CN110766627A - Speckle interference image noise reduction method and device - Google Patents
Speckle interference image noise reduction method and device Download PDFInfo
- Publication number
- CN110766627A CN110766627A CN201910980587.XA CN201910980587A CN110766627A CN 110766627 A CN110766627 A CN 110766627A CN 201910980587 A CN201910980587 A CN 201910980587A CN 110766627 A CN110766627 A CN 110766627A
- Authority
- CN
- China
- Prior art keywords
- phase diagram
- cosine
- sine
- phase
- noise reduction
- 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
- 230000009467 reduction Effects 0.000 title claims abstract description 105
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000010587 phase diagram Methods 0.000 claims abstract description 189
- 230000009466 transformation Effects 0.000 claims abstract description 49
- 238000000354 decomposition reaction Methods 0.000 claims description 59
- 238000012545 processing Methods 0.000 claims description 18
- 230000008033 biological extinction Effects 0.000 claims description 2
- 238000001914 filtration Methods 0.000 abstract description 47
- 238000010586 diagram Methods 0.000 description 18
- 230000000694 effects Effects 0.000 description 6
- 238000005259 measurement Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 235000002566 Capsicum Nutrition 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 235000016761 Piper aduncum Nutrition 0.000 description 1
- 235000017804 Piper guineense Nutrition 0.000 description 1
- 244000203593 Piper nigrum Species 0.000 description 1
- 235000008184 Piper nigrum Nutrition 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The application discloses a speckle interference image noise reduction method and device, and the method comprises the following steps: converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation; respectively denoising the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transformation to obtain a sine denoising phase diagram and a cosine denoising phase diagram; and determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram. The speckle interference image noise filtering method and device can effectively and quickly filter noise in speckle interference images.
Description
Technical Field
The application relates to the technical field of digital speckle interference, in particular to a speckle interference image noise reduction method and device.
Background
The speckle interference technology records a speckle field generated by interference between reference light and object light reflected from the rough surface of an object, and can calculate the tiny deformation of the surface of the object according to the change of a phase diagram of the speckle field before and after the deformation of the object. The speckle interference technology has the advantages of full field, non-contact, high resolution and the like, and is widely applied to engineering measurement such as strain measurement, three-dimensional deformation measurement, vibration evaluation and the like. In the digital speckle interference technology, the speckle phase related to the deformation of an object is modulated in a [0, 2 pi ] interval, and black and white stripes are represented in a speckle interference image. In general, it is necessary to unwrapp the wrapped phase of [0, 2 pi ] in the black and white stripes to obtain a full-field continuous phase field, and then linearly convert the phase into full-field continuous deformation. These wrapped phases in the 0, 2 pi range are often heavily contaminated with noise, and therefore it is necessary to de-noise the speckle interference image prior to unwrapping.
In recent decades, the speckle interference image noise reduction methods mainly include: the method has the advantages that the sine and cosine mean filtering is adopted, the denoising is realized by performing mean filtering on a sine and cosine phase diagram, the method has the characteristics of simplicity and effectiveness, and is widely applied, but the sine and cosine mean denoising effect is closely related to the filtering times, the denoising effect of one-time denoising treatment is poor and usually needs 20-40 times, and the filtering speed is slower as the denoising times are more; and secondly, windowing Fourier transform filtering, which discards small-window Fourier coefficients in a 2D window Fourier transform domain to realize denoising, has strong denoising capability but has a filtering effect influenced by the size of a window and consumes long time.
In summary, it is difficult for the current commonly used noise reduction methods for two types of speckle interference images to achieve good effects on noise reduction capability and noise reduction speed.
Disclosure of Invention
In view of this, the present application provides a method and an apparatus for reducing noise of speckle interference images. The following presents a simplified summary of the application in order to provide a basic understanding of some aspects of the application. It should be understood that this summary is not an exhaustive overview of the present application. It is not intended to identify key or critical elements of the application or to delineate the scope of the application. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later.
According to a first aspect of the present application, there is provided a method for reducing noise of a speckle interference image, including: converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation; respectively carrying out noise reduction processing on the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transform to obtain a sine noise reduction diagram and a cosine noise reduction diagram; and determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Optionally, the converting the discontinuous original phase map into continuous sine phase map and cosine phase map by sine transform and cosine transform includes: performing sinusoidal transformation on a phase value of any point in the original phase diagram to obtain a sinusoidal phase diagram; and performing cosine transformation on the phase value of any point in the original image to obtain the cosine phase diagram.
Optionally, the resolution of the original phase map is M × N, and the phase value of any point (M, N) in the original phase map is M × NWherein M is more than or equal to M and more than or equal to 1, and N is more than or equal to N and more than or equal to 1; the sinusoidal phase map S (m, n) is:the cosine phase diagram C (m, n) is:
optionally, the performing, by using stationary wavelet transform, noise reduction processing on the sine phase map and the cosine phase map respectively to obtain a sine noise reduction phase map and a cosine noise reduction phase map, includes: respectively performing stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram to obtain a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram; and respectively carrying out threshold denoising and wavelet inverse transformation on the wavelet decomposition result corresponding to the sine phase diagram and the wavelet decomposition result corresponding to the cosine phase diagram to obtain the sine denoising phase diagram and the cosine denoising phase diagram.
Optionally, the performing a stationary wavelet decomposition on the sine phase map and the cosine phase map respectively comprises: respectively carrying out 5-layer stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram by adopting a Daubechies wavelet function with an extinction moment of 4 through the following formulas:
wherein a is a detail coefficient, aAj、aHj、aVjAnd aDjApproximate detail, horizontal detail, vertical detail and diagonal detail of the stationary wavelet J (J ═ 1, 2, … J) layer decomposition respectively,
in the form of a low-pass digital filter,
for a high-pass digital filter, the upper corner marks col and row represent the operation in the column direction and row direction of the two-dimensional image, respectively.
Optionally, the performing threshold denoising and wavelet inverse transform on the wavelet decomposition result corresponding to the sine phase map and the wavelet decomposition result corresponding to the cosine phase map respectively to obtain the sine denoising phase map and the cosine denoising phase map includes: determining a threshold lambda by adopting a birgBe-Massart strategy with high penalty parameters, wherein the birgBe-Massart strategy is as follows:
λ＝|ct*|，
wherein α is a high penalty parameter, 2.5 & lt α & lt 10, sigma is the standard deviation of Gaussian noise in the error model, ckWavelet coefficients with absolute values arranged according to descending order, wherein l is the sum of the number of the coefficients, and t is the minimum value obtained by variable t; according to the threshold lambda, threshold denoising is carried out on a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram by using a soft threshold function, and stationary wavelet inverse transformation is carried out on a denoised image to obtain the sine denoising phase diagram and the cosine denoising phase diagram, wherein the soft threshold function is as follows:
ajoriginal detail coefficients for stationary wavelet transform, cjThe detail coefficient after the threshold value.
Optionally, the determining a target noise reduction phase map according to the sine noise reduction phase map and the cosine noise reduction phase map includes: and determining the target noise reduction phase diagram by a four-quadrant arc tangent algorithm according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Optionally, the determining the target denoising phase map by a four-quadrant arc tangent algorithm according to the sine denoising phase map and the cosine denoising phase map comprises: calculating the target noise reduction phase diagram phi (m, n) by the following formula:
wherein S '(m, n) is the sine noise reduction phase map, and C' (m, n) is the cosine noise reduction phase map.
According to a second aspect of the present application, there is provided a speckle interference image noise reduction device, including: the first processing module is used for converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation; the noise reduction module is used for respectively carrying out noise reduction processing on the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transform to obtain a sine noise reduction phase diagram and a cosine noise reduction phase diagram; and the second processing module is used for determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation; respectively denoising the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transformation to obtain a sine denoising phase diagram and a cosine denoising phase diagram; and determining a target noise reduction phase image according to the sine noise reduction phase image and the cosine noise reduction phase image, thereby effectively and quickly filtering the noise in the speckle interference image.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
The function of the present invention and its advantages will be more apparent from the following detailed description of the preferred embodiments of the present application taken in conjunction with the accompanying drawings.
Drawings
The present application may be better understood by reference to the following description taken in conjunction with the accompanying drawings. The accompanying drawings, which are incorporated in and form a part of this specification, illustrate preferred embodiments of the present application and, together with the detailed description, serve to further explain the principles and advantages of the application. In the drawings:
fig. 1 is a schematic flowchart of a method for reducing noise of a speckle interference image according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of a method for reducing noise of a speckle interference image according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of an original phase diagram provided by an embodiment of the present application;
FIG. 4 is a schematic diagram of a sinusoidal phase diagram obtained by performing sinusoidal transformation on the original phase diagram shown in FIG. 3 according to an embodiment of the present disclosure;
fig. 5 is a schematic diagram of a cosine phase diagram after cosine transforming the original phase diagram shown in fig. 3 according to an embodiment of the present application;
fig. 6 is a schematic diagram of a sinusoidal noise reduction phase map obtained based on the sinusoidal phase map shown in fig. 4 according to an embodiment of the present application;
fig. 7 is a schematic diagram of a cosine noise reduction phase map obtained based on the cosine phase map shown in fig. 5 according to an embodiment of the present application;
fig. 8 is a schematic diagram of a target noise reduction phase map obtained based on the sine noise reduction phase map shown in fig. 4 and the cosine noise reduction phase map shown in fig. 5 according to an embodiment of the present application;
fig. 9 is a schematic diagram of a result obtained by filtering 10 times with the sine and cosine mean filtering method, a result obtained by filtering 40 times with the sine and cosine mean filtering method, a filtering result obtained by filtering with a window size of 10 with the window fourier filtering method, and a filtering result obtained by filtering with a window size of 20 with the window fourier filtering method according to the embodiment of the present application;
FIG. 10 is a histogram comparing the filtering time consumption of the four filtering methods shown in FIG. 8 and the filtering method of the present application, provided in the embodiment of the present application;
fig. 11 is a schematic structural diagram of a noise reduction apparatus for speckle interference images according to an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present application will be described in detail below with reference to the accompanying drawings. In the interest of clarity and conciseness, not all features of an actual implementation are described in the specification. However, the present invention is not limited to the exemplary embodiments disclosed below; it can be implemented in different forms. The nature of the description is merely to assist those skilled in the relevant art in a comprehensive understanding of the specific details of the invention.
The following detailed description of embodiments of the present application will be made with reference to the accompanying drawings and examples. The following examples are intended to illustrate the present application but are not intended to limit the scope of the present application.
Fig. 1 is a flowchart of a method for reducing noise of a speckle interference image according to an embodiment of the present disclosure. As shown in fig. 1, the method may include:
step S11, the discontinuous original phase map is converted into continuous sine phase map and cosine phase map by sine transform and cosine transform.
And step S12, respectively carrying out threshold denoising treatment on the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transformation to obtain a sine denoising phase diagram and a cosine denoising phase diagram.
And step S13, determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Fig. 2 is a schematic diagram of a method for reducing noise of a speckle interference image according to an embodiment of the present disclosure. As shown in fig. 2, firstly, the speckle interference fringes containing noise (i.e. the original phase diagram containing noise obtained by speckle interference) are subjected to sine transformation and cosine transformation.
Optionally, the converting the discontinuous original phase map into continuous sine phase map and cosine phase image by sine transform and cosine transform includes: performing sinusoidal transformation on a phase value of any point in an original phase diagram to obtain a sinusoidal phase diagram; and performing cosine transformation on the phase value of any point in the original image to obtain a cosine phase diagram.
Optionally, the resolution of the original phase map is M × N, and the phase value of any point (M, N) in the original phase map is M × N
Wherein M is more than or equal to M and more than or equal to 1, and N is more than or equal to N and more than or equal to 1; the sinusoidal phase diagram S (m, n) is:
the cosine phase diagram C (m, n) is:
fig. 3 is a schematic diagram of an original phase diagram provided in an embodiment of the present application. The raw phase map in fig. 3 is a discontinuous phase fringe map with a resolution of 1280 × 1024. The phase value at any point (m, n) on the original phase map is
Wherein 1280 is more than or equal to m is more than or equal to 1, 1024 is more than or equal to n is more than or equal to 1, and the original phase diagram is respectively subjected to sine transformation and cosine transformation to obtain a sine phase diagram and a cosine phase diagram shown in fig. 4 and fig. 5. Fig. 4 is a sinusoidal phase diagram obtained by performing sinusoidal transformation on the original phase diagram shown in fig. 3 according to an embodiment of the present application. Fig. 5 is a cosine phase diagram obtained by performing cosine transform on the original phase diagram shown in fig. 3 according to an embodiment of the present application.
In FIG. 4, the phase values at (m, n) areIn FIG. 5, the phase values at (m, n) are
Still taking the above fig. 2 as an example, after performing sine transformation and cosine transformation on the speckle interference fringes containing noise, the obtained sine phase diagram and cosine phase diagram are sequentially subjected to wavelet decomposition, threshold denoising, and wavelet inverse transformation.
Optionally, the denoising processing is performed on the sine phase diagram and the cosine phase diagram respectively by using stationary wavelet transform to obtain a sine denoising phase diagram and a cosine denoising phase diagram, and the denoising processing includes: respectively performing stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram to obtain a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram; and respectively carrying out threshold denoising and wavelet inverse transformation on the wavelet decomposition result corresponding to the sine phase diagram and the wavelet decomposition result corresponding to the cosine phase diagram to obtain a sine denoising phase diagram and a cosine denoising phase diagram.
Optionally, the stationary wavelet decomposition is performed on the sine phase map and the cosine phase map respectively, and comprises: adopting a Daubechies wavelet function with the vanishing moment of 4, and respectively carrying out 5-layer stationary wavelet decomposition on a sine phase diagram and a cosine phase diagram by the following formulas:
wherein a is a detail coefficient, aAj、aHj、aVjAnd aDjApproximate detail, horizontal detail, vertical detail and diagonal detail of the stationary wavelet J (J ═ 1, 2, … J) level decomposition respectively,
in the form of a low-pass digital filter,
for a high-pass digital filter, the upper corner marks col and row represent the operation in the column direction and row direction of the two-dimensional image, respectively.
The decomposed components of each layer have the same scale with the original image, and the number of wavelet coefficients of each sub-band after decomposition is equal to the number of pixels of the original image.
The noise amplitude decreases with increasing decomposition level and the signal amplitude increases with increasing decomposition level, at the choice of decomposition level. Theoretically, the maximum possible decomposition level is
N represents the length of the signal. The decomposition level of image denoising is 3-5 layers generally, and for speckle interference images, the prior art generally adopts 3 layers of stable wavelet decomposition, so that the denoising effect is poor and not smooth enough. This application adopts 5 layers of steady wavelet decomposition, can effectively improve the noise reduction effect.
In selecting the wavelet function, the commonly used wavelet functions are Daubechies and Symlets, the performance is similar, and the wavelet order is N (N ═ 2, 3, …, 8). For a given N-th order vanishing moment orthogonal wavelet basis, when N is greater than 5, the amount of calculation is greatly increased, and the real-time performance is deteriorated. The Daubechies wavelet function with the vanishing moment of 4 is adopted, and the Daubechies wavelet function has the characteristic of minimum asymmetry, so that the visual smoothness of a filtering result and minimum data redundancy can be realized. In addition, the Daubechies wavelet function can be applied to the case without a specific model structure, i.e., to a variety of speckle interference images, is not limited to specific noise (e.g., white gaussian noise, salt and pepper noise, speckle noise, etc.), nor is it limited to a specific speckle fringe distribution (e.g., horizontal stripes, vertical stripes, circular stripes, butterfly spots, etc.).
Optionally, the performing threshold denoising and wavelet inverse transform on the wavelet decomposition result corresponding to the sine phase map and the wavelet decomposition result corresponding to the cosine phase map respectively to obtain a sine denoising phase map and a cosine denoising phase map includes: determining a threshold lambda by adopting a birgBe-Massart strategy with high penalty parameters, wherein the birgBe-Massart strategy is as follows:
λ＝|ct*|，
wherein, αIs a high penalty parameter, 2.5 < α < 10, sigma is the standard deviation of Gaussian noise in the error model, ckWavelet coefficients with absolute values arranged according to descending order, wherein l is the sum of the number of the coefficients, and t is the minimum value obtained by variable t; according to the threshold lambda, threshold denoising is carried out on a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram by utilizing a soft threshold function, and stationary wavelet inverse transformation is carried out on the denoised image to obtain a sine denoising phase diagram and a cosine denoising phase diagram, wherein the soft threshold function is as follows:
ajoriginal detail coefficients for stationary wavelet transform, cjThe detail coefficient after the threshold value.
And (2) self-adaptive layer correlation estimation determined by the layer coefficients of the stationary wavelet decomposition, namely, the corresponding threshold of each decomposition layer is different, and a birge-Massart strategy based on layer correlation is adopted to determine a high penalty threshold lambda, so that the high penalty threshold lambda has self-adaptive capacity to the noise level, and a proper threshold can be obtained by changing and adjusting parameters.
In terms of the threshold function, existing noise reduction techniques hard threshold non-decimated detail coefficients. Hard thresholding refers to preserving wavelet coefficients whose absolute values are greater than or equal to a threshold, while other wavelet coefficients are set to zero as noise. When the hard threshold is used for noise reduction, sudden change can be generated in a wavelet domain, so that the threshold is discontinuous, and the usable information can be lost due to the de-noising result. The method and the device adopt the soft threshold function, avoid the pseudo Gibbs artifact of the hard threshold function at the discontinuous part of the threshold point, have better smoothness, and are suitable for the alternation of black and white stripes in sine/cosine images.
Fig. 6 is a schematic diagram of a sinusoidal noise reduction phase map obtained based on the sinusoidal phase map shown in fig. 4 according to an embodiment of the present application. Fig. 7 is a schematic diagram of a cosine noise reduction phase map obtained based on the cosine phase map shown in fig. 5 according to an embodiment of the present application.
Optionally, determining a target noise reduction phase map according to the sine noise reduction phase map and the cosine noise reduction phase map, including: and determining a target noise reduction phase diagram by a four-quadrant arc tangent algorithm according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Optionally, determining the target noise reduction phase map by a four-quadrant arc tangent algorithm according to the sine noise reduction phase map and the cosine noise reduction phase map, including: calculating the target noise reduction phase diagram phi (m, n) by the following formula:
wherein, S '(m, n) is a sine noise reduction phase diagram, and C' (m, n) is a cosine noise reduction phase diagram.
Still taking the above fig. 2 as an example, after performing sine transformation and cosine transformation on the speckle interference fringes containing noise, and sequentially performing wavelet decomposition, threshold denoising, and wavelet inverse transformation on the obtained sine phase map and cosine phase map, a target denoising phase map (the speckle interference fringes after denoising) is determined by four-quadrant arc tangent transformation according to the obtained sine denoising phase map and cosine denoising phase map. Fig. 8 is a schematic diagram of a target noise reduction phase map obtained based on the sine noise reduction phase map shown in fig. 4 and the cosine noise reduction phase map shown in fig. 5 according to an embodiment of the present application.
Compared with a sine and cosine mean value denoising method needing to determine denoising times according to experience and a windowed Fourier denoising method needing to determine the window size in advance in the prior art, all parameters in the denoising method can be preset without changing, and the denoising method has adaptivity and practicability.
Compared with a windowing Fourier filtering method, the denoising method can obtain the denoising image with the same quality at a higher running speed.
Fig. 9 is a schematic diagram of a result obtained by filtering 10 times with the sine and cosine mean filtering method, a result obtained by filtering 40 times with the sine and cosine mean filtering method, a filtering result obtained by filtering with a window size of 10 with the window fourier filtering method, and a filtering result obtained by filtering with a window size of 20 with the window fourier filtering method according to the embodiment of the present application. Fig. 9 (a) shows the result of 10 times of sine-cosine mean filtering, fig. 9 (B) shows the result of 40 times of sine-cosine mean filtering, fig. 9 (C) shows the result of 10 times of window fourier filtering, and fig. 9 (D) shows the result of 20 times of window fourier filtering. As shown in fig. 8 and 9, the filtering method according to the present application can obtain a high-quality filtering result.
Fig. 10 is a histogram comparing the filtering time consumption of the four filtering methods shown in fig. 9 and the filtering method of the present application, provided in this embodiment of the present application. As shown in fig. 10, compared with the filtering method in the prior art, the filtering method of the present application can reduce filtering time consumption and improve filtering efficiency.
Converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation; respectively denoising the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transformation to obtain a sine denoising phase diagram and a cosine denoising phase diagram; and determining a target noise reduction phase image according to the sine noise reduction phase image and the cosine noise reduction phase image, thereby effectively and quickly filtering the noise in the speckle interference image.
Fig. 11 is a schematic structural diagram of a noise reduction apparatus for speckle interference images according to an embodiment of the present disclosure. As shown in fig. 11, the apparatus 110 includes:
a first processing module 111, configured to convert the discontinuous original phase map into continuous sine phase map and cosine phase map through sine transform and cosine transform;
the denoising module 112 is configured to perform denoising processing on the sine phase diagram and the cosine phase diagram respectively by using stationary wavelet transform to obtain a sine denoising phase diagram and a cosine denoising phase diagram;
and the second processing module 113 is configured to determine a target noise reduction phase map according to the sine noise reduction phase map and the cosine noise reduction phase map.
Optionally, the first processing module 111 is specifically configured to:
performing sinusoidal transformation on a phase value of any point in an original phase diagram to obtain a sinusoidal phase diagram;
and performing cosine transformation on the phase value of any point in the original image to obtain a cosine phase diagram.
Optionally, the resolution of the original phase map is M × N, and the phase value of any point (M, N) in the original phase map is M × N
Wherein M is more than or equal to M and more than or equal to 1, and N is more than or equal to N and more than or equal to 1;
the sinusoidal phase diagram S (m, n) is:
the cosine phase diagram C (m, n) is:
optionally, the noise reduction module 112 comprises:
the stationary wavelet decomposition sub-module is used for respectively performing stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram to obtain a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram;
and the denoising submodule is used for respectively carrying out threshold denoising and wavelet inverse transformation on the wavelet decomposition result corresponding to the sine phase diagram and the wavelet decomposition result corresponding to the cosine phase diagram to obtain a sine denoising phase diagram and a cosine denoising phase diagram.
Optionally, the stationary wavelet decomposition sub-module is specifically configured to:
adopting a Daubechies wavelet function with the vanishing moment of 4, and respectively carrying out 5-layer stationary wavelet decomposition on a sine phase diagram and a cosine phase diagram by the following formulas:
wherein a is a detail coefficient, aAj、aHj、aVjAnd aDjRespectively a stationary wavelet j (j ═ 1,2, … J) approximate, horizontal, vertical and diagonal details of the layer decomposition,
in the form of a low-pass digital filter,
for a high-pass digital filter, the upper corner marks col and row represent the operation in the column direction and row direction of the two-dimensional image, respectively.
Optionally, the noise reduction sub-module is specifically configured to:
determining a threshold lambda by adopting a birgBe-Massart strategy with high penalty parameters, wherein the birgBe-Massart strategy is as follows:
λ＝|ct*|，
wherein α is a high penalty parameter, 2.5 & lt α & lt 10, sigma is the standard deviation of Gaussian noise in the error model, ckWavelet coefficients with absolute values arranged according to descending order, wherein l is the sum of the number of the coefficients, and t is the minimum value obtained by variable t;
according to the threshold lambda, threshold denoising is carried out on a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram by utilizing a soft threshold function, and stationary wavelet inverse transformation is carried out on the denoised image to obtain a sine denoising phase diagram and a cosine denoising phase diagram, wherein the soft threshold function is as follows:
ajoriginal detail coefficients for stationary wavelet transform, cjThe detail coefficient after the threshold value.
Optionally, the second processing module 113 is specifically configured to:
and determining a target noise reduction phase diagram by a four-quadrant arc tangent algorithm according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Optionally, the second processing module 113 is specifically configured to:
calculating the target noise reduction phase diagram phi (m, n) by the following formula:
wherein, S '(m, n) is a sine noise reduction phase diagram, and C' (m, n) is a cosine noise reduction phase diagram.
Those skilled in the art will appreciate that the drawings are only schematic illustrations of preferred embodiments, and the above-described embodiments of the present invention are merely provided for description and do not represent the merits of the embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A speckle interference image noise reduction method is characterized by comprising the following steps:
converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation;
respectively carrying out noise reduction processing on the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transformation to obtain a sine noise reduction phase diagram and a cosine noise reduction phase diagram;
and determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
2. The method of claim 1, wherein converting the discontinuous original phase map into continuous sine phase map and cosine phase map by sine transform and cosine transform comprises:
performing sinusoidal transformation on a phase value of any point in the original phase diagram to obtain a sinusoidal phase diagram;
and performing cosine transformation on the phase value of any point in the original image to obtain the cosine phase diagram.
3. The method of claim 2, wherein the resolution of the original phase map is M x N, and the phase value at any point (M, N) in the original phase map is M x N
Wherein M is more than or equal to M and more than or equal to 1, and N is more than or equal to N and more than or equal to 1;
the sinusoidal phase map S (m, n) is:
the cosine phase diagram C (m, n) is:
4. the method according to claim 1, wherein the denoising processing is performed on the sine phase map and the cosine phase map by using stationary wavelet transform to obtain a sine denoising phase map and a cosine denoising phase map, and the method comprises:
respectively performing stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram to obtain a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram;
and respectively carrying out threshold denoising and wavelet inverse transformation on the wavelet decomposition result corresponding to the sine phase diagram and the wavelet decomposition result corresponding to the cosine phase diagram to obtain the sine denoising phase diagram and the cosine denoising phase diagram.
5. The method of claim 4, wherein the performing a stationary wavelet decomposition on the sine phase map and the cosine phase map respectively comprises:
respectively carrying out 5-layer stationary wavelet decomposition on the sine phase diagram and the cosine phase diagram by adopting a Daubechies wavelet function with an extinction moment of 4 through the following formulas:
wherein a is a detail coefficient, aAj、aHj、aVjAnd aDjApproximate detail, horizontal detail, vertical detail and diagonal detail of the stationary wavelet J (J ═ 1, 2, … J) layer decomposition respectively,
in the form of a low-pass digital filter,
for a high-pass digital filter, the upper corner marks col and row represent the operation in the column direction and row direction of the two-dimensional image, respectively.
6. The method according to claim 4, wherein the performing threshold denoising and wavelet inverse transformation on the wavelet decomposition result corresponding to the sine phase map and the wavelet decomposition result corresponding to the cosine phase map to obtain the sine denoising phase map and the cosine denoising phase map respectively comprises:
determining a threshold lambda by adopting a birgBe-Massart strategy with high penalty parameters, wherein the birgBe-Massart strategy is as follows:
wherein α is a high penalty parameter, 2.5 & lt α & lt 10, sigma is the standard deviation of Gaussian noise in the error model, ckWavelet coefficients whose absolute values are arranged in descending order, l being the sum of the number of coefficients, the variable t being such that t*Obtaining a minimum value;
according to the threshold lambda, threshold denoising is carried out on a wavelet decomposition result corresponding to the sine phase diagram and a wavelet decomposition result corresponding to the cosine phase diagram by using a soft threshold function, and stationary wavelet inverse transformation is carried out on a denoised image to obtain the sine denoising phase diagram and the cosine denoising phase diagram, wherein the soft threshold function is as follows:
ajoriginal detail coefficients for stationary wavelet transform, cjThe detail coefficient after the threshold value.
7. The method of claim 1, wherein determining a target noise reduced phase map from the sine noise reduced phase map and the cosine noise reduced phase map comprises:
and determining the target noise reduction phase diagram by a four-quadrant arc tangent algorithm according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
8. The method of claim 7, wherein determining the target denoising phase map by a four-quadrant arctangent algorithm from the sine denoising phase map and the cosine denoising phase map comprises:
calculating the target noise reduction phase diagram phi (m, n) by the following formula:
wherein S '(m, n) is the sine noise reduction phase map, and C' (m, n) is the cosine noise reduction phase map.
9. A speckle interference image noise reduction device, comprising:
the first processing module is used for converting the discontinuous original phase diagram into a continuous sine phase diagram and a continuous cosine phase diagram through sine transformation and cosine transformation;
the noise reduction module is used for respectively carrying out noise reduction processing on the sine phase diagram and the cosine phase diagram by utilizing stationary wavelet transform to obtain a sine noise reduction phase diagram and a cosine noise reduction phase diagram;
and the second processing module is used for determining a target noise reduction phase diagram according to the sine noise reduction phase diagram and the cosine noise reduction phase diagram.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910980587.XA CN110766627A (en) | 2019-10-16 | 2019-10-16 | Speckle interference image noise reduction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910980587.XA CN110766627A (en) | 2019-10-16 | 2019-10-16 | Speckle interference image noise reduction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110766627A true CN110766627A (en) | 2020-02-07 |
Family
ID=69331392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910980587.XA Pending CN110766627A (en) | 2019-10-16 | 2019-10-16 | Speckle interference image noise reduction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110766627A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113160088A (en) * | 2021-04-30 | 2021-07-23 | 河南大学 | Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101527037A (en) * | 2009-04-13 | 2009-09-09 | 天津大学 | Method for denoising stationary wavelet image based on neighborhood windowing |
| CN103854264A (en) * | 2014-03-28 | 2014-06-11 | 中国石油大学(华东) | Improved threshold function-based wavelet transformation image denoising method |
| CN104715461A (en) * | 2015-04-02 | 2015-06-17 | 哈尔滨理工大学 | Image noise reduction method |
| CN107563969A (en) * | 2017-08-03 | 2018-01-09 | 天津大学 | DSPI phase filtering methods based on variation mode decomposition |
| CN109242799A (en) * | 2018-09-19 | 2019-01-18 | 安徽理工大学 | A kind of Wavelet noise-eliminating method of variable threshold value |
-
2019
- 2019-10-16 CN CN201910980587.XA patent/CN110766627A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101527037A (en) * | 2009-04-13 | 2009-09-09 | 天津大学 | Method for denoising stationary wavelet image based on neighborhood windowing |
| CN103854264A (en) * | 2014-03-28 | 2014-06-11 | 中国石油大学(华东) | Improved threshold function-based wavelet transformation image denoising method |
| CN104715461A (en) * | 2015-04-02 | 2015-06-17 | 哈尔滨理工大学 | Image noise reduction method |
| CN107563969A (en) * | 2017-08-03 | 2018-01-09 | 天津大学 | DSPI phase filtering methods based on variation mode decomposition |
| CN109242799A (en) * | 2018-09-19 | 2019-01-18 | 安徽理工大学 | A kind of Wavelet noise-eliminating method of variable threshold value |
Non-Patent Citations (4)
| Title |
|---|
| LEBRON_XUEYAN: "软阈值小波变换 Birge-Massart策略整理", 《HTTPS://BLOG.CSDN.NET/ZXY115019/ARTICLE/DETAILS/80407087》 * |
| 王慧琴: "《小波理论与应用》", 31 January 2011 * |
| 矫媛，赵志刚，黄斌文: "基于平稳小波变换的邻域依赖自适应软阈值图像降噪", 《科技信息》 * |
| 肖启阳，李 健，吴思进，杨连祥，董明利，曾周末: "基于能量的DSPI 相位图正余弦降噪", 《光学精密工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113160088A (en) * | 2021-04-30 | 2021-07-23 | 河南大学 | Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy |
| CN113160088B (en) * | 2021-04-30 | 2022-08-12 | 河南大学 | Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wadhwa et al. | Phase-based video motion processing | |
| Gao et al. | An improved Sobel edge detection | |
| CN105205788B (en) | A kind of denoising method for high-throughput gene sequencing image | |
| CN112254801B (en) | Micro-vibration vision measurement method and system | |
| JP2001266140A (en) | Method and device for judging modulation and phase of two-dimensional pattern | |
| CN109087348B (en) | Single-pixel imaging method based on adaptive area projection | |
| Fernandez et al. | Automatic window size selection in Windowed Fourier Transform for 3D reconstruction using adapted mother wavelets | |
| Ning et al. | Fast phase denoising using stationary wavelet transform in speckle pattern interferometry | |
| CN105550998B (en) | Image enchancing method and Image Intensified System based on two generation small echo integer transforms | |
| CN114779343A (en) | Seismic data denoising method based on curvelet transform-joint bilateral filtering | |
| CN110766627A (en) | Speckle interference image noise reduction method and device | |
| CN105976340A (en) | Improved spin filtering algorithm based on wavelet decomposition | |
| CN114355348A (en) | SAR interferogram wavelet denoising processing method and processing device thereof | |
| CN110930433A (en) | High-speed moving object machine vision target detection method based on image restoration | |
| CN110060220A (en) | Based on the image de-noising method and system for improving BM3D algorithm | |
| Priyadharsini et al. | Underwater acoustic image enhancement using wavelet and KL transform | |
| CN110221349B (en) | Transient electromagnetic signal noise reduction method based on wavelet transformation and sine wave estimation | |
| Sadaghiani et al. | Image interpolation based on 2D-DWT with novel regularity-preserving algorithm using RLS adaptive filters | |
| CN106469438B (en) | Neighborhood based on card side's unbiased esti-mator shrinks MRI denoising method | |
| Fu et al. | Evaluation of motion blur parameter based on cepstrum domain of the intentional restored image | |
| Hu et al. | A new method and implementation of blind restoration algorithm for moving fuzzy license plate image based on frequency-domain characteristics | |
| CN109697704A (en) | Adaptive full variation ESPI image denoising method and system based on BM3D algorithm | |
| CN113506212B (en) | Improved hyperspectral image super-resolution reconstruction method based on POCS | |
| Yavariabdi et al. | 3D Medical Image Enhancement based on Wavelet Transforms. | |
| CN111028152B (en) | Super-resolution reconstruction method of sonar image based on terrain matching |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200207 |
|
| WD01 | Invention patent application deemed withdrawn after publication |