CN108257160B - A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation - Google Patents
A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation Download PDFInfo
- Publication number
- CN108257160B CN108257160B CN201810057999.1A CN201810057999A CN108257160B CN 108257160 B CN108257160 B CN 108257160B CN 201810057999 A CN201810057999 A CN 201810057999A CN 108257160 B CN108257160 B CN 108257160B
- Authority
- CN
- China
- Prior art keywords
- change detection
- image
- scale segmentation
- pixels
- remote sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000012805 post-processing Methods 0.000 title claims abstract description 11
- 230000011218 segmentation Effects 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims abstract description 4
- 238000012545 processing Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/277—Analysis of motion involving stochastic approaches, e.g. using Kalman filters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/215—Motion-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/254—Analysis of motion involving subtraction of 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
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
本发明公开了基于多尺度分割‑最大期望的遥感影像变化检测后处理方法，步骤a：对滑坡区域的滑坡前影像和滑坡后影像进行空间位置配准，求解得到变化的影像；步骤b：基于滑坡后影像，进行多尺度分割，获取其多尺度分割影像对象集合S，步骤c：取S中的第i个对象Oi，与步骤a中初始变化检测结果进行空间叠置分析，分别统计对象Oi内变化像素和未变化像素的数量；步骤d：利用最大期望的算法，对对象Oi内的像素属性进行重新精细化；步骤e：令j＝i+1；如果j≤n,则取S中的第j个对象Oj,依次执行步骤c与步骤d，直至j>n；步骤f：得到最终的变化检测结果。该方法普适性较强，效果明显的基于多尺度分割‑最大期望的遥感影像变化检测后处理方法。
The invention discloses a post-processing method for detecting remote sensing image changes based on multi-scale segmentation-maximum expectation. After the landslide image, perform multi-scale segmentation to obtain the multi-scale segmentation image object set S, step c: take the ith object Oi in S, perform spatial overlay analysis with the initial change detection result in step a, and count the objects Oi separately The number of internal changed pixels and unchanged pixels; Step d: use the algorithm of maximum expectation to re-refine the pixel attributes in the object Oi; Step e: let j=i+1; if j≤n, take the middle of S For the j-th object Oj, step c and step d are performed in sequence until j>n; step f: obtain the final change detection result. This method has strong universality and obvious effect, and is a post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation.
Description
Technical Field
The invention belongs to the technical field of remote sensing image processing, and particularly relates to a remote sensing image change detection post-processing method based on multi-scale segmentation-maximum expectation.
Background
The information of the ground features on the surface of the landed ball is continuously changed by human activities, natural disasters and the like, the change information on the surface of the earth is timely and effectively acquired, and the method has important significance on environmental monitoring, urban management, disaster emergency and other aspects. Although with the rapid development of satellite and aviation remote sensing technologies, the time resolution and the spatial resolution of images are greatly improved, so that the rapid and effective acquisition of the earth surface coverage change information becomes possible. However, the detection result obtained by the conventional change detection method still has more noise, and the reliability and the detection precision of the detection result are greatly reduced. Under the background, the post-processing method for the change detection result of the remote sensing image with high resolution is provided, and has important practical significance.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a remote sensing image change detection post-processing method based on multi-scale segmentation-maximum expectation, which has strong universality and obvious effect, aiming at the defects of the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is that the method for detecting and post-processing the change of the remote sensing image based on multi-scale segmentation-maximum expectation comprises the following steps:
step a: carrying out spatial position registration on the image before landslide and the image after landslide in the landslide area, and solving to obtain a changed image, namely an initial change detection result;
step b: performing multi-scale segmentation based On the image after landslide to obtain a multi-scale segmentation image object set S, wherein the object set S comprises n objects O1 and O2 … … Oi … … On, wherein n is a natural number greater than 0; the values of i are as follows: i is more than or equal to 1 and less than or equal to n; the object O1, O2 … … Oi … … On is a collection of spectrally-characterized approximate and spatially-adjacent pixels;
step c: taking the ith object Oi in the S, carrying out spatial superposition analysis on the ith object Oi and the initial change detection result in the step a, respectively counting the number of changed pixels and unchanged pixels in the object Oi, and determining which state is large in number of pixels;
step d: the pixel attributes within the object Oi are re-refined using the most desirable algorithm, as follows: taking the state of a large number of pixels in the Oi as the state of all pixels in the object Oi;
step e: let j equal i + 1; if j is less than or equal to n, the jth object Oj in S is taken, and the step c and the step d are sequentially executed until j is greater than n;
step f: and obtaining the states of all corresponding pixels in all the objects, namely the final change detection result.
Further, when multi-scale segmentation is performed in the step b, a fractal evolution network multi-scale segmentation method is adopted.
Further, in the step a, an initial change detection result is obtained by an EM-MRF method, or an image difference method, or a ratio method, or a change vector analysis method.
The invention has the following advantages: by introducing multi-scale segmentation and a maximum expectation algorithm, various random noises are effectively removed, so that the internal consistency of a detected target or a background area is improved, and the change detection precision is obviously improved.
Drawings
FIG. 1 is an image selected for use in an embodiment of the present invention;
wherein: FIG. 1a is a front image of a landslide;
FIG. 1b is a rear image of a landslide;
FIG. 2 is a schematic view of overlay analysis in an embodiment of the present invention;
wherein: FIG. 2a is a schematic view of a stack analysis;
FIG. 2b shows the internal pixel change state at 1 and its distribution, with white pixels changed and black pixels unchanged;
FIG. 2c shows the internal pixel change state at 2 and its distribution, with white pixels changed and black pixels unchanged;
FIG. 3 is a schematic diagram illustrating the optimization results and accuracy variation at different scales according to an embodiment of the present invention;
wherein: FIG. 3a is a schematic diagram of an initial detection result obtained by the EM-MRF method;
FIG. 3b is a diagram illustrating the optimization result and the precision variation of the segmentation scale of 5;
FIG. 3c is a diagram showing the optimization result and the precision variation of the segmentation scale 10
FIG. 3d is a diagram illustrating the optimization result and the precision variation of the segmentation scale 15;
fig. 3e is a diagram illustrating the optimization result and the precision variation of the segmentation scale 20.
Detailed Description
The invention relates to a remote sensing image change detection post-processing method based on multi-scale segmentation-maximum expectation, which comprises the following steps:
step a: carrying out spatial position registration on the image before landslide and the image after landslide in the landslide area, and solving to obtain a changed image, namely an initial change detection result; as shown in fig. 1, arcmap10.0 software is respectively adopted for a landslide front image and a landslide rear image of a landslide area, and spatial position registration of the two images is realized through an Adjust tool; and obtaining an initial change detection result by an EM-MRF method, an image difference method, a ratio method or a change vector analysis method.
Step b: performing multi-scale segmentation based On the image after landslide to obtain a multi-scale segmentation image object set S, wherein the object set S comprises n objects O1 and O2 … … Oi … … On, wherein n is a natural number greater than 0; the values of i are as follows: i is more than or equal to 1 and less than or equal to n; the object O1, O2 … … Oi … … On is a collection of spectrally-characterized approximate and spatially-adjacent pixels; in this embodiment, what is implemented is: 5. scale division under 10, 15, 20, compactness and shape parameters are set to 0.8 and 0.9 respectively, taking scale 20 as an example, the division result is shown in fig. 2, 2a, 2b and 2 c;
step c: taking the ith object Oi in the S, carrying out spatial superposition analysis on the ith object Oi and the initial change detection result in the step a, respectively counting the number of changed pixels and unchanged pixels in the object Oi, and determining which state is large in number of pixels; as shown in fig. 2, 2a, 2b, and 2c, the irregular frame is a division target, and the number of black and white pixels in the division target is counted to determine which state has the larger number of pixels.
Step d: the pixel attributes within the object Oi are re-refined using the most desirable algorithm: using the pixel state with a large number in the Oi as the state of all pixels in the object Oi;
step e: let j equal i + 1; if j is less than or equal to n, the jth object Oj in S is taken, and the step c and the step d are sequentially executed until j is greater than n;
step f: and obtaining the states of all corresponding pixels in all the objects, namely the final change detection result. As shown in FIG. 3, no matter what parameter setting is selected, a relatively optimized result can be obtained, and the method is highly universal.
The results of comparing the invention of the present invention with the EM-MRF method are shown in Table 1:
TABLE 1 comparison of EM-MRF method and post-treatment method in the present invention
As can be seen from Table 1, the false detection rate, the missed detection rate and the total error rate are significantly reduced by the method of the present invention.
Claims (3)
1. The remote sensing image change detection post-processing method based on multi-scale segmentation-maximum expectation is characterized by comprising the following steps of:
step a: carrying out spatial position registration on the image before landslide and the image after landslide in the landslide area, and solving to obtain a changed image, namely an initial change detection result;
step b: performing multi-scale segmentation based On the image after landslide to obtain a multi-scale segmentation image object set S, wherein the object set S comprises n objects O1 and O2 … … Oi … … On, wherein n is a natural number greater than 0; the values of i are as follows: i is more than or equal to 1 and less than or equal to n; the object O1, O2 … … Oi … … On is a collection of spectrally-characterized approximate and spatially-adjacent pixels;
step c: taking the ith object Oi in the S, carrying out spatial superposition analysis on the ith object Oi and the initial change detection result in the step a, respectively counting the number of changed pixels and unchanged pixels in the object Oi, and determining which state is large in number of pixels;
step d: the pixel attributes within the object Oi are re-refined using the most desirable algorithm, as follows: taking the state of a large number of pixels in the Oi as the state of all pixels in the object Oi;
step e: let j equal i + 1; if j is less than or equal to n, the jth object Oj in S is taken, and the step c and the step d are sequentially executed until j is greater than n;
step f: and obtaining the states of all corresponding pixels in all the objects, namely the final change detection result.
2. The remote sensing image change detection post-processing method based on multi-scale segmentation-maximum expectation according to claim 1, wherein a fractal evolution network multi-scale segmentation method is adopted when multi-scale segmentation is performed in the step b.
3. The method for processing the remote sensing image change detection based on the multi-scale segmentation-maximum expectation in the step a as claimed in claim 1 or 2, wherein the initial change detection result is obtained by an EM-MRF method, an image difference method, a ratio method or a change vector analysis method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810057999.1A CN108257160B (en) | 2018-01-22 | 2018-01-22 | A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810057999.1A CN108257160B (en) | 2018-01-22 | 2018-01-22 | A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108257160A CN108257160A (en) | 2018-07-06 |
| CN108257160B true CN108257160B (en) | 2021-10-19 |
Family
ID=62741405
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810057999.1A Active CN108257160B (en) | 2018-01-22 | 2018-01-22 | A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108257160B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110060273B (en) * | 2019-04-16 | 2021-05-18 | 湖北省水利水电科学研究院 | Remote sensing image landslide mapping method based on deep neural network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103198483A (en) * | 2013-04-07 | 2013-07-10 | 西安电子科技大学 | Multiple time phase remote sensing image registration method based on edge and spectral reflectivity curve |
| CN103632363A (en) * | 2013-08-27 | 2014-03-12 | 河海大学 | Object-level high-resolution remote sensing image change detection method based on multi-scale fusion |
| US9235902B2 (en) * | 2011-08-04 | 2016-01-12 | University Of Southern California | Image-based crack quantification |
| CN107330875A (en) * | 2017-05-31 | 2017-11-07 | 河海大学 | Based on the forward and reverse heterogeneous water body surrounding enviroment change detecting method of remote sensing images |
-
2018
- 2018-01-22 CN CN201810057999.1A patent/CN108257160B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9235902B2 (en) * | 2011-08-04 | 2016-01-12 | University Of Southern California | Image-based crack quantification |
| CN103198483A (en) * | 2013-04-07 | 2013-07-10 | 西安电子科技大学 | Multiple time phase remote sensing image registration method based on edge and spectral reflectivity curve |
| CN103632363A (en) * | 2013-08-27 | 2014-03-12 | 河海大学 | Object-level high-resolution remote sensing image change detection method based on multi-scale fusion |
| CN107330875A (en) * | 2017-05-31 | 2017-11-07 | 河海大学 | Based on the forward and reverse heterogeneous water body surrounding enviroment change detecting method of remote sensing images |
Non-Patent Citations (2)
| Title |
|---|
| Time Series Corrections and Analyses in Thermal Remote Sensing;Sobrino J A 等;《Springer Netherlands》;20130515;第267-285页 * |
| 面向对象的遥感影像变化检测方法;赖继文 等;《测绘科学》;20170831;第42卷(第8期);第111-115页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108257160A (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111028277B (en) | SAR and optical remote sensing image registration method based on pseudo-twin convolution neural network | |
| WO2019144469A1 (en) | Image quality classification method, system and terminal device | |
| CN105809693B (en) | SAR image registration method based on deep neural network | |
| CN111862037A (en) | Method and system for geometric feature detection of precision hole parts based on machine vision | |
| CN107301661A (en) | High-resolution remote sensing image method for registering based on edge point feature | |
| CN106503643B (en) | Tumble detection method for human body | |
| CN105354841B (en) | A kind of rapid remote sensing image matching method and system | |
| CN108960404B (en) | Image-based crowd counting method and device | |
| CN104766320A (en) | Bernoulli smoothing weak target detection and tracking method under thresholding measuring | |
| CN106296613B (en) | A kind of Dual Energy Subtraction method based on DR machine | |
| CN105678734B (en) | A kind of heterologous test image scaling method of image matching system | |
| CN114529593A (en) | Infrared and visible light image registration method, system, equipment and image processing terminal | |
| CN108230375A (en) | Visible images and SAR image registration method based on structural similarity fast robust | |
| CN112907626B (en) | Moving target extraction method based on satellite super-time phase data multi-source information | |
| CN104517124B (en) | SAR image change detection based on SIFT feature | |
| CN106897986A (en) | A kind of visible images based on multiscale analysis and far infrared image interfusion method | |
| CN108830856B (en) | GA automatic segmentation method based on time series SD-OCT retina image | |
| CN116958132B (en) | Surgical navigation system based on visual analysis | |
| CN110660089A (en) | A satellite image registration method and device | |
| CN114494371A (en) | Optical image and SAR image registration method based on multi-scale phase consistency | |
| CN111784643B (en) | Cross line structured light-based tire tangent plane acquisition method and system | |
| CN109544608B (en) | Unmanned aerial vehicle image acquisition characteristic registration method | |
| CN107369163B (en) | Rapid SAR image target detection method based on optimal entropy dual-threshold segmentation | |
| CN108257160B (en) | A post-processing method for remote sensing image change detection based on multi-scale segmentation-maximum expectation | |
| CN103955943A (en) | Non-supervision change detection method based on fuse change detection operators and dimension driving |
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 |