CN114594054A - High-spectrum remote sensing monitoring method for wetland soil microbial community - Google Patents

High-spectrum remote sensing monitoring method for wetland soil microbial community Download PDF

Info

Publication number
CN114594054A
CN114594054A CN202210165244.XA CN202210165244A CN114594054A CN 114594054 A CN114594054 A CN 114594054A CN 202210165244 A CN202210165244 A CN 202210165244A CN 114594054 A CN114594054 A CN 114594054A
Authority
CN
China
Prior art keywords
remote sensing
microbial community
hyperspectral
image
pixel
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
Application number
CN202210165244.XA
Other languages
Chinese (zh)
Inventor
褚伟良
叶素芬
葛世玲
陈一
王俊
王佳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Garden Engineering Co ltd
Original Assignee
Shanghai Garden Engineering Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Garden Engineering Co ltd filed Critical Shanghai Garden Engineering Co ltd
Priority to CN202210165244.XA priority Critical patent/CN114594054A/en
Publication of CN114594054A publication Critical patent/CN114594054A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
    • G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques

Landscapes

  • Remote Sensing (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • General Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Electromagnetism (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention provides a wetland soil microbial community high-spectrum remote sensing monitoring method, which comprises the steps of calculating a pixel difference value of hyperspectral image data and synthetic aperture radar imaging data, correcting a pixel value of the hyperspectral image data based on the pixel difference value, and acquiring a hyperspectral remote sensing image containing a to-be-detected area; performing dimensionality reduction pretreatment on the hyperspectral remote sensing image data, and representing the whole hyperspectral remote sensing image of the microbial community by adopting the first four main components; obtaining the separability degree of a region with the microbial community and a region without the microbial community in the hyperspectral remote sensing image by using the separability index M; separating a region with a microbial community from a region without the microbial community in the hyperspectral remote sensing image, reading the region image with the microbial community, and performing fragment segmentation on the region image; and classifying the image fragments obtained after fragment segmentation by using the trained classification model so as to identify the microbial community.

Description

High-spectrum remote sensing monitoring method for wetland soil microbial community
Technical Field
The invention relates to the field of microbial community detection by utilizing hyperspectral remote sensing, in particular to a method for monitoring the microbial community of wetland soil by hyperspectral remote sensing.
Background
In the prior art, the information of the wetland soil microbial community is usually obtained by using a ground measurement method, but the obtaining mode is mainly artificial measurement, is easily influenced by non-objective factors, has high economic cost and time cost, is difficult to quickly obtain data on a large space scale, cannot ensure the accuracy and the real-time performance of the data, and cannot meet the requirements of relevant departments in time.
Compared with the traditional monitoring means, the remote sensing technology can observe and identify the target in a long distance, can objectively, accurately and timely provide various information, realizes large-area synchronous observation, makes up the defects of the traditional investigation method, and has strong advantages in the aspect of ground object monitoring and identification. When the remote sensing data with medium and low spatial resolution is used for identification in the remote sensing technology, the problem of mixed pixels is easy to occur, and the classification precision is low. In addition, the spectral characteristics of the wetland soil microbial community change with time, so that the spectral characteristics have high dynamics and complexity, and the change of the spectral characteristics of the wetland soil microbial community is difficult to capture by using single-time-phase remote sensing data, and the remote sensing information identified by the wetland soil microbial community cannot be accurately acquired. The time series remote sensing data can dynamically monitor the change of the wetland soil microbial community, can accurately capture the change rule of the spectral characteristics of the wetland soil microbial community, and can determine the key time window for remote sensing identification of the wetland soil microbial community.
The hyperspectral remote sensing can acquire fine spectral information which cannot be acquired by the multispectral sensor by virtue of the characteristics of hyperspectral resolution and strong waveband continuity. The hyperspectral remote sensing data can capture the fine spectral characteristics of the soil, thereby being beneficial to quantitatively acquiring the biochemical components of the soil. Logarithm of reciprocal spectrum (LR), first order differential spectrum (FD), second order differential Spectrum (SD), envelope elimination (CR), and depth of spectral Band (BD) are spectrum transformation techniques commonly used in hyperspectral remote sensing.
The estimation of the physiological and biochemical parameters of the wetland soil microbial community by utilizing the hyperspectral remote sensing data is an urgent need of scientific research such as ecology, agriculture, global change and the like and application industries such as precision agriculture and the like. The measuring method has the characteristics of rapidness and convenience, and the measuring area can be even expanded to the size of the whole earth. And it is a non-destructive measurement that does not have any effect on vegetation growth.
Disclosure of Invention
In order to solve the technical problems, the technical problem to be solved by the application is to provide a method for monitoring the soil microbial community of the wetland by hyperspectral remote sensing, and the method is used for estimating the space-time distribution condition of the microbial community based on a hyperspectral remote sensing detection mechanism.
A wetland soil microbial community high-spectrum remote sensing monitoring method comprises the following steps:
s1, calculating a pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data, correcting a pixel value of the hyperspectral image data based on the pixel difference value, and acquiring a hyperspectral remote sensing image containing a to-be-detected area;
s2, performing dimensionality reduction pretreatment on the hyperspectral remote sensing image data, and representing the whole hyperspectral remote sensing image of the microbial community by adopting the first four main components;
s3, obtaining the separability of the area with the microbial community and the area without the microbial community in the hyperspectral remote sensing image by using the separability index M;
s4, separating a region with a microbial community from a region without the microbial community in the hyperspectral remote sensing image, reading a region image with the microbial community, and performing fragment segmentation on the region image;
and S5, classifying the image fragments obtained after the fragment segmentation by using the trained classification model so as to identify the microbial community.
Further, in step S1, by respectively traversing each of the hyperspectral image data and the synthetic aperture radar imaging data, the maximum value and the minimum value are recorded, and the difference between the maximum value and the minimum value is used as a base number to perform data normalization processing:
Figure BDA0003519370550000021
wherein X in the formula (1) is the pixel value of the image to be normalized, and X ismaxIs a maximum value of XminIs the minimum value of the pixel, and S (X') is the normalized pixel value.
Further, pixel difference value calculation is carried out by utilizing the normalized synthetic aperture radar imaging data and the normalized hyperspectral image data, and the calculation formula is as follows:
T(X')＝S2(X')-S1(X') (2)；
s2(X ') and S1(X ') respectively represent pixel values of the hyperspectral image data and the synthetic aperture radar imaging data after normalization, and T (X ') represents a pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data;
setting a pixel difference threshold ThWhen T (X')>ThEliminating the pixel value from the hyperspectral image data, and when T (X') is less than or equal to ThAnd then, reserving the pixel value in the hyperspectral image data, thereby obtaining a hyperspectral remote sensing image containing the area to be detected.
Further, in step S2, normalizing the data of each wave band of the original hyperspectral image to obtain a normalized image matrix Xc; calculating a covariance matrix Σ c of the normalized matrix Xc; solving an eigenvector matrix Ac of the covariance matrix Σ c, the eigenvector matrix Ac being arranged from left to right according to the law of decreasing eigenvalues; performing linear transformation on the standardized image matrix Xc by using the feature vector matrix Ac to obtain a transformation matrix XP：
Figure BDA0003519370550000031
Transformation matrix XPThe first line of data represents the first principal component of the hyperspectral remote sensing image, and the second line of data represents the second principal component of the hyperspectral remote sensing imageAnd the data represents the second principal component of the hyperspectral remote sensing image, and so on.
Further, in step S3, the separability index M is:
Figure BDA0003519370550000032
in the formula, mupmReflectance mean, u, for wetland soil with microbial communitiesi(i ═ 1,2, …, n) are reflectance means for different types of objects in wetland soil without microbial communities, respectively; deltapmReflectance standard deviation, δ, for wetland soil with microbial communitiesi(i ═ 1,2, …, n) are the standard deviation of reflectance of different types of objects in wetland soil without microbial communities, respectively.
Further, the step S4 includes the following steps:
s41, constructing a weighted directed graph according to the image to be segmented: the method comprises the following steps of taking pixels in an image to be segmented as nodes in a directed graph, taking edges between adjacent pixels as edges for connecting the nodes, defining a cost function on the edges, and enabling the cost value formula of two adjacent pixels to be as shown in a formula (5):
t(p，q)＝ωG×fG(q)+ωZ×fZ(q)+ωD×fD(p，q) (5)；
where t (p, q) is the local cost of a pixel p to its neighboring pixel q; omegaG、ωZAnd ωDIs a weighting coefficient; f. ofG(q)、fZ(q) and fD(q) is a gradient characteristic function, a Laplace zero-crossing characteristic function and a smoothness constraint function at the position of a corresponding pixel q;
and S42, searching the shortest path between two points in the weighted directed graph by adopting an optimal path search algorithm to serve as a boundary segment of the object, and performing edge segmentation.
Further, in step S5, an SVM is used to train the training set, and the radial basis function is used as a kernel function for training the SVM.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application.
FIG. 1 is a schematic flow chart of the wetland soil microbial community high-spectrum remote sensing monitoring method of the invention;
FIG. 2 is a flow chart of a method of the present invention for performing fragment segmentation on images of a region having microbial communities;
FIG. 3 is a diagram showing the relationship between the classification accuracy of the present invention and the number of iterations.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The wetland soil microbial community coverage is a process which is gradually completed in a certain time period, the color tone of the wetland soil with the microbial community is different from that of the wetland soil without the microbial community on a remote sensing image in different periods, and the remote sensing characteristics of the microbial community are changed correspondingly along with the change of time. The remote sensing image with short revisit period and high spatial resolution is selected, the change condition of the microbial community can be continuously observed, and the remote sensing characteristics of the microbial community can be accurately captured.
As shown in fig. 1, a schematic flow chart of the method for monitoring wetland soil microbial community by high-spectrum remote sensing of the invention specifically includes the following steps:
step 1, analyzing a pixel difference value of hyperspectral image data and synthetic aperture radar imaging data, correcting a pixel value of the hyperspectral image data based on the pixel difference value, and eliminating missing or insufficient remote sensing data, thereby obtaining a hyperspectral remote sensing image containing a to-be-detected area.
In the embodiment, a Sentinel-2 single-star is preferably adopted as the remote sensing detector, the Sentinel-2 single-star revisiting period is 10 days, the A/B double-star revising period is 5 days, the main effective load is a multi-spectral imager (MSI), 13 wave bands are in total, the spectral range is 0.4-2.4 mu m, and visible light, near infrared and short wave infrared are covered.
The all-weather synthetic aperture radar imaging system acquires synthetic aperture radar imaging data all day long, two satellites of the synthetic aperture radar imaging system adopt sun synchronous orbits, the orbit height is 693km, the inclination angle is 98.180, the orbit period is 99min, and the revisit period of a single satellite is 12 days. In this embodiment, it is preferable to use single-view complex image data as the imaging data of the synthetic aperture radar, and the imaging mode is dual polarization and is an interference wide mode.
The pixel value of the synthetic aperture radar imaging data represents the reflectivity of the ground object, and the pixel value of the hyperspectral image data represents the backscattering intensity of the ground target to radar waves. In order to correct the pixel value of the hyperspectral image data, the pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data is analyzed from the angle of the numerical value, namely all data in the hyperspectral image data and the synthetic aperture radar imaging data are normalized respectively, and the pixel value of all the data is converted to be between 0 and 1.
Specifically, the maximum value and the minimum value are recorded by respectively traversing each piece of hyperspectral image data and synthetic aperture radar imaging data, and the data normalization processing is performed by taking the difference between the maximum value and the minimum value as a base number, wherein the calculation formula of the normalization processing is shown as formula (1).
Figure BDA0003519370550000051
Wherein X in the formula (1) is the pixel value of the image to be normalized, and X ismaxIs a maximum value of XminIs the minimum value of the pixel, and S (X') is the normalized pixel value.
And calculating a pixel difference value by using the normalized synthetic aperture radar imaging data and the normalized hyperspectral image data, wherein the calculation formula is as follows:
T(X')＝S2(X')-S1(X') (2)；
wherein, S2(X '), S1(X ') represent normalized pixel values of the hyperspectral image data and the synthetic aperture radar imaging data, respectively, and T (X ') represents a pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data.
Setting a pixel difference threshold ThWhen T (X')>ThEliminating the pixel value from the hyperspectral image data, and when T (X') is less than or equal to ThAnd then, reserving the pixel value in the hyperspectral image data, thereby obtaining a hyperspectral remote sensing image containing the area to be detected.
And 2, performing dimensionality reduction preprocessing on the high remote sensing spectrum image data.
The difference of the spectral reflectance curves caused by the difference of the basic components of the microbial community is a physical basis for analyzing and classifying the ground objects based on the hyperspectral remote sensing image, and images of certain wave bands may have larger noise due to the quality problem of the remote sensing detector. Therefore, the high remote sensing spectrum image data is subjected to dimensionality reduction preprocessing.
In this embodiment, the algorithm for performing dimension reduction preprocessing on the high-light remote sensing spectrum image data adopts a component analysis method, utilizes K-L transformation, and uses the variance of the data to describe the information quantity, so as to try to make the transformed data distributed in a descending manner according to the information quantity.
And projecting the original data into a new coordinate space by adopting a linear projection method, wherein the first principal component has the largest information content in the new coordinate space, the second principal component is irrelevant to the first principal component data, the information content in the rest components is the largest, and the like. For the hyperspectral images with dozens of wave bands, the first four and five principal components can generally contain more than 90% of information content of the images, so that the first few principal components can be used for replacing the whole hyperspectral image, and the purpose of reducing the dimension is achieved.
The specific method of the dimensionality reduction pretreatment is as follows:
standardizing data of each wave band of an original hyperspectral remote sensing image to obtain a standardized image matrix Xc;
calculating a covariance matrix Σ c of the normalized matrix Xc;
solving a characteristic vector matrix Ac of the covariance matrix Σ c, wherein the characteristic vector matrix Ac is arranged from left to right according to a decreasing rule of characteristic values;
carrying out linear transformation on the standardized image matrix Xc by utilizing the obtained eigenvector matrix Ac to obtain a transformation matrix XPThe calculation formula is as follows:
Figure BDA0003519370550000061
transformation matrix XPThe data in the first row represents a first principal component of the hyperspectral remote sensing image, the data in the second row represents a second principal component of the hyperspectral remote sensing image, and the like. In this embodiment, the first four principal components are preferably used to represent the entire hyperspectral image of the microbial community.
And 3, adopting the separability index M, and selecting an index which can fully distinguish the area with the microbial community from the area without the microbial community by quantitatively analyzing the separation degree of the area with the microbial community and the area without the microbial community.
The separability index M mainly represents the difference value between the mean values of the spectral reflectivities of the two regions and the normalization processing of the sum of standard deviations delta thereof, and the mathematical expression is as follows:
Figure BDA0003519370550000062
in the formula, mupmReflectance mean, u, for wetland soil with microbial communitiesi(i ═ 1,2, …, n) are reflectance means for different types of objects in wetland soil without microbial communities, respectively; deltapmReflection of wetland soil for microbial communitiesStandard deviation of rate, deltai(i ═ 1,2, …, n) are the standard deviation of reflectance of different types of objects in wetland soil without microbial communities, respectively.
When M >1, it means that the separation between the region having a microbial community and the region not having a microbial community is good; if M <1, it means that the separation of the region having the microbial community from the region not having the microbial community is poor. A larger M value indicates better separability between the area with the microbial community and the area without the microbial community. Therefore, the separability index M is used to obtain the separability between the region having the microbial community and the region not having the microbial community in the wetland soil.
And 4, separating the area with the microbial community in the wetland soil from the area without the microbial community, reading the hyperspectral image of the area with the microbial community in the wetland soil, and performing fragment segmentation on the hyperspectral image of the area to obtain a plurality of segmented image fragments:
the fragment segmentation algorithm is an interactive image segmentation algorithm, can effectively avoid excessive segmentation on the premise of not losing image information, and has better performance in segmenting the complex microbial community image. The method specifically comprises the following substeps:
and S41, constructing a weighted directed graph according to the image to be segmented.
And taking the pixels in the image to be segmented as nodes in the directed graph, and taking edges between adjacent pixels as edges for connecting the nodes. Defining a cost function on the edge, wherein the value of the cost function is used as the weight of the edge, the strong edge has smaller cost value, the non-edge has larger cost value, and the cost value of the edge connecting the non-adjacent pixels is + ∞.
The cost value formula of two adjacent pixels is as follows (5):
t(p，q)＝ωG×fG(q)+ωZ×fZ(q)+ωD×fD(p，q) (5)；
where t (p, q) is the local cost of a pixel p to its neighboring pixel q; omegaG、ωZAnd ωDIs a weighting coefficient; f. ofG(q)、fZ(q) and fD(q) is a gradient characteristic function, a Laplace zero-crossing characteristic function and a smoothness constraint function at the corresponding pixel q, and the functions are calculated according to equations (6) to (8):
Figure BDA0003519370550000071
Figure BDA0003519370550000072
Figure BDA0003519370550000073
where G (q) and L (q) are the magnitude of the gradient at pixel q and the Laplace value, respectively; d (q) is the unit normal vector at pixel q in the image.
And S42, searching the shortest path between two points in the weighted directed graph as the boundary segment of the object by adopting an optimal path search algorithm, and carrying out edge segmentation.
And 5, using the existing confirmed microbial community image as a training set, training the training set by using an SVM (support vector machine) to obtain a microbial community classification model, and classifying the image fragments obtained in the step 2 by using the trained classification model so as to identify the microbial community.
In the preferred embodiment, the radial basis function has strong locality in SVM training, so that the method can show good performance on small sample data, and the radial basis function is selected as the kernel function for SVM training on the basis of the reason that the number of features of image fragments is small. As shown in fig. 3, the classification accuracy of the training using the existing microbial community images that have been confirmed as training sets is a function of the number of iterations.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.

Claims (7)

1. A wetland soil microbial community high-spectrum remote sensing monitoring method is characterized by comprising the following steps:
s1, calculating a pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data, correcting a pixel value of the hyperspectral image data based on the pixel difference value, and acquiring a hyperspectral remote sensing image containing a to-be-detected area;
s2, performing dimensionality reduction pretreatment on the hyperspectral remote sensing image data, and representing the whole hyperspectral remote sensing image of the microbial community by adopting the first four main components;
s3, obtaining the separability of the area with the microbial community and the area without the microbial community in the hyperspectral remote sensing image by using the separability index M;
s4, separating a region with a microbial community from a region without the microbial community in the hyperspectral remote sensing image, reading a region image with the microbial community, and performing fragment segmentation on the region image;
and S5, classifying the image fragments obtained after the fragment segmentation by using the trained classification model so as to identify the microbial community.
2. The method for hyperspectral remote sensing monitoring of a microbial community in soil of a wetland according to claim 1, wherein in step S1, the maximum value and the minimum value are recorded by traversing each of the hyperspectral image data and the synthetic aperture radar imaging data, and the data is normalized by using the difference between the maximum value and the minimum value as a base:
Figure FDA0003519370540000011
wherein X in the formula (1) is the pixel value of the image to be normalized, and X ismaxIs a maximum value of XminIs the minimum value of the pixel, and S (X') is the normalized pixel value.
3. The wetland soil microbial community hyperspectral remote sensing monitoring method according to claim 2, characterized in that pixel difference calculation is performed by using normalized synthetic aperture radar imaging data and normalized hyperspectral image data, and the calculation formula is as follows:
T(X')＝S2(X')-S1(X') (2)；
s2(X ') and S1(X ') respectively represent pixel values of the hyperspectral image data and the synthetic aperture radar imaging data after normalization, and T (X ') represents a pixel difference value of the hyperspectral image data and the synthetic aperture radar imaging data;
setting a pixel difference threshold ThWhen T (X')>ThEliminating the pixel value from the hyperspectral image data, and when T (X') is less than or equal to ThAnd then, reserving the pixel value in the hyperspectral image data, thereby obtaining a hyperspectral remote sensing image containing the area to be detected.
4. The wetland soil microbial community high-spectrum remote sensing monitoring method according to claim 1, characterized in that in step S2, the data of each wave band of the original hyperspectral image are standardized to obtain a standardized image matrix Xc; calculating a covariance matrix Σ c of the normalized matrix Xc; solving an eigenvector matrix Ac of the covariance matrix Σ c, the eigenvector matrix Ac being arranged from left to right according to the law of decreasing eigenvalues; performing linear transformation on the standardized image matrix Xc by using the feature vector matrix Ac to obtain a transformation matrix XP：
Figure FDA0003519370540000021
Transformation matrix XPThe data in the first row represents a first principal component of the hyperspectral remote sensing image, the data in the second row represents a second principal component of the hyperspectral remote sensing image, and the like.
5. The method for remote sensing and monitoring of wetland soil microbial community at high spectrum according to claim 1, wherein in the step S3, separability index M is:
Figure FDA0003519370540000022
in the formula, mupmReflectance mean, u, for wetland soil with microbial communitiesi(i ═ 1,2, …, n) are reflectance means for different types of objects in wetland soil without microbial communities, respectively; deltapmReflectance standard deviation, δ, for wetland soil with microbial communitiesi(i ═ 1,2, …, n) are the standard deviation of reflectance of different types of objects in wetland soil without microbial communities, respectively.
6. The wetland soil microbial community high-spectrum remote sensing monitoring method according to claim 1, characterized in that the step S4 comprises the following steps:
s41, constructing a weighted directed graph according to the image to be segmented: the method comprises the following steps of taking pixels in an image to be segmented as nodes in a directed graph, taking edges between adjacent pixels as edges for connecting the nodes, defining a cost function on the edges, and enabling the cost value formula of two adjacent pixels to be as shown in a formula (5):
t(p，q)＝ωG×fG(q)+ωZ×fZ(q)+ωD×fD(p，q) (5)；
where t (p, q) is the local cost of pixel p to its neighboring pixel q; omegaG、ωZAnd ωDIs a weighting coefficient; f. ofG(q)、fZ(q) and fD(q) is a gradient characteristic function, a Laplace zero-crossing characteristic function and a smoothness constraint function at the position of a corresponding pixel q;
and S42, searching the shortest path between two points in the weighted directed graph by adopting an optimal path search algorithm to serve as a boundary segment of the object, and performing edge segmentation.
7. The method for remote sensing and monitoring of wetland soil microbial community high spectrum according to claim 1, characterized in that in step S5, an SVM is used to train the training set, and the radial basis function is used as the kernel function for SVM training.
CN202210165244.XA 2022-02-24 2022-02-24 High-spectrum remote sensing monitoring method for wetland soil microbial community Pending CN114594054A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210165244.XA CN114594054A (en) 2022-02-24 2022-02-24 High-spectrum remote sensing monitoring method for wetland soil microbial community

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210165244.XA CN114594054A (en) 2022-02-24 2022-02-24 High-spectrum remote sensing monitoring method for wetland soil microbial community

Publications (1)

Publication Number Publication Date
CN114594054A true CN114594054A (en) 2022-06-07

Family

ID=81805783

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210165244.XA Pending CN114594054A (en) 2022-02-24 2022-02-24 High-spectrum remote sensing monitoring method for wetland soil microbial community

Country Status (1)

Country Link
CN (1) CN114594054A (en)

Similar Documents

Publication Publication Date Title
Zhang et al. A tensor decomposition-based anomaly detection algorithm for hyperspectral image
Coburn et al. A multiscale texture analysis procedure for improved forest stand classification
Amorós-López et al. Multitemporal fusion of Landsat/TM and ENVISAT/MERIS for crop monitoring
Halme et al. Utility of hyperspectral compared to multispectral remote sensing data in estimating forest biomass and structure variables in Finnish boreal forest
Lv et al. Object-oriented key point vector distance for binary land cover change detection using VHR remote sensing images
US7184890B2 (en) Cloud shadow detection: VNIR-SWIR
CN105930772A (en) City impervious surface extraction method based on fusion of SAR image and optical remote sensing image
Lin et al. Classification of tree species in overstorey canopy of subtropical forest using QuickBird images
CN106339674A (en) Hyperspectral image classification method based on edge preservation and graph cut model
CN108765359A (en) A kind of fusion method of target in hyperspectral remotely sensed image and full-colour image based on JSKF models and NSCT technologies
CN111144250B (en) Land coverage classification method integrating radar and optical remote sensing data
Peng et al. Object-based change detection from satellite imagery by segmentation optimization and multi-features fusion
Zhang et al. Extraction of tree crowns damaged by Dendrolimus tabulaeformis Tsai et Liu via spectral-spatial classification using UAV-based hyperspectral images
CN110100262A (en) For removing image processing equipment, method and the storage medium of cloud from image
CN111666900A (en) Method and device for acquiring land cover classification map based on multi-source remote sensing image
CN112669363A (en) Urban green land three-dimensional green volume calculation method
Cheng et al. A spatial-spectral clustering-based algorithm for endmember extraction and hyperspectral unmixing
Kim et al. Automatic pseudo-invariant feature extraction for the relative radiometric normalization of hyperion hyperspectral images
Erener et al. A methodology for land use change detection of high resolution pan images based on texture analysis
CN111751295A (en) Modeling method and application of wheat powdery mildew severity detection model based on imaging hyperspectral data
CN114594054A (en) High-spectrum remote sensing monitoring method for wetland soil microbial community
Siok et al. A simulation approach to the spectral quality of multispectral images enhancement
CN115272873A (en) Hyperspectral image nonlinear feature preprocessing system and method based on gradient domain
Chandana Sri et al. Using Convolution Networks to Remove Stripes Noise from Infrared Cloud Images
CN113689414A (en) Method and device for generating long-time sequence high-frequency NDVI in alpine region

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