CN108961284A - SAR image building extracting method, equipment and the storage medium of side lobe effect pollution - Google Patents
SAR image building extracting method, equipment and the storage medium of side lobe effect pollution Download PDFInfo
- Publication number
- CN108961284A CN108961284A CN201810598678.2A CN201810598678A CN108961284A CN 108961284 A CN108961284 A CN 108961284A CN 201810598678 A CN201810598678 A CN 201810598678A CN 108961284 A CN108961284 A CN 108961284A
- Authority
- CN
- China
- Prior art keywords
- sar image
- side lobe
- polarization
- building
- lobe effect
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30184—Infrastructure
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of SAR image building extracting methods for side lobe effect pollution, specifically include step: (1) by SAR image plane pixel to be processed labeled as marginal point and non-edge point；(2) using the marginal point and non-edge point as starting point, region growing is carried out according to polarization similarity and merges to obtain object unit；(3) label that is dominant is scattered to the object unit pixel, being dominant to mark according to the scattering carries out building extraction.This method realizes simple, easy to implement, building extraction accuracy height, with high robust, can be directly used for the polarimetric SAR image processing that various polarization SAR systems obtain.
Description
Technical field
The present invention relates to remote sensing image process field, especially the SAR image building extracting method of side lobe effect pollution,
Equipment and storage medium.
Background technique
SAR (synthetic aperture radar) has round-the-clock, round-the-clock and wear compared to other traditional remote sensing images obtaining means
The characteristics of cloud Penetrating Fog, and there is the significant advantage for being observed and detecting over the ground in real time and obtain the following information of earth's surface.Polarity combination
Aperture radar is that the different Polarization scattering channel scatterings that can receive for developing to come on the basis of original single polarization synthetic aperture radar are believed
The New Type Radar of breath.The SAR image of high-resolution difference polarization mode is the significant data for studying city atural object, variation monitoring
Source.Needs are built to meet urban economy rapid development and digital city, building information is extracted as increasingly increased city
Geospatial information provides accurately and effectively data support, and the space distribution information analysis of building can be City expending and ring
Border Changeement etc. provides necessary basic data, these are to mapping geography information, geographical national conditions monitoring, urban planning, soil
It is had great significance using fields such as investigation.The remote sensing image that the research of building extracting method is utilized dependent on it, remote sensing
The raising of image resolution, which is realized, extracts the transformation that position detection and shape describe from block for building information acquisition level.
Since SAR video imaging is using the particularity of radar wave and the limitation of particular distance imaging mode, SAR image building information solution
It is slower to translate progress, in High-resolution SAR Images, building, structures side lobe effect make the extraction result of building
Accuracy, how side lobe effect pollution SAR image in more acurrate acquisition building information be still SAR image atural object solution
The difficult point and hot spot translated.
Summary of the invention
Drawbacks described above based on the prior art, the embodiment of the present invention provide that a kind of target information extraction accuracy is high, has Shandong
Stick high efficiency, SAR image building extracting method, equipment and the storage medium for realizing simple side lobe effect pollution.
The present invention can realize in many ways, including method, system, unit or computer-readable medium, under
Discuss several embodiments of the present invention in face.
A kind of SAR image building extracting method for side lobe effect pollution, specifically includes step:
(1) SAR image plane pixel to be processed is labeled as marginal point and non-edge point；
(2) using the marginal point and non-edge point as starting point, region growing is carried out according to polarization similarity and merges to obtain pair
As unit；
(3) label that is dominant is scattered to the object unit pixel, being dominant to mark according to the scattering carries out building
It extracts.
Further, step (1) specifically includes:
(1-1) calculates the polarization general power of SAR image to be processed；
(1-2) calculates the gradient vector of the polarization general power, and it is general to calculate gradient amplitude accumulation according to the gradient vector
Rate and confidence level；
(1-3) carries out non-maxima suppression and magnetic hysteresis threshold operation to cumulative probability-confidence interval, by the SAR shadow
As planar pixel is labeled as marginal point and non-edge point.
Further, the SAR image polarization general power are as follows:
Span=| SHH|2+|SHV|2+|SVH|2+|SVV|2
Wherein, | S**|2Represent the intensity of tetra- POLARIZATION CHANNEL complex datas of HH, HV, VH, VV.
Further, the method for confidence level being calculated according to the gradient vector are as follows:
ζ=| tTγ|
Wherein, γ indicates that normalization data vector, t are the standard edge template on gradient direction.
Further, in step (1-3) according to side lobe effect contaminated area profile set confidence level height judgment condition, and root
Magnetic hysteresis threshold is carried out to through non-maxima suppression treated cumulative probability-confidence interval according to the confidence level height judgment condition
Value Operations.
Further, polarization similarity expression formula is in step (2)
Wherein, σ indicates the standard deviation of input data, and n indicates the weight of input data, σmiIndicate m-th of ith feature
The standard deviation of pixel.
Further, step (3) carries out the incoherent operation splitting of four components to the object unit complete polarization complex data,
The scattering mark that is dominant is carried out to pixel after incoherent decomposition, building in object unit is carried out according to the scattering mark that is dominant and is thrown
Ticket, and atural object polarization scattering characteristics priori knowledge is combined to carry out building coarse extraction.
Further, Polarization scattering power threshold is set, it is dirty that side lobe effect is carried out from the building coarse extraction result
Area is contaminated to reject.
A kind of storage equipment, wherein storing a plurality of instruction, described instruction is suitable for being loaded by processor and being executed such as right
It is required that the step of 1 to 8 described in any item extracting methods.
A kind of SAR image building extract equipment for side lobe effect pollution, including processor, are adapted for carrying out each finger
It enables；And storage equipment, it is suitable for storing a plurality of instruction, described instruction is suitable for by processor load and perform claim requires 1 to 8
Extracting method described in one.
SAR image building extracting method of the present invention for side lobe effect pollution, by by SAR shadow to be processed
Picture planar pixel is labeled as marginal point and non-edge point, the edge detection of construction zone is realized, with marginal point and non-edge point
Region growing is carried out for starting point to merge to obtain object unit, and pixel is carried out in object unit and scatters the label that is dominant, and according to scattered
It penetrates the label that is dominant and carries out building extraction, realize simply, easy to implement, building extraction accuracy is high, has high robust, can be straight
Connect the polarimetric SAR image processing obtained for various polarization SAR systems.
Other aspects and advantages of the present invention become obviously according to detailed description with reference to the accompanying drawing, the attached drawing
The principle of the present invention is illustrated by way of example.
Detailed description of the invention
Examples of the present invention will be described by way of reference to the accompanying drawings, in which:
Fig. 1 is building extracting method flow chart provided in an embodiment of the present invention.
Specific embodiment
All features disclosed in this specification or disclosed all methods or in the process the step of, in addition to mutually exclusive
Feature and/or step other than, can combine in any way.
Any feature disclosed in this specification unless specifically stated can be equivalent or with similar purpose by other
Alternative features are replaced.That is, unless specifically stated, each feature is an example in a series of equivalent or similar characteristics
?.
Fig. 1 is the SAR image building extracting method flow chart provided in an embodiment of the present invention for side lobe effect pollution,
As shown in Figure 1, specifically including step:
(1) SAR image plane pixel to be processed is labeled as marginal point and non-edge point；
(2) using the marginal point and non-edge point as starting point, region growing is carried out according to polarization similarity and merges to obtain pair
As unit；
(3) label that is dominant is scattered to the object unit pixel, being dominant to mark according to the scattering carries out building
It extracts.
Wherein, SAR image to be processed is the polarization SAR image for having completed Speckle noise removal, specifically be can be used
Simitest algorithm carries out phase separation immunoassay processing.
Wherein, step (1) specifically includes:
(1-1) calculates the polarization general power of SAR image to be processed；
(1-2) calculates the gradient vector of the polarization general power, and it is general to calculate gradient amplitude accumulation according to the gradient vector
Rate and confidence level；
(1-3) carries out non-maxima suppression and magnetic hysteresis threshold operation to cumulative probability-confidence interval, by the SAR shadow
As planar pixel is labeled as marginal point and non-edge point.
The SAR image polarization general power are as follows:
Span=| SHH|2+|SHV|2+|SVH|2+|SVV|2
Wherein, | S**|2The intensity of tetra- POLARIZATION CHANNEL complex datas of HH, HV, VH, VV is represented, i.e. four POLARIZATION CHANNELs are multiple
Square of the mould of number data.
The gradient vector of SAR image polarization general power is calculated, SAR image polarization general power is most fast along gradient direction variation,
Change rate is maximum, and wherein change rate is the modulus value of gradient vector.The method for calculating gradient amplitude cumulative probability according to gradient vector
ForWhereinThe amplitude for indicating gradient vector, gradient vector amplitude is arranged from small to large,
Indicate n-th of gradient vector amplitude, the numerical value in Prob function return area falls in the probability in specified section.According to gradient vector
Calculate confidence level method be ζ=| tTγ |, wherein γ indicates that normalization data vector, t are the standard edge on gradient direction
Template, T are transposition operator, tTIt indicates to carry out t matrix transposition operation, confidence level is for characterizing normalization data vector and mark
Correlation between prospective edge template.
Non-maxima suppression processing is carried out to cumulative probability-confidence interval in step (1-3), by inhibiting non-maximum
Element, search for local maximum to realize that target object location detects, non-maxima suppression window size builds according to SAR image
Build the setting of object average-size.After non-maxima suppression processing, continue to carry out magnetic hysteresis threshold value to the cumulative probability-confidence interval
Operation.Specifically, according to side lobe effect contaminated area profile set confidence level height judgment condition, and according to the confidence level height
Judgment condition carries out magnetic hysteresis threshold operation to through non-maxima suppression treated cumulative probability-confidence interval, by SAR image
Planar pixel is labeled as marginal point and non-edge point.In magnetic hysteresis thresholding process, confidence level is higher than the height and adjudicates item
The pixel of part is set as marginal point, and the pixel by confidence level lower than the height judgment condition is set as non-edge point, confidence level
The pixel among the height judgment condition is fallen in, if having marginal point in its neighborhood, which is set as marginal point, it is no
It is then set as non-edge point, magnetic hysteresis threshold process can realize the degree for being adjusted flexibly according to height judgment condition and obtaining marginal point.It is logical
Setting confidence level height judgment condition is crossed, practical construction zone and side lobe effect Polluted area are separated, tentatively solved
The problem of side lobe effect contaminated area is mistaken for building when using high-resolution polarization SAR Extraction of Image building.
In magnetic hysteresis thresholding process, the low judgment condition setting of confidence level is too low, and the marginal point that will lead to label is more and miscellaneous
Disorderly, the separation of side lobe effect contaminated area Yu non-polluting area is influenced；If the high judgment condition setting of confidence level is excessively high, label will lead to
Marginal point is very few, same to influence separation.The setting of confidence level height judgment condition need to guarantee side lobe effect contaminated area with non-dirt
The confidence level for contaminating most of pixel of area's zone of transition is higher than high judgment condition, or in its vicinity.
The marginal point and non-edge point obtained using step (1) carries out region growing merging as starting point, according to polarization similarity
Object unit is obtained, wherein polarization similarity expression formula is
Wherein, σ indicates the standard deviation of input data, and n indicates the weight of input data, σmiIndicate m-th of ith feature
The standard deviation of pixel.In object unit growth course, the determination of growth standard is one of the major issue concerning algorithm validity,
Homogeney based on object unit, the pixel for being incorporated to same target unit meet same criterion, for example, with marginal point and non-
Marginal point is starting point, the pixel that the similarity that polarizes meets threshold condition is incorporated into the same area, or multispectral SAR is schemed
As regarding hyperspace as, the Euclidean distance measured between pixel is used as growth standard, using marginal point and non-edge point as
The pixel that pixel Euclidean distance meets threshold distance condition is incorporated into the same area by point, and the standard is raw for combining unit
Criterion that is long and terminating.When merging criterion as growth using the similarity that polarizes, the polarization phase of zoning and pixel
Like degree, the similarity difference that polarizes then merges within the scope of user's given threshold, and otherwise nonjoinder terminates growth, executes this hair
When the bright step, threshold value can't excessively influence as a result, but functioning only as the effects of initial combining objects.
The incoherent operation splitting of four components is carried out to the object unit complete polarization complex data that step (2) obtains, obtains surprise
Secondary scattering, even scattering, volume scattering and spiral volume scattering power, carry out the scattering mark that is dominant, root to pixel after incoherent decomposition
Building ballot in object unit is carried out according to the scattering mark that is dominant, and atural object polarization scattering characteristics priori knowledge is combined to carry out
Building coarse extraction.The scattering mark that is dominant be compare pixel even scattering, volume scattering, the scattered power value of surface scattering it is big
It is small, and pixel is worth maximum scattering type labeled as scattered power, it determines after pixel is dominant and scatters type, it can be according to building
The polarization scattering characteristics that the scattering of object even is dominant are built to complete building coarse extraction.Polarization Characteristics Similarity meets scale invariability, only
The Polarization scattering feature that target can be expressed cannot express the echo strength information of target scattering, special for comprehensive utilization Polarization scattering
It seeks peace echo strength information, using be dominant scattering mechanism (Dominant scattering Mechanism) and the scattering that is dominant
Intensity (Dominant scattering Power) two dimension variable describes target scattering, and using being dominant, scattering strength will be dominant
Pixel in scattering classification is divided into multiple initial clusterings, and the pixel number in initial clustering is approximately equal.Calculate each cluster
Average coherence matrix, in each classification according between class Whishart distance carry out initial clustering merging, merging standard is poly-
Class is dominant, and to scatter classification identical, and between class distance is less than threshold distance, achievable in this way that initial clustering is incorporated into final classification
Required classification number, the expectation coherence matrix for then calculating each cluster after merging is cluster centre, and iteration cluster 2-4 times complete
At cluster process.
Optimally, setting Polarization scattering power threshold slightly mentions under conditions of polarizing total power constraint from the building
Progress side lobe effect contaminated area rejecting in result is taken, the accuracy for further increasing building extraction is conducive to.Specifically, usually
Building Polarization scattering power is higher, and it is poor that the Polarization scattering power and building Polarization scattering power of side lobe effect contaminated area exist
It is different, the two separation can be realized based on the difference given threshold.
SAR image building extracting method described in the embodiment of the present invention for side lobe effect pollution is equally applicable to polarize
The Image Edge-Detections fields such as SAR Road Detection, tidal saltmarsh, crop edge extraction.
A kind of storage equipment, wherein storing a plurality of instruction, described instruction is suitable for being loaded by processor and being executed such as right
It is required that the step of 1 to 8 described in any item extracting methods.
A kind of SAR image building extract equipment for side lobe effect pollution, including processor, are adapted for carrying out each finger
It enables；And storage equipment, it is suitable for storing a plurality of instruction, described instruction is suitable for by processor load and perform claim requires 1 to 8
Extracting method described in one.
Different aspect, embodiment, embodiment or feature of the invention can be used alone or be used in any combination.
The present invention can also be realized preferably by software realization with the combination of hardware or hardware and software.The present invention
The computer-readable code that can be implemented as on computer-readable medium.Computer-readable medium is can be by after capable of storing
Any data storage device for the data that computer system is read.The example of computer-readable medium include: read-only memory, with
Machine stores memory, CD-ROM, DVD, tape, optical data storage and carrier wave.Computer-readable medium can also be distributed in
In the computer system of network connection, to store and execute computer-readable code in a distributed way.
The invention is not limited to specific embodiments above-mentioned.The present invention, which expands to, any in the present specification to be disclosed
New feature or any new combination, and disclose any new method or process the step of or any new combination.
Claims (10)
1. a kind of SAR image building extracting method for side lobe effect pollution, it is characterised in that comprising steps of
(1) SAR image plane pixel to be processed is labeled as marginal point and non-edge point；
(2) using the marginal point and non-edge point as starting point, region growing is carried out according to polarization similarity and merges to obtain object list
Member；
(3) label that is dominant is scattered to the object unit pixel, being dominant to mark according to the scattering carries out building extraction.
2. a kind of SAR image building extracting method for side lobe effect pollution according to claim 1, feature exist
In step (1) specifically includes:
(1-1) calculates the polarization general power of SAR image to be processed；
(1-2) calculate it is described polarization general power gradient vector, according to the gradient vector calculate gradient amplitude cumulative probability and
Confidence level；
(1-3) carries out non-maxima suppression and magnetic hysteresis threshold operation to cumulative probability-confidence interval, and the SAR image is put down
Face pixel is labeled as marginal point and non-edge point.
3. a kind of SAR image building extracting method for side lobe effect pollution according to claim 2, feature exist
In the SAR image polarization general power are as follows:
Span=| SHH|2+|SHV|2+|SVH|2+|SVV|2
Wherein, | S**|2Represent the intensity of tetra- POLARIZATION CHANNEL complex datas of HH, HV, VH, VV.
4. a kind of SAR image building extracting method for side lobe effect pollution according to claim 2, feature exist
In the method for calculating confidence level according to the gradient vector are as follows:
ζ=| tTγ|
Wherein, γ indicates that normalization data vector, t are the standard edge template on gradient direction.
5. a kind of SAR image building extracting method for side lobe effect pollution according to claim 2, feature exist
In, according to side lobe effect contaminated area profile set confidence level height judgment condition in step (1-3), and it is high according to the confidence level
Low judgment condition carries out magnetic hysteresis threshold operation to through non-maxima suppression treated cumulative probability-confidence interval.
6. a kind of SAR image building extracting method for side lobe effect pollution according to claim 1, feature exist
In polarization similarity expression formula is in step (2)
Wherein, σ indicates the standard deviation of input data, and n indicates the weight of input data, σmiIndicate m-th of pixel of ith feature
Standard deviation.
7. a kind of SAR image building extracting method for side lobe effect pollution according to claim 1, feature exist
In step (3) carries out the incoherent operation splitting of four components to the object unit complete polarization complex data, after incoherent decomposition
Pixel carries out the scattering mark that is dominant, and carries out building in object unit according to the scattering mark that is dominant and votes, and combines atural object
Polarization scattering characteristics priori knowledge carries out building coarse extraction.
8. a kind of SAR image building extracting method for side lobe effect pollution according to claim 7, feature exist
In setting Polarization scattering power threshold carries out the rejecting of side lobe effect contaminated area from the building coarse extraction result.
9. a kind of storage equipment, wherein storing a plurality of instruction, described instruction is suitable for being loaded by processor and executing right such as wanting
The step of seeking 1 to 8 described in any item extracting methods.
10. a kind of SAR image building extract equipment for side lobe effect pollution, it is characterised in that including processor, be suitable for
Realize each instruction；And storage equipment, it is suitable for storing a plurality of instruction, described instruction is suitable for by processor load and perform claim is wanted
Seek 1 to 8 described in any item extracting methods.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810598678.2A CN108961284A (en) | 2018-06-12 | 2018-06-12 | SAR image building extracting method, equipment and the storage medium of side lobe effect pollution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810598678.2A CN108961284A (en) | 2018-06-12 | 2018-06-12 | SAR image building extracting method, equipment and the storage medium of side lobe effect pollution |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108961284A true CN108961284A (en) | 2018-12-07 |
Family
ID=64488233
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810598678.2A Withdrawn CN108961284A (en) | 2018-06-12 | 2018-06-12 | SAR image building extracting method, equipment and the storage medium of side lobe effect pollution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108961284A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615025A (en) * | 2018-12-28 | 2019-04-12 | 内蒙古工业大学 | Quick atural object classification method |
| CN110853050A (en) * | 2019-10-21 | 2020-02-28 | 中国电子科技集团公司第二十九研究所 | SAR image river segmentation method, device and medium |
| CN112305538A (en) * | 2020-09-22 | 2021-02-02 | 中国电子科技集团公司第二十九研究所 | Synthetic aperture radar image sidelobe suppression processing method |
| CN113705582A (en) * | 2021-08-04 | 2021-11-26 | 南京林业大学 | Method for extracting edge feature key points of building facade |
| US20220230364A1 (en) * | 2021-01-19 | 2022-07-21 | Si Analytics Co., Ltd. | Method And Apparatus For Processing Radar Image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104166856A (en) * | 2014-07-30 | 2014-11-26 | 西安电子科技大学 | Polarization SAR image classification method based on neighbor propagation clustering and region growing |
| CN104240234A (en) * | 2014-08-19 | 2014-12-24 | 中国测绘科学研究院 | Complete-polarization SAR image segmentation method and device based on polarization similarity combination |
| CN104376330A (en) * | 2014-11-19 | 2015-02-25 | 西安电子科技大学 | Polarization SAR image ship target detection method based on superpixel scattering mechanism |
| CN105374017A (en) * | 2015-11-30 | 2016-03-02 | 西安电子科技大学 | Polarimetric SAR image filtering method through combination of polarimetric decomposition vector statistical distribution |
| US20160170019A1 (en) * | 2014-07-09 | 2016-06-16 | Bae Systems Information & Electronic Systems Integration Inc. | Radar detection of endo-clutter high-value targets using tracker feedback |
| CN106291550A (en) * | 2016-07-27 | 2017-01-04 | 西安电子科技大学 | The polarization SAR Ship Detection of core is returned based on local scattering mechanism difference |
-
2018
- 2018-06-12 CN CN201810598678.2A patent/CN108961284A/en not_active Withdrawn
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160170019A1 (en) * | 2014-07-09 | 2016-06-16 | Bae Systems Information & Electronic Systems Integration Inc. | Radar detection of endo-clutter high-value targets using tracker feedback |
| CN104166856A (en) * | 2014-07-30 | 2014-11-26 | 西安电子科技大学 | Polarization SAR image classification method based on neighbor propagation clustering and region growing |
| CN104240234A (en) * | 2014-08-19 | 2014-12-24 | 中国测绘科学研究院 | Complete-polarization SAR image segmentation method and device based on polarization similarity combination |
| CN104376330A (en) * | 2014-11-19 | 2015-02-25 | 西安电子科技大学 | Polarization SAR image ship target detection method based on superpixel scattering mechanism |
| CN105374017A (en) * | 2015-11-30 | 2016-03-02 | 西安电子科技大学 | Polarimetric SAR image filtering method through combination of polarimetric decomposition vector statistical distribution |
| CN106291550A (en) * | 2016-07-27 | 2017-01-04 | 西安电子科技大学 | The polarization SAR Ship Detection of core is returned based on local scattering mechanism difference |
Non-Patent Citations (3)
| Title |
|---|
| 牛蕾: "SAR图像舰船目标去旁瓣处理", 《雷达科学与技术》 * |
| 郑长利: "综合极化信息和改进FNEA的PolSAR影像建筑物提取", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
| 闫丽丽: "基于散射特征的极化SAR影像建筑物提取研究", 《中国优秀博硕士学位论文全文数据库（博士） 基础科学辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109615025A (en) * | 2018-12-28 | 2019-04-12 | 内蒙古工业大学 | Quick atural object classification method |
| CN110853050A (en) * | 2019-10-21 | 2020-02-28 | 中国电子科技集团公司第二十九研究所 | SAR image river segmentation method, device and medium |
| CN110853050B (en) * | 2019-10-21 | 2023-05-26 | 中国电子科技集团公司第二十九研究所 | SAR image river segmentation method, device and medium |
| CN112305538A (en) * | 2020-09-22 | 2021-02-02 | 中国电子科技集团公司第二十九研究所 | Synthetic aperture radar image sidelobe suppression processing method |
| CN112305538B (en) * | 2020-09-22 | 2023-06-20 | 中国电子科技集团公司第二十九研究所 | Synthetic aperture radar image sidelobe suppression processing method |
| US20220230364A1 (en) * | 2021-01-19 | 2022-07-21 | Si Analytics Co., Ltd. | Method And Apparatus For Processing Radar Image |
| CN113705582A (en) * | 2021-08-04 | 2021-11-26 | 南京林业大学 | Method for extracting edge feature key points of building facade |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108961284A (en) | SAR image building extracting method, equipment and the storage medium of side lobe effect pollution | |
| Huang et al. | A new building extraction postprocessing framework for high-spatial-resolution remote-sensing imagery | |
| CN105427314B (en) | SAR image object detection method based on Bayes's conspicuousness | |
| CN105512683B (en) | Object localization method and device based on convolutional neural networks | |
| Feng et al. | Robust non-local fuzzy c-means algorithm with edge preservation for SAR image segmentation | |
| CN105389589B (en) | A kind of chest X ray piece rib cage detection method returned based on random forest | |
| Zhan et al. | Unsupervised scale-driven change detection with deep spatial–spectral features for VHR images | |
| CN102945374A (en) | Method for automatically detecting civil aircraft in high-resolution remote sensing image | |
| CN106295498A (en) | Remote sensing image target area detection apparatus and method | |
| Craciun et al. | Joint detection and tracking of moving objects using spatio-temporal marked point processes | |
| Abdulraqeb et al. | A novel segmentation algorithm for MRI brain tumor images | |
| Al-Ghrairi et al. | Classification of satellite images based on color features using remote sensing | |
| Huang et al. | Superpixel-based change detection in high resolution sar images using region covariance features | |
| An et al. | A modified PSO algorithm for remote sensing image template matching | |
| Su et al. | Which CAM is better for extracting geographic objects? A perspective from principles and experiments | |
| Reichman et al. | Learning improved pooling regions for the histogram of oriented gradient (HOG) feature for Buried Threat Detection in Ground Penetrating Radar | |
| Samanta et al. | Spatial-resolution independent object detection framework for aerial imagery | |
| CN108805186A (en) | A kind of SAR image circle oil house detection method based on multidimensional notable feature cluster | |
| Zhang et al. | Robust DEM co-registration method for terrain changes assessment using least trimmed squares estimator | |
| Naeem et al. | A new approach for image detection based on refined Bag of Words algorithm | |
| Surya et al. | Super-pixel Segmentation based Skin texture pattern recognition | |
| Tu et al. | Automatic target recognition scheme for a high-resolution and large-scale synthetic aperture radar image | |
| Liu et al. | An unsupervised automatic change detection approach based on visual attention mechanism | |
| Jenicka et al. | Distributed texture-based land cover classification algorithm using hidden Markov model for multispectral data | |
| Gu et al. | Fast iterative censoring CFAR algorithm for ship detection from SAR 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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181207 |
|
| WW01 | Invention patent application withdrawn after publication |