CN109658340A - The SAR image rapid denoising method saved based on RSVD and histogram - Google Patents
The SAR image rapid denoising method saved based on RSVD and histogram Download PDFInfo
- Publication number
- CN109658340A CN109658340A CN201811206810.7A CN201811206810A CN109658340A CN 109658340 A CN109658340 A CN 109658340A CN 201811206810 A CN201811206810 A CN 201811206810A CN 109658340 A CN109658340 A CN 109658340A
- Authority
- CN
- China
- Prior art keywords
- image
- histogram
- matrix
- denoising
- sar image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 239000011159 matrix material Substances 0.000 claims abstract description 43
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 15
- 239000000654 additive Substances 0.000 claims abstract description 7
- 230000000996 additive effect Effects 0.000 claims abstract description 7
- 238000004321 preservation Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 claims description 4
- 230000009466 transformation Effects 0.000 abstract description 5
- 238000013459 approach Methods 0.000 abstract description 2
- 238000007796 conventional method Methods 0.000 abstract 1
- 230000002708 enhancing effect Effects 0.000 abstract 1
- 230000000007 visual effect Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000014759 maintenance of location Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of SAR image rapid denoising methods saved based on RSVD and histogram, this method carries out logarithmic transformation to SAR image first, multiplicative noise is converted to additive noise, then non local similar image Block- matching is carried out, then low-rank matrix is carried out to the low-rank matrix that non local similar image block forms using random singular value decomposition to approach, texture enhancing is carried out to image using the method that histogram of gradients saves again, finally by image block reset, the quick denoising to SAR image is realized.It is on MSTAR database the experimental results showed that, compared with the conventional method, method proposed by the present invention be obviously improved edge keep index while, denoising speed accelerate three times.
Description
Technical Field
The invention belongs to the technical field of radar image processing, and particularly relates to an SAR image fast denoising method based on RSVD and histogram storage.
Background
Synthetic Aperture Radar (SAR) images provide useful information for the fields of remote sensing mapping, ground monitoring, automatic target identification, and the like. But the echo phases of many scattering points in a resolution unit of a radar radiation area are different due to the difference of the distances from the scattering points to the radar. The echoes are added in a coherent mode, so that speckle noise is inevitably generated in the SAR image. Due to the influence of speckle noise, the visual quality of the observed SAR image is reduced, and the edge information and other aspects are weakened. Therefore, removing speckle noise is a key task before subsequent segmentation, detection and classification of images. The SAR image denoising method is characterized in that the SAR image denoising method is a traditional spatial-domain-based filtering algorithm, such as a Lee, Kuan, Frost, Gamma MAP, an enhanced Lee filter, an enhanced Frost filter and the like, denoising is carried out by using local small block information of an image basically, and the problem that texture details are lost due to over-smoothing exists easily. With the improvement of a signal processing method, wavelet transformation is applied to SAR image denoising, but the method cannot effectively express the edge information of an image; then multi-scale analysis methods like the Contourlet transform have emerged. In recent years, better denoising effects are achieved by denoising SAR images of models such as a Markov random field, a Gibbs random field, BLS-GSM (Bayes Least Squares-Gaussian scales hybrids), and the like. In addition, Buads et al apply a non-local model to image denoising, and design a non-local mean (NL-means) denoising method, which has a poor denoising effect in an image edge region. Dabov proposes a three-dimensional Block Matching (BM 3D) algorithm, which has a good denoising effect but high algorithm complexity.
In 2006, Terrence Tao et al proposed a Low Rank Matrix Approximation (LRMA), which was introduced into image denoising, i.e., recovering the original Low Rank matrix from the matrix contaminated by noise. Methods of low rank matrix approximation can generally be divided into two categories: low Rank matrix decomposition (LRMF) and Nuclear Norm Minimization (NNM). Since many similar image blocks exist in the SAR image, the similar blocks have similar structural features and data features, and a matrix formed by the similar blocks can be considered to be approximate low rank, the NNM can be applied to the SAR image denoising.
The disadvantage of the NNM algorithm is that singular values are treated equally in the calculation process, resulting in large deviations. A Spatially Adaptive Iterative Singular Value Thresholding (SAIST) combining image non-local similarity and a low-rank model is proposed in the literature to improve this method. Since the real information of the signal is mainly concentrated on the larger singular value, and the noise is mainly reflected on the small singular value, a Weighted Nuclear Norm Minimization (WNNM) method is proposed, that is, for the large singular value, a small weight is adopted, and for the small singular value, a large weight is adopted, so that the algorithm can better approach the low-rank matrix. However, the WNNM algorithm has a disadvantage that each iteration of the WNNM algorithm needs to be performed with singular value decomposition, thereby consuming a large amount of operation time. Additionally, WNNM can overly smooth texture details. Therefore, it is considered that the singular value decomposition is replaced by RSVD (random singular value decomposition) with a smaller number of iterations, so that the denoising efficiency is improved, and the GHP (gradient histogram preservation) is adopted to perform texture enhancement on the image in the denoising process.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a SAR image fast denoising method based on RSVD and histogram storage, so as to solve the problems of the denoising algorithm in the prior art; the method comprises the steps of firstly carrying out logarithmic transformation on an SAR image, converting multiplicative noise into additive noise, then carrying out non-local similar image block matching, then carrying out low-rank matrix approximation on a low-rank matrix formed by non-local similar image blocks by adopting random singular value decomposition, carrying out texture enhancement on the image by adopting a gradient histogram storage method, and finally resetting the image blocks to realize rapid de-noising of the SAR image. Experimental results on an MSTAR database show that compared with the existing method, the method provided by the invention has the advantages that the denoising speed is increased by three times while the edge retention index is obviously improved.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the invention discloses a rapid SAR image denoising method based on RSVD and histogram preservation, which comprises the following steps:
(1) establishing a speckle noise model of the SAR image, namely converting multiplicative noise into additive noise through logarithmic transformation;
(2) carrying out block matching on the transformed image by using non-local similarity;
(3) decomposing a low-rank matrix composed of non-local similar image blocks by adopting RSVD (remote singular value decomposition), and realizing low-rank matrix approximation;
(4) performing texture enhancement on the SAR image by adopting a gradient histogram storage method;
(5) and resetting the image block to obtain the denoised SAR image.
Further, the step (1) specifically includes: establishing a SAR image multiplicative noise model, namely Y (X, Y) is X (X, Y) N (X, Y), wherein Y (X, Y) is the SAR image added with speckle noise, namely a final observation image; x (X, y) is an original SAR image; n (x, y) is speckle noise; x (X, y) and N (X, y) are mutually independent random processes, speckle noise N (X, y) follows a Gamma distribution with a mean value of 1 and a variance of 1/L, and the probability density function is as follows:wherein, L is an equivalent view, and e is a natural base number; taking logarithms on two sides of Y (X, Y) ═ X (X, Y) N (X, Y) at the same time, then converting the multiplicative noise into the commonly used additive noise: lg (Y (X, Y)) + lg (X, Y)) + lg (N (X, Y)).
Further, the step (2) specifically includes:
(21) dividing an original image into a plurality of image blocks with the same size, and searching non-local similar image blocks according to specified image blocks;
(22) and integrating the non-local similar image blocks into a matrix to form a low-rank matrix.
Further, the step (3) specifically includes:
(31) generating an n multiplied by l-dimensional Gaussian random matrix omega;
(32) multiplying a Gaussian random matrix omega by an m × n original matrix A to be decomposed to construct an m × l sample matrix Y which is A omega;
(33) carrying out QR decomposition on the matrix Y to obtain an m multiplied by l dimension orthogonal matrix Q;
(34) constructing a matrix of dimension Lxn, B ═ QTX A, and finally, to matrix BLine singular value decomposition, i.e. B ═ sa VT；
(35) Let QS ═ U obtain singular value decomposition of A, i.e. A ═ UΛ VT；
Where l is much smaller than the smaller of m and n, S is a matrix of l × l, Λ is a diagonal matrix of l × n, the elements on the diagonal are called singular values, VTIs the transpose of V, which is an n × n square matrix.
Further, the step (4) specifically includes:
(41) first, a gradient histogram h of an original image x is estimatedrTaking the reference gradient as a reference gradient, and taking the updated image close to the reference gradient histogram to the maximum extent as a constraint condition to obtain a result image; wherein, histogram of gradient hrThe solving formula of (2) is as follows:
wherein d is a constant, R (h)x) For regularization terms based on prior information of the gradient histogram of the natural image,is the gradient of the noise n, hy、hx and hgHistograms of the noisy image, the original image and the gradient g, respectively;
(42) to denoise an imageIs approximated by a reference histogram hrThe denoising model stored based on the gradient histogram is as follows:
s.t.hf＝hr
wherein ,for gradient operations, F is an odd function that monotonically increases over (0, + ∞), and hfFor transformed gradient imagesHistogram, hrIs a histogram of the original image, R (x) is a regularization term, λ is a normal number, σn 2μ is a constant for the noise variance.
The invention has the beneficial effects that:
1. the SAR image denoising method can improve the denoising efficiency of the SAR image and reduce the denoising time;
2. the invention improves the peak signal-to-noise ratio of the denoised image;
3. the invention improves the edge retention capability of the denoised image;
4. the invention improves the equivalent visual number of the de-noised image.
Drawings
FIG. 1 is a schematic block diagram of an SAR image denoising method according to the present invention;
fig. 2a is a schematic diagram of a raw 128 × 128SAR image;
FIG. 2b is a diagram of the effect of an original 128 × 128SAR image after SAIST denoising;
FIG. 2c is a WNNM denoised effect diagram of an original 128 x 128SAR image;
FIG. 2d is a graph of the effect of the original 128 × 128SAR image after denoising by the method of the present invention.
Detailed Description
In order to facilitate understanding of those skilled in the art, the present invention will be further described with reference to the following examples and drawings, which are not intended to limit the present invention.
Referring to fig. 1, the method for quickly denoising an SAR image based on RSVD and histogram preservation of the present invention includes the following steps:
(1) establishing a speckle noise model of the SAR image, namely converting multiplicative noise into additive noise through logarithmic transformation;
(2) carrying out block matching on the transformed image by using non-local similarity;
(3) decomposing a low-rank matrix composed of non-local similar image blocks by adopting RSVD (remote singular value decomposition), and realizing low-rank matrix approximation;
(4) introducing a gradient histogram storage method into an iteration regular term for SAR image denoising, and realizing texture enhancement of the SAR image;
(5) and finally, resetting the image block to obtain the denoised SAR image.
The SAR images of T72_ SN132 (main warfare tank), BMP _ SN9563 (armored car) and BTR70_ SNC71 (armored car) in data sets disclosed by MSTAR projects jointly subsidized by the United states State defense advanced research planning office and the United states air force research laboratory are adopted as experimental data.
Firstly, Gaussian white noise with the variance of 70 is added to the original SAR image in FIG. 2a, then the fast SAR image denoising algorithm provided by the invention is used for denoising by a WNNM method and an SAIST method respectively, and the denoising results are compared. Fig. 2b, 2c, and 2d are denoising results of the SAR image of T72_ SN132 denoised by three algorithms.
As can be seen from fig. 2b, fig. 2c, and fig. 2d, compared with WNNM algorithm and SAIST algorithm, the method of the present invention better retains the edge and texture details of the target while denoising, and has better visual effect.
In order to describe the performance of the denoising algorithm more accurately, Equivalent visual numbers (ENL), Edge Preserving Indexes (EPI), and Peak Signal to noise Ratio (PSNR) are selected to analyze the denoising performance of the three algorithms. The larger the ENL is, the better the visual effect of the denoised image is; the larger the EPI, the stronger the edge-preserving capability of the algorithm; the larger the PSNR, the stronger the denoising capability of the representation algorithm. Table 1 lists the results of 3 objective evaluation indexes of the SAR images of three targets after denoising by the three denoising algorithms. Table 1 is as follows:
TABLE 1
From each evaluation index of table 1, the PSNR of the image denoised by the method of the present invention is slightly improved compared with other two algorithms, which indicates that the denoising ability of the method of the present invention is relatively strong; the ENL of the image denoised by the method is improved to a certain extent compared with other two algorithms, which shows that the visual effect of the image denoised by the method is better than that of the other two algorithms; the EPI of the image denoised by the method is higher than that of the other two algorithms, which shows that the edge retention capability of the method is obviously better than that of the other two algorithms.
In order to accurately compare the denoising efficiencies of the three algorithms, table 2 lists the time t consumed by the SAR image denoising of the three targets through the SAIST algorithm, the WNNM algorithm and the fast SAR image denoising algorithm with minimized nuclear norm provided by the invention. Table 2 is as follows:
TABLE 2
As can be seen from Table 2, for the same target, the denoising speed of the method is increased by 3 times compared with the SAIST algorithm and 4 times compared with the WNNM algorithm, and the denoising efficiency is obviously improved. In conclusion, the analysis shows that compared with the SAIST algorithm and the WNNM algorithm, the method provided by the invention has better edge holding capability while greatly improving the denoising efficiency, and is improved in both peak signal-to-noise ratio and equivalent visual number.
While the invention has been described in terms of its preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.
Claims (5)
1. A SAR image fast denoising method based on RSVD and histogram preservation is characterized by comprising the following steps:
(1) establishing a speckle noise model of the SAR image, namely converting the logarithm of multiplicative noise into additive noise;
(2) carrying out block matching on the transformed image by using non-local similarity;
(3) decomposing a low-rank matrix composed of non-local similar image blocks by adopting RSVD (remote singular value decomposition), and realizing low-rank matrix approximation;
(4) performing texture enhancement on the SAR image by adopting a gradient histogram storage method to realize SAR image denoising;
(5) and resetting the image block to obtain the denoised SAR image.
2. The method for rapidly denoising SAR image stored based on RSVD and histogram according to claim 1, wherein the step (1) specifically comprises: establishing a SAR image multiplicative noise model, namely Y (X, Y) is X (X, Y) N (X, Y), wherein Y (X, Y) is the SAR image added with speckle noise, namely a final observation image; x (X, y) is an original SAR image; n (x, y) is speckle noise; x (X, y) and N (X, y) are mutually independent random processes, speckle noise N (X, y) follows a Gamma distribution with a mean value of 1 and a variance of 1/L, and the probability density function is as follows:wherein, L is an equivalent view, and e is a natural base number; taking logarithms on two sides of Y (X, Y) ═ X (X, Y) N (X, Y) at the same time, then converting the multiplicative noise into the commonly used additive noise: lg (Y (X, Y)) + lg (X, Y)) + lg (N (X, Y)).
3. The method for rapidly denoising SAR image stored based on RSVD and histogram according to claim 1, wherein the step (2) specifically comprises:
(21) dividing an original image into a plurality of image blocks with the same size, and searching non-local similar image blocks according to specified image blocks;
(22) and integrating the non-local similar image blocks into a matrix to form a low-rank matrix.
4. The method for rapidly denoising SAR image stored based on RSVD and histogram according to claim 1, wherein the step (3) specifically comprises:
(31) generating an n multiplied by l-dimensional Gaussian random matrix omega;
(32) multiplying a Gaussian random matrix omega by an m × n original matrix A to be decomposed to construct an m × l sample matrix Y which is A omega;
(33) carrying out QR decomposition on the original matrix Y to obtain an m multiplied by l dimension orthogonal matrix Q;
(34) constructing a matrix of dimension Lxn, B ═ QTX A, and finally performing singular value decomposition on the matrix B, namely B ═ S Λ VT；
(35) Let QS ═ U obtain singular value decomposition of A, i.e. A ═ UΛ VT；
Where l is much smaller than the smaller of m and n, QTFor the transpose of Q, S is a square matrix of l x l, Λ is a diagonal matrix of l x n, the elements on the diagonal being called singular values, VTIs the transpose of V, which is an n × n square matrix.
5. The method for rapidly denoising SAR image stored based on RSVD and histogram according to claim 1, wherein the step (4) specifically comprises:
(41) first, a gradient histogram h of an original image x is estimatedrTaking the reference gradient as a reference gradient, and taking the updated image close to the reference gradient histogram to the maximum extent as a constraint condition to obtain a result image; wherein, histogram of gradient hrThe solving formula of (2) is as follows:
wherein d is a constant, R (h)x) For regularization terms based on prior information of the gradient histogram of the natural image,is the gradient of the noise n, hy、hx and hgHistograms of the noisy image, the original image and the gradient g, respectively;
(42) to denoise an imageIs approximated by a reference histogram hrThe denoising model stored based on the gradient histogram is as follows:
s.t.hf＝hr
where ▽ is a gradient operation, F is an odd function that monotonically increases over (0, + ∞), and h isfFor the transformed gradient image | F (▽ x) | histogram, hrIs a histogram of the original image, R (x) is a regularization term, λ is a normal number, σn 2μ is a constant for the noise variance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811206810.7A CN109658340B (en) | 2018-10-17 | 2018-10-17 | SAR image rapid denoising method based on RSVD and histogram preservation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811206810.7A CN109658340B (en) | 2018-10-17 | 2018-10-17 | SAR image rapid denoising method based on RSVD and histogram preservation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109658340A true CN109658340A (en) | 2019-04-19 |
| CN109658340B CN109658340B (en) | 2023-04-25 |
Family
ID=66110305
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811206810.7A Active CN109658340B (en) | 2018-10-17 | 2018-10-17 | SAR image rapid denoising method based on RSVD and histogram preservation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109658340B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109709586A (en) * | 2018-12-19 | 2019-05-03 | 中铁第四勘察设计院集团有限公司 | The method for building up and application method of GPS reference station net coordinate time sequence three-dimensional noise model based on singular value decomposition |
| CN110675331A (en) * | 2019-08-13 | 2020-01-10 | 南京人工智能高等研究院有限公司 | Image denoising method and device, computer readable storage medium and electronic device |
| CN112287943A (en) * | 2020-09-28 | 2021-01-29 | 北京航空航天大学 | Anti-attack defense method based on image enhancement technology |
| CN114004254A (en) * | 2021-10-27 | 2022-02-01 | 云南电网有限责任公司昆明供电局 | Power cable partial discharge signal white noise and pulse type interference filtering algorithm |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323989A (en) * | 2011-09-16 | 2012-01-18 | 西安电子科技大学 | Singular value decomposition non-local mean-based polarized synthetic aperture radar (SAR) data speckle suppression method |
| CN103839237A (en) * | 2014-02-26 | 2014-06-04 | 西安电子科技大学 | SAR image despeckling method based on SVD dictionary and linear minimum mean square error estimation |
| CN104732493A (en) * | 2015-03-18 | 2015-06-24 | 西安电子科技大学 | SAR image de-noising algorithm based on Primal Sketch classification and SVD domain improvement MMSE estimation |
| CN107301631A (en) * | 2017-06-28 | 2017-10-27 | 重庆大学 | A kind of SAR image method for reducing speckle that sparse constraint is weighted based on non-convex |
| CN107358589A (en) * | 2017-07-17 | 2017-11-17 | 桂林电子科技大学 | A kind of combination histogram of gradients and the denoising method of low-rank constraint |
-
2018
- 2018-10-17 CN CN201811206810.7A patent/CN109658340B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102323989A (en) * | 2011-09-16 | 2012-01-18 | 西安电子科技大学 | Singular value decomposition non-local mean-based polarized synthetic aperture radar (SAR) data speckle suppression method |
| CN103839237A (en) * | 2014-02-26 | 2014-06-04 | 西安电子科技大学 | SAR image despeckling method based on SVD dictionary and linear minimum mean square error estimation |
| CN104732493A (en) * | 2015-03-18 | 2015-06-24 | 西安电子科技大学 | SAR image de-noising algorithm based on Primal Sketch classification and SVD domain improvement MMSE estimation |
| CN107301631A (en) * | 2017-06-28 | 2017-10-27 | 重庆大学 | A kind of SAR image method for reducing speckle that sparse constraint is weighted based on non-convex |
| CN107358589A (en) * | 2017-07-17 | 2017-11-17 | 桂林电子科技大学 | A kind of combination histogram of gradients and the denoising method of low-rank constraint |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109709586A (en) * | 2018-12-19 | 2019-05-03 | 中铁第四勘察设计院集团有限公司 | The method for building up and application method of GPS reference station net coordinate time sequence three-dimensional noise model based on singular value decomposition |
| CN109709586B (en) * | 2018-12-19 | 2020-09-08 | 中铁第四勘察设计院集团有限公司 | Method for establishing GPS reference station network coordinate time sequence three-dimensional noise model based on singular value decomposition and using method |
| CN110675331A (en) * | 2019-08-13 | 2020-01-10 | 南京人工智能高等研究院有限公司 | Image denoising method and device, computer readable storage medium and electronic device |
| CN112287943A (en) * | 2020-09-28 | 2021-01-29 | 北京航空航天大学 | Anti-attack defense method based on image enhancement technology |
| CN114004254A (en) * | 2021-10-27 | 2022-02-01 | 云南电网有限责任公司昆明供电局 | Power cable partial discharge signal white noise and pulse type interference filtering algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109658340B (en) | 2023-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112200750B (en) | Ultrasonic image denoising model establishing method and ultrasonic image denoising method | |
| CN109658340B (en) | SAR image rapid denoising method based on RSVD and histogram preservation | |
| CN105809198B (en) | SAR image target recognition method based on depth confidence network | |
| Wang et al. | Generative adversarial network-based restoration of speckled SAR images | |
| CN109919870B (en) | SAR image speckle suppression method based on BM3D | |
| CN105894476B (en) | SAR image noise reduction process method based on dictionary learning fusion | |
| CN109949256B (en) | Astronomical image fusion method based on Fourier transform | |
| CN104036461B (en) | A kind of Infrared Complex Background suppressing method based on Federated filter | |
| CN115082336A (en) | SAR image speckle suppression method based on machine learning | |
| CN114240990B (en) | SAR image point target segmentation method | |
| CN103793889B (en) | SAR image based on dictionary learning and PPB algorithm removes spot method | |
| CN103077507A (en) | Beta algorithm-based multiscale SAR (Synthetic Aperture Radar) image denoising method | |
| CN117671540A (en) | Method and system for detecting small target of attention aerial image based on multispectral frequency channel | |
| CN114529518B (en) | Image pyramid and NLM-based frozen electron microscope image enhancement method | |
| CN102722878A (en) | SAR (synthetic aperture radar) image despeckle method based on target extraction and PPB (probabilistic patch-based filter) algorithm | |
| Bhonsle et al. | Medical Image De-Noising Using Combined Bayes Shrink and Total Variation Techniques | |
| Kondaveeti et al. | Abridged shape matrix representation for the recognition of aircraft targets from 2D ISAR imagery | |
| Liu et al. | Infrared aerial small target detection with NSCT and two-dimensional property histogram | |
| Fang et al. | De-noising of SAR images based on Wavelet-Contourlet domain and PCA | |
| Zhang et al. | Bayesian-based speckle suppression for SAR image using contourlet transform | |
| Chen et al. | Ultrasound image denoising with multi-shape patches aggregation based non-local means | |
| Wang et al. | LPMsDE: Multi-Scale Denoising and Enhancement Method Based on Laplacian Pyramid Framework for Forward-Looking Sonar Image | |
| Yang et al. | Research on image denoising algorithm based on non-local block matching | |
| CN112927169A (en) | Remote sensing image denoising method based on wavelet transformation and improved weighted nuclear norm minimization | |
| Ye et al. | Quantitative effects of discrete wavelet transforms and wavelet packets on aerial digital image denoising |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |