CN108109123B - Image denoising method - Google Patents
Image denoising method Download PDFInfo
- Publication number
- CN108109123B CN108109123B CN201711391066.8A CN201711391066A CN108109123B CN 108109123 B CN108109123 B CN 108109123B CN 201711391066 A CN201711391066 A CN 201711391066A CN 108109123 B CN108109123 B CN 108109123B
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- processed
- pixels
- noise
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001914 filtration Methods 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims description 10
- 239000000126 substance Substances 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 abstract description 4
- 230000008030 elimination Effects 0.000 abstract description 2
- 238000003379 elimination reaction Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 9
- 238000012805 post-processing Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000007613 environmental effect 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
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- 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 by non-spatial domain filtering
Abstract
The invention discloses an image denoising method, which comprises the following steps: s01: in the image to be processed, an A × A region centered on a pixel F (i, j) is formed and in the regionThe variance σ between the pixel F (i, j) and the surrounding pixels is determined in the domain2 S(i,j)And S02: in an A × A region centered on a pixel F (i, j), a variance σ between a noise estimation value of the pixel F (i, j) and noise estimation values of surrounding pixels is obtained2 N(i,j)And S03: determining the area range of the pixel F (i, j), S04: if the pixel F (i, j) is located in the flat area, filtering the pixel and outputting a processed pixel value; if the pixel F (i, j) is located in the detail area, the original pixel value of the pixel is directly output without processing the pixel; and outputting the pixels which are not processed in the image to be processed according to the original values to form a processed image. The invention solves the problem that the existing image noise elimination technology destroys the image texture details while filtering noise.
Description
Technical Field
The invention relates to an image processing technology, in particular to an image denoising method.
Background
In current production life, digital images play an extremely important role. In the signal processing and computer vision fields, the data acquired from the image sensor is often contaminated by various noises, such as undesired equipment, data acquisition quantification and external interference, which degrade the data degradation and bring noises during transmission, reception and processing, thereby greatly hindering the analysis and understanding of subsequent images. In order to improve the image quality and improve the effectiveness and reliability of image post-processing, it is necessary to employ noise reduction techniques to filter out image noise. Since existing noise reduction methods often cause image distortion and blurring, noise cancellation remains a very challenging problem. Signals or images in reality are often interfered by various noises in the processes of generation, transmission, reception and processing. Noise distorts the signal or image, severely affecting post-processing and analysis, and even making it difficult to achieve the intended goal. The noise of an image can be classified into system noise and environmental noise from the source, the system noise is noise from the system itself, and the environmental noise is noise from an external environment. In order to obtain a good quality image, it is usually necessary to process and optimize the image by an image denoising algorithm. The conventional image noise reduction methods include the following methods:
(1) median filtering: the method is suitable for filtering salt and pepper noise of the image. However, the filtering effect of image noise with more details, especially with more details of points, lines and spires, is not obvious.
(2) And (3) mean filtering: the mean filtering using the neighborhood averaging method is suitable for removing grain noise in an image obtained by scanning, and although strongly suppressing the noise, it also causes a blurring phenomenon due to averaging.
(3) Adaptive wiener filtering: the method adjusts the output of the filter according to the local variance of the image, the larger the local variance is, the stronger the smoothing effect of the filter is, the better the filtering effect of the adaptive wiener filtering is than the average filtering effect, and the method is useful for keeping the details of the image and other high-frequency parts, but has larger calculation amount.
(4) Wavelet denoising: most of the wavelet coefficients containing signals are reserved, and the details of the image can be well maintained.
In the image denoising process, the traditional algorithms usually adopt the same filtering algorithm for all images, that is, the noise-free part of the images is also processed, so that the noise-free images are interfered by the filtering algorithm, the image quality is affected, meanwhile, the operation amount is increased, and the image processing efficiency is reduced.
Disclosure of Invention
The technical problem to be solved by the invention is to provide an image denoising method, which can distinguish a flat area and a detail area in an image to be processed, and solve the problem that the existing image noise elimination technology destroys image texture details while filtering noise.
In order to achieve the purpose, the invention adopts the following technical scheme: an image denoising method comprises the following steps:
s01: in an image to be processed, an A × A region centered on a pixel F (i, j) is formed, and a variance σ between the pixel F (i, j) and surrounding pixels is found in the region2 S(i,j)Wherein, the image to be processed is composed of M rows and N columns of pixels, M is an integer greater than or equal to 3, N is an integer greater than or equal to 3, A is an odd number greater than 3 and smaller than M and N,
s02: in an A × A region centered on a pixel F (i, j), a variance σ between a noise estimation value of the pixel F (i, j) and noise estimation values of surrounding pixels is obtained2 N(i,j)The method comprises the following specific steps:
s0201: in the image to be processed, an A × A region centered on a pixel F (i, j) is formed, and A in the A × A region centered on the pixel F (i, j) is calculated2The noise estimation value image is obtained by replacing the pixel value of the pixel F (i, j) in the image to be processed with the noise estimation value of the pixel F (i, j);
s0202: in the noise estimation value image, an A × A region centered on a pixel F ″ (i, j) is formed, and a variance σ between the pixel F ″ (i, j) and surrounding pixels is obtained in the A × A region2 N(i,j)；
S03: judging the area range of the pixel F (i, j), and adopting the following formula:
s04: if the pixel F (i, j) is located in the flat area, filtering the pixel and outputting a processed pixel value; if the pixel F (i, j) is located in the detail area, the original pixel value of the pixel is directly output without processing the pixel; and outputting the pixels which are not processed in the image to be processed according to the original values to form a complete processed image.
Further, before the noise estimation value image is obtained in step S02, noise is added to each pixel in the image to be processed to obtain a noise image in which an a × a region centered on the pixel F '(i, j) is formed, and an a in the a × a region centered on the pixel F' (i, j) is calculated2Replacing the pixel value of the pixel F '(i, j) in the noise image by the noise estimation value of the pixel F' (i, j) to obtain a noise estimation value image;
further, white gaussian noise having a standard deviation of 30 was added to each pixel in the image to be processed.
Further, in step S04, a gaussian template is used to perform filtering processing on the flat region pixels.
Further, the variance of the pixel F (i, j) and the surrounding pixels in the step S01
Wherein the content of the first and second substances,
is the pixel value, X, corresponding to the pixel F (i, j) in the image to be processedkIs the pixel value corresponding to the kth pixel in an A multiplied by A area with the pixel F (i, j) as the center, wherein k is more than or equal to 1 and less than or equal to A2。
Further, the variance of the pixel F ″ (i, j) in step S0202 with the surrounding pixels
Wherein the content of the first and second substances,
for the pixel value, X, corresponding to pixel F ″ (i, j) in the noise estimate imagek"is the first in the A × A region centered on the pixel F" (i, j)Pixel values corresponding to k pixels, wherein k is more than or equal to 1 and less than or equal to A2。
Further, a equals 3.
Further, pixels which are not processed in the image to be processed comprise a first row of pixels, an Mth row of pixels, a first column of pixels and an Mth column of pixels.
Further, the calculation of each pixel in the image to be processed and the noise estimate image is performed according to the order of
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
The train follows a serpentine path of the same kind until the completion of the first
And denoising the line pixels.
Further, the range of the region control threshold T1 in the step S04 is 1.0-1.8.
The invention has the beneficial effects that: the invention provides an image denoising method for protecting a detail region, which judges a flat region and the detail region by using a noise estimation value, has high processing speed, low hardware implementation resource overhead and convenient implementation, and processes the detail region and the flat region separately, thereby not only obtaining a better effect of the flat region, but also reserving the detail region in the original image to be processed, reserving the texture detail in the image to be processed, and further improving the image denoising effect.
Drawings
FIG. 1 is a flowchart of an image denoising method according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the present invention provides an image denoising method, which comprises the following steps:
s01: : in an image to be processed, an A × A region centered on a pixel F (i, j) is formed, and a variance σ between the pixel F (i, j) and surrounding pixels is found in the region2 S(i,j)Wherein, the image to be processed is composed of M rows and N columns of pixels, M is an integer greater than or equal to 3, N is an integer greater than or equal to 3, A is an odd number greater than 3 and smaller than M and N,
in the invention, A is an odd number such as 3, 5, 7 and the like, and the denoising method of the invention is not influenced by different values of A, but the smaller the value of A is, the better the denoising effect finally obtained by the invention is. In the invention, in order to obtain a denoised image with a better effect, A is equal to 3, and when A takes other odd values, the subsequent denoising step is not influenced.
In this embodiment, the variance between the pixel F (i, j) and the surrounding pixels is described by taking a as an example of 3
In order to facilitate hardware implementation, the variance calculation formula is simplified in the invention, and the following calculation formula is adopted:
wherein the content of the first and second substances,
is the pixel value, X, corresponding to the pixel F (i, j) in the image to be processedkIs the pixel value corresponding to the kth pixel in a 3 x 3 region with the pixel F (i, j) as the center, wherein k is more than or equal to 1 and less than or equal to 9.
In the step, except that the first row of pixels, the Mth row of pixels, the first column of pixels and the Mth column of pixels in the image to be processed are not calculated, each of the rest pixels corresponds to a variance σ2 S(i,j)And the order of computation does not affect the post-processing results of the present invention. To order the computation, the present invention preferably takes a serpentine path from row 2, column 2 to row 2, column N-1, row 3, column N-1 to row 3, column 2, row 4, column 2 to row 4, column N-1, … …, and so on, of the image to be processed until the pixel computation of row M-1 is completed.
When A takes other values, the same snake-shaped track is adopted for pixel denoising treatment, which specifically comprises the following steps: the calculation of each pixel in the image to be processed, the noise image and the noise estimation value image is performed according to the order of
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
The train follows a serpentine path of the same kind until the completion of the first
And denoising the line pixels.
S02: in a 3 × 3 region centered on a pixel F (i, j), a variance σ between a noise estimation value of the pixel F (i, j) and noise estimation values of surrounding pixels is obtained2 N(i,j)The method comprises the following specific steps:
s0201: in the image to be processed, a 3 × 3 region centered on the pixel F (i, j) is formed, a mean value of pixel values of 3 × 3 pixels in the 3 × 3 region centered on the pixel F (i, j) is calculated, an absolute value of a difference between the mean value and the pixel value of the pixel F (i, j) is a noise estimation value of the pixel F (i, j), and the noise estimation value of the pixel F (i, j) in the image to be processed is substituted for the pixel value of the pixel F (i, j) in the image to be processed, so that a noise estimation value image is obtained.
Meanwhile, the invention can also adopt the following method to obtain the noise estimation value image: adding noise to each pixel in the image to be processed to obtain a noise image, forming a 3 × 3 area with a pixel F '(i, j) as a center in the noise image, calculating a mean value of pixel values of 3 × 3 pixels in the 3 × 3 area with the pixel F' (i, j) as the center, wherein an absolute value of a difference between the mean value and the pixel value of the pixel F '(i, j) is a noise estimation value of the pixel F' (i, j), and replacing the pixel value of the pixel F '(i, j) in the noise image with the noise estimation value of the pixel F' (i, j) to obtain a noise estimation value image. The noise added to the image to be processed in the present invention can be any noise in the prior art, and in the present embodiment, gaussian white noise with a standard deviation of 30 is added.
It should be noted that when a is 3, the relationship between the noise estimation value image and the image to be processed is: and replacing pixels except for the first row of pixels, the Mth row of pixels, the first column of pixels and the Mth column of pixels in the image to be processed with corresponding noise estimation values, namely forming noise estimation value pixels. When A takes other values, pixels except pixels which are not processed in the image to be processed are replaced by corresponding noise estimation values, namely noise estimation value pixels are formed. Therefore, the image to be processed in the invention is a true image, and the corresponding analog image, noise image and noise estimation value image are all images virtualized for the denoising method in the invention. In order to make the present invention easy to understand, the present invention designates a pixel in the noise estimate value image corresponding to the pixel F (i, j) in the image to be processed as F "(i, j).
S0202: in the noise estimation value image, a 3 × 3 region centered on a pixel F ″ (i, j) is formed, and an image is obtained in the regionThe variance σ of the pixel F "(i, j) with the surrounding pixels2 N(i,j)。
The calculation method thereof is similar to the calculation in step S01. Variance of pixel F "(i, j) with surrounding pixels
In order to facilitate hardware implementation, the variance calculation formula is simplified in the invention, and the following calculation formula is adopted:
wherein the content of the first and second substances,
for the pixel value, X, corresponding to pixel F ″ (i, j) in the noise estimate imagek"is a pixel value corresponding to the kth pixel in a 3 × 3 region centered on pixel F" (i, j), where 1. ltoreq. k.ltoreq.9.
S03: judging the area range of the pixel F (i, j), and the image to be processed has similarity in a flat area and abrupt change in a detailed area. Because of this property, the variance of the gray value of the image in a flat area is small, and the variance in a detail area is large, and according to this characteristic, we judge the area where the image is located according to the following formula.
Wherein, the range of the region control threshold T1 is 1.0-1.8 according to a large amount of test data, and the region control threshold T1 adopted in the invention is 1.4.
S04: if the pixel F (i, j) is located in the flat area, filtering the pixel and outputting a processed pixel value; if the pixel F (i, j) is located in the detail area, the original pixel value of the pixel is directly output without processing the pixel; and outputting the pixels which are not processed in the image to be processed according to the original values.
The flat area can be processed by adopting the filtering technology in the prior art because the flat area has no too many details and edges, the invention adopts a 3 multiplied by 3 Gaussian template for filtering, firstly, the Gaussian template has better effect of protecting image details compared with the mean value filtering, secondly, the Gaussian template is easier to realize by hardware for other filtering, occupies less resources, and secondly, the drying is mainly aimed at Gaussian white noise. The detail regions separated by the method basically cover most of detail textures of the whole image, and in order to protect the detail information and ensure that the separated detail regions are more accurate, the detail regions are not denoised, and the original pixel value output is kept.
The pixels in the flat area in the image to be processed are output after filtering processing, the pixels in the detail area are output without processing, and the pixels which are not processed in the image to be processed are output without processing, wherein the pixels which are not processed comprise the pixels in the first row, the pixels in the Mth row, the pixels in the first column and the pixels in the Mth column in the image to be processed; and finally forming a processed image.
The above description is only a preferred embodiment of the present invention, and the embodiment is not intended to limit the scope of the present invention, so that all equivalent structural changes made by using the contents of the specification and the drawings of the present invention should be included in the scope of the appended claims.
Claims (10)
1. An image denoising method is characterized by comprising the following steps:
s01: in an image to be processed, an A × A region centered on a pixel F (i, j) is formed, and a variance σ between the pixel F (i, j) and surrounding pixels is found in the region2 S(i,j)Wherein, the image to be processed is composed of M rows and N columns of pixels, M is an integer greater than or equal to 3, N is an integer greater than or equal to 3, A is an odd number greater than 3 and smaller than M and N,
s02: in an A × A region centered on a pixel F (i, j), a variance σ between a noise estimation value of the pixel F (i, j) and noise estimation values of surrounding pixels is obtained2 N(i,j)The method comprises the following specific steps:
s0201: in the image to be processed, an A × A region centered on a pixel F (i, j) is formed, and A in the A × A region centered on the pixel F (i, j) is calculated2The noise estimation value image is obtained by replacing the pixel value of the pixel F (i, j) in the image to be processed with the noise estimation value of the pixel F (i, j);
s0202: in the noise estimation value image, an A × A region centered on a pixel F ″ (i, j) is formed, and a variance σ between the pixel F ″ (i, j) and surrounding pixels is obtained in the A × A region2 N(i,j)；
S03: judging the area range of the pixel F (i, j), and adopting the following formula:
s04: if the pixel F (i, j) is located in the flat area, filtering the pixel and outputting a processed pixel value; if the pixel F (i, j) is located in the detail area, the original pixel value of the pixel is directly output without processing the pixel; and outputting the pixels which are not processed in the image to be processed according to the original values to form a complete processed image.
2. An image denoising method according to claim 1, wherein before the noise estimation value image is obtained in step S02, noise is added to each pixel in the image to be processed to obtain a noise image, an a × a region centered on the pixel F '(i, j) is formed in the noise image, and a in the a × a region centered on the pixel F' (i, j) is calculated2Image of a pixelAnd replacing the pixel value of the pixel F '(i, j) in the noise image by the noise estimation value of the pixel F' (i, j) to obtain a noise estimation value image.
3. An image denoising method according to claim 2, wherein a white gaussian noise with a standard deviation of 30 is added to each pixel in the image to be processed.
4. An image denoising method according to any one of claims 1 to 3, wherein the step S04 is performed by filtering the flat region pixels using a Gaussian template.
5. An image denoising method according to claim 1, wherein the variance of the pixel F (i, j) and the surrounding pixels in step S01 is
Wherein the content of the first and second substances,
is the pixel value, X, corresponding to the pixel F (i, j) in the image to be processedkIs the pixel value corresponding to the kth pixel in an A multiplied by A area with the pixel F (i, j) as the center, wherein k is more than or equal to 1 and less than or equal to A2。
6. An image denoising method according to claim 1, wherein the variance of the pixel F "(i, j) in step S0202 with the surrounding pixels
Wherein the content of the first and second substances,
for the pixel value, X, corresponding to pixel F ″ (i, j) in the noise estimate imagekIs "in pixels F"(i, j) as the center, wherein k is more than or equal to 1 and less than or equal to A2。
7. An image denoising method according to any one of claims 1, 2, 5 and 6, wherein a is equal to 3.
8. The method of claim 7, wherein the unprocessed pixels in the image to be processed comprise a first row of pixels, an Mth row of pixels, a first column of pixels, and an Mth column of pixels.
9. An image denoising method according to claim 1, wherein the calculation of each pixel in the image to be processed and the noise estimate image is performed according to the order of the second order
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
Line from the second
Go to the first
Is listed to the first
Go to the first
The train follows a serpentine path of the same kind until the completion of the first
And denoising the line pixels.
10. An image denoising method according to claim 1, wherein the region control threshold T1 in step S04 is in the range of 1.0-1.8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711391066.8A CN108109123B (en) | 2017-12-21 | 2017-12-21 | Image denoising method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711391066.8A CN108109123B (en) | 2017-12-21 | 2017-12-21 | Image denoising method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108109123A CN108109123A (en) | 2018-06-01 |
| CN108109123B true CN108109123B (en) | 2020-05-15 |
Family
ID=62210809
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711391066.8A Active CN108109123B (en) | 2017-12-21 | 2017-12-21 | Image denoising method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108109123B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636743A (en) * | 2018-11-26 | 2019-04-16 | 上海集成电路研发中心有限公司 | A kind of method and device removing picture noise |
| CN110097511B (en) * | 2019-04-11 | 2021-06-15 | 成都微光集电科技有限公司 | Image noise reduction method |
| CN110136085B (en) * | 2019-05-17 | 2022-03-29 | 凌云光技术股份有限公司 | Image noise reduction method and device |
| CN111507923B (en) * | 2020-04-21 | 2023-09-12 | 浙江大华技术股份有限公司 | Noise processing method, device, equipment and medium for video image |
| CN115829874A (en) * | 2022-03-31 | 2023-03-21 | 南通电博士自动化设备有限公司 | Noise processing method based on image smoothing |
| CN117423113B (en) * | 2023-12-18 | 2024-03-05 | 青岛华正信息技术股份有限公司 | Adaptive denoising method for archive OCR (optical character recognition) image |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103971343A (en) * | 2014-05-21 | 2014-08-06 | 浙江宇视科技有限公司 | Image denoising method based on similar pixel detection |
| CN104134191B (en) * | 2014-07-11 | 2017-11-28 | 三星电子（中国）研发中心 | Image de-noising method and its device |
| CN111915535B (en) * | 2015-12-30 | 2022-09-27 | 展讯通信(天津)有限公司 | Image denoising method and device |
| CN106447633B (en) * | 2016-09-26 | 2019-05-14 | 华南理工大学 | The denoising method of Microfocus X-ray radioscopic image towards integrated antenna package detection |
-
2017
- 2017-12-21 CN CN201711391066.8A patent/CN108109123B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108109123A (en) | 2018-06-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108109123B (en) | Image denoising method | |
| Ramesh et al. | THE MANAGEMENT AND REDUCTION OF DIGITAL NOISE IN VIDEO IMAGE PROCESSING BY USING TRANSMISSION BASED NOISE ELIMINATION SCHEME. | |
| CN109377450B (en) | Edge protection denoising method | |
| CN110796615A (en) | Image denoising method and device and storage medium | |
| CN115984148B (en) | Denoising enhancement method for high-throughput gene sequencing data | |
| CN108492268A (en) | Enhancement algorithm for low-illumination image based on wavelet coefficient fusion | |
| Arastehfar et al. | An enhanced median filter for removing noise from MR images | |
| CN111242854A (en) | Image denoising method | |
| CN100367770C (en) | Method for removing isolated noise point in video | |
| CN105809633A (en) | Color noise removing method and device | |
| Ali | A new method to remove salt & pepper noise in Magnetic Resonance Images | |
| CN108629740B (en) | Image denoising processing method and device | |
| Zhang et al. | Diffusion scheme using mean filter and wavelet coefficient magnitude for image denoising | |
| CN116912115A (en) | Underwater image self-adaptive enhancement method, system, equipment and storage medium | |
| CN111147804B (en) | Video frame reconstruction method | |
| CN111652810A (en) | Image denoising method based on wavelet domain singular value differential model | |
| Gao et al. | Image denoising based on edge detection and prethresholding Wiener filtering of multi-wavelets fusion | |
| CN114648469B (en) | Video image denoising method, system, device and storage medium thereof | |
| CN109685729A (en) | A kind of Poisson denoising method of Pixel-level threshold adaptive | |
| Al_taie | Restoration for blurred noisy images based on guided filtering and inverse filter | |
| CN109615597B (en) | Image speckle noise suppression method based on central affine filtering | |
| CN113487505A (en) | Infrared image mixed noise reduction method based on noise identification | |
| CN109191391B (en) | Attenuation parameter self-adaptive non-local mean image noise reduction method | |
| CN112465719A (en) | Transform domain image denoising method and system | |
| Victoria et al. | A survey on impulse noise removal techniques in image processing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |