CN113610783A - Time sequence SAR intensity image variation coefficient-based change detection method and device - Google Patents

Time sequence SAR intensity image variation coefficient-based change detection method and device Download PDF

Info

Publication number
CN113610783A
CN113610783A CN202110830747.XA CN202110830747A CN113610783A CN 113610783 A CN113610783 A CN 113610783A CN 202110830747 A CN202110830747 A CN 202110830747A CN 113610783 A CN113610783 A CN 113610783A
Authority
CN
China
Prior art keywords
variation coefficient
samples
time sequence
sample
spatial
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110830747.XA
Other languages
Chinese (zh)
Inventor
蒋弥
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sun Yat Sen University
Original Assignee
Sun Yat Sen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sun Yat Sen University filed Critical Sun Yat Sen University
Priority to CN202110830747.XA priority Critical patent/CN113610783A/en
Publication of CN113610783A publication Critical patent/CN113610783A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • 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
    • G06T2207/10044—Radar image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a method and a device for detecting change of a variation coefficient of an image based on time sequence SAR intensity, wherein the method comprises the following steps: acquiring SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence; calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient graph according to the variation coefficient of each spatial pixel; extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence; performing spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph; and carrying out threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region. By implementing the embodiment of the invention, the accuracy of change detection can be improved.

Description

Time sequence SAR intensity image variation coefficient-based change detection method and device
Technical Field
The invention relates to the technical field of remote sensing image processing, in particular to a method and a device for detecting variation coefficient of time sequence SAR intensity image.
Background
Remote sensing change detection techniques have proven to be a powerful tool for acquiring spatial information at multiple scales using high spatiotemporal coverage satellites and sufficiently high image resolution. Compared with an optical system, the all-weather observation capability and the high space-time resolution ensure the availability of SAR (Synthetic Aperture Radar) data in the occurrence period of important events with different spatial scales, and the SAR can flexibly solve the spectrum ambiguity caused by surface transition aiming at the surface roughness and the high sensitivity of constructed surface attributes.
The traditional double-temporal SAR change detection is to generate a difference map based on two SAR images at the beginning and the end so as to realize change detection, but on one hand, the resolution of an image is damaged due to local adaptive statistics driven by a spatial neighborhood in the traditional double-temporal SAR change detection, and on the other hand, only the two SAR images at the beginning and the end are used for detection, so that the information amount is too small, some change insensitive regions can be easily ignored, and the error early warning probability of irrelevant events such as vehicle passing is increased. The two aspects result in the traditional double-time-phase SAR change detection, and the accuracy is low.
Disclosure of Invention
The embodiment of the invention provides a change detection method and device based on a time sequence SAR intensity image variation coefficient, which can improve the accuracy of regional change detection.
The embodiment of the invention provides a change detection method based on a time sequence SAR intensity image variation coefficient, which comprises the steps of obtaining SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence;
calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient graph according to the variation coefficient of each spatial pixel;
extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence;
performing spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph;
and carrying out threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
Further, before calculating a variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, the method further includes: and detecting and removing outlier samples in the intensity time sequence samples of each spatial pixel point through outliers.
Further, the acquiring the SAR intensity image of a plurality of time nodes of the to-be-detected region within the preset time specifically includes:
the SAR image data of a plurality of time nodes of the region to be detected in a preset time period are collected, then all the SAR image data are registered, and SAR intensity images of the plurality of time nodes are generated.
Further, outlier samples in the intensity time sequence samples of the spatial pixel points are removed in the following mode:
sequencing the strength time sequence samples of the space pixel points according to time sequence to obtain ordered samples;
obtaining E samples to be tested from the ordered samples, and then testing the E sample to be tested according to a preset hypothesis testing formula;
if an invalid hypothesis is obtained under a preset confidence level, determining that the 1 st to E th samples to be detected are all outlier samples, then removing the 1 st to E th samples to be detected, and otherwise, detecting the E-1 st samples to be detected; wherein E is a non-zero natural number.
Further, after rejecting outlier samples in the intensity time sequence samples of one spatial pixel, the method further includes:
and checking the remaining non-outlier samples in the time sequence sample through single sample AD (analog-to-digital) checking, and if the zero-flipping hypothesis is carried out under a preset second confidence level, adding the rejected outlier samples into the time sequence sample again.
Further, the coefficient of variation of the spatial pixel is calculated by the following method:
calculating the sample mean value and the standard deviation of the intensity time sequence sample of the space pixel point;
and taking the quotient of the standard deviation and the sample mean value as the coefficient of variation of the spatial pixel point.
Furthermore, a statistical same-distribution pixel sample corresponding to each spatial pixel point is extracted through BWS hypothesis test.
On the basis of the above method item embodiments, the present invention correspondingly provides apparatus item embodiments;
the invention provides a change detection device based on a time sequence SAR intensity image variation coefficient, which comprises a data acquisition module, a variation coefficient image generation module, a statistical same-distribution pixel sample acquisition module, a filtering module and a threshold segmentation module, wherein the data acquisition module is used for acquiring a variation coefficient of a time sequence SAR intensity image;
the data acquisition module is used for acquiring SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence;
the variation coefficient map generation module is used for calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient map according to the variation coefficient of each spatial pixel;
the statistical same-distribution pixel sample acquisition module is used for extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence;
the filtering module is used for carrying out spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph;
and the threshold segmentation module is used for performing threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
The embodiment of the invention has the following beneficial effects:
the embodiment of the invention provides a change detection method and a device based on a time sequence SAR intensity image variation coefficient.
Drawings
Fig. 1 is a schematic flow chart of a variation detection method based on a time series SAR intensity image variation coefficient according to an embodiment of the present invention.
FIG. 2 is a variation coefficient diagram according to an embodiment of the present invention.
FIG. 3 is a diagram showing the results of the method for detecting the real surface variations and different variations of the region to be detected according to the present invention.
Fig. 4 is a schematic structural diagram of a change detection apparatus based on a variation coefficient of an SAR intensity image according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a method for detecting a change of a variation coefficient of an image based on a time series SAR intensity, which at least includes the following steps:
s101, acquiring SAR intensity images of a plurality of time nodes of a to-be-detected area within a preset time length to obtain a time sequence SAR intensity image sequence.
Step S102: and calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient graph according to the variation coefficient of each spatial pixel.
Step S103: and extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence.
Step S104: and carrying out spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph.
Step S105: and carrying out threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
For step S101, in a preferred embodiment, the acquiring the SAR intensity images of a plurality of time nodes of the to-be-detected region within a preset time period specifically includes:
the SAR image data of a plurality of time nodes of the region to be detected in a preset time period are collected, then all the SAR image data are registered, and SAR intensity images of the plurality of time nodes are generated.
Schematically, in order to qualitatively and quantitatively analyze the effectiveness of the change detection method based on the time sequence SAR image variation coefficient, the SAR image data of the area to be detected is acquired at a plurality of time nodes within a preset time duration by adopting an X-band German TerrraSAR-X radar satellite based on a single polarization acquisition mode, and then the time sequence SAR intensity image sequence is obtained by registering by using a traditional cross-correlation maximization method. It should be noted that the preset time length and the number of time nodes in the preset time length may be adaptively adjusted according to actual conditions, and in addition, other types of satellite-borne and airborne data acquisition may also be adopted.
Schematically, the characteristic table of the SAR image data acquired by the present invention is shown in table 1:
Figure BDA0003175367990000061
TABLE 1
For step S102: in a preferred embodiment, before calculating a variation coefficient of each spatial pixel according to an intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, the method further includes: and detecting and removing outlier samples in the intensity time sequence samples of each spatial pixel point through outliers.
The method for rejecting outlier spotting in intensity time sequence samples of one spatial pixel point comprises the following steps: sequencing the strength time sequence samples of the space pixel points according to time sequence to obtain ordered samples;
obtaining E samples to be tested from the ordered samples, and then testing the E sample to be tested according to a preset hypothesis testing formula;
if an invalid hypothesis is obtained under a preset confidence level, determining that the 1 st to E th samples to be detected are all outlier samples, then removing the 1 st to E th samples to be detected, and otherwise, detecting the E-1 st samples to be detected; wherein E is a non-zero natural number. The outlier detector is used to remove extraneous pixels due to the problem of robustness of the coefficient of variation to outliers.
Preferably, after the outlier sample in the intensity time sequence sample of the spatial pixel is eliminated, the method further includes: and checking the remaining non-outlier samples in the time sequence sample through single sample AD (analog-to-digital) checking, and if the zero-flipping hypothesis is carried out under a preset second confidence level, adding the rejected outlier samples into the time sequence sample again.
Preferably, the coefficient of variation of the spatial pixel is calculated by the following method:
calculating the sample mean value and the standard deviation of the intensity time sequence sample of the space pixel point; and taking the quotient of the standard deviation and the sample mean value as the coefficient of variation of the spatial pixel point.
In this embodiment, to solve the problem that the variation coefficient has no robustness to the outlier sample, the outlier detector is used to remove irrelevant pixels, taking a spatial pixel point P corresponding to a position point in the region to be detected (in the present invention, all pixel points corresponding to a position point in the region to be detected in each SAR intensity image are the same spatial pixel point, and all pixel points corresponding to a position point in the region to be detected in each SAR intensity image form an intensity time sequence sample corresponding to the spatial pixel point) as an example, and a sequential detector aiming at exponential statistical distribution is adopted to perform spatial image processing on the spatial imageIntensity time sequence sample I (P) ═ I of prime point Pt(p)}1≤t≤NAnd (3) arranging according to the time sequence to obtain ordered samples: i '(P) = { I't(P)}1≤t≤N(ii) a N is the total number of samples (i.e. the number of acquired images), t is the sample number (i.e. the t-th time node), It(P) represents the SAR intensity of pixel point P in the t-th time node space. Obtaining E samples to be tested from the ordered samples, preferably E is 5% by N; firstly, carrying out hypothesis testing on each sample to be tested by adopting the following hypothesis testing formula:
Figure BDA0003175367990000071
Figure BDA0003175367990000072
Figure BDA0003175367990000073
the test is performed, and if an invalid hypothesis is obtained at a preset confidence level α (preferably, α ═ 5%), then the 1 st to E th samples (i.e., all samples to be tested) are determined as outlier samples, and then culling is performed. Otherwise, the E-1 sample to be tested is tested, if the invalid hypothesis is obtained under the preset confidence level alpha, the 1 st sample to the E-1 sample are determined as outlier samples and then removed, otherwise, the E-2 sample is tested, and the like.
In the formula, k is the serial number k belonging to [1, E ] of the sample to be detected]N is the total number of samples, Λk(p) is the test statistic, I ', of the kth sample to be tested of the spatial pixel point p'N+1-k(p) is SAR intensity, l 'of the N +1-k th sample of the ordered samples't(p) the SAR intensity of the t-th sample of the ordered sample; o iskIs a critical value of hypothesis testing;
after outlier samples in the ordered samples are removed by the method, non-outlier samples reserved in the ordered samples are checked by a single sample AD test method. If the null hypothesis is rolled over at a given significance level, the non-outlier samples remaining in the ordered sample are considered not to obey the exponential distribution, and the outlier samples that have been rejected are re-added to the ordered sample.
Then, after outlier samples in the intensity time sequence samples of all spatial pixels in the time sequence SAR intensity image sequence are removed in the above mode, the variation coefficient of each spatial pixel is calculated, and then a variation coefficient graph is generated.
Specifically, the coefficient of variation of each spatial pixel is calculated by the following formula:
Figure BDA0003175367990000081
wherein the content of the first and second substances,
Figure BDA0003175367990000082
cv (p) is the coefficient of variation of the spatial pixel p,
Figure BDA0003175367990000083
to remove the standard deviation of the temporal samples of spatial pixel p after outlier samples,
Figure BDA0003175367990000084
the mean value of the time sequence samples of the space pixel point p after the outlier samples are removed;
Figure BDA0003175367990000085
the sample is the t-th sample in the time sequence sample of the space pixel point p after the outlier sample is removed; it should be noted that, if the operation of removing outlier samples is not performed and the variation coefficient of spatial pixel points is directly calculated, the variance coefficient in the formula can be calculated
Figure BDA0003175367990000086
Is replaced by ItAnd (p) calculating the variation coefficient of each spatial pixel.
And calculating the variation coefficient of each spatial pixel point according to the method to generate a variation coefficient graph.
For step S103, in a preferred embodiment, a statistical homographic pixel sample corresponding to each of the spatial pixels is extracted through BWS hypothesis test.
Specifically, for the intensity time sequence samples i (p) and i (q) of the spatial pixel point p and the spatial pixel point q, the following statistics are adopted to compare the distribution FpAnd Fq：
Figure BDA0003175367990000091
In the above formula, Gt(p) in the intensity time series sample I (p), the intensity value is less than or equal to the reference sample ItNumber of samples of (p), Ht(p) in the intensity time series samples I (q), the intensity value is less than or equal to the reference sample It(q) number of samples. If the B value is smaller than the critical threshold value of the BWS test, the BWS test determines that the two spatial pixel points are taken from the same statistical distribution, the confidence level (namely, the second confidence level) of the BWS test is also set to be 5%, the critical threshold value of the BWS test can refer to a public table, then all the statistical homodistribution pixel points directly or indirectly connected with the spatial pixel points p are extracted, the statistical homodistribution pixel samples of the spatial pixel points p are obtained, and the statistical homodistribution pixel samples of all the spatial pixel points are obtained according to the method.
For step S104, taking the spatial pixel p as an example, if the statistical identically distributed pixel sample of the spatial pixel p is Ω, and the sample includes L pixels, then after filtering is performed according to the statistical identically distributed pixel sample, the coefficient of variation cv of the spatial pixel p after filtering is performed2(p) is:
Figure BDA0003175367990000092
wherein J is a pixel point in the statistical same-distribution pixel sample of the spatial pixel point p, and cv (J) is a coefficient of variation of the spatial pixel point J. And calculating the coefficient of variation of all the spatial pixels after filtering according to the formula, and then generating a filtered coefficient of variation graph according to the coefficient of variation of all the spatial pixels after filtering. An exemplary filtered coefficient of variation plot generated in the present invention is shown in fig. 2.
In step S105, a threshold is obtained from the filtered variation coefficient map according to the KI criterion, threshold segmentation is performed, and a change detection result of the region to be detected is extracted. Fig. 3(a) is a result schematic diagram of a conventional log-ratio method, which only uses the first and last SAR images for detection, fig. 3(b) is a result schematic diagram of a generalized likelihood ratio method, which directly uses a time sequence SAR image data set for detection, fig. 3(c) is a detection result schematic diagram of a change detection method based on a time sequence SAR intensity image variation coefficient, and fig. 3(d) is a schematic diagram of a real change of the earth surface of a region to be detected.
In order to quantitatively analyze the effect of the variation detection method based on the variation coefficient of the time series SAR intensity image disclosed by the invention in FIG. 3(c), the accuracy of the detection result is evaluated by adopting overall error, false alarm rate and false alarm rate indexes, and compared with the methods corresponding to FIG. 3(a) and FIG. 3(b), the comparison result is shown in Table 2:
Figure BDA0003175367990000101
TABLE 2
As can be seen from table 2 and fig. 3, compared with other methods, the change detection method based on the time series SAR intensity image variation coefficient provided by the present invention shows higher detection accuracy in three indexes of the overall error, the false alarm rate and the false alarm rate; in the detection result obtained by the method, the details of the ground features are better reserved, and meanwhile, the change area of the ground surface is better extracted and is closer to the real change of the ground surface.
As shown in fig. 4, on the basis of the above embodiment of the method, the present invention correspondingly provides an embodiment of an apparatus;
the invention provides a change detection device based on a time sequence SAR intensity image variation coefficient, which comprises a data acquisition module, a variation coefficient image generation module, a statistical same-distribution pixel sample acquisition module, a filtering module and a threshold segmentation module, wherein the data acquisition module is used for acquiring a variation coefficient of a time sequence SAR intensity image;
the data acquisition module is used for acquiring SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence;
the variation coefficient map generation module is used for calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient map according to the variation coefficient of each spatial pixel;
the statistical same-distribution pixel sample acquisition module is used for extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence;
the filtering module is used for carrying out spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph;
and the threshold segmentation module is used for performing threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
It should be noted that the foregoing apparatus embodiments correspond to the method embodiments of the present invention, and the method for detecting a change based on a time series SAR intensity image variation coefficient according to any embodiment of the present invention can be implemented
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.

Claims (8)

1. A change detection method based on time sequence SAR intensity image variation coefficient is characterized by comprising the following steps:
acquiring SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence;
calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient graph according to the variation coefficient of each spatial pixel;
extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence;
performing spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph;
and carrying out threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
2. The method as claimed in claim 1, wherein before calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, the method further comprises:
and detecting and removing outlier samples in the intensity time sequence samples of each spatial pixel point through outliers.
3. The method for detecting changes based on time series SAR intensity image variation coefficients as claimed in claim 2, wherein said obtaining the SAR intensity images of a plurality of time nodes of the to-be-detected region within a preset time period specifically comprises:
the SAR image data of a plurality of time nodes of the region to be detected in a preset time period are collected, then all the SAR image data are registered, and SAR intensity images of the plurality of time nodes are generated.
4. The method of claim 3, wherein outlier samples in intensity time series samples of a spatial pixel are rejected by:
sequencing the strength time sequence samples of the space pixel points according to time sequence to obtain ordered samples;
obtaining E samples to be tested from the ordered samples, and then testing the E sample to be tested according to a preset hypothesis testing formula;
if an invalid hypothesis is obtained under a preset confidence level, determining that the 1 st to E th samples to be detected are all outlier samples, then removing the 1 st to E th samples to be detected, and otherwise, detecting the E-1 st samples to be detected; wherein E is a non-zero natural number.
5. The method as claimed in claim 4, wherein after removing outlier samples in the intensity time series samples of the spatial pixel, the method further comprises:
and checking the remaining non-outlier samples in the time sequence sample through single sample AD (analog-to-digital) checking, and if the zero-flipping hypothesis is carried out under a preset second confidence level, adding the rejected outlier samples into the time sequence sample again.
6. The method of claim 4, wherein the variation coefficient of the spatial pixel is calculated by:
calculating the sample mean value and the standard deviation of the intensity time sequence sample of the space pixel point;
and taking the quotient of the standard deviation and the sample mean value as the coefficient of variation of the spatial pixel point.
7. The method as claimed in claim 6, wherein the statistical same-distribution pixel samples corresponding to each spatial pixel point are extracted through BWS hypothesis testing.
8. A change detection device based on time sequence SAR intensity image variation coefficient is characterized by comprising: the device comprises a data acquisition module, a variation coefficient map generation module, a statistical same-distribution pixel sample acquisition module, a filtering module and a threshold segmentation module;
the data acquisition module is used for acquiring SAR intensity images of a plurality of time nodes of a to-be-detected region within a preset time length to obtain a time sequence SAR intensity image sequence;
the variation coefficient map generation module is used for calculating the variation coefficient of each spatial pixel according to the intensity time sequence sample of each spatial pixel in the time sequence SAR intensity image sequence, and then generating a variation coefficient map according to the variation coefficient of each spatial pixel;
the statistical same-distribution pixel sample acquisition module is used for extracting a statistical same-distribution pixel sample corresponding to each spatial pixel point according to the time sequence SAR intensity image sequence;
the filtering module is used for carrying out spatial filtering on the variation coefficient graph according to the statistical same-distribution pixel samples of each spatial pixel point to obtain a filtered variation coefficient graph;
and the threshold segmentation module is used for performing threshold segmentation on the filtered variation coefficient graph to obtain a change detection result of the to-be-detected region.
CN202110830747.XA 2021-07-22 2021-07-22 Time sequence SAR intensity image variation coefficient-based change detection method and device Pending CN113610783A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110830747.XA CN113610783A (en) 2021-07-22 2021-07-22 Time sequence SAR intensity image variation coefficient-based change detection method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110830747.XA CN113610783A (en) 2021-07-22 2021-07-22 Time sequence SAR intensity image variation coefficient-based change detection method and device

Publications (1)

Publication Number Publication Date
CN113610783A true CN113610783A (en) 2021-11-05

Family

ID=78305145

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110830747.XA Pending CN113610783A (en) 2021-07-22 2021-07-22 Time sequence SAR intensity image variation coefficient-based change detection method and device

Country Status (1)

Country Link
CN (1) CN113610783A (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108062767A (en) * 2018-02-11 2018-05-22 河海大学 Statistics based on sequential SAR image is the same as distribution space pixel selecting method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108062767A (en) * 2018-02-11 2018-05-22 河海大学 Statistics based on sequential SAR image is the same as distribution space pixel selecting method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MI JIANG 等: "Delineation of built-up land change from SAR stack by analysing the coefficient of variation", 《ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING》 *

Similar Documents

Publication Publication Date Title
Rosin Thresholding for change detection
Lin et al. Radiometric normalization and cloud detection of optical satellite images using invariant pixels
CN110703244A (en) Method and device for identifying urban water body based on remote sensing data
CN115236655A (en) Landslide identification method, system, equipment and medium based on fully-polarized SAR
Jaiswal et al. Prediction of tropical cyclogenesis using scatterometer data
Pasquarella et al. Comprehensive quality assessment of optical satellite imagery using weakly supervised video learning
CN112270675B (en) Urban waterlogging area detection method based on polarized radar remote sensing image
CN110688923A (en) Sentinel 1A SAR data-based urban inland inundation risk area extraction method
US7058511B2 (en) Sub-visible cloud cover assessment: VNIR-SWIR
CN106960443B (en) Unsupervised change detection method and device based on full-polarization time sequence SAR image
CN113610783A (en) Time sequence SAR intensity image variation coefficient-based change detection method and device
Zhao et al. An automatic SAR-based change detection method for generating large-scale flood data records: the UK as a test case
Reulke et al. SNR evaluation of the RapidEye space-borne cameras
Oller et al. Correlation and similarity measures for SAR image matching
Tahraoui et al. Land change detection using multivariate alteration detection and Chi squared test thresholding
Lee et al. A Signal-Symbol Approach to Change Detection.
Lin et al. Hybrid analysis for SAR change detection based on time series data
Zou et al. New building detection using SAR images with different resolutions
Chen et al. Urban damage estimation using statistical processing of satellite images: 2003 bam, iran earthquake
Lin et al. On the improvement of ASCAT wind data assimilation in global NWP
CN117541611A (en) SAR image edge detection device and method based on fractional order cumulant spectrum
Hedir et al. Radar echoes classification based on local descriptor
US20230260248A1 (en) Feces color-based remote sensing estimation method and apparatus for antarctic bird population size
Wu et al. Change detection and analysis with radarsat-1 SAR image
CN111552919B (en) Signal-to-noise ratio evaluation method for linear array push-broom remote sensing imager

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20211105

RJ01 Rejection of invention patent application after publication