CN108846832A - A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data - Google Patents
A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data Download PDFInfo
- Publication number
- CN108846832A CN108846832A CN201810538861.3A CN201810538861A CN108846832A CN 108846832 A CN108846832 A CN 108846832A CN 201810538861 A CN201810538861 A CN 201810538861A CN 108846832 A CN108846832 A CN 108846832A
- Authority
- CN
- China
- Prior art keywords
- spot
- water body
- period
- vegetation
- oriented
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- 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
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of change detecting method and system based on multi-temporal remote sensing image and GIS data, this method are dedicated to solving the problems, such as exist in segmentation, classification and change detecting method based on multidate image and GIS data.Including following steps：Step 1, Remote Sensing Image Segmentation and Pixel-level extracting change information mainly include Remote Sensing Image Segmentation and Pixel-level variation detection；Step 2, the classification of remote-sensing images of object-oriented mainly includes changing graphic decision, automated sample selection and object-oriented SVM Decision-Tree Classifier Model；Step 3, change detection after object oriented classification.With it is existing be changed the method for information extraction using remote sensing image and GIS data compared with, the present invention can be realized automation and can large-area applications change detection method flow, the correlation of remote sensing image and GIS data can sufficiently be excavated, the attribute informations such as the classification using GIS data improve the accuracy of extracting change information.
Description
Technical field
The present invention relates to the field rs and gis (GIS, Geographic Information System),
In particular to remote sensing image and GIS data change information automatic testing method more particularly to a kind of multi-temporal remote sensing image is utilized
The method that the change information for carrying out object-oriented with GIS data detects automatically.
Background technique
In remote sensing fields, variation detection is the remote sensing image using areal difference phase, anti-by comparison spectrum
Value, the difference of Texture eigenvalue are penetrated, the process of region of variation is obtained.Detection is changed using remote sensing image to be widely used
In environmental monitoring, agricultural investigation, urban sprawl research, disaster monitoring, destroy the fields such as assessment.In the field GIS, GIS data
It is the symbolic formulation of topographical features landforms and atural object, is the result of field survey and image interpretation.GIS data not only contains several
What location information, and further comprise the semantic informations such as atural object classification abundant, attribute.Development and GIS with remote sensing technology
Be related to the extensive use of spatial information industry in environment, agricultural, city, disaster relief mitigation, geographical national conditions etc., using remote sensing image and
GIS data is changed information extraction and gets growing concern for.
Traditional is changed the method for information extraction mainly by new, old two phase using remote sensing image and GIS data
Remote sensing image utilizes the change detecting method based on image processing techniques, removes not changed region, generates by that may send out
The image of the pixel composition for changing.Grid and vector processing is carried out to the modified-image, and is laid out point with GIS data
Analysis, obtains final variation testing result.The above method combines remote sensing image change detection techniques and GIS analytical technology,
Result of the detection accuracy dependent on remote sensing image variation detection.Since this method does not make full use of the category attribute of GIS data
Information, it is lower for the correlation utilization rate between remote sensing image and GIS data, it is difficult to ensure that extracting change information is accurate
Property.
Summary of the invention
The deficiency of extracting change information method for tradition based on multi-temporal remote sensing image and GIS data, mesh of the present invention
Be propose a kind of automation and can large-area applications multi-temporal remote sensing image and GIS data change detection method
Process.
The technical scheme adopted by the invention is that：A kind of variation detection side based on multi-temporal remote sensing image and GIS data
Method specifically comprises the following steps：
Step 1, Remote Sensing Image Segmentation and Pixel-level extracting change information, including following sub-step；
Step 1.1, according to T1 period GIS vector data and T2 period raster data, GIS vector data figure spot classification is utilized
Attribute information and geometry information carry out fitting to T2 period raster data, obtain and vector figure spot in T1 period GIS data
Corresponding grid cell, raster cell set；
Step 1.2, to each grid cell, raster cell set of acquisition, divided again using multi-scale division algorithm, obtained
T2 period figure spot, as T2 period Image Segmentation result；
Step 1.2, to T1 period and T2 period raster data, it is changed detection using Pixel-level change detection algorithm,
Obtain Pixel-level variation testing result；
Step 2, the classification of remote-sensing images of object-oriented, including following sub-step；
Step 2.1, analysis is laid out to T2 period Image Segmentation result and Pixel-level variation testing result, counted each
Change pixel in a figure spot and do not change the number of pixel, when variation number of pixels in figure spot and the ratio for not changing number of pixels
When greater than threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,；
Step 2.2, characteristic statistics are carried out to each T2 period of acquisition non-changing graphic, and according to specifically species
Not, it selects different features to combine, obtains suitable training sample；
Step 2.3, the SVM Decision-Tree Classifier Model for constructing object-oriented, determines to SVM by training sample in step 2.2
Plan tree classification model is trained, and is then classified using trained model to all doubtful changing graphics, is obtained each figure
The classification results of spot；
Step 3, the classification of vector figure spot in T1 period GIS data and the T2 period sorted figure of doubtful changing graphic are utilized
Spot classification changes detection after carrying out object oriented classification, obtains two period classifications inconsistent figure spot and its category attribute information,
As final variation testing result.
Further, the SVM Decision-Tree Classifier Model of the object-oriented constructed in step 2.3 be used for water body, plough garden grass,
This 5 class atural object of forest land, artificial works and bare area is more accurately divided, and specific implementation is as follows,
Step 2.3.1 schemes T2 period to the normalization water body index characteristic value of each figure spot according to certain threshold value
Spot is divided into water body figure spot and non-water body figure spot；
Step 2.3.2 selects mean value, the variance in spectral signature, utilizes to water body figure spot obtained in step 2.3.1
Svm classifier model carries out reclassification, obtains more accurate water body figure spot and non-water body figure spot；
Step 2.3.3, to the normalized differential vegetation index characteristic value of non-water body figure spot obtained in step 2.3.1 and 2.3.2,
According to certain threshold value, it is divided into vegetation figure spot and non-vegetation figure spot；
Step 2.3.4 selects mean value, the variance in spectral signature, utilizes to vegetation figure spot obtained in step 2.3.3
Svm classifier model is classified, and obtains ploughing garden sketch spot and forest land figure spot；
Step 2.3.5 selects mean value, the variance in spectral signature, benefit to non-vegetation figure spot obtained in step 2.3.3
Classified with svm classifier model, obtains artificial works figure spot and bare area figure spot.
Further, Pixel-level change detection algorithm is k Change in Mean detection algorithm in step 1.2.
Further, it is carried out in step 2.3.1 using normalization water body index of the k means clustering algorithm to T2 period figure spot
Clustering obtains the threshold value of water body figure spot Yu non-water body figure spot；Using k means clustering algorithm to non-water body in step 2.3.3
The normalized differential vegetation index of figure spot carries out clustering, obtains the threshold value of vegetation figure spot and non-vegetation figure spot.
The present invention also provides a kind of object-oriented change informations based on multi-temporal remote sensing image and GIS data to detect automatically
System comprises the following modules：
Image Segmentation and extracting change information module, for the extraction of Remote Sensing Image Segmentation and Pixel-level change information, packet
Include following submodule；
Grid and vector data conflation submodule, for utilizing according to T1 period GIS vector data and T2 period raster data
GIS vector data figure spot category attribute information and geometry information carry out fitting to T2 period raster data, when obtaining with T1
The corresponding grid cell, raster cell set of vector figure spot in phase GIS data, i.e. T2 period figure spot；
Fitting data divide submodule again, for each T2 period figure spot to acquisition, utilize multi-scale division algorithm
Divided again, obtains the subgraph spot of each figure spot, as T2 period Image Segmentation result；
Pixel-level changes detection sub-module, for being changed using Pixel-level and being detected to T1 period and T2 period raster data
Algorithm is changed detection, obtains Pixel-level variation testing result；
Classification of remote-sensing images module, for the classification of remote-sensing images of object-oriented, including following submodule；
Changing graphic decision submodule, for being folded to T2 period Image Segmentation result and Pixel-level variation testing result
Set analysis, count variation pixel in each figure spot and do not change the number of pixel, when in figure spot variation number of pixels with it is unchanged
When changing the ratio of number of pixels greater than threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,；
Samples selection submodule, for each T2 period of acquisition non-changing graphic carry out characteristic statistics, and according to
Specifically species are other, and different features is selected to combine, and obtain suitable training sample；
SVM decision tree classification submodule passes through samples selection for constructing the SVM Decision-Tree Classifier Model of object-oriented
Training sample is trained SVM Decision-Tree Classifier Model in submodule, then using trained model to all doubtful changes
Change figure spot to classify, obtains the classification results of each figure spot；
Change detection module, for the classification and T2 period doubtful variation diagram using vector figure spot in T1 period GIS data
The sorted figure spot classification of spot changes detection after carrying out object oriented classification, obtain the inconsistent figure spot of two period classifications and its
Category attribute information, as final variation testing result.
Further, the SVM Decision-Tree Classifier Model of the object-oriented constructed in SVM decision tree classification submodule for pair
Water body is ploughed garden grass, forest land, artificial works and bare area this 5 class atural object and is more accurately divided, and specifically includes such as lower unit,
Water body threshold cell, for the normalization water body index characteristic value to each figure spot according to certain threshold value,
T2 period figure spot is divided into water body figure spot and non-water body figure spot；
Water body svm classifier unit, for selecting equal in spectral signature to water body figure spot obtained in water body threshold cell
Value, variance carry out reclassification using svm classifier model, obtain more accurate water body figure spot and non-water body figure spot；
Vegetation threshold cell, for non-water body figure spot obtained in water body threshold cell and water body svm classifier unit
Normalized differential vegetation index characteristic value is divided into vegetation figure spot and non-vegetation figure spot according to certain threshold value；
Vegetation svm classifier unit, for selecting equal in spectral signature to vegetation figure spot obtained in vegetation threshold cell
Value, variance, are classified using svm classifier model, obtain ploughing garden sketch spot and forest land figure spot；
Non- vegetation svm classifier unit, for selecting in spectral signature to non-vegetation figure spot obtained in vegetation threshold cell
Mean value, variance, classified using svm classifier model, obtain artificial works figure spot and bare area figure spot.
Further, it is k Change in Mean detection algorithm that fitting data divide Pixel-level change detection algorithm in submodule again.
Further, utilize k means clustering algorithm to the normalization water body index of T2 period figure spot in water body threshold cell
Clustering is carried out, the threshold value of water body figure spot Yu non-water body figure spot is obtained；K means clustering algorithm pair is utilized in vegetation threshold cell
The normalized differential vegetation index of non-water body figure spot carries out clustering, obtains the threshold value of vegetation figure spot and non-vegetation figure spot.
The advantages of the present invention：
The present invention proposes what a kind of object-oriented change information based on multi-temporal remote sensing image and GIS data detected automatically
Method, including Remote Sensing Image Segmentation, Pixel-level variation detection, feature extraction and characteristic statistics, automated sample selection, disaggregated model
The contents such as variation detection after building and classification.Its main technical flows includes that Remote Sensing Image Segmentation is mentioned with Pixel-level change information
It takes, variation detection three phases after the classification of remote-sensing images and object oriented classification of object-oriented.At the data of the three phases
Reason, ensure that the precision and accuracy of extracting change information in the present invention.
The present invention is a kind of side that the object-oriented change information based on multi-temporal remote sensing image and GIS data detects automatically
Method, most important feature are available with computer system and handle multi-temporal remote sensing image and GIS data, obtain automatically
Take change information.With it is existing be changed the method for information extraction using remote sensing image and GIS data compared with, the present invention can
Realize automation and can large-area applications change detection method flow, can sufficiently excavate remote sensing image and GIS number
According to correlation improve the accuracy of extracting change information using attribute informations such as the classifications of GIS data.
Detailed description of the invention
Fig. 1 is the flow chart of the embodiment of the present invention.
Fig. 2 is object oriented classification model in the embodiment of the present invention.
Specific embodiment
Below in conjunction with attached drawing and the specific algorithm specific embodiment that the present invention will be described in detail, as shown in Figure 1, one kind is based on
The object-oriented change information automatic testing method of multi-temporal remote sensing image and GIS data, its step are as follows：
1. Remote Sensing Image Segmentation and Pixel-level extracting change information
The step mainly utilizes GIS vector data and multi-scale division algorithm, is split to T2 period raster data,
Figure spot information is obtained, meanwhile, detection is changed to T1 period and T2 period raster data using Pixel-level change detection algorithm,
Obtain Pixel-level variation testing result.
1.1. Remote Sensing Image Segmentation
1.1.1. grid and vector data conflation
According to T1 period GIS vector data and T2 period raster data, GIS vector data figure spot category attribute information is utilized
Fitting is carried out to T2 period raster data with geometry information, obtains grid corresponding with vector figure spot in T1 period GIS data
Lattice pixel set.
1.1.2. dividing again based on fitting data
It to each grid cell, raster cell set of acquisition, is split using multi-scale division algorithm, obtains T2 period figure spot,
As T2 period Image Segmentation result.
1.2. Pixel-level variation detection
To T1 period and T2 period raster data, it is changed detection using k Change in Mean detection algorithm, obtains Pixel-level
Change testing result.
2. the classification of remote-sensing images of object-oriented
The step is mainly changed figure spot decision to T2 period Image Segmentation result and Pixel-level variation testing result,
Obtain T2 period doubtful changing graphic and T2 period non-changing graphic.Then feature extraction algorithm and sample selection algorithm pair are utilized
T2 period, non-changing graphic carried out characteristic statistics and selection, obtained suitable training sample and then pass through to train to obtain final face
To object SVM Decision-Tree Classifier Model.Finally the segmentation result of T2 period doubtful changing graphic is obtained using the model.The step
Major advantage be that unchanged figure spot can be gone out to preliminary screening by changing graphic decision, reduce the mistake in post-processing
Difference, while characteristic statistics and samples selection, available more accurate training sample are carried out to non-changing graphic.Except this it
Outside, the object-oriented SVM Decision-Tree Classifier Model that the embodiment of the present invention proposes takes full advantage of different characteristic value and different samples
Combined advantage, to obtain more accurate disaggregated model.
Category division is five classes in the present embodiment, i.e. water body, cultivated garden grass, forest land, artificial works and bare area.
2.1. changing graphic decision
Changing graphic decision screens T2 period Image Segmentation result, obtains T2 period doubtful changing graphic and not
Changing graphic.The main method of changing graphic decision is to change testing result to T2 period Image Segmentation result and Pixel-level to carry out
Overlap Analysis counts the interior variation pixel of each figure spot and does not change the number of pixel.When variation number of pixels in figure spot and not
When changing the ratio of number of pixels greater than threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,.This implementation
The value of threshold value m is 1 in example.
2.2. automated sample selects
2.2.1. characteristic statistics and selection
Characteristic statistics, including spectral signature (such as mean value, variance are carried out to each T2 period of acquisition non-changing graphic
Deng), textural characteristics (such as contrast, entropy), shape feature (such as length-width ratio).It is other according to specifically species, it selects different
Feature combination.The present embodiment is referred to counting mean value in spectral signature and variance and normalization water body index, normalization vegetation
Number is illustrated.
2.2.2. training sample selection
According to Principle of Statistics, for a large amount of sample, the characteristic value of same atural object answers Normal Distribution.With
Certain confidence level screens the mean value and variance of figure spot, obtains the figure spot that can represent each type of ground objects, as instructs
Practice sample.
2.3. object-oriented SVM Decision-Tree Classifier Model
The step mainly designs and training object-oriented SVM Decision-Tree Classifier Model.The model mainly includes being based on returning
One changes water body index threshold method, the object-oriented svm classifier model based on water body, is based on normalized differential vegetation index threshold method, base
Object-oriented svm classifier model in vegetation and five parts of object-oriented svm classifier model based on non-vegetation.
2.3.1. based on normalization water body index threshold method
It is normalization water body index characteristic value to each figure spot according to certain based on normalization water body index threshold method
Threshold value, T2 period figure spot is divided into water body figure spot and non-water body figure spot.When utilizing k means clustering algorithm to T2 in the present embodiment
The normalization water body index of phase figure spot carries out clustering, obtains the threshold value of water body figure spot Yu non-water body figure spot.
2.3.2. the object-oriented svm classifier model based on water body
Certain error is commonly present in step 2.3.1 by the classification results that threshold method obtains, therefore is utilized in this step
Object-oriented svm classifier model carries out reclassification, has obtained higher nicety of grading.Object-oriented svm classifier based on water body
Model is to select mean value, the variance in spectral signature to water body figure spot obtained in step 2.3.1, using svm classifier model into
Row reclassification obtains more accurate water body figure spot and non-water body figure spot.
2.3.3. being based on normalized differential vegetation index threshold method
Based on the normalizing that normalized differential vegetation index threshold method is to non-water body figure spot obtained in step 2.3.1 and 2.3.2
Change vegetation index characteristic value, according to certain threshold value, the method that is divided into vegetation figure spot and non-vegetation figure spot.It is sharp in the present embodiment
Clustering is carried out with normalized differential vegetation index of the k means clustering algorithm to non-water body figure spot, obtains vegetation figure spot and non-vegetation
The threshold value of figure spot.
2.3.4. the object-oriented svm classifier model based on vegetation
It mainly include ploughing garden sketch spot and forest land figure spot in vegetation figure spot.Object-oriented svm classifier model based on vegetation
It is that is selected by mean value, the variance in spectral signature, is divided using svm classifier model for vegetation figure spot obtained in step 2.3.3
Class obtains ploughing garden sketch spot and forest land figure spot.
2.3.5. the object-oriented svm classifier model based on non-vegetation
It mainly include artificial works figure spot and bare area figure spot in non-vegetation figure spot.Object-oriented SVM based on non-vegetation
Disaggregated model is to select mean value, the variance in spectral signature to non-vegetation figure spot obtained in step 2.3.3, utilize svm classifier
Model is classified, and artificial works figure spot and bare area figure spot are obtained.
2.3.6. disaggregated model training
By step 2.2 it is available suitably include multiple features training sample, using the training sample to towards
Object SVM Decision-Tree Classifier Model is trained the disaggregated model that can be obtained suitable for the data.
2.4. object oriented classification
Using the object-oriented SVM Decision-Tree Classifier Model of acquisition, classify to all figure spots to be sorted, obtains each
The classification results of figure spot.
3. changing detection after object oriented classification
Utilize the classification and the sorted figure spot class of T2 period doubtful changing graphic of vector figure spot in T1 period GIS data
Not, change detection after carrying out object oriented classification, obtain two period classifications inconsistent figure spot and its category attribute information, as
Final variation testing result.
The embodiment of the present invention also provide a kind of object-oriented change information based on multi-temporal remote sensing image and GIS data from
Dynamic detection system, comprises the following modules：
Image Segmentation and extracting change information module, for the extraction of Remote Sensing Image Segmentation and Pixel-level change information, packet
Include following submodule；
Grid and vector data conflation submodule, for utilizing according to T1 period GIS vector data and T2 period raster data
GIS vector data figure spot category attribute information and geometry information carry out fitting to T2 period raster data, when obtaining with T1
The corresponding grid cell, raster cell set of vector figure spot in phase GIS data, i.e. T2 period figure spot；
Fitting data divide submodule again, for each T2 period figure spot to acquisition, utilize multi-scale division algorithm
Divided again, obtains the subgraph spot of each figure spot, as T2 period Image Segmentation result；
Pixel-level changes detection sub-module, for being changed using Pixel-level and being detected to T1 period and T2 period raster data
Algorithm is changed detection, obtains Pixel-level variation testing result；
Classification of remote-sensing images module, for the classification of remote-sensing images of object-oriented, including following submodule；
Changing graphic decision submodule, for being folded to T2 period Image Segmentation result and Pixel-level variation testing result
Set analysis, count variation pixel in each figure spot and do not change the number of pixel, when in figure spot variation number of pixels with it is unchanged
When changing the ratio of number of pixels greater than threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,；
Samples selection submodule, for each T2 period of acquisition non-changing graphic carry out characteristic statistics, and according to
Specifically species are other, and different features is selected to combine, and obtain suitable training sample；
SVM decision tree classification submodule passes through samples selection for constructing the SVM Decision-Tree Classifier Model of object-oriented
Training sample is trained SVM Decision-Tree Classifier Model in submodule, then using trained model to all doubtful changes
Change figure spot to classify, obtains the classification results of each figure spot；
Change detection module, for the classification and T2 period doubtful variation diagram using vector figure spot in T1 period GIS data
The sorted figure spot classification of spot changes detection after carrying out object oriented classification, obtain the inconsistent figure spot of two period classifications and its
Category attribute information, as final variation testing result.
Wherein, the SVM Decision-Tree Classifier Model of the object-oriented constructed in SVM decision tree classification submodule is used for water
Body is ploughed garden grass, forest land, artificial works and bare area this 5 class atural object and is more accurately divided, and specifically includes such as lower unit,
Water body threshold cell, for the normalization water body index characteristic value to each figure spot according to certain threshold value,
T2 period figure spot is divided into water body figure spot and non-water body figure spot；
Water body svm classifier unit, for selecting equal in spectral signature to water body figure spot obtained in water body threshold cell
Value, variance carry out reclassification using svm classifier model, obtain more accurate water body figure spot and non-water body figure spot；
Vegetation threshold cell, for non-water body figure spot obtained in water body threshold cell and water body svm classifier unit
Normalized differential vegetation index characteristic value is divided into vegetation figure spot and non-vegetation figure spot according to certain threshold value；
Vegetation svm classifier unit, for selecting equal in spectral signature to vegetation figure spot obtained in vegetation threshold cell
Value, variance, are classified using svm classifier model, obtain ploughing garden sketch spot and forest land figure spot；
Non- vegetation svm classifier unit, for selecting in spectral signature to non-vegetation figure spot obtained in vegetation threshold cell
Mean value, variance, classified using svm classifier model, obtain artificial works figure spot and bare area figure spot.
Wherein, it is k Change in Mean detection algorithm that fitting data divide Pixel-level change detection algorithm in submodule again.
Wherein, it is carried out in water body threshold cell using normalization water body index of the k means clustering algorithm to T2 period figure spot
Clustering obtains the threshold value of water body figure spot Yu non-water body figure spot；Using k means clustering algorithm to non-aqueous in vegetation threshold cell
The normalized differential vegetation index of body figure spot carries out clustering, obtains the threshold value of vegetation figure spot and non-vegetation figure spot.
Specific embodiment described herein is only an example for the spirit of the invention.The neck of technology belonging to the present invention
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (8)
1. a kind of object-oriented change information automatic testing method based on multi-temporal remote sensing image and GIS data, feature exist
In this approach includes the following steps：
Step 1, Remote Sensing Image Segmentation and Pixel-level extracting change information, including following sub-step；
Step 1.1, according to T1 period GIS vector data and T2 period raster data, GIS vector data figure spot category attribute is utilized
Information and geometry information carry out fitting to T2 period raster data, obtain corresponding with vector figure spot in T1 period GIS data
Grid cell, raster cell set；
Step 1.2, to each grid cell, raster cell set of acquisition, divided again using multi-scale division algorithm, when obtaining T2
Phase figure spot, as T2 period Image Segmentation result；
Step 1.2, to T1 period and T2 period raster data, it is changed detection using Pixel-level change detection algorithm, is obtained
Pixel-level changes testing result；
Step 2, the classification of remote-sensing images of object-oriented, including following sub-step；
Step 2.1, analysis is laid out to T2 period Image Segmentation result and Pixel-level variation testing result, counts each figure
Change pixel in spot and do not change the number of pixel, is greater than when changing number of pixels in figure spot with the ratio for not changing number of pixels
When threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,；
Step 2.2, characteristic statistics are carried out to each T2 period of acquisition non-changing graphic, and other according to specifically species, choosing
Different feature combinations is selected, suitable training sample is obtained；
Step 2.3, the SVM Decision-Tree Classifier Model for constructing object-oriented, by training sample in step 2.2 to SVM decision tree
Disaggregated model is trained, and is then classified using trained model to all doubtful changing graphics, is obtained each figure spot
Classification results；
Step 3, the classification of vector figure spot in T1 period GIS data and the sorted figure spot class of T2 period doubtful changing graphic are utilized
Not, change detection after carrying out object oriented classification, obtain two period classifications inconsistent figure spot and its category attribute information, as
Final variation testing result.
2. the object-oriented change information as described in claim 1 based on multi-temporal remote sensing image and GIS data detects automatically
Method, it is characterised in that：The SVM Decision-Tree Classifier Model of the object-oriented constructed in step 2.3 be used for water body, plough garden grass,
This 5 class atural object of forest land, artificial works and bare area is more accurately divided, and specific implementation is as follows,
Step 2.3.1, to the normalization water body index characteristic value of each figure spot according to certain threshold value, T2 period figure spot point
For water body figure spot and non-water body figure spot；
Step 2.3.2 selects mean value, the variance in spectral signature, utilizes SVM points to water body figure spot obtained in step 2.3.1
Class model carries out reclassification, obtains more accurate water body figure spot and non-water body figure spot；
Step 2.3.3, to the normalized differential vegetation index characteristic value of non-water body figure spot obtained in step 2.3.1 and 2.3.2, according to
Certain threshold value is divided into vegetation figure spot and non-vegetation figure spot；
Step 2.3.4 selects mean value, the variance in spectral signature, utilizes SVM points to vegetation figure spot obtained in step 2.3.3
Class model is classified, and obtains ploughing garden sketch spot and forest land figure spot；
Step 2.3.5 selects mean value, the variance in spectral signature, utilizes SVM to non-vegetation figure spot obtained in step 2.3.3
Disaggregated model is classified, and artificial works figure spot and bare area figure spot are obtained.
3. the object-oriented change information as described in claim 1 based on multi-temporal remote sensing image and GIS data detects automatically
Method, it is characterised in that：Pixel-level change detection algorithm is k Change in Mean detection algorithm in step 1.2.
4. the object-oriented change information as claimed in claim 2 based on multi-temporal remote sensing image and GIS data detects automatically
Method, it is characterised in that：It is carried out in step 2.3.1 using normalization water body index of the k means clustering algorithm to T2 period figure spot
Clustering obtains the threshold value of water body figure spot Yu non-water body figure spot；Using k means clustering algorithm to non-water body in step 2.3.3
The normalized differential vegetation index of figure spot carries out clustering, obtains the threshold value of vegetation figure spot and non-vegetation figure spot.
5. a kind of object-oriented change information automatic checkout system based on multi-temporal remote sensing image and GIS data, feature exist
In comprising the following modules：
Image Segmentation and extracting change information module, for the extraction of Remote Sensing Image Segmentation and Pixel-level change information, including such as
Lower submodule；
Grid and vector data conflation submodule, for utilizing GIS according to T1 period GIS vector data and T2 period raster data
Vector data figure spot category attribute information and geometry information carry out fitting to T2 period raster data, obtain and T1 period
The corresponding grid cell, raster cell set of vector figure spot in GIS data, i.e. T2 period figure spot；
Fitting data divide submodule again, for each T2 period figure spot to acquisition, are carried out using multi-scale division algorithm
Divide again, obtains the subgraph spot of each figure spot, as T2 period Image Segmentation result；
Pixel-level changes detection sub-module, for utilizing Pixel-level change detection algorithm to T1 period and T2 period raster data
It is changed detection, obtains Pixel-level variation testing result；
Classification of remote-sensing images module, for the classification of remote-sensing images of object-oriented, including following submodule；
Changing graphic decision submodule, for being laid out point to T2 period Image Segmentation result and Pixel-level variation testing result
Analysis counts and changes pixel in each figure spot and do not change the number of pixel, do not change when variation number of pixels in figure spot and picture
When the ratio of plain number is greater than threshold value m, then otherwise it is non-changing graphic that the figure spot, which is doubtful changing graphic,；
Samples selection submodule, for carrying out characteristic statistics to each T2 period of acquisition non-changing graphic, and according to specific
Atural object classification selects different features to combine, and obtains suitable training sample；
SVM decision tree classification submodule passes through samples selection submodule for constructing the SVM Decision-Tree Classifier Model of object-oriented
Training sample is trained SVM Decision-Tree Classifier Model in block, then using trained model to all doubtful variation diagrams
Spot is classified, and the classification results of each figure spot are obtained；
Change detection module, for the classification and T2 period doubtful changing graphic point using vector figure spot in T1 period GIS data
Figure spot classification after class changes detection after carrying out object oriented classification, obtains two period classifications inconsistent figure spot and its classification
Attribute information, as final variation testing result.
6. the object-oriented change information as claimed in claim 5 based on multi-temporal remote sensing image and GIS data detects automatically
System, it is characterised in that：The SVM Decision-Tree Classifier Model of the object-oriented constructed in SVM decision tree classification submodule for pair
Water body is ploughed garden grass, forest land, artificial works and bare area this 5 class atural object and is more accurately divided, and specifically includes such as lower unit,
Water body threshold cell, for the normalization water body index characteristic value to each figure spot according to certain threshold value, when T2
Phase figure spot is divided into water body figure spot and non-water body figure spot；
Water body svm classifier unit, for water body figure spot obtained in water body threshold cell, select mean value in spectral signature,
Variance carries out reclassification using svm classifier model, obtains more accurate water body figure spot and non-water body figure spot；
Vegetation threshold cell, for the normalizing to non-water body figure spot obtained in water body threshold cell and water body svm classifier unit
Change vegetation index characteristic value and is divided into vegetation figure spot and non-vegetation figure spot according to certain threshold value；
Vegetation svm classifier unit, for vegetation figure spot obtained in vegetation threshold cell, select mean value in spectral signature,
Variance is classified using svm classifier model, obtains ploughing garden sketch spot and forest land figure spot；
Non- vegetation svm classifier unit, for selecting equal in spectral signature to non-vegetation figure spot obtained in vegetation threshold cell
Value, variance, are classified using svm classifier model, obtain artificial works figure spot and bare area figure spot.
7. the object-oriented change information as claimed in claim 5 based on multi-temporal remote sensing image and GIS data detects automatically
System, it is characterised in that：It is k Change in Mean detection algorithm that fitting data divide Pixel-level change detection algorithm in submodule again.
8. the object-oriented change information as claimed in claim 6 based on multi-temporal remote sensing image and GIS data detects automatically
System, it is characterised in that：In water body threshold cell using k means clustering algorithm to the normalization water body index of T2 period figure spot into
Row clustering obtains the threshold value of water body figure spot Yu non-water body figure spot；Using k means clustering algorithm to non-in vegetation threshold cell
The normalized differential vegetation index of water body figure spot carries out clustering, obtains the threshold value of vegetation figure spot and non-vegetation figure spot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810538861.3A CN108846832B (en) | 2018-05-30 | 2018-05-30 | Multi-temporal remote sensing image and GIS data based change detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810538861.3A CN108846832B (en) | 2018-05-30 | 2018-05-30 | Multi-temporal remote sensing image and GIS data based change detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108846832A true CN108846832A (en) | 2018-11-20 |
| CN108846832B CN108846832B (en) | 2021-06-15 |
Family
ID=64210020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810538861.3A Active CN108846832B (en) | 2018-05-30 | 2018-05-30 | Multi-temporal remote sensing image and GIS data based change detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108846832B (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109655583A (en) * | 2018-12-14 | 2019-04-19 | 天津珞雍空间信息研究院有限公司 | A kind of atmospheric environment ground monitoring website based on satellite remote sensing is deployed to ensure effective monitoring and control of illegal activities network-building method |
| CN109657610A (en) * | 2018-12-18 | 2019-04-19 | 北京航天泰坦科技股份有限公司 | A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images |
| CN109816749A (en) * | 2019-02-14 | 2019-05-28 | 中国测绘科学研究院 | Figure spot map symbol fill method, device, computer equipment and storage medium |
| CN109829425A (en) * | 2019-01-31 | 2019-05-31 | 沈阳农业大学 | A kind of small scale terrain classification method and system of Farmland Landscape |
| CN109901240A (en) * | 2019-04-03 | 2019-06-18 | 中国矿业大学 | A method of detecting residual coal spontaneous combustion region of appearing |
| CN109934154A (en) * | 2019-03-08 | 2019-06-25 | 北京科技大学 | A kind of remote sensing image variation detection method and detection device |
| CN110263627A (en) * | 2019-05-09 | 2019-09-20 | 生态环境部卫星环境应用中心 | A kind of mankind's activity variation clue monitoring method towards protecting field supervision |
| CN110348483A (en) * | 2019-06-06 | 2019-10-18 | 重庆工商大学融智学院 | Change detection and classification method based on the spatial information of spatial information and statistical learning |
| CN110473205A (en) * | 2019-07-10 | 2019-11-19 | 北京吉威数源信息技术有限公司 | Remote sensing image information extracting method and system based on arrow bar phantom |
| CN110738679A (en) * | 2019-10-21 | 2020-01-31 | 山东省国土测绘院 | Geographical provincial monitoring method and system based on automatic extraction of remote sensing image change |
| CN110826454A (en) * | 2019-10-30 | 2020-02-21 | 北京科技大学 | Remote sensing image change detection method and device |
| CN111739067A (en) * | 2020-05-28 | 2020-10-02 | 中国国土勘测规划院 | Remote sensing image change detection method and device |
| CN111897907A (en) * | 2020-08-12 | 2020-11-06 | 国家基础地理信息中心 | Method and device for filtering pseudo-change information of surface coverage data and storage medium |
| CN112101159A (en) * | 2020-09-04 | 2020-12-18 | 国家林业和草原局中南调查规划设计院 | Multi-temporal forest remote sensing image change monitoring method |
| CN112215135A (en) * | 2020-10-10 | 2021-01-12 | 东华理工大学 | Mining area mining and treatment effect monitoring method and device |
| CN112580510A (en) * | 2020-12-18 | 2021-03-30 | 广州市城市规划设计所 | Permeable ground area rate estimation method, permeable ground area rate estimation device, permeable ground area rate estimation equipment and storage medium |
| CN112801170A (en) * | 2021-01-25 | 2021-05-14 | 河南省国土资源调查规划院(河南省国土资源厅不动产登记中心) | Multi-temporal change detection and ground feature identification and classification method based on remote sensing image |
| CN113298755A (en) * | 2021-04-13 | 2021-08-24 | 生态环境部卫星环境应用中心 | Method and device for rapidly detecting ecological environment change patch based on time sequence image |
| CN113484245A (en) * | 2021-07-05 | 2021-10-08 | 重庆市规划和自然资源调查监测院 | Remote sensing rapid monitoring method and system for paddy field planting pattern in hilly and mountainous areas and computer readable storage medium |
| CN113822900A (en) * | 2021-07-09 | 2021-12-21 | 武汉大学 | Vector constraint object-oriented new image sample automatic selection method and system |
| CN113989657A (en) * | 2021-10-11 | 2022-01-28 | 中国测绘科学研究院 | Method and device for detecting farmland range change based on invariant information sample screening |
| CN113989652A (en) * | 2021-12-27 | 2022-01-28 | 中国测绘科学研究院 | Method and system for detecting farmland change under layered multiple judgment rules |
| CN114119575A (en) * | 2021-11-30 | 2022-03-01 | 二十一世纪空间技术应用股份有限公司 | Spatial information change detection method and system |
| CN114882084A (en) * | 2022-05-07 | 2022-08-09 | 安徽农业大学 | Land use change pattern spot automatic identification method based on artificial intelligence |
| CN117612020A (en) * | 2024-01-24 | 2024-02-27 | 西安宇速防务集团有限公司 | SGAN-based detection method for resisting neural network remote sensing image element change |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050041277A (en) * | 2003-10-30 | 2005-05-04 | 이준환 | Automatic analysing method for high precision satellite image |
| CN102496034A (en) * | 2011-11-29 | 2012-06-13 | 南京师范大学 | High-spatial resolution remote-sensing image bag-of-word classification method based on linear words |
| CN102622345A (en) * | 2011-01-26 | 2012-08-01 | 杨存建 | High-precision land-utilization remote sensing updating technology with synergistic multisource spatio-temporal data |
| 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 |
| CN103729653A (en) * | 2014-01-21 | 2014-04-16 | 武汉大学 | Supervision change detection method of high-resolution remote sensing images |
| CN103971115A (en) * | 2014-05-09 | 2014-08-06 | 中国科学院遥感与数字地球研究所 | Automatic extraction method for newly-increased construction land image spots in high-resolution remote sensing images based on NDVI and PanTex index |
| CN106803097A (en) * | 2016-12-28 | 2017-06-06 | 中国科学院深圳先进技术研究院 | A kind of utilization remote sensing image extracts method and device, the user equipment of vegetation sunshade net classification |
-
2018
- 2018-05-30 CN CN201810538861.3A patent/CN108846832B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20050041277A (en) * | 2003-10-30 | 2005-05-04 | 이준환 | Automatic analysing method for high precision satellite image |
| CN102622345A (en) * | 2011-01-26 | 2012-08-01 | 杨存建 | High-precision land-utilization remote sensing updating technology with synergistic multisource spatio-temporal data |
| CN102496034A (en) * | 2011-11-29 | 2012-06-13 | 南京师范大学 | High-spatial resolution remote-sensing image bag-of-word classification method based on linear words |
| 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 |
| CN103729653A (en) * | 2014-01-21 | 2014-04-16 | 武汉大学 | Supervision change detection method of high-resolution remote sensing images |
| CN103971115A (en) * | 2014-05-09 | 2014-08-06 | 中国科学院遥感与数字地球研究所 | Automatic extraction method for newly-increased construction land image spots in high-resolution remote sensing images based on NDVI and PanTex index |
| CN106803097A (en) * | 2016-12-28 | 2017-06-06 | 中国科学院深圳先进技术研究院 | A kind of utilization remote sensing image extracts method and device, the user equipment of vegetation sunshade net classification |
Non-Patent Citations (5)
| Title |
|---|
| LINRUI SONG ET AL: "LAND COVER CHANGE AND EFFECT ON AGRICULTURAL IRRIGATION BASED ON RS AND GIS FOR 20 YEARS IN YANQING COUNTY, BEIJING", 《2011 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM 》 * |
| ZHENXUAN LI ET AL: "Change Detection Based on Gabor Wavelet Features for Very High Resolution Remote Sensing Images", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| 崔伟宏 等: "基于特征的时空数据模型研究及在土地利用变化动态监测中的应用", 《测绘学报》 * |
| 张晓东 等: "遥感影像与GIS分析相结合的变化检测方法", 《武汉大学学报 信息科学版》 * |
| 赵珍珍: "高分辨率遥感影像与矢量数据结合的变化检测方法研究", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109655583A (en) * | 2018-12-14 | 2019-04-19 | 天津珞雍空间信息研究院有限公司 | A kind of atmospheric environment ground monitoring website based on satellite remote sensing is deployed to ensure effective monitoring and control of illegal activities network-building method |
| CN109657610A (en) * | 2018-12-18 | 2019-04-19 | 北京航天泰坦科技股份有限公司 | A kind of land use change survey detection method of high-resolution multi-source Remote Sensing Images |
| CN109829425A (en) * | 2019-01-31 | 2019-05-31 | 沈阳农业大学 | A kind of small scale terrain classification method and system of Farmland Landscape |
| CN109816749A (en) * | 2019-02-14 | 2019-05-28 | 中国测绘科学研究院 | Figure spot map symbol fill method, device, computer equipment and storage medium |
| CN109934154B (en) * | 2019-03-08 | 2021-06-01 | 北京科技大学 | Remote sensing image change detection method and detection device |
| CN109934154A (en) * | 2019-03-08 | 2019-06-25 | 北京科技大学 | A kind of remote sensing image variation detection method and detection device |
| CN109901240A (en) * | 2019-04-03 | 2019-06-18 | 中国矿业大学 | A method of detecting residual coal spontaneous combustion region of appearing |
| CN110263627A (en) * | 2019-05-09 | 2019-09-20 | 生态环境部卫星环境应用中心 | A kind of mankind's activity variation clue monitoring method towards protecting field supervision |
| CN110348483A (en) * | 2019-06-06 | 2019-10-18 | 重庆工商大学融智学院 | Change detection and classification method based on the spatial information of spatial information and statistical learning |
| CN110473205A (en) * | 2019-07-10 | 2019-11-19 | 北京吉威数源信息技术有限公司 | Remote sensing image information extracting method and system based on arrow bar phantom |
| CN110738679A (en) * | 2019-10-21 | 2020-01-31 | 山东省国土测绘院 | Geographical provincial monitoring method and system based on automatic extraction of remote sensing image change |
| CN110826454B (en) * | 2019-10-30 | 2022-06-28 | 北京科技大学 | Remote sensing image change detection method and device |
| CN110826454A (en) * | 2019-10-30 | 2020-02-21 | 北京科技大学 | Remote sensing image change detection method and device |
| CN111739067A (en) * | 2020-05-28 | 2020-10-02 | 中国国土勘测规划院 | Remote sensing image change detection method and device |
| CN111897907B (en) * | 2020-08-12 | 2024-03-01 | 国家基础地理信息中心 | Method, device and storage medium for filtering pseudo-change information of earth surface coverage data |
| CN111897907A (en) * | 2020-08-12 | 2020-11-06 | 国家基础地理信息中心 | Method and device for filtering pseudo-change information of surface coverage data and storage medium |
| CN112101159A (en) * | 2020-09-04 | 2020-12-18 | 国家林业和草原局中南调查规划设计院 | Multi-temporal forest remote sensing image change monitoring method |
| CN112215135A (en) * | 2020-10-10 | 2021-01-12 | 东华理工大学 | Mining area mining and treatment effect monitoring method and device |
| CN112215135B (en) * | 2020-10-10 | 2022-07-08 | 东华理工大学 | Mining area mining and treatment effect monitoring method and device |
| CN112580510A (en) * | 2020-12-18 | 2021-03-30 | 广州市城市规划设计所 | Permeable ground area rate estimation method, permeable ground area rate estimation device, permeable ground area rate estimation equipment and storage medium |
| CN112580510B (en) * | 2020-12-18 | 2024-05-21 | 广州市城市规划设计有限公司 | Method, device, equipment and storage medium for estimating permeable ground area rate |
| CN112801170A (en) * | 2021-01-25 | 2021-05-14 | 河南省国土资源调查规划院(河南省国土资源厅不动产登记中心) | Multi-temporal change detection and ground feature identification and classification method based on remote sensing image |
| CN113298755A (en) * | 2021-04-13 | 2021-08-24 | 生态环境部卫星环境应用中心 | Method and device for rapidly detecting ecological environment change patch based on time sequence image |
| CN113484245B (en) * | 2021-07-05 | 2022-11-22 | 重庆市规划和自然资源调查监测院 | Remote sensing rapid monitoring method and system for paddy field planting pattern in hilly and mountainous areas and computer readable storage medium |
| CN113484245A (en) * | 2021-07-05 | 2021-10-08 | 重庆市规划和自然资源调查监测院 | Remote sensing rapid monitoring method and system for paddy field planting pattern in hilly and mountainous areas and computer readable storage medium |
| CN113822900B (en) * | 2021-07-09 | 2023-09-22 | 武汉大学 | Method and system for automatically selecting new image sample based on vector constraint object-oriented |
| CN113822900A (en) * | 2021-07-09 | 2021-12-21 | 武汉大学 | Vector constraint object-oriented new image sample automatic selection method and system |
| CN113989657A (en) * | 2021-10-11 | 2022-01-28 | 中国测绘科学研究院 | Method and device for detecting farmland range change based on invariant information sample screening |
| CN113989657B (en) * | 2021-10-11 | 2024-05-10 | 中国测绘科学研究院 | Cultivated land range change detection method and device based on invariable information sample screening |
| CN114119575A (en) * | 2021-11-30 | 2022-03-01 | 二十一世纪空间技术应用股份有限公司 | Spatial information change detection method and system |
| CN113989652A (en) * | 2021-12-27 | 2022-01-28 | 中国测绘科学研究院 | Method and system for detecting farmland change under layered multiple judgment rules |
| CN114882084A (en) * | 2022-05-07 | 2022-08-09 | 安徽农业大学 | Land use change pattern spot automatic identification method based on artificial intelligence |
| CN114882084B (en) * | 2022-05-07 | 2024-04-05 | 安徽农业大学 | Land utilization change pattern automatic identification method based on artificial intelligence |
| CN117612020A (en) * | 2024-01-24 | 2024-02-27 | 西安宇速防务集团有限公司 | SGAN-based detection method for resisting neural network remote sensing image element change |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108846832B (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108846832A (en) | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data | |
| Huang et al. | A new building extraction postprocessing framework for high-spatial-resolution remote-sensing imagery | |
| Mallinis et al. | Object-based classification using Quickbird imagery for delineating forest vegetation polygons in a Mediterranean test site | |
| CN108830870B (en) | Satellite image high-precision farmland boundary extraction method based on multi-scale structure learning | |
| Qian et al. | Comparison of pixel-based and object-oriented classification methods for extracting built-up areas in arid zone | |
| Walker et al. | An object-oriented approach to urban forest mapping in Phoenix | |
| Walde et al. | From land cover-graphs to urban structure types | |
| CN110263717A (en) | It is a kind of incorporate streetscape image land used status determine method | |
| CN107330875A (en) | Based on the forward and reverse heterogeneous water body surrounding enviroment change detecting method of remote sensing images | |
| Cheng et al. | DESTIN: A new method for delineating the boundaries of crop fields by fusing spatial and temporal information from WorldView and Planet satellite imagery | |
| Hecht et al. | Automatic derivation of urban structure types from topographic maps by means of image analysis and machine learning | |
| Peng et al. | Object-based change detection from satellite imagery by segmentation optimization and multi-features fusion | |
| Costa et al. | Combining per-pixel and object-based classifications for mapping land cover over large areas | |
| CN106295498A (en) | Remote sensing image target area detection apparatus and method | |
| Truax | Comparing spectral and object based approaches for classification and transportation feature extraction from high resolution multispectral imagery | |
| Genitha et al. | Classification of satellite images using new fuzzy cluster centroid for unsupervised classification algorithm | |
| Han et al. | Integration of texture and landscape features into object-based classification for delineating Torreya using IKONOS imagery | |
| Fauvel et al. | Detection of hedges in a rural landscape using a local orientation feature: from linear opening to path opening | |
| Karydas et al. | Development of an IKONOS image classification rule-set for multi-scale mapping of Mediterranean rural landscapes | |
| Ji et al. | Discriminating saltcedar (Tamarix ramosissima) from sparsely distributed cottonwood (Populus euphratica) using a summer season satellite image | |
| Shumilo | Automatic anomaly detection methodology for crop classification data using morphological features | |
| Forghani et al. | Extracting terrain categories from multi-source satellite imagery | |
| Kato | Integrating openstreetmap data in object based landcover and landuse classification for disaster recovery | |
| Pluto-Kossakowska | Automatic detection of grey infrastructure based on VHR image | |
| Awuah | Effects of spatial resolution, land-cover heterogeneity and different classification methods on accuracy of land-cover mapping |
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 |