CN116486258A - Alfalfa identification method and system based on SAR image and crop growth characteristics - Google Patents
Alfalfa identification method and system based on SAR image and crop growth characteristics Download PDFInfo
- Publication number
- CN116486258A CN116486258A CN202310320498.9A CN202310320498A CN116486258A CN 116486258 A CN116486258 A CN 116486258A CN 202310320498 A CN202310320498 A CN 202310320498A CN 116486258 A CN116486258 A CN 116486258A
- Authority
- CN
- China
- Prior art keywords
- alfalfa
- data
- sar image
- sar
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 235000017587 Medicago sativa ssp. sativa Nutrition 0.000 title claims abstract description 165
- 238000000034 method Methods 0.000 title claims abstract description 40
- 240000004658 Medicago sativa Species 0.000 title description 6
- 241000219823 Medicago Species 0.000 claims abstract description 159
- 238000004458 analytical method Methods 0.000 claims abstract description 34
- 238000012216 screening Methods 0.000 claims abstract description 31
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000010287 polarization Effects 0.000 claims description 58
- 238000000605 extraction Methods 0.000 claims description 9
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 238000003786 synthesis reaction Methods 0.000 claims description 5
- 238000013075 data extraction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012300 Sequence Analysis Methods 0.000 claims description 2
- 230000002708 enhancing effect Effects 0.000 abstract description 4
- 230000001360 synchronised effect Effects 0.000 abstract description 4
- 235000009024 Ceanothus sanguineus Nutrition 0.000 description 9
- 240000003553 Leptospermum scoparium Species 0.000 description 9
- 235000015459 Lycium barbarum Nutrition 0.000 description 9
- 238000011161 development Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 239000004576 sand Substances 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003306 harvesting Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003746 surface roughness Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000008216 herbs Nutrition 0.000 description 1
- 244000144972 livestock Species 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 239000008267 milk Substances 0.000 description 1
- 210000004080 milk Anatomy 0.000 description 1
- 235000013336 milk Nutrition 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009304 pastoral farming Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000009331 sowing Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Remote Sensing (AREA)
- Image Processing (AREA)
Abstract
The application provides a alfalfa identification method and system based on SAR images and crop growth characteristics, which belong to the field of crop identification and comprise the following steps: acquiring a time sequence remote sensing image of a region to be identified and preprocessing to obtain a first SAR image; establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, and extracting the first SAR image based on an analysis result to obtain key image data for alfalfa land block identification; and sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring an intersection of the screened images to obtain the alfalfa land parcel range in the area to be identified. The method mainly comprises the steps of starting from SAR images, extracting land block features of alfalfa by utilizing the reflection of growth features of the alfalfa in time sequence image features, comparing with other synchronous growth crops, screening proper time phase data for enhancing features of synchronous crops or alfalfa land blocks, and enhancing separability.
Description
Technical Field
The invention belongs to the field of crop identification, and particularly relates to an alfalfa identification method and system based on SAR images and crop growth characteristics.
Background
With the popularization of (Synthetic Aperture Radar; SAR) images, the SAR images are widely applied to ground object identification and change detection. In addition to the land block identification and monitoring of rice crops, the application of SAR images is widely studied in land block identification of rape, wheat and the like. In northwest regions, the earth surface mainly takes sand as a main material, the texture of the sand surface is relatively gentle, the formed back scattering intensity is low, and the earth surface appears as dark tone in SAR images. The land block planted on the sand area has the backward scattering strength changed along with the growth of crops due to the change of roughness formed by the growth of crops, the water content and the surface roughness of the planted land block are generally increased along with the growth of crops, and the backward scattering strength starts to be reduced along with the maturation of the crops. The features of the planted land block are obvious on the sand background.
Alfalfa is an important feed source for the development of livestock industries in areas such as inner mongolia and Gansu. Along with the support of alfalfa development policies and the like in the plain milk industry, the alfalfa patch area is enlarged from 50 mu to 100 mu in 2019, the fund is increased from 3 to 10 hundred million yuan, and the alfalfa artificial grassland planting area monitoring has important significance. In addition, in the northern semi-agriculture and semi-grazing area, desertification and further development of desertification are hindered, and alfalfa is not only used as a feed source, but also plays an important role in ecological environment protection.
The prior art discloses a tea tree planting area extraction method and system (patent application number: CN 202211089743.1), wherein three different remote sensing data sources including satellite multispectral remote sensing images, SAR satellite polarized remote sensing images and DEM data are introduced, so that diversity of the remote sensing data is increased, a multi-level remote sensing feature set is constructed by simultaneously acquiring primary features, secondary features and tertiary features related to the growth of tea trees, and in addition, geographic environment information of a target area and tea tree growth characteristic information are introduced when the features are acquired, important factors such as the growth environment, growth characteristics and the like of tea trees in the growth process are fully considered, so that the classification result precision of a final tea tree planting area classification model is greatly improved.
The scheme is applied to tea tree planting area identification, great differences exist between tea trees and alfalfa planting areas and between the tea trees and growth characteristics, SAR satellite polarized remote sensing images are introduced in the scheme, so that diversity of remote sensing data is increased, and crop identification is not performed by fully combining SAR image features and crop growth features.
Disclosure of Invention
The application provides a alfalfa identification method and system based on SAR images and crop growth characteristics, and aims to solve the problems that the existing alfalfa identification technology does not consider alfalfa growth characteristics, the extraction precision is poor, and crop identification cannot be performed by fully combining SAR image characteristics and crop growth characteristics in the existing tea tree planting area identification technology, so that the crop identification precision cannot be improved to the greatest extent.
In order to achieve the above purpose, the present solution adopts the following technical solution, including:
acquiring a time sequence remote sensing image of a region to be identified and preprocessing to obtain a first SAR image;
establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, and extracting the first SAR image based on an analysis result to obtain key image data for alfalfa land block identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
and sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring an intersection of the screened images to obtain the alfalfa land parcel range in the area to be identified.
Preferably, alfalfa growth feature analysis is performed based on the second SAR image, specifically:
and calculating the NDVI of the alfalfa sample plot based on the second SAR image, drawing an alfalfa plot NDVI time sequence graph, and determining the growth characteristics of alfalfa according to the date corresponding to the lowest value of the NDVI in the alfalfa plot NDVI time sequence graph.
Preferably, the SAR image feature analysis is performed based on the second SAR image, specifically:
drawing a first backward scattering characteristic curve in the alfalfa plot growth period based on VV polarization data and VH polarization data in the second SAR image;
and acquiring a third SAR image of other crops in the same period of alfalfa in the region to be identified, drawing a second backscattering characteristic curve of the other crops based on VV polarization data and VH polarization data in the third SAR image, and carrying out time sequence analysis by combining the first backscattering characteristic curve and the second backscattering characteristic curve to obtain SAR image characteristics.
Preferably, the band screening is sequentially performed on the key image data, specifically:
substituting 4-5 month time series data in VV polarization data, 6-9 month time series data in VV polarization data, and 4-5 month time series data in VH polarization data into B1=VV_NRCS (in 4 months at bottom-5 months) median B2=vv_nrcs (6-9 months) median B3=vh_nrcs (4 months bottom-5 months middle) median And obtaining characteristic wave bands, wherein B1, B2 and B3 respectively represent characteristic wave bands of different periods, VV_NRCS (in 4 months at the bottom of 4 months and 5 months) represents 4-5 months of time sequence data in the VV polarization data, VV_NRCS (in 6 months and 9 months) represents 6-9 months of time sequence data in the VV polarization data, and VH_NRCS (in 4 months at the bottom of 4 months and 5 months) represents 4-5 months of time sequence data in the VH polarization data.
Preferably, the preset threshold screening is specifically:
RGB image synthesis is respectively carried out based on the characteristic wave bands, and SAR image characteristic synthesis images are obtained;
and B1, B2 and B3 are respectively substituted into E1=B2 < B, E2=B1 > a and E3=B3 > c for screening, and alfalfa block images in three periods after screening are obtained, wherein E1, E2 and E3 respectively represent alfalfa block images in different periods, and a, B and c respectively represent thresholds for eliminating non-alfalfa crops.
Preferably, the alfalfa field images of three periods are substituted into a formula E=E1 n E2 n E3, and the alfalfa field ranges in the area to be identified are obtained after intersection, wherein E is the alfalfa field range in the area to be identified.
An alfalfa identification system based on SAR images and crop growth characteristics, comprising:
a first SAR image acquisition module: the method comprises the steps of acquiring a time sequence remote sensing image of a region to be identified, and preprocessing to obtain a first SAR image;
the key image data extraction module: the method comprises the steps of establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, extracting a first SAR image based on an analysis result to obtain key image data for alfalfa land block identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
alfalfa field zone extraction module: the method is used for sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring the alfalfa land parcel range in the area to be identified after the screened images are intersected.
An electronic device comprising a memory and a processor, the memory for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement a alfalfa identification method based on SAR images and crop growth characteristics as in any of the above.
A computer readable storage medium storing a computer program which when executed by a computer implements a alfalfa identification method based on SAR images and crop growth characteristics as described in any one of the above.
The invention has the following beneficial effects:
(1) The method mainly comprises the steps of starting from SAR images, extracting land block characteristics of alfalfa by utilizing the reflection of growth characteristics of the alfalfa in time sequence image characteristics, comparing with other synchronous growth crops, screening proper time phase data for enhancing the characteristics of synchronous crops or alfalfa land blocks, and enhancing the separability;
(2) According to the method, the growth characteristics of the alfalfa are analyzed, the growth period, the growth speed and the characteristics of multiple mowing of the alfalfa are described, an information basis is provided for SAR image characteristic description, the growth characteristics of the alfalfa are fully considered, and the accuracy of alfalfa land parcel identification is improved;
(3) In the scheme, after the back scattering intensity of alfalfa and other main two ground features is compared and analyzed, three key image data for identifying alfalfa are determined according to alfalfa growth characteristics and SAR image characteristics, wherein the key image data are respectively 4-5 month synthetic data in VV polarization data, 6-9 month synthetic data in VV polarization data and 4-5 month synthetic data in VH polarization data, and the growth characteristics and the image characteristics are fully combined to achieve the highest alfalfa crop identification accuracy.
Drawings
FIG. 1 is a flow chart of a method for identifying alfalfa based on SAR images and crop growth characteristics in the present invention
FIG. 2 is a schematic diagram of a specific embodiment of example 1 of the present invention
FIG. 3 is a timing diagram of the NDVI of alfalfa field plots in example 1 of the present invention
FIG. 4 is a graph showing NRCS values of alfalfa plots VV and VH polarization data in example 1 of the present invention
FIG. 5 is a comparison of NRCS timing and harmonic fitting for different features in example 1 of the present invention
FIG. 6 is a synthetic image of SAR image features in embodiment 1 of the present invention
FIG. 7 is an exemplary image of the B1 feature of embodiment 1 of the present invention
FIG. 8 is an exemplary image of the B2 feature of embodiment 1 of the present invention
FIG. 9 is an exemplary image of the B3 feature of embodiment 1 of the present invention
FIG. 10 is a schematic diagram of a alfalfa identification system according to the present invention
Detailed Description
Example 1
A alfalfa identification method based on SAR images and crop growth characteristics comprises the following steps:
s11, acquiring a time sequence remote sensing image of a region to be identified and preprocessing the time sequence remote sensing image to obtain a first SAR image;
s12, establishing an alfalfa sample plot and acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, and extracting the first SAR image based on an analysis result to obtain key image data for alfalfa plot identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
and S13, sequentially performing band screening and preset threshold screening on the key image data, and acquiring intersection of the screened images to obtain the alfalfa land parcel range in the area to be identified.
According to the technical scheme, firstly, growth characteristics of alfalfa are analyzed from common optical image data, and an information basis is provided for SAR image characteristic description; and then comparing the alfalfa with other two main crops in the same growth period of the alfalfa to obtain a time period suitable for distinguishing the alfalfa from other crops and corresponding SAR image features, combining the VV and VH time sequence data features, establishing a relevant threshold value to extract the alfalfa in the images of different time periods, and finally overlapping the alfalfa planting land parcel ranges correspondingly extracted in different time periods to obtain a final alfalfa land parcel identification result after intersection. The thinking diagram of the scheme is shown in fig. 2, and the specific scheme flow is as follows:
1. alfalfa growth profiling
Taking alfalfa in a semi-desert area as an example (i.e. an alfalfa sample plot is established for illustration), alfalfa artificial plots serving as perennial herbs, particularly as feed sources, exist in field management operations of multiple mowings in an annual period. Alfalfa grows faster and has a short growth cycle, and the possible time from sowing to harvesting is shorter than 1 month. The method is described in connection with common optical image data, and is used for comparing and analyzing alfalfa field growth characteristics in SAR images (namely second SAR images). By taking Sentinel-2 optical data as an example, in the alfalfa growth process, the corresponding vegetation index is utilized to perform growth characteristic description, and then feature selection based on SAR images (namely second SAR images) is established.
Alfalfa is sown in 3 months each year, 4-5 months is a growth and development stage, 6, 8 and 10 months are generally harvest stages due to multiple mowing, and 7 and 9 months are stages of regrowth and development after mowing. The alfalfa growth was demonstrated using the NDVI index of Sentinel-2 data (ndvi= (Nir-R)/(nir+r), where Nir, R are the spectral reflectivities of the near-infrared and red bands, respectively). The revisiting period of the Sentinel-2 satellite is 5 days, in the processing process (the pretreatment process), image data with cloud content lower than 10% is firstly screened, and in the northwest area, drought and little rain are generally carried out, and clear sky is the main, so that the cloud content threshold can be properly improved. Cloud parameters are provided by a 'QA60' wave band of Sentinel-2, bit10 and bit11 of the QA60 wave band correspond to opaque clouds and cirruscloads respectively, and NDVI calculation is carried out on the basis of cloud masking on images. Taking 2021-2022 as an example, the corresponding NDVI time sequence curve of alfalfa plots is shown in FIG. 3.
Plots in FIG. 3 were mowed around 6/14, 7/29, 9/30 for the lowest NDVI values, respectively, i.e., the date (corresponding to alfalfa growth characteristics), which were consistent with alfalfa growth climates.
2. SAR image feature analysis
The scheme adopts Sentinel-1 data as SAR image data source to analyze the image characteristics of alfalfa plots. The revisitation period of a single satellite of the Sentinel-1 satellite is 12 days, and the scheme selects an interference broad-width (IW) mode product of the Sentinel-1 satellite, and derating data is used for alfalfa monitoring. The SAR data source is geocoded and radiometric scaled to form an intensity map.
In the land area corresponding to fig. 3, VV and VH time sequence polarization data are obtained respectively, as shown in fig. 4, it can be found that in the SAR image (i.e., the second SAR image), the backward scattering intensity (NRCS) rapidly rises in 4-5 months (i.e., the growth and development stage of alfalfa), and after the last mowing, i.e., about 10 months, the NRCS significantly drops, which indicates that the SAR image has a corresponding response to alfalfa growth. In particular, in VH polarization data, there is a pronounced valley response in alfalfa mowing-growth phase. It is shown that VH polarization mode is more sensitive to variations in surface roughness, dielectric constant, etc. due to mowing during alfalfa mowing.
The main plants in the semi-desert area grow mainly in one year. The comparison between alfalfa and other features of the same period of alfalfa is selected (here, NRCS changes of other main crops in the alfalfa planting area are analyzed, and a semi-desert area, namely, an alfalfa sample land is taken as a research area), and NRCS changes of the main features in the SAR image (third SAR image) are shown in fig. 5. In VV polarization, crop 2 has higher image similarity with alfalfa after 5 months 31 days, but there is a significant difference between the periods of 4 months 26 to 5 months 20 days; crop 1 differs from alfalfa mainly during 6-9 months of growth, with crop 1 having significantly higher NRCS values than alfalfa plots. Crop differentiation can be performed step by selecting images of different time periods. In the VH polarization data comparison, it was found that alfalfa plot NRCS values were clearly differentiated from the contemporaneous crop by day 4, 26 to day 5, 20 days ago. The alfalfa can be distinguished from crops 1 and 2 during the period. In fig. 5, the first back scattering characteristic curve is a curve corresponding to VV and VH of alfalfa, and the second back scattering characteristic curve is a curve corresponding to VV and VH of crop 1 and crop 2.
Meanwhile, the NRCS curves of different crops are subjected to harmonic fitting respectively, and the comparison of the different crops under the polarization data of VV and VH is shown in figure 5. In VV polarization, crop 1 is more amplitude than alfalfa and crop 2, and the alfalfa peak occurs earlier than crop 3; in VH polarization, after harmonic treatment, the growth characteristics of alfalfa in 4-5 months are weakened, and the whole alfalfa land is close to the crop 1.
Combining with time sequence NRCS comparison and harmonic fitting analysis of different ground objects in a research area, the main distinguishable characteristics (namely SAR image characteristics) of the alfalfa ground are obtained as follows:
VV polarization data: in the cultivated land range, the crop 1 can be preferentially extracted in the period of 6-9 months, and the data of the last ten days of 4 months to 5 months can be selected to distinguish the crop 2 on the basis of removing the crop 1. In addition, the feature of the alfalfa land parcels can be found out that certain differences exist in amplitude and phase of the alfalfa land parcels through fitting of the VV polarization data, and the feature can be used for alfalfa recognition.
Vh polarization data: the alfalfa plots are mainly distinguished from crops 1 and 2 in the period of 4 months to 5 months, and images in the time period can be selected for alfalfa identification in the cultivated land range.
3. Alfalfa land parcel identification
Through the analysis of alfalfa growth characteristics and corresponding image characteristics (the analysis of 1 and 2 is based on the characteristic analysis of alfalfa sample plots in a semi-desert region), (the following process is to extract key image data of an area to be identified according to the type of key image data determined after the analysis so as to realize the identification of alfalfa plots in the area to be identified), SAR images of the area to be identified (first SAR images after the same pretreatment in 1) are firstly obtained, then 4-5 month synthetic data, 6-9 month synthetic data and 4-5 month synthetic data in VH polarization data in VV polarization data are respectively extracted for alfalfa extraction, and specific wave bands are screened as follows.
B1 =vv_nrcs (4 months bottom-5 months middle) median ；
B2 =vv_nrcs (6-9 months) median ；
B3 =vh_nrcs (4 months bottom-5 months middle) median 。
On the basis of the above-mentioned wave bands, RGB synthesis is adopted, and the obtained characteristics are shown in fig. 6. The white boundary range in the figure is alfalfa plots.
In the characteristics, a threshold is established for screening, and an alfalfa land parcel extraction flow is constructed:
e1 B2< B (b= -9); (crop removal 1);
e2 =b1 > a (a= -15); (crop removal 2);
e3 =b3 > c (c= -22); (extracting alfalfa);
wherein a, b and c respectively represent threshold values for eliminating non-alfalfa crops, and specific values are obtained by comparing and analyzing curves among different crops in FIG. 5; finally, the alfalfa field scope E=E1 n E2 n E3, namely the intersection of E1, E2 and E3, is extracted from the alfalfa field scope in the area to be identified through the scheme in 3, and the alfalfa identification is completed.
Examples of B1, B2, and B3 are shown in table 1.
TABLE 1
The medium (median) in the specific band screening is to screen a plurality of time series data (3 time phase data are included in the marking frame at the lower left corner in fig. 5, each point represents one time phase data) in the bottom-5 months of 4 months, select the median of three pixel values at the same pixel position in the time series data, and further process the three-scene image into a scene image, wherein the scene image is taken as a characteristic band B; this section describes a procedure corresponding to the above-mentioned "procedure for band-specific screening", and the specific procedures for three different types of band screening are identical, and one of them is merely taken as an example for illustration.
In the characteristic band B, the two classifications are performed according to the threshold values a, B, c, and the B region after the threshold value is taken as E. E is the result after two classifications. The section description corresponds to the "establishing a threshold value for screening and constructing an alfalfa land parcel extraction flow", the specific processing procedures of three different types of characteristic wave bands are the same, and only one of the three types of characteristic wave bands is taken as an example for illustration.
Example 2
As shown in fig. 10, an alfalfa identification system based on SAR images and crop growth characteristics, comprising:
the first SAR image acquisition module 10: the method comprises the steps of acquiring a time sequence remote sensing image of a region to be identified, and preprocessing to obtain a first SAR image;
the key image data extraction module 20: the method comprises the steps of establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, extracting a first SAR image based on an analysis result to obtain key image data for alfalfa land block identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
alfalfa field parcel scope extraction module 30: the method is used for sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring the alfalfa land parcel range in the area to be identified after the screened images are intersected.
In one embodiment of the system, in a first SAR image acquisition module 10, a time-series remote sensing image of a region to be identified is acquired and preprocessed to obtain a first SAR image, in a key image data extraction module 20, an alfalfa sample land block is established and a second SAR image is acquired, alfalfa growth feature analysis and SAR image feature analysis are respectively performed based on the second SAR image, the first SAR image is extracted based on the analysis result to obtain key image data for alfalfa land block identification, the key image data comprises 4-5 month time-series data in VV polarization data, 6-9 month time-series data in VV polarization data and 4-5 month time-series data in VH polarization data, in an alfalfa land block range extraction module 30, band screening and preset threshold screening are sequentially performed on the key image data, and after the screened images are acquired, an alfalfa land block range in the region to be identified is obtained.
Example 3
On the basis of the above embodiments, the present embodiment provides an electronic device.
Example 4
On the basis of the above embodiments, the present embodiment provides a storage medium.
The above embodiments are merely illustrative embodiments of the present invention, but the technical features of the present invention are not limited thereto, and any changes or modifications made by those skilled in the art within the scope of the present invention are included in the scope of the present invention.
Claims (9)
1. The alfalfa identification method based on SAR images and crop growth characteristics is characterized by comprising the following steps of:
acquiring a time sequence remote sensing image of a region to be identified and preprocessing to obtain a first SAR image;
establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, and extracting the first SAR image based on an analysis result to obtain key image data for alfalfa land block identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
and sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring an intersection of the screened images to obtain the alfalfa land parcel range in the area to be identified.
2. The alfalfa identification method based on the SAR image and the crop growth characteristics according to claim 1, wherein alfalfa growth characteristic analysis is performed based on the second SAR image, specifically:
and calculating the NDVI of the alfalfa sample plot based on the second SAR image, drawing an alfalfa plot NDVI time sequence graph, and determining the growth characteristics of alfalfa according to the date corresponding to the lowest value of the NDVI in the alfalfa plot NDVI time sequence graph.
3. The alfalfa identification method based on the SAR image and the crop growth characteristics according to claim 1, wherein the SAR image characteristic analysis is performed based on the second SAR image, specifically:
drawing a first backward scattering characteristic curve in the alfalfa plot growth period based on VV polarization data and VH polarization data in the second SAR image;
and acquiring a third SAR image of other crops in the same period of alfalfa in the region to be identified, drawing a second backscattering characteristic curve of the other crops based on VV polarization data and VH polarization data in the third SAR image, and carrying out time sequence analysis by combining the first backscattering characteristic curve and the second backscattering characteristic curve to obtain SAR image characteristics.
4. The alfalfa identification method based on the SAR image and the crop growth characteristics according to claim 1, wherein the band screening is sequentially performed on the key image data, specifically:
substituting 4-5 month time series data in VV polarization data, 6-9 month time series data in VV polarization data, and 4-5 month time series data in VH polarization data into B1=VV_NRCS (in 4 months at bottom-5 months) median B2=vv_nrcs (6-9 months) median B3=vh_nrcs (4 months bottom-5 months middle) median And obtaining characteristic wave bands, wherein B1, B2 and B3 respectively represent characteristic wave bands of different periods, VV_NRCS (in 4 months at the bottom of 4 months and 5 months) represents 4-5 months of time sequence data in the VV polarization data, VV_NRCS (in 6 months and 9 months) represents 6-9 months of time sequence data in the VV polarization data, and VH_NRCS (in 4 months at the bottom of 4 months and 5 months) represents 4-5 months of time sequence data in the VH polarization data.
5. The alfalfa identification method based on SAR images and crop growth characteristics according to claim 4, wherein the preset threshold screening is specifically:
RGB image synthesis is respectively carried out based on the characteristic wave bands, and SAR image characteristic synthesis images are obtained;
and B1, B2 and B3 are respectively substituted into E1=B2 < B, E2=B1 > a and E3=B3 > c for screening, and alfalfa block images in three periods after screening are obtained, wherein E1, E2 and E3 respectively represent alfalfa block images in different periods, and a, B and c respectively represent thresholds for eliminating non-alfalfa crops.
6. The alfalfa identification method based on SAR images and crop growth characteristics according to claim 5, wherein alfalfa block images of three periods are substituted into a formula E=E1 n E2 n E3, and an alfalfa block range in the area to be identified is obtained after intersection, wherein E is the alfalfa block range in the area to be identified.
7. An alfalfa identification system based on SAR images and crop growth characteristics, comprising:
a first SAR image acquisition module: the method comprises the steps of acquiring a time sequence remote sensing image of a region to be identified, and preprocessing to obtain a first SAR image;
the key image data extraction module: the method comprises the steps of establishing an alfalfa sample land block, acquiring a second SAR image, respectively carrying out alfalfa growth feature analysis and SAR image feature analysis based on the second SAR image, extracting a first SAR image based on an analysis result to obtain key image data for alfalfa land block identification, wherein the key image data comprises 4-5 month time sequence data in VV polarization data, 6-9 month time sequence data in VV polarization data and 4-5 month time sequence data in VH polarization data;
alfalfa field zone extraction module: the method is used for sequentially carrying out band screening and preset threshold screening on the key image data, and acquiring the alfalfa land parcel range in the area to be identified after the screened images are intersected.
8. An electronic device comprising a memory and a processor, the memory configured to store one or more computer instructions, wherein the one or more computer instructions are executable by the processor to implement a method of alfalfa identification based on SAR images and crop growth characteristics of any one of claims 1-6.
9. A computer-readable storage medium storing a computer program, wherein the computer program when executed by a computer implements a method for alfalfa identification based on SAR images and crop growth characteristics according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310320498.9A CN116486258A (en) | 2023-03-29 | 2023-03-29 | Alfalfa identification method and system based on SAR image and crop growth characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310320498.9A CN116486258A (en) | 2023-03-29 | 2023-03-29 | Alfalfa identification method and system based on SAR image and crop growth characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116486258A true CN116486258A (en) | 2023-07-25 |
Family
ID=87214665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310320498.9A Pending CN116486258A (en) | 2023-03-29 | 2023-03-29 | Alfalfa identification method and system based on SAR image and crop growth characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116486258A (en) |
-
2023
- 2023-03-29 CN CN202310320498.9A patent/CN116486258A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109345555B (en) | Method for identifying rice based on multi-temporal multi-source remote sensing data | |
| CN109948596B (en) | Method for identifying rice and extracting planting area based on vegetation index model | |
| George et al. | Forest tree species discrimination in western Himalaya using EO-1 Hyperion | |
| Yang et al. | Greenness identification based on HSV decision tree | |
| Peña-Barragán et al. | Object-based crop identification using multiple vegetation indices, textural features and crop phenology | |
| CN112183209B (en) | Regional crop classification method and system based on multidimensional feature fusion | |
| CN113033670B (en) | Rice planting area extraction method based on Sentinel-2A/B data | |
| Sarkate et al. | Application of computer vision and color image segmentation for yield prediction precision | |
| CN111345214A (en) | Xinjiang cotton region identification method and system based on satellite image data | |
| CN113221806B (en) | Cloud platform fusion multi-source satellite image and tea tree phenological period based automatic tea garden identification method | |
| Roth et al. | Automated detection of individual clove trees for yield quantification in northeastern Madagascar based on multi-spectral satellite data | |
| Ruiz et al. | Automated classification of crop types and condition in a mediterranean area using a fine-tuned convolutional neural network | |
| CN115861629A (en) | High-resolution farmland image extraction method | |
| Tridawati et al. | Estimation the oil palm age based on optical remote sensing image in Landak Regency, West Kalimantan Indonesia | |
| CN117079152A (en) | Fine crop classification extraction method and system based on satellite remote sensing image | |
| CN111950361A (en) | Beet identification method based on single-time-sequence NDVI | |
| Duan et al. | Mapping saffron fields and their ages with Sentinel-2 time series in north-east Iran | |
| Roy et al. | Comparative analysis of object based and pixel based classification for mapping of mango orchards in Sitapur district of Uttar Pradesh | |
| Kaur et al. | Automatic crop furrow detection for precision agriculture | |
| CN114332596A (en) | Overwintering crop identification method and device | |
| CN115861844A (en) | Rice early-stage remote sensing identification method based on planting probability | |
| CN116486258A (en) | Alfalfa identification method and system based on SAR image and crop growth characteristics | |
| CN115063690A (en) | Vegetation classification method based on NDVI (normalized difference vegetation index) time sequence characteristics | |
| Dakir et al. | Crop type mapping using optical and radar images: A review | |
| Hashim et al. | Analysis of Oil Palm Tree Recognition using Drone-Based Remote Sensing Images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |