CN103116879A - Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method - Google Patents
Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method Download PDFInfo
- Publication number
- CN103116879A CN103116879A CN2013100839229A CN201310083922A CN103116879A CN 103116879 A CN103116879 A CN 103116879A CN 2013100839229 A CN2013100839229 A CN 2013100839229A CN 201310083922 A CN201310083922 A CN 201310083922A CN 103116879 A CN103116879 A CN 103116879A
- Authority
- CN
- China
- Prior art keywords
- pixel
- denoising
- search
- local mean
- weights
- 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
Images
Landscapes
- Apparatus For Radiation Diagnosis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method for mainly improving original CT projected image data by adopting a search region and weight calculation in a non-local mean value technology. The main implementation process of the neighborhood windowing based non-local mean value CT imaging de-noising method comprises: (1) setting a mean square error of a Gaussian window for a collected original CT projected image; (2) finding all the similar blocks in the search region; (3) calculating a Gaussian Euclidean distance between the similar blocks and a block where a current spot locates; (4) calculating a similarity weight by utilizing a negative index function, wherein the level of similarity increases along with the increase of the weight; (5) obtaining a product of the similarity weight and a distance weight to obtain a mixed weight; (6) performing weighted average on all the pixel point values in the search region by using the mixed weight to obtain a corrected pixel point value; and (7) reconstructing de-noised projection image data to form a final chromatographic X-ray image. By the adoption of the neighborhood windowing based non-local mean value CT imaging de-noising method disclosed by the invention, the purpose of restoring the original image better and improving the de-noising performances for the image can be realized.
Description
Technical field
The invention belongs to the CT technical field of imaging, relate to a kind of non-local mean CT imaging denoising method based on neighborhood window addition.
Background technology
Along with the development of science and technology, the application of CT technology becomes more and more universal, and CT detects and plays an important role in daily life and research and production.But in the CT imaging process, tend to be subject to inevitable noise, irregular because noise is random, Gaussian noise is to occur maximum noises in imaging process usually.
CT imaging denoising can be carried out also can carrying out after rebuilding before rebuilding, and before rebuilding, denoising can make noise spread without reconstruction, and denoising before the CT imaging is carried out denoising to data for projection exactly.
Multiple known noise-removed technology is arranged in modern technologies, and traditional image de-noising method has gaussian filtering, mean filter.They are all the hypothesis of directly utilizing image space continuity or piecewise continuity, think that the noise at the pixel in the neighborhood centered by certain pixel and center pixel in image place is obeyed identical distribution.These methods are carried out the smoothing processing of same type to level and smooth district, marginarium and the texture area in image, so the detailed structure of meeting blurred picture, the resolution of image after the reduction denoising.More classical filtering and noise reduction method had appearred again afterwards, such as: medium filtering, its denoising effect is better, but the meeting Inhibitory signal; Field filtering, it can effectively solve " fuzzy " phenomenon in Gaussian smoothing, but can introduce false edge and staircase effect; Total variation method (TV) can be gone noise more thoroughly, but resolution is descended.
The non-local mean method that is proposed by Buades in recent years is one of very outstanding method in image denoising field, its basic thought be if in image two pixels around field structure similar, these two pixels are also similar so; These two pixels may be arranged in any position of image, so, search such pixel and also should launch in whole image range.In the non-local mean method, for the current denoising pixel for the treatment of, all pixels similar to this dot structure of search in image, as weights, the gradation of image value after denoising is obtained by these pixel weightings with the similarity of block structure.The contribution of the method for non-local mean is that the regional area pixel that in the past method thinks that substantially all the pixel of image only is adjacent is similar, it thinks that pixel and the pixel in its non-conterminous zone also may be similar, as long as these two pixels have similar field structure.But it also has the following disadvantages: 1. weights calculate not accurate enough; 2. processing speed is slow; 3. image resolution ratio is not high.
Summary of the invention
In view of this, technical matters to be solved by this invention is to have proposed a kind of non-local mean CT imaging denoising method based on neighborhood window addition, added the impact of distance on similarity on the basis of originally only considering the gray scale similarity, thereby the accuracy to the image block similarity measurement is improved, and has improved the effect of image denoising.
The object of the present invention is achieved like this:
A kind of non-local mean CT imaging denoising method based on neighborhood window addition provided by the invention comprises the following steps:
S1: obtain the original CT projecting image data;
S2: parameter is carried out initialization;
S3: calculate the similarity weights, adopt in non-local mean method search projected image and the current dot structure piece for the treatment of that the denoising dot structure is similar, calculate the similarity of similar dot structure piece and as the current similarity weights for the treatment of the denoising pixel;
S4: the similarity weights are obtained revising weights by adding the correction of Gauss's window function;
S5: treat that to current the denoising pixel is weighted on average according to revising weights, obtain the current correction gray-scale value for the treatment of the denoising pixel:
S6: utilize and revise the gray-scale value that gray-scale value replaces pixel in the original CT projecting image data of inputting, obtain the denoising perspective view.
Further, described parameter comprises dot structure block size t, the Gaussian window parameter of positive pixel to be repaired
, the rate of decay
, the described rate of decay
Standard deviation according to Gaussian noise
Determine.
Further, described calculating similarity weights are realized by following steps:
S32: utilize following formula to noisy data for projection should in erect image vegetarian refreshments i to be repaired and region of search
The similarity of interior pixel j is calculated, and obtains the region of search
The weights of interior all pixels:
Wherein,
Normalization coefficient,
,
In the difference presentation video, with i, the size centered by j is
The gray-scale value of image block, the rate of decay of parameter h control characteristic function; Weights
Satisfy
And
Further, described correction weights are realized by following steps:
S41: to the similarity weights by adding the correction of Gauss's window function:
Wherein, x is i, the Euclidean distance between j.
Further, describedly treat that to current it is on average to realize by following steps that the denoising pixel is weighted:
According to the weights of all pixels in the region of search that calculates, to all pixel weighted means in the region of search, obtain the revised gray-scale value of pixel:
Further, described region of search
For length is the square region of 6 times of mean square deviations, wherein, the region of search
Size by the mean square deviation of Gauss function
Determine.
The invention has the advantages that: the present invention adopts the non-local mean denoising method of neighborhood window addition that the projecting image data of CT imaging is rebuild, and has the following advantages:
1. the present invention can calculate the similarity between pixel in noisy perspective view more accurately, and then can calculate more accurately the gray-scale value of erect image vegetarian refreshments to be repaired;
2. the present invention can more accurately calculate the gray-scale value of erect image vegetarian refreshments to be repaired, and then can better keep image resolution preferably in denoising.
Adopt the inventive method, the projecting image data after denoising is reconstructed into final chromatographical X-ray image, can reach and recover better original image, improves the purpose to the Denoising performance.
Description of drawings
In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing, wherein:
Fig. 1 is process flow diagram of the present invention;
Fig. 2 is the projected image that the present invention uses;
Fig. 3 is the reconstructed image that the present invention uses;
Fig. 4 is the noisy projected image that the present invention uses;
Fig. 5 is the noisy reconstruction from projections imaging that the present invention uses;
Fig. 6 is with the reconstructed image after present non-local mean filtering method denoising;
Fig. 7 is with the reconstructed image after the inventive method denoising.
Embodiment
Below with reference to accompanying drawing, the preferred embodiments of the present invention are described in detail; Should be appreciated that preferred embodiment only for the present invention is described, rather than in order to limit protection scope of the present invention.
Embodiment 1
Fig. 1 is process flow diagram of the present invention, Fig. 2 is the projected image that the present invention uses, Fig. 3 is the reconstructed image that the present invention uses, Fig. 4 is the noisy projected image that the present invention uses, Fig. 5 is the noisy reconstruction from projections imaging that the present invention uses, and Fig. 6 is with the reconstructed image after present non-local mean filtering method denoising, and Fig. 7 is with the reconstructed image after the inventive method denoising, as shown in the figure: noisy reconstruction from projections imaging provided by the invention neighborhood window addition non-local mean denoising method comprises the following steps:
Step 1. is obtained the original CT projecting image data, and the noise in image is all usually take Gauss's additive white noise as main;
Step 2. pair parametric t (positive pixel tile size to be repaired),
(Gaussian window parameter),
(rate of decay) carries out initialization, gets respectively
, 1.67,0.1,
Standard deviation by Gaussian noise
Determine,
Step 3. is determined the region of search
, length is that the square region of 6 times of mean square deviations is the region of search, wherein
Size by the mean square deviation of Gauss function
Determine,
Step 4. utilize following formula to noisy data for projection should in erect image vegetarian refreshments i to be repaired and region of search the similarity of pixel j calculate, obtain the weights of all pixels in the region of search;
Wherein,
Normalization coefficient,
,
In the difference presentation video, with i, the size centered by j is
The gray-scale value of image block, the rate of decay of parameter h control characteristic function.
Step 5. pair similarity weights are by adding the correction of Gauss's window function;
Wherein, x is i, the Euclidean distance between j.
Step 6. to all pixel weighted means in the region of search, obtains the revised gray-scale value of pixel according to the weights of all pixels in the region of search that calculates:
The gray-scale value of pixel after step 7. is used and revised
The gray-scale value that replaces pixel in the noisy projected image of input obtains the perspective view after denoising;
Step 8. is rebuild final image to realize global de-noising by the projecting image data after denoising.
Embodiment 2
The difference of the present embodiment and embodiment 1 only is:
The present embodiment illustrates the non-local mean CT imaging denoising method based on neighborhood window addition provided by the invention to the processing procedure of projecting image data:
Obtain original CT projecting image data such as Fig. 2, Fig. 4 is that to add noise criteria poor to Fig. 2 be 0.1 noisy projecting image data.Fig. 3 is the image after Fig. 2 is rebuild, and Fig. 5 is the image after Fig. 4 is rebuild.
Under above-mentioned experiment condition, use directly respectively and rebuild, present non-local mean filtering and noise reduction is rebuild and the inventive method is rebuild the data for projection denoising.
Experimental result is to have shown noisy CT data for projection is directly rebuild effect as shown in Figure 5.
With rebuilding effect as shown in Figure 6 after present non-local mean filtering and noise reduction, wherein
,
Poor for the contained noise criteria of image, the size of search window is
, similar window size is
, as can be seen from Figure 6: the method noise inhibiting ability is relatively good, but can not well keep the edge of image.
Denoising result of the present invention as shown in Figure 7, as can be seen from Figure 7: its denoising effect is better than traditional non-local mean filtering method, has kept the edge of image in fine denoising, has strengthened resolution;
It is 0.01 Gaussian noise that test projected image in Fig. 2 is added variance, with the evaluation index of MSE as denoising effect, and with noisy, the non-local mean denoising, after denoising of the present invention, reconstructed image compares, in MSE value such as table 1.
Table 1 is denoising not, non-local mean denoising, reconstructed results contrast after denoising of the present invention.
Table 1
Method | Not denoising | NLM | The present invention |
MSE | 0.0137 | 0.0090 | 0.0040 |
The above is only the preferred embodiments of the present invention, is not limited to the present invention, and obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of claim of the present invention and equivalent technologies thereof, the present invention also is intended to comprise these changes and modification interior.
Claims (6)
1. non-local mean CT imaging denoising method based on neighborhood window addition is characterized in that: comprise the following steps:
S1: obtain the original CT projecting image data;
S2: parameter is carried out initialization;
S3: calculate the similarity weights, adopt in non-local mean method search projected image and the current dot structure piece for the treatment of that the denoising dot structure is similar, calculate the similarity of similar dot structure piece and as the current similarity weights for the treatment of the denoising pixel;
S4: the similarity weights are obtained revising weights by adding the correction of Gauss's window function;
S5: treat that to current the denoising pixel is weighted on average according to revising weights, obtain the current correction gray-scale value for the treatment of the denoising pixel:
S6: utilize and revise the gray-scale value that gray-scale value replaces pixel in the original CT projecting image data of inputting, obtain the denoising perspective view.
2. the non-local mean CT imaging denoising method based on neighborhood window addition according to claim 1, it is characterized in that: described parameter comprises dot structure block size t, the Gaussian window parameter of positive pixel to be repaired
, the rate of decay
, the described rate of decay
Standard deviation according to Gaussian noise
Determine.
3. the non-local mean CT imaging denoising method based on neighborhood window addition according to claim 1, it is characterized in that: described calculating similarity weights are realized by following steps:
S32: utilize following formula to noisy data for projection should in erect image vegetarian refreshments i to be repaired and region of search
The similarity of interior pixel j is calculated, and obtains the region of search
The weights of interior all pixels:
4. the non-local mean CT imaging denoising method based on neighborhood window addition according to claim 1, it is characterized in that: described correction weights are realized by following steps:
S41: to the similarity weights by adding the correction of Gauss's window function:
Wherein, x is i, the Euclidean distance between j.
5. the non-local mean CT imaging denoising method based on neighborhood window addition according to claim 1 is characterized in that: describedly treat that to current it is on average to realize by following steps that the denoising pixel is weighted:
According to the weights of all pixels in the region of search that calculates, to all pixel weighted means in the region of search, obtain the revised gray-scale value of pixel:
6. the non-local mean CT imaging denoising method based on neighborhood window addition according to claim 1, is characterized in that: described region of search
For length is the square region of 6 times of mean square deviations, wherein, the region of search
Size by the mean square deviation of Gauss function
Determine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100839229A CN103116879A (en) | 2013-03-15 | 2013-03-15 | Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100839229A CN103116879A (en) | 2013-03-15 | 2013-03-15 | Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103116879A true CN103116879A (en) | 2013-05-22 |
Family
ID=48415243
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013100839229A Pending CN103116879A (en) | 2013-03-15 | 2013-03-15 | Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103116879A (en) |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103489161A (en) * | 2013-09-12 | 2014-01-01 | 南京邮电大学 | Gray level image colorizing method and device |
CN103839234A (en) * | 2014-02-21 | 2014-06-04 | 西安电子科技大学 | Double-geometry nonlocal average image denoising method based on controlled nuclear |
CN105046662A (en) * | 2015-07-06 | 2015-11-11 | 嘉恒医疗科技(上海)有限公司 | CT image denoising method based on principal component analysis |
CN106408616A (en) * | 2016-11-23 | 2017-02-15 | 山西大学 | Method of correcting projection background inconsistency in CT imaging |
CN106815818A (en) * | 2017-01-17 | 2017-06-09 | 中国科学院上海高等研究院 | A kind of image de-noising method |
CN106910165A (en) * | 2015-12-23 | 2017-06-30 | 通用电气公司 | Repair method and device, the CT imaging systems of original CT data for projection |
CN106910163A (en) * | 2015-12-23 | 2017-06-30 | 通用电气公司 | The data prosthetic device and method of original CT data for projection, CT imaging systems |
CN107025640A (en) * | 2017-04-28 | 2017-08-08 | 南京觅踪电子科技有限公司 | The Medical Image Denoising method of non-local mean based on optimization |
CN107341495A (en) * | 2017-07-17 | 2017-11-10 | 深圳森阳环保材料科技有限公司 | A kind of effective underground coal dust detecting system |
CN107389516A (en) * | 2017-07-17 | 2017-11-24 | 陈剑桃 | A kind of efficient factory floor dust monitoring system |
CN107615331A (en) * | 2015-05-15 | 2018-01-19 | 深圳市大疆创新科技有限公司 | System and method for supporting the image denoising based on neighborhood block dimensionality reduction |
CN108447037A (en) * | 2018-03-26 | 2018-08-24 | 上海顺久电子科技有限公司 | A kind of method and apparatus of dynamic range of images enhancing |
CN110136088A (en) * | 2019-05-23 | 2019-08-16 | 大连理工大学 | A kind of human embryos cardiac ultrasound images denoising method |
CN110599563A (en) * | 2019-09-11 | 2019-12-20 | 河南工业大学 | CT reconstruction method and device for adaptive NLM correction |
CN110738612A (en) * | 2019-09-27 | 2020-01-31 | 深圳市安健科技股份有限公司 | Method for reducing noise of X-ray perspective image and computer readable storage medium |
CN111724325A (en) * | 2020-06-24 | 2020-09-29 | 湖南国科微电子股份有限公司 | Trilateral filtering image processing method and device |
CN116912102A (en) * | 2023-05-11 | 2023-10-20 | 上海宇勘科技有限公司 | Edge-preserving image denoising method and system based on non-local mean value |
CN117788570A (en) * | 2024-02-26 | 2024-03-29 | 山东济矿鲁能煤电股份有限公司阳城煤矿 | Bucket wheel machine positioning method and system based on machine vision |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101661611A (en) * | 2009-09-25 | 2010-03-03 | 西安电子科技大学 | Realization method based on bayesian non-local mean filter |
US20110311154A1 (en) * | 2010-06-17 | 2011-12-22 | Canon Kabushiki Kaisha | Method and device for enhancing a digital image |
-
2013
- 2013-03-15 CN CN2013100839229A patent/CN103116879A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101661611A (en) * | 2009-09-25 | 2010-03-03 | 西安电子科技大学 | Realization method based on bayesian non-local mean filter |
US20110311154A1 (en) * | 2010-06-17 | 2011-12-22 | Canon Kabushiki Kaisha | Method and device for enhancing a digital image |
Non-Patent Citations (3)
Title |
---|
BART GOOSSENS ET AL.: "An Improved Non-local Denoising Algorithm", 《2008 INTERNATIONAL WORKSHOP ON LOCAL AND NON-LOCAL APPROXIMATION IN IMAGE PROCESSING》, 31 December 2008 (2008-12-31), pages 143 - 156 * |
JOSE V. MANJON ET AL.: "MRI denoising using Non-Local Means", 《MEDICAL IMAGE ANALYSIS》, vol. 12, no. 4, 31 August 2008 (2008-08-31), pages 514 - 523, XP022834275, DOI: 10.1016/j.media.2008.02.004 * |
刘晓明等: "一种改进的非局部均值图像去噪算法", 《计算机工程》, vol. 38, no. 4, 29 January 2012 (2012-01-29) * |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103489161A (en) * | 2013-09-12 | 2014-01-01 | 南京邮电大学 | Gray level image colorizing method and device |
CN103839234A (en) * | 2014-02-21 | 2014-06-04 | 西安电子科技大学 | Double-geometry nonlocal average image denoising method based on controlled nuclear |
CN103839234B (en) * | 2014-02-21 | 2016-09-14 | 西安电子科技大学 | A kind of double geometry non-local mean image de-noising methods based on controlled core |
CN107615331A (en) * | 2015-05-15 | 2018-01-19 | 深圳市大疆创新科技有限公司 | System and method for supporting the image denoising based on neighborhood block dimensionality reduction |
CN107615331B (en) * | 2015-05-15 | 2021-03-02 | 深圳市大疆创新科技有限公司 | System and method for supporting image denoising based on neighborhood block dimension reduction |
CN105046662A (en) * | 2015-07-06 | 2015-11-11 | 嘉恒医疗科技(上海)有限公司 | CT image denoising method based on principal component analysis |
CN106910163B (en) * | 2015-12-23 | 2022-06-21 | 通用电气公司 | Data restoration device and method for original CT projection data and CT imaging system |
CN106910165B (en) * | 2015-12-23 | 2023-06-30 | 通用电气公司 | Method and device for repairing original CT projection data and CT imaging system |
CN106910165A (en) * | 2015-12-23 | 2017-06-30 | 通用电气公司 | Repair method and device, the CT imaging systems of original CT data for projection |
CN106910163A (en) * | 2015-12-23 | 2017-06-30 | 通用电气公司 | The data prosthetic device and method of original CT data for projection, CT imaging systems |
CN106408616A (en) * | 2016-11-23 | 2017-02-15 | 山西大学 | Method of correcting projection background inconsistency in CT imaging |
CN106408616B (en) * | 2016-11-23 | 2019-02-26 | 山西大学 | The inconsistent bearing calibration of perspective view background in a kind of CT imaging |
CN106815818A (en) * | 2017-01-17 | 2017-06-09 | 中国科学院上海高等研究院 | A kind of image de-noising method |
CN107025640A (en) * | 2017-04-28 | 2017-08-08 | 南京觅踪电子科技有限公司 | The Medical Image Denoising method of non-local mean based on optimization |
CN107389516A (en) * | 2017-07-17 | 2017-11-24 | 陈剑桃 | A kind of efficient factory floor dust monitoring system |
CN107341495A (en) * | 2017-07-17 | 2017-11-10 | 深圳森阳环保材料科技有限公司 | A kind of effective underground coal dust detecting system |
CN108447037A (en) * | 2018-03-26 | 2018-08-24 | 上海顺久电子科技有限公司 | A kind of method and apparatus of dynamic range of images enhancing |
CN108447037B (en) * | 2018-03-26 | 2022-02-18 | 上海顺久电子科技有限公司 | Method and device for enhancing dynamic range of image |
CN110136088A (en) * | 2019-05-23 | 2019-08-16 | 大连理工大学 | A kind of human embryos cardiac ultrasound images denoising method |
CN110136088B (en) * | 2019-05-23 | 2022-12-13 | 大连理工大学 | Human embryo heart ultrasonic image denoising method |
CN110599563B (en) * | 2019-09-11 | 2022-10-18 | 河南工业大学 | CT reconstruction method and device for adaptive NLM correction |
CN110599563A (en) * | 2019-09-11 | 2019-12-20 | 河南工业大学 | CT reconstruction method and device for adaptive NLM correction |
CN110738612B (en) * | 2019-09-27 | 2022-04-29 | 深圳市安健科技股份有限公司 | Method for reducing noise of X-ray perspective image and computer readable storage medium |
CN110738612A (en) * | 2019-09-27 | 2020-01-31 | 深圳市安健科技股份有限公司 | Method for reducing noise of X-ray perspective image and computer readable storage medium |
CN111724325A (en) * | 2020-06-24 | 2020-09-29 | 湖南国科微电子股份有限公司 | Trilateral filtering image processing method and device |
CN111724325B (en) * | 2020-06-24 | 2023-10-31 | 湖南国科微电子股份有限公司 | Trilateral filtering image processing method and trilateral filtering image processing device |
CN116912102A (en) * | 2023-05-11 | 2023-10-20 | 上海宇勘科技有限公司 | Edge-preserving image denoising method and system based on non-local mean value |
CN116912102B (en) * | 2023-05-11 | 2024-04-09 | 上海宇勘科技有限公司 | Edge-preserving image denoising method and system based on non-local mean value |
CN117788570A (en) * | 2024-02-26 | 2024-03-29 | 山东济矿鲁能煤电股份有限公司阳城煤矿 | Bucket wheel machine positioning method and system based on machine vision |
CN117788570B (en) * | 2024-02-26 | 2024-05-07 | 山东济矿鲁能煤电股份有限公司阳城煤矿 | Bucket wheel machine positioning method and system based on machine vision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103116879A (en) | Neighborhood windowing based non-local mean value CT (Computed Tomography) imaging de-noising method | |
CN107578430B (en) | Stereo matching method based on self-adaptive weight and local entropy | |
US20240037732A1 (en) | Method for enhancing quality and resolution of ct images based on deep learning | |
CN102184533B (en) | Non-local-restriction-based total variation image deblurring method | |
CN108921800A (en) | Non-local mean denoising method based on form adaptive search window | |
CN103020918B (en) | Shape-adaptive neighborhood mean value based non-local mean value denoising method | |
CN103440644B (en) | A kind of multi-scale image weak edge detection method based on minimum description length | |
CN105913396A (en) | Noise estimation-based image edge preservation mixed de-noising method | |
CN102567973B (en) | Image denoising method based on improved shape self-adaptive window | |
CN104331863B (en) | A kind of image filtering denoising method | |
CN109377450B (en) | Edge protection denoising method | |
CN104252698A (en) | Semi-inverse method-based rapid single image dehazing algorithm | |
CN103455991A (en) | Multi-focus image fusion method | |
CN102968770A (en) | Method and device for eliminating noise | |
CN102298773A (en) | Shape-adaptive non-local mean denoising method | |
CN108470343A (en) | A kind of improved method for detecting image edge | |
Zhang et al. | Decision-based non-local means filter for removing impulse noise from digital images | |
CN106663315A (en) | Method for denoising noisy image | |
Chen et al. | A color-guided, region-adaptive and depth-selective unified framework for Kinect depth recovery | |
CN103971345A (en) | Image denoising method based on improved bilateral filtering | |
CN107451986B (en) | Single infrared image enhancement method based on fusion technology | |
CN107025640A (en) | The Medical Image Denoising method of non-local mean based on optimization | |
CN107424121A (en) | A kind of blurred picture ultra-resolution ratio reconstructing method based on fractional order differential | |
CN108510510A (en) | Method for detecting image edge based on gradient direction | |
Tallón et al. | Upsampling and denoising of depth maps via joint-segmentation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20130522 |