CN117934518B - Remote sensing image segmentation method and system for ocean homeland space detailed planning - Google Patents
Remote sensing image segmentation method and system for ocean homeland space detailed planning Download PDFInfo
- Publication number
- CN117934518B CN117934518B CN202410323137.4A CN202410323137A CN117934518B CN 117934518 B CN117934518 B CN 117934518B CN 202410323137 A CN202410323137 A CN 202410323137A CN 117934518 B CN117934518 B CN 117934518B
- Authority
- CN
- China
- Prior art keywords
- local
- remote sensing
- ocean
- scale
- sensing 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000003709 image segmentation Methods 0.000 title claims abstract description 25
- 230000006870 function Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000004927 fusion Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a remote sensing image segmentation method and a remote sensing image segmentation system for marine homeland space detailed planning, comprising the following steps: acquiring satellite remote sensing images of the ocean to be developed and a plurality of local remote sensing images of the ocean to be developed; according to the significance of each characteristic point of the local remote sensing image of each ocean to be developed, acquiring all target characteristic points of the local remote sensing image of each ocean to be developed; acquiring a target matching scale of each local area according to the target degree of each local area under each scale; and planning ocean resources in detail according to the clear satellite remote sensing image of the ocean to be developed. The method obtains clear satellite remote sensing images and realizes planning of ocean resources.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a remote sensing image segmentation method and a remote sensing image segmentation system for marine homeland space detailed planning.
Background
Resource information in the ocean can be obtained through the satellite remote sensing image, but the resolution of the remote sensing image obtained by the satellite is low although the remote sensing image is wide, so that planning of ocean resources is affected; although a high-definition remote sensing image can be obtained through the unmanned aerial vehicle with high resolution, the shooting range is smaller, and the planning of ocean resources is not facilitated; the two acquired images can be fused, the high-resolution image is fused into a large-range satellite remote sensing image, the resolution of the satellite remote sensing image is improved, and the planning of ocean resources is more in line with the actual situation.
When the image fusion is carried out, due to the fact that the sizes of the sea areas shot by the high-resolution remote sensing image and the satellite remote sensing image are different, scale scaling is not considered by the existing image fusion algorithm, so that a large amount of redundancy exists on the high-resolution extracted corner points and the low-resolution extracted corner points when the corner points are detected, matching accuracy is affected, and feature point matching accuracy in the high-resolution remote sensing image is reduced.
Disclosure of Invention
In order to solve the problems, the invention provides a remote sensing image segmentation method and a remote sensing image segmentation system for marine homeland space detailed planning.
The embodiment of the invention provides a remote sensing image segmentation method for marine homeland space detailed planning, which comprises the following steps of:
acquiring satellite remote sensing images of the ocean to be developed and a plurality of local remote sensing images of the ocean to be developed; dividing a satellite remote sensing image of the ocean to be developed into a plurality of local areas, wherein each local remote sensing image of the ocean to be developed corresponds to one local area of the satellite remote sensing image of the ocean to be developed;
detecting the characteristic points of each local remote sensing image to obtain all the characteristic points of each local remote sensing image and each local remote sensing image under a plurality of different scales; according to the area change of the local remote sensing images under different scales, the saliency of each characteristic point of each local remote sensing image is obtained; screening according to the significance of the feature points to obtain all target feature points of each local remote sensing image;
Detecting characteristic points of each local area of the satellite remote sensing image to obtain all characteristic points of each local area and each local area under a plurality of different scales; according to the gray level change of each feature point in the neighborhood of the local area under each scale, obtaining the remarkable feature of each feature point in the local area under each scale; obtaining the target degree of the local area under each scale according to the area difference between the local area under each scale and the local area under the original scale and the obvious characteristic difference of the characteristic points; screening according to the target degree of the local area under each scale to obtain the target matching scale of each local area;
according to all target feature points of the local remote sensing image of each ocean to be developed and all feature points of each local area under the target matching scale, matching and fusing are carried out, so that a clear satellite remote sensing image of the ocean to be developed is obtained; the clear satellite remote sensing image of the ocean to be developed is divided into an ocean unit and a land unit.
Preferably, the obtaining the saliency of each feature point of each local remote sensing image according to the area change of the local remote sensing image under different scales comprises the following specific methods:
Acquiring the effective degree of each local remote sensing image of the ocean to be developed under each scale according to the area change of the local remote sensing image under different scales; obtaining local characteristic values of each characteristic point in each local remote sensing image of ocean to be developed under each scale; then the first The/>, of the local remote sensing image of each ocean to be developedThe calculation method of the significance of each feature point comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of the local remote sensing image of each ocean to be developedSignificance of individual feature points; Represents the/>, at all scales The total number of local remote sensing images of the ocean to be developed; /(I)Represents the/>/>, At the individual scaleThe effectiveness degree of the local remote sensing images of the ocean to be developed; /(I)Represents the/>The characteristic points are at the/>/>, At the individual scaleLocal characteristic values in local remote sensing images of the ocean to be developed; /(I)Representing a linear normalization function.
Preferably, the obtaining the effective degree of each local remote sensing image of ocean to be developed under each scale according to the area change of the local remote sensing image under different scales comprises the following specific methods:
Will be the first A local remote sensing image of each ocean to be developed corresponds to a local area of a satellite remote sensing image of the ocean to be developed as the/>, of the satellite remote sensing image of the ocean to be developedLocal area, the second/>, of the satellite remote sensing image of the ocean to be developedArea of the local area as the/>Target area of local remote sensing image of ocean to be developed, then the/>/>, At the individual scaleThe method for calculating the effective degree of the local remote sensing images of the ocean to be developed comprises the following steps:
；
In the method, in the process of the invention, Represents the/>/>, At the individual scaleAreas of local remote sensing images of the ocean to be developed; /(I)Represents the/>Target areas of local remote sensing images of the ocean to be developed; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant is represented.
Preferably, the obtaining the local feature value in the local remote sensing image of each ocean to be developed of each feature point under each scale includes the following specific steps:
Presetting two parameters Couple/>, using SIFT algorithmDetecting characteristic points of local remote sensing images of the ocean to be developed, and judging the/>, under each scale, of each characteristic pointWhether the local remote sensing images of the ocean to be developed belong to local extremum points or not, if the feature points belong to the local extremum points, the local feature values of the feature points are/>If the feature point does not belong to the local extremum point, the local feature value of the feature point is/>。
Preferably, the method for screening according to the significance of the feature points to obtain all target feature points of each local remote sensing image includes the following specific steps:
Presetting a threshold value If/>The/>, of the local remote sensing image of each ocean to be developedThe significance of each feature point is greater than or equal to a threshold/>Will be/>The feature points are marked as the/>Target feature points of the local remote sensing image of the ocean to be developed.
Preferably, the method for obtaining the salient features of each feature point in the local area under each scale according to the gray level variation of each feature point in the neighborhood of the local area under each scale includes the following specific steps:
Acquisition of the first The gray level difference degree of each eight neighborhood directions of the characteristic points and the gray level characteristic difference of each neighborhood direction are the/>The/>, of each feature point under the original scaleThe method for calculating the salient features in the local areas comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; Represents the/> The/>, of each feature point under the original scaleMaximum values of all neighborhood-oriented gray feature differences in the local regions; /(I)Represents the/>The/>, of each feature point under the original scaleVariance of gray scale difference degree in all eight neighborhood directions in each local area; /(I)An exponential function based on a natural constant; /(I)Representing a linear normalization function.
Preferably, the acquiring a firstThe specific method for the gray level difference degree of each eight neighborhood directions of the characteristic points and the gray level characteristic difference of each neighborhood direction comprises the following steps:
For the first The/>, of each feature point under the original scaleIn the local area, by the/>Establishing eight neighborhood windows with the feature points as central points, and setting the/>Pixel point and/>, in any eight neighborhood directions, of each feature pointAbsolute value of difference of gray values of each feature point as/>The gray level difference degree of the eight neighborhood directions of the characteristic points is as followsThe absolute value of the difference value of gray level difference degrees in any two adjacent eight neighborhood directions of the characteristic points is taken as the/>Gray scale characteristic difference in a neighborhood direction of each characteristic point.
Preferably, the specific formula for obtaining the target degree of the local area under each scale according to the area difference between the local area under each scale and the local area under the original scale and the significant characteristic difference of the feature points is as follows:
；
In the method, in the process of the invention, Represents the/>/>, At the individual scaleTarget degree of the individual local areas; /(I)Represents the first/>, at the original scaleThe area of the individual local areas; /(I)Represents the/>/>, At the individual scaleThe area of the individual local areas; /(I)Represent the firstThe number of all feature points of the local area; /(I)Represents the/>The characteristic points are at the/>Scale/>Significant features in the individual local regions; /(I)Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant is represented.
Preferably, the target matching scale of each local area is obtained by screening according to the target degree of the local area under each scale, which comprises the following specific steps:
If at first />, At the individual scaleThe target degree of the local area is the largest, the/>The individual scale is taken as the/>The target under the individual local regions matches the scale.
The invention also provides a remote sensing image segmentation system for the marine homeland space detailed planning, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one of the steps of the remote sensing image segmentation method for the marine homeland space detailed planning when executing the computer program.
The technical scheme of the invention has the beneficial effects that: according to the method, all target characteristic points of each local remote sensing image are obtained through screening according to the salience of the characteristic points, so that the characteristic points with high salience are accurately found out in the local remote sensing image, and the accuracy of characteristic point matching in the local remote sensing image is improved; and screening according to the target degree of the local area under each scale to obtain the target matching scale of each local area, so that the problem of scaling of the image scale is solved, a clear satellite remote sensing image is obtained, and planning of ocean resources is realized.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flow chart of steps of a remote sensing image segmentation method for marine homeland space detailed planning of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of the remote sensing image segmentation method and system for marine homeland space detailed planning according to the invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a remote sensing image segmentation method and a remote sensing image segmentation system for marine homeland space detailed planning, which are specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a remote sensing image segmentation method for marine homeland space detailed planning according to an embodiment of the present invention is shown, the method includes the following steps:
Step S001: and acquiring satellite remote sensing images of the ocean to be developed and a plurality of local remote sensing images of the ocean to be developed.
Specifically, in order to implement the remote sensing image segmentation method for marine homeland space detailed planning provided in this embodiment, firstly, a satellite remote sensing image of a sea to be developed and a plurality of local remote sensing images of the sea to be developed need to be acquired, and the specific process is as follows:
Presetting a parameter Wherein the present embodiment is described as/>To describe the example, the present embodiment is not particularly limited, wherein/>Depending on the particular implementation.
The satellite remote sensing images of the ocean to be developed are acquired through satellites, and a plurality of high-resolution local ocean remote sensing images of the ocean to be developed are acquired through unmanned aerial vehicles, so that the size is reducedThe multiplied high-resolution local ocean remote sensing image of the ocean to be developed is recorded as the local ocean remote sensing image of the ocean to be developed.
The ocean region to be developed, which is shot by the satellite remote sensing image of the ocean to be developed, is uniformly divided intoThe unmanned aerial vehicle comprises a plurality of local areas, wherein each local area is shot by adopting the same height and angle for each local area unmanned aerial vehicle; thus, each local remote sensing image of the ocean to be developed corresponds to a local area of the satellite remote sensing image of the ocean to be developed.
So far, the satellite remote sensing image of the ocean to be developed and a plurality of local remote sensing images of the ocean to be developed are obtained through the method.
Step S002: and acquiring all target characteristic points of the local remote sensing image of each ocean to be developed according to the significance of each characteristic point of the local remote sensing image of each ocean to be developed.
It should be noted that, the two kinds of collected image information are fused, because the sizes of the areas of the ocean to be developed, which are captured by the two kinds of collected images, are inconsistent, the collected high-resolution local remote sensing image needs to be scaled, the scaled image information is matched with the whole low-resolution satellite remote sensing, the matched images are fused, the fusion is not simply carried out according to the gray weighting of the matched pixel points, but the low-resolution image information is interpolated through the high-resolution remote sensing information to improve the resolution of the images, for example, the sizes of the two kinds of images are consistent and are both images with the size of 500×1000, the size of the sea area, which is captured by the satellite image, is 1000 m×1000 m, the size of the sea area, which is captured by the unmanned aerial vehicle is 10m×10m, the image, which is captured by the unmanned aerial vehicle needs to be scaled to the size of 5×10m, and then the remote sensing data, which is captured by the satellite, is matched.
It should be further noted that, when the high resolution image and the low resolution image are matched, the two images can be matched through feature points with unchanged scales, and when the collected image information is detected, since the high resolution local remote sensing image contains a large amount of detail texture information relative to the low resolution remote sensing image, the detail texture information does not exist in the extracted feature points in the image with low resolution, and the redundant feature points can influence the accuracy of the image matching to be detected.
Presetting two parametersWherein the present embodiment is described as/>To describe the example, the present embodiment is not particularly limited, wherein/>Depending on the particular implementation.
Specifically, in this embodiment, the first of a plurality of local remote sensing images of the ocean to be developedThe local remote sensing image of the ocean to be developed is described as an example, and the/>A local remote sensing image of each ocean to be developed corresponds to a local area of a satellite remote sensing image of the ocean to be developed as the/>, of the satellite remote sensing image of the ocean to be developedLocal area, the second/>, of the satellite remote sensing image of the ocean to be developedArea of the local area as the/>Target areas of local remote sensing images of the ocean to be developed are converted into SIFT (Scale-INVARIANT FEATURE TRANSFORM) algorithm by utilizing Scale-invariant features, and the method is used for the first/>Detecting characteristic points of local remote sensing images of the ocean to be developed, constructing a scale pyramid, and obtaining the/>All characteristic points of local remote sensing images of ocean to be developed and the/>, under a plurality of different scalesLocal remote sensing images of the ocean to be developed and judging the/>, under each scale, of each characteristic pointWhether the local remote sensing images of the ocean to be developed belong to local extreme points or not, and for the first/>, under any scale, of any one feature pointIn the local remote sensing images of the ocean to be developed, if the characteristic points belong to local extremum points, the local characteristic values of the characteristic points are/>If the feature points do not belong to local extreme points, the local feature values of the feature points are/>; Then/>The/>, of the local remote sensing image of each ocean to be developedThe calculation method of the significance of each feature point comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of the local remote sensing image of each ocean to be developedSignificance of individual feature points; Represents the/>, at all scales The total number of local remote sensing images of the ocean to be developed; /(I)Represents the/>/>, At the individual scaleAreas of local remote sensing images of the ocean to be developed; /(I)Represents the/>Target areas of local remote sensing images of the ocean to be developed; /(I)Represents the/>The characteristic points are at the/>/>, At the individual scaleLocal characteristic values in local remote sensing images of the ocean to be developed; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant; /(I)Representing a linear normalization function.
It should be noted that the number of the substrates,Represents the/>/>, At the individual scaleThe greater the value of the effective degree of the local remote sensing images of the ocean to be developed is, the higher the contribution degree of the scale transformation in the whole level is, and the level is closer to the scale of the satellite image; obtaining the/>, using SIFT algorithmAll feature points and the/>, under a plurality of scales, of local remote sensing images of a sea to be developedLocal remote sensing image of each ocean to be developed and the/>, under each scale, of each characteristic pointWhether the local remote sensing images of the ocean to be developed belong to the local extreme points is the prior art, and the embodiment is not repeated here.
Presetting a threshold valueWherein the present embodiment is described as/>To describe the example, the present embodiment is not particularly limited, wherein/>Depending on the particular implementation.
Specifically, if the firstThe/>, of the local remote sensing image of each ocean to be developedThe significance of each feature point is greater than or equal to a threshold/>Will be/>The feature points are marked as the/>Target feature points of the local remote sensing image of the ocean to be developed; and further obtaining all target characteristic points of the local remote sensing image of each ocean to be developed.
So far, all target characteristic points of the local remote sensing image of each ocean to be developed are obtained through the method.
Step S003: and acquiring the target matching scale of each local area according to the target degree of each local area under each scale.
It should be noted that, when the image fusion is performed, the number of pixel points in the two images with the scale changed is inconsistent, and the direct gray weighting of the pixel level is performed to recalculate the gray value, so that a clear satellite remote sensing image cannot be obtained; the matched satellite remote sensing image can be interpolated through the existing high-resolution unmanned aerial vehicle remote sensing image, so that the number of pixels of the matched part of the satellite remote sensing image is increased, and the resolution of the part of the area is increased; when the matching position interpolation calculation is performed, as the image shot by the unmanned aerial vehicle is not in a perfect splicing state at the joint of the two images, each image of the unmanned aerial vehicle can be overlapped for ensuring the integrity, and a satellite image in an overlapped area can be provided with a plurality of unmanned aerial vehicle images in an area matching manner, so that the interpolation calculation is not facilitated.
Specifically, the scale-invariant feature transform SIFT algorithm is utilized to develop the satellite remote sensing image of the oceanFeature point detection is carried out on each local area, a scale pyramid is constructed, and the/> isobtainedAll feature points of the local regions and the/>, at several different scalesLocal regions.
For the firstThe/>, of each feature point under the original scaleIn the local area, by the/>Establishing eight neighborhood windows with the feature points as central points, and setting the/>Pixel point and/>, in any eight neighborhood directions, of each feature pointAbsolute value of difference of gray values of each feature point as/>The gray level difference degree of the eight neighborhood directions of the characteristic points is as followsThe absolute value of the difference value of gray level difference degrees in any two adjacent eight neighborhood directions of the characteristic points is taken as the/>One neighborhood direction gray scale characteristic difference of each characteristic point is the/>The/>, of each feature point under the original scaleThe method for calculating the salient features in the local areas comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; Represents the/> The/>, of each feature point under the original scaleMaximum values of all neighborhood-oriented gray feature differences in the local regions; /(I)Represents the/>The/>, of each feature point under the original scaleVariance of gray scale difference degree in all eight neighborhood directions in each local area; /(I)An exponential function based on a natural constant; /(I)Representing a linear normalization function.
Further, the first/>, At the individual scaleThe method for calculating the target degree of each local area comprises the following steps:
；
In the method, in the process of the invention, Represents the/>/>, At the individual scaleTarget degree of the individual local areas; /(I)Represents the first/>, at the original scaleThe area of the individual local areas; /(I)Represents the/>/>, At the individual scaleThe area of the individual local areas; /(I)Represents the/>The number of all feature points of the local area; /(I)Represents the/>The characteristic points are at the/>Scale/>Significant features in the individual local regions; /(I)Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant is represented.
It should be noted that the number of the substrates,Indicating that all feature points are at the/>Scale/>The greater the value of the degree of change in the salient features in the individual local regions, the less likely it is that the greater the cumulative difference in the salient changes will match as a target scale.
Specifically, if the first/>, At the individual scaleThe target degree of the local area is the largest, the/>The individual scale is taken as the/>The target under the individual local regions matches the scale.
So far, the target matching scale of each local area is obtained through the method.
Step S004: and planning ocean resources in detail according to the clear satellite remote sensing image of the ocean to be developed.
Specifically, the K-nearest neighbor matching algorithm is utilized to pass throughAll target feature points of local remote sensing images of the ocean to be developed and the/>, under the target matching scaleMatching all characteristic points of each local area to obtain all matching point pairs, and performing image fusion by using a Laplacian pyramid algorithm according to all the matching point pairs to obtain a fused/>And forming a fused image by using all the fused local areas as a clear satellite remote sensing image of the ocean to be developed.
Presetting a threshold valueWherein the present embodiment is described as/>To describe the example, the present embodiment is not particularly limited, wherein/>Depending on the particular implementation.
Further, the gray value in the clear satellite remote sensing image of the ocean to be developed is smaller than or equal to the threshold valueThe region formed by all pixel points is used as an ocean unit, and the gray value in the clear satellite remote sensing image of the ocean to be developed is larger than a threshold value/>The area formed by all the pixels is used as a land unit.
The K-nearest neighbor matching algorithm and the laplacian pyramid algorithm are related to the prior art, and are not described in detail here.
Through the steps, the remote sensing image segmentation for the marine homeland space detailed planning is completed.
Presetting two flow parametersWherein the present embodiment is described as/>To describe the example, the present embodiment is not particularly limited, wherein/>Depending on the particular implementation.
Taking the edge line between any adjacent ocean unit and land unit as one coastline of the ocean to be developed, and acquiring all coastlines of the ocean to be developed; for any coastline, marking the highway closest to the coastline in the coastline corresponding land unit as a target highway of the coastline; acquiring the traffic flow of a target highway of the coastline in the last two years through a vehicle-mounted GPS, acquiring the total number of all ships passing through the ocean units of the coastline in the last two years through the ship-mounted GPS, marking the total number as a first number, taking the ratio of the first number to the total number of two years as the ship flow, and if the traffic flow is larger than the flow parameterAnd the ship flow is greater than the flow parameter/>Marking the coastline as a coastline to be planned; acquiring all coastlines to be planned of the ocean to be developed; and carrying out sea-land integrated detailed planning on all coastlines to be planned of the sea to be developed.
It should be noted that, the process of performing the detailed planning of all coastlines to be planned of the ocean to be developed is a well-known content of "basic problem and path thinking in land, space, land and sea overall planning", and this embodiment is not described in detail here.
The invention also provides a remote sensing image segmentation system for the marine homeland space detailed planning, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the steps of the remote sensing image segmentation method for the marine homeland space detailed planning in the steps S001 to S004 are realized when the processor executes the computer program.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (4)
1. The remote sensing image segmentation method for the marine homeland space detailed planning is characterized by comprising the following steps of:
acquiring satellite remote sensing images of the ocean to be developed and a plurality of local remote sensing images of the ocean to be developed; dividing a satellite remote sensing image of the ocean to be developed into a plurality of local areas, wherein each local remote sensing image of the ocean to be developed corresponds to one local area of the satellite remote sensing image of the ocean to be developed;
detecting the characteristic points of each local remote sensing image to obtain all the characteristic points of each local remote sensing image and each local remote sensing image under a plurality of different scales; according to the area change of the local remote sensing images under different scales, the saliency of each characteristic point of each local remote sensing image is obtained; screening according to the significance of the feature points to obtain all target feature points of each local remote sensing image;
Detecting characteristic points of each local area of the satellite remote sensing image to obtain all characteristic points of each local area and each local area under a plurality of different scales; according to the gray level change of each feature point in the neighborhood of the local area under each scale, obtaining the remarkable feature of each feature point in the local area under each scale; obtaining the target degree of the local area under each scale according to the area difference between the local area under each scale and the local area under the original scale and the obvious characteristic difference of the characteristic points; screening according to the target degree of the local area under each scale to obtain the target matching scale of each local area;
According to all target feature points of the local remote sensing image of each ocean to be developed and all feature points of each local area under the target matching scale, matching and fusing are carried out, so that a clear satellite remote sensing image of the ocean to be developed is obtained; dividing a clear satellite remote sensing image of a sea to be developed into a sea unit and a land unit;
The method for acquiring the saliency of each characteristic point of each local remote sensing image according to the area change of the local remote sensing image under different scales comprises the following specific steps:
Acquiring the effective degree of each local remote sensing image of the ocean to be developed under each scale according to the area change of the local remote sensing image under different scales; obtaining local characteristic values of each characteristic point in each local remote sensing image of ocean to be developed under each scale; then the first The/>, of the local remote sensing image of each ocean to be developedThe calculation method of the significance of each feature point comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of the local remote sensing image of each ocean to be developedSignificance of individual feature points; Represents the/>, at all scales The total number of local remote sensing images of the ocean to be developed; /(I)Represents the/>/>, At the individual scaleThe effectiveness degree of the local remote sensing images of the ocean to be developed; /(I)Represents the/>The characteristic points are at the/>/>, At the individual scaleLocal characteristic values in local remote sensing images of the ocean to be developed; /(I)Representing a linear normalization function;
according to the area change of the local remote sensing image under different scales, the effective degree of each local remote sensing image of the ocean to be developed under each scale is obtained, and the method comprises the following specific steps:
Will be the first A local remote sensing image of each ocean to be developed corresponds to a local area of a satellite remote sensing image of the ocean to be developed as the/>, of the satellite remote sensing image of the ocean to be developedLocal area, the first satellite remote sensing image of ocean to be developedArea of the local area as the/>Target area of local remote sensing image of ocean to be developed, then the/>/>, At the individual scaleThe method for calculating the effective degree of the local remote sensing images of the ocean to be developed comprises the following steps:
；
In the method, in the process of the invention, Represents the/>/>, At the individual scaleAreas of local remote sensing images of the ocean to be developed; /(I)Represents the/>Target areas of local remote sensing images of the ocean to be developed; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant;
the method for acquiring the local characteristic value of each local remote sensing image of each ocean to be developed of each characteristic point under each scale comprises the following specific steps:
Presetting two parameters Couple/>, using SIFT algorithmDetecting characteristic points of local remote sensing images of the ocean to be developed, and judging the/>, under each scale, of each characteristic pointWhether the local remote sensing images of the ocean to be developed belong to local extremum points or not, if the feature points belong to the local extremum points, the local feature values of the feature points are/>If the feature point does not belong to the local extremum point, the local feature value of the feature point is/>；
According to the gray level change of each feature point in the neighborhood of the local area under each scale, the remarkable feature of each feature point in the local area under each scale is obtained, and the specific method comprises the following steps:
Acquisition of the first The gray level difference degree of each eight neighborhood directions of the characteristic points and the gray level characteristic difference of each neighborhood direction are the/>The/>, of each feature point under the original scaleThe method for calculating the salient features in the local areas comprises the following steps:
；
In the method, in the process of the invention, Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; Represents the/> The/>, of each feature point under the original scaleMaximum values of all neighborhood-oriented gray feature differences in the local regions; /(I)Represents the/>The/>, of each feature point under the original scaleVariance of gray scale difference degree in all eight neighborhood directions in each local area; /(I)An exponential function based on a natural constant; /(I)Representing a linear normalization function;
The acquisition of the first The specific method for the gray level difference degree of each eight neighborhood directions of the characteristic points and the gray level characteristic difference of each neighborhood direction comprises the following steps:
For the first The/>, of each feature point under the original scaleIn the local area, by the/>Establishing eight neighborhood windows with the feature points as central points, and setting the/>Pixel point and/>, in any eight neighborhood directions, of each feature pointAbsolute value of difference of gray values of each feature point as/>The gray level difference degree of the eight neighborhood directions of the characteristic points is as followsThe absolute value of the difference value of gray level difference degrees in any two adjacent eight neighborhood directions of the characteristic points is taken as the/>Gray scale characteristic difference in one neighborhood direction of each characteristic point;
The specific formula for obtaining the target degree of the local area under each scale according to the area difference between the local area under each scale and the local area under the original scale and the obvious characteristic difference of the characteristic points is as follows:
；
In the method, in the process of the invention, Represents the/>/>, At the individual scaleTarget degree of the individual local areas; /(I)Representing the first dimension at the original scaleThe area of the individual local areas; /(I)Represents the/>/>, At the individual scaleThe area of the individual local areas; /(I)Represents the/>The number of all feature points of the local area; /(I)Represents the/>The characteristic points are at the/>Scale/>Significant features in the individual local regions; /(I)Represents the/>The/>, of each feature point under the original scaleSignificant features in the individual local regions; /(I)The representation takes absolute value; /(I)An exponential function based on a natural constant is represented.
2. The remote sensing image segmentation method for marine homeland space detailed planning of claim 1, wherein the screening according to the saliency of the feature points obtains all target feature points of each local remote sensing image, and the specific method comprises the following steps:
Presetting a threshold value If/>The/>, of the local remote sensing image of each ocean to be developedThe significance of each feature point is greater than or equal to a threshold/>Will be/>The feature points are marked as the/>Target feature points of the local remote sensing image of the ocean to be developed.
3. The remote sensing image segmentation method for marine homeland space detailed planning according to claim 1, wherein the screening according to the size of the target degree of the local area under each scale to obtain the target matching scale of each local area comprises the following specific steps:
If at first />, At the individual scaleThe target degree of the local area is the largest, the/>The individual scale is taken as the/>The target under the individual local regions matches the scale.
4. Remote sensing image segmentation system for marine homeland space detailed planning, comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that the processor, when executing the computer program, realizes the steps of the remote sensing image segmentation method for marine homeland space detailed planning according to any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410323137.4A CN117934518B (en) | 2024-03-21 | 2024-03-21 | Remote sensing image segmentation method and system for ocean homeland space detailed planning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410323137.4A CN117934518B (en) | 2024-03-21 | 2024-03-21 | Remote sensing image segmentation method and system for ocean homeland space detailed planning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117934518A CN117934518A (en) | 2024-04-26 |
| CN117934518B true CN117934518B (en) | 2024-06-04 |
Family
ID=90766736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410323137.4A Active CN117934518B (en) | 2024-03-21 | 2024-03-21 | Remote sensing image segmentation method and system for ocean homeland space detailed planning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117934518B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105469415A (en) * | 2015-12-28 | 2016-04-06 | 云南师范大学 | Multi-view remote sensing image fusion method |
| CN110866926A (en) * | 2019-10-24 | 2020-03-06 | 北京航空航天大学 | Infrared remote sensing image rapid and fine sea-land segmentation method |
| WO2021077847A1 (en) * | 2019-10-23 | 2021-04-29 | 北京建筑大学 | Seawater-polluted area identification method based on high-resolution remote-sensing image, and device |
| KR102594363B1 (en) * | 2022-09-30 | 2023-10-26 | (주) 아이렘기술개발 | Method and system for providing marine debris information using images of satellite and drone |
| CN117522884A (en) * | 2024-01-05 | 2024-02-06 | 武汉理工大学三亚科教创新园 | Ocean remote sensing image semantic segmentation method and device and electronic equipment |
| CN117541654A (en) * | 2024-01-09 | 2024-02-09 | 广东泰一高新技术发展有限公司 | Detail enhancement method for high-resolution remote sensing image |
-
2024
- 2024-03-21 CN CN202410323137.4A patent/CN117934518B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105469415A (en) * | 2015-12-28 | 2016-04-06 | 云南师范大学 | Multi-view remote sensing image fusion method |
| WO2021077847A1 (en) * | 2019-10-23 | 2021-04-29 | 北京建筑大学 | Seawater-polluted area identification method based on high-resolution remote-sensing image, and device |
| CN110866926A (en) * | 2019-10-24 | 2020-03-06 | 北京航空航天大学 | Infrared remote sensing image rapid and fine sea-land segmentation method |
| KR102594363B1 (en) * | 2022-09-30 | 2023-10-26 | (주) 아이렘기술개발 | Method and system for providing marine debris information using images of satellite and drone |
| CN117522884A (en) * | 2024-01-05 | 2024-02-06 | 武汉理工大学三亚科教创新园 | Ocean remote sensing image semantic segmentation method and device and electronic equipment |
| CN117541654A (en) * | 2024-01-09 | 2024-02-09 | 广东泰一高新技术发展有限公司 | Detail enhancement method for high-resolution remote sensing image |
Non-Patent Citations (1)
| Title |
|---|
| 面向海洋应用的无人机遥感图像配准研究;王飞;高小伟;高宁;赵建华;吴合风;孟庆辉;;测绘通报;20171125(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117934518A (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110298298B (en) | Target detection and target detection network training method, device and equipment | |
| Yahyanejad et al. | A fast and mobile system for registration of low-altitude visual and thermal aerial images using multiple small-scale UAVs | |
| WO2017071160A1 (en) | Sea-land segmentation method and system for large-size remote-sensing image | |
| CN109712071B (en) | Unmanned aerial vehicle image splicing and positioning method based on track constraint | |
| CN112163995B (en) | Splicing generation method and device for oversized aerial strip images | |
| CN114396945B (en) | Unmanned ship edge cleaning path planning method, device, equipment and storage medium | |
| Lee et al. | Accurate registration using adaptive block processing for multispectral images | |
| CN115205118A (en) | Underwater image splicing method and device, computer equipment and storage medium | |
| CN112883971A (en) | SAR image ship target detection method based on deep learning | |
| CN114265427A (en) | Inspection unmanned aerial vehicle auxiliary navigation system and method based on infrared image matching | |
| CN107169412B (en) | Remote sensing image harbor-berthing ship detection method based on mixed model decision | |
| CN110174109B (en) | Unmanned ship multi-element combined navigation method based on sea-air coordination | |
| CN114266947A (en) | Classification method and device based on fusion of laser point cloud and visible light image | |
| CN117934518B (en) | Remote sensing image segmentation method and system for ocean homeland space detailed planning | |
| CN117291790A (en) | SAR image registration method, SAR image registration device, SAR image registration equipment and SAR image registration medium | |
| WO2022156652A1 (en) | Vehicle motion state evaluation method and apparatus, device, and medium | |
| CN111582013A (en) | Ship retrieval method and device based on gray level co-occurrence matrix characteristics | |
| Bailey et al. | Determining large scale sandbar behaviour | |
| CN113033578B (en) | Image calibration method, system, terminal and medium based on multi-scale feature matching | |
| CN113592744B (en) | Geometric fine correction method suitable for high-resolution remote sensing image | |
| CN115760933A (en) | SAR image registration method, device, equipment and medium | |
| CN116128919A (en) | Multi-temporal image abnormal target detection method and system based on polar constraint | |
| CN112967360A (en) | Synthetic aperture radar image Voronoi polygon mosaic method considering multiple dimensions | |
| CN114998630A (en) | Ground-to-air image registration method from coarse to fine | |
| CN113255405B (en) | Parking space line identification method and system, parking space line identification equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |