CN116612032A - Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD - Google Patents
Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD Download PDFInfo
- Publication number
- CN116612032A CN116612032A CN202310587521.0A CN202310587521A CN116612032A CN 116612032 A CN116612032 A CN 116612032A CN 202310587521 A CN202310587521 A CN 202310587521A CN 116612032 A CN116612032 A CN 116612032A
- Authority
- CN
- China
- Prior art keywords
- image
- gray
- representing
- row
- modal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000001914 filtration Methods 0.000 title claims abstract description 34
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 16
- 238000012216 screening Methods 0.000 claims abstract description 6
- 230000003044 adaptive effect Effects 0.000 claims description 22
- 230000008569 process Effects 0.000 claims description 15
- 230000014759 maintenance of location Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 9
- 238000012545 processing Methods 0.000 description 8
- 230000009466 transformation Effects 0.000 description 5
- 238000009499 grossing Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 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
- G06T5/20—Image enhancement or restoration using local operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
A sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD belong to the technical field of sonar image denoising. The invention solves the problems of poor edge and detail feature retaining capability, low denoising efficiency and poor denoising effect of the existing sonar image denoising method. The technical scheme adopted by the invention is as follows: performing modal decomposition on the acquired original sonar image, and decomposing the original sonar image into K modal components; step two, screening out effective modal components from the K modal components obtained in the step one; step three, filtering the effective modal components screened in the step two to obtain filtered effective modal components; and step four, reconstructing the filtered effective modal components to obtain a denoised sonar image. The method can be applied to sonar image denoising.
Description
Technical Field
The invention belongs to the technical field of sonar image denoising, and particularly relates to a sonar image denoising method and equipment based on self-adaptive wiener filtering and 2D-VMD (two-dimensional variation modal decomposition).
Background
Sonar is the most dominant way to feed back ocean information, and because of the influence of sea wind, ocean currents, water temperature, impurities, imaging equipment and the like, images obtained by using a sonar detection technology are far more severely interfered by noise than optical images, usually contain various types of noise, and mainly show granular spots. Speckle noise is multiplicative noise, so that problems such as image blurring, edge detail loss and the like are caused, the image quality is seriously reduced, and the subsequent image segmentation, image recognition and the like are adversely affected. Algorithms commonly used in recent years for sonar image denoising include: wavelet transformation, ridge wave transformation, curvelet transformation and the like, and wavelet denoising cannot well obtain edge and contour characteristics; the ridge wave transformation can only represent linear singular features, and curve singular features in signals cannot be effectively described, so that detail features are lost; the curvelet transformation structure is complex and the operation amount is large. In summary, these methods have the problems that the image edge and detail feature cannot be captured accurately during denoising, the denoising efficiency is low, and the denoising effect is poor.
Therefore, how to effectively remove noise in a sonar image and improve the edge and detail feature retaining capability under the condition of serious noise interference, and meanwhile, improving the denoising efficiency is a problem to be solved urgently.
Disclosure of Invention
The invention aims to solve the problems of poor edge and detail feature retaining capability, low denoising efficiency and poor denoising effect of the existing sonar image denoising method, and provides a sonar image denoising method based on adaptive wiener filtering and 2D-VMD.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the sonar image denoising method based on the adaptive wiener filtering and the 2D-VMD specifically comprises the following steps:
performing modal decomposition on the acquired original sonar image, and decomposing the original sonar image into K modal components;
step two, screening out effective modal components from the K modal components obtained in the step one;
step three, filtering the effective modal components screened in the step two to obtain filtered effective modal components;
and step four, reconstructing the filtered effective modal components to obtain a denoised sonar image.
The sonar image denoising device based on the adaptive wiener filtering and the 2D-VMD comprises a processor and a memory, wherein at least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to realize a sonar image denoising method based on the adaptive wiener filtering and the 2D-VMD.
The beneficial effects of the invention are as follows:
the invention decomposes the noisy image through the 2D-VMD, then screens out the effective modal component through the correlation coefficient and the structural similarity, and then removes the noise signal in the effective modal component by the self-adaptive wiener filtering, and retains the detail information; and finally reconstructing the processed modal components to obtain a denoised image. The invention combines the self-adaptive wiener filtering and the 2D-VMD, and can furthest improve the denoising effect, the edge and the detail feature retaining capability of the sonar image. The invention can also improve the problems of low contrast of the sonar image and serious noise interference, simultaneously reduce the operand and improve the denoising efficiency.
Drawings
FIG. 1 is a flow chart of a sonar image denoising method based on adaptive wiener filtering and 2D-VMD of the present invention;
FIG. 2 is a noisy sonar image;
fig. 3 is a denoised sonar image.
Detailed Description
Detailed description of the inventionin the first embodiment, this embodiment will be described with reference to fig. 1. The sonar image denoising method based on the adaptive wiener filtering and the 2D-VMD, which is disclosed by the embodiment, specifically comprises the following steps of:
performing modal decomposition on the acquired original sonar image, and decomposing the original sonar image into K modal components;
step two, screening out effective modal components from the K modal components obtained in the step one; the other modal components except the effective modal component are ineffective modal components, and zero value processing is carried out on the ineffective modal components, namely the ineffective modal components do not need to participate in subsequent reconstruction.
Step three, filtering the effective modal components screened in the step two to obtain filtered effective modal components;
and step four, reconstructing the filtered effective modal components by using a 2D-VMD reconstruction algorithm to obtain a denoised sonar image.
The denoising object of the present embodiment is an underwater sonar image. In addition, the invention can obtain good denoising effect and edge holding capability under the condition of serious image noise pollution.
The second embodiment is as follows: the difference between the embodiment and the specific embodiment is that the mode decomposition of the collected original sonar image adopts 2D-VMD, and the center frequencies of K mode components obtained by decomposition are different.
The two-dimensional variational modal decomposition minimizes the bandwidth of the constituent sub-signals while maintaining data fidelity. The problem of two-dimensional variational modal decomposition is classified into finding a modal u k Which can optimally reconstruct a given input signalWhile each mode is limited to an online estimated center frequency +.>The specific process of the two-dimensional variational modal decomposition algorithm can be described as:
step one by one, determining an input signalNumber of eigenmodes K, parameter alpha k ，τ，ε；
Step two, determining the output as an eigenmode functionCenter frequency->
Step one, three, initializing parametersLet iteration variable n=0, k= 1:K;
step four, a two-dimensional mask is established by utilizing an analysis signal Fourier multiplier:
step one five, update
Step one, updating
Step seven, update u k ：
Step one, eight, updating
Step one, nine, judging whether the components meet constraint conditionsIf yes, stopping iteration, if not, making n=n+1, and returning to the step one.
In the first step, the number K of mode components is different, and the signal u is analyzed in two dimensions k And a center frequency omega k The optimization of (c) yields different results, thereby affecting the effect of the image decomposition. While determining the number K of image modal component decomposition, the values of bandwidth limit alpha, lagrange multiplier double rise time step tau and tolerance tol should be adjusted appropriately so that the decomposition result is an ideal effect.
And a third specific embodiment: the first difference between the present embodiment and the specific embodiment is that the specific process of the second step is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
step two, calculating similarity coefficients of the image A and the image B:
wherein ,Amn Representing the gray value of the nth row and column pixels in image A, B mn Representing the gray values of the mth row and column pixels in image B, m=1, 2, …, M, n=1, 2, …, N, M representing the width of image a (image B), N representing the height of image a (image B),represents the average value of the gray scale of the pixels in image a, is->Representing the average value of the pixel gray scale in image B, CC is the correlation between image A and image BA number;
step two, calculating the structural similarity of the image A and the image B;
the brightness relationship L (X, Y) between the image a and the image B is as follows:
wherein ,μX Is the average value of the brightness of the pixels in image A, mu Y Is the average value of the brightness of the pixels in the image B, C 1 Is a constant;
the contrast relation C (X, Y) between the image A and the image B is as follows:
wherein ,C2 Is constant, sigma X Is the standard deviation of the pixels of image a (i.e., the standard deviation of the brightness of the pixels in image a), σ Y Is the standard deviation of the pixels of image B (i.e., the standard deviation of the brightness of the pixels in image B), σ XY Is the cross-covariance between image a and image B;
the structural relation S (X, Y) between the image A and the image B is as follows:
wherein ,C3 Is a constant;
and step two, three and four, the structural similarity SSIM (X, Y) of the image A and the image B is as follows:
SSIM(X,Y)＝[L(X,Y)]·[C(X,Y)]·[S(X,Y)]
step two, for each modal component obtained in step one, executing the processes from step two to step three;
step five, selecting 1 effective modal component from the K modal components obtained in the step one according to the calculated similarity coefficient and the structural similarity;
the effective modal components satisfy: the similarity coefficient corresponding to the effective modal component is larger than that of any other modal component, and the structural similarity corresponding to the effective modal component is larger than that of any other modal component. I.e. the modal component with the largest similarity coefficient, the structural similarity is necessarily the largest.
The specific embodiment IV is as follows: the first difference between the present embodiment and the specific embodiment is that the specific process of the second step is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
wherein ,Amn Is the gray value of the nth row and nth column pixels in image A, B mn Is the gray value of the nth row and nth column pixels in image B, M is the width of image a (image B), N is the height of image a (image B), i·i represents the 2 norm, and PSNR is the peak signal-to-noise ratio;
step two, executing the process of step two for each modal component obtained in step one;
step two, selecting 1 effective modal component from the K modal components obtained in the step one according to the peak signal-to-noise ratio corresponding to each modal component;
the effective modal components satisfy: the peak signal-to-noise ratio corresponding to the effective modal component is greater than the peak signal-to-noise ratio corresponding to any other modal component.
Fifth embodiment: the first difference between the present embodiment and the specific embodiment is that the specific process of the second step is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
wherein EPI represents edge retention index, A m,n Representing the gray value of the nth row and column pixels in image A, A (m-1,n-1) Representing the gray value of the (m-1) th row (n-1) th column (A) pixel in the image (A) (m-1,n) Representing the gray value of the nth pixel of the m-1 th row and the nth pixel of the image A (m-1,n+1) Represents the gray value of the (m-1) th row and (n+1) th column pixels in the image A, A (m,n-1) Representing the gray value of the nth row and nth column pixels in image A, A (m,n+1) Representing the gray value of the (n+1) th column pixel of the m-th row in the image A, A (m+1,n-1) Represents the gray value of the (m+1) -th row and (n-1) -th column pixels in the image A, A (m+1,n) Representing the gray value of the (m+1) -th row and (n) -th column pixels in the image A, A (m+1,n+1) Representing the gray value of the (m+1) -th row and (n+1) -th column pixels in the image A; b (B) m,n Representing the gray value of the nth row and column pixels in image B, B (m-1,n-1) Representing the gray value of the (m-1) th row (n-1) th column (B) pixel in the image (B) (m-1,n) Representing the gray value of the nth pixel of the m-1 th row and the nth pixel of the image B (m-1,n+1) Representing the gray value of the (n+1) th column pixel of the (m-1) th row in the image B, B (m,n-1) Representing the gray value of the nth row and nth column pixels in image B, B (m,n+1) Representing the gray value of the (n+1) th column pixel of the m-th row in the image B, B (m+1,n-1) Representing the gray value of the (m+1) -th row and (n-1) -th column pixels in the image B, B (m+1,n) Representing the gray value of the (m+1) -th row and (n) -th column pixels in the image B, B (m+1,n+1) Representing the gray value of the (m+1) -th row and (n+1) -th column pixels in the image B;
step two, executing the process of step two for each modal component obtained in step one;
step two, selecting 1 effective modal component from the K modal components obtained in the step one according to the calculated edge retention index;
the effective modal components satisfy: the edge retention index corresponding to the effective modal component is greater than the edge retention index corresponding to any other modal component.
Specific embodiment six: the first difference between this embodiment and the specific embodiment is that adaptive wiener filtering is adopted in the third step.
Seventh embodiment: the sixth embodiment is different from the sixth embodiment in that, in the image corresponding to the filtered effective modal component, a pixel point (t 1 ,t 2 ) The gray values of (2) are:
wherein ,P(t1 ,t 2 ) Is the pixel point (t) in the gray image corresponding to the filtered effective modal component 1 ,t 2 ) Gray value, mu Z Is the pixel point (t 1 ,t 2 ) Is a local gray average value of sigma Z 2 Is the pixel point (t 1 ,t 2 ) Is a (t) 1 ,t 2 ) Is the pixel point (t 1 ,t 2 ) Gray value of n a 2 Is the variance of the noise.
Eighth embodiment: the seventh difference between the present embodiment and the specific embodiment is that the pixel point (t 1 ,t 2 ) Is a local gray average mu Z And local gray variance sigma Z 2 The method comprises the following steps:
wherein ,is the pixel point (t) 1 ,t 2 ) R X S neighborhood of (t' 1 ,t′ 2 ) Is the pixel point (t) 1 ,t 2 ) Pixel points in R×S neighborhood of (a), (t' 1 ,t′ 2 ) Is the pixel point (t 'in the gray image corresponding to the effective modal component' 1 ,t′ 2 ) Is a gray value of (a).
Detailed description nine: this embodiment differs from the specific embodiment eight in that the variance n of the noise a 2 The calculation mode of (a) is as follows:
wherein ,σZ 2 (t′ 1 ,t′ 2 ) Is the pixel point (t 'in the gray image corresponding to the effective modal component' 1 ,t′ 2 ) Is used for the local gray variance of (a).
And respectively processing each pixel point in the gray image corresponding to the effective modal component to obtain a filtered effective modal component. In this embodiment, the specific value of r×s may be set according to the actual situation, and small smoothing is performed by the filter in the case where the image variance is large, and more smoothing is performed in the case where the image variance is small. The size of the filtering template influences the denoising effect, and the window with proper size is selected to process effective modal components aiming at noise with different degrees so as to achieve the optimal denoising effect. For any boundary pixel of an image, only the pixels within the image are considered when calculating using pixels within its r×s neighborhood.
The tenth embodiment is a sonar image denoising device based on adaptive wiener filtering and 2D-VMD, the device comprises a processor and a memory, it should be understood that any device comprising a processor and a memory described in the present invention can also comprise other units and modules for performing display, interaction, processing, control, etc. and other functions by signals or instructions;
at least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement a sonar image denoising method based on adaptive wiener filtering and 2D-VMD.
Examples
The underwater sonar image denoising method based on image processing in the embodiment comprises the following steps: decomposing an acquired sonar image by using a 2D-VMD, screening out effective modal components by using a correlation coefficient and structural similarity, reconstructing the effective modal components to obtain an image formed by the effective modal components, processing the image by using a self-adaptive wiener filter, removing image noise, and retaining detail information; zero value processing is carried out on invalid modal components; and finally reconstructing the processed modal components to obtain a denoised image.
Taking a wreck sonar image as an example for a denoising experiment, the wreck image containing noise is shown in fig. 2, and the problems of serious noise interference and low contrast exist in the image, so that the overall quality and visual effect of the image are affected. The embodiment is processed by a sonar image denoising method based on adaptive wiener filtering and 2D-VMD. The specific process is as follows:
performing modal decomposition on a noisy sonar image by adopting a 2D-VMD to obtain 5 components with different modal center frequencies;
calculating each modal component image by using two indexes of the correlation coefficient and the structural similarity, dividing the modal components into an effective modal component and an ineffective modal component, and screening out the component with the correlation coefficient and the maximum structural similarity as an effective modal component IMF1;
step three, processing IMF1 modal components by utilizing self-adaptive wiener filtering, removing noise, and performing zero value processing on invalid IMF components;
and step four, reconstructing the filtered effective modal component to obtain a denoised image.
The invention carries out denoising treatment on the underwater wreck image, can improve the overall quality and visual effect of the image, enhance the image edge holding capacity and detail information, and the denoised image is shown in figure 3.
The above examples of the present invention are only for describing the calculation model and calculation flow of the present invention in detail, and are not limiting of the embodiments of the present invention. Other variations and modifications of the above description will be apparent to those of ordinary skill in the art, and it is not intended to be exhaustive of all embodiments, all of which are within the scope of the invention.
Claims (10)
1. The sonar image denoising method based on the adaptive wiener filtering and the 2D-VMD is characterized by comprising the following steps of:
performing modal decomposition on the acquired original sonar image, and decomposing the original sonar image into K modal components;
step two, screening out effective modal components from the K modal components obtained in the step one;
step three, filtering the effective modal components screened in the step two to obtain filtered effective modal components;
and step four, reconstructing the filtered effective modal components to obtain a denoised sonar image.
2. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the mode decomposition of the collected original sonar image adopts 2D-VMD, and the center frequencies of K mode components obtained by decomposition are different.
3. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the specific process of the step two is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
step two, calculating similarity coefficients of the image A and the image B:
wherein ,Amn Representing the gray value of the nth row and column pixels in image A, B mn Representing the gray values of the nth row and nth column pixels in image B, m=1, 2, …, M, n=1, 2, …, N, M representing the width of image a, N representing the height of image a,represents the average value of the gray scale of the pixels in image a, is->Representing the average value of the pixel gray scale in the image B, wherein CC is the correlation coefficient of the image A and the image B;
step two, calculating the structural similarity of the image A and the image B;
the brightness relationship L (X, Y) between the image a and the image B is as follows:
wherein ,μX Is the average value of the brightness of the pixels in image A, mu Y Is the average value of the brightness of the pixels in the image B, C 1 Is a constant;
the contrast relation C (X, Y) between the image A and the image B is as follows:
wherein ,C2 Is constant, sigma X Standard deviation, sigma, of image a pixels Y Standard deviation, sigma, of image B pixels XY For image A and mapCross covariance between images B;
the structural relation S (X, Y) between the image A and the image B is as follows:
wherein ,C3 Is a constant;
and step two, three and four, the structural similarity SSIM (X, Y) of the image A and the image B is as follows:
SSIM(X,Y)＝[L(X,Y)]·[C(X,Y)]·[S(X,Y)]
step two, for each modal component obtained in step one, executing the processes from step two to step three;
step five, selecting 1 effective modal component from the K modal components obtained in the step one according to the calculated similarity coefficient and the structural similarity;
the effective modal components satisfy: the similarity coefficient corresponding to the effective modal component is larger than that of any other modal component, and the structural similarity corresponding to the effective modal component is larger than that of any other modal component.
4. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the specific process of the step two is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
wherein ,Amn Is the gray value of the nth row and nth column pixels in image A, B mn Is the gray value of the nth row and nth column pixels in the image B, M is the width of the image A, N is the height of the image A, I I.I. represents the 2 norm, and PSNR is the peak signal to noise ratio;
step two, executing the process of step two for each modal component obtained in step one;
step two, selecting 1 effective modal component from the K modal components obtained in the step one according to the peak signal-to-noise ratio corresponding to each modal component;
the effective modal components satisfy: the peak signal-to-noise ratio corresponding to the effective modal component is greater than the peak signal-to-noise ratio corresponding to any other modal component.
5. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the specific process of the step two is:
step two, for any modal component, marking a gray image corresponding to the modal component as an image A, and marking a gray image corresponding to an original sonar image as an image B;
wherein EPI represents edge retention index, A m,n Representing the gray value of the nth row and column pixels in image A, A (m-1,n-1) Representing the gray value of the (m-1) th row (n-1) th column (A) pixel in the image (A) (m-1,n) Representing the gray value of the nth pixel of the m-1 th row and the nth pixel of the image A (m-1,n+1) Represents the gray value of the (m-1) th row and (n+1) th column pixels in the image A, A (m,n-1) Representing the gray value of the nth row and nth column pixels in image A, A (m,n+1) Representing the gray value of the (n+1) th column pixel of the m-th row in the image A, A (m+1,n-1) Represents the gray value of the (m+1) -th row and (n-1) -th column pixels in the image A, A (m+1,n) Representing the gray value of the (m+1) -th row and (n) -th column pixels in the image A, A (m+1,n+1) M-th in representative image AGray values of +1th row and n+1th column pixels; b (B) m,n Representing the gray value of the nth row and column pixels in image B, B (m-1,n-1) Representing the gray value of the (m-1) th row (n-1) th column (B) pixel in the image (B) (m-1,n) Representing the gray value of the nth pixel of the m-1 th row and the nth pixel of the image B (m-1,n+1) Representing the gray value of the (n+1) th column pixel of the (m-1) th row in the image B, B (m,n-1) Representing the gray value of the nth row and nth column pixels in image B, B (m,n+1) Representing the gray value of the (n+1) th column pixel of the m-th row in the image B, B (m+1,n-1) Representing the gray value of the (m+1) -th row and (n-1) -th column pixels in the image B, B (m+1,n) Representing the gray value of the (m+1) -th row and (n) -th column pixels in the image B, B (m+1,n+1) Representing the gray value of the (m+1) -th row and (n+1) -th column pixels in the image B;
step two, executing the process of step two for each modal component obtained in step one;
step two, selecting 1 effective modal component from the K modal components obtained in the step one according to the calculated edge retention index;
the effective modal components satisfy: the edge retention index corresponding to the effective modal component is greater than the edge retention index corresponding to any other modal component.
6. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the adaptive wiener filtering is adopted in the third step.
7. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 1, wherein the pixel point (t 1 ,t 2 ) The gray values of (2) are:
wherein ,P(t1 ,t 2 ) Is the filtered effective modal componentPixel point (t) in corresponding gray scale image 1 ,t 2 ) Gray value, mu Z Is the pixel point (t 1 ,t 2 ) Is a local gray average value of sigma Z 2 Is the pixel point (t 1 ,t 2 ) Is a (t) 1 ,t 2 ) Is the pixel point (t 1 ,t 2 ) Gray value of n a 2 Is the variance of the noise.
8. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 7, wherein the pixel point (t 1 ,t 2 ) Is a local gray average mu Z And local gray variance sigma Z 2 The method comprises the following steps:
wherein ,is the pixel point (t) 1 ,t 2 ) R X S neighborhood of (t' 1 ,t′ 2 ) Is the pixel point (t) 1 ,t 2 ) Pixel points in R×S neighborhood of (a), (t' 1 ,t′ 2 ) Is the pixel point (t 'in the gray image corresponding to the effective modal component' 1 ,t′ 2 ) Is a gray value of (a).
9. The sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to claim 8, wherein the variance n of the noise a 2 The calculation mode of (a) is as follows:
wherein ,σZ 2 (t′ 1 ,t′ 2 ) Is the pixel point (t 'in the gray image corresponding to the effective modal component' 1 ,t′ 2 ) Is used for the local gray variance of (a).
10. Sonar image denoising device based on adaptive wiener filtering and 2D-VMD, characterized in that it comprises a processor and a memory, in which at least one instruction is stored, loaded and executed by the processor to implement the sonar image denoising method based on adaptive wiener filtering and 2D-VMD according to one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310587521.0A CN116612032A (en) | 2023-05-23 | 2023-05-23 | Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310587521.0A CN116612032A (en) | 2023-05-23 | 2023-05-23 | Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116612032A true CN116612032A (en) | 2023-08-18 |
Family
ID=87679584
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310587521.0A Pending CN116612032A (en) | 2023-05-23 | 2023-05-23 | Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116612032A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117522724A (en) * | 2023-10-16 | 2024-02-06 | 江苏海洋大学 | Ad-Wiener and Sobel-based 2D-TV-VMD sonar image denoising method |
-
2023
- 2023-05-23 CN CN202310587521.0A patent/CN116612032A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117522724A (en) * | 2023-10-16 | 2024-02-06 | 江苏海洋大学 | Ad-Wiener and Sobel-based 2D-TV-VMD sonar image denoising method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111681174B (en) | Joint filtering method for inhibiting salt-pepper/Gaussian noise mixing target image | |
| CN109671029B (en) | Image denoising method based on gamma norm minimization | |
| CN111028163A (en) | Convolution neural network-based combined image denoising and weak light enhancement method | |
| CN109242799B (en) | Variable-threshold wavelet denoising method | |
| CN108416740B (en) | Iterative adaptive median filtering method for eliminating salt and pepper noise | |
| CN108765330B (en) | Image denoising method and device based on global and local prior joint constraint | |
| CN112991199B (en) | Image high-low frequency decomposition noise removal method based on residual dense network | |
| CN113808036B (en) | Low-illumination image enhancement and denoising method based on Retinex model | |
| CN116612032A (en) | Sonar image denoising method and device based on self-adaptive wiener filtering and 2D-VMD | |
| CN112750090A (en) | Underwater image denoising method and system for improving wavelet threshold | |
| CN114723630A (en) | Image deblurring method and system based on cavity double-residual multi-scale depth network | |
| CN111242854A (en) | Image denoising method | |
| CN110349106B (en) | Wavelet soft threshold image denoising method based on Renyi entropy | |
| CN115082336A (en) | SAR image speckle suppression method based on machine learning | |
| Pham | Kriging-weighted laplacian kernels for grayscale image sharpening | |
| CN111461999B (en) | SAR image speckle suppression method based on super-pixel similarity measurement | |
| CN117011192A (en) | NSCT domain sonar image denoising method and device based on neutral set and bilateral filtering | |
| CN115760773A (en) | Noise suppression method and system for hyperspectral medical image acquisition | |
| Pang | Improved image denoising based on Haar wavelet transform | |
| CN115409872A (en) | Underwater camera image optimization method | |
| CN112927169A (en) | Remote sensing image denoising method based on wavelet transformation and improved weighted nuclear norm minimization | |
| CN113888405A (en) | Denoising and demosaicing method based on clustering self-adaptive expansion convolutional neural network | |
| CN113379641A (en) | Single image rain removing method and system based on self-coding convolutional neural network | |
| CN113822823B (en) | Point neighbor restoration method and system for aerodynamic optical effect image space-variant fuzzy core | |
| Shakeel et al. | Efficient blind image deblurring method for palm print images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |