CN116012334A - Construction land unsupervised change detection method based on domain knowledge constraint - Google Patents
Construction land unsupervised change detection method based on domain knowledge constraint Download PDFInfo
- Publication number
- CN116012334A CN116012334A CN202310005807.3A CN202310005807A CN116012334A CN 116012334 A CN116012334 A CN 116012334A CN 202310005807 A CN202310005807 A CN 202310005807A CN 116012334 A CN116012334 A CN 116012334A
- Authority
- CN
- China
- Prior art keywords
- change
- change detection
- construction land
- domain knowledge
- mask
- 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
- 230000008859 change Effects 0.000 title claims abstract description 114
- 238000001514 detection method Methods 0.000 title claims abstract description 73
- 238000010276 construction Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 claims abstract description 23
- 238000012937 correction Methods 0.000 claims abstract description 8
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 230000000877 morphologic effect Effects 0.000 claims abstract description 4
- 238000012847 principal component analysis method Methods 0.000 claims abstract description 4
- 230000005855 radiation Effects 0.000 claims abstract description 4
- 238000012805 post-processing Methods 0.000 claims abstract 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 27
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000002310 reflectometry Methods 0.000 claims description 4
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 230000000452 restraining effect Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 1
- 230000001932 seasonal effect Effects 0.000 abstract 1
- 238000011160 research Methods 0.000 description 6
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000000873 masking effect Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to a construction land unsupervised change detection method of domain knowledge constraint, and belongs to the technical field of remote sensing geochemistry application. The method mainly comprises the following steps: 1) Performing geometric correction and radiation correction pretreatment on two-stage Sentinel-2 remote sensing images (only using blue, green, red and near infrared bands with spatial resolution of 10 meters); 2) Identifying all types of changes by using four non-supervision change detection methods, namely a change vector analysis method, an iterative weighted multivariate change detection method, a multivariate change detection method and a principal component analysis method, and generating an initial change detection result based on ensemble learning; 3) Constraining the real change and all pseudo changes of the non-construction land by using domain knowledge, and only reserving the real change of the construction land; 4) The connectivity and integrity of the construction site-variation map spots is enhanced using mathematical morphological post-processing. The method provided by the invention can be used for rapidly and effectively extracting the change of the large-scale construction land, and has stronger robustness to seasonal change.
Description
Technical Field
The invention relates to a construction land unsupervised change detection method of domain knowledge constraint, and belongs to the technical field of remote sensing geochemistry application.
Background
At present, china enters a key period of high-quality urbanization, and a construction land refers to the land of a building and a structure, and is one of important indexes for measuring the urbanization level. The method for accurately monitoring the construction land change has great significance for comprehensively judging urban trend and supporting sustainable city decision.
The traditional method for investigating the change of construction land on the artificial ground is time-consuming and labor-consuming, and is difficult to meet the current demands. Remote sensing earth observation can realize long-time, large-range and periodic observation of the earth surface, and technology for acquiring construction land change information is changed. The construction land change detection method can be divided into two major categories, namely a direct comparison method and a post-classification comparison method according to the processing process of the image. The direct comparison method calculates the difference of the features of the two-stage images to find change information, and the classified comparison method firstly separately classifies each stage of images to generate category pattern spots, and then calculates and analyzes category changes.
With the development of satellite and sensor technologies, images with different spatial resolutions are used adequately in construction land change detection, and application of large-scale areas is a key point of research. High spatial resolution multispectral images (the spatial resolution is less than or equal to 5 meters) such as GaoFen-2, worldView-3 and the like can provide fine construction land change information, but the phenomenon of 'homomeric foreign matters and homospectral foreign matters' is serious, and the cost is high when the multispectral visible images are applied to large-scale area change detection due to the high price. Landsat, MODIS and other low-spatial-resolution multispectral images (the spatial resolution is more than or equal to 30 meters) are free and easy to obtain and have large widths, but the detail characteristics of ground objects are obviously insufficient. In contrast, the Sentinel-2 multispectral image with medium spatial resolution (the highest spatial resolution is 10 meters) has the advantages that the data is easy to obtain, the abundant ground feature details can be provided, and the Sentinel-2 multispectral image is an ideal data source for detecting the ground change of a large-scale area construction. However, most of the current Sentinel-2 image construction land change detection is focused on supervision researches, and although the construction land change types can be provided, the training sample acquisition is time-consuming and labor-consuming, so that the practical application is lack of competitiveness. The development of an efficient Sentinel-2 image construction land unsupervised change detection method is urgent.
Disclosure of Invention
The invention aims to solve the technical problems that: the method solves the problems of low recall ratio, poor generalization capability, low automation degree and the like in the large-scale application of the existing change detection method, provides a construction land non-supervision change detection method with field knowledge constraint, and achieves the purpose of quickly and accurately extracting the change of the large-scale construction land.
In order to solve the technical problems, the invention provides a construction land unsupervised change detection method of domain knowledge constraint, which comprises the following steps:
step 1, performing geometric correction and radiation correction pretreatment on two-stage Sentinel-2 remote sensing images;
step 2, identifying all types of changes by using four non-supervision change detection methods, namely a change vector analysis method, an iterative weighted multivariate change detection method, a multivariate change detection method and a principal component analysis method, and generating an initial change detection result J based on ensemble learning;
step 3, restraining real changes and all pseudo changes of the non-construction land by using domain knowledge, and only keeping the real changes of the construction land, wherein the method specifically comprises the following steps:
3.1, calculating a water-vegetation change index ICWV of the two-stage images, and accordingly obtaining a Mask image Mask of water-vegetation change 1 ；
3.2, calculating the normalized difference vegetation index of the two-stage images, and obtaining the normalized difference vegetation indexObtaining Mask image Mask of vegetation intra-class change and pseudo-change 2 ；
3.3, calculating normalized difference water body indexes of the two-stage images, and accordingly obtaining Mask image Mask of change and pseudo-change in the water body 3 ；
3.4, calculating the brightness components of the two-phase images, and accordingly obtaining a Mask image Mask with pseudo-change of the construction land 4 ；
And 3.5, calculating a construction land change detection result after domain knowledge constraint by using the following formula:
wherein J is the initial change detection result, mask i Calculating an ith mask image for domain knowledge constraint;
and 4, enhancing connectivity and integrity of the construction land change pattern spots by using mathematical morphological post-treatment.
The invention has the beneficial effects that:
the construction land unsupervised change detection method based on the domain knowledge constraint can realize large-area, high-precision and high-efficiency construction land change detection, and better serve for urban dynamic monitoring and urban sustainable development.
Drawings
The following drawings are intended to illustrate and explain the present invention and are not intended to limit the scope of the invention. Wherein:
FIG. 1 is a truth chart of two-phase Sentinel-2 images and land changes for construction of an example study area.
Fig. 2 is a flow chart of the present invention.
Fig. 3 is a graph of the results of the application of the present invention in an example study area.
FIG. 4 is a partial exemplary graph of the results of the present invention applied in an exemplary study area.
FIG. 5 is a partial exemplary graph of the detection results before and after the mask of the true variation between the water body and the vegetation.
FIG. 6 is a partial illustration of the detection of intra-vegetation class variation and spurious variation masking.
FIG. 7 is a partial exemplary graph of the detection results before and after masking of changes and spurious changes in a body of water.
Fig. 8 is a partial example diagram of detection results before and after construction land pseudo-change masks.
Detailed Description
The technical route and operation steps of the present invention will be more apparent from the following detailed description of the present invention with reference to the accompanying drawings. The research area of the embodiment of the invention is a main urban area of Nanjing city and surrounding areas, and the true value of the construction land change of two-stage Sentinel-2 image and manual visual interpretation is shown in figure 1. The shooting time of the two-stage images is respectively 2019, 10, 19 and 2021, 10, 3, 5000 pixels×5000 pixels, and the spatial resolution is 10 meters, and the two-stage images comprise four wave bands of blue, green, red and near infrared.
It should be understood by those skilled in the art that the following embodiments are not limited to the only embodiments of the present invention, and any equivalent changes or modifications made under the spirit of the embodiments of the present invention should be considered as the protection scope of the present invention.
Fig. 2 is a flowchart of an unsupervised change detection method of construction land for domain knowledge constraint, fig. 3 is a result of detection of change of construction land for example research area, and a partial example is shown in fig. 4.
The method of the present invention mainly comprises the following steps, which are described in detail below.
And step 1, performing geometric correction and radiation correction pretreatment on two-stage Sentinel-2 remote sensing images (only using blue, green, red and near infrared bands with spatial resolution of 10 meters).
Step 2, identifying all types of changes by using four non-supervision change detection methods, namely a change vector analysis method, an iterative weighted multivariate change detection method, a multivariate change detection method and a principal component analysis method, and generating an initial change detection result by utilizing the following formula for integrated learning:
where x (i, j) is the label of the ith row and jth column pixels, and C (i, j) is the total number of times the ith row and jth column pixels are identified as a class of change by the four change detection methods.
And 3, constraining the real change and all pseudo changes of the non-construction land by using domain knowledge, and only keeping the real change of the construction land. The method specifically comprises the following steps:
(a) Constructing a water-vegetation change index (Index of Change Between Water Body and Vegetation, ICWV) removes the true change between water-vegetation. ICWV is first calculated using the following formula:
wherein f 01 NDWI is a maximum and minimum normalization operation (to remove dimension effects) 2 Normalized differential water index for second phase images, NDVI 2 Normalized differential vegetation index, NDWI, for second-phase images 1 Normalized differential water index for first-phase images, NDVI 1 Normalized differential vegetation index for the first-stage image; in ICWV, the high value region is the vegetation to water variation, the low value region is the water to vegetation variation, and the construction land variation is located in the median region.
Then based on ICWV, the variable pixels are clustered into N classes (N of the research area of the example is set to 4 through parameter adjustment) by using a fuzzy C-means clustering method, and a Mask image Mask for the water-vegetation change is generated by the following formula 1 ：
Where x (i, j) is the label of the ith row and jth column of pixels, ICWV (i, j) is the ICWV value of the ith row and jth column of pixels, M 1 And M 2 The minimum and maximum class center intensity values, L (i, j), are the label values of the ith row and jth column image elements in the initial change detection result, respectively.
By using the step (a), the actual change between the water body and the vegetation in the initial mask change detection result can be effectively detected, and a partial example of the mask front-back change detection result is shown in fig. 5.
(b) And removing the intra-vegetation variation and the pseudo-variation by using the normalized difference vegetation index. Specifically, calculating normalized difference vegetation index of two-stage images, and generating Mask image Mask of vegetation intra-class change and pseudo-change by using the following formula 2 ：
Where x (i, j) is the label of the ith row and jth column pels, NDVI 1 (i, j) is the normalized difference vegetation index value NDVI of the ith row and jth column pixels of the first-period image 1 ，NDVI 2 (i, j) is the normalized difference vegetation index value NDVI of the ith row and jth column pixels of the second-stage image 2 ，N 1 And N 2 NDVI respectively 1 And NDVI 2 L (i, j) is the label value of the ith row and jth column pels in the initial change detection result.
By using the step (b), the vegetation type internal change and the pseudo-change in the initial mask change detection result can be effectively detected, and a partial example of the mask front and rear detection result is shown in fig. 6.
(c) And removing the internal change and the pseudo change of the water body by using the normalized difference water body index. Specifically, a normalized difference water body index of the two-phase image is calculated, and a Mask image Mask for the change and pseudo change in the water body is generated by using the following formula 3 ：
In NDWI 1 (i, j) is the normalized difference water index value NDWI of the ith row and jth column pixels of the first-stage image 1 ，NDWI 2 (i, j) is the normalized difference water index value NDWI of the ith row and jth column pixels of the second-stage image 2 ，Z 1 And Z 2 NDWI, respectively 1 And NDWI 2 L (i, j) is the label value of the ith row and jth column pels in the initial change detection result. Considering that in rapidly developing urban areas, the water body area is often small in proportion, and NDWI 1 And NDWI 2 Only a small peak is present on the right side and the optimal segmentation threshold should be the largest valley. Since the OTSU thresholding method is more suitable for histogram double peak case, the histogram filtering method is used to calculate Z 1 And Z 2 . Specifically, first for NDWI 1 And NDWI 2 Performing local filtering from left to right to generate a plurality of local minima, and then taking their maxima as the final segmentation threshold.
By using the step (c), the initial mask change detection result can be effectively used to detect the internal change and the pseudo-change of the water body, and a partial example of the detection result before and after the mask is shown in fig. 7.
(d) And removing the pseudo-change of the construction land by utilizing the brightness component. Specifically, the luminance component V of the two-phase image is calculated using the following formula 1 、V 2 Generating Mask image Mask of pseudo-change of construction land by combining OTSU threshold segmentation method 4 ：
V 1 ＝max(R 1 ,G 1 ,B 1 )
Wherein R is 1 、G 1 、B 1 The reflectivity of red, green and blue bands of the first-stage image are respectively.
V 2 ＝max(R 2 ,G 2 ,B 2 )
Wherein R is 2 、G 2 、B 2 The reflectivity of red, green and blue bands of the second-stage image are respectively.
Wherein V is 1 (i, j) is the brightness component value of the ith row and jth column pixels of the first-stage image, V 2 (i, j) is the brightness component value of the ith row and jth column pixels of the second-stage image, Y 1 And Y 2 V is respectively 1 And V 2 L (i, j) is the value of the ith row and jth column pixel in the initial change detection result.
By using the above step (d), the pseudo-change of the construction land in the initial mask change detection result can be effectively performed, and a partial example of the detection result before and after the mask is shown in fig. 8.
(e) Calculating a construction land change detection result after the field constraint by using the following formula:
wherein J is the initial change detection result, mask i An ith mask image (generated by steps (a) - (d) above).
And 4, enhancing connectivity and integrity of the construction land change pattern spots by using mathematical morphological post-treatment. Specifically, small patches with a pixel count of less than 10 are first removed, and then a closed operation (a square with 7×7 structural elements) and a hole filling operation are used to produce the final result.
Through the steps 1-4, the precision, recall ratio, F1 fraction and overall accuracy of the example research area construction land change detection result are divided into 68.48%, 91.01%,78.15% and 95.52%, which shows that the invention can have high recall ratio and higher precision when used for large-scale Sentinel-2 image construction land unsupervised change detection.
In addition to the embodiments described above, other embodiments of the invention are possible. All technical schemes formed by equivalent substitution or equivalent transformation fall within the protection scope of the invention.
Claims (9)
1. An unsupervised change detection method for construction land of domain knowledge constraint comprises the following steps:
step 1, performing geometric correction and radiation correction pretreatment on two-stage Sentinel-2 remote sensing images;
step 2, identifying all types of changes by using four non-supervision change detection methods, namely a change vector analysis method, an iterative weighted multivariate change detection method, a multivariate change detection method and a principal component analysis method, and generating an initial change detection result J based on ensemble learning;
step 3, restraining real changes and all pseudo changes of the non-construction land by using domain knowledge, and only keeping the real changes of the construction land, wherein the method specifically comprises the following steps:
3.1, calculating a water-vegetation change index ICWV of the two-stage images, and accordingly obtaining a Mask image Mask of water-vegetation change 1 ；
3.2, calculating normalized difference vegetation indexes of the two-stage images, and accordingly obtaining Mask image Mask of intra-vegetation change and pseudo-change 2 ；
3.3, calculating normalized difference water body indexes of the two-stage images, and accordingly obtaining Mask image Mask of change and pseudo-change in the water body 3 ；
3.4, calculating the brightness components of the two-phase images, and accordingly obtaining a Mask image Mask with pseudo-change of the construction land 4 ；
And 3.5, calculating a construction land change detection result after domain knowledge constraint by using the following formula:
wherein J is the initial change detection result, mask i Calculating an ith mask image for domain knowledge constraint;
and 4, enhancing connectivity and integrity of the construction land change pattern spots by using mathematical morphological post-treatment.
2. The construction land unsupervised change detection method of domain knowledge constraint according to claim 1, wherein: the ensemble learning rule employed for generating the initial change detection result is defined as follows:
where x (i, j) is the label of the ith row and jth column of pixels, and C (i, j) is the total number of times the ith row and jth column of pixels are identified as a change class by the four change detection methods.
3. The construction land unsupervised change detection method of domain knowledge constraint according to claim 1, wherein: the calculation formula of the water body-vegetation change index ICWV in the step 3.1 is as follows:
wherein f 01 NDWI for maximum and minimum normalization operations 2 Normalized differential water index for second phase images, NDVI 2 Normalized differential vegetation index, NDWI, for second-phase images 1 Normalized differential water index for first-phase images, NDVI 1 Normalized differential vegetation index for the first-stage image;
then based on ICWV, the variable pixels in the initial detection result are clustered into preset N classes by using a fuzzy C-means clustering method, and the Mask image Mask for the water-vegetation change is generated by using the following formula 1 ：
Where x (i, j) is the label of the ith row and jth column of pixels, ICWV (i, j) is the ICWV value of the ith row and jth column of pixels, M 1 And M 2 The minimum and maximum class center intensity values, L (i, j), are the label values of the ith row and jth column image elements in the initial change detection result, respectively.
4. A construction land unsupervised change detection method according to the domain knowledge constraint of claim 3, characterized in that: the specific method of the step 3.2 is as follows: calculating normalized difference vegetation indexes of the two-stage images, and generating Mask image Mask of vegetation intra-class change and pseudo-change by using the following formula 2 ：
Where x (i, j) is the label of the ith row and jth column pels, NDVI 1 (i, j) is the normalized difference vegetation index value NDVI of the ith row and jth column pixels of the first-period image 1 ，NDVI 2 (i, j) is the normalized difference vegetation index value NDVI of the ith row and jth column pixels of the second-stage image 2 ，N 1 And N 2 NDVI respectively 1 And NDVI 2 L (i, j) is the label value of the ith row and jth column pels in the initial change detection result.
5. The construction land unsupervised change detection method of domain knowledge constraint according to claim 4, wherein: the specific method of step 3.3 is as follows: calculating normalized difference water body indexes of two-phase images, and generating Mask image Mask of change and pseudo change in water body by using the following formula 3 ：
In NDWI 1 (i, j) is the normalized difference water index value NDWI of the ith row and jth column pixels of the first-stage image 1 ，NDWI 2 (i, j) is the normalized difference water index value NDWI of the ith row and jth column pixels of the second-stage image 2 ，Z 1 And Z 2 NDWI, respectively 1 And NDWI 2 L (i, j) is the label value of the ith row and jth column pels in the initial change detection result.
6. The construction land unsupervised change detection method of domain knowledge constraint according to claim 5, characterized in that: in step 3.3, for NDWI 1 And NDWI 2 Performing local filtering from left to right to generate a plurality of local minima, and then taking their maxima as the final segmentation threshold.
7. The construction land unsupervised change detection method of domain knowledge constraint according to claim 5, characterized in that: the specific method of the step 3.4 is as follows: the brightness component V of the two-phase image is calculated by the following formula 1 、V 2 Generating Mask image Mask of pseudo-change of construction land by combining OTSU threshold segmentation method 4 ：
V 1 ＝max(R 1 ,G 1 ,B 1 )
Wherein R is 1 、G 1 、B 1 The reflectivity of red, green and blue wave bands of the first-stage image are respectively;
V 2 ＝max(R 2 ,G 2 ,B 2 )
wherein R is 2 、G 2 、B 2 The reflectivity of red, green and blue wave bands of the second-stage image are respectively;
wherein V is 1 (i, j) is the brightness component value of the ith row and jth column pixels of the first-stage image, V 2 (i, j) is the brightness component value of the ith row and jth column pixels of the second-stage image, Y 1 And Y 2 V is respectively 1 And V 2 L (i, j) is the value of the ith row and jth column pixel in the initial change detection result.
8. The construction land unsupervised change detection method of domain knowledge constraint according to claim 1, wherein: the mathematical morphology post-processing step used is to first remove small spots with pixel numbers less than 10, and then use a closed operation and a hole filling operation to produce the final result.
9. The construction land unsupervised change detection method of domain knowledge constraint according to claim 8, wherein: the structural elements in the closed operation are 7×7 squares.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005807.3A CN116012334B (en) | 2023-01-04 | 2023-01-04 | Construction land unsupervised change detection method based on domain knowledge constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310005807.3A CN116012334B (en) | 2023-01-04 | 2023-01-04 | Construction land unsupervised change detection method based on domain knowledge constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116012334A true CN116012334A (en) | 2023-04-25 |
| CN116012334B CN116012334B (en) | 2024-06-25 |
Family
ID=86025268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310005807.3A Active CN116012334B (en) | 2023-01-04 | 2023-01-04 | Construction land unsupervised change detection method based on domain knowledge constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116012334B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855487A (en) * | 2012-08-27 | 2013-01-02 | 南京大学 | Method for automatically extracting newly added construction land change image spot of high-resolution remote sensing image |
| CN103366373A (en) * | 2013-07-10 | 2013-10-23 | 昆明理工大学 | Multi-time-phase remote-sensing image change detection method based on fuzzy compatible chart |
| 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 |
| CN109598701A (en) * | 2018-10-29 | 2019-04-09 | 同济大学 | A kind of non-supervisory change detecting method of multi-spectrum remote sensing image based on Information expansion |
| CN109614942A (en) * | 2018-12-14 | 2019-04-12 | 中国科学院遥感与数字地球研究所 | A kind of forest disturbance long-term sequence monitoring method based on cloud computing platform |
| CN111507303A (en) * | 2020-04-28 | 2020-08-07 | 同济大学 | Wetland plant species detection method |
| CN112817954A (en) * | 2021-01-27 | 2021-05-18 | 胡安民 | Missing value interpolation method based on multi-method ensemble learning |
| CN113963222A (en) * | 2021-10-28 | 2022-01-21 | 中国电子科技集团公司第五十四研究所 | High-resolution remote sensing image change detection method based on multi-strategy combination |
| CN114066876A (en) * | 2021-11-25 | 2022-02-18 | 北京建筑大学 | Construction waste change detection method based on classification result and CVA-SGD method |
| 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 |
| WO2022252799A1 (en) * | 2021-06-04 | 2022-12-08 | 成都数之联科技股份有限公司 | Model training method, woodland change detection method, system, and apparatus, and medium |
-
2023
- 2023-01-04 CN CN202310005807.3A patent/CN116012334B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855487A (en) * | 2012-08-27 | 2013-01-02 | 南京大学 | Method for automatically extracting newly added construction land change image spot of high-resolution remote sensing image |
| CN103366373A (en) * | 2013-07-10 | 2013-10-23 | 昆明理工大学 | Multi-time-phase remote-sensing image change detection method based on fuzzy compatible chart |
| 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 |
| CN109598701A (en) * | 2018-10-29 | 2019-04-09 | 同济大学 | A kind of non-supervisory change detecting method of multi-spectrum remote sensing image based on Information expansion |
| CN109614942A (en) * | 2018-12-14 | 2019-04-12 | 中国科学院遥感与数字地球研究所 | A kind of forest disturbance long-term sequence monitoring method based on cloud computing platform |
| CN111507303A (en) * | 2020-04-28 | 2020-08-07 | 同济大学 | Wetland plant species detection method |
| CN112817954A (en) * | 2021-01-27 | 2021-05-18 | 胡安民 | Missing value interpolation method based on multi-method ensemble learning |
| WO2022252799A1 (en) * | 2021-06-04 | 2022-12-08 | 成都数之联科技股份有限公司 | Model training method, woodland change detection method, system, and apparatus, and medium |
| CN113963222A (en) * | 2021-10-28 | 2022-01-21 | 中国电子科技集团公司第五十四研究所 | High-resolution remote sensing image change detection method based on multi-strategy combination |
| CN114066876A (en) * | 2021-11-25 | 2022-02-18 | 北京建筑大学 | Construction waste change detection method based on classification result and CVA-SGD method |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| 杜培军: "一种基于集成学习的城市新增建设用地快速提取 方法", 《环境监控与预警》, vol. 11, no. 5, 30 September 2019 (2019-09-30) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116012334B (en) | 2024-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | DeepCrack: A deep hierarchical feature learning architecture for crack segmentation | |
| CN108573276B (en) | Change detection method based on high-resolution remote sensing image | |
| Zhou et al. | Multiscale water body extraction in urban environments from satellite images | |
| Yin et al. | Hot region selection based on selective search and modified fuzzy C-means in remote sensing images | |
| Li et al. | FoSA: F* seed-growing approach for crack-line detection from pavement images | |
| CN101408942B (en) | Method for locating license plate under a complicated background | |
| CN103400151B (en) | The optical remote sensing image of integration and GIS autoregistration and Clean water withdraw method | |
| CN103077515B (en) | Multi-spectral image building change detection method | |
| CN106815583B (en) | Method for positioning license plate of vehicle at night based on combination of MSER and SWT | |
| CN107992856B (en) | High-resolution remote sensing building shadow detection method under urban scene | |
| CN109635726B (en) | Landslide identification method based on combination of symmetric deep network and multi-scale pooling | |
| CN109978032A (en) | Bridge Crack detection method based on spatial pyramid cavity convolutional network | |
| CN103971377A (en) | Building extraction method based on prior shape level set segmentation | |
| CN112396612A (en) | Vector information assisted remote sensing image road information automatic extraction method | |
| CN116630971B (en) | Wheat scab spore segmentation method based on CRF_Resunate++ network | |
| CN111353371A (en) | Coastline extraction method based on satellite-borne SAR image | |
| CN112907626A (en) | Moving object extraction method based on satellite time-exceeding phase data multi-source information | |
| CN115731257A (en) | Leaf form information extraction method based on image | |
| CN112101283A (en) | Intelligent identification method and system for traffic signs | |
| CN114332644A (en) | Large-view-field traffic density acquisition method based on video satellite data | |
| CN109785318B (en) | Remote sensing image change detection method based on facial line primitive association constraint | |
| CN112232249A (en) | Remote sensing image change detection method and device based on depth features | |
| CN116012334B (en) | Construction land unsupervised change detection method based on domain knowledge constraint | |
| CN116682024A (en) | Rapid cloud detection method based on four-band remote sensing image | |
| Alshehhi et al. | Change detection using multi-scale convolutional feature maps of bi-temporal satellite high-resolution images |
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 |