CN107480634A - A kind of geographical national conditions ground mulching monitoring method based on multistage target classification - Google Patents
A kind of geographical national conditions ground mulching monitoring method based on multistage target classification Download PDFInfo
- Publication number
- CN107480634A CN107480634A CN201710688315.3A CN201710688315A CN107480634A CN 107480634 A CN107480634 A CN 107480634A CN 201710688315 A CN201710688315 A CN 201710688315A CN 107480634 A CN107480634 A CN 107480634A
- Authority
- CN
- China
- Prior art keywords
- classification
- image
- ground mulching
- period
- national conditions
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of geographical national conditions ground mulching monitoring method based on multistage target classification, including：Current image is divided into the primitive of object-oriented analysis with reference to the image feature of ground mulching vector data bin information and current period high-definition remote sensing image data for the first period, multiclass classification information early stage is obtained using the classification information of ground mulching vector data for the first period, the imaged object of all categories included in early stage multiclass classification information is screened to obtain sample object of all categories according to the image feature of current period image, and current period image is exercised supervision step by step classification using the sample, obtain current period classification results, difference by comparing imaged object classification early stage and current class obtains ground mulching result of variations.The present invention realizes the whole-course automation of monitoring process, improves classification and the change accuracy of detection of group, realizes the initial identification of automatic, efficient ground mulching change, reduce the artificial workload of interior industry.
Description
Technical field
The invention belongs to remote sensing image information extraction and change detection techniques field, and in particular to one kind is based on multistage target
The geographical national conditions ground mulching monitoring method of classification.
Background technology
2 months 2013, the General Office of the State Council issued《On carrying out the notice of the geographical national conditions generaI investigation in the whole nation for the first time》, ground
Manage national conditions generaI investigation to start to carry out comprehensively, to 2015, the geographical national conditions generaI investigation in the whole nation for the first time was basically completed.The ground from 2016
Reason national conditions monitoring has been enter into the normalization monitoring stage.At present, the ground mulching of geographical national conditions monitoring and its change information with
Geographical national conditions key element and its change information are applied in natural, humane, society, economic dispatch various fields, to formulate country
Reliable basis are provided with Regional development planning and decision-making, the promotion sustainable development of socio-economy.At this stage, geographical national conditions monitoring
Middle ground mulching monitoring mainly utilizes high-resolution remote sensing image, and the method combined using interior field operation is carried out, according to " interior industry
Based on, supplemented by field operation " principle arrangement task, main technical flows be interior industry remote Sensing Interpretation-field operation verification-interior industry editor it is whole
Reason.Wherein interior industry remote Sensing Interpretation is mainly interpreted by artificial visual and realized, interpretation process consume substantial amounts of human and material resources and when
Between.Therefore, there is an urgent need to industry work in being reduced using the automatic/semi-automatic remote sensing image interpretation method of computer for each relevant departments
Measure and improve interior industry operating efficiency and precision.
Detection method research based on remote sensing image change is always the research of remote sensing information process and analysis technical field
Focus.With the continuous development of remote sensing technology, remote sensing image variation detection method is also increasingly abundant, and has obtained widely should
With.However, so far and in the absence of the remote sensing image variation detection method with universality, it usually needs designed according to particular demands
With targetedly change detection model and method.Meanwhile with the raising of remote sensing image spatial resolution, high-definition remote sensing
Image is used widely in every profession and trade.But the raising of remote sensing image spatial resolution is providing more rich, detailed earth's surface
While information, the difficulty of remote sensing image automation interpretation is also considerably increased.
The content of the invention
It is an object of the invention to provide it is a kind of based on multistage target classification geographical national conditions ground mulching monitoring method, with
Reduce the workload of geographical national conditions monitoring surveyor and improve the efficiency of ground mulching monitoring.
Therefore, technical scheme is as follows：
A kind of geographical national conditions ground mulching monitoring method based on multistage target classification, comprises the following steps：
A) Image Segmentation：Input the high-resolution remote sensing image of ground mulching vector data and current period for the first period
Data, with reference to ground mulching vector data and the image feature of current period high-resolution remote sensing image for the first period, to working as
The high-resolution remote sensing image in preceding period carries out Image Segmentation, obtains base image object set；
B) previous phase ground mulching classification:Utilize the classification of the ground mulching vector data for the first period of step a inputs
Information obtains multistage classification information, and assigns all imaged objects in base image object set；
C) screening sample：The multistage classification obtained according to basic geographical national conditions Contents for Monitoring and index system and step b
Information is screened to imaged object, respectively obtains the sample object of each rank and classification；
D) current image hierarchical classification：Using the sample object that step c is filtered out to current period high-definition remote sensing shadow
As being classified step by step；
E) ground mulching mutation analysis：The classification results for the current period that comparison step d is obtained obtain previous with step b
Period classification results, identify the imaged object that classification changes.
In above-mentioned step a), the imaged object is the set of the adjacent pixel with feature homogeney；The image
Object set is the set for all imaged objects that image includes, and by Image Segmentation, image is divided into non-overlapping sub-district
Domain and obtain；The base image object set is the smallest dimension object set for carrying out object oriented classification.
In above-mentioned step a), the method for the Image Segmentation comprises the following steps：
1) the bin positional information segmentation image of ground mulching vector data for the first period is utilized；
2) it is comprehensive current using multi-scale division algorithm using the imaged object that above-mentioned segmentation result obtains as initial object
The image feature of image, obtains the imaged object set of smaller yardstick, the scale size of imaged object by changing graphic yardstick
Size determines.
In above-mentioned step b), the method for carrying out multiclass classification is：According to basic geographical national conditions Contents for Monitoring and index
System carries out three-level classification, wherein, the low other classification of one-level should be included in the corresponding high other classification of one-level.
In above-mentioned step c), using maximum variance between clusters threshold value, the image included respectively to all types
Object is screened, sample pair of the imaged object that member-retaining portion classification in current image does not change as all types
As.
In above-mentioned step d), by the use of the sample object that step c) is obtained as training set, entered using Random Forest model
Row training and classification.
In above-mentioned step d), the process classified step by step is：The other category classification of high one-level is carried out first, further according to
High-level classification results enter the other category classification of places one-level to its all kinds of included imaged object respectively.
For the actual demand of geographical national conditions ground mulching monitoring, the present invention proposes the geography based on multistage target classification
National conditions ground mulching monitoring method.This method utilizes geography national conditions monitoring result early stage and current period high-resolution remote sensing image
Data realize the automatic monitoring of geographical national conditions ground mulching, using object-oriented remote sensing image processing and analytical technology,
According to geographical national conditions Contents for Monitoring and the classification subordinate relation of index system, the automation point of high-resolution remote sensing image is realized
Class detects with change.
The key of the algorithm is to utilize ground mulching findings of the survey for the first period and current period high-resolution remote sensing image
Data realize the automatically screening to sample, training and the classification to current image, pass through the policy control point of hierarchical classification
The stability and precision of class, so as to realize that what species over the ground did not changed accurately identifies.The present invention provides change for human interpretation
The data of atural object, the change detection process realized has higher automaticity and change testing result recall ratio is higher,
Interior industry operating efficiency, the workload of less human interpretation can be greatly improved on the basis of monitoring accuracy is ensured.
Compared with prior art, the invention has the advantages that：
The present invention obtains sample by carrying out automatically screening to the vector data of previous phase geography national conditions ground mulching, really
The whole-course automation of monitoring process is protected；The classification strategy proposed can improve classification and the change detection essence of group
Degree, it is achieved thereby that automatic, the efficient initial identification of ground mulching change, reduces the work of geographical national conditions monitoring surveyor
Amount.
Brief description of the drawings
Fig. 1 is the flow chart of the monitoring method of the present invention.
Embodiment
Technical scheme for a better understanding of the present invention, the present invention is done with reference to the accompanying drawings and examples further
Describe in detail.
The present invention be based on multistage target classification geographical national conditions ground mulching monitoring method, the embodiment of reference picture 1, this
The method and step of invention is as follows：
A) Image Segmentation
Input the high-definition remote sensing image data of ground mulching vector data and current period for the first period.First with
The bin positional information of ground mulching vector data is split to current period remote sensing image for the first period, obtain one it is larger
The imaged object collection of yardstick.Then use multi-scale division algorithm using the imaged object under the yardstick to integrate to mark off as initial object
The imaged object of smaller yardstick, as subsequent classification and the analysis primitive of change detection.
Present invention progress segmentation twice is the atural object changed because of front and rear period be present, inevitable in current period remote sensing image
In the presence of the atural object inconsistent with previous phase ground mulching, it need to consider that current characteristics of remote sensing image marks off the atural object changed and formed
New imaged object.
B) previous phase ground mulching classification
At present, geographical national conditions Contents for Monitoring and index are classified for three-level, as shown in table 1.
According to geographical national conditions Contents for Monitoring and index and the classification information of vector data, the image pair obtained for step a)
As assigning multistage classification information.As certain pel classification information is the shadow in shrub orchard, then the pel correspondence position in vector data
As the classification information of object includes plantation soil, orchard, shrub orchard three-level classification information.Subordinate pass between different stage be present
System, i.e., the low other classification of one-level should be contained in the corresponding high other classification of one-level, such as：Classification information is the shadow in shrub orchard
As object falls within orchard class and plantation soil class.
The earth's surface national conditions of table 1 monitoring ground mulching collection content example
| One-level | Two level | Three-level |
| Plant soil | ||
| Paddy field | Paddy field | |
| Dry land | Dry land | |
| Orchard | ||
| Qiao fills orchard | ||
| Liana orchard | ||
| Draft orchard | ||
| …… | …… | |
| Building construction | ||
| Multilayer and above building construction area | ||
| High-density multi-layered and above building construction area | ||
| Low-density multilayer and above building construction area | ||
| …… | …… | |
| Structures | ||
| Harden earth's surface | ||
| …… | ||
| Solidifying pool | ||
| …… | ||
| …… | …… | …… |
C) screening sample
Screening sample is carried out to the imaged object of all categories included respectively, it is ensured that the sample retained is stablizes sample, i.e.,
In the imaged object that current period does not change, the sample object of each rank and classification is respectively obtained.The present invention is using most
Big Ostu method analysis determines screening threshold value, and the stable sample of a member-retaining portion is used for next step classification based training.
The purpose of present invention progress screening sample is to ensure that the precision of image classification.Because atural object changes, directly
Utilize the imaged object classification information that ground mulching vector data for the first period obtains and the actual feelings of current period imaged object
There is some difference for condition, and the sample information of mistake certainly will reduce nicety of grading, therefore need to be to the imaged object that is obtained in step b)
Stable imaged object is filtered out on the basis of classification information as classification samples.
D) current image hierarchical classification
Using random forests algorithm, it is trained using the obtained sample objects of step c, and the image of current period is entered
Row classification, assorting process are carried out step by step.According to geographical national conditions Contents for Monitoring and index system, the training of one-level class is carried out first and is divided
Class, then included imaged objects all kinds of to one-level class carry out secondary classifications, included images finally all kinds of to two level class
Object carries out three-level classification.Such as the taxonomic hierarchies that table 1 provides, plantation soil, building construction, structure are carried out according to sample object first
Build the training and classification of the categories such as thing.Then the imaged object included respectively to above-mentioned classification is trained and classified, such as
The imaged object that plantation soil class is included, the classes such as paddy field, dry land, orchard need to be carried out according to the two level classification information of imaged object
Two level classification training and classification.Finally all kinds of to two level the classification training of carry out three-level classification and classification respectively, such as to orchard class
Comprising imaged object carry out the tall classification for filling the classes such as orchard, liana orchard, draft orchard.
The present invention carries out hierarchical classification according to the physical characteristic between classification system classifications at different levels with subordinate relation can be effective
Control tactics precision simultaneously improves classification effectiveness.
E) ground mulching mutation analysis
The classification results for the first period that the classification results for the current period that comparison step d) is obtained obtain with step b), classification
Inconsistent imaged object is the imaged object to change.
Claims (7)
1. a kind of geographical national conditions ground mulching monitoring method based on multistage target classification, it is characterised in that comprise the following steps：
A) Image Segmentation：Input the high-resolution remote sensing image number of ground mulching vector data and current period for the first period
According to reference to ground mulching vector data and the image feature of current period high-resolution remote sensing image for the first period, to current
The high-resolution remote sensing image in period carries out Image Segmentation, obtains base image object set；
B) previous phase ground mulching classification:Utilize the classification information of the ground mulching vector data for the first period of step a inputs
Multistage classification information is obtained, and assigns all imaged objects in base image object set；
C) screening sample：The multistage classification information obtained according to basic geographical national conditions Contents for Monitoring and index system and step b
Imaged object is screened, respectively obtains the sample object of each rank and classification；
D) current image hierarchical classification：Using the sample object that step c is filtered out to current period high-resolution remote sensing image by
Level is classified；
E) ground mulching mutation analysis：The classification results for the current period that comparison step d is obtained obtain for the first period with step b
Classification results, identify the imaged object that classification changes.
2. the geographical national conditions ground mulching monitoring method according to claim 1 based on multistage target classification, its feature exist
In：In step a),
The imaged object is the set of the adjacent pixel with feature homogeney；
The imaged object collection is the set for all imaged objects that image includes, and by Image Segmentation, image is divided into mutually
Not overlapping subregion and obtain；
The base image object set is the smallest dimension object set for carrying out object oriented classification.
3. the geographical national conditions ground mulching monitoring method according to claim 1 based on multistage target classification, its feature exist
In in step a), the method for the Image Segmentation comprises the following steps：
1) the bin positional information segmentation image of ground mulching vector data for the first period is utilized；
2) using the imaged object that above-mentioned segmentation result obtains as initial object, using multi-scale division algorithm, comprehensive current image
Image feature, obtain the imaged object set of smaller yardstick, the scale size of imaged object by changing graphic scale size
Determine.
4. the geographical national conditions ground mulching monitoring method according to claim 1 based on multistage target classification, its feature exist
In in step b), the method for carrying out multiclass classification is：Three-level is carried out according to basic geographical national conditions Contents for Monitoring and index system
Classification, wherein, the low other classification of one-level should be included in the corresponding high other classification of one-level.
5. the geographical national conditions ground mulching monitoring method according to claim 1 based on multistage target classification, its feature exist
In：In step c), using maximum variance between clusters threshold value, the imaged object included respectively to all types sieves
Choosing, sample object of the imaged object that member-retaining portion classification in current image does not change as all types.
6. the geographical national conditions ground mulching monitoring method according to claim 1 based on multistage target classification, its feature exist
In in step d), by the use of the sample object that step c) is obtained as training set, being trained and divided using Random Forest model
Class.
7. the geographical national conditions ground mulching monitoring method according to claim 6 based on multistage target classification, its feature exist
In in step d), the process classified step by step is：The other category classification of high one-level is carried out first, further according to high-level classification
As a result the other category classification of places one-level is entered to its all kinds of included imaged object respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710688315.3A CN107480634A (en) | 2017-08-12 | 2017-08-12 | A kind of geographical national conditions ground mulching monitoring method based on multistage target classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710688315.3A CN107480634A (en) | 2017-08-12 | 2017-08-12 | A kind of geographical national conditions ground mulching monitoring method based on multistage target classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107480634A true CN107480634A (en) | 2017-12-15 |
Family
ID=60599535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710688315.3A Pending CN107480634A (en) | 2017-08-12 | 2017-08-12 | A kind of geographical national conditions ground mulching monitoring method based on multistage target classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107480634A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108197328A (en) * | 2018-02-08 | 2018-06-22 | 江苏省基础地理信息中心 | Geographical national conditions data variation type automatic identifying method |
| CN113359133A (en) * | 2021-06-03 | 2021-09-07 | 电子科技大学 | Object-oriented change detection method for collaborative optical and radar remote sensing data |
| CN113762070A (en) * | 2021-07-26 | 2021-12-07 | 中国测绘科学研究院 | Surface coverage classification sample collection method for deep learning |
| CN114216445A (en) * | 2021-12-08 | 2022-03-22 | 中国电建集团成都勘测设计研究院有限公司 | Water and soil conservation monitoring method for rapidly determining vegetation coverage in field |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254319A (en) * | 2011-04-19 | 2011-11-23 | 中科九度(北京)空间信息技术有限责任公司 | Method for carrying out change detection on multi-level segmented remote sensing image |
| CN104881677A (en) * | 2015-05-08 | 2015-09-02 | 北京师范大学 | Optimum segmentation dimension determining method for remotely-sensed image land cover classification |
| CN105005782A (en) * | 2014-04-24 | 2015-10-28 | 中国科学院遥感与数字地球研究所 | Fine method for global vegetation classification based on multi-temporal remote sensing data and spectroscopic data |
| CN105260738A (en) * | 2015-09-15 | 2016-01-20 | 武汉大学 | Method and system for detecting change of high-resolution remote sensing image based on active learning |
| CN105405133A (en) * | 2015-11-04 | 2016-03-16 | 河海大学 | Remote sensing image alteration detection method |
| CN105447494A (en) * | 2015-12-01 | 2016-03-30 | 二十一世纪空间技术应用股份有限公司 | Vegetable field monitoring method based on multi-source multi-temporal remote sensing image data |
| CN105809177A (en) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | Method used for actuating remote sensing image classification |
| CN106355143A (en) * | 2016-08-25 | 2017-01-25 | 中国农业大学 | Seed maize field identification method and system based on multi-source and multi-temporal high resolution remote sensing data |
| CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
-
2017
- 2017-08-12 CN CN201710688315.3A patent/CN107480634A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254319A (en) * | 2011-04-19 | 2011-11-23 | 中科九度(北京)空间信息技术有限责任公司 | Method for carrying out change detection on multi-level segmented remote sensing image |
| CN105005782A (en) * | 2014-04-24 | 2015-10-28 | 中国科学院遥感与数字地球研究所 | Fine method for global vegetation classification based on multi-temporal remote sensing data and spectroscopic data |
| CN105809177A (en) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | Method used for actuating remote sensing image classification |
| CN104881677A (en) * | 2015-05-08 | 2015-09-02 | 北京师范大学 | Optimum segmentation dimension determining method for remotely-sensed image land cover classification |
| CN105260738A (en) * | 2015-09-15 | 2016-01-20 | 武汉大学 | Method and system for detecting change of high-resolution remote sensing image based on active learning |
| CN105405133A (en) * | 2015-11-04 | 2016-03-16 | 河海大学 | Remote sensing image alteration detection method |
| CN105447494A (en) * | 2015-12-01 | 2016-03-30 | 二十一世纪空间技术应用股份有限公司 | Vegetable field monitoring method based on multi-source multi-temporal remote sensing image data |
| CN106355143A (en) * | 2016-08-25 | 2017-01-25 | 中国农业大学 | Seed maize field identification method and system based on multi-source and multi-temporal high resolution remote sensing data |
| CN106650689A (en) * | 2016-12-30 | 2017-05-10 | 厦门理工学院 | Coastal city time sequence land utilization information extracting method |
Non-Patent Citations (1)
| Title |
|---|
| 关海鹰: "基于多尺度分割方法的地理国情变化检测研究", 《科技创新导报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108197328A (en) * | 2018-02-08 | 2018-06-22 | 江苏省基础地理信息中心 | Geographical national conditions data variation type automatic identifying method |
| CN108197328B (en) * | 2018-02-08 | 2020-12-22 | 江苏省基础地理信息中心 | Automatic identification method for geographical national condition data change types |
| CN113359133A (en) * | 2021-06-03 | 2021-09-07 | 电子科技大学 | Object-oriented change detection method for collaborative optical and radar remote sensing data |
| CN113359133B (en) * | 2021-06-03 | 2022-03-15 | 电子科技大学 | Object-oriented change detection method for collaborative optical and radar remote sensing data |
| CN113762070A (en) * | 2021-07-26 | 2021-12-07 | 中国测绘科学研究院 | Surface coverage classification sample collection method for deep learning |
| CN114216445A (en) * | 2021-12-08 | 2022-03-22 | 中国电建集团成都勘测设计研究院有限公司 | Water and soil conservation monitoring method for rapidly determining vegetation coverage in field |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829399B (en) | Vehicle-mounted road scene point cloud automatic classification method based on deep learning | |
| Keshtkar et al. | Land-cover classification and analysis of change using machine-learning classifiers and multi-temporal remote sensing imagery | |
| CN108846832A (en) | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data | |
| Bauer et al. | Per-parcel land use classification in urban areas applying a rule-based technique | |
| Reger et al. | Identifying patterns of land-cover change and their physical attributes in a marginal European landscape | |
| CN107480634A (en) | A kind of geographical national conditions ground mulching monitoring method based on multistage target classification | |
| US20160003792A1 (en) | Functional soil maps | |
| Tiede | A new geospatial overlay method for the analysis and visualization of spatial change patterns using object-oriented data modeling concepts | |
| Gong et al. | Effect of terrain on landscape patterns and ecological effects by a gradient-based RS and GIS analysis | |
| CN113223042A (en) | Intelligent acquisition method and equipment for remote sensing image deep learning sample | |
| CN102867183A (en) | Method and device for detecting littered objects of vehicle and intelligent traffic monitoring system | |
| Suárez‐Seoane et al. | Scale dependence of spatial patterns and cartography on the detection of landscape change: relationships with species’ perception | |
| CN109447111A (en) | A kind of remote sensing supervised classification method based on subclass training sample | |
| Tormos et al. | Object-based image analysis for operational fine-scale regional mapping of land cover within river corridors from multispectral imagery and thematic data | |
| CN114398951A (en) | Land use change driving factor mining method based on random forest and crowd-sourced geographic information | |
| CN102938069A (en) | Pure and mixed pixel automatic classification method based on information entropy | |
| CN104268557B (en) | Polarization SAR sorting technique based on coorinated training and depth S VM | |
| CN113516060A (en) | Building space fusion discrimination method and device based on multi-temporal high-resolution data | |
| CN112906627A (en) | Green pricklyash peel identification method based on semantic segmentation | |
| Pazúr et al. | The distribution of selected CORINE land cover classes in different natural landscapes in Slovakia: Methodological framework and applications | |
| CN114170441B (en) | Roadside tree automatic extraction method based on geographic national condition data and image classification | |
| Droj | GIS and remote sensing in environmental management | |
| Yadahalli et al. | Accuracy Assessment of Supervised and Unsupervised Classification using Landsat-8 Imagery of D-7 Shahapur Branch Canal of UKP Command Area Karnataka, India | |
| CN115457386A (en) | Village land informatization generation method | |
| Ernstrom et al. | Enhanced soils information systems from advances in computer technology |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171215 |
|
| RJ01 | Rejection of invention patent application after publication |