CN112330700A - Cultivated land plot extraction method of satellite image - Google Patents
Cultivated land plot extraction method of satellite image Download PDFInfo
- Publication number
- CN112330700A CN112330700A CN202011279619.2A CN202011279619A CN112330700A CN 112330700 A CN112330700 A CN 112330700A CN 202011279619 A CN202011279619 A CN 202011279619A CN 112330700 A CN112330700 A CN 112330700A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- sensing image
- node
- cultivated land
- closed curve
- 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
- 238000000605 extraction Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000000007 visual effect Effects 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 3
- 230000011218 segmentation Effects 0.000 description 9
- 230000006978 adaptation Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012272 crop production Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002262 irrigation Effects 0.000 description 1
- 238000003973 irrigation Methods 0.000 description 1
- 230000035764 nutrition Effects 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000002310 reflectometry Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10036—Multispectral image; Hyperspectral image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
-
- 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
Abstract
The invention discloses a cultivated land plot extraction method of satellite images, which comprises the following steps: acquiring a panchromatic remote sensing image and a multispectral remote sensing image; segmenting the multispectral remote sensing image based on a visual interpretation method to generate a plurality of first interior point sets; constructing an energy function of the T-Snake model based on the panchromatic remote sensing image; extracting a second interior point set corresponding to the panchromatic remote sensing image based on the first interior point set; calculating a plot boundary curve corresponding to the second interior point set based on the second interior point set, the panchromatic remote sensing image and the energy function; and repeating the steps until all the land block boundary curves corresponding to all the first internal point sets in the panchromatic remote sensing image are traversed, and generating a plurality of cultivated land block images based on all the land block boundary curves. The method is not limited by the size and the shape of the cultivated land plot, can obtain a better extraction effect at a weak boundary, effectively improves the accuracy of the identification of the boundary of the cultivated land plot, and reduces the generation of broken pattern spots.
Description
Technical Field
The invention relates to the technical field of image processing for land segmentation, in particular to a cultivated land plot extraction method of a satellite image.
Background
The satellite remote sensing technology plays more and more important roles in the aspects of crop classification identification, growth monitoring and the like, and the traditional crop extraction method based on the low-and-medium-resolution satellite images mainly uses an image classification method based on pixels and has no requirement on land boundary information. However, with the improvement of the spatial resolution of the satellite images, how to accurately extract the land parcel boundary based on the satellite images becomes an urgent problem to be solved in future agricultural development, and the solution of the problem can provide more accurate cultivated land parcel vector data for the acquisition of information such as crop classification, crop production condition monitoring, pest and disease damage and nutrition diagnosis and the like.
The current means for extracting farmland plot information by using high-resolution satellite images comprises the following steps: a region segmentation method, a region growing method, a watershed algorithm, a supervision classification method, a mean shift segmentation algorithm, a multi-scale combination aggregation segmentation algorithm and the like. However, the high-resolution satellite images have abundant details, and the phenomena of 'same-object different spectrum' and 'same-spectrum foreign matter' exist, which brings challenges to the accuracy rate of farmland boundary extraction. The size and the shape of the cultivated land are inconsistent in many times under the influence of factors such as irrigation modes, topographic relief and the like, and the traditional algorithm has more or less phenomena of over-segmentation and under-segmentation, so that the segmentation scale is difficult to grasp. And the traditional algorithm is only suitable for a specific small-area farmland area, the algorithm has high calculation complexity, and the boundary missing detection phenomenon can be generated in an area with a weak boundary of a land block generally.
Therefore, the existing methods for extracting farmland land block information by using high-resolution satellite images have the problem of low boundary segmentation accuracy.
Disclosure of Invention
In view of the above, the invention provides a cultivated land plot extraction method of a satellite image, which solves the problem that the boundary segmentation accuracy rate is low in the existing method for extracting cultivated land plot information by a high-resolution satellite image by improving an image detection method.
In order to solve the problems, the technical scheme of the invention is a cultivated land plot extraction method adopting a satellite image, which comprises the following steps: s1: acquiring a panchromatic remote sensing image and a multispectral remote sensing image; s2: segmenting the multispectral remote sensing image based on a visual interpretation method to generate a plurality of first interior point sets; s3: constructing an energy function of a T-Snake model based on the panchromatic remote sensing image; s4: extracting a second interior point set corresponding to the panchromatic remote sensing image based on the first interior point set; s5: calculating a plot boundary curve corresponding to the second interior point set based on the second interior point set, the panchromatic remote sensing image and the energy function; s6: and repeating the steps S4-S5 until all the corresponding plot boundary curves of all the first interior point sets in the panchromatic remote sensing image are traversed, and generating a plurality of cultivated land plot images based on all the plot boundary curves.
Optionally, the S2 includes: segmenting the multispectral remote sensing image based on the visual interpretation method to generate a plurality of cultivated land plot outlines; traversing all internal points of the cultivated land parcel profile to form the first internal point set P { (x) corresponding to the cultivated land parcel profilesn,yn)|n=1,2,...,N-1},(xn,yn) Is the coordinate of an inner point of the nth parcel.
Optionally, segmenting the multispectral remote sensing image based on the visual interpretation method to generate a plurality of arable land parcel profiles, comprising: judging a cultivated land area based on the shape, color and texture difference between a cultivated land parcel and other background land features in the multispectral remote sensing image, and extracting a planar area with uniform texture and uniform color in the multispectral remote sensing image as the cultivated land area; judging whether boundary lines exist among different arable land areas or not, if so, respectively extracting the contours of the boundary lines to generate a plurality of arable land block contours, and if not, generating a single arable land block contour based on the plurality of arable land areas.
Optionally, the S3 includes: constructing a closed curve function S ═ V under a T-Snake model1,V2,...,VmCharacterizing a closed curve in the panchromatic remote sensing image, wherein Vi(xi,yi) Is a curve node, i ═ 1, 2.., m; constructing the energy function based on the closed curve functionWherein eta, gamma and lambda are proportionality coefficients, C is the geometric center of the current closed curve, and sigma and delta are the gray scale standard deviation and the range of the node neighborhood in the full-color image respectively,is node ViFull color image gradient of (a).
Optionally, the S5 includes: randomly extracting coordinate points (x) in the second set of interior pointsn,yn) And extracting four pixel points of which the coordinate is at a distance of r in the upper, lower, left and right directions, and sequentially connecting the four pixel points clockwise to construct an initial closed curve S { (x)n+r,yn,0),(xn,yn+r,1),(xn-r,yn,2),(xn,yn-r, 3) }, wherein the third attribute of each node represents the direction information of the node, i.e. 0 to the right, 1 to the bottom, 2 to the left and 3 to the top in clockwise order; traversing the node sequence of the initial closed curve S ', splitting nodes, calculating new node coordinates, inserting the new node coordinates into the node sequence of the initial closed curve S' and generating a second closed curve; traversing the node sequence of the second closed curve, updating the node coordinates in the node sequence after the node movement and generating a second closed curveA three-closed curve; using formulasCalculating the energy value of the third closed curve and judging whether the energy value is smaller than the minimum curve energy value EminIf yes, then use formula Emin=EShdeAnd updating the minimum curve energy value, and repeatedly performing node splitting and node movement based on the third closed curve until the updated energy value of the closed curve is greater than the minimum curve energy value, if the energy value of the third closed curve is greater than the minimum energy function value, indicating that the curve has reached the target edge of the plot, and outputting the third closed curve as the plot boundary curve corresponding to the second internal point set.
Optionally, the node splitting the node sequence of the initial closed curve S' includes: calculating each point V of the node sequencei(xi,yi,zi) Two adjacent direction values of theta1=(zi=0)?3:(zi-1) and θ2=(zi=3)?0:(zi+1) if z is satisfiedi0 and xi>xi-1Or satisfy zi1 and yi>yi-1Or satisfy zi2 and xi<xi-1Or satisfy zi3 and yi<yi-1Condition (2) indicates the direction theta1Points to the outside of the curve to generate new node coordinates P1=(xi,yi,θ1) If z is satisfiedi0 and xi>xi+1Or satisfy zi1 and yi>yi+1Or satisfy zi2 and xi<xi+1Or satisfy zi3 and yi<yi+1Condition (2) indicates the direction theta2Points to the outside of the curve to generate new node coordinates P2=(xi,yi,θ2)。
Optionally, the node moving the node sequence of the second closed curve includes: for each of the node sequencesPoint Vi(xi,yi,zi) If z isiWhen the value is 0, then Vi′=(xi+r,yi0), if ziWhen 1, then Vi′=(xi,yi+ r, 1), if ziWhen 2, then Vi′=(xi-r,yi2) if z isiWhen it is 3, then Vi′=(xi,yi-r, 3); calculating ViAnd Vi' energy values E (i) and E (i) ', if E (i) ' < E (i), using the formula Vi=Vi', E (i) ═ E (i)' updates the node coordinates and their energy values.
Optionally, the S1 further includes: and after the panchromatic remote sensing image and the multispectral remote sensing image are obtained, image preprocessing is carried out on the panchromatic remote sensing image and the multispectral remote sensing image.
Optionally, the image preprocessing is performed on the panchromatic remote sensing image, namely, the panchromatic remote sensing image is subjected to orthorectification by using the DEM data.
Optionally, the image preprocessing the multispectral remote sensing image comprises: orthonormal, radiometric, and atmospheric corrections.
The method has the advantages that the first internal point set in the multispectral image is accurately acquired by using a visual interpretation method and is converted into the second internal point set in the panchromatic remote sensing image, the segmentation accuracy of the land parcel profile in the early stage is improved, and the smooth closed curve which is accurately segmented is generated through a T-Snake model based on the second internal point set and the panchromatic remote sensing image, so that the method is not limited by the size and the shape of the land parcel, a good extraction effect can be obtained at a weak boundary, the identification accuracy of the land parcel boundary is effectively improved, and the generation of broken graphic spots is reduced.
Drawings
FIG. 1 is a simplified flowchart of the method for extracting cultivated land and land parcel of satellite image of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood by those skilled in the art, the present invention will be further described in detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, a method for extracting cultivated land and land blocks by satellite images includes: s1: acquiring a panchromatic remote sensing image and a multispectral remote sensing image; s2: segmenting the multispectral remote sensing image based on a visual interpretation method to generate a plurality of first interior point sets; s3: constructing an energy function of a T-Snake model based on the panchromatic remote sensing image; s4: extracting a second interior point set corresponding to the panchromatic remote sensing image based on the first interior point set; s5: calculating a plot boundary curve corresponding to the second interior point set based on the second interior point set, the panchromatic remote sensing image and the energy function; s6: and repeating the steps S4-S5 until all the corresponding plot boundary curves of all the first interior point sets in the panchromatic remote sensing image are traversed, and generating a plurality of cultivated land plot images based on all the plot boundary curves.
Extracting a second interior point set corresponding to the panchromatic remote sensing image based on the first interior point set comprises two conditions: when the panchromatic remote sensing image and the multispectral remote sensing image are obtained by using the same satellite unit, the first interior point set can directly represent the second interior point set due to the fact that image data sources are the same; when different satellite units are used for acquiring the panchromatic remote sensing image and the multispectral remote sensing image, due to the fact that image data sources are different, the situations that satellite coordinate systems are different and visual angles are different exist, a coordinate transformation matrix needs to be calculated based on a plurality of satellite coordinate systems of a plurality of satellite units, and the coordinate transformation matrix and the first internal point set are calculated to generate the second internal point set, so that the problem that when certain acquired image accuracy of a certain satellite unit is low and is not credible is effectively solved, the accuracy of the extracted second internal point set is improved by introducing different satellite units for image acquisition. Meanwhile, by introducing a plurality of acquired images of satellite units, the accuracy of the extracted second internal point set can be improved by a cross-computing averaging method, such as: the method comprises the steps of obtaining a first panchromatic remote sensing image and a first multispectral remote sensing image by using a first satellite unit, obtaining a second panchromatic remote sensing image and a second multispectral remote sensing image by using a second satellite unit, calculating a first group of second interior point sets based on the first panchromatic remote sensing image and the second multispectral remote sensing image, calculating a second group of second interior point sets based on the second panchromatic remote sensing image and the first multispectral remote sensing image, and performing one-to-one corresponding fitting between interior points based on the first group of second interior point sets and the second group of second interior point sets to generate a final second interior point set, so that the accuracy of the extracted second interior point set is improved.
According to the method, the first internal point set in the multispectral image is accurately acquired by using a visual interpretation method and is converted into the second internal point set in the panchromatic remote sensing image, the segmentation accuracy of the land parcel profile in the early stage is improved, and the smooth closed curve which is accurately segmented is generated on the basis of the second internal point set and the panchromatic remote sensing image through the T-Snake model, so that the method is not limited by the size and the shape of the cultivated land parcel, a good extraction effect can be obtained at a weak boundary, the identification accuracy of the boundary of the cultivated land parcel is effectively improved, and the generation of broken image spots is reduced.
Further, the S2 includes: segmenting the multispectral remote sensing image based on the visual interpretation method to generate a plurality of cultivated land plot outlines; traversing all internal points of the cultivated land parcel profile to form the first internal point set P { (x) corresponding to the cultivated land parcel profilesn,yn)|n=1,2,...,N-1},(xn,yn) Is the coordinate of an inner point of the nth parcel. The visual interpretation method is used for segmenting the multispectral remote sensing image to generate a plurality of cultivated land plot outlines, and comprises the following steps: judging a cultivated land area based on the shape, color and texture difference between a cultivated land parcel and other background land features in the multispectral remote sensing image, and extracting a planar area with uniform texture and uniform color in the multispectral remote sensing image as the cultivated land area; judging whether boundary lines exist among different farmland regions, if so, extracting the contours of the boundary lines to generate a plurality of farmland land block contours, and if so, extracting the contours of the boundary lines to generate a plurality of farmland land block contoursIf not, generating a single arable land block profile based on a plurality of the arable land areas.
Further, the S3 includes: constructing a closed curve function S ═ V under a T-Snake model1,V2,...,VmCharacterizing a closed curve in the panchromatic remote sensing image, wherein Vi(xi,yi) Is a curve node, i ═ 1, 2.., m; constructing the energy function based on the closed curve functionWherein eta, gamma and lambda are proportionality coefficients, C is the geometric center of the current closed curve, and sigma and delta are the gray scale standard deviation and the range of the node neighborhood in the full-color image respectively,is node ViFull color image gradient of (a).
Further, the S5 includes: randomly extracting coordinate points (x) in the second set of interior pointsn,yn) And extracting four pixel points of which the coordinate is at a distance of r in the upper, lower, left and right directions, and sequentially connecting the four pixel points clockwise to construct an initial closed curve S { (x)n+r,yn,0),(xn,yn+r,1),(xn-r,yn,2),(xn,yn-r, 3) }, wherein the third attribute of each node represents the direction information of the node, i.e. 0 to the right, 1 to the bottom, 2 to the left and 3 to the top in clockwise order; traversing the node sequence of the initial closed curve S ', splitting nodes, calculating new node coordinates, inserting the new node coordinates into the node sequence of the initial closed curve S' and generating a second closed curve; traversing the node sequence of the second closed curve, updating the node coordinates in the node sequence after the node moves, and generating a third closed curve; using formulasCalculating the energy value of the third closed curve and judging the energy valueWhether or not it is less than the minimum curve energy value EminIf yes, then use formula Emin=ESnakeAnd updating the minimum curve energy value, and repeatedly performing node splitting and node movement based on the third closed curve until the updated energy value of the closed curve is greater than the minimum curve energy value, if the energy value of the third closed curve is greater than the minimum energy function value, indicating that the curve has reached the target edge of the plot, and outputting the third closed curve as the plot boundary curve corresponding to the second internal point set. In the first iteration, after the energy value of the third closed curve is calculated, the second iteration, namely the second node splitting and the node moving, is directly performed on the basis of the third closed curve, wherein the energy value of the third closed curve calculated in the first iteration is used as the minimum curve energy value E in the second iterationmin。
Specifically, the energy function is defined according to the gray level characteristics of the imageThe first term is called the spring force, the second term is the expansion force, and the third term is the image force, and the trend of the curve is controlled by the three forces after each iteration calculation. When the node is inside the parcel and away from the edge,smaller and e-ε(σ+δ)Larger, where the expansion force dominates, as the curve expands outward, the distance | V between the node and the geometric center of the curvei-C | becomes larger and thus the energy function value becomes smaller, so that the curve continues to move; when the node is at the boundary of the land parcel, e-ε(σ+δ)Rapidly decays to zeroAnd suddenly increasing, wherein the image force is dominant, the energy function value is not reduced continuously, and the curve is prompted to stop moving, so that the curve stays at the boundary of the land parcel. Therefore, the smooth closed curve is easily generated by utilizing the T-Snake modelAnd the curve is terminated at the image gradient abrupt change position, thereby effectively improving the accuracy of recognizing the boundary of the farmland block and breaking through the limitation of the size and the shape of the farmland block.
To facilitate understanding how to perform node splitting, specifically, performing node splitting on the node sequence of the initial closed curve S' includes: calculating each point V of the node sequencei(xi,yi,zi) Two adjacent direction values of theta1=(zi=0)?3:(zi-1) and θ2=(zi=3)?0:(zi+1) if z is satisfiedi0 and xi>xi-1Or satisfy zi1 and yi>yi-1Or satisfy zi2 and xi<xi-1Or satisfy zi3 and yi<yi-1Condition (2) indicates the direction theta1Points to the outside of the curve to generate new node coordinates P1=(xi,yi,θ1) If z is satisfiedi0 and xi>xi+1Or satisfy zi1 and yi>yi+1Or satisfy zi2 and xi<xi+1Or satisfy zi3 and yi<yi+1Condition (2) indicates the direction theta2Points to the outside of the curve to generate new node coordinates P2=(xi,yi,θ2). Wherein, theta1=(zi=0)?3:(zi-1) represents when ziWhen equal to 0 (i.e., the current node is to the right), let θ1Equal to 3, otherwise let θ1Is equal to zi-1;θ2=(zi=3)?0:(zi+1) denotes when z isiWhen the value is equal to 3 (i.e. the current node is upward), let θ2Equal to 0, otherwise let θ2Is equal to zi+1。
To facilitate understanding how to perform node movement, in particular, performing node movement on the node sequence of the second closed curve includes: for each point V of the sequence of nodesi(xi,yi,zi) If z isi=0,Then Vi′=(xi+r,yiO), if ziWhen 1, then Vi′=(xi,yi+ r, 1), if ziWhen 2, then Vi′=(xi-r,yi2) if z isiWhen it is 3, then Vi′=(xi,yi-r, 3); calculating ViAnd Vi' energy values E (i) and E (i) ', if E (i) ' < E (i), using the formula Vi=Vi', E (i) ═ E (i)' updates the node coordinates and their energy values.
Further, the S1 further includes: and after the panchromatic remote sensing image and the multispectral remote sensing image are obtained, image preprocessing is carried out on the panchromatic remote sensing image and the multispectral remote sensing image. Performing image preprocessing on the panchromatic remote sensing image, namely performing orthorectification on the panchromatic remote sensing image by adopting DEM data; the image preprocessing of the multispectral remote sensing image comprises the following steps: orthonormal, radiometric, and atmospheric corrections. The radiometric calibration can convert the DN value of the image into earth surface radiance data, and then the earth surface reflectivity data can be obtained through atmospheric correction.
The above is only a preferred embodiment of the present invention, and it should be noted that the above preferred embodiment should not be considered as limiting the present invention, and the protection scope of the present invention should be subject to the scope defined by the claims. It will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the spirit and scope of the invention, and these modifications and adaptations should be considered within the scope of the invention.
Claims (10)
1. A cultivated land plot extraction method of a satellite image is characterized by comprising the following steps:
s1: acquiring a panchromatic remote sensing image and a multispectral remote sensing image;
s2: segmenting the multispectral remote sensing image based on a visual interpretation method to generate a plurality of first interior point sets;
s3: constructing an energy function of a T-Snake model based on the panchromatic remote sensing image;
s4: extracting a second interior point set corresponding to the panchromatic remote sensing image based on the first interior point set;
s5: calculating a plot boundary curve corresponding to the second interior point set based on the second interior point set, the panchromatic remote sensing image and the energy function;
s6: and repeating the steps S4-S5 until all the corresponding plot boundary curves of all the first interior point sets in the panchromatic remote sensing image are traversed, and generating a plurality of cultivated land plot images based on all the plot boundary curves.
2. The cultivated land parcel extraction method according to claim 1, wherein said S2 comprises:
segmenting the multispectral remote sensing image based on the visual interpretation method to generate a plurality of cultivated land plot outlines;
and traversing all the interior points of the cultivated land block contour to form a first interior point set corresponding to the cultivated land block contours.
3. The arable land parcel extraction method of claim 2, wherein segmenting the multispectral remote sensing image based on the visual interpretation method to generate a plurality of arable land parcel profiles comprises:
judging a cultivated land area based on the shape, color and texture difference between a cultivated land parcel and other background land features in the multispectral remote sensing image, and extracting a planar area with uniform texture and uniform color in the multispectral remote sensing image as the cultivated land area;
judging whether boundary lines exist among different arable land areas or not, if so, respectively extracting the contours of the boundary lines to generate a plurality of arable land block contours, and if not, generating a single arable land block contour based on the plurality of arable land areas.
4. The cultivated land parcel extraction method according to claim 3, wherein said S3 comprises:
constructing a closed curve function S ═ V under a T-Snake model1,V2,...,VmCharacterizing a closed curve in the panchromatic remote sensing image, wherein Vi(xi,yi) Is a curve node, i ═ 1, 2.., m;
constructing the energy function based on the closed curve functionWherein η, γ, λ are proportionality coefficients, C is the geometric center of the current closed curve, and σ and δ are the gray scale standard deviation and range of the node neighborhood in the full-color image, respectively.Is node ViFull color image gradient of (a).
5. The cultivated land parcel extraction method according to claim 4, wherein said S5 comprises:
randomly extracting coordinate points (x) in the second set of interior pointsn,yn) And extracting four pixel points of which the coordinate is at a distance of r in the upper, lower, left and right directions, and sequentially connecting the four pixel points clockwise to construct an initial closed curve S { (x)n+r,yn,0),(xn,yn+r,1),(xn-r,yn,2),(xn,yn-r, 3) }, wherein the third attribute of each node represents the direction information of the node, i.e. 0 to the right, 1 to the bottom, 2 to the left and 3 to the top in clockwise order;
traversing the node sequence of the initial closed curve S ', splitting nodes, calculating new node coordinates, inserting the new node coordinates into the node sequence of the initial closed curve S' and generating a second closed curve;
traversing the node sequence of the second closed curve, updating the node coordinates in the node sequence after the node moves, and generating a third closed curve;
and calculating whether the energy value of the third closed curve is smaller than the minimum curve energy value, if so, updating the minimum curve energy value, and repeatedly performing node splitting and node movement on the basis of the third closed curve until the updated energy value of the closed curve is larger than the minimum curve energy value, and if so, outputting the third closed curve as a plot boundary curve corresponding to the second internal point set.
6. The cultivated land parcel extraction method according to claim 5, characterized in that node splitting of the node sequence of the initial closed curve S' comprises:
calculating each point V of the node sequencei(xi,yi,zi) Two adjacent direction values of theta1=(zi=0)?3:(zi-1) and θ2=(zi==3)?0:(zi+1),
If z is satisfiedi0 and xi>xi-1Or satisfy zi1 and yi>yi-1Or satisfy zi2 and xi<xi-1Or satisfy zi3 and yi<yi-1Condition (2), then new node coordinates P are generated1=(xi,yi,θ1),
If z is satisfiedi0 and xi>xi+1Or satisfy zi1 and yi>yi+1Or satisfy zi2 and xi<xi+1Or satisfy zi3 and yi<yi+1Condition (2), then new node coordinates P are generated2=(xi,yi,θ2)。
7. The arable land parcel extraction method of claim 5, wherein performing a node movement on the second closed curve's sequence of nodes comprises:
for each point V of the sequence of nodesi(xi,yi,zi) If z isi0, then V'i=(xi+r,yi0), if zi1, then V'i=(xi,yi+ r, 1), if zi2, then V'i=(xi-r,yi2) if z isi3, then V'i=(xi,yi-r,3);
Calculating ViAnd V'iIf E (i)' < E (i), using the formula Vi=V′iAnd E (i) ═ E (i)' updating the coordinates of the nodes and the energy values of the nodes.
8. The cultivated land parcel extraction method according to claim 1, wherein said S1 further comprises:
and after the panchromatic remote sensing image and the multispectral remote sensing image are obtained, image preprocessing is carried out on the panchromatic remote sensing image and the multispectral remote sensing image.
9. The cultivated land parcel extraction method according to claim 8, characterized in that the image pre-processing of said panchromatic remote sensing image is an orthorectification of said panchromatic remote sensing image using DEM data.
10. The cultivated land parcel extraction method according to claim 8, wherein image preprocessing of said multispectral remote sensing image comprises: orthonormal, radiometric, and atmospheric corrections.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011279619.2A CN112330700A (en) | 2020-11-16 | 2020-11-16 | Cultivated land plot extraction method of satellite image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011279619.2A CN112330700A (en) | 2020-11-16 | 2020-11-16 | Cultivated land plot extraction method of satellite image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112330700A true CN112330700A (en) | 2021-02-05 |
Family
ID=74318634
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011279619.2A Pending CN112330700A (en) | 2020-11-16 | 2020-11-16 | Cultivated land plot extraction method of satellite image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112330700A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117496374A (en) * | 2024-01-02 | 2024-02-02 | 天津财经大学 | Ecological tourism resource satellite remote sensing data batch processing and downloading system |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010087112A1 (en) * | 2009-01-27 | 2010-08-05 | 国立大学法人大阪大学 | Image analysis apparatus, image analysis method, image analysis program and recording medium |
CN105374024A (en) * | 2015-09-24 | 2016-03-02 | 四川航天系统工程研究所 | A method for extracting bridges over water from high resolution satellite images |
CN106548141A (en) * | 2016-11-01 | 2017-03-29 | 南京大学 | A kind of object-oriented farmland information extraction method based on the triangulation network |
CN107480706A (en) * | 2017-07-24 | 2017-12-15 | 中国农业大学 | A kind of seed production corn field remote sensing recognition method and device |
CN108171210A (en) * | 2018-01-18 | 2018-06-15 | 中国地质科学院矿产资源研究所 | Method and system for extracting remote sensing abnormal information of alteration of covered area of planting |
US20180189954A1 (en) * | 2016-12-30 | 2018-07-05 | International Business Machines Corporation | Method and system for crop recognition and boundary delineation |
CN109344810A (en) * | 2018-11-26 | 2019-02-15 | 国智恒北斗科技集团股份有限公司 | A kind of arable land use change monitoring method and system based on high score satellite remote sensing date |
CN109657610A (en) * | 2018-12-18 | 2019-04-19 | 北京航天泰坦科技股份有限公司 | A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images |
CN109933639A (en) * | 2019-03-22 | 2019-06-25 | 合肥工业大学 | A kind of multispectral image towards map overlay and full-colour image method for self-adaption amalgamation |
CN109960972A (en) * | 2017-12-22 | 2019-07-02 | 北京航天泰坦科技股份有限公司 | A kind of farm-forestry crop recognition methods based on middle high-resolution timing remotely-sensed data |
CN110310246A (en) * | 2019-07-05 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | A kind of cane -growing region remote sensing information extracting method based on three-line imagery |
CN110660089A (en) * | 2019-09-25 | 2020-01-07 | 云南电网有限责任公司电力科学研究院 | Satellite image registration method and device |
CN111882573A (en) * | 2020-07-31 | 2020-11-03 | 北京师范大学 | Cultivated land plot extraction method and system based on high-resolution image data |
-
2020
- 2020-11-16 CN CN202011279619.2A patent/CN112330700A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2010087112A1 (en) * | 2009-01-27 | 2010-08-05 | 国立大学法人大阪大学 | Image analysis apparatus, image analysis method, image analysis program and recording medium |
CN105374024A (en) * | 2015-09-24 | 2016-03-02 | 四川航天系统工程研究所 | A method for extracting bridges over water from high resolution satellite images |
CN106548141A (en) * | 2016-11-01 | 2017-03-29 | 南京大学 | A kind of object-oriented farmland information extraction method based on the triangulation network |
US20180189954A1 (en) * | 2016-12-30 | 2018-07-05 | International Business Machines Corporation | Method and system for crop recognition and boundary delineation |
CN107480706A (en) * | 2017-07-24 | 2017-12-15 | 中国农业大学 | A kind of seed production corn field remote sensing recognition method and device |
CN109960972A (en) * | 2017-12-22 | 2019-07-02 | 北京航天泰坦科技股份有限公司 | A kind of farm-forestry crop recognition methods based on middle high-resolution timing remotely-sensed data |
CN108171210A (en) * | 2018-01-18 | 2018-06-15 | 中国地质科学院矿产资源研究所 | Method and system for extracting remote sensing abnormal information of alteration of covered area of planting |
CN109344810A (en) * | 2018-11-26 | 2019-02-15 | 国智恒北斗科技集团股份有限公司 | A kind of arable land use change monitoring method and system based on high score satellite remote sensing date |
CN109657610A (en) * | 2018-12-18 | 2019-04-19 | 北京航天泰坦科技股份有限公司 | A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images |
CN109933639A (en) * | 2019-03-22 | 2019-06-25 | 合肥工业大学 | A kind of multispectral image towards map overlay and full-colour image method for self-adaption amalgamation |
CN110310246A (en) * | 2019-07-05 | 2019-10-08 | 广西壮族自治区基础地理信息中心 | A kind of cane -growing region remote sensing information extracting method based on three-line imagery |
CN110660089A (en) * | 2019-09-25 | 2020-01-07 | 云南电网有限责任公司电力科学研究院 | Satellite image registration method and device |
CN111882573A (en) * | 2020-07-31 | 2020-11-03 | 北京师范大学 | Cultivated land plot extraction method and system based on high-resolution image data |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117496374A (en) * | 2024-01-02 | 2024-02-02 | 天津财经大学 | Ecological tourism resource satellite remote sensing data batch processing and downloading system |
CN117496374B (en) * | 2024-01-02 | 2024-03-29 | 天津财经大学 | Ecological tourism resource satellite remote sensing data batch processing and downloading system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111798475B (en) | Indoor environment 3D semantic map construction method based on point cloud deep learning | |
Lucieer et al. | Existential uncertainty of spatial objects segmented from satellite sensor imagery | |
KR101404640B1 (en) | Method and system for image registration | |
CN112686935B (en) | Airborne sounding radar and multispectral satellite image registration method based on feature fusion | |
CN109871823B (en) | Satellite image ship detection method combining rotating frame and context information | |
CN104574347A (en) | On-orbit satellite image geometric positioning accuracy evaluation method on basis of multi-source remote sensing data | |
CN108428220A (en) | Satellite sequence remote sensing image sea island reef region automatic geometric correction method | |
CN112052783A (en) | High-resolution image weak supervision building extraction method combining pixel semantic association and boundary attention | |
CN109323697B (en) | Method for rapidly converging particles during starting of indoor robot at any point | |
CN105335965B (en) | Multi-scale self-adaptive decision fusion segmentation method for high-resolution remote sensing image | |
CN115049925B (en) | Field ridge extraction method, electronic device and storage medium | |
CN110728706B (en) | SAR image fine registration method based on deep learning | |
CN112084871B (en) | High-resolution remote sensing target boundary extraction method based on weak supervised learning | |
CN112766155A (en) | Deep learning-based mariculture area extraction method | |
CN112837320B (en) | Remote sensing image semantic segmentation method based on parallel hole convolution | |
CN113298742A (en) | Multi-modal retinal image fusion method and system based on image registration | |
CN108428236B (en) | Multi-target SAR image segmentation method based on feature fair integration | |
CN112330700A (en) | Cultivated land plot extraction method of satellite image | |
CN109741337B (en) | Region merging watershed color remote sensing image segmentation method based on Lab color space | |
CN109741358B (en) | Superpixel segmentation method based on adaptive hypergraph learning | |
CN113409332B (en) | Building plane segmentation method based on three-dimensional point cloud | |
Cai et al. | Improving agricultural field parcel delineation with a dual branch spatiotemporal fusion network by integrating multimodal satellite data | |
CN110276270B (en) | High-resolution remote sensing image building area extraction method | |
CN111862005A (en) | Method and system for accurately positioning tropical cyclone center by using synthetic radar image | |
CN115457022B (en) | Three-dimensional deformation detection method based on live-action three-dimensional model front-view image |
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 |