Lodging monitoring method, device, electronic equipment and storage medium for target crops

Abstract

The application provides a lodging monitoring method and device for target crops, electronic equipment and storage medium, and relates to the technical field of crop monitoring. The method comprises the following steps: extracting the target ground reflectivity of the target crops in the visible light green wave band in the monitoring range from a satellite remote sensing image; the target crops comprise at least one of corn and rice; inputting the target ground reflectivity into a decision tree model to obtain the lodging state monitoring result of the target crops output by the decision tree model; and the decision tree model is determined based on the ground reflectivity in the visible light green wave band in the target crop sample area. Through the above method, the significant difference in the target reflectivity of the target crops in the visible light green wave band is utilized to determine the target ground reflectivity in the visible light green wave band, the target ground reflectivity is input into the decision tree model, the lodging state of the target crops can be accurately judged, and the lodging monitoring precision of the target crops is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of crop monitoring, and in particular to a target crop lodging monitoring method and device, an electronic device, and a storage medium. BACKGROUND

[0002] Crop lodging monitoring is of great significance to post-disaster agricultural production management, agricultural insurance, subsidies, and the like.

[0003] In related technologies, the lodging monitoring technology for corn and rice mainly uses a typical vegetation index, a ratio vegetation index, an enhanced vegetation index, a red edge position index, and a texture mean value of three bands of short-wave infrared, red light, and red edge to construct a lodging crop remote sensing extraction model. However, such a monitoring method is easily affected by other crops around, and has low monitoring result precision and unobvious effect.

[0004] Therefore, how to improve the lodging monitoring precision for corn and rice is a problem to be solved at present. SUMMARY

[0005] To solve the problems in the prior art, the present application provides a target crop lodging monitoring method and device, an electronic device, and a storage medium.

[0006] The present application provides a target crop lodging monitoring method, comprising:

[0007] extracting a target ground surface reflectivity of a target crop in a monitoring range from a satellite remote sensing image, the target ground surface reflectivity being a ground surface reflectivity of a visible light green band; the target crop including at least one of corn and rice;

[0008] inputting the target ground surface reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crop output by the decision tree model; the decision tree model being constructed based on the ground surface reflectivity of the visible light green band in a corn and rice sample area.

[0009] Optionally, the decision tree model is constructed in the following manner:

[0010] dividing the target crop sample area into a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a fifth region of interest, and a sixth region of interest; the first region of interest being a rice lodging area, the second region of interest being a rice semi-lodging area, the third region of interest being a rice non-lodging area, the fourth region of interest being a corn lodging area, the fifth region of interest being a corn semi-lodging area, and the sixth region of interest being a corn non-lodging area;

[0011] determine at least one reflectance interval based on the ground reflectance of the visible green band of each of the regions of interest; each of the reflectance intervals is used to reflect the lodging state of the corn or the rice;

[0012] construct the decision tree model based on the reflectance intervals.

[0013] Optionally, the determining at least one reflectance interval based on the ground reflectance of the visible green band of each of the regions of interest comprises at least one of:

[0014] determine a first reflectance interval based on the maximum value and the minimum value of the ground reflectance of the visible green band in the first region of interest; the first reflectance interval is used to reflect that the rice is in a lodging state;

[0015] determine a second reflectance interval based on the maximum value and the minimum value of the ground reflectance of the visible green band in the second region of interest; the second reflectance interval is used to reflect that the rice is in a semi-lodging state;

[0016] determine a third reflectance interval based on the maximum value of the ground reflectance of the visible green band in the third region of interest; the third reflectance interval is used to reflect that the rice is in a non-lodging state;

[0017] determine a fourth reflectance interval based on the maximum value and the minimum value of the ground reflectance of the visible green band in the fourth region of interest; the fourth reflectance interval is used to reflect that the corn is in a lodging state;

[0018] determine a fifth reflectance interval based on the maximum value and the minimum value of the ground reflectance of the visible green band in the fifth region of interest; the fifth reflectance interval is used to reflect that the corn is in a semi-lodging state;

[0019] determine a sixth reflectance interval based on the maximum value of the ground reflectance of the visible green band in the sixth region of interest; the sixth reflectance interval is used to reflect that the corn is in a non-lodging state.

[0020] Optionally, the lodging state monitoring result comprises at least one of:

[0021] a first monitoring result, indicating that the rice is in a lodging state; in a case where the lodging state monitoring result is the first monitoring result, the target ground reflectance belongs to a first reflectance interval;

[0022] a second monitoring result, indicating that the rice is in a semi-lodging state; in a case where the lodging state monitoring result is the second monitoring result, the target ground reflectance belongs to a second reflectance interval;

[0023] The third monitoring result indicates that the rice is in an un-lodging state; in a case where the lodging state monitoring result is the third monitoring result, the target ground surface reflectivity belongs to a third reflectivity interval;

[0024] The fourth monitoring result indicates that the corn is in a lodging state; in a case where the lodging state monitoring result is the fourth monitoring result, the target ground surface reflectivity belongs to a fourth reflectivity interval;

[0025] The fifth monitoring result indicates that the corn is in a semi-lodging state; in a case where the lodging state monitoring result is the fifth monitoring result, the target ground surface reflectivity belongs to a fifth reflectivity interval;

[0026] The sixth monitoring result indicates that the corn is in an un-lodging state; in a case where the lodging state monitoring result is the sixth monitoring result, the target ground surface reflectivity belongs to a sixth reflectivity interval.

[0027] Optionally, the extracting of the target ground surface reflectivity of the target crop in the monitoring range in a visible light green wave band from the satellite remote sensing image comprises:

[0028] preprocessing the satellite remote sensing image to obtain apparent reflectivity of all visible light wave bands in the monitoring range;

[0029] determining ground surface reflectivity of all visible light wave bands in the monitoring range based on the apparent reflectivity;

[0030] extracting the target ground surface reflectivity from the ground surface reflectivity of all visible light wave bands.

[0031] Optionally, the preprocessing of the satellite remote sensing image to obtain apparent reflectivity of all visible light wave bands in the monitoring range comprises:

[0032] sequentially performing a target operation on the satellite remote sensing image to obtain apparent reflectivity of all visible light wave bands in the monitoring range; the target operation comprises at least one of the following:

[0033] radiation correction, atmospheric correction, image mosaicking, orthorectification, and geometric fine correction.

[0034] Optionally, after the obtaining of the lodging state monitoring result of the corn and the rice output by the decision tree model, the method further comprises:

[0035] determining a Kappa coefficient based on the lodging state monitoring result; the Kappa coefficient is used to represent the correctness of the lodging state monitoring result;

[0036] verifying the lodging state monitoring result based on the Kappa coefficient.

[0037] The application also provides a lodging monitoring device for target crops, comprising:

[0038] An extraction module is configured to extract target ground surface reflectivity of target crops in a monitoring range from satellite remote sensing images, the target ground surface reflectivity being ground surface reflectivity in a visible light green wave band; and the target crops include at least one of corn and rice.

[0039] An input module is configured to input the target ground surface reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crops output by the decision tree model; and the decision tree model is constructed based on ground surface reflectivity in a visible light green wave band in a target crop sample area.

[0040] The application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the lodging monitoring method for target crops according to any one of the above when executing the program.

[0041] The application also provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the lodging monitoring method for target crops according to any one of the above.

[0042] The application also provides a computer program product, comprising a computer program, and the computer program is executable on a processor to implement the lodging monitoring method for target crops according to any one of the above.

[0043] The lodging monitoring method for target crops, device, electronic device, and storage medium provided by the application can extract target ground surface reflectivity of target crops in a monitoring range from satellite remote sensing images, wherein the target crops include at least one of corn and rice; and the target ground surface reflectivity is in a visible light green wave band; the lodging state of the target crops has a significant difference in the target reflectivity in the visible light green wave band; the target ground surface reflectivity is input into a decision tree model, the lodging state of the target crops can be accurately determined, the lodging monitoring precision of the target crops is improved, the problem of fuzzy boundary division in previous lodging extraction is solved; and the anti-interference capability is high during the monitoring process, the accuracy and reliability of the monitoring result are ensured. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0045] Figure 1 is one of the flowcharts of the lodging monitoring method of the target crop provided by the present application;

[0046] Figure 2 is one of the flowcharts of the lodging monitoring method of the target crop provided by the present application;

[0047] Figure 3 is a schematic diagram of the multi-spectral curve of rice of different lodging types provided by the present application;

[0048] Figure 4 is a schematic diagram of the multi-spectral curve of corn of different lodging types provided by the present application;

[0049] Figure 5 is a schematic diagram of the rice lodging monitoring result of cultivated land A provided by the present application;

[0050] Figure 6 is a schematic diagram of the rice lodging monitoring result of cultivated land B provided by the present application;

[0051] Figure 7 is a schematic diagram of the rice lodging monitoring result of cultivated land C provided by the present application;

[0052] Figure 8 is a schematic diagram of the corn lodging monitoring result of cultivated land D provided by the present application;

[0053] Figure 9 is a schematic diagram of the structure of the lodging monitoring device of the target crop provided by the present application;

[0054] Figure 10 is a schematic diagram of the structure of the electronic device provided by the present application. DETAILED DESCRIPTION

[0055] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below in conjunction with the drawings in the present application. Obviously, the described embodiments are some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0056] The technical solutions provided by the present application will be described below in conjunction with Figures 1 to 8 The lodging monitoring method of the target crop provided by the present application will be described in detail. Figure 1 is one of the flowcharts of the lodging monitoring method of the target crop provided by the present application, referring to Figure 1 , the method comprises steps 101-102, wherein:

[0057] Step 101, extracting target ground reflectivity of target crops in a monitoring range from satellite remote sensing images, the target ground reflectivity being ground reflectivity in a visible light green wave band; the target crops including at least one of corn and rice.

[0058] First of all, it needs to be pointed out that the execution subject of the present application can be any electronic device capable of realizing the lodging monitoring of corn and rice, for example, it can be a smart phone, a smart watch, a desktop computer, a laptop computer or any other electronic device.

[0059] In the embodiment of the present application, the satellite remote sensing images include ground reflectivity in all visible light wave bands within the satellite monitoring range. Since the lodging state of rice and corn is most significant in the difference of target ground reflectivity in the visible light green wave band, it is necessary to extract the target ground reflectivity in the visible light green wave band of corn and rice from the ground reflectivity in all visible light wave bands.

[0060] It needs to be pointed out that in the related art, the data monitoring is usually obtained by aerial photography with a close-range unmanned aerial vehicle. Such monitoring method is easily affected by other crops around, has low monitoring result precision and unobvious effect, and has small application coverage for different types of crops, and the unmanned aerial vehicle monitoring has high cost and is only suitable for lodging of crops in a small range.

[0061] In the embodiment of the present application, the target ground reflectivity in the visible light green wave band of corn and rice in the monitoring range is directly extracted from the satellite remote sensing images, which can more intuitively and quickly extract the lodging area, greatly improve the lodging monitoring result precision, reduce the monitoring cost, and provide sufficient decision support information for crop disaster monitoring.

[0062] Step 102, inputting the target ground reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crops output by the decision tree model; the decision tree model being constructed based on the ground reflectivity in the visible light green wave band in a target crop sample area.

[0063] In the embodiment of the present application, the target crop sample area includes lodging areas, semi-lodging areas and non-lodging areas of corn and rice. Based on the ground reflectivity in the visible light green wave band in the corn and rice sample area, the ground reflectivity interval range and the critical value of full lodging, semi-lodging and non-lodging of rice and corn in the visible light green wave band can be statistically analyzed, and the lodging state of corn and rice in the monitoring range of satellite remote sensing images can be judged by using the interval range and the critical value.

[0064] The application provides a lodging monitoring method for target crops, which comprises the following steps: extracting target ground reflectivity of a visible light green wave band of the target crops in a monitoring range from satellite remote sensing images, wherein the target crops comprise at least one of corn and rice; inputting the target ground reflectivity into a decision tree model by using the significant difference of the target reflectivity of the lodging state of the target crops in the visible light green wave band, so as to accurately judge the lodging state of the target crops, thereby improving the lodging monitoring precision of the target crops, solving the problem of fuzzy boundary division in the lodging extraction in the prior art, and improving the anti-interference capability in the monitoring process and the accuracy and reliability of the monitoring results.

[0065] Optionally, the target ground reflectivity of the visible light green wave band of the target crops in the monitoring range is extracted from the satellite remote sensing images, and the extraction can be realized through the following steps.

[0066] Step 1), pre-processing the satellite remote sensing images to obtain apparent reflectivity of all visible light wave bands in the monitoring range.

[0067] Optionally, the satellite remote sensing images are pre-processed to obtain apparent reflectivity of all visible light wave bands in the monitoring range, and the pre-processing can be realized through the following steps.

[0068] The target operation is sequentially performed on the satellite remote sensing images to obtain apparent reflectivity of all visible light wave bands in the monitoring range; the target operation comprises at least one of the following operations:

[0069] a) radiation correction; b) atmospheric correction; c) image mosaic; d) ortho correction; and e) geometric fine correction.

[0070] Specifically, first, the satellite remote sensing images are subjected to radiation correction to convert digital quantization values (DN) of the satellite remote sensing images into radiance values. Then, the satellite remote sensing images are subjected to atmospheric correction to convert the radiance values into remote sensing image radiance. Then, the satellite remote sensing images are subjected to image mosaic, ortho correction, geometric fine correction and other pre-processing, and then apparent reflectivity of all visible light wave bands in the monitoring range is calculated based on the remote sensing image radiance by using the following formula (1):

[0071]

[0072] wherein ρ represents apparent reflectivity of all visible light wave bands; L represents remote sensing image radiance (image value after atmospheric correction), and the unit is W.m-2sr-1um-1; d represents the distance between the earth and the sun; F0 represents solar irradiance (unit: W / m2*um-1) outside the atmosphere; and cosθ0 represents the solar irradiation angle.

[0073] Step 2), determining the ground reflectance of all visible light bands in the monitoring range based on the apparent reflectance.

[0074] Optionally, the ground reflectance of all visible light bands in the monitoring range is calculated by using the following formula (2):

[0075]

[0076] wherein Rrs represents the ground reflectance of all visible light bands; p represents the apparent reflectance of all visible light bands; t0=exp[-0.5τ r / cos(θ0)], τ r represents the Rayleigh optical thickness, τ r (λ) = 0.008569λ -4 (1+0.0113λ -2 +0.00013λ -4 )*(λ: μm), λ represents the wavelength, and is converted to μm during calculation.

[0077] Step 3), extracting the target ground reflectance from the ground reflectance of all visible light bands.

[0078] In the embodiment of the present application, the sensors of different satellites are all provided with band length introductions, and thus the target ground reflectance of the visible light green band can be extracted from the ground reflectance of all visible light bands according to the band length introductions of the sensors of different satellites.

[0079] In the above embodiment, the target ground reflectance of the visible light green band of corn and rice in the monitoring range is extracted from the satellite remote sensing image, and the target ground reflectance is input into the decision tree model by using the significant difference in the target reflectance of the visible light green band between the lodging state of rice and corn, so that the lodging state of rice and corn can be accurately judged, thereby improving the lodging monitoring precision of corn and rice.

[0080] Optionally, after the target ground reflectance is input into the decision tree model, the lodging state monitoring result of the target crop output by the decision tree model is obtained, which includes at least one of the following:

[0081] a) a first monitoring result, indicating that the rice is in a lodging state; in the case that the lodging state monitoring result is the first monitoring result, the target ground reflectance belongs to a first reflectance interval.

[0082] In the embodiment of the present application, the first reflectance interval is used to reflect that the rice is in a lodging state.

[0083] b) a second monitoring result indicating that the rice is in a semi-lodging state; in a case where the lodging state monitoring result is the second monitoring result, the target ground surface reflectivity belongs to a second reflectivity interval.

[0084] In an embodiment of the present application, the second reflectivity interval is used to reflect that the rice is in a semi-lodging state.

[0085] c) a third monitoring result indicating that the rice is in a non-lodging state; in a case where the lodging state monitoring result is the third monitoring result, the target ground surface reflectivity belongs to a third reflectivity interval.

[0086] In an embodiment of the present application, the third reflectivity interval is used to reflect that the rice is in a non-lodging state.

[0087] d) a fourth monitoring result indicating that the corn is in a lodging state; in a case where the lodging state monitoring result is the fourth monitoring result, the target ground surface reflectivity belongs to a fourth reflectivity interval.

[0088] In an embodiment of the present application, the fourth reflectivity interval is used to reflect that the corn is in a lodging state.

[0089] e) a fifth monitoring result indicating that the corn is in a semi-lodging state; in a case where the lodging state monitoring result is the fifth monitoring result, the target ground surface reflectivity belongs to a fifth reflectivity interval.

[0090] In an embodiment of the present application, the fifth reflectivity interval is used to reflect that the corn is in a semi-lodging state.

[0091] f) a sixth monitoring result indicating that the corn is in a non-lodging state; in a case where the lodging state monitoring result is the sixth monitoring result, the target ground surface reflectivity belongs to a sixth reflectivity interval.

[0092] In an embodiment of the present application, the sixth reflectivity interval is used to reflect that the corn is in a non-lodging state.

[0093] Optionally, the decision tree model is constructed by the following way:

[0094] Step 1), dividing the target crop sample area into a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a fifth region of interest and a sixth region of interest; the first region of interest is a rice lodging region, the second region of interest is a rice semi-lodging region, the third region of interest is a rice non-lodging region, the fourth region of interest is a corn lodging region, the fifth region of interest is a corn semi-lodging region, and the sixth region of interest is a corn non-lodging region.

[0095] In the embodiment of the present application, the decision tree model is determined based on the ground reflectivity of the visible light green band in the target crop sample area.

[0096] Specifically, by artificial visual interpretation method, the target crop sample area is divided into six interest regions of rice and corn lodging, semi-lodging, and non-lodging. Among them, the division principle of each interest region meets the following conditions:

[0097] a) The most serious full lodging of rice and corn and the best growth of non-lodging in the target crop sample area are determined as the interest region.

[0098] In the embodiment of the present application, the most serious lodging area of rice is taken as the first interest region, the best growth area of rice is taken as the third interest region, and the other area of rice is taken as the second interest region, that is, the semi-lodging area of rice.

[0099] The most serious lodging area of corn is taken as the fourth interest region, the best growth area of corn is taken as the sixth interest region, and the other area of corn is taken as the fifth interest region, that is, the semi-lodging area of corn.

[0100] b) Four boundaries of "lodging and semi-lodging boundary, semi-lodging and non-lodging boundary" of rice and corn, each boundary has 1 interest region of different levels.

[0101] In the embodiment of the present application, there is a first boundary value of the ground reflectivity of the visible light green band between the first interest region and the second interest region of rice, there is a second boundary value of the ground reflectivity of the visible light green band between the second interest region and the third interest region of rice, there is a third boundary value of the ground reflectivity of the visible light green band between the fourth interest region and the fifth interest region of corn, and there is a fourth boundary value of the ground reflectivity of the visible light green band between the fifth interest region and the sixth interest region of corn.

[0102] After dividing the target crop sample area into six interest regions, the ground reflectivity of the visible light green band of each interest region is extracted.

[0103] Step 2), at least one reflectivity interval is determined based on the ground reflectivity of the visible light green band of each interest region; each reflectivity interval is used to reflect the lodging state of the corn or the rice.

[0104] Step 3), the decision tree model is constructed based on each reflectivity interval.

[0105] Optionally, the determination of at least one reflectivity interval based on the ground reflectivity of the visible light green band of each interest region comprises at least one of the following:

[0106] a) determining a first reflectance interval based on the maximum and minimum of the ground reflectance of the visible green band in the first region of interest; the first reflectance interval is used to reflect that the rice is in the flat state.

[0107] In the embodiment of the present application, the minimum of the ground reflectance of the visible green band in the first region of interest, that is, the first critical value of the first region of interest and the second region of interest.

[0108] b) determining a second reflectance interval based on the maximum and minimum of the ground reflectance of the visible green band in the second region of interest; the second reflectance interval is used to reflect that the rice is in the semi-flat state.

[0109] In the embodiment of the present application, the minimum of the ground reflectance of the visible green band in the second region of interest, that is, the second critical value of the second region of interest and the third region of interest.

[0110] It should be noted that the maximum of the ground reflectance of the visible green band in the second region of interest and the minimum of the ground reflectance of the visible green band in the first region of interest can be equal or not equal.

[0111] c) determining a third reflectance interval based on the maximum of the ground reflectance of the visible green band in the third region of interest; the third reflectance interval is used to reflect that the rice is in the non-flat state.

[0112] In the embodiment of the present application, the maximum of the ground reflectance of the visible green band in the third region of interest, and the minimum of the ground reflectance of the visible green band in the second region of interest can be equal or not equal.

[0113] d) determining a fourth reflectance interval based on the maximum and minimum of the ground reflectance of the visible green band in the fourth region of interest; the fourth reflectance interval is used to reflect that the corn is in the flat state.

[0114] In the embodiment of the present application, the minimum of the ground reflectance of the visible green band in the fourth region of interest, that is, the third critical value of the fourth region of interest and the fifth region of interest.

[0115] e) determining a fifth reflectance interval based on the maximum and minimum of the ground reflectance of the visible green band in the fifth region of interest; the fifth reflectance interval is used to reflect that the corn is in the semi-flat state.

[0116] In the embodiment of the present application, the minimum of the ground reflectance of the visible green band in the fifth region of interest, that is, the fourth critical value of the fifth region of interest and the sixth region of interest.

[0117] It should be noted that the maximum value of the surface reflectivity of the visible light green band in the fifth region of interest can be equal to or different from the minimum value of the surface reflectivity of the visible light green band in the fourth region of interest.

[0118] f) determining a sixth reflectivity interval based on the maximum value of the surface reflectivity of the visible light green band in the sixth region of interest; the sixth reflectivity interval is used to reflect that the corn is in the non-lodged state.

[0119] In the embodiments of the present application, the maximum value of the surface reflectivity of the visible light green band in the sixth region of interest can be equal to or different from the minimum value of the surface reflectivity of the visible light green band in the fifth region of interest.

[0120] Optionally, after obtaining the lodging state monitoring result of the target crop output by the decision tree model, it is also necessary to verify the lodging state monitoring result of the target crop, which can be realized by the following steps:

[0121] Step 1), determining a Kappa coefficient based on the lodging state monitoring result; the Kappa coefficient is used to represent the correctness of the lodging state monitoring result;

[0122] Step 2), verifying the lodging state monitoring result based on the Kappa coefficient.

[0123] In the embodiments of the present application, the Kappa coefficient is an index for verifying whether the model prediction result and the actual classification result are consistent, which can be used to measure the classification effect of the decision tree model.

[0124] Specifically, first, the overall accuracy is determined based on the lodging state monitoring result of the corn and the rice, wherein the overall accuracy is the percentage of the number of all correct classification inspection points (lodging, semi-lodging and non-lodging) in the total number of extracted inspection points, that is, the sum of all values on the diagonal line in the confusion matrix divided by the total sum of all samples.

[0125] The calculation of the Kappa coefficient is based on the confusion matrix, and the general value range is [0, 1]. When Kappa<0.4, it indicates that the classification result accuracy is poor; when 0.4≤Kappa<0.75, it indicates that the classification result accuracy is general; when 0.75≤Kappa<0.85, it indicates that the classification result is high; and when Kappa>0.85, it indicates that the classification result accuracy is high.

[0126] When the classification result accuracy of the decision tree model is poor, manual correction is needed to ensure the correctness and reliability of the classification of the decision tree model.

[0127] Figure 2is the second flowchart of the lodging monitoring method of the target crop provided by the application, see Figure 2 As shown, comprising steps 201-215, wherein:

[0128] Step 201, acquiring high-resolution remote sensing images.

[0129] Step 202, performing radiation correction on the remote sensing images to convert the digital quantization value of the remote sensing images into a radiation brightness value.

[0130] Step 203, performing atmospheric correction on the remote sensing images to convert the radiation brightness value into a remote sensing image radiation rate.

[0131] Step 204, performing image mosaic processing, orthographic correction processing and geometric fine correction processing on the remote sensing images.

[0132] Step 205, determining the apparent reflectivity of all visible light bands within the satellite monitoring range based on the remote sensing image radiation rate.

[0133] Step 206, determining the ground surface reflectivity of all visible light bands within the satellite monitoring range according to the apparent reflectivity of all visible light bands within the satellite monitoring range.

[0134] Step 207, extracting the ground surface reflectivity of the visible light green band from the ground surface reflectivity of all visible light bands.

[0135] Step 208, making the lodging, semi-lodging, and non-lodging regions of interest of rice and the lodging, semi-lodging, and non-lodging regions of interest of corn by artificial visual interpretation method.

[0136] Step 209, setting the ground surface reflectivity interval of the visible light green band for the lodging, semi-lodging, and non-lodging states of rice and corn based on the ground surface reflectivity of the visible light green band in each region of interest.

[0137] Step 210, constructing a decision tree model based on the ground surface reflectivity interval of each visible light green band.

[0138] Step 211, monitoring the lodging, semi-lodging, and non-lodging regions of corn and rice based on the ground surface reflectivity of the visible light green band within the corn and rice farmland range in the remote sensing images and the decision tree model.

[0139] For example, based on the decision tree model, it is determined whether the ground reflectivity of the visible light green band in the corn and rice cultivated land range is in the [A, B] ground reflectivity interval, if yes, the rice is determined to be in the lodging state; if not, it is determined whether the ground reflectivity of the visible light green band in the corn and rice cultivated land range is in the [C, D] ground reflectivity interval, if yes, the rice is determined to be in the semi-lodging state; if not, it is determined whether the ground reflectivity of the visible light green band in the corn and rice cultivated land range is in the [E, F] ground reflectivity interval, if yes, the corn is determined to be in the lodging state; if not, it is determined whether the ground reflectivity of the visible light green band in the corn and rice cultivated land range is in the [G, H] ground reflectivity interval, if yes, the corn is determined to be in the semi-lodging state; if not, the rice or corn is determined to be in the non-lodging state.

[0140] Step 212, manually correcting the monitoring result.

[0141] Step 213, extracting the monitoring result after manual correction.

[0142] Step 214, verifying the accuracy of the monitoring result after manual correction.

[0143] Optionally, the accuracy verification includes office accuracy verification and field accuracy verification.

[0144] Step 215, outputting the verification result.

[0145] In order to facilitate clearer understanding of the lodging monitoring method of the target crop provided in the embodiments of the present application, the lodging monitoring method of the target crop provided in the present application is further explained and described below in combination with specific embodiments.

[0146] Remote sensing monitoring of crop lodging mainly monitors the disaster range and disaster level according to the differences in spectrum, hue and texture and other characteristics of the lodging crop and the normally growing crop in remote sensing images. When the crop is lodged, the crop population canopy structure and morphology change greatly, which is reflected in the change of pixel values on the remote sensing image. By analyzing the changes of remote sensing images before and after lodging, the monitoring of the disaster range and the disaster degree of the crop can be realized.

[0147] In the present embodiment, the lodging monitoring method selects satellite images with a resolution of 4 meters. The satellite has programming capability and can realize 2 times coverage per day for any area, and can collect up to 50 million square kilometers of data every day. It has four spectral bands of R, G, B and NIR, a spatial resolution of 3 meters, a width of 30 km, a coordinate projection of RPC / WGS84UTM / WGS84, an orbital height of 508 kilometers, and an orbital inclination of 55°.

[0148] The embodiment of the present application extracts crop lodging information of different farms by using the significant difference of the green wave band based on the farmland data of the farms, improves the precision of the monitoring result, and realizes high-precision extraction of large-scale rice and corn lodging crops.

[0149] The specific implementation steps of the embodiment of the present application are as follows:

[0150] Step one: rice and corn have different reflectivity in different wave bands, the most obvious wave band is selected according to the difference of different reflectivity, and a decision tree classification method is used to monitor rice and corn lodging by setting a threshold value.

[0151] Step two: sample selection is performed according to rice survey data, 250 lodging samples, 280 half-lodging samples and 300 non-lodging samples are selected in the test area, the reflectivity of normal rice, corn and half-lodging rice and corn in each wave band is counted, and then the reflectivity curves of normal, lodging rice and corn are drawn according to the position of the wave band center wavelength, and the results are shown in Figure 3 、 Figure 4 . Figure 3 is the multi-spectral curve schematic diagram of different lodging types of rice provided by the present application. Figure 4 is the multi-spectral curve schematic diagram of different lodging types of corn provided by the present application.

[0152] Step three: it can be seen from Figure 3 、 Figure 4 that the reflectivity of lodging and half-lodging rice and corn in the range of 456-853 nm is obviously increased compared with that of non-lodging rice and corn. Among them, in the visible light wave band, the difference of green light reflectivity is the most significant, and the green light reflectivity of non-lodging rice increases by 0.067 and 0.057 from non-lodging rice to lodging rice, and the green light reflectivity of non-lodging corn increases by 0.031 and 0.023 from non-lodging corn to lodging corn. Figure 3 、 Figure 4 It can be seen from Figure 3 、 Figure 4 that the separation of lodging rice and lodging corn, half-lodging rice and half-lodging corn, and non-lodging rice and non-lodging corn in the green light wave band is the largest, that is, the green wave band can be used as an index to distinguish lodging, half-lodging and non-lodging.

[0153] In order to detect the monitoring precision of the embodiment of the present application, the precision verification of the indoor and outdoor is carried out, and the overall classification precision and Kappa coefficient are used to verify the monitoring precision.

[0154] 1) The overall precision is the percentage of the number of all correct classification of lodging, half-lodging and non-lodging test points in the total number of extracted test points, that is, the sum of all values in the diagonal line in the confusion matrix divided by the total sum of all samples, which is obtained by multiplying the total number of all true reference pixels by the sum of the diagonal line in the confusion matrix.

[0155] 2) Kappa coefficient is an index for consistency test (so-called consistency is whether the model prediction result and the actual classification result are consistent) and can be used to measure the effect of classification. The calculation of Kappa coefficient is based on the confusion matrix, and the general value range is [0, 1]. When Kappa < 0.4, it indicates that the classification result accuracy is poor; when 0.4≤Kappa < 0.75, it indicates that the classification result accuracy is general; when 0.75≤Kappa < 0.85, it indicates that the classification result is higher; when Kappa > 0.85, it indicates that the classification result accuracy is high.

[0156] Table 1 is a lodging monitoring indoor precision verification table, and Table 2 is a field (outdoor) precision verification table. After verification, the extraction precision indoor precision of the present case is as high as 90%, and the outdoor precision is as high as 80%.

[0157] Table 1

[0158]

[0159] Table 2

[0160]

[0161] Figures 5 to 8 It is a lodging monitoring result schematic diagram of rice and corn in different farmlands provided by the present application. Figure 5 It is a lodging monitoring result schematic diagram of rice in farmland A provided by the present application.

[0162] Figure 6 It is a lodging monitoring result schematic diagram of rice in farmland B provided by the present application. Figure 7 It is a lodging monitoring result schematic diagram of rice in farmland C provided by the present application. Figure 8 It is a lodging monitoring result schematic diagram of corn in farmland D provided by the present application.

[0163] The lodging monitoring device of the target crop provided by the present application is described below, and the lodging monitoring device of the target crop described below can be correspondingly referred to the lodging monitoring method of the target crop described above. Figure 9 It is a structure schematic diagram of the lodging monitoring device of the target crop provided by the present application, as shown in Figure 9 The lodging monitoring device 900 of the target crop provided by the present application includes an extraction module 901 and an input module 902, wherein:

[0164] The extraction module 901 is used for extracting the target ground reflectivity of the target crop in the monitoring range from the satellite remote sensing image, and the target ground reflectivity is the ground reflectivity of the visible light green wave band; the target crop includes at least one of corn and rice;

[0165] The input module 902 is configured to input the target ground reflectivity into the decision tree model to obtain a lodging state monitoring result of the target crop output by the decision tree model.

[0166] The lodging monitoring device for the target crop provided by the application can extract the target ground reflectivity of the visible light green wave band of the target crop in the monitoring range from the satellite remote sensing image, wherein the target crop includes at least one of corn and rice; the target ground reflectivity is input into the decision tree model by using the significant difference of the target reflectivity of the visible light green wave band of the lodging state of the target crop, the lodging state of the target crop can be accurately judged, the lodging monitoring precision of the target crop is improved, the problem of fuzzy boundary division in the previous lodging extraction is solved; and the anti-interference capability is high during the monitoring process, and the accuracy and reliability of the monitoring result are ensured.

[0167] Optionally, the device further comprises:

[0168] The division module is configured to divide the target crop sample area into a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a fifth region of interest and a sixth region of interest; the first region of interest is a rice lodging area, the second region of interest is a rice semi-lodging area, the third region of interest is a rice non-lodging area, the fourth region of interest is a corn lodging area, the fifth region of interest is a corn semi-lodging area, and the sixth region of interest is a corn non-lodging area.

[0169] The first determination module is configured to determine at least one reflectivity interval based on the ground reflectivity of the visible light green wave band of each region of interest; each reflectivity interval is used to reflect the lodging state of the corn or the rice.

[0170] The construction module is configured to construct the decision tree model based on each reflectivity interval.

[0171] Optionally, the determination module is further configured to perform at least one of the following:

[0172] determine a first reflectivity interval based on the maximum value and the minimum value of the ground reflectivity of the visible light green wave band in the first region of interest; the first reflectivity interval is used to reflect that the rice is in a lodging state;

[0173] determine a second reflectivity interval based on the maximum value and the minimum value of the ground reflectivity of the visible light green wave band in the second region of interest; the second reflectivity interval is used to reflect that the rice is in a semi-lodging state;

[0174] determine a third reflectivity interval based on the maximum value of the ground reflectivity of the visible light green band in the third region of interest; the third reflectivity interval is used to reflect that the rice is in an un-lodged state;

[0175] determine a fourth reflectivity interval based on the maximum value and the minimum value of the ground reflectivity of the visible light green band in the fourth region of interest; the fourth reflectivity interval is used to reflect that the corn is in a lodged state;

[0176] determine a fifth reflectivity interval based on the maximum value and the minimum value of the ground reflectivity of the visible light green band in the fifth region of interest; the fifth reflectivity interval is used to reflect that the corn is in a semi-lodged state;

[0177] determine a sixth reflectivity interval based on the maximum value of the ground reflectivity of the visible light green band in the sixth region of interest; the sixth reflectivity interval is used to reflect that the corn is in an un-lodged state.

[0178] Optionally, the lodged state monitoring result includes at least one of:

[0179] a first monitoring result, indicating that the rice is in a lodged state; in a case where the lodged state monitoring result is the first monitoring result, the target ground reflectivity belongs to a first reflectivity interval;

[0180] a second monitoring result, indicating that the rice is in a semi-lodged state; in a case where the lodged state monitoring result is the second monitoring result, the target ground reflectivity belongs to a second reflectivity interval;

[0181] a third monitoring result, indicating that the rice is in an un-lodged state; in a case where the lodged state monitoring result is the third monitoring result, the target ground reflectivity belongs to a third reflectivity interval;

[0182] a fourth monitoring result, indicating that the corn is in a lodged state; in a case where the lodged state monitoring result is the fourth monitoring result, the target ground reflectivity belongs to a fourth reflectivity interval;

[0183] a fifth monitoring result, indicating that the corn is in a semi-lodged state; in a case where the lodged state monitoring result is the fifth monitoring result, the target ground reflectivity belongs to a fifth reflectivity interval;

[0184] a sixth monitoring result, indicating that the corn is in an un-lodged state; in a case where the lodged state monitoring result is the sixth monitoring result, the target ground reflectivity belongs to a sixth reflectivity interval.

[0185] Optionally, the extraction module 901 is further used for:

[0186] Preprocessing the satellite remote sensing image to obtain apparent reflectance of all visible light bands in the monitoring range;

[0187] Based on the apparent reflectance, determine the ground reflectance of all visible light bands in the monitoring range;

[0188] Extract the target ground reflectance from the ground reflectance of all visible light bands.

[0189] Optionally, the extraction module 901 is further used for:

[0190] Performing target operations on the satellite remote sensing image in turn to obtain the apparent reflectance of all visible light bands in the monitoring range; the target operation includes at least one of the following:

[0191] Radiation correction, atmospheric correction, image mosaic, orthorectification and geometric fine correction.

[0192] Optionally, the device further comprises:

[0193] The second determination module is configured to determine a Kappa coefficient based on the lodging state monitoring result; the Kappa coefficient is used to represent the correctness of the lodging state monitoring result;

[0194] The verification module is configured to verify the lodging state monitoring result based on the Kappa coefficient.

[0195] Figure 10 is a structural schematic diagram of an electronic device provided by the application, as Figure 10 As shown in the figure, the electronic device can include: a processor 1010, a communications interface 1020, a memory 1030 and a communications bus 1040, wherein the processor 1010, the communications interface 1020, the memory 1030 complete mutual communication through the communications bus 1040. The processor 1010 can call the logical instructions in the memory 1030 to execute the lodging monitoring method of the target crop, which includes: extracting the target ground reflectance of the target crop in the monitoring range from the satellite remote sensing image, the target ground reflectance being the ground reflectance of the visible light green band; the target crop includes at least one of corn and rice; inputting the target ground reflectance into a decision tree model to obtain the lodging state monitoring result of the target crop output by the decision tree model; the decision tree model is constructed based on the ground reflectance of the visible light green band in the target crop sample area.

[0196] In addition, the logic instructions in the memory 1030 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0197] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the lodging monitoring method of the target crop provided by the above-mentioned methods. The method comprises: extracting the target ground reflectance of the target crop in the monitoring range from the satellite remote sensing image, the target ground reflectance being the ground reflectance of the visible green wave band; the target crop comprising at least one of corn and rice; inputting the target ground reflectance into a decision tree model to obtain the lodging state monitoring result of the target crop output by the decision tree model; and the decision tree model being constructed based on the ground reflectance of the visible green wave band in the target crop sample area.

[0198] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the lodging monitoring method of the target crop provided by the above-mentioned methods. The method comprises: extracting the target ground reflectance of the target crop in the monitoring range from the satellite remote sensing image, the target ground reflectance being the ground reflectance of the visible green wave band; the target crop comprising at least one of corn and rice; inputting the target ground reflectance into a decision tree model to obtain the lodging state monitoring result of the target crop output by the decision tree model; and the decision tree model being constructed based on the ground reflectance of the visible green wave band in the target crop sample area.

[0199] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0200] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0201] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of monitoring lodging of a target crop, characterized by, The method comprises the following steps: extracting target ground reflectivity of target crops in a monitoring range from satellite remote sensing images, the target ground reflectivity being ground reflectivity of a visible light green wave band; the target crops include at least one of corn and rice; inputting the target ground reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crops output by the decision tree model; the decision tree model is constructed based on ground reflectivity of a visible light green wave band in a target crop sample area; the decision tree model is constructed by the following method: dividing the target crop sample area into a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a fifth region of interest and a sixth region of interest; the first region of interest is a rice lodging area, the second region of interest is a rice semi-lodging area, the third region of interest is a rice non-lodging area, the fourth region of interest is a corn lodging area, the fifth region of interest is a corn semi-lodging area, and the sixth region of interest is a corn non-lodging area; determining at least one reflectivity interval based on ground reflectivity of a visible light green wave band of each region of interest; each reflectivity interval is used to reflect the lodging state of the corn or the rice; constructing the decision tree model based on each reflectivity interval.

2. The method of lodging monitoring of a target crop according to claim 1, characterized in that, The method of determining at least one reflectivity interval based on ground reflectivity of a visible light green wave band of each region of interest comprises at least one of the following: determining a first reflectivity interval based on the maximum and minimum values of the ground reflectivity of the visible light green wave band in the first region of interest; the first reflectivity interval is used to reflect that the rice is in a lodging state; determining a second reflectivity interval based on the maximum and minimum values of the ground reflectivity of the visible light green wave band in the second region of interest; the second reflectivity interval is used to reflect that the rice is in a semi-lodging state; determining a third reflectivity interval based on the maximum value of the ground reflectivity of the visible light green wave band in the third region of interest; the third reflectivity interval is used to reflect that the rice is in a non-lodging state; determining a fourth reflectivity interval based on the maximum and minimum values of the ground reflectivity of the visible light green wave band in the fourth region of interest; the fourth reflectivity interval is used to reflect that the corn is in a lodging state; determining a fifth reflectivity interval based on the maximum and minimum values of the ground reflectivity of the visible light green wave band in the fifth region of interest; the fifth reflectivity interval is used to reflect that the corn is in a semi-lodging state; determining a sixth reflectivity interval based on the maximum value of the ground reflectivity of the visible light green wave band in the sixth region of interest; the sixth reflectivity interval is used to reflect that the corn is in a non-lodging state.

3. The lodging monitoring method of the target crop according to claim 1 or 2, characterized by, The lodging state monitoring result comprises at least one of the following: a first monitoring result indicating that the rice is in a lodging state; in the case that the lodging state monitoring result is the first monitoring result, the target ground reflectivity belongs to a first reflectivity interval; a second monitoring result indicating that the rice is in a semi-lodging state; In a case where the lodging state monitoring result is the second monitoring result, the target ground surface reflectivity belongs to a second reflectivity interval; A third monitoring result indicates that the rice is in an un-lodging state; In a case where the lodging state monitoring result is the third monitoring result, the target ground surface reflectivity belongs to a third reflectivity interval; A fourth monitoring result indicates that the corn is in a lodging state; In a case where the lodging state monitoring result is the fourth monitoring result, the target ground surface reflectivity belongs to a fourth reflectivity interval; A fifth monitoring result indicates that the corn is in a semi-lodging state; In a case where the lodging state monitoring result is the fifth monitoring result, the target ground surface reflectivity belongs to a fifth reflectivity interval; A sixth monitoring result indicates that the corn is in an un-lodging state; in a case where the lodging state monitoring result is the sixth monitoring result, the target ground surface reflectivity belongs to a sixth reflectivity interval.

4. The method of lodging monitoring of a target crop according to claim 1 or 2, characterized in that, The target ground surface reflectivity of the target crop in the monitoring range in the visible green wave band is extracted from the satellite remote sensing image, including: Pretreating the satellite remote sensing image to obtain the apparent reflectivity of all visible light wave bands in the monitoring range; Based on the apparent reflectivity, determining the ground surface reflectivity of all visible light wave bands in the monitoring range; Extracting the target ground surface reflectivity from the ground surface reflectivity of all visible light wave bands.

5. The method of lodging monitoring of a target crop according to claim 4, characterized in that, The pretreatment of the satellite remote sensing image to obtain the apparent reflectivity of all visible light wave bands in the monitoring range includes: Sequentially performing a target operation on the satellite remote sensing image to obtain the apparent reflectivity of all visible light wave bands in the monitoring range; the target operation includes at least one of the following: Radiation correction, atmospheric correction, image mosaic, orthorectification, and geometric fine correction.

6. The method of lodging monitoring of a target crop according to claim 1 or 2, characterized in that, The method further includes: Based on the lodging state monitoring result, determining a Kappa coefficient; the Kappa coefficient is used to represent the correctness of the lodging state monitoring result; Based on the Kappa coefficient, verifying the lodging state monitoring result.

7. A lodging monitoring device for a target crop, characterized by, Including: An extraction module is configured to extract a target ground surface reflectivity of a target crop in a monitoring range from a satellite remote sensing image, the target ground surface reflectivity being a ground surface reflectivity in a visible green wave band; the target crop includes at least one of corn and rice; An input module is configured to input the target ground surface reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crop output by the decision tree model; the decision tree model is constructed based on the ground surface reflectivity in the visible green wave band in a target crop sample area; An input module is configured to input the target ground surface reflectivity into a decision tree model to obtain a lodging state monitoring result of the target crop output by the decision tree model; the decision tree model is constructed based on the ground surface reflectivity in the visible green wave band in a target crop sample area; The decision tree model is built by dividing the target crop sample area into a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a fifth region of interest and a sixth region of interest; the first region of interest is a rice lodging region, the second region of interest is a rice semi-lodging region, the third region of interest is a rice non-lodging region, the fourth region of interest is a corn lodging region, the fifth region of interest is a corn semi-lodging region, and the sixth region of interest is a corn non-lodging region; at least one reflectance interval is determined based on the surface reflectance of the visible light green wave band of each region of interest; each reflectance interval is used to reflect the lodging state of the corn or the rice; and the decision tree model is built based on each reflectance interval.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the lodging monitoring method of the target crop according to any one of claims 1 to 6. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the lodging monitoring method of the target crop according to any one of claims 1 to 6.

Images