Upright corn stalk extraction method, device, equipment, medium and program product

Abstract

The application provides a kind of standing corn stalk extraction method, device, equipment, medium and program product, it is related to agricultural remote sensing technical field, the method comprises: in autumn and winter, obtain the first reflectivity of the ground object to be identified in corn planting area in SWIR1 wave band and the second reflectivity in SWIR2 wave band;With the first reflectivity and the second reflectivity as the coordinates of the point to be identified, the first distance and the second distance between the bare soil line and the straw residue line are calculated respectively, the ratio of the first distance and the second distance is used as the remote sensing spectral index corresponding to the ground object to be identified;The maximum and minimum value normalization calculation is carried out on the first reflectivity and the second reflectivity respectively, and the larger normalized value is used as the shadow index corresponding to the ground object to be identified;Determine the standing corn stalk area based on the remote sensing spectral index and the shadow index corresponding to the ground object to be identified.The application enhances the identification ability of standing corn stalk, and improves the remote sensing monitoring precision of standing corn stalk.

Description

Technical Field

[0001] This invention relates to the field of agricultural remote sensing technology, and in particular to a method, apparatus, equipment, medium, and program product for extracting upright corn stalks. Background Technology

[0002] With the development of China's economy and society, the demand for traditional uses of corn stalks has decreased, leading to serious problems of discarding and burning. To achieve precise monitoring and efficient utilization of corn stalks, there is an urgent need for a technological means to quickly, dynamically, and cost-effectively acquire information on the distribution of upright corn stalks in a large area of ​​the field.

[0003] Satellite remote sensing has natural advantages such as wide coverage, high timeliness, and convenient data acquisition, making it an ideal choice for monitoring the spatial distribution of upright corn stalks. By utilizing satellite remote sensing technology, convenient information can be provided for stalk supervision and guiding enterprises in harvesting and utilizing stalks.

[0004] However, similar ground features to upright corn stalks include straw residue, soil, and sparse wheat with low cover in the early growing season. These features appear yellow, bright yellow, or light yellow. In multispectral imagery, especially images containing only visible and near-infrared bands, they appear as "different objects with the same spectrum," causing confusion with upright corn stalks and resulting in low accuracy in remote sensing monitoring of upright corn stalks. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, medium, and program for extracting upright corn stalks, which addresses the shortcomings of low remote sensing monitoring accuracy of upright corn stalks in existing technologies and improves the accuracy of remote sensing monitoring of upright corn stalks.

[0006] This invention provides a method for extracting upright corn stalks, comprising:

[0007] In autumn and winter, the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band were obtained in the corn planting area.

[0008] In a two-dimensional coordinate plane constructed using the reflectance of shortwave infrared band 1 and shortwave infrared band 2, a first distance between the point to be identified and the bare soil line, with the first and second reflectances as coordinates, and a second distance between the point to be identified and the straw residue line are calculated. The ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified. The bare soil line is obtained by fitting the reflectance of the bare soil sample in shortwave infrared band 1 and shortwave infrared band 2; the straw residue line is obtained by fitting the reflectance of the straw residue sample in shortwave infrared band 1 and shortwave infrared band 2.

[0009] The first reflectance and the second reflectance are respectively normalized to their maximum and minimum values, and the larger of the two normalized values ​​is taken as the shadow index corresponding to the ground feature to be identified.

[0010] Based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined.

[0011] In some embodiments, extracting upright corn stalks and determining the area of ​​upright corn stalks based on the remote sensing spectral index and shading index corresponding to the land cover to be identified includes:

[0012] The remote sensing spectral index corresponding to the land cover to be identified is compared with the first threshold of the remote sensing spectral index corresponding to upright corn stalks, the second threshold of the remote sensing spectral index corresponding to straw residues, and the third threshold of the remote sensing spectral index corresponding to bare soil to obtain a first comparison result; wherein the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order.

[0013] The shadow index corresponding to the feature to be identified is compared with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain a second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order;

[0014] Based on the first comparison result and the second comparison result, the type of the land feature to be identified is determined;

[0015] If the type of the feature to be identified is determined to be an upright corn stalk, the area corresponding to the feature to be identified is extracted.

[0016] In some embodiments, determining the type of the feature to be identified based on the first comparison result and the second comparison result includes:

[0017] If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the first threshold of the shadow index and less than the third threshold of the shadow index, then the ground object to be identified is determined to be an upright straw.

[0018] If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the first threshold of the remote sensing spectral index and less than the second threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground object to be identified is determined to be straw residue.

[0019] If the first comparison result is that the remote sensing spectral index corresponding to the ground feature to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground feature to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground feature to be identified is determined to be bare soil.

[0020] In some embodiments, the method further includes:

[0021] Based on the threshold values ​​of the first enhanced vegetation index and the first red edge position index corresponding to corn in early September, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of cultivated land area in early September are masked to obtain the first masked area.

[0022] Based on the second enhanced vegetation index threshold and the second red edge position index threshold corresponding to corn in mid-October, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of the cultivated land area in mid-October are masked to obtain the second masked area.

[0023] The overlapping area of ​​the first mask area and the second mask area is taken as the corn planting area.

[0024] In some embodiments, the method further includes:

[0025] The cultivated land area was determined based on the land cover dataset and the land survey dataset.

[0026] The present invention also provides an upright corn stalk extraction device, comprising:

[0027] The acquisition module is used to acquire the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area during autumn and winter.

[0028] The first calculation module is used to calculate, in a two-dimensional coordinate plane constructed using the reflectance of the shortwave infrared 1 band and the shortwave infrared 2 band, a first distance between the point to be identified and the bare soil line, and a second distance between the point to be identified and the straw residue line, using the first reflectance and the second reflectance as coordinates; the ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified; wherein, the bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and the shortwave infrared 2 band.

[0029] The second calculation module is used to perform maximum and minimum value normalization calculations on the first reflectance and the second reflectance respectively, and take the larger of the two normalized values ​​as the shadow index corresponding to the ground feature to be identified.

[0030] The determination module is used to extract upright corn stalks and determine the area of ​​upright corn stalks based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified.

[0031] In some embodiments, the determining module includes:

[0032] The first comparison unit is used to compare the remote sensing spectral index corresponding to the land cover to be identified with the first threshold of the remote sensing spectral index corresponding to the upright corn stalk, the second threshold of the remote sensing spectral index corresponding to the straw residue, and the third threshold of the remote sensing spectral index corresponding to the bare soil, and to obtain a first comparison result; wherein the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order.

[0033] The second comparison unit is used to compare the shadow index corresponding to the feature to be identified with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain a second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order.

[0034] A determining unit is configured to determine the type of the land feature to be identified based on the first comparison result and the second comparison result;

[0035] The extraction unit is used to extract the area corresponding to the land feature to be identified when it is determined that the type of the land feature to be identified is an upright corn stalk.

[0036] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the upright corn stalk extraction method as described above.

[0037] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the upright corn stalk extraction method as described above.

[0038] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the upright corn stalk extraction method as described above.

[0039] The method, apparatus, equipment, medium, and program products for extracting upright corn stalks provided by this invention construct remote sensing spectral indices and shadow indices by measuring the reflectance of the ground object to be identified in the SWIR1 and SWIR2 bands. The remote sensing spectral indices effectively amplify the spectral differences between upright corn stalks and other ground objects, while the shadow indices effectively compensate for the shortcomings of the remote sensing spectral indices in identifying upright corn stalks in high-coverage areas, thereby enhancing the ability to identify upright corn stalks and improving the accuracy of remote sensing monitoring of upright corn stalks. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a schematic flowchart of the method for extracting upright corn stalks provided by the present invention.

[0042] Figure 2 This is an example diagram of the feature space of the three-part model jointly driven by the remote sensing spectral index and the shadow index provided by the present invention.

[0043] Figure 3 This is a schematic diagram of the structure of the upright corn stalk extraction device provided by the present invention.

[0044] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0045] my country is a major agricultural country with abundant crop straw resources. In the single-cropping system areas of Northeast China and the water-saving crop planting belts in the underground funnel zone of the North China Plain, farmland becomes fallow after corn matures. Some mature corn is directly returned to the field by crushing the corn stalks with agricultural machinery, but a large amount of corn stalks are left standing upright in the field after manual harvesting.

[0046] In the past, farmers used corn stalks standing upright in the fields during autumn and winter as feed for cattle and sheep or as firewood. However, with the development of China's economy and society, the demand for traditional uses of straw has decreased, leading to serious problems of discarding and burning. To achieve precise monitoring and efficient utilization of straw, there is an urgent need for a technology that can quickly, dynamically, and at low cost acquire information on the distribution of standing corn stalks in a wide area of ​​the field. Satellite remote sensing has natural advantages such as wide coverage, high timeliness, and convenient data acquisition, making it an ideal choice for monitoring the spatial distribution of straw. Using satellite remote sensing technology, convenient information can be provided for straw monitoring and guiding enterprises in harvesting and utilizing straw.

[0047] Upright corn stalks are non-photosynthetic vegetation left in the field after corn ears are harvested. They include dried corn stalks, leaves, and terminal flowers, and their chemical composition is mainly cellulose and lignin. In autumn and winter, upright corn stalks that have not been removed from the field are about 1-2 meters tall, with a coverage of 70% or more. Their upright shape and row arrangement provide a relatively obvious shade characteristic when the sun's altitude is low in autumn and winter.

[0048] Similar ground features to upright corn stalks include straw residue, soil, and sparse wheat with low cover in the early growing season. These features appear yellow, bright yellow, or light yellow. In multispectral imagery, especially images containing only visible and near-infrared bands, they appear as "different objects with the same spectrum," causing confusion with upright corn stalks and resulting in low accuracy in remote sensing monitoring of upright corn stalks.

[0049] To address the shortcomings of low remote sensing accuracy in detecting upright corn stalks, this invention provides a method for extracting upright corn stalks. Based on the spectral characteristics of the shortwave infrared band, a remote sensing spectral index and shadow index are constructed that can effectively distinguish between upright corn stalks, stalk residues, and bare soil, thereby improving the remote sensing accuracy of upright corn stalks in autumn and winter.

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0051] Figure 1This is a schematic flowchart of the method for extracting upright corn stalks provided by the present invention, as shown below. Figure 1 As shown, the present invention provides a method for extracting upright corn stalks, comprising the following steps:

[0052] Step 110: In autumn and winter, obtain the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area.

[0053] Specifically, since the reflectance of upright corn stalks in the short-wave infrared (SWIR) band is usually between that of bare soil and high-coverage straw residue, and the shadowing effect caused by the structural characteristics of upright corn stalks makes their reflectance significantly lower than that of straw residue in the same coverage condition, the reflectance of upright corn stalks in the short-wave infrared band has unique spectral response characteristics.

[0054] Field measurements and remote sensing analysis show that upright corn stalks, high-coverage straw residues, and bare soil exhibit a high degree of correlation and regularity in the shortwave infrared 1 (SWIR1) and shortwave infrared 2 (SWIR2) bands. Therefore, during autumn and winter (e.g., October, November, and December), the first reflectance of the target features in SWIR1 and the second reflectance in SWIR2 in the corn-growing area should be obtained.

[0055] The specific acquisition method can be as follows: Obtain medium-to-high resolution remote sensing data (mainly Sentinel-2A / BMSI data) of the corn-growing area. This data can be downloaded through the Google Earth Engine big data cloud platform, including at least one remote sensing image per month in October, November, and December. The images should be atmospherically corrected using FLAASH. Using the medium-to-high resolution remote sensing data of the corn-growing area, obtain the first reflectance of the features to be identified in the corn-growing area in the SWIR1 band and the second reflectance in the SWIR2 band.

[0056] Step 120: In a two-dimensional coordinate plane constructed using the reflectance of shortwave infrared band 1 and shortwave infrared band 2, calculate the first distance between the point to be identified and the bare soil line, and the second distance between the point to be identified and the straw residue line, using the first and second reflectance as coordinates. The ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified. The bare soil line is obtained by fitting the reflectance of the bare soil sample in shortwave infrared band 1 and shortwave infrared band 2; the straw residue line is obtained by fitting the reflectance of the straw residue sample in shortwave infrared band 1 and shortwave infrared band 2.

[0057] Specifically, a two-dimensional coordinate plane is constructed with the reflectance of the SWIR1 band and the reflectance of the SWIR2 band as the horizontal and vertical axes. Within this two-dimensional plane, the reflectance of bare soil samples in the SWIR1 and SWIR2 bands is fitted to obtain the bare soil line; similarly, the reflectance of straw residue samples in the SWIR1 and SWIR2 bands is fitted to obtain the straw residue line. In other words, the bare soil line is obtained by fitting a straight line to the bare soil sample within a Cartesian coordinate system constructed using the reflectance of the shortwave infrared 1 band and shortwave infrared 2 band; the straw residue line is obtained by fitting a straight line to the straw residue sample within a Cartesian coordinate system constructed using the reflectance of the shortwave infrared 1 band and shortwave infrared 2 band.

[0058] Within this two-dimensional plane, points with coordinates based on a first reflectance and a second reflectance are selected as the points to be identified. A first distance is calculated between the point to be identified and the bare soil line, and a second distance is calculated between the point to be identified and the straw residue line. The ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the feature to be identified.

[0059] For example, the expression for the straw residue line is: The expression for the bare soil line is: ,in This represents the x-coordinate value in a two-dimensional coordinate plane with the reflectivity of the SWIR1 band and the reflectivity of the SWIR2 band as the x and y axes. The ordinate value represents the vertical coordinate of the point to be identified in a two-dimensional coordinate plane with the reflectance of the SWIR1 band and the reflectance of the SWIR2 band as the horizontal and vertical axes. The expression for the SUMI remote sensing spectral index corresponding to the land feature to be identified is as follows:

[0060]

[0061] In the formula, The remote sensing spectral indices corresponding to the ground features to be identified. This represents the first distance between the point to be identified and the bare soil line. This represents the second distance between the point to be identified and the straw residue line. Let the slope of the straw residue line be denoted as . The intercept of the straw residue line, The slope of the bare soil line. The intercept of the bare soil line. The x-coordinate value of the first reflectance in a two-dimensional coordinate plane with the reflectance of the SWIR1 band and the reflectance of the SWIR2 band as the x and y axes. The value of the second reflectance is the ordinate value in a two-dimensional coordinate plane with the reflectance of the SWIR1 band and the reflectance of the SWIR2 band as the horizontal and vertical axes.

[0062] Remote sensing spectral indices can measure the positional relationship of a target feature relative to the bare soil line and the straw residue line in a two-dimensional coordinate plane with the reflectance of SWIR1 and SWIR2 as the horizontal and vertical axes. If the positional relationship is close to the bare soil line, the target feature is preliminarily identified as bare soil; if the positional relationship is close to the straw residue line, the target feature is preliminarily identified as straw residue; if the positional relationship is neither close to the bare soil line nor close to the straw residue line, the target feature is preliminarily identified as upright corn stalks or straw residue with similar coverage to upright corn stalks. This significantly enhances the spectral separability between upright corn stalks and background features (bare soil, low-coverage straw residue).

[0063] Step 130: Perform maximum and minimum value normalization calculations on the first reflectance and the second reflectance respectively, and take the larger of the two normalized values ​​as the shadow index corresponding to the ground feature to be identified.

[0064] Specifically, while remote sensing spectral indices amplify the spectral differences between upright corn stalks and other ground features, they can lead to oversaturation of upright corn stalks, making it impossible to distinguish them from flat-lying stalk residues with similar coverage.

[0065] Under the same illumination but different levels of shading, the spectral curves of straw with the same coverage exhibit a highly consistent trend, and pixels with higher overall brightness also show a higher amplitude of spectral reflectance. This phenomenon is particularly pronounced in the SWIR band, indicating that short-wave infrared is quite sensitive to the degree of shading and can be used to characterize the shading caused by the canopy structure of upright corn stalks, distinguishing between upright corn stalks and straw residue under the same coverage. Upright corn stalks exhibit a stronger shading effect in the short-wave infrared region due to their canopy structure, resulting in lower reflectance than straw residue with the same coverage.

[0066] The reflectance of both SWIR1 and SWIR2 bands increases with increasing illumination, and can be used to quantitatively characterize the proportion of shadows in upright corn stalks. To address the oversaturation problem of remote sensing spectral indices in distinguishing upright corn stalks from straw residue with similar coverage, two shadow indices were constructed based on the reflectance of SWIR1 and SWIR2 bands. These two indices were then optimized to obtain a more sensitive shadow index to aid in distinguishing upright corn stalks from flat-lying straw residue. The shadow index effectively compensates for the shortcomings of remote sensing spectral indices in identifying upright corn stalks in highly covered areas, enhancing the ability to distinguish between upright corn stalks, straw residue, and bare soil. The process of constructing the shadow index is as follows:

[0067] The first reflectivity and the second reflectivity are respectively normalized to their maximum and minimum values ​​to obtain the first normalized value corresponding to the first reflectivity and the second normalized value corresponding to the second reflectivity.

[0068] The expression for the first normalized value is shown below:

[0069]

[0070] In the formula, The first normalized value, The first reflectivity, The minimum reflectance of all pixels in the SWIR1 band in a remote sensing image containing upright corn stalks, straw residue, and bare soil. The maximum reflectance of all pixels in the SWIR1 band in a remote sensing image containing upright corn stalks, straw residue, and bare soil.

[0071] The expression for the second normalized value is as follows:

[0072]

[0073] In the formula, This is the second normalized value. The second reflectivity, The minimum reflectance of all pixels in the SWIR2 band in a remote sensing image containing upright corn stalks, straw residue, and bare soil. The maximum reflectance of all pixels in the SWIR2 band in a remote sensing image containing upright corn stalks, straw residue, and bare soil.

[0074] Normalization of reflectance can eliminate differences in illumination and environment, making the spectral characteristics of shadow areas more prominent, thus effectively reflecting the presence of shadows. Therefore, the first normalized value is... Second normalized value The larger value is used as the shadow index corresponding to the feature to be identified. The expression for the shadow index corresponding to the feature to be identified is as follows:

[0075]

[0076] In the formula, The shadow index corresponding to the feature to be identified. The first normalized value, This is the second normalized value.

[0077] Step 140: Based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified, extract the upright corn stalks and determine the area of ​​upright corn stalks.

[0078] Specifically, the remote sensing spectral index and shadow index are combined to identify the land features to be identified, and to determine whether the land features to be identified are upright stalks, stalk residues or bare soil, so as to extract upright corn stalks and identify the area of ​​upright corn stalks.

[0079] The method for extracting upright corn stalks provided by this invention constructs remote sensing spectral indices and shadow indices by measuring the reflectance of the ground object to be identified in the SWIR1 and SWIR2 bands. The remote sensing spectral indices effectively amplify the spectral differences between upright corn stalks and other ground objects, while the shadow indices effectively compensate for the shortcomings of the remote sensing spectral indices in identifying upright corn stalks in high-coverage areas, thereby enhancing the ability to identify upright corn stalks and improving the accuracy of remote sensing monitoring of upright corn stalks.

[0080] In some embodiments, based on the remote sensing spectral index and shading index corresponding to the land cover to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined, including:

[0081] The remote sensing spectral index corresponding to the land cover to be identified is compared with the first threshold of the remote sensing spectral index corresponding to upright corn stalks, the second threshold of the remote sensing spectral index corresponding to straw residues, and the third threshold of the remote sensing spectral index corresponding to bare soil to obtain the first comparison result; wherein, the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order.

[0082] The shadow index corresponding to the feature to be identified is compared with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain the second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order;

[0083] Based on the first comparison result and the second comparison result, determine the type of the land feature to be identified;

[0084] If the type of feature to be identified is determined to be upright corn stalks, the feature to be identified is extracted and the area of ​​upright corn stalks is determined.

[0085] Specifically, for the three components of upright corn stalks, straw residue, and bare soil in multispectral images, a three-part model feature space jointly driven by remote sensing spectral indices and shadow indices is constructed based on the linear mixed spectral analysis theory. In the system constructed by remote sensing spectral indices and shadow indices, upright stalks, straw residue, and bare soil are concentrated in three locations of the system, and the line connecting the center points of the three locations constitutes the three-part model, thereby distinguishing upright stalks, straw residue, and bare soil.

[0086] Figure 2 This is an example diagram of the feature space of the three-part model jointly driven by the remote sensing spectral index and the shadow index provided by the present invention, as shown in the figure. Figure 2As shown, with the shadow index as the horizontal axis and the remote sensing spectral index as the vertical axis, the first threshold S1 of the remote sensing spectral index corresponding to upright corn stalks, the second threshold S2 of the remote sensing spectral index corresponding to straw residue, and the third threshold S3 of the remote sensing spectral index corresponding to bare soil are determined. The first threshold N1 of the shadow index corresponding to upright corn stalks, the second threshold N2 of the shadow index corresponding to straw residue, and the third threshold N3 of the shadow index corresponding to bare soil are also determined.

[0087] The first, second, and third thresholds of the remote sensing spectral index are denoted as S1, S2, and S3, respectively; the first, second, and third thresholds of the shadow index are denoted as N1, N2, and N3, respectively. Figure 2 As shown, the values ​​of the three remote sensing spectral index thresholds, arranged from smallest to largest, are: Remote Sensing Spectral Index Third Threshold S3, Remote Sensing Spectral Index First Threshold S1, and Remote Sensing Spectral Index Second Threshold S2. The values ​​of the three shadow index thresholds, arranged from smallest to largest, are: Shadow Index First Threshold N1, Shadow Index Third Threshold N3, and Shadow Index Second Threshold N2.

[0088] The remote sensing spectral index corresponding to the ground feature to be identified is compared with the first threshold S1, the second threshold S2, and the third threshold S3 of the remote sensing spectral index to obtain the first comparison result. The shadow index corresponding to the ground feature to be identified is compared with the first threshold N1, the second threshold N2, and the third threshold N3 of the shadow index to obtain the second comparison result.

[0089] Based on the first and second comparison results, the type of the feature to be identified is determined, specifically whether it is upright corn stalks, straw residue, or bare soil. If the type of feature to be identified is upright corn stalks, the feature is extracted to determine the area containing the upright corn stalks.

[0090] The method for extracting upright corn stalks provided by this invention compares the remote sensing spectral index corresponding to the feature to be identified with the remote sensing spectral index thresholds corresponding to upright corn stalks, stalk residue, and bare soil, respectively. It also compares the shadow index corresponding to the feature to be identified with the shadow index thresholds corresponding to upright corn stalks, stalk residue, and bare soil, respectively. Based on the first comparison result and the second comparison result, the type of the feature to be identified is determined, thereby improving the accuracy of feature type identification.

[0091] In some embodiments, determining the type of the feature to be identified based on a first comparison result and a second comparison result includes:

[0092] When the first comparison result is that the remote sensing spectral index corresponding to the object to be recognized is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the object to be recognized is greater than the first threshold of the shadow index and less than the third threshold of the shadow index, determine that the object to be recognized is standing straw.

[0093] When the first comparison result is that the remote sensing spectral index corresponding to the object to be recognized is greater than the first threshold of the remote sensing spectral index and less than the second threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the object to be recognized is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, determine that the object to be recognized is straw residue.

[0094] When the first comparison result is that the remote sensing spectral index corresponding to the object to be recognized is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the object to be recognized is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, determine that the object to be recognized is bare soil.

[0095] Specifically, when S3 < SUMI < S1 and N1 < NUSI < N3, determine that the type of the object to be recognized is standing corn straw. When S1 < SUMI < S2 and N3 < NUSI < N2, determine that the type of the object to be recognized is straw residue. When S3 < SUMI < S1 and N3 < NUSI < N2, determine that the type of the object to be recognized is bare soil.

[0096] The method for extracting standing corn straw provided by the present invention gradually establishes classification rules by using the method of multi-index threshold division to ensure the accuracy and robustness of the classification results.

[0097] In some embodiments, the method for extracting standing corn straw provided by the present invention further includes:

[0098] According to the enhanced vegetation index threshold and red edge position index threshold corresponding to early September of corn, mask the enhanced vegetation index and red edge position index corresponding to the remote sensing image of the cultivated area in early September to obtain the first masked area;

[0099] According to the enhanced vegetation index threshold and red edge position index threshold corresponding to mid-October of corn, mask the enhanced vegetation index and red edge position index corresponding to the remote sensing image of the cultivated area in mid-October to obtain the second masked area;

[0100] Take the overlapping area of the first masked area and the second masked area as the corn planting area.

[0101] Specifically, since upright corn stalks originate from summer-planted corn areas, the summer corn planting range can be used as prior knowledge for extracting farmland stalks in autumn and winter. This invention selects the Enhanced Vegetation Index (EVI) and the Red Edge Position Index (REP) as the main spectral indices for extraction from corn-growing areas. The Enhanced Vegetation Index is used to assess vegetation productivity, reduce soil reflectance and air pollution, enhance the contrast between bare land and vegetation, and improve extraction accuracy. The Red Edge Position Index is defined as the wavelength of the maximum slope (inflection point) in the red edge region (680-780nm) of the vegetation spectrum.

[0102] The formula for calculating the Enhanced Vegetation Index is as follows:

[0103]

[0104] In the formula, To enhance the vegetation index, This is the gain factor (usually 2.5). The reflectance in the near-infrared band (865nm) is... The reflectivity is in the red band (665nm). and This is the atmospheric correction factor. The reflectivity is in the blue band (496nm). This is the soil adjustment coefficient (usually 1).

[0105] The formula for calculating the red-edge position index is as follows:

[0106]

[0107] In the formula, The red-edge position index. The reflectance is for the red-edge 3 band (780nm). The reflectivity is in the red band (665nm). The reflectance is for the red-edge band 1 (705nm). The reflectance is the red-edge band 2 (740nm).

[0108] In early September, corn stalks have relatively high chlorophyll content, resulting in a strong green vegetation signal and a significant contrast with built-up land. By mid-October, the green woodland contrasts sharply with the withered, upright corn stalks, making the distinction between the two highly pronounced. Therefore, to extract corn-growing areas from cultivated land, remote sensing images of cultivated land from early September and mid-October were used to calculate the Enhanced Vegetation Index and the Red Edge Location Index, respectively.

[0109] Using the enhanced vegetation index (EVI) threshold and red-edge position index threshold corresponding to corn in early September, the EVI and red-edge position index of remote sensing images of cultivated land areas in early September are masked to obtain the first masked area. After removing built-up land through this first masking process, the first masked area includes corn stalk areas and woodland. For example, the EVI threshold is 0.5 and the red-edge position index threshold is 725, which can be adjusted according to the actual situation.

[0110] Using the enhanced vegetation index (VEI) threshold and red-edge position index threshold corresponding to corn in mid-October, the VEI and red-edge position indices corresponding to remote sensing images of cultivated land areas in mid-October were masked to obtain a second masked region. Woodland was then removed through this second masking process, and the second masked region includes corn stalk areas.

[0111] To improve the accuracy of corn planting area extraction, the overlapping area of ​​the first mask area and the second mask area is taken as the corn planting area.

[0112] The method for extracting upright corn stalks provided by this invention enhances the targeting and regional positioning capabilities of upright corn stalk extraction. It utilizes remote sensing images from early September and mid-October, when corn is in its vigorous growth period, to calculate the Enhanced Vegetation Index (EVI) and Red Edge Position Index (REP). By setting thresholds, a corn planting range mask is constructed to spatially limit the area where upright corn stalks may appear, thereby reducing the risk of misidentification.

[0113] In some embodiments, the method for extracting upright corn stalks provided by the present invention further includes:

[0114] Farmland areas were identified based on land cover datasets and land survey datasets.

[0115] Specifically, land cover datasets are primarily used to extract initial arable land information and to remove land types such as sandy land, roads, residential areas, and water bodies that affect the extraction of information on upright corn stalk land. Land survey datasets, such as the dataset from the Third National Land Survey, contain detailed land resource data obtained through field surveys or high-resolution image interpretation. Combining land cover datasets and land survey datasets helps to identify arable land areas.

[0116] The method for extracting upright corn stalks provided by this invention determines cultivated land areas through land cover datasets and land survey datasets, thereby improving the accuracy of cultivated land area extraction. In order to better extract corn planting areas in the future, non-target feature areas are removed.

[0117] The upright corn stalk extraction device provided by the present invention is described below. The upright corn stalk extraction device described below and the upright corn stalk extraction method described above can be referred to in correspondence.

[0118] Figure 3 This is a schematic diagram of the structure of the upright corn stalk extraction device provided by the present invention, as shown below. Figure 3 As shown, the present invention provides an upright corn stalk extraction device, comprising:

[0119] The acquisition module 310 is used to acquire the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area during autumn and winter.

[0120] The first calculation module 320 is used to calculate, in a two-dimensional coordinate plane constructed using the reflectance of the shortwave infrared 1 band and the shortwave infrared 2 band, a first distance between the point to be identified and the bare soil line, and a second distance between the point to be identified and the straw residue line, using the first reflectance and the second reflectance as coordinates; and to use the ratio of the first distance to the second distance as the remote sensing spectral index corresponding to the ground feature to be identified; wherein, the bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and the shortwave infrared 2 band; and the straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and the shortwave infrared 2 band.

[0121] The second calculation module 330 is used to perform maximum and minimum value normalization calculations on the first reflectance and the second reflectance respectively, and take the larger of the two normalized values ​​as the shadow index corresponding to the ground feature to be identified.

[0122] The determination module 340 is used to extract upright corn stalks and determine the area of ​​upright corn stalks based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified.

[0123] In some embodiments, the determining module 340 includes:

[0124] The first comparison unit is used to compare the remote sensing spectral index corresponding to the land cover to be identified with the first threshold of the remote sensing spectral index corresponding to the upright corn stalk, the second threshold of the remote sensing spectral index corresponding to the straw residue, and the third threshold of the remote sensing spectral index corresponding to the bare soil, and to obtain a first comparison result; wherein the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order.

[0125] The second comparison unit is used to compare the shadow index corresponding to the feature to be identified with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain a second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order.

[0126] A determining unit is configured to determine the type of the land feature to be identified based on the first comparison result and the second comparison result;

[0127] An extraction unit is used to extract the feature to be identified and determine the area of ​​the upright corn stalks when it is determined that the type of the feature to be identified is an upright corn stalk.

[0128] In some embodiments, the determining unit is used to:

[0129] If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the first threshold of the shadow index and less than the third threshold of the shadow index, then the ground object to be identified is determined to be an upright straw.

[0130] If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the first threshold of the remote sensing spectral index and less than the second threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground object to be identified is determined to be straw residue.

[0131] If the first comparison result is that the remote sensing spectral index corresponding to the ground feature to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground feature to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground feature to be identified is determined to be bare soil.

[0132] In some embodiments, the apparatus further includes a first extraction module, the first extraction module being configured to:

[0133] Based on the enhanced vegetation index threshold and red edge position index threshold corresponding to corn in early September, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of cultivated land area in early September are masked to obtain the first masked area.

[0134] Based on the enhanced vegetation index threshold and the red edge position index threshold corresponding to corn in mid-October, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of the cultivated land area in mid-October are masked to obtain the second masked area.

[0135] The overlapping area of ​​the first mask area and the second mask area is taken as the corn planting area.

[0136] In some embodiments, the apparatus further includes a second extraction module, the second extraction module being used for:

[0137] The cultivated land area was determined based on the land cover dataset and the land survey dataset.

[0138] It should be noted that the upright corn stalk extraction device provided by the present invention can realize all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0139] Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a method for extracting upright corn stalks. This method includes: obtaining, in autumn and winter, the first reflectance of the target feature in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area; calculating, in a two-dimensional coordinate plane constructed using the reflectance of the shortwave infrared 1 band and the shortwave infrared 2 band, a first distance between the target point and the bare soil line, and a second distance between the target point and the straw residue line, using the first distance and the second distance as coordinates; and using the ratio of the first distance to the second distance as the... The remote sensing spectral index corresponding to the land feature to be identified is described; wherein, the bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the first reflectance and the second reflectance are respectively normalized to their maximum and minimum values, and the larger of the two normalized values ​​is taken as the shadow index corresponding to the land feature to be identified; based on the remote sensing spectral index and shadow index corresponding to the land feature to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined.

[0140] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0141] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the upright corn stalk extraction method provided by the above methods. The method includes: obtaining the first reflectance of the target feature in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area during autumn and winter; calculating, in a two-dimensional coordinate plane constructed with the reflectance of the shortwave infrared 1 band and the reflectance of the shortwave infrared 2 band, the first distance between the target feature and the bare soil line, and the distance between the target feature and the bare soil line, using the first reflectance and the second reflectance as coordinates. The second distance between the straw residue lines is used as the ratio of the first distance to the second distance, which is taken as the remote sensing spectral index corresponding to the feature to be identified. The bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and shortwave infrared 2 band. The straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and shortwave infrared 2 band. The first and second reflectances are normalized to their maximum and minimum values, respectively, and the larger of the two normalized values ​​is taken as the shadow index corresponding to the feature to be identified. Based on the remote sensing spectral index and shadow index corresponding to the feature to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined.

[0142] Furthermore, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for extracting upright corn stalks provided by the methods described above. This method includes: obtaining, in autumn and winter, the first reflectance of the feature to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in a corn-growing area; calculating, in a two-dimensional coordinate plane constructed using the reflectance of the shortwave infrared 1 band and the shortwave infrared 2 band, a first distance between the point to be identified, with the first reflectance and the second reflectance as coordinates, and a second distance between the point to be identified and the line of bare soil, and the line of straw residue, and then... The ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified; wherein, the bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the first reflectance and the second reflectance are respectively normalized to their maximum and minimum values, and the larger of the two normalized values ​​is used as the shadow index corresponding to the ground feature to be identified; based on the remote sensing spectral index and shadow index corresponding to the ground feature to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined.

[0143] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0144] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0145] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting from upright corn stalks, characterized in that, include: In autumn and winter, the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band were obtained in the corn planting area. In a two-dimensional coordinate plane constructed using the reflectance of shortwave infrared band 1 and shortwave infrared band 2, a first distance between the point to be identified and the bare soil line, with the first and second reflectances as coordinates, and a second distance between the point to be identified and the straw residue line are calculated. The ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified. The bare soil line is obtained by fitting the reflectance of the bare soil sample in shortwave infrared band 1 and shortwave infrared band 2; the straw residue line is obtained by fitting the reflectance of the straw residue sample in shortwave infrared band 1 and shortwave infrared band 2. The first reflectance and the second reflectance are respectively normalized to their maximum and minimum values, and the larger of the two normalized values ​​is taken as the shadow index corresponding to the ground feature to be identified. Based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified, upright corn stalks are extracted and the area of ​​upright corn stalks is determined.

2. The method for extracting upright corn stalks according to claim 1, characterized in that, The step of extracting upright corn stalks and determining the area of ​​upright corn stalks based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified includes: The remote sensing spectral index corresponding to the land cover to be identified is compared with the first threshold of the remote sensing spectral index corresponding to upright corn stalks, the second threshold of the remote sensing spectral index corresponding to straw residues, and the third threshold of the remote sensing spectral index corresponding to bare soil to obtain a first comparison result; wherein the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order. The shadow index corresponding to the feature to be identified is compared with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain a second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order; Based on the first comparison result and the second comparison result, the type of the land feature to be identified is determined; If the type of the feature to be identified is determined to be an upright corn stalk, the feature to be identified is extracted and the area of ​​the upright corn stalk is determined.

3. The method for extracting upright corn stalks according to claim 2, characterized in that, Determining the type of the feature to be identified based on the first comparison result and the second comparison result includes: If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the first threshold of the shadow index and less than the third threshold of the shadow index, then the ground object to be identified is determined to be an upright straw. If the first comparison result is that the remote sensing spectral index corresponding to the ground object to be identified is greater than the first threshold of the remote sensing spectral index and less than the second threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground object to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground object to be identified is determined to be straw residue. If the first comparison result is that the remote sensing spectral index corresponding to the ground feature to be identified is greater than the third threshold of the remote sensing spectral index and less than the first threshold of the remote sensing spectral index, and the second comparison result is that the shadow index corresponding to the ground feature to be identified is greater than the third threshold of the shadow index and less than the second threshold of the shadow index, then the ground feature to be identified is determined to be bare soil.

4. The method for extracting upright corn stalks according to claim 1, characterized in that, The method further includes: Based on the enhanced vegetation index threshold and red edge position index threshold corresponding to corn in early September, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of cultivated land area in early September are masked to obtain the first masked area. Based on the enhanced vegetation index threshold and the red edge position index threshold corresponding to corn in mid-October, the enhanced vegetation index and red edge position index corresponding to the remote sensing image of the cultivated land area in mid-October are masked to obtain the second masked area. The overlapping area of ​​the first mask area and the second mask area is taken as the corn planting area.

5. The method for extracting upright corn stalks according to claim 4, characterized in that, The method further includes: The cultivated land area was determined based on the land cover dataset and the land survey dataset.

6. A vertical corn stalk extraction device, characterized in that, include: The acquisition module is used to acquire the first reflectance of the ground features to be identified in the shortwave infrared 1 band and the second reflectance in the shortwave infrared 2 band in the corn planting area during autumn and winter. The first calculation module is used to calculate, in a two-dimensional coordinate plane constructed using the reflectance of the shortwave infrared 1 band and the shortwave infrared 2 band, a first distance between the point to be identified and the bare soil line, and a second distance between the point to be identified and the straw residue line, using the first reflectance and the second reflectance as coordinates; the ratio of the first distance to the second distance is used as the remote sensing spectral index corresponding to the ground feature to be identified; wherein, the bare soil line is obtained by fitting the reflectance of the bare soil sample in the shortwave infrared 1 band and the shortwave infrared 2 band; the straw residue line is obtained by fitting the reflectance of the straw residue sample in the shortwave infrared 1 band and the shortwave infrared 2 band. The second calculation module is used to perform maximum and minimum value normalization calculations on the first reflectance and the second reflectance respectively, and take the larger of the two normalized values ​​as the shadow index corresponding to the ground feature to be identified. The determination module is used to extract upright corn stalks and determine the area of ​​upright corn stalks based on the remote sensing spectral index and shadow index corresponding to the land cover to be identified.

7. The upright corn stalk extraction device according to claim 6, characterized in that, The determining module includes: The first comparison unit is used to compare the remote sensing spectral index corresponding to the land cover to be identified with the first threshold of the remote sensing spectral index corresponding to the upright corn stalk, the second threshold of the remote sensing spectral index corresponding to the straw residue, and the third threshold of the remote sensing spectral index corresponding to the bare soil, and to obtain a first comparison result; wherein the values ​​of the third threshold of the remote sensing spectral index, the first threshold of the remote sensing spectral index, and the second threshold of the remote sensing spectral index are arranged in ascending order. The second comparison unit is used to compare the shadow index corresponding to the feature to be identified with the first threshold of shadow index corresponding to upright corn stalks, the second threshold of shadow index corresponding to straw residue, and the third threshold of shadow index corresponding to bare soil to obtain a second comparison result; the values ​​of the first threshold of shadow index, the third threshold of shadow index, and the second threshold of shadow index are arranged in ascending order. A determining unit is configured to determine the type of the land feature to be identified based on the first comparison result and the second comparison result; An extraction unit is used to extract the feature to be identified and determine the area of ​​the upright corn stalks when it is determined that the type of the feature to be identified is an upright corn stalk.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the upright corn stalk extraction method as described in any one of claims 1 to 5.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for extracting upright corn stalks as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for extracting upright corn stalks as described in any one of claims 1 to 5.

Images