CN111968041A - Self-adaptive image enhancement method - Google Patents
Self-adaptive image enhancement method Download PDFInfo
- Publication number
- CN111968041A CN111968041A CN202010630421.8A CN202010630421A CN111968041A CN 111968041 A CN111968041 A CN 111968041A CN 202010630421 A CN202010630421 A CN 202010630421A CN 111968041 A CN111968041 A CN 111968041A
- Authority
- CN
- China
- Prior art keywords
- image
- color
- enhancement
- frequency component
- component
- 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 41
- 238000011084 recovery Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 16
- 230000003044 adaptive effect Effects 0.000 claims abstract description 13
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 13
- 238000012937 correction Methods 0.000 claims abstract description 6
- 230000009466 transformation Effects 0.000 claims description 8
- 230000001131 transforming effect Effects 0.000 claims description 6
- IXKSXJFAGXLQOQ-XISFHERQSA-N WHWLQLKPGQPMY Chemical compound C([C@@H](C(=O)N[C@@H](CC=1C2=CC=CC=C2NC=1)C(=O)N[C@@H](CC(C)C)C(=O)N[C@@H](CCC(N)=O)C(=O)N[C@@H](CC(C)C)C(=O)N1CCC[C@H]1C(=O)NCC(=O)N[C@@H](CCC(N)=O)C(=O)N[C@@H](CC(O)=O)C(=O)N1CCC[C@H]1C(=O)N[C@@H](CCSC)C(=O)N[C@@H](CC=1C=CC(O)=CC=1)C(O)=O)NC(=O)[C@@H](N)CC=1C2=CC=CC=C2NC=1)C1=CNC=N1 IXKSXJFAGXLQOQ-XISFHERQSA-N 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 abstract description 13
- 238000011156 evaluation Methods 0.000 description 14
- 238000005286 illumination Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 238000004088 simulation Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 230000002087 whitening effect Effects 0.000 description 1
Images
Classifications
-
- G06T5/90—
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- 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
- G06T2207/10024—Color image
Abstract
The invention relates to the technical field of image enhancement, in particular to a self-adaptive image enhancement method, which comprises the following steps: s1, converting the original image to be processed from RGB into HSV color space and decomposing the HSV color space into H, S, V components; s2, carrying out single-scale decomposition on the V component to obtain low-frequency component and high-frequency component information of the V component of the image; s3, performing enhancement processing on the low frequency component in step S2, and then performing adaptive correction; s4, performing blur enhancement on the high frequency component in step S2; s5, obtaining an enhanced brightness V component by wavelet reconstruction of the low-frequency component and the high-frequency component after the enhancement processing in the steps S3 and S4; s6, converting the HSV image back to RGB color space; s7, color recovery is performed in the RGB space. The invention avoids color distortion, improves the contrast and detail information of the image, better accords with the visual effect of people and has self-adaptability.
Description
Technical Field
The invention relates to the technical field of image enhancement, in particular to a self-adaptive image enhancement method.
Background
The image enhancement technology is a basic and important technology in the field of image processing, and is a long-standing irrecoverable research subject in the field of image processing. How to properly enhance the image, so that the characteristics of the image can be well protected while the image is denoised, and the improvement of the visual effect of the image and the improvement of the definition of the image are realized, which is the problem to be solved by image enhancement. The traditional histogram equalization method is a mature algorithm in a spatial domain, is relatively simple to implement, but can cause excessive enhancement of images, cannot keep mean brightness and entropy of the images, and is rarely applied to practical engineering. For this reason, many improved histogram equalization methods have been developed, such as a representative Limited Contrast adaptive histogram equalization (CLAHE), which can effectively enhance the image Contrast but has a color distortion problem. At present, scholars at home and abroad develop different image enhancement methods from the aspects of image characteristics, human eye visual effect and different mathematical theories. In the aspect of improving the image definition, methods such as a sparse expression and learning dictionary method for noise removal, a Smart Deblur restoration method for motion blur removal, a dark primary color prior theory for haze removal and the like exist, but the methods have a blur effect on edge characteristics and detail information of an image. In the aspect of image edge feature extraction, there are difference methods such as Roberts, Canny, Laplacian and the like, but the difference methods can enhance the influence of noise on the image. There are also various image enhancement algorithms based on Retinex theory for uneven illumination correction, but most of the algorithms produce the phenomena of whitening, graying, color distortion and low contrast. The above image enhancement methods have certain limitations due to different application requirements, that is, different image enhancement methods are suitable for images with different characteristics. How to design an intelligent image enhancement method, the images with different characteristics are adaptively enhanced, visual effects such as the definition, characteristic details and the like of the enhanced images are achieved, and the method has important research significance.
Based on the human visual characteristics, the invention provides a self-adaptive image enhancement method, which is used for preprocessing an image by adopting wavelet transformation in an HSV color space and realizing self-adaptive enhancement of the image by combining a Retinex theory, an Otsu threshold method and a Pal-King fuzzy enhancement algorithm.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a self-adaptive image enhancement method, which avoids color distortion, improves the contrast and detail information of an image, better conforms to the visual effect of a human and has self-adaptability.
In order to realize the purpose of the invention, the invention adopts the technical scheme that:
the invention discloses a self-adaptive image enhancement method, which comprises the following steps:
s1, converting the original image to be processed from RGB into HSV color space and decomposing the HSV color space into H, S, V components;
s2, carrying out single-scale decomposition on the V component to obtain low-frequency component and high-frequency component information of the V component of the image;
s3, performing enhancement processing on the low frequency component in step S2, and then performing adaptive correction;
s4, performing blur enhancement on the high frequency component in step S2;
s5, obtaining an enhanced brightness V component by wavelet reconstruction of the low-frequency component and the high-frequency component after the enhancement processing in the steps S3 and S4;
s6, converting the HSV image back to the RGB color space.
In step S3, a multi-scale retinex (MSR) is used to perform enhancement processing on the low-frequency component a1 obtained by wavelet decomposition, where the MSR formula is as follows:
wherein i ∈ R, G, B represent three color channels, wjA weighting factor representing the scale, K representing the number of scales used, usually taken as wj＝1/3，K＝3。
In step S3, the estimated luminance component is adaptively corrected by an Otsu threshold method, and the luminance component L is first stretched to obtain a luminance image L', which is expressed as follows:
wherein, the alpha factor determines the degree of image stretching, and the beta factor is used for adjusting the stretching interval;
the Otsu threshold method obtains a gray threshold T by calculating the maximum inter-class variance of the image; let β be 255T, then:
in step S4, a King-Pal fuzzy algorithm is used to perform image detail enhancement and noise filtering on the high-frequency component, where the Pal-King fuzzy algorithm is:
for an M × N image with a gray level of L, firstly, transforming the image from a spatial domain to a fuzzy characteristic domain, wherein a fuzzy matrix is represented as P:
in the formula, muij/pijRepresenting a pixel point (i, j) in an image relative to a particular gray level pijDegree of membership mu ofijUsually 0. ltoreq. muijLess than or equal to 1; the membership functions defined by Pal-King are as follows:
in the formula, pmaxIs the maximum gray level, FeFor exponential ambiguity factors, usually take Fe＝2，FdIs the reciprocal ambiguity factor. To muijFor fuzzy enhancement, the Pal-King defines a transformation function as:
μ′ij＝Er(μij)＝E1(Er-1(μij)),r＝1,2,... (6)
wherein:
and finally, transforming the enhanced image from the fuzzy feature domain to a spatial domain:
the method further comprises a step S7 of eliminating color deviation of the image by adding a color recovery factor C and using an MSRCR algorithm of the multi-scale Retinex with color recovery, wherein the MSRCR algorithm formula is as follows:
wherein the content of the first and second substances,
in the formula, the first step is that,
and
respectively the output after the MSRCR algorithm and the MSR algorithm are enhanced, i belongs to R, G and B represent three color channels, Ci(x, y) is the color recovery factor for the ith channel, f () is the mapping function, Si(x, y) is an image of the ith channel;
as can be seen from the formula, the MSRCR algorithm adjusts the color proportion of three color channels (R, G, B) of the image through the color recovery factor of each channel, thereby eliminating color distortion;
the color recovery factor, as used herein, is formulated as follows:
where β is a gain parameter, γ is a parameter for adjusting the brightness of the image, and gi(i ═ 1,2,3) for images to be color restored，
The image is the image after color recovery.
The invention has the beneficial effects that:
(1) the invention adopts wavelet transformation to preprocess the image to be processed, combines Retinex theory, Otsu threshold method and Pal-King fuzzy enhancement algorithm to adaptively enhance the image, and compares the adaptive enhancement image with the related algorithm.
Drawings
FIG. 1 is a schematic flow chart of the algorithm of the present invention;
FIG. 2 is a comparison graph of the outdoor daytime image enhancement effect of the present invention;
FIG. 3 is a comparison graph of the enhancement effect of the outdoor night image according to the present invention;
FIG. 4 is a comparison graph of the indoor daytime image enhancement effect of the present invention.
Detailed Description
The invention is further illustrated below:
referring to figures 1-4 of the drawings,
the invention discloses a self-adaptive image enhancement method, because HSV color space is closest to human visual characteristics, firstly converting an image to be processed into HSV color space and decomposing the HSV color space into H, S, V components, and carrying out single-scale wavelet decomposition on a V component by adopting wavelet transformation to obtain low-frequency and high-frequency component information of the V component of the image; and processing the low-frequency component by adopting multi-scale Retinex (MSR), performing Otsu threshold value self-adaptive correction on the illumination component estimated by the Retinex model, and performing Pal-King fuzzy enhancement on the high-frequency component. And finally, performing color recovery processing on the fused image to realize the self-adaptive enhancement of the image, wherein the flow of the algorithm is shown in figure 1.
(1) Wavelet decomposition of V-component
The independence of each component in the HSV color space is strong, and compared with the change of saturation H and hue S, a person is more sensitive to the change of brightness V, and the V component is processed independently, so that the color distortion can be eliminated or weakened, and the algorithm is simpler and faster.
Wavelet decomposition is performed on the extracted brightness V component by adopting wavelet transformation through decomposing the HSV color space. In order to avoid the blurring of the processed image caused by the wavelet decomposition of higher layers, the single-scale wavelet decomposition is adopted for the V component, and a low-frequency approximate component A1, a high-frequency horizontal detail component H1, a high-frequency vertical detail component V1 and a high-frequency diagonal detail component D1 are obtained. The low-frequency and high-frequency components are processed by different enhancement methods, and then wavelet reconstruction is carried out, so that the purpose of image enhancement can be achieved.
(2) Low frequency component enhancement
In the scheme, multi-scale Retinex (MSR) is adopted to perform enhancement processing on a low-frequency component A1 obtained by wavelet decomposition.
MSR was developed from SSR and its basic formula is as follows:
wherein i ∈ R, G, B represent three color channels, wjA weighting factor representing the scale, K representing the number of scales used, usually taken as wj1/3, K3. The formula reflects that the MSR enhancement algorithm can take two characteristics of tone reproduction and dynamic range compression into account.
When the MSR algorithm estimates the illumination component of the image, the illumination intensity is not sufficiently estimated, so that the phenomenon that the obtained reflection component is excessively enhanced in a bright area is caused, and if a gamma correction method is adopted, the dark noise of the image is amplified, so that the estimated illumination component is corrected by adopting an Otsu threshold method. The method is suitable for different illumination conditions and has self-adaptive capacity.
Firstly, stretching the illumination component L to obtain an illumination image L', wherein the formula is as follows:
wherein, the alpha factor determines the degree of image stretching, and the beta factor is used for adjusting the stretching interval.
The Otsu threshold method obtains a gray threshold T by calculating the maximum inter-class variance of the image; let β be 255T, then:
as can be seen from the above equation, the algorithm finally realizes adaptive stretching of the illumination component and adjustment of the image contrast.
(3) High frequency component Pal-King blur enhancement
Aiming at that the coefficient with a larger absolute value in the high-frequency components obtained by wavelet decomposition represents image texture and detail information, and the coefficient with a smaller absolute value may be noise, the image detail enhancement and the noise filtering are carried out on the high-frequency components by adopting a Pal-King fuzzy algorithm.
The idea of the Pal-King fuzzy enhancement algorithm is as follows:
for an M × N image with a gray level of L, firstly, transforming the image from a spatial domain to a fuzzy characteristic domain, wherein a fuzzy matrix is represented as P:
in the formula, muij/pijRepresenting a pixel point (i, j) in an image relative to a particular gray level pijDegree of membership mu ofijUsually 0. ltoreq. muijLess than or equal to 1. The membership functions defined by Pal-King are as follows:
in the formula, pmaxIs the maximum gray level, FeFor exponential ambiguity factors, usually take Fe＝2，FdIs the reciprocal ambiguity factor.
To muijFor fuzzy enhancement, the Pal-King defines a transformation function as:
μ′ij＝Er(μij)＝E1(Er-1(μij)),r＝1,2,... (6)
wherein:
and finally, transforming the enhanced image from the fuzzy feature domain to a spatial domain:
(4) MSRCR-based color recovery
And respectively carrying out enhancement processing on the low-frequency component and the high-frequency component after wavelet decomposition, obtaining an enhanced brightness V component through wavelet reconstruction, and finally converting the image back to an RGB color space. Practice proves that only the brightness V component is processed in the HSV space, and color deviation can be generated due to the lack of processing of the tone information. Also, color distortion results from MSR algorithms that may add noise to the image and lack of consideration for the color scale of the image. On the basis of MSR, a Multi-Scale Retinex with Color retrieval (MSRCR) algorithm with Color retrieval is adopted to eliminate the Color deviation of the image by adding a Color retrieval factor C.
The MSRCR algorithm formula is as follows:
wherein the content of the first and second substances,
in the formula, the first step is that,
and
respectively the output after the MSRCR algorithm and the MSR algorithm are enhanced, i belongs to R, G and B represent three color channels, Ci(x, y) is the color recovery factor for the ith channel, f () is the mapping function, Si(x, y) is the image of the ith channel.
As can be seen from the formula, the MSRCR algorithm adjusts the color proportion of three color channels (R, G, B) of the image through the color recovery factor of each channel, thereby eliminating color distortion;
the color recovery factor, as used herein, is formulated as follows:
where β is a gain parameter, γ is a parameter for adjusting the brightness of the image, and gi(i-1, 2,3) is an image to be color-restored,
the image is the image after color recovery.
Example (b):
in order to verify the self-adaptive enhancement performance of the algorithm, different scene images acquired in outdoor daytime, outdoor night and indoor daytime are selected to carry out simulation experiments, and are compared with the current commonly used Contrast-Limited adaptive histogram equalization method (CLAHE) and HSV-based MSRCR algorithm (MSRCR algorithm processing is carried out on the V component of the image only in HSV space, called HSV-MSRCR for short) and evaluated through subjective visual effect and objective quality standard.
The algorithm parameters are set as: in the Otsu threshold method, the image stretching factor alpha is 1.05, and the 3 gauss function scales of the MSRCR algorithm are sigma1＝15、σ2＝80、σ3250, the color recovery parameter beta is 0.75,γ ═ 6.5, the parameters relevant for the comparison algorithms HSV-MSRCR and CLAHE remain the same as those of the algorithms herein. Simulation experiment results of different scene images adopting different enhancement algorithms are respectively shown in fig. 2-4, and the experiment results are analyzed from the aspect of subjective visual effect, so that the CLAHE algorithm effectively enhances the contrast of the images, but has the problem of hue deviation, the phenomenon of color inversion occurs in partial areas, and the brightness of the images is also reduced. The HSV-MSRCR algorithm has good tone retention, but the image is overall dull, the layering sense is not rich enough, and the visual effect is poor. The algorithm optimizes the color of the image, further enhances the contrast, is excessively natural in the light and shade mutation area, greatly improves the brightness and the details of the darker area of the image, and limits the excessive enhancement of the lighter area. Therefore, the algorithm can effectively keep the detail information of the image and enrich the color of the image while improving the overall brightness of the image.
The experimental result is evaluated through an objective Quality standard, and the enhanced Image is objectively evaluated through a Natural Image Quality Evaluation (NIQE) standard. The NIQE evaluation method does not depend on the information of the original image and has better consistency with subjective evaluation of human eyes. The evaluation method measures the quality of the image by calculating the distance between the fitting parameters of a Multivariate Gaussian (MVG) model of a natural image and a degraded image, and the evaluation formula is as follows:
in the formula, v1、v2、∑1Sum Σ2Mean vectors and variance matrices of MVG models for natural and degraded images, respectively. The smaller the distance D, i.e. the smaller the NIQE value, the higher the image visual quality.
The NIQE evaluation formula is adopted to calculate the NIQE evaluation values of simulation experiment results of the different scene images by adopting different enhancement algorithms, which are respectively shown in the following tables 1-3.
TABLE 1 NIQE evaluation value for outdoor daytime image enhancement
| Type of algorithm | NIQE evaluation value |
| Original drawing | 6.3256 |
| CLAHE algorithm | 5.9761 |
| HSV-MSRCR algorithm | 5.5635 |
| Text algorithm | 2.9325 |
TABLE 2 NIQE evaluation value for outdoor night image enhancement
| Type of algorithm | NIQE evaluation value |
| Original drawing | 7.8376 |
| CLAHE algorithm | 6.3753 |
| HSV-MSRCR algorithm | 4.9592 |
| Text algorithm | 3.8453 |
TABLE 3 indoor daytime image enhanced NIQE evaluation values
| Type of algorithm | NIQE evaluation value |
| Original drawing | 4.3378 |
| CLAHE algorithm | 3.8213 |
| HSV-MSRCR algorithm | 2.8959 |
| Text algorithm | 2.7164 |
As can be seen from the table above, the NIQE value obtained by the algorithm after the enhancement processing is performed on the images of different indoor and outdoor scenes is the minimum, which indicates that the image quality is optimal, and the algorithm effect is the best.
In conclusion, subjective evaluation and objective evaluation are carried out, and results show that the algorithm optimizes the colors of the images, retains more details of the images, has the best visual effect and has self-adaptability under different indoor and outdoor environments.
The image to be processed is preprocessed by adopting wavelet transformation, the image is adaptively enhanced by combining Retinex theory, Otsu threshold method and Pal-King fuzzy enhancement algorithm, and the adaptive enhancement is compared with related algorithm. The experimental result shows that the algorithm avoids color distortion, improves the contrast and detail information of the image, better accords with the visual effect of people, and has self-adaptability.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all equivalent modifications made by the present invention and the contents of the drawings or directly or indirectly applied to the related technical fields are included in the scope of the present invention.
Claims (5)
1. An adaptive image enhancement method, comprising the steps of:
s1, converting the original image to be processed from RGB into HSV color space and decomposing the HSV color space into H, S, V components;
s2, carrying out single-scale decomposition on the V component to obtain low-frequency component and high-frequency component information of the V component of the image;
s3, performing enhancement processing on the low frequency component in step S2, and then performing adaptive correction;
s4, performing blur enhancement on the high frequency component in step S2;
s5, obtaining an enhanced brightness V component by wavelet reconstruction of the low-frequency component and the high-frequency component after the enhancement processing in the steps S3 and S4;
s6, converting the HSV image back to the RGB color space.
2. The adaptive image enhancement method according to claim 1, wherein in step S3, the low frequency component obtained by wavelet decomposition is enhanced by using multi-scale retinex (MSR), and the MSR formula is as follows:
wherein i ∈ R, G, B represent three color channels, wjA weighting factor representing the scale, K representing the number of scales used, usually taken as wj＝1/3，K＝3。
3. An adaptive image enhancement method according to claim 2, wherein in step S3, the estimated luminance component is adaptively corrected by Otsu threshold method, and the luminance component L is first stretched to obtain a luminance image L', which is as follows:
wherein, the alpha factor determines the degree of image stretching, and the beta factor is used for adjusting the stretching interval;
the Otsu threshold method obtains a gray threshold T by calculating the maximum inter-class variance of the image; let β be 255T, then:
4. the adaptive image enhancement method according to claim 1, wherein in step S4, image detail enhancement and noise filtering are performed on the high frequency components by using a Pal-King blurring algorithm, where the Pal-King blurring algorithm is:
for an M × N image with a gray level of L, firstly, transforming the image from a spatial domain to a fuzzy characteristic domain, wherein a fuzzy matrix is represented as P:
in the formula, muij/pijRepresenting a pixel point (i, j) in an image relative to a particular gray level pijDegree of membership mu ofijUsually 0. ltoreq. muijLess than or equal to 1; membership defined by Pal-KingThe degree function is as follows:
in the formula, pmaxIs the maximum gray level, FeFor exponential ambiguity factors, usually take Fe＝2，FdIs the reciprocal ambiguity factor.
To muijFor fuzzy enhancement, the Pal-King defines a transformation function as:
μij’＝Er(μij)＝E1(Er-1(μij)),r＝1,2,... (6)
wherein:
and finally, transforming the enhanced image from the fuzzy feature domain to a spatial domain:
5. the adaptive image enhancement method according to claim 1, further comprising step S7, wherein the color recovery factor C is added to eliminate the color bias of the image by using the MSRCR algorithm of multi-scale Retinex with color recovery, and the formula of the MSRCR algorithm is as follows:
wherein the content of the first and second substances,
in the formula, the first step is that,
and
respectively the output after the MSRCR algorithm and the MSR algorithm are enhanced, i belongs to R, G and B represent three color channels, Ci(x, y) is the color recovery factor for the ith channel, f () is the mapping function, Si(x, y) is an image of the ith channel;
as can be seen from the formula, the MSRCR algorithm adjusts the color proportion of three color channels (R, G, B) of the image through the color recovery factor of each channel, thereby eliminating color distortion;
the color recovery factor, as used herein, is formulated as follows:
where β is a gain parameter, γ is a parameter for adjusting the brightness of the image, and gi(i-1, 2,3) is an image to be color-restored,
the image is the image after color recovery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010630421.8A CN111968041A (en) | 2020-07-03 | 2020-07-03 | Self-adaptive image enhancement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010630421.8A CN111968041A (en) | 2020-07-03 | 2020-07-03 | Self-adaptive image enhancement method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111968041A true CN111968041A (en) | 2020-11-20 |
Family
ID=73360920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010630421.8A Pending CN111968041A (en) | 2020-07-03 | 2020-07-03 | Self-adaptive image enhancement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111968041A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112581400A (en) * | 2020-12-22 | 2021-03-30 | 安徽圭目机器人有限公司 | Tuning image enhancement method based on Gaussian standard deviation and contrast |
| CN112785532A (en) * | 2021-01-12 | 2021-05-11 | 安徽大学 | Singular value equalization image enhancement algorithm based on weighted histogram distribution gamma correction |
| CN112801925A (en) * | 2021-01-12 | 2021-05-14 | 扬州大学 | Underwater image enhancement method for maximally eliminating water ripple influence based on information entropy |
| CN112907470A (en) * | 2021-02-05 | 2021-06-04 | 北京理工大学 | Underwater image recovery method based on Lab color gamut transformation, classification and white balance |
| CN113256533A (en) * | 2021-06-15 | 2021-08-13 | 北方民族大学 | Self-adaptive low-illumination image enhancement method and system based on MSRCR |
| CN113554572A (en) * | 2021-08-13 | 2021-10-26 | 中国矿业大学 | Image enhancement method and system based on improved Retinex |
| CN113658086A (en) * | 2021-08-06 | 2021-11-16 | 桂林日盛水务有限公司 | CLAHE and histogram stretching underwater image enhancement method based on wavelet fusion |
| CN114972074A (en) * | 2022-04-27 | 2022-08-30 | 广东鉴面智能科技有限公司 | Night vision image analysis system based on low-light-level environment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303532A (en) * | 2015-10-21 | 2016-02-03 | 北京工业大学 | Wavelet domain Retinex image defogging method |
| CN106530250A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | Low illumination color image enhancement method based on improved Retinex |
-
2020
- 2020-07-03 CN CN202010630421.8A patent/CN111968041A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105303532A (en) * | 2015-10-21 | 2016-02-03 | 北京工业大学 | Wavelet domain Retinex image defogging method |
| CN106530250A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | Low illumination color image enhancement method based on improved Retinex |
Non-Patent Citations (3)
| Title |
|---|
| 崔昊杨;许永鹏;孙岳;孙旭日;盛戈;江秀臣;: "基于自适应遗传算法的变电站红外图像模糊增强", 高电压技术, vol. 41, no. 03, pages 902 - 907 * |
| 张军国;程浙安;胡春鹤;陈宸;鲍伟东;: "野生动物监测光照自适应Retinex图像增强算法", 农业工程学报, vol. 34, no. 15, pages 183 - 189 * |
| 赵小川 编著: "《MATLAB图像处理：能力提高与应用案例》", 31 January 2019, 北京航空航天大学出版社, pages: 173 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112581400A (en) * | 2020-12-22 | 2021-03-30 | 安徽圭目机器人有限公司 | Tuning image enhancement method based on Gaussian standard deviation and contrast |
| CN112785532A (en) * | 2021-01-12 | 2021-05-11 | 安徽大学 | Singular value equalization image enhancement algorithm based on weighted histogram distribution gamma correction |
| CN112801925A (en) * | 2021-01-12 | 2021-05-14 | 扬州大学 | Underwater image enhancement method for maximally eliminating water ripple influence based on information entropy |
| CN112785532B (en) * | 2021-01-12 | 2022-11-18 | 安徽大学 | Singular value equalization image enhancement algorithm based on weighted histogram distribution gamma correction |
| CN112801925B (en) * | 2021-01-12 | 2023-08-18 | 扬州大学 | Underwater image enhancement method for maximally eliminating influence of water ripple based on information entropy |
| CN112907470A (en) * | 2021-02-05 | 2021-06-04 | 北京理工大学 | Underwater image recovery method based on Lab color gamut transformation, classification and white balance |
| CN113256533A (en) * | 2021-06-15 | 2021-08-13 | 北方民族大学 | Self-adaptive low-illumination image enhancement method and system based on MSRCR |
| CN113256533B (en) * | 2021-06-15 | 2022-08-09 | 北方民族大学 | Self-adaptive low-illumination image enhancement method and system based on MSRCR |
| CN113658086A (en) * | 2021-08-06 | 2021-11-16 | 桂林日盛水务有限公司 | CLAHE and histogram stretching underwater image enhancement method based on wavelet fusion |
| CN113554572A (en) * | 2021-08-13 | 2021-10-26 | 中国矿业大学 | Image enhancement method and system based on improved Retinex |
| CN113554572B (en) * | 2021-08-13 | 2024-03-26 | 中国矿业大学 | Image enhancement method and system based on improved Retinex |
| CN114972074A (en) * | 2022-04-27 | 2022-08-30 | 广东鉴面智能科技有限公司 | Night vision image analysis system based on low-light-level environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111968041A (en) | Self-adaptive image enhancement method | |
| Wang et al. | Adaptive image enhancement method for correcting low-illumination images | |
| Lee et al. | Adaptive multiscale retinex for image contrast enhancement | |
| Wang et al. | Simple low-light image enhancement based on Weber–Fechner law in logarithmic space | |
| CN108765336B (en) | Image defogging method based on dark and bright primary color prior and adaptive parameter optimization | |
| CN107798661B (en) | Self-adaptive image enhancement method | |
| CN109919859B (en) | Outdoor scene image defogging enhancement method, computing device and storage medium thereof | |
| Zotin | Fast algorithm of image enhancement based on multi-scale retinex | |
| CN106709504B (en) | High fidelity tone mapping method for detail preservation | |
| CN111968065B (en) | Self-adaptive enhancement method for image with uneven brightness | |
| CN110298792B (en) | Low-illumination image enhancement and denoising method, system and computer equipment | |
| CN111476725A (en) | Image defogging enhancement algorithm based on gradient domain oriented filtering and multi-scale Retinex theory | |
| CN110473152B (en) | Image enhancement method based on improved Retinex algorithm | |
| CN116188339A (en) | Retinex and image fusion-based scotopic vision image enhancement method | |
| CN108550124B (en) | Illumination compensation and image enhancement method based on bionic spiral | |
| CN114187222A (en) | Low-illumination image enhancement method and system and storage medium | |
| Feng et al. | Low-light color image enhancement based on Retinex | |
| CN110580690B (en) | Image enhancement method for identifying peak value transformation nonlinear curve | |
| Zhao et al. | An adaptive low-illumination image enhancement algorithm based on weighted least squares optimization | |
| Wen et al. | A survey of image dehazing algorithm based on retinex theory | |
| CN112308793A (en) | Novel method for enhancing contrast and detail of non-uniform illumination image | |
| CN113256533B (en) | Self-adaptive low-illumination image enhancement method and system based on MSRCR | |
| CN115760630A (en) | Low-illumination image enhancement method | |
| CN109859138A (en) | A kind of infrared image enhancing method based on human-eye visual characteristic | |
| CN112435184A (en) | Haze sky image identification method based on Retinex and quaternion |
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 |