CN111709887B - Image rain removing method based on sparse blind detection and image multiple feature restoration - Google Patents
Image rain removing method based on sparse blind detection and image multiple feature restoration Download PDFInfo
- Publication number
- CN111709887B CN111709887B CN202010464988.2A CN202010464988A CN111709887B CN 111709887 B CN111709887 B CN 111709887B CN 202010464988 A CN202010464988 A CN 202010464988A CN 111709887 B CN111709887 B CN 111709887B
- Authority
- CN
- China
- Prior art keywords
- image
- rain
- restoration
- method based
- blind detection
- 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 48
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 6
- 230000008439 repair process Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 5
- 101100173586 Schizosaccharomyces pombe (strain 972 / ATCC 24843) fft2 gene Proteins 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013179 statistical model Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image processing, and discloses an image rain removing method based on sparse blind detection and image multi-feature restoration, which comprises the following steps: (1) Detecting the distribution position of rain in the image through a rain mark detection model, and generating a rain mark distribution template; (2) And carrying out rain removing treatment and image information restoration treatment on the image area with rain through the image rain removing and restoration model by utilizing the rain trace distribution template, and retaining the original information on the area without rain pollution so as to realize the image rain removing and restoration. Compared with the prior art, the invention can realize higher-performance image rain removal and image information protection.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an image rain removing method based on sparse blind detection and image multi-feature restoration.
Background
At present, the photoelectric imaging system is widely used for outdoor information acquisition sensing, for example: traffic monitoring, public safety, satellite remote sensing, etc. When the imaging system works outdoors, the imaging system inevitably needs to image in a rainy day environment, however, the rainy day environment is generally influenced by raindrops formed in the rainy falling process, so that the imaging quality of the imaging system in the rainy day is reduced, and part of information of the image is shielded by the raindrops and even lost. This presents great difficulties for later analysis and use of the image, which requires a rain removal treatment of the image obtained in a rainy environment. The existing image rain removing method mainly comprises an image priori knowledge-based rain removing method and a deep learning-based image rain removing method; the method based on the priori knowledge of the image mainly uses the image decomposition technology to erase the rain layer information from the degraded image to realize rain removal, but the method has the problems of incomplete rain removal or excessive smoothness of the image when rain is removed, because the intensity distribution of the rain in the image is random; the deep learning method needs a sufficient number of clear/rain degradation image training sample sets, and the rain removal can only be effective for removing rain to a certain degree and a certain type; and these methods do not take into account the raindrop distribution characteristics in the image.
Aiming at the problem of rain mark removal of images, the invention provides an image rain removal method based on sparse blind detection and image information restoration, which discovers that rain marks are distributed in images with sparsity (shown in figure 1), and utilizes the characteristic of rain space distribution to firstly detect the distribution position of rain in the images by using a sparse statistical model, and then carries out rain removal treatment and image information restoration treatment on the image areas with rain, thereby realizing higher-performance image rain removal and image information protection.
Disclosure of Invention
The invention aims to: aiming at the problem of rain mark removal of images, the invention provides an image rain removal method based on sparse blind detection and image multi-feature restoration, which comprises the steps of firstly detecting the distribution position of rain in the images by utilizing a rain mark vertical gradient based on a sparse statistical model, and generating a function distribution template; then, the rain trace distribution template is utilized to carry out rain removing treatment and image information restoration treatment on the image area with rain, original information is reserved on the area without rain pollution, and the image rain removing and image information protection with higher performance are realized.
The technical scheme is as follows: the invention provides an image rain removing method based on sparse blind detection and image multi-feature restoration, which comprises the following steps:
(1) Detecting the distribution position of rain in the image through a rain mark detection model, and generating a rain mark distribution template;
(2) And carrying out rain removing treatment and image information restoration treatment on the image area with rain through the image rain removing and restoration model by utilizing the rain trace distribution template, and retaining the original information on the area without rain pollution so as to realize the image rain removing and restoration.
Further, in the step (1), the raindrop detection model is:
where s is a rain trace image to be detected, r is a rain pollution image, and I.I L0 In order to be an L0 norm,and->Is a horizontal and vertical gradient operator; lambda (lambda) 1 ,λ 2 ,λ 3 ,λ 4 And lambda (lambda) 5 Is a non-negative regularization coefficient.
Further, the solving process of the raindrop detection model is as follows:
introducing an auxiliary variable d 1 ,d 2 ,d 3 ,d 4 ,d 5 And cause
According to the principle of half-quadratic division, the raindrop detection model is converted into the following unconstrained condition problem:
wherein alpha is 1 ,α 2 ,α 3 ,α 4 ,α 5 Is a non-negative iteration coefficient introduced in the case of a half-quadratic split.
Further, regarding s k+1 Solving the sub-problems:
solving the formula (3) to obtain:
sorting and utilizing fourier transform and inverse fourier transform are:
wherein, fft2 (·) and ifft2 (·) represent the fourier transform and the inverse fourier transform, respectively.
Further, regardingAnd->Solving the sub-problems:
the following iterative formula can be obtained by using the hard threshold iterative principle:
further, according to the iterative formula, the image rain detection process can be summarized as follows:
step1, initializing the setting,iteration stop condition error limit tol=10 -5 Initial error = 0; maximum iteration number itermax=100, and initial iteration number iter=1;
step2 while item < itermax or error > tol
And (3) calculating:
step3: outputting a rain detection graph s, and generating a rain distribution position template P by binarization ^ 。
Further, in the step (2), the image rain removal and repair model is:
in the above formula, D is a tight wavelet frame sparse transform operator, u is a clear image to be restored, and β is a non-negative real coefficient.
Further, the solving process of the model formula (16) is as follows:
let p=du, with half-quadratic splitting, formula (16) can be converted into:
then there is u k+1 Is to be solved:
and (3) solving to obtain:
u k+1 =(P ^ +γI) -1 (P ^ f+γD T p k )
p k+1 the optimization problem of (2) is:
the hard threshold filtering is utilized as follows:
further, the rain removing process and the image information restoration process are carried out on the image area with rain through the image rain removing and restoration model as follows:
step1: initializing settings, p 0 =u 0 =0, iteration stop condition error limit tol=10 -4 Initial error = 0; maximum iteration number itermax=100, and initial iteration number iter=1;
step2: while iter < itermax or error > tol
And (3) calculating:
u k+1 =(P ^ +γI) -1 (P ^ f+γD T p k )
step3: and outputting a repair image result u.
The beneficial effects are that: aiming at the problem of rain mark removal of images, the invention provides an image rain removal method based on sparse blind detection and image multi-feature restoration, which comprises the steps of firstly detecting the distribution position of rain in the images by utilizing a rain mark vertical gradient based on a sparse statistical model, and generating a function distribution template; then, the rain trace distribution template is utilized to carry out rain removing treatment and image information restoration treatment on the image area with rain, original information is reserved on the area without rain pollution, and the image rain removing and image information protection with higher performance are realized.
Drawings
Fig. 1 shows the spatial distribution characteristics of rain: (a) A plot of intensity information for rain, (b) a spatial distribution plot of rain;
fig. 2 is a process of summarizing an image rain removal method based on sparse blind detection and image multi-feature restoration: the method comprises the steps of (a) an image needing rain removal, (b) a rain distribution template detected in the first step, (c) an image area to be repaired after mapping by using the rain distribution template, and (d) an image after rain removal and repair;
fig. 3 is a comparison of rain removal and image restoration: the first column: rain drop degradation map, second column: the neural network rain removal result, third column: the fourth column is based on the rain removal and restoration results of the total variation of directivity: the method of the invention removes rain and restores the result.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
Embodiment 1:
the embodiment provides an image rain removing method based on sparse blind detection and image multi-feature restoration, which comprises the following steps:
(1) And detecting the distribution position of rain in the image through the rain mark detection model, and generating a rain mark distribution template.
The rain mark detection model is as follows:
where s is a rain trace pattern to be detected (as in FIG. 1 (a)), r is a rain pollution image (as in figure 2 (a)), I.I L0 In order to be an L0 norm,and->Is a horizontal and vertical gradient operator; lambda (lambda) 1 ,λ 2 ,λ 3 ,λ 4 And lambda (lambda) 5 Is a non-negative regularization coefficient.
The solution process of the rain trace detection model is as follows:
introducing an auxiliary variable d 1 ,d 2 ,d 3 ,d 4 ,d 5 And cause
The above model can be transformed into the following unconstrained problem, based on the principle of half-quadratic splitting:
wherein alpha is 1 ,α 2 ,α 3 ,α 4 ,α 5 Is a non-negative iteration coefficient introduced in the case of a half-quadratic split.
Equation (1) is a multivariate optimization problem that can be transformed into the following univariate optimization sub-problem:
(1) with respect to s k+1 Solving the sub-problems:
solving the formula (3) to obtain:
sorting and utilizing fourier transform and inverse fourier transform are:
wherein, fft2 (·) and ifft2 (·) represent the fourier transform and the inverse fourier transform, respectively.
(2) With respect toAnd->Solving the sub-problems:
the following iterative formula can be obtained by using the hard threshold iterative principle:
according to the iterative formula of the variables, the image rain detection process can be summarized as follows:
step1, initializing the setting,iteration stop condition error limit tol=10 -5 Initial error = 0; the maximum iteration number itermax=100 and the iteration number initial value iter=1.
Step2 while item < itermax or error > tol
And (3) calculating:
step3, outputting a rain detection diagram s, and generating a rain distribution position template P by binarization ^ (as in fig. 2 (b)).
(2) By using the rain trace distribution template, the image rain removing and repairing model is used for carrying out rain removing treatment and image information repairing treatment on the image area with rain, and original information is reserved on the area without rain pollution, so that the image rain removing and repairing are realized (as shown in fig. 2 (d)).
The image rain removal and restoration model is as follows:
in the above formula, D is a tight wavelet frame sparse transform operator, u is a clear image to be restored, and β is a non-negative real coefficient.
The model form (16) solving process is as follows:
let p=du, with half-quadratic splitting, formula (16) can be converted into:
then there is u k+1 Is to be solved:
and (3) solving to obtain:
u k+1 =(P ^ +γI) -1 (P ^ f+γD T p k )
p k+1 the optimization problem of (2) is:
the hard threshold filtering is utilized as follows:
the image rain removal and repair process in the second step can be summarized as follows:
step1, initializing the settings, p 0 =u 0 =0, iteration stop condition error limit tol=10 -4 Initial error = 0; the maximum iteration number itermax=100 and the iteration number initial value iter=1.
Step2 while item < itermax or error > tol
And (3) calculating:
u k+1 =(P ^ +γI) -1 (P ^ f+γD T p k )
step3, outputting a repair image result u.
The above image rain detection parameters are valued in the experiment as follows: lambda (lambda) 1 =0.01,λ 2 =0.001,λ 3 =0.004,λ 4 =3×10 -4 ,λ 5 =0.19; the parameter values in the image restoration stage are as follows: beta=0.3, gamma=3×10 -4 . The above parameters may be preferable to be ideal values according to practical applications.
The result of the rain removal process performed on the picture shown in the first column of fig. 3 according to the above method is compared with that of fig. 3 (where the second column of the image is the rain removal result of the deep neural network, the third column is the rain removal result based on the directional total variation method, and the last column is the rain removal result of the present invention). The images after rain removal based on the deep neural network and the directional total variation method still have a certain amount of rain marks which are not removed completely, and some images are blurred after rain removal, so that the target characteristics are not protected enough; the method of the invention can remove rain marks more fully after removing rain, and the processed image target information is well protected. Therefore, the method only carries out rain removal processing and image information restoration processing on the image area with rain, and retains the original information on the area without rain pollution, so that the method can realize higher-performance image rain removal and image information protection.
The foregoing embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the present invention and to implement the same, not to limit the scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of the present invention.
Claims (8)
1. An image rain removing method based on sparse blind detection and image multi-feature restoration is characterized by comprising the following steps of:
(1) Detecting the distribution position of rain in the image through a rain mark detection model, and generating a rain mark distribution template;
(2) The rain mark distribution template is utilized, the image rain removing and repairing model is utilized to carry out rain removing treatment and image information repairing treatment on the image area with rain, original information is reserved on the area without rain pollution, and the image rain removing and repairing are realized;
wherein, the rain mark detection model is:
in the above formula, s is a rain trace image to be detected, r is a rain pollution image, and I.I.I. L0 For L0 norm, v x and v y are horizontal and vertical gradient operators; lambda (lambda) 1 ,λ 2 ,λ 3 ,λ 4 And lambda (lambda) 5 Is a non-negative regularization coefficient.
2. The image rain removal method based on sparse blind detection and image multi-feature restoration according to claim 1, wherein the solving process of the raindrop detection model is as follows:
introducing an auxiliary variable d 1 ,d 2 ,d 3 ,d 4 ,d 5 And make d 1 =▽xs,d 2 =▽ys,d 3 =▽x(r-s),d 4 =▽y(r-s),d 5 =s;
According to the principle of half-quadratic division, the raindrop detection model is converted into the following unconstrained condition problem:
wherein alpha is 1 ,α 2 ,α 3 ,α 4 ,α 5 Is a non-negative iteration coefficient introduced in the case of a half-quadratic split.
3. The image rain removal method based on sparse blind detection and image multi-feature restoration of claim 2, wherein regarding s k+1 Solving the sub-problems:
solving the formula (3) to obtain:
sorting and utilizing fourier transform and inverse fourier transform are:
wherein, fft2 (·) and ifft2 (·) represent the fourier transform and the inverse fourier transform, respectively.
4. The image rain removal method based on sparse blind detection and image multi-feature restoration of claim 3, wherein, with respect toAnd->Solving the sub-problems:
the following iterative formula can be obtained by using the hard threshold iterative principle:
5. the method for image rain removal based on sparse blind detection and image multi-feature restoration according to claim 4, wherein according to the iterative formula, the image rain detection process can be summarized as follows:
step1, initializing the setting,iteration stop condition error limit tol=10 -5 Initial error = 0; maximum iteration number itermax=100, and initial iteration number iter=1;
step2:while iter<itermax or error > tol
And (3) calculating:
step3: and outputting a rain detection graph s, and generating a rain distribution position template P-by binarization.
6. The image rain removal method based on sparse blind detection and image multi-feature restoration of any one of claims 1 to 5, wherein in step (2), the image rain removal and restoration model is:
in the above formula, D is a tight wavelet frame sparse transform operator, u is a clear image to be restored, and β is a non-negative real coefficient.
7. The image rain removal method based on sparse blind detection and image multi-feature restoration of claim 6, wherein the solution process of the image rain removal and restoration model (16) is as follows:
let p=du, with half-quadratic splitting, the image rain removal and repair model (16) can be transformed into:
then there is u k+1 Is to be solved:
and (3) solving to obtain:
u k+1 =(P^+γI) -1 (P^f+γD T p k )
p k+1 the optimization problem of (2) is:
the hard threshold filtering is utilized as follows:
8. the image rain removal method based on sparse blind detection and image multi-feature restoration according to claim 6, wherein the processes of rain removal processing and image information restoration processing for the image area with rain by the image rain removal and restoration model are as follows:
step1: initializing settings, p 0 =u 0 =0, iteration stop condition error limit tol=10 -4 Initial error = 0; maximum iteration number itermax=100, and initial iteration number iter=1;
step2: while iter < itermax or error > tol
And (3) calculating:
u k+1 =(P^+γI) -1 (P^f+γD T p k )
step3: and outputting a repair image result u.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010464988.2A CN111709887B (en) | 2020-05-28 | 2020-05-28 | Image rain removing method based on sparse blind detection and image multiple feature restoration |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010464988.2A CN111709887B (en) | 2020-05-28 | 2020-05-28 | Image rain removing method based on sparse blind detection and image multiple feature restoration |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111709887A CN111709887A (en) | 2020-09-25 |
CN111709887B true CN111709887B (en) | 2023-09-22 |
Family
ID=72538044
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010464988.2A Active CN111709887B (en) | 2020-05-28 | 2020-05-28 | Image rain removing method based on sparse blind detection and image multiple feature restoration |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111709887B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113902931B (en) * | 2021-09-17 | 2022-07-12 | 淮阴工学院 | Image rain removing method based on learning type convolution sparse coding |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108520501A (en) * | 2018-03-30 | 2018-09-11 | 西安交通大学 | A kind of video and removes rain snow method based on multiple dimensioned convolution sparse coding |
CN110544215A (en) * | 2019-08-23 | 2019-12-06 | 淮阴工学院 | traffic monitoring image rain removing method based on anisotropic sparse gradient |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9734601B2 (en) * | 2014-04-04 | 2017-08-15 | The Board Of Trustees Of The University Of Illinois | Highly accelerated imaging and image reconstruction using adaptive sparsifying transforms |
-
2020
- 2020-05-28 CN CN202010464988.2A patent/CN111709887B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108520501A (en) * | 2018-03-30 | 2018-09-11 | 西安交通大学 | A kind of video and removes rain snow method based on multiple dimensioned convolution sparse coding |
CN110544215A (en) * | 2019-08-23 | 2019-12-06 | 淮阴工学院 | traffic monitoring image rain removing method based on anisotropic sparse gradient |
Also Published As
Publication number | Publication date |
---|---|
CN111709887A (en) | 2020-09-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108230264B (en) | Single image defogging method based on ResNet neural network | |
CN108520501B (en) | Video rain and snow removing method based on multi-scale convolution sparse coding | |
CN111862194B (en) | Deep learning plant growth model analysis method and system based on computer vision | |
CN111428748A (en) | Infrared image insulator recognition and detection method based on HOG characteristics and SVM | |
CN107038416B (en) | Pedestrian detection method based on binary image improved HOG characteristics | |
CN110544215B (en) | Traffic monitoring image rain removing method based on anisotropic sparse gradient | |
CN110830043B (en) | Image compressed sensing reconstruction method based on mixed weighted total variation and non-local low rank | |
CN113989257A (en) | Electric power comprehensive pipe gallery settlement crack identification method based on artificial intelligence technology | |
CN111127354A (en) | Single-image rain removing method based on multi-scale dictionary learning | |
CN111815526B (en) | Rain image rainstrip removing method and system based on image filtering and CNN | |
CN111709887B (en) | Image rain removing method based on sparse blind detection and image multiple feature restoration | |
CN115908354A (en) | Photovoltaic panel defect detection method based on double-scale strategy and improved YOLOV5 network | |
Wang et al. | Effective segmentation approach for solar photovoltaic panels in uneven illuminated color infrared images | |
Cruz-Rojas et al. | A novel comparison of image semantic segmentation techniques for detecting dust in photovoltaic panels using machine learning and deep learning | |
CN106023097A (en) | Iterative-method-based flow field image preprocessing algorithm | |
Kim et al. | ADOM: ADMM-based optimization model for stripe noise removal in remote sensing image | |
CN116309407A (en) | Method for detecting abnormal state of railway contact net bolt | |
Kwok et al. | Adaptive scale adjustment design of unsharp masking filters for image contrast enhancement | |
CN112862729B (en) | Remote sensing image denoising method based on characteristic curve guidance | |
CN106934344B (en) | quick pedestrian detection method based on neural network | |
CN111709962B (en) | Image contour and texture feature decomposition method based on anisotropic L0 gradient sparse representation and DCT (discrete cosine transform) | |
CN116029957A (en) | Insulator image pollution identification method based on Markov chain Monte Carlo | |
CN113989260A (en) | Photovoltaic panel edge segmentation method based on unmanned aerial vehicle infrared image | |
CN116402725B (en) | Oblique strip removing method, device, equipment and medium | |
CN113902931B (en) | Image rain removing method based on learning type convolution sparse coding |
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 |