CN107609526A - Rule-based fine dimension city impervious surface rapid extracting method - Google Patents
Rule-based fine dimension city impervious surface rapid extracting method Download PDFInfo
- Publication number
- CN107609526A CN107609526A CN201710856065.XA CN201710856065A CN107609526A CN 107609526 A CN107609526 A CN 107609526A CN 201710856065 A CN201710856065 A CN 201710856065A CN 107609526 A CN107609526 A CN 107609526A
- Authority
- CN
- China
- Prior art keywords
- wave band
- impervious surface
- road
- rule
- band
- 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
Abstract
Rule-based fine dimension city impervious surface rapid extracting method category Information Trend of Urban Planning applied technical field, the present invention comprise the following steps:Original high resolution multispectral image is pre-processed, obtains fusion evaluation;Easy first and difficult later, the obvious principle of characters of ground object otherness between node is followed, establishes the decision tree of extraction;Road and non-rice habitats part are distinguished on image pixel layer;Non-rice habitats part is split;Feature description is carried out to different atural objects, according to categorised decision tree, classifying rules collection is built, obtains urban area fine dimension impervious surface distribution results;The precision test of classification results is carried out with Kappa index methods by being evaluated after classification.Good classification effect of the present invention, " spiced salt phenomenon " in being extracted efficiently against remote sensing image information, and " same object different images " and " different spectrum jljl " phenomenon can be reduced well, improve efficiency and accuracy that city impervious surface obtains.
Description
Technical field
The invention belongs to Information Trend of Urban Planning applied technical field, and in particular to rule-based fine dimension city is impermeable
Water surface rapid extracting method.
Background technology
City impervious surface type and distribution be urban green system planning, drainage waterlogging prevention planning concern important content,
City planning design requires also higher to the fine degree of impervious surface spatial distribution present situation.Existing impervious surface evaluation method
Estimated that construction land plot is typically using urban road as border frequently with construction land plot property.Circular
It is as follows:In the range of statistical estimation, land character is " park green land ", the area in " green area for environmental pro tection " plot.Existing data acquisition
Mode has the following disadvantages, and space scale is more rough first, and existing city impervious surface analysis is based on construction land, land used
Plot have ignored the permeable EDS maps inside plot, such as the ground for residential estate property typically using urban road as border
Block, the greening situation inside residential quarter are not taken into account, furthermore the update cycle is grown, are built used in urban planning and construction department
If land used data renewal time is indefinite, and gather material, verification link expend substantial amounts of manpower.
Remotely-sensed data spatial resolution, temporal resolution, spectral resolution and radiometric resolution improve constantly, data class
Type is also continuously increased, and is sharply increased from the remotely-sensed data amount acquired in the remote sensing platforms such as space flight, aviation, near space, remote sensing number
According to having obvious big data feature.Rule-based remote sensing information extractive technique can not only utilize the spectrum of image to believe
Breath, its be more using the spatial information of imaged object, texture information, domain object related information etc..Due to towards right
Image information extracting method takes into full account the correlation between pixel and pixel from image entirety, can be efficiently against distant
Feel " spiced salt phenomenon " in image information extraction, and " same object different images " and " different spectrum jljl " phenomenon can be reduced well, because
This has become a hot research of remote sensing image process field.Establishing the general interpretation flow of automation not only can be very big
Personnel and time input cost are saved in degree, or remote sensing is combined with Other subjects carries out follow-up study offer preferably
Data basis.Therefore, interpretation automation is always to have popular problem to be solved.At present still without for extracting fine dimension city
The rule set of city's impervious surface improves city, it is necessary to establish the city fine dimension underlying surface extracting rule collection of a set of universality
City's planning field obtains the ageing of this item data, accuracy, economy.
The content of the invention
The purpose of the present invention is to propose to rule-based fine dimension city impervious surface rapid extracting method, according to maximum
Separating degree principle, according to geometric properties, textural characteristics, user-defined feature, class correlated characteristic, spectral signature structure classifying rules
Collection, systematization technique flow is formed, improve efficiency and accuracy that city impervious surface obtains.
1. a kind of rule-based fine dimension city impervious surface rapid extracting method, passes through the remote sensing image to input
Data are handled, and are drawn the distribution results of city impervious surface, are comprised the following steps:
1.1 pairs of original high resolution multispectral images carry out atmospheric correction, fusion, geometric correction pretreatment, are had
Abundant spectral information, high-resolution fusion evaluation;
1.2 follow easy first and difficult later, the obvious principle of characters of ground object otherness between node, establish the decision tree of extraction;
1.3 utilize chessboard partitioning algorithm on image pixel layer, view picture image is split by the road vectors data of input;
Feature description is carried out for road and non-rice habitats, i.e. road is more than 0 to be overlapping with vector;Non-rice habitats be region in be not road
Part, distinguish road and non-rice habitats part;
Split using heterogeneous minimum region merging algorithm 1.4 pairs of non-rice habitats parts;
1.5 spectral signature, geometric properties, the topological characteristic according to type of ground objects, feature description is carried out to different atural objects, if
Determine the feature or combinations of features of type of ground objects, establish complete categorised decision tree, high-resolution can be completely applied to by constructing
The classifying rules collection of rate or middle low resolution remote sensing image;Imaged object identification is carried out on feature space according to classifying rules collection
And mark, the extraction of remote sensing information is completed, obtains urban area impervious surface distribution results;
1.6 carry out the precision test of classification results by being evaluated after classification with Kappa index methods.
Original high resolution multispectral image described in step 1.1 is high-resolution WorldView-3 images, has 8
Multi light spectrum hands image, i.e. blue wave band, green band, red band, near infrared band, seashore wave band, yellow band, red
The image of edge wave band and the wave band of near-infrared 2, multispectral spatial resolution are 1.6 meters, and panchromatic wave-band spatial resolution is 0.4
Rice.
Heterogeneous minimum region merging algorithm segmentation described in step 1.4, it is 80~100 that it, which recommends scale parameter scope,
Form factor weight is arranged to 0.1, i.e. spectrum Factor Weight is 0.9;Degree of the compacting factor and smoothness factor are 0.5, each wave band
The weight for participating in segmentation is 1.
Atural object described in step 1.2 is building, road, vegetation, exposed soil, water body, artificial hard ground and arable land.
Decision tree described in step 1.2, it is divided into five levels, eight nodes.
Classifying rules collection described in step 1.5, it is according to being characterized as spectral signature, geometric properties, textural characteristics, self-defined
Feature and class correlated characteristic structure;
Described classifying rules collection, its feature selecting foundation is maximum separation degree index;
Described classifying rules collection, its road, non-rice habitats characteristic of division are length-width ratio, numerical value more than 3 for road;Plant
Quilt, non-vegetation classification are characterized as normalized differential vegetation index NDVI, blue wave band, and NDVI is more than 0.3 and blue wave band reflectivity is small
It is vegetation in 300;Low-light level body, high brightness body, which are divided into, is characterized as the wave band of near-infrared 1, the wave band of near-infrared 2, maximum difference index,
The wave band of near-infrared 1 and the wave band reflectivity of near-infrared 2 are less than 300, and maximum difference index is low-light level body more than 0.4;It is permeable, no
Permeable characteristic of division is blue wave band, normalization water body index NDWI, and blue wave band reflectivity is less than 0 less than 300 and NDWI
Permeable atural object;Exposed soil, arable land characteristic of division are area, rectangle adaptability, and area is more than 100,000 pixels and rectangle adaptability is more than
0.5 is arable land;Hard ground, classification of buildings are characterized as degree of compacting, soil index WVSI, and degree of compacting is house less than 1.5 or so;
It is hard ground that NVSI, which is less than 0 more than -0.05 or so,;Water body, non-water body characteristic of division are and waterproof atural object distance, gold-tinted ripple
Section, modified water body index MNDWI, apart from waterproof atural object be less than 200 pixels and yellow band reflectivity be less than 300 and
MDNWI is non-water body less than 0.3.
The present invention proposes rule-based fine dimension city impervious surface rapid extracting method, former according to maximum separation degree
Then, according to geometric properties, textural characteristics, user-defined feature, class correlated characteristic, spectral signature structure classifying rules collection, system is formed
Systemization techniqueflow, efficiency and accuracy that city impervious surface obtains are improved, and extraction result is extracted with topographic map data
Sample is analyzed, and Main Conclusions is as follows:
Classification results and sampled point classification have a higher uniformity, and classifying quality is preferable.Overall accuracy, user or production
The accuracy of precision depends on sample data, causes the small variation of pixel classification and its percentage may be caused to change.Kappa
Coefficient overcomes the gatherer process that other method excessively relies on sample data and sample, more objective evaluation classification quality.Utilize 1 ratio
500 topographic maps are uniformly middle in the range of research area to choose 3136 sampled points, and the Kappa coefficients asked for according to confusion matrix are
0.82, overall precision 0.86, from the point of view of every a kind of individually Kappa numbers, preferably (road extraction has used existing water body effect
Data, not in contrast range), Kappa coefficients are more than 0.9;Exposed soil, vegetation are also all more than 0.8.
Rule-based extractive technique can not only utilize the spectral information of image, and it is more the sky using imaged object
Between information, texture information, the related information etc. of domain object.Because extracting method is from image entirety, pixel is taken into full account
Correlation between pixel, " spiced salt phenomenon " in being extracted efficiently against remote sensing image information, and can be fine
Reduce " same object different images " and " different spectrum jljl " phenomenon in ground.
Brief description of the drawings
Fig. 1 is general technical route map
Fig. 2 is the first hierarchical classification result figure
Fig. 3 is the second hierarchical classification result figure
Fig. 4 is third layer level classification result figure
Fig. 5 is the 4th hierarchical classification result figure
Fig. 6 is impervious surface classification final result figure
Embodiment
Embodiments of the invention are described in further detail below in conjunction with accompanying drawing.
As shown in figure 1, the Technology Roadmap for the present invention.
1. tentatively being pre-processed to image first, pretreatment described here includes atmospheric correction, image co-registration, geometry
Correction.
1.1 atmospheric correction:The various radiation energies that remote sensing is utilized are intended to that interaction-occurs with earth atmosphere or dissipated
Penetrate or absorb, and make energy attenuation, and spatial distribution is changed.Therefore in order to eliminate influence of the air to remote sensing image,
Need to carry out atmospheric correction, the present invention carries out atmospheric correction using FLAASH models.
Image co-registration:At present for panchromatic and multi light spectrum hands image co-registration method, mainly include principal component change,
Brovey changes, HIS conversion, Gram-Schmidt changes and Wavelet transformation etc., the present invention uses Gram-Schmidt
Spectral Sharpening methods, the fusion method can preferably keep the colourity and saturation degree of former image, while preferably
Remain spectral information and spatial information.
Geometric correction:Geometric correction causes to eliminate sensor imaging mode, attitude of satellite change and flying height etc.
Geometric distortion.Geometric correction uses 1:500 topographic maps are significantly put as control point as benchmark, artificial selection, utilize two
Order polynomial carries out geometric correction to image, and the control point of selection can be distributed in image as far as possible, and quantity is no less than 30, always
Control errors are in a pixel.
1.2 selections include seven kinds of typical cases such as building, road, vegetation, exposed soil, water body, artificial hard ground and arable land
Ground class is as extraction target.The principle that taxonomic hierarchies is established follows from easy to difficult, therefore considers to establish seven kinds of atural objects point first
The sequencing of class;
1.3 according to the taxonomic hierarchies flow that is pre-designed, the first step to be gone out according to road vectors data separation road with it is non-
Road two parts.In order that vector segmentation of the view picture image by input, will be calculated on raw video pixel layer using chessboard segmentation
Method, the segmentation scale-value of setting are more than the maximal side pixel value of image, and simultaneous selection road vectors participate in segmentation.It is directed to respectively
Road carries out feature description with non-rice habitats, i.e. road is more than 0 to be overlapping with vector;The portion that it is not road in region that non-rice habitats, which are,
Point.Afterwards, the feature description of both road and non-rice habitats is applied in the New Level figure layers just split using chessboard,
The classification results of second layer road and non-rice habitats are as shown in Figure 2.
1.4 distinguish vegetation and non-vegetation in non-rice habitats classification.First have to split non-rice habitats part.Towards right
As the classification interpretation work of method first has to establish the multiple dimensioned Internet of different levels according to the terrestrial object information classification to be extracted.It is more
Multi-scale segmentation using heterogeneous minimum region merging algorithm, in image the merging of pixel start from any one picture in image
Member.Single pixel is first merged into less imaged object by it, is afterwards again merged into less imaged object larger polygon
Shape object, and in order to more accurately extract vegetation tiny in city, using less 80 segmentation chis on non-rice habitats object layer
Degree, simultaneously because identification vegetation is arranged to 0.1, i.e. spectrum Factor Weight mainly by spectral information, therefore by form factor weight
For 0.9;Degree of the compacting factor and smoothness Factor minute are not set to 0.5.Each wave band participate in splitting weight is set to 1.
1.5 spectral signature, geometric properties, the topological characteristic according to type of ground objects, feature description is carried out to different atural objects, if
Determine the feature or combinations of features of type of ground objects, establish complete categorised decision tree, high-resolution can be completely applied to by constructing
The classifying rules collection of rate or middle low resolution remote sensing image;Imaged object identification and mark are carried out on feature space according to rule set
Know, complete the extraction of remote sensing information, obtain urban area impervious surface distribution results.
After choosing a part of vegetation and non-vegetation sample, its corresponding NDVI and each wave band reflectance value are exported to be formed respectively
From two species very histogram of the sheet on different-waveband and NDVI features, separating degree J-M (Jeffries- are calculated
Matusita) distance.Contrast understands that both have preferable separation property in NDVI features, and BShi distances are 2.27, J 1.79.
Therefore NDVI is chosen as the optimal characteristics for distinguishing vegetation and non-vegetation, and its corresponding characteristic threshold value T is 0.32.
It is that normalized differential vegetation index NDVI is more than 0.32 to set vegetation characteristics, simultaneously because the NDVI values on blue roof are also very
Greatly, to exclude falsely dropping for blue roof, limitation blue wave band spectral value is less than 300.Rather than vegetation then directly sets condition as not
It is the part of vegetation.Both meet the part for inheriting upper level non-rice habitats simultaneously simultaneously.In non-rice habitats layer according to vegetation with it is non-
The feature description of vegetation is classified, as a result as shown in Figure 3.
After choosing a part of low-light level body and high brightness body sample, by its corresponding maximum difference and each wave band reflectance value
Export forms the respective two species other style originally histogram on different-waveband and maximum difference feature, and separating degree J-M is calculated
(Jeffries-Matusita) distance.It is preferable that contrast understands that both have in near-infrared 1, near-infrared 2 and maximum difference feature
Separation property, BShi distance respectively 3.21,1.72 and 2.37, J 1.92,1.64 and 1.81.Therefore these three features are chosen to refer to
Be denoted as to distinguish the optimal characteristics of low-light level body and high brightness body, its corresponding characteristic threshold value T be 318.91,257.46 and
0.41。
Classification is described according to the characteristic threshold value of three optimal characteristics of high brightness body and low-light level body in non-vegetable layer,
As a result it is as shown in Figure 4.At this moment it will be clear that water body and architectural shadow are separated well, the various regions thing for after
Further extraction done good preparation.
Because the atural object that does not extract of residue is more suitable for using somewhat larger segmentation yardstick, and each relatively independently,
Therefore we split high brightness body more in small, broken bits before and low-light level body merge respectively after entered again with new optimized parameter
Row segmentation.Wherein, it is divided into permeable earth's surface and waterproof earth's surface again in high brightness body.And the predominantly follow-up knowledge of the purpose specifically split
Other house, therefore split yardstick and tune up to 200, and shape index weight is changed to 0.5, degree of compacting weight is changed to 0.8, while root
Green wave band and yellow wave band weighted value are suitably increased according to result of study before.It is after choosing permeable face and impervious surface sample, its is right
The self-defined spectral signature and each wave band reflectance value answered export to form the respective two species other style originally Nogata in different characteristic
Figure, is calculated separating degree J-M (Jeffries-Matusita) distance.Contrast understands both in NDWI and blue wave band feature
Upper to have preferable separation property, BShi distances respectively 3.97 and 3.20, J are 1.96 and 1.92.Therefore the two features are chosen to refer to
It is denoted as to distinguish the optimal characteristics in permeable face and impervious surface, its corresponding characteristic threshold value T is 0.05,309.79.
Classified in high brightness layer according to the description of the feature on pervious surface and waterproof ground.It is afterwards elimination one
Misclassification misclassification, the impervious surface enclosed by permeable bread is classified as pervious surface using comprising algorithm, corrects the fruit such as Fig. 5 that finishes
It is shown.It can be seen that artificial pervious surface and non-construction land are distinguished, the good results are evident.
Mainly include roof and hard ground in impervious surface, both optimal spies are found using SEaTH algorithms before
Found during sign without the feature that substantially can distinguish both very well.
Now we can use the post-classification comparison algorithm of the small patch of removal that area is less than to the tiny segmentation of 300 pixels
Object categorization is in the classification around it, to remove classification results relatively fine in image.Wrapped in pervious surface classification
What is contained is exposed soil and both classifications of arable land, and both atural objects can be identified with larger cutting object, therefore will not
Pervious surface is split after merging with 500 segmentation yardsticks to permeable face, and form factor is arranged to 0.8, and degree of the compacting factor is set
For 0.5.
After choosing arable land and exposed soil sample, a series of geometric properties and textural characteristics corresponding to it are exported to form respective two
Species very histogram of the sheet in different characteristic, is calculated separating degree J-M (Jeffries-Matusita) distance.Contrast
Understand that both have preferable separation property on area and rectangle adaptability feature, BShi distances respectively 3.10 and 2.93, J are
1.91 with 1.82.Therefore, the two characteristic indexs are chosen as the optimal characteristics for distinguishing arable land and exposed soil, its corresponding feature threshold
Value T is 118740,0.51.
For not carrying out handling low-light level body, in low-light level body comprising water body, architectural shadow and sub-fraction very
Dark building roof.Wherein water body is easier to identify, therefore water-outlet body and non-water body are first distinguished in low-light level body
Two kinds of atural objects.After choosing water body, non-water body sample, a series of spectral signatures and geometric properties corresponding to it are exported to be formed each
Two species very histogram of the sheet in different characteristic, is calculated separating degree J-M (Jeffries-Matusita) distance.It is right
There is preferably separation in modified normalization difference water body index, rectangle adaptability and yellow band feature than understanding both
Property, BShi distances respectively 2.58,3.28 and 2.60, J 1.85,1.92 and 1.85.Therefore, these three characteristic indexs work is chosen
To distinguish the optimal characteristics of water body and non-water body, its corresponding characteristic threshold value T is 0.32,0.34 and 254.44.Simultaneously as room
Room shade one is scheduled near house, therefore is increased on the basis of the description of three of the above feature in the pixel of impervious surface 150
Imaged object further improve accuracy for non-aqueous physical efficiency.
It is last then be that non-water body is divided into house shade and a part of dark roof.Its brightness for architectural shadow
Should be more darker than roof, therefore it is shade directly to define cutting object of the brightness less than 200, remaining is then house.Together
When shade is classified as hard ground, final classification results are as shown in Figure 6.
1.6Kappa coefficients overcome the gatherer process that other method excessively relies on sample data and sample, more objective evaluation
Classification quality.3136 sampled points are chosen using 1 to 500 topographic map is uniformly middle in the range of research area, are asked according to confusion matrix
The Kappa coefficients taken are 0.82, overall precision 0.86.
Claims (6)
1. a kind of rule-based fine dimension city impervious surface rapid extracting method, it is characterised in that comprise the following steps:
1.1 pairs of original high resolution multispectral images carry out atmospheric correction, fusion, geometric correction pretreatment, obtain having and enrich
Spectral information, high-resolution fusion evaluation;
1.2 follow easy first and difficult later, the obvious principle of characters of ground object otherness between node, establish the decision tree of extraction;
1.3 utilize chessboard partitioning algorithm on image pixel layer, view picture image is split by the road vectors data of input;For
Road carries out feature description with non-rice habitats, i.e. road is more than 0 to be overlapping with vector;The portion that it is not road in region that non-rice habitats, which are,
Point, distinguish road and non-rice habitats part;
Split using heterogeneous minimum region merging algorithm 1.4 pairs of non-rice habitats parts;
1.5 spectral signature, geometric properties, the topological characteristic according to type of ground objects, different atural objects are carried out with feature description, setting ground
The feature or combinations of features of species type, establish complete categorised decision tree, construct completely can be applied to high-resolution or
The classifying rules collection of middle low resolution remote sensing image;Imaged object identification and mark are carried out on feature space according to classifying rules collection
Know, complete the extraction of remote sensing information, obtain urban area impervious surface distribution results;
1.6 carry out the precision test of classification results by being evaluated after classification with Kappa index methods.
2. the rule-based fine dimension city impervious surface rapid extracting method as described in claim 1, it is characterised in that
Original high resolution multispectral image described in step 1.1 is high-resolution WorldView-3 images, has 8 multispectral ripples
Section image, i.e. blue wave band, green band, red band, near infrared band, seashore wave band, yellow band, red edge wave band
With the image of the wave band of near-infrared 2, multispectral spatial resolution is 1.6 meters, and panchromatic wave-band spatial resolution is 0.4 meter.
3. the rule-based fine dimension city impervious surface rapid extracting method as described in claim 1, it is characterised in that
Heterogeneous minimum region merging algorithm segmentation described in step 1.4, it is 80~100 that it, which recommends scale parameter scope, form factor
Weight is arranged to 0.1, i.e. spectrum Factor Weight is 0.9;Degree of the compacting factor and smoothness factor are 0.5, and each wave band participates in segmentation
Weight be 1.
4. the rule-based fine dimension city impervious surface rapid extracting method as described in claim 1, it is characterised in that
Atural object described in step 1.2 is building, road, vegetation, exposed soil, water body, artificial hard ground and arable land.
5. the rule-based fine dimension city impervious surface rapid extracting method as described in claim 1, it is characterised in that
Decision tree described in step 1.2, it is divided into five levels, eight nodes.
6. the rule-based fine dimension city impervious surface rapid extracting method as described in claim 1, it is characterised in that
Classifying rules collection described in step 1.5, its foundation are characterized as spectral signature, geometric properties, textural characteristics, user-defined feature and class
Correlated characteristic is built;
Described classifying rules collection, its feature selecting foundation is maximum separation degree index;
Described classifying rules collection, its road, non-rice habitats characteristic of division are length-width ratio, numerical value more than 3 for road;It is vegetation, non-
Vegetation classification is characterized as normalized differential vegetation index NDVI, blue wave band, and NDVI is more than 0.3 and blue wave band reflectivity is less than 300
For vegetation;Low-light level body, high brightness body, which are divided into, is characterized as the wave band of near-infrared 1, the wave band of near-infrared 2, maximum difference index, near-infrared
1 wave band and the wave band reflectivity of near-infrared 2 are less than 300, and maximum difference index is low-light level body more than 0.4;Permeable, impermeable moisture
Category feature is blue wave band, normalization water body index NDWI, and blue wave band reflectivity is less than 300 and NDWI and is less than 0 for dankly
Thing;Exposed soil, arable land characteristic of division are area, rectangle adaptability, and it is cultivated that area, which is more than 100,000 pixels and rectangle adaptability more than 0.5,
Ground;Hard ground, classification of buildings are characterized as degree of compacting, soil index WVSI, and degree of compacting is house less than 1.5 or so;NVSI is small
More than -0.05 or so it is hard ground in 0;Water body, non-water body characteristic of division be with waterproof atural object distance, yellow band, change
Enter type water body index MNDWI, be less than 200 pixels apart from waterproof atural object and yellow band reflectivity is less than less than 300 and MDNWI
0.3 is non-water body.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710856065.XA CN107609526A (en) | 2017-09-21 | 2017-09-21 | Rule-based fine dimension city impervious surface rapid extracting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710856065.XA CN107609526A (en) | 2017-09-21 | 2017-09-21 | Rule-based fine dimension city impervious surface rapid extracting method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107609526A true CN107609526A (en) | 2018-01-19 |
Family
ID=61061307
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710856065.XA Pending CN107609526A (en) | 2017-09-21 | 2017-09-21 | Rule-based fine dimension city impervious surface rapid extracting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107609526A (en) |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108537169A (en) * | 2018-04-09 | 2018-09-14 | 吉林大学 | A kind of high-resolution remote sensing image method for extracting roads based on center line and detection algorithm of having a lot of social connections |
CN109448016A (en) * | 2018-11-02 | 2019-03-08 | 三亚中科遥感研究所 | It is a kind of based on object-oriented and its be subordinate to rule remote sensing image tidal saltmarsh method |
CN109584284A (en) * | 2018-12-13 | 2019-04-05 | 宁波大学 | A kind of seashore wetland ground object sample extracting method of hierarchical decision making |
CN109598202A (en) * | 2018-11-05 | 2019-04-09 | 中国科学院遥感与数字地球研究所 | A kind of object-based satellite image multi objective built-up areas extraction method |
CN109657598A (en) * | 2018-12-13 | 2019-04-19 | 宁波大学 | Seashore wetland Classification in Remote Sensing Image method based on Stratified Strategy |
CN109740645A (en) * | 2018-12-20 | 2019-05-10 | 核工业北京地质研究院 | A kind of CART Decision-Tree Method suitable for high score No.1 image |
CN109753916A (en) * | 2018-12-28 | 2019-05-14 | 厦门理工学院 | A kind of vegetation index spatial scaling model building method and device |
CN110309780A (en) * | 2019-07-01 | 2019-10-08 | 中国科学院遥感与数字地球研究所 | High resolution image houseclearing based on BFD-IGA-SVM model quickly supervises identification |
CN110427914A (en) * | 2019-08-14 | 2019-11-08 | 西南交通大学 | A kind of transmission of electricity corridor vegetation threat early warning method based on satellite remote sensing |
CN110929739A (en) * | 2019-11-21 | 2020-03-27 | 成都理工大学 | Automatic impervious surface range remote sensing iterative extraction method |
CN111157524A (en) * | 2020-01-09 | 2020-05-15 | 北京观澜智图科技有限公司 | Road material identification method and device based on high-resolution image and electronic equipment |
CN111199236A (en) * | 2020-01-06 | 2020-05-26 | 国家卫星气象中心(国家空间天气监测预警中心) | Method, equipment and medium for extracting water body in satellite image by using decision tree |
CN111199195A (en) * | 2019-12-26 | 2020-05-26 | 中科禾信遥感科技(苏州)有限公司 | Pond state full-automatic monitoring method and device based on remote sensing image |
CN111222536A (en) * | 2019-11-19 | 2020-06-02 | 南京林业大学 | City green space information extraction method based on decision tree classification |
CN111709379A (en) * | 2020-06-18 | 2020-09-25 | 谢国雪 | Remote sensing image-based hilly area citrus planting land plot monitoring method and system |
CN111783625A (en) * | 2020-06-29 | 2020-10-16 | 盐城工学院 | Rapid extraction method for information of water-impermeable surface of plain river network area |
CN112329790A (en) * | 2020-10-27 | 2021-02-05 | 厦门理工学院 | Rapid extraction method for urban impervious surface information |
CN112818749A (en) * | 2020-12-31 | 2021-05-18 | 中国电子科技集团公司第二十七研究所 | Multi-cropping mode remote sensing monitoring method for bulk grain and oil crops in double cropping area of one year |
CN113592770A (en) * | 2021-06-23 | 2021-11-02 | 中国科学院南京地理与湖泊研究所 | Algal bloom remote sensing identification method for removing influence of aquatic weeds |
CN114120133A (en) * | 2021-12-03 | 2022-03-01 | 中国科学院地理科学与资源研究所 | Urban road information extraction method for aggregating multiple factors |
CN114266968A (en) * | 2021-12-16 | 2022-04-01 | 河南大学 | Remote sensing automatic interpretation method for different land coverage types of city |
CN114842356A (en) * | 2022-07-01 | 2022-08-02 | 江西师范大学 | High-resolution earth surface type sample automatic generation method, system and equipment |
CN115620043A (en) * | 2022-05-18 | 2023-01-17 | 上海航遥信息技术有限公司 | Hyperspectral and spatial data fused geographic entity semantic annotation method |
CN117456378A (en) * | 2023-12-20 | 2024-01-26 | 山东锋士信息技术有限公司 | Water conservancy digital twin base element realization method and system based on satellite remote sensing |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101840581A (en) * | 2010-02-05 | 2010-09-22 | 北京交通大学 | Method for extracting profile of building from satellite remote sensing image |
CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
CN105046242A (en) * | 2015-08-24 | 2015-11-11 | 山东省农业可持续发展研究所 | Asparagus planting area extraction method based on Landsat 8 image two-dimensional feature space |
CN105404753A (en) * | 2015-12-08 | 2016-03-16 | 中国科学院东北地理与农业生态研究所 | Marsh wetland mapping method based on object-oriented random forest classification method and medium-resolution remote sensing image |
CN106971156A (en) * | 2017-03-22 | 2017-07-21 | 中国地质科学院矿产资源研究所 | Rare earth mining area remote sensing information extraction method based on object-oriented classification |
-
2017
- 2017-09-21 CN CN201710856065.XA patent/CN107609526A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101840581A (en) * | 2010-02-05 | 2010-09-22 | 北京交通大学 | Method for extracting profile of building from satellite remote sensing image |
CN103364781A (en) * | 2012-04-11 | 2013-10-23 | 南京财经大学 | Remote sensing data and geographical information system-based grainfield ground reference point screening method |
CN105046242A (en) * | 2015-08-24 | 2015-11-11 | 山东省农业可持续发展研究所 | Asparagus planting area extraction method based on Landsat 8 image two-dimensional feature space |
CN105404753A (en) * | 2015-12-08 | 2016-03-16 | 中国科学院东北地理与农业生态研究所 | Marsh wetland mapping method based on object-oriented random forest classification method and medium-resolution remote sensing image |
CN106971156A (en) * | 2017-03-22 | 2017-07-21 | 中国地质科学院矿产资源研究所 | Rare earth mining area remote sensing information extraction method based on object-oriented classification |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108537169A (en) * | 2018-04-09 | 2018-09-14 | 吉林大学 | A kind of high-resolution remote sensing image method for extracting roads based on center line and detection algorithm of having a lot of social connections |
CN108537169B (en) * | 2018-04-09 | 2022-01-25 | 吉林大学 | High-resolution remote sensing image road extraction method based on center line and road width detection algorithm |
CN109448016A (en) * | 2018-11-02 | 2019-03-08 | 三亚中科遥感研究所 | It is a kind of based on object-oriented and its be subordinate to rule remote sensing image tidal saltmarsh method |
CN109598202A (en) * | 2018-11-05 | 2019-04-09 | 中国科学院遥感与数字地球研究所 | A kind of object-based satellite image multi objective built-up areas extraction method |
CN109584284A (en) * | 2018-12-13 | 2019-04-05 | 宁波大学 | A kind of seashore wetland ground object sample extracting method of hierarchical decision making |
CN109657598A (en) * | 2018-12-13 | 2019-04-19 | 宁波大学 | Seashore wetland Classification in Remote Sensing Image method based on Stratified Strategy |
CN109740645A (en) * | 2018-12-20 | 2019-05-10 | 核工业北京地质研究院 | A kind of CART Decision-Tree Method suitable for high score No.1 image |
CN109753916A (en) * | 2018-12-28 | 2019-05-14 | 厦门理工学院 | A kind of vegetation index spatial scaling model building method and device |
CN110309780A (en) * | 2019-07-01 | 2019-10-08 | 中国科学院遥感与数字地球研究所 | High resolution image houseclearing based on BFD-IGA-SVM model quickly supervises identification |
CN110427914A (en) * | 2019-08-14 | 2019-11-08 | 西南交通大学 | A kind of transmission of electricity corridor vegetation threat early warning method based on satellite remote sensing |
CN110427914B (en) * | 2019-08-14 | 2021-09-28 | 西南交通大学 | Power transmission corridor vegetation threat early warning method based on satellite remote sensing |
CN111222536A (en) * | 2019-11-19 | 2020-06-02 | 南京林业大学 | City green space information extraction method based on decision tree classification |
CN110929739A (en) * | 2019-11-21 | 2020-03-27 | 成都理工大学 | Automatic impervious surface range remote sensing iterative extraction method |
CN111199195A (en) * | 2019-12-26 | 2020-05-26 | 中科禾信遥感科技(苏州)有限公司 | Pond state full-automatic monitoring method and device based on remote sensing image |
CN111199236A (en) * | 2020-01-06 | 2020-05-26 | 国家卫星气象中心(国家空间天气监测预警中心) | Method, equipment and medium for extracting water body in satellite image by using decision tree |
CN111157524A (en) * | 2020-01-09 | 2020-05-15 | 北京观澜智图科技有限公司 | Road material identification method and device based on high-resolution image and electronic equipment |
CN111709379A (en) * | 2020-06-18 | 2020-09-25 | 谢国雪 | Remote sensing image-based hilly area citrus planting land plot monitoring method and system |
CN111709379B (en) * | 2020-06-18 | 2023-04-18 | 广西壮族自治区农业科学院 | Remote sensing image-based hilly area citrus planting land plot monitoring method and system |
CN111783625A (en) * | 2020-06-29 | 2020-10-16 | 盐城工学院 | Rapid extraction method for information of water-impermeable surface of plain river network area |
CN112329790A (en) * | 2020-10-27 | 2021-02-05 | 厦门理工学院 | Rapid extraction method for urban impervious surface information |
CN112329790B (en) * | 2020-10-27 | 2024-01-23 | 厦门理工学院 | Quick extraction method for urban impervious surface information |
CN112818749B (en) * | 2020-12-31 | 2022-09-13 | 中国电子科技集团公司第二十七研究所 | Multi-cropping mode remote sensing monitoring method for bulk grain and oil crops in double cropping area of one year |
CN112818749A (en) * | 2020-12-31 | 2021-05-18 | 中国电子科技集团公司第二十七研究所 | Multi-cropping mode remote sensing monitoring method for bulk grain and oil crops in double cropping area of one year |
CN113592770A (en) * | 2021-06-23 | 2021-11-02 | 中国科学院南京地理与湖泊研究所 | Algal bloom remote sensing identification method for removing influence of aquatic weeds |
CN113592770B (en) * | 2021-06-23 | 2024-02-23 | 中国科学院南京地理与湖泊研究所 | Algal bloom remote sensing identification method for removing influence of aquatic weeds |
CN114120133A (en) * | 2021-12-03 | 2022-03-01 | 中国科学院地理科学与资源研究所 | Urban road information extraction method for aggregating multiple factors |
CN114266968A (en) * | 2021-12-16 | 2022-04-01 | 河南大学 | Remote sensing automatic interpretation method for different land coverage types of city |
CN114266968B (en) * | 2021-12-16 | 2023-01-31 | 河南大学 | Remote sensing automatic interpretation method for different land coverage types of city |
CN115620043A (en) * | 2022-05-18 | 2023-01-17 | 上海航遥信息技术有限公司 | Hyperspectral and spatial data fused geographic entity semantic annotation method |
CN114842356A (en) * | 2022-07-01 | 2022-08-02 | 江西师范大学 | High-resolution earth surface type sample automatic generation method, system and equipment |
CN114842356B (en) * | 2022-07-01 | 2022-10-04 | 江西师范大学 | High-resolution earth surface type sample automatic generation method, system and equipment |
CN117456378A (en) * | 2023-12-20 | 2024-01-26 | 山东锋士信息技术有限公司 | Water conservancy digital twin base element realization method and system based on satellite remote sensing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107609526A (en) | Rule-based fine dimension city impervious surface rapid extracting method | |
CN106384081B (en) | Slope farmland extraction method and system based on high-resolution remote sensing image | |
CN103971115B (en) | Automatic extraction method for newly-increased construction land image spots based on NDVI and PanTex index | |
Seto et al. | Monitoring land-use change in the Pearl River Delta using Landsat TM | |
CN111598045B (en) | Remote sensing farmland change detection method based on object spectrum and mixed spectrum | |
CN111832518B (en) | Space-time fusion-based TSA remote sensing image land utilization method | |
CN111340826A (en) | Single tree crown segmentation algorithm for aerial image based on superpixels and topological features | |
CN106023133B (en) | A kind of high-resolution remote sensing image Clean water withdraw method based on multiple features combining processing | |
CN112800973B (en) | Spartina alterniflora extraction method based on vegetation phenological feature decision | |
CN107392130A (en) | Classification of Multispectral Images method based on threshold adaptive and convolutional neural networks | |
WO2022067598A1 (en) | Method of individual tree crown segmentation from airborne lidar data using novel gaussian filter and energy function minimization | |
CN103839267B (en) | Building extracting method based on morphological building indexes | |
CN104851113A (en) | Urban vegetation automatic extraction method of multiple-spatial resolution remote sensing image | |
Lu et al. | Detection of urban expansion in an urban-rural landscape with multitemporal QuickBird images | |
CN106778629B (en) | Greenhouse identification method and device | |
CN108051371A (en) | A kind of shadow extraction method of ecology-oriented environment parameter remote-sensing inversion | |
CN114926748A (en) | Soybean remote sensing identification method combining Sentinel-1/2 microwave and optical multispectral images | |
CN106340005A (en) | High-resolution remote sensing image unsupervised segmentation method based on scale parameter automatic optimization | |
CN105404873A (en) | Winter wheat recognition method based on NDVI time sequence coordinate conversion | |
CN115641412A (en) | Hyperspectral data-based three-dimensional semantic map generation method | |
CN103208121B (en) | Based on the remote sensing image segmentation method that bounds constraint merges with two benches | |
Smits et al. | Individual tree identification using different LIDAR and optical imagery data processing methods | |
CN102231190B (en) | Automatic extraction method for alluvial-proluvial fan information | |
CN111882573A (en) | Cultivated land plot extraction method and system based on high-resolution image data | |
Gong et al. | Vineyard identification in an oak woodland landscape with airborne digital camera imagery |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180119 |
|
WD01 | Invention patent application deemed withdrawn after publication |