CN111899196A - Blade defect motion blurred image restoration method based on classical restoration algorithm - Google Patents
Blade defect motion blurred image restoration method based on classical restoration algorithm Download PDFInfo
- Publication number
- CN111899196A CN111899196A CN202010776748.6A CN202010776748A CN111899196A CN 111899196 A CN111899196 A CN 111899196A CN 202010776748 A CN202010776748 A CN 202010776748A CN 111899196 A CN111899196 A CN 111899196A
- Authority
- CN
- China
- Prior art keywords
- image
- noise
- restoration
- fuzzy
- blurred 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.)
- Pending
Links
- 230000033001 locomotion Effects 0.000 title claims abstract description 39
- 230000007547 defect Effects 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001914 filtration Methods 0.000 claims description 31
- 238000001228 spectrum Methods 0.000 claims description 10
- 230000000694 effects Effects 0.000 claims description 5
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000006731 degradation reaction Methods 0.000 abstract description 16
- 230000015556 catabolic process Effects 0.000 abstract description 14
- 238000011084 recovery Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000009659 non-destructive testing Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation 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
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration using non-spatial domain filtering
-
- 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
- G06T5/70—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
- G06T2207/20172—Image enhancement details
- G06T2207/20201—Motion blur correction
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention provides a blade defect motion blurred image restoration method based on a classical restoration algorithm, relates to the technical field of wind power blade nondestructive testing, and provides a restoration algorithm for image restoration and application to wind power blade defect detection. Firstly, establishing a motion blur degradation model according to motion blur images and blade defect characteristics; then, a PSF (point spread function) is estimated based on the cepstrum method. And finally, carrying out experimental analysis on the recovery algorithm and summarizing the application range of the recovery algorithm. Experimental results show that each restoration algorithm can well restore the blade blurred image under specific conditions.
Description
Technical Field
The invention relates to the technical field of nondestructive testing of wind power blades, in particular to a blade defect motion blurred image restoration method based on a classical restoration algorithm.
Background
Wind energy is one of the most rapidly-developed renewable energy sources in the last two decades, and the proportion of the wind energy to the energy structure of the world is increased year by year, so that the wind energy has great development prospect. In the wind power generation equipment, the blade is one of the most core components, and the importance of detecting the defects of the fan blade is seen. The rapid development and the gradual maturity of a digital image processing technology and an unmanned aerial vehicle remote sensing technology provide a new means for detecting the surface defects of the wind power blade. At present, the blade defects are mainly detected by adopting an unmanned aerial vehicle to carry on a camera for image shooting. During the formation, recording processing and transmission of images, the images are often blurred due to the relatively fast relative movement between the camera and the wind power blade, and the phenomenon is called image degradation. The image restoration is to process the degraded image to restore the original view of the original image as much as possible.
The motion blur image restoration is an important branch of image restoration and is also the most common problem of wind power blade defect image acquisition. In the imaging process, because the camera carried by the unmanned aerial vehicle and the fan blade have rapid relative motion, the energy of the shutter is abnormally accumulated on an imaging plane in the process of one-time exposure, so that the motion blur of the obtained image is caused. The image restoration algorithm is generally based on a degradation model, and then restores an image according to a point spread function of the degraded image. Traditional image algorithms, such as inverse filter algorithms, wiener filter algorithms and least squares algorithms, are classical restoration algorithms that can recover motion blurred images in specific situations. The restoration algorithm is mainly used for carrying out defogging day blurring and motion blurring on the wind power blade defect image, and has great theoretical value and practical significance on the future wind power blade defect detection research.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a blade defect motion blurred image restoration method based on a classical restoration algorithm, which reduces the motion blurred influence on a blade defect image caused by the shooting of an unmanned aerial vehicle, so that the defect image is more clearly displayed.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
a leaf defect motion blurred image restoration method based on a classical restoration algorithm comprises the following steps:
step 1: simulating a blade defect motion blurred image according to the motion blurred mathematical model;
step 2: calculating a fuzzy angle and a fuzzy length of the blade defect motion blurred image based on a cepstrum method, and determining a point spread function;
step 2.1, performing two-dimensional Fourier transform on the blade defect motion blurred image G (x, y) to obtain G (u, v); wherein u and v are frequency and amplitude of a frequency domain in two-dimensional Fourier transform and respectively correspond to horizontal and vertical coordinates x and y under an image coordinate system;
step 2.2, obtaining absolute value of G (u, v) to obtain | G (u, v) | and obtaining logarithm of | G (u, v) |
Step 2.3, mixing
Performing inverse Fourier transform to obtain a cepstrum of g (x, y)
Step 2.4, mixing
Dividing the four parts into four parts with equal size, and carrying out diagonal exchange;
step 2.5, according to the symmetry of the cepstrum, finding out
Minimum position (x) of0,y0) Then, a world coordinate system is established by taking the cepstrum center as an original point, the length of a connecting line between the minimum point and the original point is a fuzzy length L, and an included angle formed by the connecting line and the positive direction of the X axis is a fuzzy angle alpha;
step 2.6, determining a point spread function h (x, y) according to the fuzzy length L and the fuzzy angle alpha;
and step 3: restoring the blurred image by using inverse filtering and wiener filtering;
the inverse filtering restores a noise-free blurred image; in the noise-free case, the inverse filter formula is as follows:
F(u,v)＝G(u,v)/H(u,v)
wherein, F (u, v) is Fourier transform of the original image F (x, y), and 1/H (u, v) is an inverse filter;
the wiener filtering restores the fuzzy image containing the noise by enabling the original image f (x, y) and the restored image
The restoration method with the minimum mean square error is also called minimum mean square error filtering, and the stronger the noise is, the better the restoration effect is, and the minimum mean square error filtering is as follows:
wherein E [. cndot ] is a mathematical expectation operator, and when the image noise mean is 0 or the noise and the image are statistically independent, the estimation of the original image is as follows:
in the formula, sn(u, v) is the noise power spectrum, sf(u, v) is the original image power spectrum,
as the noise-to-signal ratio, when the noise power sn(u, v) is much smaller than the original image power sf(u, v), where wiener filtering is converted to inverse filtering; when s isf(u, v) towards 0 or sf(u, v) is much smaller than sn(u, v) in the above-mentioned order,
also tends to be 0, and the like,the specific value of the image noise-signal ratio is replaced by a constant k, and the estimation of the original image is simplified as follows:
adopt the produced beneficial effect of above-mentioned technical scheme to lie in:
the invention provides a blade defect motion blurred image restoration method based on a classical restoration algorithm, which can effectively restore motion blur generated by images shot by an unmanned aerial vehicle. Firstly, estimating fuzzy parameters of a fuzzy image by using a cepstrum method, and flexibly coping with a fuzzy angle and a fuzzy length during shooting; and restoring by using a corresponding restoration algorithm according to the estimated fuzzy parameters, thereby realizing nondestructive detection of the surface defects of the blade, finding the defects in advance, restoring in advance and enabling the generator set to operate more stably.
Drawings
FIG. 1 is a flowchart of a method for restoring a motion-blurred image of a defective blade according to an embodiment of the present invention;
FIG. 2 is an image blur degradation model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of data exchange according to an embodiment of the present invention;
FIG. 4 is a blurred image spectrum and cepstrum map according to an embodiment of the present invention;
wherein diagram (a) -spectrogram; graph (b) -cepstrum plot;
FIG. 5 is an original image of a blade according to an embodiment of the present invention;
FIG. 6 illustrates a classic restoration algorithm to restore image contrast according to an embodiment of the present invention;
wherein graph (a) -noiseless motion blur graph; (b) -a noise-free inverse filter map; (c) -a noise-free wiener filter map; (d) -adding a noise motion blur map; (e) -adding a noise inverse filter map; (f) -adding a noise wiener filter map.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
A leaf defect motion blurred image restoration method based on a classical restoration algorithm is disclosed, as shown in FIG. 1, and comprises the following steps:
step 1: simulating a blade defect motion blurred image according to the motion blurred mathematical model;
the image restoration is to reconstruct the degraded image by using a certain empirical knowledge (degradation model) of the degraded image, so it is important to find the cause of the degradation, and to establish a corresponding mathematical degradation model according to the cause of the degradation, and restore the image along the inverse process of the degradation. In this embodiment, a common image degradation model in the prior art is used to model the degradation process as a degradation function h (x, y) acting on the original image f (x, y), and then the degradation function h (x, y) cooperates with an additive noise n (x, y) to generate a blurred image g (x, y). The degradation model is shown in fig. 2, and the degradation formula is:
g(x,y)＝h(x,y)*f(x,y)+n(x,y) (1)
where ". sup." denotes convolution operation, and x and y are the rows and columns of the image coordinate system.
In the time domain, the convolution operation corresponds to the multiplication of the frequency domain, and Fourier transformation is carried out on two sides of the formula (1) to obtain a frequency domain expression:
G(u,v)＝F(u,v)H(u,v)+N(u,v) (2)
wherein G (u, v), F (u, v), H (u, v) and N (u, v) are Fourier transforms corresponding to G (x, y), F (x, y), H (x, y) and N (x, y), respectively.
Under the conditions of no noise interference and unknown fuzzy parameters, parallel black stripes appear in the uniform linear motion blurred image frequency spectrum G (u, v), which is caused by the fact that G (u, v) is equal to zero, and the fuzzy length L and the value of the fuzzy angle alpha are estimated according to the width of the bright stripes passing through the origin and the stripe angle. In practical applications, noise is inevitable, so an effective estimation method is needed. Commonly used estimation methods include image observation estimation, experimental estimation, model estimation, and cepstrum estimation. This document focuses on analytical cepstral estimation.
For a blurred image g (x, y), the cepstrum is defined as equation (3). For convenience of description, it is meaningful to say that | G (u, v) | is 0, and it is usually expressed by formula (4)
In the formula F-1Is an inverse fourier transform.
Step 2: calculating a fuzzy angle and a fuzzy length of the blade defect motion blurred image based on a cepstrum method;
in the image degradation process, the key of restoring the image is to find the PSF (Point Spread Function) h (x, y), so the important Point of image restoration is the accuracy of h (x, y) estimation. The first step in achieving image restoration is to determine the parameters of the PSF, and for motion blurred images, the blur length L and the blur angle α are the most important blur parameters. The method specifically comprises the following steps:
step 2.1, performing two-dimensional Fourier transform on the blade defect motion blurred image G (x, y) to obtain G (u, v); wherein u and v are frequency and amplitude of a frequency domain in two-dimensional Fourier transform and respectively correspond to horizontal and vertical coordinates x and y under an image coordinate system;
step 2.2, obtaining | G (u, v) | by taking absolute value of G (u, v), and obtaining formula (5) by taking logarithm of | G (u, v) |
Step 2.3, mixing
Performing inverse Fourier transform to obtain a cepstrum of g (x, y)
Step 2.4, mixing
Dividing the four parts into four parts with equal size, and carrying out diagonal exchange; as shown in fig. 3. The number of exchanges 1 and 4, 2 and 3,the origin is moved from the upper left corner to the center position.
Step 2.5, according to the symmetry of the cepstrum, finding out
Minimum position (x) of0,y0) Then, a world coordinate system is established by taking the cepstrum center as an original point, the length of a connecting line between the minimum point and the original point is a fuzzy length L, and an included angle formed by the connecting line and the positive direction of the X axis is a fuzzy angle alpha;
step 2.6, determining a point spread function h (x, y) according to the fuzzy length L and the fuzzy angle alpha;
in this embodiment, a wind power blade with cracks in a laboratory is selected as an original image, a blur angle is 45 degrees, a blur length is 20 pixels, and a blur image frequency spectrum and a cepstrum are shown in fig. 4.
And step 3: and restoring the blurred image by using inverse filtering and wiener filtering.
Because the inverse filtering can amplify the noise, a proper filtering method is selected according to the existence of the noise in the image;
inverse filtering is the estimation of the original from the fourier transform of the image, which, in the absence of noise,
F(u,v)＝G(u,v)/H(u,v) (6)
combining the formula (2) to obtain an inverse filter formula:
wherein 1/H (u, v) is an inverse filter,
to restore the image spectrum, the image is processed
Performing inverse Fourier transform to obtain restorationImage f (x, y).
As derived from the inverse filter equation, when H (u, v) is 0 or small,
this term becomes very large, which corresponds to amplifying the noise many times and the original information of the image will be covered. In practice, H (u, v) decreases rapidly as u, v increases from the origin, and N (u, v) shifts more slowly, so that the noise is amplified many times further away from the origin. So in general, the inverse filter is not exactly 1/H (u, v), but another function with respect to u, v, commonly denoted as M (u, v) and called the recovery transfer function, as follows:
in the formula, w0Is a range other than H (u, v) ═ 0. It can be seen from the formula that if the image contains noise, the image restoration effect by using the inverse filtering method is very poor. In order to restore the image as much as possible and reduce the noise and the influence of H (u, v) being 0 or small, the restoration transfer function is changed to:
in the formula, k and d are constants less than 1, and the smaller d is selected, the better.
From the above analysis, it can be found that the inverse filter restoration algorithm has the best restoration effect on a noise-free image, and has a good restoration effect on an image with a high signal-to-noise ratio even if the image contains noise. The inverse filtering algorithm has high operation speed and is also suitable for images with large image information, so that the noise-free blurred images are restored by using inverse filtering.
The wiener filtering is to make the original image f (x, y) and the restored image
Recovery method with minimum mean square error, so called minimum mean square errorDifferential filtering, namely:
where E [. cndot. ] is the mathematical expectation operator. Assuming that the average noise of the image is 0 or the noise is statistically independent from the image, the estimation of the original image is:
in the formula, sn(u, v) is the noise power spectrum, sf(u, v) is the original image power spectrum,
is the noise to signal ratio. When noise power sn(u, v) is much smaller than the original image power sf(u, v) (or the noise is negligibly small), at which point the wiener filtering is transformed to inverse filtering. When s isf(u, v) tends to 0(s)f(u, v) is much less than sn(u, v) in the above-mentioned state,
also tending towards 0. In practical situations, it is difficult to obtain a specific value of the image noise-to-signal ratio, so that the image noise-to-signal ratio is replaced by a constant k, and equation (11) is simplified as follows:
the experiment uses Python language and IDE (Integrated development Environment) Pycharm 2017.1, firstly, the original leaf image figure 4 is blurred with the length of 20 degrees and the blurring angle of 45 degrees, then the mean value is 0, the variance is 10-3The Gaussian white noise is restored by the restoration algorithm to the image, and FIG. 5 is the original image of the blade in the embodiment of the invention; the recovery result is shown in fig. 6, and the result is supported by RGB three channels.
The method applies the image restoration technology to wind power blade defect detection, estimates the blur parameters of the motion blur image through a cepstrum method, and flexibly deals with the motion blur of various angles and lengths; and calculating a point spread function according to the estimated fuzzy parameters, applying the point spread function to inverse filtering and wiener filtering, and flexibly selecting a filtering method according to the existence of noise in the image to realize the restoration of the blade defect motion fuzzy image.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present disclosure is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure are mutually replaced to form the technical solution.
Claims (3)
1. A leaf defect motion blurred image restoration method based on a classical restoration algorithm is characterized by comprising the following steps: the method comprises the following steps:
step 1: simulating a blade defect motion blurred image according to the motion blurred mathematical model;
step 2: calculating a fuzzy angle and a fuzzy length of the blade defect motion blurred image based on a cepstrum method, and determining a point spread function;
and step 3: and restoring the blurred image by using inverse filtering and wiener filtering.
2. The method for restoring a blade defect motion-blurred image based on a classical restoration algorithm according to claim 1, wherein the step 2 specifically comprises:
step 2.1, performing two-dimensional Fourier transform on the blade defect motion blurred image G (x, y) to obtain G (u, v); wherein u and v are frequency and amplitude of a frequency domain in two-dimensional Fourier transform and respectively correspond to horizontal and vertical coordinates x and y under an image coordinate system;
step 2.2, obtaining absolute value of G (u, v) to obtain | G (u, v) | and obtaining logarithm of | G (u, v) |
Step 2.3, mixing
Performing inverse Fourier transform to obtain a cepstrum of g (x, y)
Step 2.4, mixing
Dividing the four parts into four parts with equal size, and carrying out diagonal exchange;
step 2.5, according to the symmetry of the cepstrum, finding out
Minimum position (x) of0,y0) Then, a world coordinate system is established by taking the cepstrum center as an original point, the length of a connecting line between the minimum point and the original point is a fuzzy length L, and an included angle formed by the connecting line and the positive direction of the X axis is a fuzzy angle alpha;
step 2.6, determining a point spread function h (x, y) according to the fuzzy length L and the fuzzy angle alpha;
3. the method for restoring a blade defect motion blurred image based on a classical restoration algorithm according to claim 1, wherein the inverse filtering restores the noise-free blurred image in step 3; in the noise-free case, the inverse filter formula is as follows:
F(u,v)＝G(u,v)/H(u,v)
wherein F (u, v) is Fourier transform of the original image F (x, y), 1/H (u, v) is an inverse filter, wherein H (u, v) is Fourier transform of H (x, y);
the wiener filtering restores the fuzzy image containing the noise by enabling the original image f (x, y) and the restored image
The restoration method with the minimum mean square error is also called minimum mean square error filtering, and the stronger the noise is, the better the restoration effect is, and the minimum mean square error filtering is as follows:
wherein E [. cndot ] is a mathematical expectation operator, and when the image noise mean is 0 or the noise and the image are statistically independent, the estimation of the original image is as follows:
in the formula, sn(u, v) is the noise power spectrum, sf(u, v) is the original image power spectrum,
as the noise-to-signal ratio, when the noise power sn(u, v) is much smaller than the original image power sf(u, v), where wiener filtering is converted to inverse filtering; when s isf(u, v) towards 0 or sf(u, v) is much smaller than sn(u, v) in the above-mentioned order,
also tends to 0, the specific value of the image noise-to-signal ratio is replaced by a constant k, and the estimation of the original image is simplified as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010776748.6A CN111899196A (en) | 2020-08-05 | 2020-08-05 | Blade defect motion blurred image restoration method based on classical restoration algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010776748.6A CN111899196A (en) | 2020-08-05 | 2020-08-05 | Blade defect motion blurred image restoration method based on classical restoration algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111899196A true CN111899196A (en) | 2020-11-06 |
Family
ID=73246677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010776748.6A Pending CN111899196A (en) | 2020-08-05 | 2020-08-05 | Blade defect motion blurred image restoration method based on classical restoration algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111899196A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465776A (en) * | 2020-11-26 | 2021-03-09 | 常州信息职业技术学院 | Intelligent crack detection method based on fuzzy image on surface of wind turbine |
| CN114004759A (en) * | 2021-10-22 | 2022-02-01 | 贵州电网有限责任公司 | Restoration and reconstruction method for machine patrol defect picture image |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000026650A (en) * | 1998-10-22 | 2000-05-15 | Lg Electronics Inc | Apparatus and method for estimating motion blur information from image degenerated by motion blur of digital camera |
| CN103440620A (en) * | 2013-06-17 | 2013-12-11 | 中国航天科工集团第三研究院第八三五八研究所 | Method for automatically recognizing and restoring degraded images |
| CN104820969A (en) * | 2015-04-03 | 2015-08-05 | 西安交通大学 | Real-time blind image restoration method |
| CN105825480A (en) * | 2016-03-09 | 2016-08-03 | 西安电子科技大学 | Image deblurring method based on sapphire surface defect automatic detection |
| CN106373106A (en) * | 2016-09-21 | 2017-02-01 | 哈尔滨工业大学 | Cepstrum linear potential energy function-based industrial image motion blur suppression method |
| CN108305230A (en) * | 2018-01-31 | 2018-07-20 | 上海康斐信息技术有限公司 | A kind of blurred picture integrated conduct method and system |
-
2020
- 2020-08-05 CN CN202010776748.6A patent/CN111899196A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000026650A (en) * | 1998-10-22 | 2000-05-15 | Lg Electronics Inc | Apparatus and method for estimating motion blur information from image degenerated by motion blur of digital camera |
| CN103440620A (en) * | 2013-06-17 | 2013-12-11 | 中国航天科工集团第三研究院第八三五八研究所 | Method for automatically recognizing and restoring degraded images |
| CN104820969A (en) * | 2015-04-03 | 2015-08-05 | 西安交通大学 | Real-time blind image restoration method |
| CN105825480A (en) * | 2016-03-09 | 2016-08-03 | 西安电子科技大学 | Image deblurring method based on sapphire surface defect automatic detection |
| CN106373106A (en) * | 2016-09-21 | 2017-02-01 | 哈尔滨工业大学 | Cepstrum linear potential energy function-based industrial image motion blur suppression method |
| CN108305230A (en) * | 2018-01-31 | 2018-07-20 | 上海康斐信息技术有限公司 | A kind of blurred picture integrated conduct method and system |
Non-Patent Citations (1)
| Title |
|---|
| 郭瑞瑞: "运动模糊图像盲复原方法研究", 中国优秀硕士学位论文全文数据库信息科技辑, no. 3, pages 1 - 51 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112465776A (en) * | 2020-11-26 | 2021-03-09 | 常州信息职业技术学院 | Intelligent crack detection method based on fuzzy image on surface of wind turbine |
| CN112465776B (en) * | 2020-11-26 | 2023-10-31 | 常州信息职业技术学院 | Crack intelligent detection method based on wind turbine surface blurred image |
| CN114004759A (en) * | 2021-10-22 | 2022-02-01 | 贵州电网有限责任公司 | Restoration and reconstruction method for machine patrol defect picture image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9582862B2 (en) | Direction-adaptive image deblurring method | |
| Mosleh et al. | Explicit ringing removal in image deblurring | |
| CN111899196A (en) | Blade defect motion blurred image restoration method based on classical restoration algorithm | |
| Ai et al. | A refined bilateral filtering algorithm based on adaptively-trimmed-statistics for speckle reduction in SAR imagery | |
| Patel et al. | Hybrid approach for single image super resolution using ISEF and IBP | |
| Yang et al. | Remote sensing image restoration using estimated point spread function | |
| Gowri et al. | 2D Image data approximation using Savitzky Golay filter—Smoothing and differencing | |
| Karnaukhov et al. | Analysis of linear distortion characteristics in problems of restoration of multispectral images | |
| CN105825480A (en) | Image deblurring method based on sapphire surface defect automatic detection | |
| CN110929574A (en) | Infrared weak and small target rapid detection method | |
| CN113592738B (en) | Underwater distorted image restoration method | |
| Kerouh et al. | Wavelet-based blind blur reduction | |
| CN101236650B (en) | Real time rotating blurred image restoration method based on gradient loading optimization | |
| Junjie et al. | An image defocus deblurring method based on gradient difference of boundary neighborhood | |
| Shah et al. | Blind estimation of motion blur kernel parameters using Cepstral domain and Hough transform | |
| Panigrahi et al. | Joint bilateral filter for signal recovery from phase preserved curvelet coefficients for image denoising | |
| Chen et al. | Super-resolution reconstruction for underwater imaging | |
| Gatcha et al. | An anisotropic diffusion adaptive filter for image denoising and restoration applied on satellite remote sensing images: a case study | |
| Zhu et al. | Passive millimeter wave image denoising based on adaptive manifolds | |
| Yang et al. | Non-blind image blur removal method based on a Bayesian hierarchical model with hyperparameter priors | |
| Halder et al. | A centroid algorithm for stabilization of turbulence-degraded underwater videos | |
| Lata et al. | Novel method to assess motion blur kernel parameters and comparative study of restoration techniques using different image layouts | |
| Du et al. | Deblurring crack images generated by UAV camera shake | |
| Sun et al. | Blind Image Restoration Based on Power Law | |
| Huang et al. | Deblurring approach for motion camera combining FFT with α-confidence goal optimization |
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 |