CN114372938A - Image self-adaptive restoration method based on calibration - Google Patents
Image self-adaptive restoration method based on calibration Download PDFInfo
- Publication number
- CN114372938A CN114372938A CN202210040753.XA CN202210040753A CN114372938A CN 114372938 A CN114372938 A CN 114372938A CN 202210040753 A CN202210040753 A CN 202210040753A CN 114372938 A CN114372938 A CN 114372938A
- Authority
- CN
- China
- Prior art keywords
- image
- calibration
- noise
- distribution
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000003384 imaging method Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 13
- 230000003287 optical effect Effects 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 10
- 238000009792 diffusion process Methods 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000012935 Averaging Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 14
- 230000004075 alteration Effects 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000012634 optical imaging Methods 0.000 description 2
- 241001081972 Arctium medians Species 0.000 description 1
- 235000002566 Capsicum Nutrition 0.000 description 1
- 241000282414 Homo sapiens Species 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
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010351 charge transfer process Methods 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- 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
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the field of image restoration processing, and particularly discloses an image self-adaptive restoration method based on calibration, which comprises the following steps: noise calibration, shooting and storing a set target, and performing noise calibration; calibrating an imaging system; carrying out image denoising treatment; image de-blurring processing; and outputting the final restored image. The invention calibrates the noise and the fuzzy coefficient of the actual optical system, performs parameterized mathematical modeling, applies the calibrated result to the subsequent image restoration algorithm, obtains the optimal image restoration result for the set of imaging system, and greatly improves the image restoration effect.
Description
Technical Field
The invention relates to the field of image restoration processing, in particular to an image self-adaptive restoration method based on calibration.
Background
An image of one point of the object space through an ideal optical imaging system is still one point. The actual optical imaging system is affected by geometric aberration and optical diffraction limit, the image is no longer a clear perfect image, and certain optical blurring exists. Meanwhile, noise signals are always accompanied in the imaging process, and the noise mainly comes from poisson noise related to light intensity, transfer noise introduced in the charge transfer process, dark current noise generated on a silicon substrate by thermal excitation, non-uniformity noise and the like.
These factors cause degradation of image quality, and thus image denoising and deblurring are very important steps in image restoration. Image denoising generally starts from smoothness of an image, high frequency of a frequency domain and randomness of distribution, and a corresponding mathematical model is constructed to remove noise; the image deblurring utilizes the cognition of human beings on a clear image to add prior to a degradation model to solve the model, so that the image deblurring function is achieved.
Most current algorithms artificially make certain assumptions about image processing, such as that noise follows gaussian distribution or default values of blur kernel, which makes image restoration undesirable for different imaging systems.
Disclosure of Invention
The present invention aims to provide a calibration-based image adaptive restoration method to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme: a calibration-based image self-adaptive restoration method comprises the following steps:
s1: noise calibration, shooting and storing a set target, and performing noise calibration;
s2: calibrating an imaging system;
s3: carrying out image denoising treatment;
s4: image de-blurring processing;
s5: and outputting the final restored image.
Preferably, the noise calibration in S1 includes global noise calibration and local noise calibration, and the noise calibration specifically includes:
s101: setting a target, shooting the target by adopting an actual optical system, and storing a shot image;
s102: selecting two frames of target images of the same scene at the same time;
s103: calculating global image noise distribution;
s104: a local image noise distribution is calculated.
Preferably, the global image noise distribution calculation process in S103 specifically includes:
s103 a: carrying out difference operation on the two frames of target images to obtain a difference image;
s103 b: performing distribution fitting of a probability density function on the difference image, and determining the distribution type and specific parameter values of the difference image;
s103 c: and storing the result into a parameter file.
Preferably, the local image noise distribution calculation process in S104 specifically includes:
s104 a: intercepting N image blocks with different gray values on a target image, wherein N is defaulted to be 8;
s104 b: respectively drawing a histogram for each image block;
s104 c: then, performing distribution fitting on the probability density function to determine the distribution type and specific parameter values of the probability density function;
s104 d: averaging the N image block results;
s104 e: and storing all results into a parameter file.
Preferably, the imaging system calibration process of S2 specifically includes:
s201: selecting a frame of target image;
s202: intercepting a region ROI with black and white bevel edges in the image;
s203: sequentially projecting data of different rows in the ROI on the same pixel grid to obtain an edge diffusion function ESF;
s204: deriving the edge diffusion function ESF to obtain a linear change rate line diffusion function LSF of a straight line;
s205: carrying out Fourier FFT (fast Fourier transform) on the LSF to obtain a response value SFR under each spatial frequency;
s206: analyzing SFR data to obtain imaging system fuzzy parameters;
s207: and storing the result into a parameter file.
Preferably, the image denoising process in S3 is implemented based on a distribution algorithm, which includes, but is not limited to, median filtering, non-local mean algorithm, and R-L algorithm.
Preferably, the image deblurring processing in S4 is implemented based on deblurring calculation, and the deblurred picture is the final restored image.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, the noise distribution and the imaging system calibration are carried out through the current optical system, the calibrated file is stored and applied to a subsequent algorithm as a prior parameter, and different algorithms are selected according to the noise distribution, so that the denoising effect is optimal; the result calibrated by the imaging system is used as an initial value based on an edge deblurring algorithm, so that the accuracy of blind deblurring fuzzy kernel estimation is further improved, and the deblurring effect can be greatly improved.
Drawings
FIG. 1 is a block flow diagram of the present invention as a whole;
FIG. 2 is a flow chart of noise calibration in an embodiment of the present invention;
FIG. 3 is a flow chart of imaging system calibration in an embodiment of the present invention;
FIG. 4 is a flowchart illustrating image denoising according to an embodiment of the present invention;
FIG. 5 is a flow chart of image deblurring in an embodiment of the present invention;
FIG. 6 is a schematic illustration of a calibration target in an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides a technical solution: a calibration-based image self-adaptive restoration method comprises the following steps:
s1: noise calibration, shooting and storing the set target, and performing noise calibration (the calibration target is shown as figure 6);
s2: calibrating an imaging system;
s3: carrying out image denoising treatment;
s4: image de-blurring processing;
s5: and outputting the final restored image.
Referring to fig. 2, in the present embodiment, the noise calibration in S1 includes global noise calibration and local noise calibration, and the noise calibration specifically includes the following steps:
s101: setting a target, shooting the target by adopting an actual optical system, and storing a shot image;
s102: selecting two frames of target images of the same scene at the same time;
s103: calculating global image noise distribution;
s104: a local image noise distribution is calculated.
In this embodiment, the process of calculating the global image noise distribution in S103 specifically includes:
s103 a: carrying out difference operation on the two frames of target images to obtain a difference image;
s103 b: performing distribution fitting of a probability density function on the difference image, and determining the distribution type and specific parameter values of the difference image;
s103 c: and storing the result into a parameter file.
In this embodiment, the local image noise distribution calculation process in S104 specifically includes:
s104 a: intercepting N image blocks with different gray values on a target image, wherein N is defaulted to be 8;
s104 b: respectively drawing a histogram for each image block;
s104 c: then, performing distribution fitting on the probability density function to determine the distribution type and specific parameter values of the probability density function;
s104 d: averaging the N image block results;
s104 e: and storing all results into a parameter file.
Referring to fig. 3, in the present embodiment, the imaging system calibration process of S2 specifically includes:
s201: selecting a frame of target image;
s202: intercepting a region ROI with black and white bevel edges in the image;
s203: sequentially projecting data of different rows in the ROI on the same pixel grid to obtain an edge diffusion function ESF;
s204: deriving the edge diffusion function ESF to obtain a linear change rate line diffusion function LSF of a straight line;
s205: carrying out Fourier FFT (fast Fourier transform) on the LSF to obtain a response value SFR under each spatial frequency;
s206: analyzing SFR data to obtain imaging system fuzzy parameters; s207: and storing the result into a parameter file.
Referring to fig. 4, in the present embodiment, the image denoising process in S3 is implemented based on a distribution algorithm, which includes, but is not limited to, median filtering, a non-local mean algorithm, and an R-L algorithm.
The step S3 specifically includes:
s301: importing a parameter file;
s302: selecting an input image to be processed;
s303: according to the noise distribution calibrated by the parameter file, self-adaptive algorithm selection is carried out, and the algorithm provides self-adaptive algorithm selection; the algorithm comprises the following steps:
a. median filtering (salt and pepper noise), which is a nonlinear signal processing technology based on a sequencing statistical theory and capable of effectively inhibiting noise, wherein the basic principle of the median filtering is to replace the value of one point in a digital image or a digital sequence by the median of all point values in a neighborhood of the point, so that an isolated noise point is eliminated;
b. non-local mean algorithm (gaussian noise), which is a method of smoothly averaging in a region around a target pixel, so non-local mean filtering means that it uses all pixels in an image, which are weighted-averaged according to some similarity. The image after filtering has high definition and does not lose details; the non-local mean algorithm removes noise by using redundant information commonly existing in a natural image, and is different from filtering by using image local information such as bilinear filtering, median filtering and the like, the noise removal is performed by using the whole image, namely similar areas are searched in the image by taking an image block as a unit, and then the areas are averaged, so that Gaussian noise in the image is well filtered;
c. R-L algorithm (Poisson noise);
s304: processing an input image;
s305: and outputting the processed image, and then performing image de-blurring processing.
Referring to fig. 5, in the present embodiment, the image deblurring processing in S4 is implemented based on deblurring calculation, and the deblurred picture is the final restored image.
In this embodiment, S4 specifically includes:
s401: reading the imaging system fuzzy parameter calibrated by the parameter file to construct a two-dimensional kernel function;
s402: starting image deblurring by taking the kernel obtained in the S401 as an initial value based on an edge deblurring algorithm;
s403: and (5) finishing restoration operation after the image is deblurred.
In the embodiment, the noise and the fuzzy coefficient of the actual optical system are calibrated, parameterized mathematical modeling is performed, and the calibration result is applied to a subsequent image restoration algorithm to obtain the optimal image restoration result for the set of imaging system.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (7)
1. An image self-adaptive restoration method based on calibration is characterized by comprising the following steps:
s1: noise calibration, shooting and storing a set target, and performing noise calibration;
s2: calibrating an imaging system;
s3: carrying out image denoising treatment;
s4: image de-blurring processing;
s5: and outputting the final restored image.
2. The method for adaptive image restoration based on calibration according to claim 1, wherein the noise calibration in S1 includes global noise calibration and local noise calibration, and the specific steps of noise calibration include:
s101: setting a target, shooting the target by adopting an actual optical system, and storing a shot image;
s102: selecting two frames of target images of the same scene at the same time;
s103: calculating global image noise distribution;
s104: a local image noise distribution is calculated.
3. The calibration-based image adaptive restoration method according to claim 2, wherein the global image noise distribution calculation process in S103 specifically includes:
s103 a: carrying out difference operation on the two frames of target images to obtain a difference image;
s103 b: performing distribution fitting of a probability density function on the difference image, and determining the distribution type and specific parameter values of the difference image;
s103 c: and storing the result into a parameter file.
4. The calibration-based image adaptive restoration method according to claim 2, wherein the local image noise distribution calculation process in S104 specifically includes:
s104 a: intercepting N image blocks with different gray values on a target image, wherein N is defaulted to be 8;
s104 b: respectively drawing a histogram for each image block;
s104 c: then, performing distribution fitting on the probability density function to determine the distribution type and specific parameter values of the probability density function;
s104 d: averaging the N image block results;
s104 e: and storing all results into a parameter file.
5. The image adaptive restoration method based on calibration according to claim 1, wherein the imaging system calibration process of S2 specifically includes:
s201: selecting a frame of target image;
s202: intercepting a region ROI with black and white bevel edges in the image;
s203: sequentially projecting data of different rows in the ROI on the same pixel grid to obtain an edge diffusion function ESF;
s204: deriving the edge diffusion function ESF to obtain a linear change rate line diffusion function LSF of a straight line;
s205: carrying out Fourier FFT (fast Fourier transform) on the LSF to obtain a response value SFR under each spatial frequency;
s206: analyzing SFR data to obtain imaging system fuzzy parameters;
s207: and storing the result into a parameter file.
6. The calibration-based image adaptive restoration method according to claim 1, wherein the image denoising process in S3 is implemented based on a distribution algorithm, which includes but is not limited to median filtering, non-local mean algorithm, and R-L algorithm.
7. The calibration-based image adaptive restoration method according to claim 1, wherein the image deblurring processing in S4 is implemented based on a deblurring algorithm, and the deblurred image is a final restored image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210040753.XA CN114372938A (en) | 2022-01-14 | 2022-01-14 | Image self-adaptive restoration method based on calibration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210040753.XA CN114372938A (en) | 2022-01-14 | 2022-01-14 | Image self-adaptive restoration method based on calibration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114372938A true CN114372938A (en) | 2022-04-19 |
Family
ID=81144359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210040753.XA Pending CN114372938A (en) | 2022-01-14 | 2022-01-14 | Image self-adaptive restoration method based on calibration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114372938A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823938A (en) * | 2023-08-28 | 2023-09-29 | 荣耀终端有限公司 | Method for determining spatial frequency response, electronic device and storage medium |
-
2022
- 2022-01-14 CN CN202210040753.XA patent/CN114372938A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823938A (en) * | 2023-08-28 | 2023-09-29 | 荣耀终端有限公司 | Method for determining spatial frequency response, electronic device and storage medium |
| CN116823938B (en) * | 2023-08-28 | 2023-11-17 | 荣耀终端有限公司 | Method for determining spatial frequency response, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110246089B (en) | Bayer domain image noise reduction system and method based on non-local mean filtering | |
| CN110706174B (en) | Image enhancement method, terminal equipment and storage medium | |
| CN104021532B (en) | A kind of image detail enhancement method of infrared image | |
| CN111986120A (en) | Low-illumination image enhancement optimization method based on frame accumulation and multi-scale Retinex | |
| CN109978774B (en) | Denoising fusion method and device for multi-frame continuous equal exposure images | |
| WO2016139260A9 (en) | Method and system for real-time noise removal and image enhancement of high-dynamic range images | |
| CN102970464A (en) | Information processing apparatus and information processing method | |
| WO2017100971A1 (en) | Deblurring method and device for out-of-focus blurred image | |
| CN111415317B (en) | Image processing method and device, electronic equipment and computer readable storage medium | |
| CN111340732B (en) | Low-illumination video image enhancement method and device | |
| WO2023273868A1 (en) | Image denoising method and apparatus, terminal, and storage medium | |
| WO2021139635A1 (en) | Method and apparatus for generating super night scene image, and electronic device and storage medium | |
| CN110942427A (en) | Image noise reduction method and device, equipment and storage medium | |
| Tsutsui et al. | Halo artifacts reduction method for variational based realtime retinex image enhancement | |
| Buades et al. | Enhancement of noisy and compressed videos by optical flow and non-local denoising | |
| Tanbakuchi et al. | Adaptive pixel defect correction | |
| CN115578286A (en) | High dynamic range hybrid exposure imaging method and apparatus | |
| Teranishi et al. | Improvement of robustness blind image restoration method using failing detection process | |
| CN114372938A (en) | Image self-adaptive restoration method based on calibration | |
| CN108898561B (en) | Defogging method, server and system for foggy image containing sky area | |
| CN117408886A (en) | Gas image enhancement method, gas image enhancement device, electronic device and storage medium | |
| Kumar | Satellite image denoising using local spayed and optimized center pixel weights | |
| Bertalmio et al. | Movie denoising by average of warped lines | |
| Tigga et al. | Image deblurring with impulse noise using alternating direction method of multipliers and Lucy-Richardson method | |
| Schöberl et al. | Sparsity-based defect pixel compensation for arbitrary camera raw images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |