CN112565718A - Underwater image processing method based on Retinex and high dynamic range image gradient compression - Google Patents
Underwater image processing method based on Retinex and high dynamic range image gradient compression Download PDFInfo
- Publication number
- CN112565718A CN112565718A CN202011385470.6A CN202011385470A CN112565718A CN 112565718 A CN112565718 A CN 112565718A CN 202011385470 A CN202011385470 A CN 202011385470A CN 112565718 A CN112565718 A CN 112565718A
- Authority
- CN
- China
- Prior art keywords
- image
- underwater
- retinex
- compression
- dynamic range
- 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
- 238000007906 compression Methods 0.000 title claims abstract description 38
- 230000006835 compression Effects 0.000 title claims abstract description 37
- 238000003672 processing method Methods 0.000 title claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 32
- 238000001914 filtration Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 33
- 238000005286 illumination Methods 0.000 abstract description 23
- 230000000694 effects Effects 0.000 abstract description 20
- 125000001475 halogen functional group Chemical group 0.000 abstract description 8
- 230000002708 enhancing effect Effects 0.000 abstract description 2
- 238000011156 evaluation Methods 0.000 description 17
- 241000251468 Actinopterygii Species 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 241000242757 Anthozoa Species 0.000 description 2
- 235000014653 Carica parviflora Nutrition 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 239000012736 aqueous medium Substances 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000001525 retina Anatomy 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000036548 skin texture Effects 0.000 description 1
- 238000002945 steepest descent method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/10—Circuitry of solid-state image sensors [SSIS]; Control thereof for transforming different wavelengths into image signals
- H04N25/11—Arrangement of colour filter arrays [CFA]; Filter mosaics
-
- 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
- G06T5/90—Dynamic range modification of images or parts thereof
-
- 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
-
- 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
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- 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
- G06T2207/20172—Image enhancement details
- G06T2207/20208—High dynamic range [HDR] image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Color Image Communication Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an underwater image processing method based on Retinex and high dynamic range image gradient compression, which comprises the steps of firstly carrying out multi-scale Retinex algorithm processing on an underwater image, then carrying out pixel value compression on the image subjected to Retinex processing by using a gradient domain high dynamic range image compression algorithm, then attenuating low-frequency noise by a high-pass filter, enhancing the edge of the image, and finally recovering the chroma by applying a color saturation adjusting function. The method can inhibit the halo effect caused after the Retinex method processes the underwater image, improve the color cast influence brought by the underwater environment, and effectively process the problems of low illumination, uneven illumination and the like of the underwater image.
Description
Technical Field
The invention relates to an image processing method, in particular to an underwater image processing method based on Retinex and high dynamic range image gradient compression.
Background
The underwater image is an image shot underwater by using an optical imaging system, is an important carrier of ocean information, and is widely applied to the fields of ocean energy exploration, ocean ecological protection, ocean military and the like at present. However, due to the fact that light rays with different wavelengths have different attenuation characteristics when being transmitted in water and the influence of suspended particles, artificial light sources and the like in an underwater scene, the underwater image often has the quality degradation problems of low contrast, fuzzy texture, color distortion, non-uniform illumination, limited visual range and the like, and great inconvenience is brought to practical application and scientific research. In order to restore clear and real underwater images, various underwater image processing methods are available at present.
Conventional underwater image processing algorithms include histogram equalization, homomorphic filtering, Retinex algorithm and the like, wherein Retinex theory is an image enhancement method established on the basis of scientific experiments and analysis and is proposed by Edwin. Liu Xiao Yang and the like replace Gaussian filtering with bilateral filtering on the basis of Retinex algorithm, so that color distortion and detail loss of images are avoided; fu et al first propose a simple underwater image color deviation correction algorithm, then propose a variable framework separating direct light and reflected light based on retina cortex theory, and finally enhance the separated illumination component by different enhancement strategies, thereby enhancing the contrast of the underwater image; li Hui Fang et al uses variation optimization technique, and adds projection normalization steepest descent method in the technique, and through the combination of the two techniques to remove the illumination component of the remote sensing image, i.e. low frequency component, it can obviously eliminate the gray level nonuniformity of the remote sensing image, and keep the texture information of the image, etc. Although the above method can solve the quality degradation phenomenon of the underwater image to a certain extent, there are still some problems, for example, the multi-scale retinex (msr) algorithm processes each channel of RGB, when processing a single channel, it is easy to cause imbalance after each channel is processed, and further other interference noise is introduced, which causes local color distortion of the image, halo effect ("halo" effect) occurs, that is, a bright boundary appears as a brighter halo at a boundary, and a dark boundary appears as a darker black ring at a bright-dark boundary of a significant edge in the image, so that the overall effect of the image will be deteriorated, and the visual effect is not good.
High Dynamic Range (HDR) images can fully record luminance information that can be perceived by human eyes, and have a larger Dynamic Range and more detailed information than low Dynamic Range images. However, the existing mainstream display device is a low dynamic range display device, and is not capable of directly displaying the HDR image, so the HDR compression algorithm is generated by mapping the dynamic range of the HDR image to the dynamic range that can be displayed by a general display device. The HDR compression algorithm is focused and studied intensively by many researchers due to its superiority in contrast reproduction and halo removal, but the existing HDR compression algorithm also compresses the color gamut in the high dynamic range image compression process, and is prone to color distortion.
Disclosure of Invention
The invention provides an underwater image processing method based on Retinex and high dynamic range image gradient compression, aiming at solving the technical problems in the prior art.
The technical solution of the invention is as follows: an underwater image processing method based on Retinex and high dynamic range image gradient compression is sequentially carried out according to the following steps:
step 1: inputting an underwater image;
step 2: carrying out multi-scale Retinex algorithm processing on the underwater image;
and step 3: high dynamic range image gradient compression is carried out on the image, namely pixel values R of the image are mapped to a compression range Rmin,RmaxThe output pixel value is R through normalization processingnewThe image of (a);
the R ismin＝avg(R)-n·var(R)、Rmax＝avg(R)+n·var(R)、
In the formula: avg (R) represents the average value of the image pixel value R, var (R) represents the variance of the image pixel value R, and n is a compression range adjustment coefficient;
and 4, step 4: and carrying out high-pass filtering processing on the image, wherein the filter function H (u, v) is as follows:
in the formula: gamma rayHAnd gammaLFor high-pass filter parameters, gammaH>1,γL<1; c is a constant for controlling the sharpness of the function slope, making the function in gammaHAnd gammaLD (u, v) is the distance of a point in the frequency domain from a central point; d0Is a characteristic value of the attenuation;
and 5: the color saturation adjustment is performed on the image, the color saturation S is adjusted by utilizing an HSL space, and the calculation formula is as follows:
said max (rgb) is the maximum of the three channel pixel values of image R, G, B, said min (rgb) is the minimum of the three channel pixel values of image R, G, B;
defining a color saturation adjustment value alpha according to the following formula, wherein the value of the auxiliary variable i is-1:
after adjustment is calculated as followsRGBnew：
And 6, outputting the processed underwater image.
The invention uses the gradient domain high dynamic range image compression algorithm to compress the pixel value of the image processed by Retinex, then attenuates low-frequency noise through a high-pass filter, enhances the image edge, and finally recovers the chroma by applying a color saturation adjusting function. The method can inhibit the halo effect caused after the Retinex method processes the underwater image, improve the color cast influence brought by the underwater environment, and effectively process the problems of low illumination, uneven illumination and the like of the underwater image.
Drawings
FIG. 1 is an underwater image artwork used by embodiments of the present invention.
Fig. 2 is a graph comparing the processing results of the entire low-illuminance image according to the present invention and the prior art.
Fig. 3 is a comparison graph of the processing result of the local low-illumination image according to the present invention and the prior art.
FIG. 4 is a graph comparing the results of the present invention and prior art processing of partially reflected underwater images.
Detailed Description
The invention relates to an underwater image processing method based on Retinex and high dynamic range image gradient compression, which is sequentially carried out according to the following steps:
step 1: inputting an underwater image;
step 2: carrying out multi-scale Retinex algorithm (MSR) processing on the underwater image, and attenuating low-frequency components in illumination components in the underwater image, wherein the processed underwater image has a halo effect;
and step 3: high dynamic range image gradient compression is carried out on the image, namely pixel values R of the image are mapped to a compression range Rmin,RmaxThe output pixel value is R through normalization processingnewThe image of (a);
the R ismin＝avg(R)-n·var(R)、Rmax＝avg(R)+n·var(R)、
In the formula: avg (R) represents the average value of the image pixel value R, var (R) represents the variance of the image pixel value R, and n is a compression range adjustment coefficient.
The halo effect is inhibited by compressing the pixel value change rate, namely the compression degree is higher in places with large change rates, and the compression degree is correspondingly reduced in places with small change rates, so that the realization result is effective, the robustness is realized, and better image details can be kept; the compression effect is more obvious when n is larger, but more noise is introduced and color is lost;
and 4, step 4: and carrying out high-pass filtering processing on the image, wherein the filter function H (u, v) is as follows:
in the formula: gamma rayHAnd gammaLFor high-pass filter parameters, gammaH>1,γL<1; c is a constant for controlling the sharpness of the function slope, making the function in gammaHAnd gammaLD (u, v) is the distance of a point in the frequency domain from a central point; d0Is a characteristic value of the attenuation, D0The greater the detail enhancement is;
and 5: the color saturation adjustment is performed on the image, the color saturation S is adjusted by utilizing an HSL space, and the calculation formula is as follows:
said max (rgb) is the maximum of the three channel pixel values of image R, G, B, said min (rgb) is the minimum of the three channel pixel values of image R, G, B;
defining a color saturation adjustment value alpha according to the following formula, wherein the value of the auxiliary variable i is-1:
after adjustment is calculated as followsRGBnew：
And 6, outputting the processed underwater image.
Experiment:
the computer parameter used in the simulation experiment is Intel (R) core (TM) M-5Y10c @0.80GHZ1.00GHZ, the used operating system is 64 bits of Chinese version Windows 8.1, and the used simulation software is MATLAB R2017a 64 bits.
The 3 underwater images shown in fig. 1 are selected, the formats are all jpg, and the problems of the images respectively correspond to three typical problems of the underwater images with uneven illumination, namely, the phenomena of overall low illumination (a), local low illumination (b), reflection (c) in the images and the like. The invention is compared with other four traditional algorithms: HE. HF, MSR, gradient domain HDR compression algorithms were compared.
Subjective evaluation:
fig. 2 is a graph comparing the processing results of the entire low-illuminance image according to the present invention and the prior art. In fig. 2, (a) to (f) are the original image, HE, HF, MSR, gradient domain HDR compression algorithm, and the present invention, respectively.
The parameters of each algorithm are chosen as follows: wherein the HF algorithm parameter is gammaH＝2,γL＝0.5,c＝2,D0256; the MSR algorithm scales are respectively 16, 32 and 64; the scale of the invention is 16, 32 and 64, the three channels of compression adjustment coefficients RGB are 20, 30 and 10 respectively, the color saturation increment i is-0.2, and the parameter of the high-pass filter is gammaH＝2,γL＝0.9,c＝1,D0256. The overall illumination of the original image is very low and has serious blue-green color cast; although the HE algorithm enhances the contrast, the HE algorithm does not improve the color cast influence because the HE method changes the distribution of image gray levels instead of increasing the gray levels, and the color cast phenomenon is also aggravated while the gray levels are homogenized; the HF method is very effective in improving contrast and noise, but has a "Holo" effect and color shift effects; the MSR method has good effect on improving color cast and color degree, but the image has serious noise and a 'Holo' effect; the gradient domain HDR compression method has a poor effect on image processing with a low dynamic range; the invention has satisfactory color cast, contrast, noise processing and color expression effects.
Fig. 3 is a comparison graph of the processing result of the local low-illumination image according to the present invention and the prior art. In fig. 3, (a) to (f) are the original image, HE, HF, MSR, gradient domain HDR compression algorithm, and the present invention, respectively.
The parameters of each algorithm are chosen as follows: HF algorithm parameter is gammaH＝2,γL＝0.25,c＝2,D0256; the MSR algorithm scales are respectively 64, 128 and 256; the invention has the scale of 64, 128 and 256, the three channels of the compression adjustment coefficients RGB are respectively 13, 14 and 19, the color saturation increment i is 0.5, and the parameter of the high-pass filter is gammaH＝2,γL＝0.99,c＝2,D0256. The distant corals in the original image are completely in the dark and difficult to distinguish; the HE method enables the corals in the dark to be completely exposed, the contrast is good, but the color is seriously inconsistent with the reality, and the over-enhancement phenomenon is shown in the bright area; the HF method is better in the details of dark places and the overall brightness, but poor in the processing of color expression and has color cast; the MSR method performs well on the color of the image, but does not handle well on the detail presentation in dark places; the gradient domain HDR compression method has good performance on dark details and image edges, but has poor effect on color performance and overall brightness; the invention has good performance on color presentation, image edge and dark detail enhancement.
FIG. 4 is a graph comparing the results of the present invention and prior art processing of partially reflected underwater images. In fig. 4, (a) to (f) are the original image, HE, HF, MSR, gradient domain HDR compression algorithm, and the present invention, respectively.
The parameters of each algorithm are chosen as follows: HF algorithm parameter is gammaH＝2,γL＝0.25,c＝2,D0256; the MSR algorithm scales are respectively 64, 128 and 256; the invention has the scale of 64, 96 and 256, the three channels of the compression adjustment coefficients RGB are respectively 30, 30 and 40, the color saturation increment i is 0.1, and the parameter of the high-pass filter is gammaH＝2,γL＝0.99,c＝2,D0256. The original image shows a reflection phenomenon on the surface of the fish skin and the whole image shows strong light and shade contrast. The HE processing result has serious over-enhancement phenomenon, and the light reflection phenomenon is intensified; the visual effect of HF treatment is good, the texture on the surface of the fish skin is clearer, but color cast and a 'Holo' effect exist; the MSR method improves the phenomenon of integral illumination nonuniformity of the image, but poor treatment of the phenomena such as color, noise, color cast and the like; gradient domain HDR compression method is expressed in contrast and color densityPoor and accompanied by color bias phenomenon; the invention has good improvement performance in the aspects of blurring, color cast and the like, the detail performance of the light reflecting area is good, the fish skin texture in the light reflecting area can be seen, and the 'Holo' effect is also inhibited.
Experimental results prove that the method has better processing results on the problems of color cast, noise and color attenuation caused by the underwater environment when processing the underwater image with uneven illumination.
Objective evaluation:
at present, image evaluation indexes are generally three types, namely a full reference type, a partial reference type and a no reference type. The full reference type refers to the existence of an ideal image, and a comparison method is carried out by comparing the image to be evaluated with the ideal image; the partial reference type is to compare and analyze the image to be evaluated by taking partial characteristic information of the ideal image as reference so as to obtain an image quality evaluation result; the no-reference type is an image quality evaluation method completely separated from the dependence on an ideal reference image. Because the underwater environment is poor when the underwater image is usually obtained, and an ideal image is difficult to obtain, the invention adopts a non-reference evaluation standard. Currently, common underwater image evaluation standards include methods such as UCIQE (offline color image quality), uiqm (offline image quality measurement), and the like, wherein the evaluation standard of the UCIQE refers to three indexes of color concentration, contrast and saturation of an image, and an image score is obtained through linear weighting; the UIQM selects indices of color density, sharpness, and contrast deteriorated by the characteristics of the aqueous medium, and evaluates them. The invention adopts the two comprehensive evaluation standards to carry out objective evaluation on each image, and also considers the operation speed of each algorithm. HE. The evaluation index values of HF, MSR, gradient domain HDR compression algorithm and the three kinds of uneven illumination image processing results shown in fig. 1 according to the present invention are shown in tables 1 to 3.
Table 1 shows the objective evaluation results of the overall low-illumination image processing results for each algorithm. The results show that: the image processed by the method has satisfactory quality, the scores of UCIQE and UIQM are the highest, and the method has good results in the aspects of color concentration, contrast and definition when processing low-illumination images.
Table 2 shows the objective evaluation results of the local low-illumination image processing results of the algorithms. The results show that: the method has ideal effect when processing the local low-illumination underwater image, has good effect on the aspects of image definition, contrast, color concentration and the like, has the highest score of UCIQE and UIQM, and shows that the image processed by the method has the best quality.
Table 3 shows the objective evaluation results of the local reflection underwater image processing results by each algorithm. The results show that: from the UCIQE score, the HE score is highest because the color density and the average brightness of the HE score are good, but in combination with the subjective impression of the image, the HE processing result is extremely serious, the image details are lost, and the light reflection phenomenon is not improved. Therefore, the underwater image enhancement effect is generally evaluated comprehensively by combining objective evaluation indexes and subjective evaluation indexes. The invention has good performance on clear images and contrast, so the UIQM has the highest score, and the invention combines subjective impression, has no over-enhancement phenomenon on color performance, good detail performance and good improvement on color cast phenomenon.
TABLE 1 evaluation index of overall low-illumination image by each method
TABLE 2 evaluation index of local low-illumination image by each method
TABLE 3 evaluation indexes of local reflection image
Claims (1)
1. An underwater image processing method based on Retinex and high dynamic range image gradient compression is characterized by comprising the following steps in sequence:
step 1: inputting an underwater image;
step 2: carrying out multi-scale Retinex algorithm processing on the underwater image;
and step 3: high dynamic range image gradient compression is carried out on the image, namely pixel values R of the image are mapped to a compression range Rmin,RmaxThe output pixel value is R through normalization processingnewThe image of (a);
the R ismin＝avg(R)-n·var(R)、Rmax＝avg(R)+n·var(R)、
In the formula: avg (R) represents the average value of the image pixel value R, var (R) represents the variance of the image pixel value R, and n is a compression range adjustment coefficient;
and 4, step 4: and carrying out high-pass filtering processing on the image, wherein the filter function H (u, v) is as follows:
in the formula: gamma rayHAnd gammaLFor high-pass filter parameters, gammaH>1,γL<1; c is a constant for controlling the sharpness of the function slope, making the function in gammaHAnd gammaLD (u, v) is the distance of a point in the frequency domain from a central point; d0Is a characteristic value of the attenuation;
and 5: the color saturation adjustment is performed on the image, the color saturation S is adjusted by utilizing an HSL space, and the calculation formula is as follows:
said max (rgb) is the maximum of the three channel pixel values of image R, G, B, said min (rgb) is the minimum of the three channel pixel values of image R, G, B;
defining a color saturation adjustment value alpha according to the following formula, wherein the value of the auxiliary variable i is-1:
RGB after adjustment is calculated as followsnew：
And 6, outputting the processed underwater image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011385470.6A CN112565718A (en) | 2020-12-01 | 2020-12-01 | Underwater image processing method based on Retinex and high dynamic range image gradient compression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011385470.6A CN112565718A (en) | 2020-12-01 | 2020-12-01 | Underwater image processing method based on Retinex and high dynamic range image gradient compression |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112565718A true CN112565718A (en) | 2021-03-26 |
Family
ID=75046960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011385470.6A Pending CN112565718A (en) | 2020-12-01 | 2020-12-01 | Underwater image processing method based on Retinex and high dynamic range image gradient compression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112565718A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070172139A1 (en) * | 2006-01-20 | 2007-07-26 | Rising Hawley K | Method and apparatus for image processing with color constancy |
| CN102289792A (en) * | 2011-05-03 | 2011-12-21 | 北京云加速信息技术有限公司 | Method and system for enhancing low-illumination video image |
| CN102752605A (en) * | 2011-09-30 | 2012-10-24 | 新奥特（北京）视频技术有限公司 | Method and system for regulating image/video color saturation degree |
| JP2013211662A (en) * | 2012-03-30 | 2013-10-10 | Eizo Corp | Gradation correction apparatus or method thereof |
| CN103907345A (en) * | 2011-11-01 | 2014-07-02 | 歌乐株式会社 | Image processing device, image pickup apparatus, and storage medium storing image processing program |
| US20150043806A1 (en) * | 2013-08-08 | 2015-02-12 | Adobe Systems Incorporated | Automatic geometry and lighting inference for realistic image editing |
| CN104680490A (en) * | 2015-02-13 | 2015-06-03 | 中科创达软件股份有限公司 | Method for enhancing visuality of text image |
| CN107578370A (en) * | 2017-09-20 | 2018-01-12 | 武汉鸿瑞达信息技术有限公司 | The quick undistorted method and apparatus for improving picture saturation degree under a kind of RGB patterns |
| CN111210395A (en) * | 2020-01-09 | 2020-05-29 | 大连海事大学 | Retinex underwater image enhancement method based on gray value mapping |
-
2020
- 2020-12-01 CN CN202011385470.6A patent/CN112565718A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070172139A1 (en) * | 2006-01-20 | 2007-07-26 | Rising Hawley K | Method and apparatus for image processing with color constancy |
| CN102289792A (en) * | 2011-05-03 | 2011-12-21 | 北京云加速信息技术有限公司 | Method and system for enhancing low-illumination video image |
| CN102752605A (en) * | 2011-09-30 | 2012-10-24 | 新奥特（北京）视频技术有限公司 | Method and system for regulating image/video color saturation degree |
| CN103907345A (en) * | 2011-11-01 | 2014-07-02 | 歌乐株式会社 | Image processing device, image pickup apparatus, and storage medium storing image processing program |
| JP2013211662A (en) * | 2012-03-30 | 2013-10-10 | Eizo Corp | Gradation correction apparatus or method thereof |
| US20150043806A1 (en) * | 2013-08-08 | 2015-02-12 | Adobe Systems Incorporated | Automatic geometry and lighting inference for realistic image editing |
| CN104680490A (en) * | 2015-02-13 | 2015-06-03 | 中科创达软件股份有限公司 | Method for enhancing visuality of text image |
| CN107578370A (en) * | 2017-09-20 | 2018-01-12 | 武汉鸿瑞达信息技术有限公司 | The quick undistorted method and apparatus for improving picture saturation degree under a kind of RGB patterns |
| CN111210395A (en) * | 2020-01-09 | 2020-05-29 | 大连海事大学 | Retinex underwater image enhancement method based on gray value mapping |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108053374B (en) | Underwater image enhancement method combining bilateral filtering and Retinex | |
| CN107507145B (en) | Underwater image enhancement method based on different color space self-adaptive histogram stretching | |
| CN111696052B (en) | Underwater image enhancement method and system based on red channel weakness | |
| CN111968041A (en) | Self-adaptive image enhancement method | |
| CN110473152B (en) | Image enhancement method based on improved Retinex algorithm | |
| CN109886885B (en) | Image enhancement method and system based on Lab color space and Retinex | |
| Peng et al. | Multi-algorithm fusion of RGB and HSV color spaces for image enhancement | |
| Zhou et al. | A multifeature fusion method for the color distortion and low contrast of underwater images | |
| Yu et al. | An improved multi-scale retinex fog and haze image enhancement method | |
| CN111861896A (en) | UUV-oriented underwater image color compensation and recovery method | |
| CN111968065A (en) | Self-adaptive enhancement method for image with uneven brightness | |
| CN104766285A (en) | Self-adapting enhancement method of underwater degraded image | |
| CN116188339A (en) | Retinex and image fusion-based scotopic vision image enhancement method | |
| Muniraj et al. | Underwater image enhancement by modified color correction and adaptive Look-Up-Table with edge-preserving filter | |
| Wen et al. | A survey of image dehazing algorithm based on retinex theory | |
| CN114037641A (en) | Low-illumination image enhancement method, device, equipment and medium | |
| CN112907470A (en) | Underwater image recovery method based on Lab color gamut transformation, classification and white balance | |
| CN112565718A (en) | Underwater image processing method based on Retinex and high dynamic range image gradient compression | |
| Wang et al. | An Improved Unsupervised Color Correction Algorithm for Underwater Image | |
| Fan et al. | Underwater image enhancement algorithm combining color correction and multi-scale fusion | |
| Huo et al. | Underwater image restoration based on color correction and red channel prior | |
| CN111028200A (en) | Image optimization method based on no-reference image quality evaluation and MSR | |
| Peicheng et al. | Backlit image enhancement based on illumination-reflection imaging model | |
| CN113554572B (en) | Image enhancement method and system based on improved Retinex | |
| Wei et al. | MULTISCALE RETINEX COLOR RESTORATION WITH ADAPTIVE GAMMA CORRECTION FOR FOGGY IMAGE ENHANCEMENT |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210326 |
|
| WD01 | Invention patent application deemed withdrawn after publication |