fire points and false fire points, wherein the false fire points include factory smoke exhaust points, and filter the temperature anomaly data to obtain real fire points, wherein the real fire points include farmland fire points and forest fire points;

[0006] Using a machine learning image recognition algorithm to identify real fire spots to classify land types and obtain farmland fire spots, obtaining fire spot images based on the farmland fire spots, each fire spot image corresponding to a farmland fire spot, identifying the outline of the fire spot image and performing denoising processing to obtain an initial fire trace outline;

[0007] Determining whether the initial fire trace contour is a regular contour, wherein the regular contour includes a rectangular contour and a circular contour;

[0008] If not, then obtaining edge data of the initial fire trace contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning;

[0009] The steps of obtaining edge data of the initial fire trace contour and screening the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning include:

[0010] Obtain edge data of the initial fire trace contour, wherein the initial fire trace contour is an irregular contour, obtain an edge contour inflection point according to the edge data, sequentially connect two adjacent edge contour inflection points to obtain a plurality of line segments, and construct a polygonal contour by the plurality of line segments to optimize the initial fire trace contour;

[0011] Get the line segment length of the polygon outline;

[0012] Judging whether there is a line segment greater than a preset length value according to the line segment length;

[0013] If there is a line segment whose length is greater than the preset value, obtain the corresponding line segment whose length is greater than the preset value as the optimizable line segment, obtain the edge data corresponding to the optimizable line segment to obtain the secondary edge contour inflection point of the edge data corresponding to the optimizable line segment;

[0014] Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold;

[0015] If it is greater than the spacing threshold, the inflection points of the secondary edge contour are connected to the two end points of the optimizable line segment to construct two secondary line segments. The two secondary line segments replace the corresponding optimizable line segments to update the polygon contour, and return to execute the step of obtaining the line segment length of the polygon contour until the length of each line segment is no greater than the preset length value, and the target fire trace contour is obtained.

[0016] The above-mentioned farmland fire and straw burning monitoring method based on multi-source remote sensing data, through screening the acquired temperature anomaly data to obtain farmland fire points, identifies the initial fire trace contour of the farmland fire points, if the initial fire trace contour is an irregular contour, then screens the initial fire trace contour to obtain a target fire trace contour that meets the calculation accuracy requirements to monitor the regional range of farmland fire / straw burning; specifically, obtains the polygonal contour of the farmland fire point, and determines whether the line segment length of the contour is greater than the preset length value to improve the contour construction accuracy, when the line segment length is greater than the preset length value, then defines the line segment as an optimizable line segment and determines whether the spacing value between the optimizable line segment and the inflection point of the secondary edge contour is greater than the spacing threshold to determine whether the optimizable line segment needs to be optimized, when the spacing value is greater than the spacing threshold, then connects the inflection point of the secondary edge contour to the two end points of the optimizable line segment to construct two secondary line segments, the two secondary line segments replace the corresponding optimizable line segments to update the polygonal contour, thereby obtaining the target fire trace contour, and improving the calculation accuracy of the burning area.

[0017] In addition, the farmland fire and straw burning monitoring method based on multi-source remote sensing data according to the present invention may also have the following additional technical features:

[0018] Furthermore, in the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes removing cloud interference from the remote sensing data. In the cloud removal method, data that meets the cloud screening logical relationship can be determined as cloud. The cloud screening logical relationship is:

[0019] P 1>0.2 and P 26 >0.02 and Td 20 <320;

[0020] In the formula, P 1 represents the reflectivity of band 1 in MODIS data; P 26 Represents the reflectivity of 26 bands in MODIS data; Td 20 Represents the brightness temperature value of the 20 bands in the MODIS data.

[0021] Furthermore, in the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes eliminating water interference from the remote sensing data. In the water body exclusion method, data that meets the water body screening logic relationship can be determined as water body. The water body screening logic relationship is:

[0022] P 2<0.15 and P 7<0.05 and (( P 2- P 1) / ( P 2+ P 1))<0;

[0023] In the formula, P 1 represents the reflectivity of band 1 in MODIS data; P 2 represents the reflectivity of band 2 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data.

[0024] Furthermore, in the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes eliminating smoke interference from the remote sensing data, and the smoke screening logical relationship in the smoke elimination method is:

[0025] ;

[0026] In the formula, P 3 represents the reflectivity of band 3 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data; P8 represents the reflectivity of band 8 in MODIS data; P 9 represents the reflectivity of band 9 in MODIS data.

[0027] Furthermore, in the step of screening the temperature anomaly data to obtain the real fire point, in the algorithm for determining the real fire point, the real fire point screening logic relationship is:

[0028] ;

[0029] In the formula, Td 20 Represents the brightness temperature value of the 20 bands in MODIS data; Td 31 Represents the brightness temperature value of band 31 in MODIS data; Td 32 Represents the brightness temperature value of 32 bands in MODIS data; Ts Indicates low temperature value, unit is K.

[0030] Another aspect of the present invention provides a farmland fire and straw burning monitoring system based on multi-source remote sensing data, the system comprising:

[0031] An acquisition module is used to acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data, wherein the temperature anomaly data includes real fire points and false fire points, wherein the false fire points include factory smoke exhaust points, and the temperature anomaly data is screened to obtain real fire points, wherein the real fire points include farmland fire points and forest fire points;

[0032] A denoising module is used to identify real fire points using a machine learning image recognition algorithm to classify land types and obtain farmland fire points, obtain fire point images based on the farmland fire points, each fire point image corresponds to a farmland fire point, identify the outline of the fire point image and perform denoising processing to obtain an initial fire trace outline;

[0033] A judging module, used for judging whether the initial fire trace contour is a regular contour, wherein the regular contour includes a rectangular contour and a circular contour;

[0034] The first execution module is used for obtaining edge data of the initial fire trace contour if the initial fire trace contour is an irregular contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning;

[0035] The steps of obtaining edge data of the initial fire trace contour and screening the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning include:

[0036] Obtain edge data of the initial fire trace contour, wherein the initial fire trace contour is an irregular contour, obtain an edge contour inflection point according to the edge data, sequentially connect two adjacent edge contour inflection points to obtain a plurality of line segments, and construct a polygonal contour by the plurality of line segments to optimize the initial fire trace contour;

[0037] Get the line segment length of the polygon outline;

[0038] Judging whether there is a line segment greater than a preset length value according to the line segment length;

[0039] If there is a line segment whose length is greater than the preset value, obtain the corresponding line segment whose length is greater than the preset value as an optimizable line segment, obtain the edge data corresponding to the optimizable line segment to obtain the secondary edge contour inflection point of the edge data corresponding to the optimizable line segment;

[0040] Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold;

[0041] If it is greater than the spacing threshold, the inflection points of the secondary edge contour are connected to the two end points of the optimizable line segment to construct two secondary line segments. The two secondary line segments replace the corresponding optimizable line segments to update the polygon contour, and return to execute the step of obtaining the line segment length of the polygon contour until the length of each line segment is no greater than the preset length value, and the target fire trace contour is obtained.

[0042] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned farmland fire and straw burning monitoring method based on multi-source remote sensing data.

[0043] On the other hand, the present invention also provides a data processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the farmland fire and straw burning monitoring method based on multi-source remote sensing data as described above is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flow chart of the farmland fire and straw burning monitoring method based on multi-source remote sensing data of the present invention;

[0045] Figure 2 A flowchart of the specific steps in the screening method of the present invention;

[0046] Figure 3 The system block diagram of the farmland fire and straw burning monitoring system based on multi-source remote sensing data of the present invention.

[0047] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0048] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0050] In order to solve the technical problem in the prior art that the irregular original edge contour affects the calculation accuracy of the burning area, the present application provides a farmland fire and straw burning monitoring method and system based on multi-source remote sensing data, by screening the acquired temperature anomaly data to obtain the farmland fire point, identifying the initial fire trace contour of the farmland fire point, if the initial fire trace contour is an irregular contour, then the initial fire trace contour is screened to obtain a target fire trace contour that meets the calculation accuracy requirements to monitor the regional range of the farmland fire / straw burning; specifically, the polygonal contour of the farmland fire point is obtained, and it is determined whether the line segment length of the contour is greater than a preset length value to improve the contour construction accuracy, when the line segment length is greater than the preset length value, then the line segment is defined as an optimizable line segment and it is determined whether the spacing value between the optimizable line segment and the inflection point of the secondary edge contour is greater than the spacing threshold to determine whether the optimizable line segment needs to be optimized, when the spacing value is greater than the spacing threshold, then the inflection point of the secondary edge contour is respectively connected to the two end points of the optimizable line segment to construct two secondary line segments, the two secondary line segments replace the corresponding optimizable line segment to update the polygonal contour, thereby obtaining the target fire trace contour, and improving the calculation accuracy of the burning area.

[0051] In order to facilitate understanding of the present invention, several embodiments of the present invention are provided below. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0052] Embodiment 1

[0053] See also Figure 1 , which shows a method for monitoring farmland fire and straw burning based on multi-source remote sensing data in a first embodiment of the present invention, and the method includes steps S101 to S105:

[0054] S101. Acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data. The temperature anomaly data includes real fire points and false fire points. False fire points include factory smoke exhaust points. Filter the temperature anomaly data to obtain real fire points. Real fire points include farmland fire points and forest fire points.

[0055] As a specific example, MODIS data can be obtained by obtaining remote sensing data through MODIS. MODIS data has the characteristics of wide band range, high monitoring frequency, and high temporal and spatial resolution. It can not only accurately reflect the location and area of ​​​​the fire point, but also continuously track and monitor the fire, and can timely feedback dynamic fire point location information, thereby quickly and accurately monitoring the straw burning area.

[0056] The monitoring of fire points is mainly affected by cloud layer and temperature anomaly data. The band characteristics of cloud layer are related to the water content above the study area. Cloud layer is highly reflective in band 1, and the reflectivity value of high cloud layer in band 26 is large. Therefore, the interference of cloud layer can be eliminated by the reflectivity of band 1 and band 26. Specifically, in the cloud layer elimination method, data that meets the cloud layer screening logic relationship can be determined as cloud layer. The cloud layer screening logic relationship is:

[0057] P 1>0.2 and P 26 >0.02 and Td 20 <320;

[0058] In the formula, P 1 represents the reflectivity of band 1 in MODIS data; P 26 Represents the reflectivity of 26 bands in MODIS data; Td 20 Represents the brightness temperature value of the 20 bands in the MODIS data.

[0059] Another interference factor for fire point monitoring is water interference. Therefore, it is necessary to exclude water interference in remote sensing data. In the water body exclusion method, data that meets the water body screening logic relationship can be determined as water body. The water body screening logic relationship is:

[0060] P 2<0.15 and P 7<0.05 and (( P 2- P 1) / ( P 2+ P 1))<0; where, P 1 represents the reflectivity of band 1 in MODIS data; P2 represents the reflectivity of band 2 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data.

[0061] Accurate identification of smoke is an important step in identifying smoldering fire points. Therefore, the preprocessing method also includes eliminating smoke interference from remote sensing data. The following relationship is the smoke screening logic relationship. The pixels that meet the following smoke screening logic relationship can be considered as smoke pixels. The smoke screening logic relationship in the smoke elimination method is:

[0062] ;

[0063] In the formula, P 3 represents the reflectivity of band 3 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data; P 8 represents the reflectivity of band 8 in MODIS data; P 9 represents the reflectivity of band 9 in MODIS data.

[0064] Among the extracted fire points, there will inevitably be some high-temperature anomalies that are not fire points. In MODIS data, the mid-infrared band 20 and the thermal infrared band 31 are more sensitive to temperature changes. Among them, the 20 band is the most sensitive to fire point information, and tiny fire point information will be perceived through changes in radiation values. Based on this feature, the brightness temperature difference between the two channels can be used to quickly extract high-temperature anomalies, that is, temperature anomaly data. The difference between temperature anomaly data in bands 31 and 32 is relatively large. Therefore, the difference between the two can effectively eliminate the impact of temperature anomaly data. Td 31 - Td 32 <8K can eliminate abnormal temperature data. Therefore, in the algorithm for determining the true fire point, the true fire point screening logic relationship is:

[0065] ;

[0066] In the formula, Td 20 Represents the brightness temperature value of the 20 bands in MODIS data; Td 31 Represents the brightness temperature value of band 31 in MODIS data; Td 32 Represents the brightness temperature value of 32 bands in MODIS data; Ts Indicates low temperature value, unit is K.

[0067] S102. Use a machine learning image recognition algorithm to identify real fire spots to classify land types and obtain farmland fire spots. Obtain fire point images based on the farmland fire points. Each fire point image corresponds to a farmland fire point. Identify the outline of the fire point image and perform denoising to obtain the initial fire trace outline.

[0068] As a specific example, real fire points can occur in different land types, including agricultural land, which includes farmland and forest land; the fire points corresponding to agricultural land are farmland fire points, and the fire points corresponding to forest land are forest land fire points. However, this application needs to locate farmland fire points, so it is necessary to eliminate forest land fire points according to the land type and filter out farmland fire points. Furthermore, in this embodiment, the denoising process includes closing the regional contour of the fire point according to the fire point image to obtain the initial fire trace contour.

[0069] S103: Determine whether the initial fire trace contour is a regular contour.

[0070] Different fire trace contours directly affect the computer's calculation efficiency of the fire point area. For example, compared with regular contours, the more complex the fire trace contour, the more work the computer needs to do when obtaining effective pixels. Therefore, the requirements for the computer are higher and the efficiency is also affected. On the other hand, it also directly affects the system's calculation accuracy of the fire point area. Therefore, it is necessary to screen the fire point contour before calculating the fire point area to improve the system's calculation efficiency and accuracy of the fire point area. As a specific example, regular contours include rectangular contours and circular contours.

[0071] If the initial fire trace contour is a regular contour, execute step S104;

[0072] If the initial fire trace contour is an irregular contour, step S105 is executed;

[0073] S104. Monitor the area scope of farmland fire / straw burning based on the initial fire trace outline.

[0074] S105, obtaining edge data of the initial fire trace contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning.

[0075] For details, see Figure 2 In this embodiment, step S105 specifically includes steps S1051 to S1057:

[0076] S1051. Obtain edge data of an initial fire trace contour, which is an irregular contour. Obtain an inflection point of the edge contour based on the edge data, sequentially connect two adjacent inflection points of the edge contour to obtain multiple line segments, and construct a polygonal contour using the multiple line segments to optimize the initial fire trace contour.

[0077] When the initial fire trace contour is an irregular contour, it means that the initial fire trace contour at this time is a polygonal contour. Due to the characteristics of farmland fire points during burning, the fire trace contour is generally an arc contour. Therefore, in order to improve the system's calculation efficiency and accuracy of the farmland fire point area, the initial fire trace contour needs to be optimized to obtain a contour that is as regular as possible.

[0078] S1052. Obtain the line segment length of the polygon outline.

[0079] As a specific example, the arc contour is used as the edge data of the farmland fire point. Each line segment in the polygonal contour is obtained by connecting the corresponding inflection points of two adjacent arc contours. Therefore, in the arc contour with a larger span, in order to avoid the situation where a large area is optimized when the arc contour with a larger span is optimized through the line segments obtained by two inflection points, that is, the elimination area is large, it is necessary to make a secondary judgment on the line segment and the middle area corresponding to the line segment and the arc contour (that is, the distance between the line segment and the arc) to avoid a large elimination area, so as to improve the calculation accuracy of the farmland fire point area.

[0080] S1053. Determine whether there is a line segment whose length is greater than a preset value according to the length of the line segment.

[0081] Specifically, the arc contour corresponding to the line segment is limited to a certain span according to the length of the line segment to avoid affecting the calculation accuracy of the farmland fire area due to the arc contour with a large span. As a specific example, the preset length value can be selected according to the resolution. For example, if the resolution is 720P, the preset length value can be 50,000 pixels.

[0082] If there is a line segment whose length is greater than the preset value, execute step S1054;

[0083] If there is no line segment whose length is greater than the preset value, execute step S1057;

[0084] S1054, obtaining a corresponding line segment having a length greater than a preset value as an optimizable line segment, and obtaining edge data corresponding to the optimizable line segment to obtain a secondary edge contour inflection point of the edge data corresponding to the optimizable line segment.

[0085] When a line segment with a length greater than the preset value appears, it indicates that the line segment may need further optimization. However, it is also possible that the span is simply long, and the distance between the line segment and the arc is small. In this case, even if an arc contour with a large span appears, the impact on the calculation accuracy of the farmland fire area can be ignored. Therefore, in order to determine whether it is necessary to optimize a line segment greater than the preset value, it is necessary to further determine whether the distance between the line segment and the arc meets the standard. In this embodiment, the distance between the line segment and the arc can be quantified by the distance value between the inflection point of the secondary edge contour and the optimizable line segment for comparison.

[0086] S1055: Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than a distance threshold.

[0087] As a specific example, the corresponding spacing threshold may be 3000 pixels. It should be further explained that the spacing threshold and the preset length value may be adjusted according to the resolution, and the specific selection may be made according to the actual situation.

[0088] If the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold, executing step S1056;

[0089] S1056, connecting the turning points of the secondary edge contour to the two end points of the optimizable line segment to construct two secondary line segments, and replacing the corresponding optimizable line segments with the two secondary line segments to update the polygon contour.

[0090] After updating the polygon outline to obtain the updated polygon outline, the process returns to step S1052 and executes step S1057 until the length of each line segment is not greater than the preset length value.

[0091] S1057, obtaining the target fire trace outline.

[0092] In summary, the farmland fire and straw burning monitoring method based on multi-source remote sensing data in the above-mentioned embodiment of the present invention, through screening the acquired temperature anomaly data to obtain the farmland fire point, identifies the initial fire trace contour of the farmland fire point, if the initial fire trace contour is an irregular contour, then screens the initial fire trace contour to obtain the target fire trace contour that meets the calculation accuracy requirements to monitor the regional range of farmland fire / straw burning; specifically, obtains the polygonal contour of the farmland fire point, and determines whether the line segment length of the contour is greater than the preset length value to improve the contour construction accuracy, when the line segment length is greater than the preset length value, then defines the line segment as an optimizable line segment and determines whether the spacing value between the optimizable line segment and the inflection point of the secondary edge contour is greater than the spacing threshold to determine whether the optimizable line segment needs to be optimized, when the spacing value is greater than the spacing threshold, then connects the inflection point of the secondary edge contour to the two end points of the optimizable line segment to construct two secondary line segments, the two secondary line segments replace the corresponding optimizable line segment to update the polygonal contour, thereby obtaining the target fire trace contour, and improving the calculation accuracy of the burning area.

[0093] Embodiment 2

[0094] See also Figure 3 , shown is a farmland fire and straw burning monitoring system based on multi-source remote sensing data in the second embodiment of the present invention, comprising:

[0095] An acquisition module is used to acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data, wherein the temperature anomaly data includes real fire points and false fire points, wherein the false fire points include factory smoke exhaust points, and the temperature anomaly data is screened to obtain real fire points, wherein the real fire points include farmland fire points and forest fire points;

[0096] A denoising module is used to identify real fire points using a machine learning image recognition algorithm to classify land types and obtain farmland fire points, obtain fire point images based on the farmland fire points, each fire point image corresponds to a farmland fire point, identify the outline of the fire point image and perform denoising processing to obtain an initial fire trace outline;

[0097] A judging module, used for judging whether the initial fire trace contour is a regular contour, wherein the regular contour includes a rectangular contour and a circular contour;

[0098] The first execution module is used for obtaining edge data of the initial fire trace contour if the initial fire trace contour is an irregular contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning;

[0099] The steps of obtaining edge data of the initial fire trace contour and screening the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning include:

[0100] Obtain edge data of the initial fire trace contour, wherein the initial fire trace contour is an irregular contour, obtain an edge contour inflection point according to the edge data, sequentially connect two adjacent edge contour inflection points to obtain a plurality of line segments, and construct a polygonal contour by the plurality of line segments to optimize the initial fire trace contour;

[0101] Get the line segment length of the polygon outline;

[0102] Judging whether there is a line segment greater than a preset length value according to the line segment length;

[0103] If there is a line segment whose length is greater than the preset value, obtain the corresponding line segment whose length is greater than the preset value as an optimizable line segment, obtain the edge data corresponding to the optimizable line segment to obtain the secondary edge contour inflection point of the edge data corresponding to the optimizable line segment;

[0104] Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold;

[0105] If it is greater than the spacing threshold, the inflection points of the secondary edge contour are connected to the two end points of the optimizable line segment to construct two secondary line segments. The two secondary line segments replace the corresponding optimizable line segments to update the polygon contour, and return to execute the step of obtaining the line segment length of the polygon contour until the length of each line segment is no greater than the preset length value, and the target fire trace contour is obtained.

[0106] In summary, the farmland fire and straw burning monitoring system based on multi-source remote sensing data in the above-mentioned embodiment of the present invention, through screening the acquired temperature anomaly data to obtain the farmland fire point, identifies the initial fire trace contour of the farmland fire point, if the initial fire trace contour is an irregular contour, then screens the initial fire trace contour to obtain the target fire trace contour that meets the calculation accuracy requirements to monitor the regional range of farmland fire / straw burning; specifically, obtains the polygonal contour of the farmland fire point, and determines whether the line segment length of the contour is greater than the preset length value to improve the contour construction accuracy, when the line segment length is greater than the preset length value, then defines the line segment as an optimizable line segment and determines whether the spacing value between the optimizable line segment and the inflection point of the secondary edge contour is greater than the spacing threshold to determine whether the optimizable line segment needs to be optimized, when the spacing value is greater than the spacing threshold, then connects the inflection point of the secondary edge contour to the two end points of the optimizable line segment to construct two secondary line segments, the two secondary line segments replace the corresponding optimizable line segment to update the polygonal contour, thereby obtaining the target fire trace contour, thereby improving the calculation accuracy of the burning area.

[0107] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0108] In addition, an embodiment of the present invention further provides a data processing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method in the above embodiment when executing the program.

[0109] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0110] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or processing in another suitable manner as necessary, and then stored in a computer memory.

[0111] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0112] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0113] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A method for monitoring farmland fire and straw burning based on multi-source remote sensing data, characterized in that: include: Acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data, wherein the temperature anomaly data includes real fire points and false fire points, wherein the false fire points include factory smoke exhaust points, and filter the temperature anomaly data to obtain real fire points, wherein the real fire points include farmland fire points and forest fire points; Using a machine learning image recognition algorithm to identify real fire spots to classify land types and obtain farmland fire spots, obtaining fire spot images based on the farmland fire spots, each fire spot image corresponding to a farmland fire spot, identifying the outline of the fire spot image and performing denoising processing to obtain an initial fire trace outline; Determining whether the initial fire trace contour is a regular contour, wherein the regular contour includes a rectangular contour and a circular contour; If not, then obtaining edge data of the initial fire trace contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning; The steps of obtaining edge data of the initial fire trace contour and screening the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning include: Obtain edge data of the initial fire trace contour, wherein the initial fire trace contour is an irregular contour, obtain an edge contour inflection point according to the edge data, sequentially connect two adjacent edge contour inflection points to obtain a plurality of line segments, and construct a polygonal contour by the plurality of line segments to optimize the initial fire trace contour; Get the line segment length of the polygon outline; Judging whether there is a line segment greater than a preset length value according to the line segment length; If there is a line segment whose length is greater than the preset value, obtain the corresponding line segment whose length is greater than the preset value as an optimizable line segment, obtain the edge data corresponding to the optimizable line segment to obtain the secondary edge contour inflection point of the edge data corresponding to the optimizable line segment; Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold; If it is greater than the spacing threshold, the inflection points of the secondary edge contour are connected to the two end points of the optimizable line segment to construct two secondary line segments. The two secondary line segments replace the corresponding optimizable line segments to update the polygon contour, and return to execute the step of obtaining the line segment length of the polygon contour until the length of each line segment is no greater than the preset length value, and the target fire trace contour is obtained.

2. The method for monitoring farmland fire and straw burning based on multi-source remote sensing data according to claim 1, characterized in that: In the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes removing cloud interference from the remote sensing data. In the cloud removal method, data that meets the cloud screening logical relationship can be determined as cloud. The cloud screening logical relationship is: P 1>0.2 and P 26 >0.02 and Td 20 <320; In the formula, P 1 represents the reflectivity of band 1 in MODIS data; P 26 Represents the reflectivity of 26 bands in MODIS data; Td 20 Represents the brightness temperature value of the 20 bands in the MODIS data.

3. The method for monitoring farmland fire and straw burning based on multi-source remote sensing data according to claim 1, characterized in that: In the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes eliminating water interference from the remote sensing data. In the water body exclusion method, data that meets the water body screening logic relationship can be determined as water body. The water body screening logic relationship is: P 2<0.15 and P 7<0.05 and (( P 2- P 1) / ( P 2+ P 1))<0; In the formula, P 1 represents the reflectivity of band 1 in MODIS data; P 2 represents the reflectivity of band 2 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data.

4. The method for monitoring farmland fire and straw burning based on multi-source remote sensing data according to claim 1, characterized in that: In the step of preprocessing the remote sensing data to obtain temperature anomaly data, the preprocessing method includes eliminating smoke interference from the remote sensing data. The smoke screening logic relationship in the smoke elimination method is: ； In the formula, P 3 represents the reflectivity of band 3 in MODIS data; P 7 represents the reflectivity of band 7 in MODIS data; P 8 represents the reflectivity of band 8 in MODIS data; P 9 represents the reflectivity of band 9 in MODIS data.

5. The method for monitoring farmland fire and straw burning based on multi-source remote sensing data according to claim 1, characterized in that: In the step of screening the temperature anomaly data to obtain the real fire point, in the algorithm for determining the real fire point, the real fire point screening logic relationship is: ； In the formula, Td 20 Represents the brightness temperature value of the 20 bands in MODIS data; Td 31 Represents the brightness temperature value of band 31 in MODIS data; Td 32 Represents the brightness temperature value of 32 bands in MODIS data; Ts Indicates low temperature value, unit is K.

6. A farmland fire and straw burning monitoring system based on multi-source remote sensing data, characterized in that: The system comprises: An acquisition module is used to acquire remote sensing data, and pre-process the remote sensing data to obtain temperature anomaly data, wherein the temperature anomaly data includes real fire points and false fire points, wherein the false fire points include factory smoke exhaust points, and the temperature anomaly data is screened to obtain real fire points, wherein the real fire points include farmland fire points and forest fire points; A denoising module is used to identify real fire points using a machine learning image recognition algorithm to classify land types and obtain farmland fire points, obtain fire point images based on the farmland fire points, each fire point image corresponds to a farmland fire point, identify the outline of the fire point image and perform denoising processing to obtain an initial fire trace outline; A judging module, used for judging whether the initial fire trace contour is a regular contour, wherein the regular contour includes a rectangular contour and a circular contour; The first execution module is used for obtaining edge data of the initial fire trace contour if the initial fire trace contour is an irregular contour, and filtering the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning; The steps of obtaining edge data of the initial fire trace contour and screening the target fire trace contour according to the edge data to monitor the area range of farmland fire / straw burning include: Obtain edge data of the initial fire trace contour, wherein the initial fire trace contour is an irregular contour, obtain an edge contour inflection point according to the edge data, sequentially connect two adjacent edge contour inflection points to obtain a plurality of line segments, and construct a polygonal contour by the plurality of line segments to optimize the initial fire trace contour; Get the line segment length of the polygon outline; Judging whether there is a line segment greater than a preset length value according to the line segment length; If there is a line segment whose length is greater than the preset value, obtain the corresponding line segment whose length is greater than the preset value as an optimizable line segment, obtain the edge data corresponding to the optimizable line segment to obtain the secondary edge contour inflection point of the edge data corresponding to the optimizable line segment; Determine whether the distance between the inflection point of the secondary edge contour and the optimizable line segment is greater than the distance threshold; If it is greater than the spacing threshold, the inflection points of the secondary edge contour are connected to the two end points of the optimizable line segment to construct two secondary line segments. The two secondary line segments replace the corresponding optimizable line segments to update the polygon contour, and return to execute the step of obtaining the line segment length of the polygon contour until the length of each line segment is no greater than the preset length value, and the target fire trace contour is obtained.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for monitoring farmland fire and straw burning based on multi-source remote sensing data as described in any one of claims 1 to 5 is implemented.

8. A data processing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the farmland fire and straw burning monitoring method based on multi-source remote sensing data as described in any one of claims 1 to 5 is implemented.

Images