CN113177887A - Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis - Google Patents
Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis Download PDFInfo
- Publication number
- CN113177887A CN113177887A CN202110411206.3A CN202110411206A CN113177887A CN 113177887 A CN113177887 A CN 113177887A CN 202110411206 A CN202110411206 A CN 202110411206A CN 113177887 A CN113177887 A CN 113177887A
- Authority
- CN
- China
- Prior art keywords
- noise
- wavelet packet
- local mean
- insar
- frequency analysis
- 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
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000004458 analytical method Methods 0.000 title claims abstract description 22
- 230000009466 transformation Effects 0.000 claims abstract description 31
- 230000000694 effects Effects 0.000 claims abstract description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 11
- 239000000654 additive Substances 0.000 claims description 9
- 230000000996 additive effect Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005305 interferometry Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 229940124913 IPOL Drugs 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis, which combines the multi-scale time-frequency analysis characteristic of wavelet packet transformation and the good filtering effect of non-local mean filtering to denoise an InSAR phase signal to smooth noise, wherein the method comprises sine-cosine transformation, wavelet transformation, non-local mean filtering, inverse wavelet transformation and the like.
Description
Technical Field
The invention belongs to the technical field of satellite remote sensing application, and particularly relates to a fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis.
Technical Field
The synthetic aperture radar interferometry is a very common remote sensing observation means, is characterized by large range, all weather and much time, and is almost not interfered by cloud layers due to the penetrability of microwaves, so that the synthetic aperture radar interferometry plays an important role in ground surface deformation monitoring, such as post-earthquake monitoring, volcanic motion, landslide monitoring, frozen soil in glaciers, underground water exploitation and the like, and provides good technical support for overall planning and disaster prevention of governments. The InSAR technology acquires earth surface deformation for a period of time by utilizing interference phase signals between radar images in the same region but different time, acquiring a differential phase related to the deformation signal through the difference of interference phases after image registration, performing phase unwrapping on the differential phase to acquire a relative deformation signal, and finally converting the deformation of a line of sight (LOS) into deformation information in a vertical direction, so that the phase signal is particularly important.
However, in the signal transmission process, the SAR signal is inevitably affected by various noises, such as various phase loss coherence phenomena, both temporally and spatially, such as the delay effect of the atmosphere on the signal, some thermal noises of the system, and so on, and thus it is important to remove the noise of the phase signal. The signal filtering is a common denoising means, the signal-to-noise ratio of the phase signal can be improved through filtering processing, and the phase unwrapping is performed on the filtered interference phase, so that the interference measurement precision is improved. During the research process, a plurality of researches are carried out on the filtration of InSAR signals. The literature data (Wanlu, Wang Guannan, Ma Liuping. InSAR interference image filtering method based on wavelet transformation and median filtering [ J ] the Proc. Megaging, 2005, 34 (2): 108-; a non-local mean de-noising algorithm based on wavelet packet transformation for common images is provided for literature data (Longjunyu, lovely people, a non-local mean de-noising method [ J ] based on wavelet packet transformation, computer and modernization, 2013, 000 (011): 13-16.). Wavelet packet transformation and rapid non-local mean filtering methods are two common algorithms, how to process the InSAR image by combining the advantages of the two algorithms is a direction worth researching, and no report that a similar method is adopted to process the InSAR image is found after the prior art is retrieved. Therefore, the invention provides an InSAR phase denoising method based on wavelet packet transformation and a fast non-local mean filtering algorithm.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis, the algorithm combines the multi-scale time-frequency analysis characteristic of wavelet packet transformation and the high-precision characteristic of a non-local mean filtering algorithm to filter the phase noise of the InSAR phase with fringe characteristic, and the interference measurement precision is improved.
In order to achieve the purpose, the invention adopts the following technical scheme:
a fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis combines the multi-scale time-frequency analysis characteristic of wavelet packet transformation and the good filtering effect of non-local mean filtering to denoise an InSAR phase signal to smooth noise, and specifically comprises the following steps:
s1: performing sine and cosine transformation processing on the InSAR phase signal;
s2: performing wavelet packet transformation on the InSAR phase signal subjected to sine and cosine transformation in the step S1 to obtain a wavelet packet coefficient containing noise, converting the phase signal into the wavelet packet coefficient by utilizing the linear characteristic of the wavelet packet transformation and the multi-time-frequency analysis characteristic, wherein the noise of the phase signal subjected to the wavelet packet transformation is Gaussian white noise with the average value of 0;
s3: carrying out non-local mean filtering on the wavelet packet coefficient containing the noise obtained in the step S2 to obtain a filtered wavelet packet coefficient, and removing InSAR phase noise by using the characteristics of white Gaussian noise;
s4: and carrying out inverse wavelet packet transformation on the filtered wavelet packet coefficient obtained in the step S3 to obtain a filtered InSAR phase signal.
Further, the sine-cosine transform processing in step S1 is performed in the complex plane, and the sine-cosine transform is used to avoid the phase jump problem of the InSAR phase in the complex domain.
Further, the formula of the sine-cosine transform in step S1 is:
wherein phizRepresenting the true phase signal.
Further, the InSAR phase signal noise after the wavelet packet transform processing in step S2 is an additive noise model, and satisfies the following formula:
wherein N iscDenotes the mean value of cos (v), vcAnd vsDenotes the additive noise with a mean value of zero and v denotes the phase signal noise term.
Further, the noisy wavelet packet coefficient in step S2 contains InSAR phase noise and satisfies the following formula:
wherein, WPT2D{. represents the wavelet packet coefficient containing noise, 2iThe scale of the wavelet is represented by,
in order to be a noise-free phase signal,
representing zero mean noise.
Further, the InSAR phase noise included in the filtered wavelet packet coefficient in step S3 is additive noise, and satisfies the following formula:
x(i，j)＝s(i，j)+n(i，j)；
wherein x (i, j) represents the filtered wavelet packet coefficient, s (i, j) represents the real signal, and n (i, j) represents the noise signal.
Further, the non-local mean filtering algorithm in step S3 is accelerated by using an integral image method to reduce the complexity of the algorithm.
Further, the filtered InSAR phase signal value is-pi.
Compared with the prior art, the invention has the following beneficial effects:
(1) the invention creatively combines wavelet packet transformation with a non-local mean filtering algorithm for InSAR phase filtering processing, and avoids the jumping problem of interference phase signals by carrying out multi-scale time-frequency analysis on real part signals and imaginary part signals; the phase noise is better inhibited and the signal detail information is kept through the non-local mean filtering algorithm, the whole filtering method is simple and easy to implement, the denoising effect is good, and the denoising precision of the traditional InSAR phase denoising algorithm is improved.
(2) Because the complexity of the traditional non-local mean filtering algorithm is higher on the basis of the non-local mean filtering algorithm, the integral image-based acceleration algorithm is adopted to improve the further calculation efficiency.
(3) Aiming at the filtering method provided by the invention, the denoising performance of different filtering methods is analyzed through comparing the simulation data with the actual data, the effectiveness and superiority of the method are verified, and the method has better application value.
Drawings
Fig. 1 is an algorithm diagram of wavelet packet transformation in the present invention.
Fig. 2 is an algorithm schematic diagram of non-local mean filtering in the present invention.
FIG. 3 is a graph showing the effects of embodiment 1 of the present invention. The graphs respectively represent simulation data, a simulation data noise adding graph, a filtered simulation graph, real data and a real data filtering graph.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings and examples. It should be noted that the following examples are only for illustrating the present invention and are not intended to limit the scope of the present invention.
Example 1
Referring to fig. 1, wavelet packet decomposition may provide a more refined method, so that time-frequency plane decomposition is more refined, a concept of optimal basis selection is introduced while high-frequency signals are decomposed again, and signal analysis capability is improved. The invention utilizes the advantages of wavelet packet transformation in multi-scale time-frequency analysis as a basis to carry out interference phase signal denoising, and carries out signal filtering on InSAR phase signals through a fast non-local mean value filtering algorithm after the wavelet packet transformation. Where s represents the original signal, a, D represent the low and high frequency signals, respectively, and each row represents the number of layers of wavelet packet transforms.
Referring to fig. 2, the green portion represents the target pixel, with x at the center; the red part represents a neighborhood block, the center of the red part is y, a search frame containing a target pixel and the neighborhood block, the gray part of the neighborhood block is the target pixel, the center of the search frame is also x, the neighborhood block slides in the search frame, the weight is determined according to the similarity with the target region, and finally the purpose of non-local mean filtering is achieved, so that noise is smoothed.
The invention provides a rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis, which combines the multi-scale time-frequency analysis characteristic of wavelet packet transformation and the good filtering effect of non-local mean filtering to denoise an InSAR phase signal to smooth noise, and specifically comprises the following steps:
s1: performing sine-cosine transform processing on the InSAR phase signal in a complex plane; the formula of the sine-cosine transform is as follows:
wherein phizRepresenting the true phase signal.
S2: performing wavelet packet transformation on the InSAR phase signal subjected to sine and cosine transformation in the step S1 to obtain a wavelet packet coefficient containing noise; wherein wavelet packets are transformed into prior art, the present invention is not specifically described, and the specific implementation can be found in document 1(Zha X, Fu R, Dai Z, et al. noise reduction in interference mapping and Wiener filtering [ J ]. IEEE Geoscience and remove Sensing Letters, 2008, 5 (3): 404-. 408.).
The InSAR phase signal noise after wavelet packet transformation processing in the step of S2 is an additive noise model and satisfies the following formula:
wherein N iscDenotes the mean value of cos (v), vcAnd vsDenotes the additive noise with a mean value of zero and v denotes the phase signal noise term.
The wavelet packet coefficient containing noise in the step S2 contains InSAR phase noise, and satisfies the following formula:
wherein, WPT2D{. represents the wavelet packet coefficient containing noise, 2iThe scale of the wavelet is represented by,
in order to be a noise-free phase signal,
representing zero mean noise.
S3: carrying out rapid non-local mean filtering on the wavelet packet coefficient containing the noise obtained in the step S2 to obtain a filtered wavelet packet coefficient, wherein the non-local mean filtering adopts an integral image method to accelerate a classic non-local mean filtering algorithm, and filtering parameters realize self-adaptation by evaluating the noise level; the integral Image method belongs to the prior art, the invention is not specifically described, and the specific implementation can be found in literature 2 (FRONTJ. parameter-Free Fast Pixel Non-Local Means removing [ J/OL ]. Image Processing On Line, 2014, 4: 300-326. DOI: 10.5201/ipol.2014.120.).
The InSAR phase noise contained in the filtered wavelet packet coefficient is additive noise and meets the following formula:
x(i，j)＝s(i，j)+n(i，j)；
wherein x (i, j) represents the filtered wavelet packet coefficient, s (i, j) represents the real signal, and n (i, j) represents the noise signal.
S4: and performing inverse wavelet packet transformation (the specific implementation can be shown in literature data 1) on the filtered wavelet packet coefficient obtained in the step S3 to obtain a filtered InSAR phase signal, wherein the value of the filtered InSAR phase signal is-pi to pi.
The data were analyzed using the algorithm described in the above example, and the results are shown in fig. 3.
The above description is only a preferred embodiment of the invention, and should not be taken as limiting the invention, and any modifications, equivalent substitutions and improvements made within the spirit and design principles of the invention should be included within the scope of the invention.
Claims (8)
1. A fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis is characterized in that the method combines the multi-scale time-frequency analysis characteristic of wavelet packet transformation and the good filtering effect of non-local mean filtering to denoise an InSAR phase signal to smooth noise, and specifically comprises the following steps:
s1: performing sine and cosine transformation processing on the InSAR phase signal;
s2: performing wavelet packet transformation on the InSAR phase signal subjected to sine and cosine transformation in the step S1 to obtain a wavelet packet coefficient containing noise;
s3: carrying out rapid non-local mean filtering on the wavelet packet coefficient containing the noise obtained in the step of S2 to obtain a filtered wavelet packet coefficient;
s4: and carrying out inverse wavelet packet transformation on the filtered wavelet packet coefficient obtained in the step S3 to obtain a filtered InSAR phase signal.
2. The method according to claim 1, wherein the n-cosine transform processing in step S1 is performed in a complex plane.
3. The fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis according to claim 2, wherein the formula of the sine and cosine transform in step S1 is:
wherein phizRepresenting the true phase signal.
4. The fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis according to claim 1, wherein InSAR phase signal noise after wavelet packet transform processing in step S2 is an additive noise model and satisfies the following formula:
wherein N iscDenotes the mean value of cos (v), vcAnd vsDenotes the additive noise with a mean value of zero and v denotes the phase signal noise term.
5. The fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis according to claim 4, wherein the wavelet packet coefficient containing noise in step S2 contains InSAR phase noise and satisfies the following formula:
wherein, WPT2D{. represents the wavelet packet coefficient containing noise, 2iThe scale of the wavelet is represented by,
in order to be a noise-free phase signal,
representing zero mean noise.
6. The method according to claim 1, wherein the InSAR phase noise contained in the filtered wavelet packet coefficients in step S3 is additive noise and satisfies the following formula:
x(i,j)＝s(i,j)+n(i,j)；
wherein x (i, j) represents the filtered wavelet packet coefficient, s (i, j) represents the real signal, and n (i, j) represents the noise signal.
7. The fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis as claimed in claim 1, wherein the non-local mean filtering algorithm in step S3 is accelerated by means of integral image to reduce algorithm complexity.
8. The fast non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis according to claim 1, characterized in that the filtered InSAR phase signal value is-pi.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110411206.3A CN113177887A (en) | 2021-04-16 | 2021-04-16 | Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110411206.3A CN113177887A (en) | 2021-04-16 | 2021-04-16 | Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113177887A true CN113177887A (en) | 2021-07-27 |
Family
ID=76923969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110411206.3A Pending CN113177887A (en) | 2021-04-16 | 2021-04-16 | Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113177887A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459633A (en) * | 2014-12-01 | 2015-03-25 | 中国科学院电子学研究所 | Wavelet domain InSAR interferometric phase filtering method combined with local frequency estimation |
| CN109598680A (en) * | 2018-10-19 | 2019-04-09 | 浙江工业大学 | Shearing wave conversion medicine CT image denoising method based on quick non-local mean and TV-L1 model |
-
2021
- 2021-04-16 CN CN202110411206.3A patent/CN113177887A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459633A (en) * | 2014-12-01 | 2015-03-25 | 中国科学院电子学研究所 | Wavelet domain InSAR interferometric phase filtering method combined with local frequency estimation |
| CN109598680A (en) * | 2018-10-19 | 2019-04-09 | 浙江工业大学 | Shearing wave conversion medicine CT image denoising method based on quick non-local mean and TV-L1 model |
Non-Patent Citations (3)
| Title |
|---|
| 曾峦 等: "《侦察图像获取与融合技术》", 31 May 2015 * |
| 王志勇 等: "InSAR干涉条纹图去噪方法的研究", 《测绘科学》 * |
| 龙钧宇 等: "基于小波包变换的非局部均值去噪方法", 《计算机与现代化》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tania et al. | A comparative study of various image filtering techniques for removing various noisy pixels in aerial image | |
| CN100510773C (en) | Single satellite remote sensing image small target super resolution ratio reconstruction method | |
| WO2016086699A1 (en) | Wavelet domain insar interferometric phase filtering method in combination with local frequency estimation | |
| CN103077508B (en) | Transform domain non local and minimum mean square error-based SAR (Synthetic Aperture Radar) image denoising method | |
| CN108961181B (en) | Shearlet transform-based ground penetrating radar image denoising method | |
| Zha et al. | Noise reduction in interferograms using the wavelet packet transform and wiener filtering | |
| CN104182941A (en) | Hyperspectral image band noise removing method | |
| CN110852969B (en) | Star map wavelet denoising method based on local abnormal factor | |
| Li et al. | Contourlet based seismic reflection data non-local noise suppression | |
| CN104036461A (en) | Infrared complicated background inhibiting method based on combined filtering | |
| Zhang et al. | A reverberation noise suppression method of sonar image based on shearlet transform | |
| CN105976340A (en) | Improved spin filtering algorithm based on wavelet decomposition | |
| CN112907458A (en) | F-XY domain improved non-local mean de-noising method and equipment for seismic exploration | |
| CN116385312A (en) | Low-illumination image denoising method based on phase correlation | |
| CN113177887A (en) | Rapid non-local mean InSAR phase filtering method based on multi-scale time-frequency analysis | |
| CN106845448B (en) | Infrared weak and small target detection method based on non-negative constraint 2D variational modal decomposition | |
| Menaka et al. | Change detection in deforestation using high resolution satellite image with Haar wavelet transforms | |
| Xu et al. | A novel SAR imaging method based on morphological component analysis | |
| Zhang et al. | Bayesian-based speckle suppression for SAR image using contourlet transform | |
| Saied et al. | Digital Elevation Model Enhancement using CNN-Based Despeckled SAR Images | |
| Wang et al. | Study on denoising and enhancement method in SAR image based on wavelet packet and fuzzy set | |
| CN117274094B (en) | Deconvolution reconstruction method for improving wearable ultrasonic imaging quality | |
| Fazel et al. | Speckle reduction of SAR images using curvelet and wavelet transforms based on spatial features characteristics | |
| Choi et al. | Effect of pre-processing on satellite image fusion | |
| Aiazzi et al. | An experimental setup for multiresolution despeckling of COSMO-SkyMed image products |
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 |