Planting area extraction method and device based on satellite data, equipment and medium

Through the acquisition and processing of hyperspectral satellite data, combined with tree growth characteristics and spectral difference information, the tree planting area is identified, and the problem of mis-extraction in the existing technology is solved, and the precise extraction of tree planting area is achieved.

CN119048914APending Publication Date: 2024-11-29ZHUHAI ORBIT SATELLITE BIG DATA CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411209927.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

When obtaining information about tree planting areas, the prior art cannot effectively consider the geographical environment and growth phenological period of the trees, resulting in the incorrect extraction of non-tree planting areas into tree planting areas, resulting in the extraction area being greater than the actual area.

Method used

Using a satellite data-based method, through hyperspectral satellite data acquisition and preprocessing, tree growth characteristics and spectral differences are obtained, tree flowering period is determined, spectral index and texture characteristics are constructed, extracted samples and classification systems are established, and tree planting areas are identified.

Benefits of technology

Accurate data acquisition in tree planting areas is realized, the extraction accuracy is improved, the error extraction of non-tree areas is avoided, and the accuracy of the extraction area is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048914A_ABST
    Figure CN119048914A_ABST
Patent Text Reader

Abstract

The invention provides a planting area extraction method based on satellite data. The method comprises the steps that a hyperspectral satellite data collection task and a shooting plan are made, image information of a planting area is obtained and preprocessed, tree growth characteristic information and spectral difference information of the planting area are obtained, and the flowering phase of trees and hyperspectral satellite data are determined; obtaining topographic features of the planting area, extracting spectral index features and texture features of the flowering phase of the tree, constructing a feature data set, and establishing an extraction sample of the planting area and a classification system to identify the tree planting area in the image information; constructing and extracting a plurality of tree flowering phase characteristic data sets to obtain a peach blossom extraction result; and obtaining an intersection of a result union set of the peach blossom extraction results and a preset tree flowering phase characteristic data set, and obtaining a tree distribution condition of the tree planting area. According to the technical scheme of the embodiment, accurate data of the tree planting area can be effectively obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of satellite remote sensing technology, and in particular to a method, device, equipment and medium for extracting a planting area based on satellite data. Background Art

[0002] In the prior art, the traditional way of obtaining tree spatial distribution information requires field measurements by professionals. However, during the measurement process, the characteristics of satellite data and the feature data selection are single, and the influence of the geographical environment and the growth phenological period of trees cannot be effectively considered. When extracting samples of tree planting areas and non-tree planting areas, non-tree planting areas are mistakenly extracted as tree planting areas, resulting in the area of ​​the extracted tree planting area being larger than the actual area of ​​the tree planting area. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a planting area extraction method, device, equipment and medium based on satellite data, which can obtain accurate data of tree planting areas.

[0004] In a first aspect, an embodiment of the present invention provides a method for extracting a planting area based on satellite data, comprising:

[0005] Develop hyperspectral satellite data collection tasks and shooting plans to obtain image information of the planting area;

[0006] Performing a preprocessing operation on the image information to obtain tree growth characteristic information and spectral difference information in the planting area;

[0007] Determine the tree flowering period and hyperspectral satellite data according to the tree growth characteristic information and the spectral difference information;

[0008] Acquire the topographic features of the planting area, extract the spectral index features and texture features of the tree flowering period according to the hyperspectral satellite data, and construct the hyperspectral satellite data, the spectral index features, the topographic features and the texture features into a feature data set;

[0009] Establishing an extraction sample and a classification system for the planting area according to the hyperspectral satellite data and the feature data set, and identifying the tree planting area in the image information according to the extraction sample and the classification system;

[0010] Constructing a plurality of tree flowering period characteristic data sets according to the hyperspectral satellite data, extracting each of the tree flowering period characteristic data sets to obtain peach blossom extraction results;

[0011] The intersection of the result union of the peach blossom extraction result and a preset tree flowering period feature data set is obtained to obtain the tree distribution situation in the tree planting area.

[0012] In some embodiments of the present invention, the preprocessing operation on the image information includes:

[0013] Acquire mission planning information, orbit prediction information and data reception information according to the hyperspectral satellite acquisition mission and the shooting plan;

[0014] Decoding the hyperspectral satellite data acquisition mission according to the mission planning information, the orbit prediction information and the data reception information to generate auxiliary data, relative radiation correction data and RPC parameters;

[0015] Acquire radiation calibration information and correction information according to the auxiliary data, the relative radiation correction data and the RPC parameters, wherein the correction information at least includes atmospheric correction and orthorectification;

[0016] The original grayscale value obtained by the hyperspectral satellite data acquisition task is converted into a radiation brightness value according to the radiation calibration information to determine the target radiation value of the hyperspectral satellite;

[0017] The radiation brightness value and deformation data of the hyperspectral satellite are obtained, the atmospheric correction converts the radiation brightness value into the actual reflectivity of the surface, and the orthorectification generates a planar orthophoto according to the deformation data.

[0018] In some embodiments of the present invention, determining the tree flowering period and the hyperspectral satellite data according to the tree growth characteristic information and the spectral difference information includes:

[0019] Acquiring terrain difference information and tree variety information of the planting area;

[0020] Determining the phenological information of the tree according to the terrain difference information and the variety information;

[0021] The image information selects the planting area that is in the flowering period of the tree according to the phenological information.

[0022] In some embodiments of the present invention, constructing the hyperspectral satellite data, the spectral index feature, the terrain feature and the texture feature into a feature data set includes:

[0023] Obtain near infrared band reflectivity, red light reflectivity, blue light reflectivity, green light reflectivity and soil adjustment coefficient of the hyperspectral satellite data;

[0024] A spectral index is calculated according to the near-infrared band reflectivity, the red light reflectivity, the blue light reflectivity, the green light reflectivity and the soil adjustment coefficient, wherein the spectral index at least includes a normalized vegetation index.

[0025] In some embodiments of the present invention, the spectral index is calculated according to the near-infrared band reflectivity, the red light reflectivity, the blue light reflectivity, the green light reflectivity and the soil adjustment coefficient, and is obtained according to the following formula:

[0026] NDVI=(ρ NIR -ρ RED ) / (ρ NIR -ρ REd );

[0027] EVI=2.5*[(ρ NIR -ρ ReD ) / (ρ NIR +6ρ RED -7.5ρ BLUE )+1];

[0028] ARVI=[ρ NIR -(2ρ RED -ρ BLUE )] / [ρ NIR +(2ρ RED -ρ BLUE )];

[0029] SAVI=[(ρ NIR -ρ RED )*(1+L)] / [ρ NIR +ρ RED +L];

[0030] RG=ρ RED / ρ GREEN ;

[0031] RPI=ρ BLUE / ρ NIR ;

[0032] VI2=[ρ NIR -ρ RED / (ρ NIR +ρ RED )]*ρ NIR =NDVI*ρ NIR ;

[0033] Pseudo CRI1 = 1 / ρ BLUE -1 / ρ GREEN ;

[0034] Pseudo CRI2 = 1 / ρ GREEN -1 / ρRED ;

[0035] Among them, NDVI is the normalized vegetation index, EVI is the enhanced vegetation index, ARVI is the atmospheric impedance vegetation index, SAVI is the soil adjusted vegetation index, RG is the red-green ratio index, RPI is the ratio residential index, VI2 is the first vegetation index, CRI1 is the first pseudo-carotenoid reflectance index, CRI2 is the second pseudo-carotenoid reflectance index, ρ NIR is the infrared band reflectivity, ρ RED is the red light reflectivity, ρ BLUE is the blue light reflectivity, ρ GREEN is the green light reflectance and L is the soil adjustment coefficient.

[0036] In some embodiments of the present invention, the extracting samples and the classification system of the planting area are established according to the hyperspectral satellite data and the feature data set, including:

[0037] Acquiring classification object information of the feature data set, and establishing interpretation marks in the hyperspectral satellite data according to the terrain features;

[0038] Selecting pure pixels in the image information according to the interpretation mark, and producing ground object classification samples according to the pure pixels;

[0039] The classification system is constructed by acquiring feature information of the ground object classification samples in the hyperspectral satellite data at different time periods.

[0040] In some embodiments of the present invention, after extracting the spectral index characteristics and texture characteristics of the tree flowering period according to the hyperspectral satellite data, the method further includes:

[0041] Acquire surface roughness information in the image information, and when the surface roughness meets the texture feature of the planting area, acquire multiple band information of the spectral index feature;

[0042] The average of the plurality of band information is calculated according to the hyperspectral data.

[0043] In a second aspect, an embodiment of the present invention provides a planting area extraction device based on satellite data, comprising at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the planting area extraction method based on satellite data as described in the first aspect above.

[0044] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a planting area extraction device based on satellite data as described in the second aspect above.

[0045] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are used to execute the method for extracting planting areas based on satellite data as described in the first aspect above.

[0046] The planting area extraction method based on satellite data according to the embodiment of the present invention has at least the following beneficial effects:

[0047] Formulate hyperspectral satellite data collection tasks and shooting plans to obtain image information of the planting area; perform preprocessing operations on the image information to obtain tree growth characteristic information and spectral difference information in the planting area; determine the tree flowering period and hyperspectral satellite data based on the tree growth characteristic information and spectral difference information; obtain the terrain characteristics of the planting area, extract the spectral index characteristics and texture characteristics of the tree flowering period based on the hyperspectral satellite data, and construct the hyperspectral satellite data, spectral index characteristics, terrain characteristics and texture characteristics into a feature data set; establish extraction samples and classification systems for the planting area based on the hyperspectral satellite data and the feature data set, and identify the tree planting area in the image information based on the extraction samples and classification system; construct multiple tree flowering period feature data sets based on the hyperspectral satellite data, extract each tree flowering period feature data set, and obtain peach blossom extraction results; obtain the intersection of the result union of the peach blossom extraction results and the preset tree flowering period feature data set to obtain the tree distribution situation in the tree planting area. According to the technical solution of this embodiment, accurate data of the tree planting area can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of a method for extracting planting areas based on satellite data provided by an embodiment of the present invention;

[0049] Figure 2 is a flow chart of performing preprocessing operations on image information provided by an embodiment of the present invention;

[0050] Figure 3 It is a flow chart of determining peach tree flowering period and hyperspectral satellite data according to peach tree growth characteristic information and spectral difference information provided by an embodiment of the present invention;

[0051] Figure 4 A flowchart of constructing a feature data set using hyperspectral satellite data, spectral index features, terrain features, and texture features is provided by an embodiment of the present invention;

[0052] Figure 5A flowchart of an extraction sample and classification system for establishing a planting area based on hyperspectral satellite data and feature data sets provided by an embodiment of the present invention;

[0053] Figure 6 It is a flow chart after extracting spectral index characteristics and texture characteristics of peach tree flowering period according to hyperspectral satellite data provided by an embodiment of the present invention;

[0054] Figure 7 It is a schematic diagram of remote sensing image interpretation signs provided by an embodiment of the present invention;

[0055] Figure 8 It is a structural diagram of a planting area extraction device based on satellite data provided by another embodiment of the present invention. DETAILED DESCRIPTION

[0056] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0057] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0058] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0059] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0060] The embodiment of the present invention provides a method for extracting planting areas based on satellite data, including formulating a hyperspectral satellite data collection task and a shooting plan to obtain image information of the planting area; performing preprocessing operations on the image information to obtain tree growth characteristic information and spectral difference information of the planting area; determining the tree flowering period and hyperspectral satellite data according to the tree growth characteristic information and spectral difference information; obtaining the terrain characteristics of the planting area, extracting the spectral index characteristics and texture characteristics of the tree flowering period according to the hyperspectral satellite data, and constructing the hyperspectral satellite data, spectral index characteristics, terrain characteristics and texture characteristics as a feature data set; establishing an extraction sample and a classification system of the planting area according to the hyperspectral satellite data and the feature data set, and identifying the tree planting area in the image information according to the extraction sample and the classification system; constructing multiple tree flowering period feature data sets according to the hyperspectral satellite data, extracting each tree flowering period feature data set to obtain a peach blossom extraction result; obtaining the intersection of the result union of the peach blossom extraction result and the preset tree flowering period feature data set to obtain the tree distribution situation in the tree planting area. According to the technical solution of this embodiment, accurate data of the tree planting area can be effectively obtained.

[0061] It should be noted that traditional tree planting area extraction is generally based on the binary classification of satellite data. Since the binary classification can only simply distinguish between tree planting area samples and non-tree planting area samples, non-tree planting area samples. Those skilled in the art can understand that the composition structure of the acquired planting area samples is relatively complex, including but not limited to cultivated land and woodland that are similar to the characteristics of tree planting areas. When classifying and extracting planting areas, non-tree planting areas are mistakenly extracted as tree planting areas, resulting in the area of ​​the extracted tree planting area being larger than the area of ​​the actual tree planting area. The technical solution of this embodiment makes full use of the tree growth characteristics and the high temporal resolution, high spatial resolution and rich spectral information of hyperspectral satellite data to improve the generalization ability and accuracy of the global tree recognition model, thereby completing the extraction of large-scale comprehensive information on trees.

[0062] The control method of the embodiment of the present invention is further described below based on the accompanying drawings.

[0063] Reference Figure 1 , Figure 1 A flowchart of a planting area extraction method based on satellite data provided by an embodiment of the present invention includes but is not limited to the following steps:

[0064] Step S11, formulating a hyperspectral satellite data collection task and shooting plan to obtain image information of the planting area;

[0065] It should be noted that data acquisition is the basis for extracting the entire tree planting area. Hyperspectral satellite data uses hyperspectral satellite data and is converted into a shooting plan based on mission requirements to obtain the original hyperspectral data of the study area. By formulating hyperspectral satellite data collection tasks and shooting plans, including mission planning, orbit prediction and data reception, image information of the planting area can be obtained.

[0066] Step S12, performing a preprocessing operation on the image information to obtain tree growth characteristic information and spectral difference information in the planting area;

[0067] It should be noted that the original hyperspectral data is decoded and produced, and the data with a cloud cover of 0 is selected for preprocessing operations, and the data is preprocessed to improve the accuracy of subsequent tree planting area extraction. In this embodiment, OpenOHS software is used to quickly preprocess the image information obtained based on the hyperspectral satellite data. The preprocessing operations include but are not limited to radiation calibration, atmospheric correction and orthorectification processes, and then the tree growth characteristic information and spectral difference information of the planting area are obtained, wherein the growth characteristic information includes but is not limited to tree height, crown shape, leaf density and fruit number, etc., thereby reflecting the planting information of the trees in the image of the trees, and the spectral difference information uses spectral imaging technology to obtain the spectral information of the trees and their surrounding environment, and the approximate planting range of the trees is determined by comparing the spectral differences between the trees and the rest of the vegetation in the same area.

[0068] Step S13, determining the tree flowering period and hyperspectral satellite data according to the tree growth characteristic information and the spectral difference information;

[0069] It should be noted that due to the differences in terrain and tree varieties in tree planting, the growth stage of trees can be inferred by obtaining the growth characteristics of trees. The spectral reflectance characteristics of leaves and flower buds of trees at different growth stages are different. The flowering period of trees lasts from late March to mid-April, and the fruiting period of trees lasts from mid-to-late June to early October. At the same time, the spectral characteristics of trees and other plants at different growth stages have a certain degree of overlap. Therefore, the flowering period of trees and hyperspectral satellite data can be determined through tree growth characteristics information and spectral difference information.

[0070] Step S14, obtaining the terrain features of the planting area, extracting the spectral index features and texture features of the tree flowering period according to the hyperspectral satellite data, and constructing the hyperspectral satellite data, spectral index features, terrain features and texture features into a feature data set;

[0071] It should be noted that the hyperspectral satellite data has a narrow spectrum and rich spectral information, which can reflect the spectral characteristics of various objects in the study area; when performing spectral feature analysis, it is usually found that the spectral characteristics of two objects are similar and difficult to extract, and the timely introduction of terrain features can achieve good extraction results. Through field surveys, it was found that trees are mostly planted in hilly terraces and shallow mountainous areas, and there are also a small number of trees in plains. Trees are not planted in areas with large slopes. Therefore, adding terrain features can better improve the distinction between trees and forests; hyperspectral satellite data can not only reflect the spectral information of tree planting areas, but also show the texture characteristics of tree planting areas. Among them, forests are concentrated in patches and have irregular shapes; the surface texture of cultivated land is smooth, block-shaped, and has obvious boundaries; the texture of tree planting areas has a granular and rough feeling, and the boundaries are obvious. Based on hyperspectral data, the relevant bands in the hyperspectral data are calculated separately, thereby improving the distinguishability of trees and other objects, and further improving the extraction accuracy of tree planting areas. The acquired spectral index features, terrain features, and texture features are combined with the corresponding hyperspectral satellite data to construct a feature data set.

[0072] Step S15, establishing an extraction sample and a classification system for the planting area based on the hyperspectral satellite data and the feature data set, and identifying the tree planting area in the image information based on the extraction sample and the classification system;

[0073] It should be noted that trees, woodlands, water bodies, buildings and cultivated land are taken as the main classification objects. Combined with field work, the characteristics of the above-mentioned objects in hyperspectral satellite data at different times are analyzed to establish corresponding object interpretation marks. According to the object interpretation marks, representative and typical pure pixels are selected to make object classification samples. According to the operability of the object classification samples and the classification method, appropriate classifiers are selected, and then the corresponding classification extraction system is formulated. The support vector machine classification method is used to analyze the characteristics of various objects in hyperspectral data at different times. The radial basis kernel (RBF) is used as the kernel function, that is, the SVM-RBF model, so as to realize the comprehensive recognition of tree planting areas in image information by spectral index, terrain characteristics, stylistic characteristics and afternoon differences.

[0074] Step S16, constructing a plurality of tree flowering period feature data sets according to the hyperspectral satellite data, extracting each tree flowering period feature data set to obtain peach blossom extraction results;

[0075] It should be noted that the constructed multiple tree flowering period feature data sets are used as data sources, combined with the produced land feature classification samples, and the classifier is used to extract the tree flowering period feature data of each period to obtain the peach blossom extraction results.

[0076] Step S17, obtaining the intersection of the result union of the peach blossom extraction result and the preset tree flowering period feature data set to obtain the tree distribution situation in the tree planting area.

[0077] It should be noted that the union of multiple peach blossom extraction results is taken, and then the constructed tree flowering period feature dataset is used as the data source, combined with the production of classified samples, the classifier is used to extract the tree planting area, and then the union of multiple peach blossom extraction results is intersected with the peach blossom extraction results, the fragrance mist confirms the tree distribution, and the small patches in the intersection results are post-processed and vector corrected to obtain the precise tree planting area.

[0078] It should be noted that this embodiment takes into account important factors such as the growth environment and growth characteristics of trees during their growth process, and verifies the annual growth phenological period of trees with the characteristics of the flowering period. This reduces the interference of non-tree planting areas with similar tree characteristics on the accuracy, thereby greatly improving the accuracy of the model extracted from the tree planting area.

[0079] In addition, in one embodiment, referring to Figure 2 ,exist Figure 1 Step S12 of the illustrated embodiment also includes but is not limited to the following steps:

[0080] Step S21, acquiring mission planning information, orbit prediction information and data reception information according to the hyperspectral satellite acquisition mission and the shooting plan;

[0081] Step S22, decoding the hyperspectral satellite data acquisition task according to the task planning information, orbit prediction information and data reception information, and generating auxiliary data, relative radiation correction data and RPC parameters;

[0082] Step S23, acquiring radiation calibration information and correction information according to the auxiliary data, relative radiation correction data and RPC parameters, wherein the correction information at least includes atmospheric correction and orthorectification;

[0083] Step S24, converting the original grayscale value obtained by the hyperspectral satellite data acquisition task into a radiation brightness value according to the radiation calibration information, and determining the target radiation value of the hyperspectral satellite;

[0084] Step S25, obtaining the radiation brightness value and deformation data of the hyperspectral satellite, atmospheric correction converting the radiation brightness value into the actual reflectivity of the surface, and orthorectification generating a planar orthophoto according to the deformation data.

[0085] It should be noted that the original grayscale value is converted into a physically meaningful radiation brightness value through radiation calibration to eliminate the error of the sensor itself and determine the accurate radiation value at the sensor entrance; the radiation brightness value is converted into the actual reflectivity of the surface through atmospheric correction to eliminate the errors caused by atmospheric scattering, absorption and reflection. Orthorectification is to eliminate the deformation caused by terrain or camera orientation, and finally generate a planar orthophoto.

[0086] In addition, in one embodiment, referring to Figure 3 ,exist Figure 1 Step S13 of the illustrated embodiment also includes but is not limited to the following steps:

[0087] Step S31, obtaining terrain difference information and tree variety information of the planting area;

[0088] Step S32, determining the phenological information of the trees according to the terrain difference information and the variety information;

[0089] Step S33: The image information selects a planting area in the flowering period of trees according to the phenological information.

[0090] It should be noted that this embodiment is illustrated using peach trees. Due to the differences in terrain and peach varieties in peach planting areas, the peach flowering period lasts from late March to mid-April, and the peach fruiting period lasts from mid-to-late June to early October. Therefore, the phenological information of peach trees is determined by obtaining terrain difference information and variety information. After comparing the phenological information with the spectrum, the data collection plan of the satellite is determined, and multiple periods of hyperspectral data are collected in late March and April when the peach tree flowering period is concentrated. Considering the data availability affected by weather and the time limit of about one week for the flowering period of peach trees of the same variety, the present embodiment selects hyperspectral satellite data on March 30, 2023, April 8, 2023, April 11, 2023, and April 19, 2023 within the peach tree flowering period to monitor the peach distribution area; 2) Select hyperspectral satellite data on February 22, 2022, April 26, 2022, July 14, 2022, September 7, 2022, October 4, 2022, and November 5, 2022 to extract the planting area that meets the annual phenological characteristics of peach trees. The distribution of peach trees is analyzed and mutually verified through two levels to improve the accuracy of analysis. (Note: Peach trees generally have a lifespan of 13-15 years, and fruit farmers will adopt an interval update method, so hyperspectral satellite data can be used in the same year or similar years.)

[0091] In addition, in one embodiment, referring to Figure 4 ,exist Figure 1 Step S14 of the illustrated embodiment also includes but is not limited to the following steps:

[0092] Step S41, obtaining near infrared band reflectivity, red light reflectivity, blue light reflectivity, green light reflectivity and soil adjustment coefficient of hyperspectral satellite data;

[0093] Step S42, calculating a spectral index according to the near-infrared band reflectivity, the red light reflectivity, the blue light reflectivity, the green light reflectivity and the soil adjustment coefficient, wherein the spectral index at least includes a normalized vegetation index.

[0094] It should be noted that the hyperspectral satellite data has a narrow spectrum and rich spectral information, which can reflect the spectral characteristics of various objects in the study area. Therefore, by analyzing the spectrum of various objects, the spectral differences between the objects can be determined, and the spectral information affected by the hyperspectral of various objects can be extracted based on the field survey information. Since the overall morphology of the spectrum of trees, forest land and cultivated land is similar, but the overall reflectivity is higher than that of forest land, the reflectivity of cultivated land and trees in the infrared band increases as a whole, so when extracting in the tree planting area, it will be affected by more cultivated land and forest land. Therefore, when selecting the spectral index, this embodiment considers the sensitivity of the band reflectivity to trees, and tries to eliminate the influence of factors such as atmosphere and soil, and finally calculates the spectral index by obtaining the near-infrared band reflectivity, red light reflectivity, blue light reflectivity, green light reflectivity and soil adjustment coefficient of the hyperspectral satellite data.

[0095] In addition, in one embodiment, Figure 4 Step S42 of the illustrated embodiment can be expressed by the following formula:

[0096] NDVI=(ρ NIR -ρ RED ) / (ρ NIR +ρ RED )

[0097] EVI=2.5×[(ρ NIR -ρ RED ) / (ρ NIR +6ρ RED -7.5ρ BLUE )+1]

[0098] ARVI=[ρ NIR -(2ρ RED -ρ BLUE )] / [ρ NIR +(2ρ RED -ρ BLUE )]

[0099] SAVI=[(ρ NIR -ρ RED )×(1+L)] / [ρ NIR +ρ RED +L)]

[0100] RG=ρ RED / ρ GREEN

[0101] RRI=ρ BELUE / ρ NIR

[0102] VI2=[(ρ NIR -ρ RED ) / (ρ NIR +ρRED )]×ρ NIR =NDVI×ρ NIR

[0103] Pseudo CRI1 = 1 / ρ BLUE -1 / ρ GREEN

[0104] Pseudo CRI2 = 1 / ρ GREEN -1 / ρ RED

[0105] Among them, ρ NIR represents the reflectivity of the near-infrared band. Here, the 26th band of the hyperspectral satellite data is taken, with a central wavelength of 880nm; ρ RED represents the reflectivity of the red light band, where the 14th band of the hyperspectral satellite data is taken, with a central wavelength of 670nm; ρ BLUE represents the reflectivity of the blue light band, where the 7th band of the hyperspectral satellite data is taken, with a central wavelength of 566 nm; ρ GREEN represents the reflectivity of the green light band, where the third band of the hyperspectral satellite data is taken, and the central wavelength is 500nm; L is the soil adjustment coefficient, which is 0.5 in this embodiment.

[0106] In addition, in one embodiment, referring to Figure 5 and Figure 7 ,exist Figure 1 Step S15 of the illustrated embodiment also includes but is not limited to the following steps:

[0107] Step S51, obtaining classification object information of the feature data set, and establishing interpretation marks in the hyperspectral satellite data according to the terrain features;

[0108] Step S52, selecting pure pixels in the image information according to the interpretation mark, and making ground object classification samples according to the pure pixels;

[0109] Step S53, acquiring feature information of ground object classification samples in hyperspectral satellite data at different time periods to construct a classification system.

[0110] It should be noted that by analyzing the characteristics of various objects in hyperspectral satellite data at different times, corresponding object interpretation marks are established. Based on the object interpretation marks, representative and typical pure pixels are selected to make object classification samples. According to the sample separability and the operability of the classifier method, a suitable classifier is selected to formulate a corresponding classification extraction system. Furthermore, when performing spectral feature analysis, it is usually found that the spectral features of two objects are similar and difficult to extract, and the timely introduction of terrain features can achieve good extraction effects. Through field surveys, it was found that trees are mostly planted in hilly terraces and shallow mountainous areas, and there are also a small number of trees in plains. Trees are not planted in areas with larger slopes. Therefore, adding terrain features can better improve the distinction between trees and woodlands.

[0111] Furthermore, the feature dataset is composed of hyperspectral satellite data (32-band reflectance, spectral range: 400nm~1000nm), spectral index features (NDVI) and texture features (eight texture features of mean, variance, homogeneity, contrast, difference, entropy, angular second moment and correlation calculated from the 3rd, 7th, 14th and 26th bands of hyperspectral satellite data) and terrain features (slope).

[0112] In addition, in one embodiment, referring to Figure 6 ,exist Figure 1 After step S14 in the illustrated embodiment, the following steps are also included but not limited to:

[0113] Step S61, obtaining surface roughness information in the image information, and when the surface roughness meets the texture characteristics of the planting area, obtaining multiple band information of the spectral index characteristics;

[0114] Step S62: Calculate the mean of multiple band information according to the hyperspectral data.

[0115] It should be noted that forest land is concentrated and distributed in patches with irregular shapes; cultivated land has smooth surface texture, block distribution, and obvious boundaries; the texture of tree planting areas has granular roughness and obvious boundaries. Based on hyperspectral data, eight texture features, namely mean, variance, homogeneity, contrast, dissimilarity, entropy, angular second moment, and correlation, are calculated for the relevant bands, which helps to improve the separability of trees from other objects and further improve the extraction accuracy.

[0116] It is understood by those skilled in the art that the texture features presented by the same objects in hyperspectral data are different, and adding texture information helps to improve the separability of objects. Since adjacent bands in hyperspectral images are often correlated and have high information redundancy, in order to reduce the dimension of the data set and improve the operation efficiency, after multiple tests and selections, the 3rd, 7th, 14th, and 26th bands of hyperspectral satellite data were selected to perform eight texture feature operations, namely, mean, variance, homogeneity, contrast, difference, entropy, angular second moment, and correlation.

[0117] like Figure 8 As shown, Figure 8 : is a structural diagram of a planting area extraction device based on satellite data provided by an embodiment of the present invention. The present invention also provides a planting area extraction device based on satellite data, comprising:

[0118] The processor 801 may be implemented by a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0119] The memory 802 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 802, and the processor 801 calls and executes the planting area extraction method based on satellite data in the embodiment of this application;

[0120] Input / output interface 803, used to implement information input and output;

[0121] The communication interface 804 is used to realize the communication interaction between the present apparatus and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0122] A bus 805 that transmits information between the various components of the device (e.g., the processor 801, the memory 802, the input / output interface 803, and the communication interface 804);

[0123] The processor 801 , the memory 802 , the input / output interface 803 and the communication interface 804 are connected to each other in communication within the device via a bus 805 .

[0124] An embodiment of the present application also provides an electronic device, including the planting area extraction device based on satellite data as described above.

[0125] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned planting area extraction method based on satellite data is implemented.

[0126] As a non-transient computer-readable storage medium, the memory can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and are implemented to be located in one place, or may also be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0127] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically include computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0128] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions under the shared conditions without violating the spirit of the present invention. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A method for extracting planting areas based on satellite data, characterized in that: include: Develop hyperspectral satellite data collection tasks and shooting plans to obtain image information of the planting area; Performing a preprocessing operation on the image information to obtain tree growth characteristic information and spectral difference information in the planting area; Determine the tree flowering period and hyperspectral satellite data according to the tree growth characteristic information and the spectral difference information; Acquire the topographic features of the planting area, extract the spectral index features and texture features of the tree flowering period according to the hyperspectral satellite data, and construct the hyperspectral satellite data, the spectral index features, the topographic features and the texture features into a feature data set; Establishing an extraction sample and a classification system for the planting area according to the hyperspectral satellite data and the feature data set, and identifying the tree planting area in the image information according to the extraction sample and the classification system; Constructing a plurality of tree flowering period characteristic data sets according to the hyperspectral satellite data, extracting each of the tree flowering period characteristic data sets to obtain peach blossom extraction results; The intersection of the result union of the peach blossom extraction result and a preset tree flowering period feature data set is obtained to obtain the tree distribution situation in the tree planting area.

2. The method for extracting planting areas based on satellite data according to claim 1, characterized in that: The preprocessing operation on the image information includes: Acquire mission planning information, orbit prediction information and data reception information according to the hyperspectral satellite acquisition mission and the shooting plan; Decoding the hyperspectral satellite data acquisition mission according to the mission planning information, the orbit prediction information and the data reception information to generate auxiliary data, relative radiation correction data and RPC parameters; Acquire radiation calibration information and correction information according to the auxiliary data, the relative radiation correction data and the RPC parameters, wherein the correction information at least includes atmospheric correction and orthorectification; The original grayscale value obtained by the hyperspectral satellite data acquisition task is converted into a radiation brightness value according to the radiation calibration information to determine the target radiation value of the hyperspectral satellite; The radiation brightness value and deformation data of the hyperspectral satellite are obtained, the atmospheric correction converts the radiation brightness value into the actual reflectivity of the surface, and the orthorectification generates a planar orthophoto according to the deformation data.

3. The method for extracting planting areas based on satellite data according to claim 1, characterized in that: The determining of tree flowering period and hyperspectral satellite data according to the tree growth characteristic information and the spectral difference information includes: Acquiring terrain difference information and tree variety information of the planting area; Determining the phenological information of the tree according to the terrain difference information and the variety information; The image information selects the planting area that is in the flowering period of the tree according to the phenological information.

4. The method for extracting planting areas based on satellite data according to claim 1, characterized in that: The step of constructing the hyperspectral satellite data, the spectral index features, the terrain features and the texture features into a feature data set includes: Obtain near infrared band reflectivity, red light reflectivity, blue light reflectivity, green light reflectivity and soil adjustment coefficient of the hyperspectral satellite data; A spectral index is calculated according to the near-infrared band reflectivity, the red light reflectivity, the blue light reflectivity, the green light reflectivity and the soil adjustment coefficient, wherein the spectral index at least includes a normalized vegetation index.

5. The method for extracting planting areas based on satellite data according to claim 4 is characterized in that the spectral index is calculated according to the near-infrared band reflectivity, the red light reflectivity, the blue light reflectivity, the green light reflectivity and the soil adjustment coefficient, and is obtained according to the following formula: NDVI=(r NIR -r RED ) / (ρ NIR -r RED ); EVI=2.5*[(ρ NIR -r RED ) / (ρ NIR +6p RED -7.5p BLUE )+1]; ARVI=[ρ NIR -(2p RED -r BLUE )] / [p NIR +(2p RED -r BLUE )]; SAVI=[(ρ NIR -r RED )*(1+L)] / [ρ NIR +r RED +L]; RG=ρ RED / r GREEN ; RPI=ρ BLUE / r NIR ; VI2=[ρ NIR -r RED / (r NIR +r RED )]*r NIR =NDVI*r NIR ; Pseudo CRI1 = 1 / ρ BLUE -1 / ρ GREEN ; Pseudo CRI2 = 1 / ρ GREEN -1 / ρ RED ; in, NDVI is the normalized vegetation index, EVI is the enhanced vegetation index, ARVI is the atmospheric impedance vegetation index, SAVI is the soil adjusted vegetation index, RG is the red-green ratio index, RPI is the ratio residential index, VI2 is the first vegetation index, CRI 1 is the first pseudo-carotenoid reflectance index, CRI2 is the second pseudo-carotenoid reflectance index, ρ NIR is the infrared band reflectivity, ρ RED is the red light reflectivity, ρ BLUE is the blue light reflectivity, ρ GREEN is the green light reflectance and L is the soil adjustment coefficient.

6. The method for extracting planting areas based on satellite data according to claim 1, characterized in that: The extracting samples and classifying system of the planting area are established according to the hyperspectral satellite data and the characteristic data set, including: Acquiring classification object information of the feature data set, and establishing interpretation marks in the hyperspectral satellite data according to the terrain features; Selecting pure pixels in the image information according to the interpretation mark, and producing ground object classification samples according to the pure pixels; The classification system is constructed by acquiring feature information of the ground object classification samples in the hyperspectral satellite data at different time periods.

7. The method for extracting planting areas based on satellite data according to claim 1, characterized in that: After extracting the spectral index characteristics and texture characteristics of the tree flowering period according to the hyperspectral satellite data, the method further includes: Acquire surface roughness information in the image information, and when the surface roughness meets the texture feature of the planting area, acquire multiple band information of the spectral index feature; The average of the plurality of band information is calculated according to the hyperspectral data.

8. A planting area extraction device based on satellite data, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the planting area extraction method based on satellite data as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: Including the planting area extraction device based on satellite data as described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the planting area extraction method based on satellite data as described in any one of claims 1 to 7.

Citation Information

Cited By

  • Estimation method for crop yield

    TWI909968B