CN111784606B - Remote sensing image classification post-processing method, storage medium and system - Google Patents
Remote sensing image classification post-processing method, storage medium and system Download PDFInfo
- Publication number
- CN111784606B CN111784606B CN202010613222.6A CN202010613222A CN111784606B CN 111784606 B CN111784606 B CN 111784606B CN 202010613222 A CN202010613222 A CN 202010613222A CN 111784606 B CN111784606 B CN 111784606B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- pixel
- binary
- image
- pixel point
- 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 30
- 238000012805 post-processing Methods 0.000 title claims abstract description 22
- 238000010586 diagram Methods 0.000 claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 15
- 230000000877 morphologic effect Effects 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 12
- 235000002566 Capsicum Nutrition 0.000 abstract description 12
- 239000006002 Pepper Substances 0.000 abstract description 12
- 235000016761 Piper aduncum Nutrition 0.000 abstract description 12
- 235000017804 Piper guineense Nutrition 0.000 abstract description 12
- 235000008184 Piper nigrum Nutrition 0.000 abstract description 12
- 150000003839 salts Chemical class 0.000 abstract description 12
- 244000203593 Piper nigrum Species 0.000 abstract 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 15
- 241000722363 Piper Species 0.000 description 11
- 238000001228 spectrum Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a remote sensing image classification post-processing method, which comprises the steps of obtaining an initial classification result diagram of a remote sensing image; converting the initial classification result diagram into a plurality of binary diagrams corresponding to the target ground object categories respectively, and endowing each pixel point on the binary diagrams with a category label; carrying out morphological algorithm processing on a plurality of binary images so as to eliminate noise in the binary images; superposing and combining all the binary images by taking the same coordinate system as a reference, and acquiring definition of class labels on each pixel point in the combined image; and determining the target ground object represented by each pixel point according to the number and definition of the class labels on each pixel point. The invention also provides a storage medium and a remote sensing image classification post-processing system, and the remote sensing image classification post-processing method, the storage medium and the remote sensing image classification post-processing system provided by the invention are used for performing morphological processing by dividing an initial classification result image into a plurality of binary images so as to effectively remove salt and pepper noise.
Description
Technical Field
The present invention relates to the field of remote sensing image processing, and in particular, to a remote sensing image classification post-processing method, a storage medium, and a system.
Background
Along with the continuous development of remote sensing technology, the diversity of remote sensing platforms and the continuous improvement of the spectral resolution, the spatial resolution and the time resolution of remote sensing images are widely applied to the fields of environmental monitoring, material analysis, urban planning, precise agriculture, military reconnaissance and the like.
The classification of the ground object is an important link of processing and applying the remote sensing image, and aims to assign a unique ground object identifier to each pixel in the image, a series of methods are currently available for classifying the ground object of the remote sensing image, and the classification is classified into non-supervision classification, semi-supervision classification and supervision classification according to whether a sample is introduced or not and how much sample is introduced, but most classification methods only consider the spectral characteristics of the image, and due to the complexity of the ground object classification and the existence of the phenomena of different spectrums and different spectrums of the image pixels, the classification result often appears a plurality of isolated points, and the classification result has the phenomena of ground object classification confusion and 'salt and pepper noise', so that classification accuracy is difficult to achieve an ideal result.
Disclosure of Invention
In view of the above, the invention provides a remote sensing image classification post-processing method, a storage medium and a system, which solve the problems of ground object category confusion and salt and pepper noise of classification results after remote sensing image classification.
In order to achieve the above object, the present invention provides a method for post-processing classification of a remote sensing image, which includes the steps of obtaining an initial classification result map of the remote sensing image; converting the initial classification result diagram into a plurality of binary diagrams corresponding to the target ground object categories respectively, and endowing each pixel point on the binary diagrams with a category label; carrying out morphological algorithm processing on a plurality of binary images so as to eliminate noise in the binary images; superposing and combining all the binary images by taking the same coordinate system as a reference, and acquiring definition of class labels on each pixel point in the combined image; and determining the target ground object represented by each pixel point according to the number and definition of the class labels on each pixel point.
The invention also provides a storage medium having a computer program stored therein, wherein the computer program is arranged to execute the remote sensing picture classification post-processing method at run-time.
The invention also provides a remote sensing image processing system which comprises a processor and a memory, wherein the memory is stored with a computer program, and when the computer program is executed by the processor, the remote sensing image classification post-processing method is realized.
Compared with the prior art, the remote sensing image classification post-processing method, the storage medium and the system provided by the invention have the following beneficial effects:
the method comprises the steps of converting an initial classification result diagram of a remote sensing image into a plurality of binary diagrams corresponding to a target ground object, eliminating noise in each binary diagram by morphological opening and closing operation, overlapping the plurality of binary diagrams into one diagram, and endowing each pixel point with a class label, so that the problem of salt and pepper noise after the classification of the traditional remote sensing image can be effectively solved.
The above-described embodiments of the present invention do not limit the scope of the present invention. Any other corresponding changes and modifications made in accordance with the technical idea of the present invention shall be included in the scope of the claims of the present invention.
Drawings
Fig. 1 is a schematic flow chart of steps of a remote sensing image classification post-processing method according to a first embodiment of the present invention;
FIG. 2 is a schematic diagram of the principle of superposition of multiple binary images after superposition in a remote sensing image classification post-processing method;
FIG. 3 is a schematic shape diagram of a rectangular region for pixel points in the overlapping region in FIG. 2;
FIG. 4 is a schematic flow chart of the substeps of step S1 in FIG. 1;
FIG. 5 is a flow chart illustrating the substep of step S2 in FIG. 1;
FIG. 6 is a flow chart illustrating the substep of step S3 in FIG. 1;
FIG. 7 is a schematic flow chart of the substep of step S4 in FIG. 1;
fig. 8 is a schematic flow chart of the substep of step S5 in fig. 1.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, the method for post-processing classification of remote sensing images provided by the invention comprises the following steps:
s1, acquiring an initial classification result diagram of a remote sensing image;
specifically, the remote sensing image refers to the difference of different ground objects represented by the difference of brightness values or pixel values and the spatial variation, and is used as a physical basis for distinguishing the ground objects with different images. The remote sensing image classification is to analyze the spectrum information and the space information of various ground objects in the remote sensing image by using a computer, select the characteristics, divide the characteristic space into mutually non-overlapping subspaces, and then classify each pixel in the image into each subspace, thereby obtaining a classification result.
Under the condition that only the spectral characteristics of the images are considered during classification, due to the complexity of the ground object types, isolated points which are uniformly distributed are usually present on the classified images, namely salt and pepper noise phenomenon is generated, and the images with the salt and pepper noise phenomenon are the initial classification result diagrams of the remote sensing images.
S2, converting the initial classification result diagram into a plurality of binary diagrams corresponding to the target ground object categories respectively, and endowing each pixel point on the plurality of binary diagrams with a category label;
specifically, after the initial classification result diagram is obtained, multiple target ground object categories are included in the image, each ground object category is respectively converted into a binary diagram through binarization processing, multiple binary diagrams respectively corresponding to different ground object categories are obtained, and therefore the outline of the target ground object category in the image is highlighted.
It can be understood that the target ground object category can be selected according to the requirement, if the target ground object category is selected as five kinds of water, vegetation, building, road and bare land, 5 binary maps corresponding to the water, vegetation, building, road and bare land are obtained after the binarization treatment.
And after obtaining the binary image corresponding to the target ground object category, assigning category labels to all pixel points in each binary image.
It will be appreciated that the role of the class label is to attach an identifier to each pixel so that each pixel has a definition corresponding thereto. For example, in the binary image corresponding to the water body, all the pixels representing the water body are given category labels of the water body, in the binary image corresponding to the vegetation, all the pixels representing the vegetation are given category labels of the vegetation, so as to push in until all the pixels representing the target ground object on the binary image are given corresponding category labels. And the pixels in the blank area are assigned with a label defined as a blank category, so that each pixel has a definition corresponding to the label. That is, a pixel with a water class label represents that the pixel represents a water on an image, a pixel with a vegetation class label represents that the pixel represents vegetation on the image, and a pixel with an empty class label represents that the pixel represents a blank area on the image.
It can be understood that the gray values of all the pixels in each binary image are only 0 or 255, wherein the pixel of one gray value represents the pixel of the target ground object, and the other pixel represents the blank area, and when the class label is assigned, the pixel of the blank area is assigned with a class label with blank definition.
It will be appreciated that the expression form of the class label definition may be selected autonomously, such as by representing the class label representing the body of water as 1, the class label representing the vegetation as 2, the class label representing the building as 3, the class label of the empty definition as 0, etc. Instead of taking the form of numerals, letters or other characters may be used instead.
S3, carrying out morphological algorithm processing on the plurality of binary images so as to eliminate noise in the binary images;
specifically, because salt and pepper noise exists in the initial classification result diagram, after the salt and pepper noise is converted into a plurality of binary diagrams, the salt and pepper noise still exists in the binary diagrams. And by using morphological opening and closing operation, salt and pepper noise in the binary image can be eliminated, so that the contour of the binary image corresponding to each target ground object classification is optimized. That is, the noise is removed by morphological operation processing, and discrete pixel points are eliminated.
S4, overlapping and combining all the binary images by taking the same coordinate system as a reference, and acquiring definition of a class label corresponding to each pixel point in the combined image;
specifically, after the two-value images are processed by the morphological algorithm, the salt and pepper noise in the two-value images is eliminated, and at the moment, the two-value images are overlapped and combined by taking the same coordinate system as a reference to form an image comprising all the target ground objects. Since the class labels are assigned to the pixels representing the target ground object in all the binary images in step S2, there may be pixels with 2 class labels superimposed as one pixel in the superimposing process, which results in that the same pixel may include a plurality of defined class labels in the superimposed image. That is, after the overlapping and merging, two or more binary images are superimposed on each other at the pixel points representing the target ground object.
When the binary image corresponding to the water body and the binary image corresponding to the vegetation are overlapped and combined, the pixel points in the overlapped area are provided with two defined class labels, namely the class label of the water body and the class label of the vegetation, as shown in fig. 2.
S5, determining a target ground object represented by each pixel point according to the definition of the class label of each pixel point;
specifically, after the definition of the class label of each pixel is obtained, when the pixel has only 1 defined class label, the pixel is directly judged to represent the target ground object defined by the class label, if the pixel has only the class label defined as the water body, the pixel is then represented as the water body.
When the class label on the pixel is empty, the pixel is determined to be a blank area.
When the pixel point has more than 2 kinds of defined class labels, taking a rectangular area with KxK as a center, counting the definition of the class labels of all the pixel points in the area, and judging the definition with the largest number in the definition of all the class labels in the area as the target ground object defined by the pixel point. If the maximum number of class labels of all the pixel points in the rectangular area with the pixel point as the center is counted, the water body represented by the pixel point is judged, as shown in fig. 3.
Through the above-mentioned decision criteria, any pixel point in the overlapped and combined images can be defined, so that the overlapped and combined images can clearly reflect the ground object distribution condition.
It will be appreciated that the value of K may be 33mm or 55mm, i.e. KxK may be 33mm or 55mm.
Referring to fig. 4, step S1 further includes the sub-steps of:
s11, acquiring a remote sensing image by using a remote sensing technology;
specifically, the remote sensing technology is a technology for collecting electromagnetic radiation information of ground object targets from artificial satellites, airplanes or other aircrafts and judging the earth environment and resources. The method is a comprehensive sensing technology which is gradually formed along with the development of aerospace technology and electronic computer technology on the basis of aerospace photography and interpretation, any object has different electromagnetic wave reflection or radiation characteristics, and remote sensing images of formulated areas can be shot by utilizing the remote sensing technology.
S12, classifying the remote sensing images to obtain an initial classification result diagram;
specifically, the classification of the remote sensing image is to analyze the spectrum information and the space information of various ground objects in the remote sensing image by using a computer, select characteristics, divide various pixels in the image into different categories according to a certain rule or algorithm, and then obtain the corresponding information of the remote sensing image and the actual ground objects, thereby realizing the classification of the remote sensing image. At present, a series of methods are used for classifying the ground features of the remote sensing image, and the classification can be classified into non-supervision classification, semi-supervision classification and supervision classification according to whether a sample is introduced or not and how much sample is introduced. By classifying the remote sensing images, an initial classification result diagram of the remote sensing images can be obtained.
Referring to fig. 5, step S2 further includes the sub-steps of:
s21, performing binarization processing on the initial classification result diagram;
specifically, after the initial classification result diagram is obtained, binarization processing is performed on the initial classification result diagram, that is, the gray value of the pixel point on the image is set to 0 or 255, so that the whole image shows a remarkable black-and-white effect.
S22, dividing the binarized image into a plurality of binary images according to the target ground object;
specifically, the initial classification result diagram includes a plurality of target ground objects, and after binarization processing, the initial classification result diagram is divided into a plurality of binary diagrams according to different target ground objects. Namely, each target ground object corresponds to a binary image.
S23, assigning the pixel points on each binary image with category labels;
specifically, with the pixel points as units, class labels are assigned to the pixel points on each binary image, and each class label is used for defining the pixel points subsequently. If the pixel has only one type of label, the type of label is the definition of the corresponding pixel. If there is only one type label on a certain pixel point, and the type label is a water body, the definition of the pixel point corresponding to the type label is the water body. Or when the class label on the pixel point is empty, defining the pixel point corresponding to the class label as a blank area.
Referring to fig. 6, step S3 further includes the sub-steps of:
s31, eliminating discrete pixel points in the binary image by using morphological opening and closing operation;
specifically, morphological open and close operations are two basic operations in morphology, which are mainly based on erosion and expansion of contours in an image, thereby eliminating noise except the contours in the image. Wherein, the open operation is: firstly corroding the image contour, and then performing expansion operation on the image contour, so that isolated points or burrs connected with the image contour are eliminated; the closing operation is as follows: the image contour is expanded first, and then the image contour is eroded, so that some noise points inside the image contour are eliminated.
It can be understood that in this embodiment, the image contour is a contour representing the category of the ground object in the image.
S32, filling pixel islands in pixels representing the target ground object;
specifically, after the noise points inside the image contour are eliminated by using the closed operation, pixel islands appear inside the image contour, that is, vacant parts appear at the places where the internal noise points of the image contour are eliminated, and the pixel islands need to be filled so as to ensure the integrity of the image contour.
It can be understood that when filling the pixel island, the pixel points of the filling part are given a category label corresponding to the target ground object of the image outline.
Referring to fig. 7, step S4 further includes a sub-step;
s41, establishing a plane coordinate system;
specifically, before integrating a plurality of binary images into one image, a unified reference plane coordinate system needs to be established, so that all binary images are referenced by the plane coordinate system.
S42, overlapping the plurality of binary images in a plane coordinate system;
specifically, after the planar coordinate system is established, the plurality of binary images are sequentially placed in the coordinate system with the planar coordinate system as a reference, so that the plurality of binary images are overlapped into one image. That is, two-value maps corresponding to a plurality of target ground objects exist simultaneously in one image.
S43, acquiring definition of category labels on each pixel point in the overlapped and combined image;
specifically, after overlapping a plurality of binary images, each pixel is taken as a unit, and a category label on each pixel in the combined image is acquired so as to facilitate definition of each pixel in the image.
Referring to fig. 8, step S5 further includes the sub-steps of:
s51, screening out pixel points with a plurality of category labels according to the acquired category labels on the pixel points;
specifically, after the multiple binary images are overlapped, there may be a case that the image outlines of multiple target ground objects are overlapped, so that all the pixel points in the overlapped images can be divided into three cases. The pixel points of the image contour overlapping areas of the plurality of target ground objects are provided with a plurality of category labels; the pixel points of the areas where the image outlines are not overlapped are provided with single category labels; while areas without image contours, i.e. blank areas, do not have category labels. And the pixel points with the plurality of class labels are independently screened out so as to be convenient for defining the pixel points with the plurality of class labels later.
S52, taking pixel points with a plurality of category labels as the center, establishing a rectangular area, and acquiring the category labels of all the pixel points in the rectangular area;
specifically, a rectangular region of KxK is established by taking a pixel point with a plurality of class labels as a center, and class labels on all the pixel points in the rectangular region are acquired by taking the pixel point as a unit.
S53, counting the number of all kinds of labels in the rectangular area, and defining a pixel point in the center of the rectangular area by one kind of label with the largest number;
specifically, all class labels in the rectangular area are ordered according to the number, and the pixel point in the center of the rectangular area is defined by one class label with the largest number. By defining each pixel point, the user can clearly know which region in the image represents the target ground object.
The invention also provides a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the above-mentioned method steps when run. The storage medium may include, for example, a floppy disk, an optical disk, a DVD, a hard disk, a flash Memory, a U-disk, a CF card, an SD card, an MMC card, an SM card, a Memory Stick (Memory Stick), an XD card, and the like.
The computer software product is stored in a storage medium and includes instructions for causing one or more computer devices (which may be personal computer devices, servers or other network devices, etc.) to perform all or part of the steps of the method of the invention.
The invention also provides a character recognition system, which comprises a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the character recognition method is realized.
Compared with the prior art, the remote sensing image classification post-processing method, the storage medium and the system provided by the invention have the following beneficial effects:
the method comprises the steps of converting an initial classification result diagram of a remote sensing image into a plurality of binary diagrams corresponding to a target ground object, eliminating noise in each binary diagram by morphological opening and closing operation, overlapping the plurality of binary diagrams into one diagram, and endowing each pixel point with a class label, so that the problem of salt and pepper noise after the classification of the traditional remote sensing image can be effectively solved.
The above-described embodiments of the present invention do not limit the scope of the present invention. Any other corresponding changes and modifications made in accordance with the technical idea of the present invention shall be included in the scope of the claims of the present invention.
Claims (6)
1. The remote sensing image classification post-processing method is characterized by comprising the following steps:
acquiring an initial classification result diagram of a remote sensing image;
converting the initial classification result diagram into a plurality of binary diagrams corresponding to the target ground object categories respectively, and endowing each pixel point on the binary diagrams with a category label;
carrying out morphological algorithm processing on a plurality of binary images so as to eliminate noise in the binary images;
superposing and combining all the binary images by taking the same coordinate system as a reference, and acquiring definition of class labels on each pixel point in the combined image; and
Determining the target ground object represented by each pixel according to the number and definition of class labels on each pixel;
the converting the initial classification result graph into a plurality of binary graphs corresponding to the target ground object categories respectively, and assigning a category label to each pixel point on the plurality of binary graphs further comprises:
performing binarization processing on the initial classification result graph;
dividing the binarized image into a plurality of binary images according to the target ground object; and
Assigning a class label to the pixel points on each binary image;
the step of superposing and combining all the binary images by taking the same coordinate system as a reference, and the step of obtaining the definition of the category label on each pixel point in the combined image further comprises the following steps:
establishing a plane coordinate system;
overlapping a plurality of binary images in the plane coordinate system; and
Acquiring definition of category labels on each pixel point in the overlapped and combined images;
the determining the target ground object represented by each pixel point according to the number and definition of the class labels on each pixel point further comprises:
according to the obtained class labels on the pixel points, the pixel points with a plurality of class labels are screened out;
setting up a rectangular area by taking pixel points with a plurality of category labels as centers, and acquiring the category labels of all the pixel points in the rectangular area; and
Counting the number of all kinds of labels in a rectangular area, and defining a pixel point in the center of the rectangular area by using one kind of label with the largest number;
the pixel points with the plurality of class labels are the pixel points of the overlapping part of the image outlines of the target ground object when the plurality of binary images are overlapped.
2. The method for post-processing remote sensing image classification as defined in claim 1, wherein:
the class label is used for attaching an identifier to each pixel point so that each pixel point has a definition corresponding to the pixel point.
3. The method of claim 1, wherein said performing morphological algorithm processing on said plurality of said binary images to eliminate noise in said binary images further comprises:
eliminating discrete pixel points in the binary image by morphological opening and closing operation; and
And filling the pixel island in the pixel representing the target ground object.
4. The method of claim 1, wherein obtaining an initial classification result map of the remote sensing image further comprises:
acquiring a remote sensing image by utilizing a remote sensing technology; and
And classifying the remote sensing images to obtain an initial classification result graph.
5. A storage medium, characterized by:
the storage medium having stored therein a computer program, wherein the computer program is arranged to perform the remote sensing image classification post-processing method of any of the claims 1-4 when run.
6. The remote sensing image classification post-processing system is characterized in that:
the remote sensing image classification post-processing system comprises a processor and a memory, wherein a computer program is stored on the memory, and the computer program is executed by the processor to realize the remote sensing image classification post-processing method according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010613222.6A CN111784606B (en) | 2020-06-30 | 2020-06-30 | Remote sensing image classification post-processing method, storage medium and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010613222.6A CN111784606B (en) | 2020-06-30 | 2020-06-30 | Remote sensing image classification post-processing method, storage medium and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111784606A CN111784606A (en) | 2020-10-16 |
| CN111784606B true CN111784606B (en) | 2023-11-24 |
Family
ID=72761260
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010613222.6A Active CN111784606B (en) | 2020-06-30 | 2020-06-30 | Remote sensing image classification post-processing method, storage medium and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111784606B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114563378B (en) * | 2022-03-03 | 2022-07-29 | 生态环境部卫星环境应用中心 | Method, device, medium and equipment for quantitatively describing space distribution of cyanobacterial bloom in lakes and reservoirs |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400151A (en) * | 2013-08-16 | 2013-11-20 | 武汉大学 | Optical remote-sensing image, GIS automatic registration and water body extraction integrated method |
| CN103824304A (en) * | 2014-03-18 | 2014-05-28 | 中国矿业大学(北京) | Method for performing fault diagnosis on ores on conveying belt during conveying process |
| CN104966295A (en) * | 2015-06-16 | 2015-10-07 | 武汉大学 | Ship extraction method based on wire frame model |
| CN109063754A (en) * | 2018-07-18 | 2018-12-21 | 武汉大学 | A kind of remote sensing image multiple features combining classification method based on OpenStreetMap |
| CN111192239A (en) * | 2019-12-18 | 2020-05-22 | 星际空间（天津）科技发展有限公司 | Method and device for detecting change area of remote sensing image, storage medium and electronic equipment |
| CN111339947A (en) * | 2020-02-26 | 2020-06-26 | 苏州中科天启遥感科技有限公司 | Method and system for extracting remote sensing image fuzzy boundary ground object, storage medium and equipment |
-
2020
- 2020-06-30 CN CN202010613222.6A patent/CN111784606B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103400151A (en) * | 2013-08-16 | 2013-11-20 | 武汉大学 | Optical remote-sensing image, GIS automatic registration and water body extraction integrated method |
| CN103824304A (en) * | 2014-03-18 | 2014-05-28 | 中国矿业大学(北京) | Method for performing fault diagnosis on ores on conveying belt during conveying process |
| CN104966295A (en) * | 2015-06-16 | 2015-10-07 | 武汉大学 | Ship extraction method based on wire frame model |
| CN109063754A (en) * | 2018-07-18 | 2018-12-21 | 武汉大学 | A kind of remote sensing image multiple features combining classification method based on OpenStreetMap |
| CN111192239A (en) * | 2019-12-18 | 2020-05-22 | 星际空间（天津）科技发展有限公司 | Method and device for detecting change area of remote sensing image, storage medium and electronic equipment |
| CN111339947A (en) * | 2020-02-26 | 2020-06-26 | 苏州中科天启遥感科技有限公司 | Method and system for extracting remote sensing image fuzzy boundary ground object, storage medium and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111784606A (en) | 2020-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xie et al. | Multilevel cloud detection in remote sensing images based on deep learning | |
| CN109740639B (en) | Wind cloud satellite remote sensing image cloud detection method and system and electronic equipment | |
| CN110728197B (en) | Single-tree-level tree species identification method based on deep learning | |
| Liu et al. | Multimorphological superpixel model for hyperspectral image classification | |
| Li et al. | Ensemble EMD-based spectral-spatial feature extraction for hyperspectral image classification | |
| de Gélis et al. | Siamese KPConv: 3D multiple change detection from raw point clouds using deep learning | |
| CN114820463A (en) | Point cloud detection and segmentation method and device, and electronic equipment | |
| Chen et al. | Object-based multi-modal convolution neural networks for building extraction using panchromatic and multispectral imagery | |
| CN113033385A (en) | Deep learning-based violation building remote sensing identification method and system | |
| CN113537180A (en) | Tree obstacle identification method and device, computer equipment and storage medium | |
| CN111784606B (en) | Remote sensing image classification post-processing method, storage medium and system | |
| CN116664954A (en) | Hyperspectral ground object classification method based on graph convolution and convolution fusion | |
| CN110310263B (en) | SAR image residential area detection method based on significance analysis and background prior | |
| CN110321867B (en) | Shielded target detection method based on component constraint network | |
| CN115100469A (en) | Target attribute identification method, training method and device based on segmentation algorithm | |
| Abbasi et al. | Fuzzy histogram equalization of hazy images: a concept using a type-2-guided type-1 fuzzy membership function | |
| CN111368865A (en) | Method and device for detecting remote sensing image oil storage tank, readable storage medium and equipment | |
| CN117710728A (en) | SAR image target recognition method, SAR image target recognition device, SAR image target recognition computer equipment and storage medium | |
| Prakash et al. | Implementation of Image Segmentation with Prewitt Edge Detection using VLSI Technique | |
| Wu et al. | Research on asphalt pavement disease detection based on improved YOLOv5s | |
| CN109977965B (en) | Method and device for determining detection target in remote sensing airport image | |
| CN115984712A (en) | Multi-scale feature-based remote sensing image small target detection method and system | |
| Sheikh et al. | Noise tolerant classification of aerial images into manmade structures and natural-scene images based on statistical dispersion measures | |
| CN112116561B (en) | Power grid transmission line detection method and device based on image processing fusion network weight | |
| CN111079797B (en) | Image classification method and device 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 |