CN104112253B - A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering - Google Patents
A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering Download PDFInfo
- Publication number
- CN104112253B CN104112253B CN201410266769.8A CN201410266769A CN104112253B CN 104112253 B CN104112253 B CN 104112253B CN 201410266769 A CN201410266769 A CN 201410266769A CN 104112253 B CN104112253 B CN 104112253B
- Authority
- CN
- China
- Prior art keywords
- image
- light
- radiation field
- level
- light radiation
- 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
Abstract
The present invention provides the low-light (level) image/video Enhancement Method filtered based on self-adapting multi-dimension, the present invention is based on air photon diffusion models, the adaptive-filtering based on region characteristic and adaption brightness repair theory, in low-light (level) video and field of image enhancement, the adaptive multi-scale filtering theory and method proposed, the brightness of image after processing can be made to get a promotion, the contrast and clarity of image or video are lifted at the same time, with low using classical mean filter, computation complexity;Inner parameter is adaptively asked for, accommodation is wider, can be adapted for coloured image or gray level image, has good versatility.
Description
Technical field
The invention belongs to technical field of image processing, more particularly, to available for the adaptive of single image or continuous videos
At a kind of image of adaptive mean filter and adapting to image brightness reparation based on area attribute of low-light (level) enhancing
Reason method.
Background technology
With the development of informatization, video technique is widely applied, but when image/video collection field
When the illumination of scape is too low, the brightness of video pictures and clarity drastically reduce, this is because what the video image of low-light (level) had
Tonal range is narrow, and the correlation between adjacent pixel is higher, and the feature such as change unobvious of gray value, causes in image
Useful information and useless noise are included within a very narrow tonal range so that people can not normally observe
Video under low-light (level), it is useful therefrom to obtain in order to improve the effect of video quality and vision in the case of low-light (level)
Information, it is necessary to the processing strengthened the video of low-light (level).
Current researcher both domestic and external is by the analysis to low-light (level) video, using the basic theories of digital image processing,
Derive many rich in innovative low-light (level) video enhancement algorithm.Studies in China person mainly uses local histogram equalization
Change, the enhancing of fuzzy logic theory, based on wavelet transformation and the methods of mathematical morphology low-light (level) image strengthened;And
Foreign study person generally employs nonlinear tone mapping, based on adaptive bilateral filtering and BP nerve spatially and temporally
The methods of network, strengthens low-light (level) image.The method on low-light (level) enhancing of these domestic and international researchers improves
The quality of image after enhancing, but there are low-light (level) enhancing algorithm complex is high, visual effect is general, can not be according to field for these methods
In place of the deficiencies of scape automatic adjusument brightness.
The content of the invention
It is an object of the invention to the shortcoming of method before overcoming, and proposes that a kind of low-light (level) adaptive in real time increases
Strong algorithms, can be right according to different scene characteristic adaptive determining relevant parameters this process employs air photon diffusion models
There is universal enhancing effect in the video and image of low-light (level) environment, robustness is higher, and enhanced image visual effect is preferable.
To achieve the above object, the following technical solutions are proposed by the present invention:
Variable is defined as follows in order to describe:
The size of pending entire image is imagesize:That is imagesize=Height*Width, wherein Height
For the height of image, Width is the width of image.
Mean filter is avefilter (N), and wherein N is the window size of mean filter.
Basic light intensity A:The intensity of atmosphere light component, the present invention use adaptive acquiring method, can be according to different
Image property judges different values.
Brightness reparative factor ω:The brightness in transmissivity distribution function is repaired, the present invention uses adaptive acquiring method,
Different values can be asked for according to the brightness of different images.
Transmissivity distribution function t (x, y):Light disturbed by atmospheric environment after not by the ratio of scattered portion, 0≤t
(x,y)≤1。
Original low-light (level) image is Ioriginal:If coloured image, the value of R, G, B triple channel is respectively Ioriginal_R、
Ioriginal_G、Ioriginal_B;If gray level image then IoriginalFor single channel.
A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering, is realized in accordance with the following steps:
1) the light radiation field of low-light (level) image is defined, takes the maximum of the triple channel of each pixel of low-light (level) image
Ioriginal, obtain light radiation field initial value w (x, y);
2) by the estimation to low-light (level) brightness, adaptive brightness reparative factor ω is prepared:Obtain the flat of w (x, y)
Average:NAmean=mean (w (x, y)), wherein, NAmeanFor the average of w (x, y), its formula isWherein its
Middle na is the sum of pixel in image;xi' be image in certain point brightness value;W (x, y) is the initial value of light radiation field,
Adaption brightness reparative factor ω=k0-(NAmean/k1)2, wherein NAmeanReflect the average brightness of entire image, k1To set
Fixed control parameter, the parameter reflect the bright-dark degree of image;
3) by the estimation to low-light (level) brightness, basic light intensity A is prepared:The size for defining entire image is imagesize
=Height*Width, wherein Height are that the high Width of image is the width of image, the histogram of statistics light radiation field w (x, y)
Histogram_w;Calculate the cumulative of histogram andWhenIt is right
The j answered1As basic light intensity A, wherein k2For the control parameter of setting;
4) two adaptive mean filter avefilter1 (N, x, y), avefilter2 (N, x, y), definition are prepared
Local window mean filter is avefilter (N, x, y), its formula is:
Wherein p, q are the coordinate of pixel in two dimensional image, just
The length of side N=2r+1 of square window;
The standard deviation of light radiation field in node-by-node algorithm low-light (level) image local window:Nstd=std (w (x, y)), wherein, w (x,
Y) it is the initial value of light radiation field;Std () is the operation for asking for standard deviation, its formula isIts
Middle n is the sum of pixel in 2r+1 length of side square windows;xiFor the pixel value of every bit in image;NmeanFor the 2r+1 length of sides
Brightness of image average value in square window;NstD is the standard deviation of image in 2r+1 length of side square windows, reflects local window
The difference of image in mouthful;
Normalized is done to the standard deviation of light radiation field:Nstd2=Nstd 2/(k3+Nstd 2), k3For the control parameter of setting,
Control the degree of crook of normalized curve;
Ask for adaptive window size N:N=2*floor (k4*Nstd2)+1, wherein, floor () is downward round numbers
Operation, k4For the control parameter of setting, the window of wave filter can be controlled, its value is bigger, and filtering window is bigger, after filtering
Image it is more smooth;
5) luminance raising processing is carried out to low-light (level) image:
Light radiation field initial value is done and is handled for the first time:Wherein
For convolution operator;Avefilter1 (N, x, y) is the mean filter template previously prepared;At the beginning of w (x, y) is light radiation field
Initial value;
Second of processing is done to light radiation field initial value,
WhereinFor convolution operator;Avefilter2 (2*N+1, x, y) is the mean filter template previously prepared;W (x, y) is light
Radiation field initial value:
The fine estimation of light radiation field:U (x, y)=ave1 (x, y) * (1-Nstd2)+ave2(x,y)*Nstd2, wherein
Nstd2Previously to prepare the normalized light radiation field standard deviation required by mean filter;Ave1 (x, y) and ave2 (x, y) is respectively
To be filtered the value tried to achieve twice to light radiation field initial value;U (x, y) is defined as the fine estimation of light radiation field;
Transmissivity distribution function s (x, y):Wherein ω is previously prepared
Adaption brightness reparative factor;B=255-A;U (x, y) is calculated light radiation field fine estimation;
According to recovery formula J (x, y)=(Ioriginal(x, y)-A) * s (x, y)+A, by gray level image brightness value Ioriginal
And value J (x, y) after recovery is calculated in basic light intensity A, transmissivity distribution function s (x, y) substitutions, you can schemes after being restored
As J (x, y).
In above-mentioned technical proposal, original low-light (level) image is coloured image, calculates the three of each pixel of low-light (level) image
The maximum of passage, you can obtain light radiation field initial value w (x, y)=max (Ioriginal_R,Ioriginal_G,Ioriginal_B), it is multiple
Former formula J (x, y)=(Ioriginal(x, y)-A) * s (x, y)+A, respectively by the value I of triple channeloriginal_R(x,y)、Ioriginal_G
(x,y)、Ioriginal_B(x, y) and basic light intensity A, transmissivity distribution function s (x, y) are substituted into and R, G, B triple channel are calculated
Disengaging value JR(x,y)、JG(x,y)、JB(x, y), you can image J (x, y) after being restored;
In above-mentioned technical proposal, if image is gray level image, light radiation field initial value w (x, y)=Ioriginal, restore formula
J (x, y)=(Ioriginal(x, y)-A) * s (x, y)+A, by gray level image brightness value IoriginalAnd basic light intensity A, transmissivity
Value J (x, y) after recovery is calculated in distribution function s (x, y) substitutions, you can image J (x, y) after being restored;
In above-mentioned technical proposal, parameter k1:Adjustable control parameter, scope are 0.7≤k1≤ 0.95, actual value is according to image
Property concrete decision.
In above-mentioned technical proposal, parameter k2:Adjustable control parameter, scope are 0≤k2≤ 0.1, actual value is according to image property
Concrete decision.
The present invention is based on air photon diffusion models, the adaptive-filtering based on region characteristic and adaption brightness reparation
Theory, in low-light (level) video and field of image enhancement, the adaptive multi-scale filtering theory and method of proposition, can make place
The brightness of image gets a promotion after reason, while lifts the contrast and clarity of image or video, has using classical average
Wave filter, computation complexity are low;Inner parameter is adaptively asked for, accommodation is wider, can be adapted for coloured image or ash
Image is spent, there is good versatility.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
To obtain other attached drawings according to these attached drawings.
Fig. 1 is the flow chart of the method for the present invention.
Fig. 2 is low-light (level) picture.
Design sketch after Fig. 3 is strengthened for Fig. 2.
Embodiment
Below in conjunction with the attached drawing of the present invention, technical scheme is clearly and completely described, it is clear that institute
The embodiment of description is only part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention,
Those of ordinary skill in the art's all other embodiments obtained without creative efforts, belong to this hair
The scope of bright protection.
According to Fig. 1, Fig. 2, Fig. 3:
Variable is defined as follows first in order to arthmetic statement:
By taking pending image graph 2 as an example, variable i magesize=Height*Widthian=600 × 400=120000.
Data of 8 pixel RGB component values as embodiment in 1 selection Fig. 2 are shown in Table, table 2 is 8 pictures of image after processing
Plain RGB component value, the position in the small square frame of location of pixels Fig. 2, Fig. 3.
Table 1 is 8 pixel RGB component values in original low-light (level) Fig. 2
Brightness reparative factor ω:The brightness in transmissivity distribution function is repaired, the present invention uses adaptive acquiring method,
Different values can be asked for according to the brightness of different images, is specifically shown in subsequent step (A1) -- (A2).
Basic light intensity A:The intensity of atmosphere light component, the present invention use adaptive acquiring method, can be according to different
Image property judges different values, referring specifically to subsequent step.
Transmissivity distribution function t (x, y):Light disturbed by atmospheric environment after not by the ratio of scattered portion, setting
Its interval is 0≤t (x, y)≤1.
Original low-light (level) image Ioriginal:Fig. 2 is coloured image, then the value of R, G, B triple channel is respectively defined as
Ioriginal_R、Ioriginal_G、Ioriginal_B。
Parameter k0:Adjustable control parameter, scope are 0.8≤k0≤ 1.1,0.95 is taken in the present embodiment.When value is larger,
The strength disposal of luminance raising is larger, and vice versa.
Parameter k1:Adjustable control parameter, scope are 100≤k1≤ 150,128 are taken in the present embodiment.It is bright when value is larger
The strength disposal of degree lifting is larger, and vice versa.
Parameter k2:Adjustable control parameter, scope are 0≤k2≤ 0.1,0.1 is taken in the present embodiment, the image when value is larger
Contrast it is stronger, vice versa.
Parameter k3:Adjustable control parameter, scope are 500≤k3≤ 1500,1000 are taken in the present embodiment.
Parameter k4:Adjustable control parameter, scope are 5≤k4≤ 20,11 are taken in the present embodiment.
First, initial processing steps are as follows:
The initial value w (x, y) of light radiation field is calculated first:
Because original low-light (level) image is coloured image, as shown in Fig. 2, calculating the triple channel of each pixel of low-light (level) image
Maximum, you can obtain light radiation field initial value w (x, y)=max (Ioriginal_R,Ioriginal_G,Ioriginal_B)。
Calculate adaptive brightness reparative factor ω:
(A1) average value of w (x, y) is obtained:NAmean=mean (w (x, y)), the size of its value reflects entire image
Monochrome information;Mean () is the operation for trying to achieve entire image average, its formula is(n=imagesize;
xi' be image element brightness value).
(A2) adaption brightness reparative factor ω is obtained:ω=k0-(NAmean/k1)2=0.95- (NAmean/128)2.Wherein
Nmean is the light radiation field average value calculated in above-mentioned steps.
Calculate adaptive basic light intensity A:
(B1) the histogram Histogram_w of light radiation field w (x, y) is counted.
(B2) calculate the cumulative of histogram andCalculate the cumulative of histogram and
J is the upper limit of cumulative sum, is changed for 8bit images j in the range of [0,255], when cumulative and be just met forWhen remember j be j1, then basic light intensity A=j1, k in the present embodiment2=0.1.
Adaptive mean filter, its step are:
Mean filter avefilter (N, x, y) is first defined, its formula is:
Wherein p, q are the amount of bias of pixel distance centre coordinate (x, y) in window, and N is window size, and r is window
Radius, r=(N-1)/2.
(C1) the light radiation field standard deviation in local window is calculated:Nstd=std (w (x, y)), wherein, w (x, y) is foregoing
The initial value of light radiation field;Std () is the operation for asking for standard deviation, its formula isWherein n
For the sum of pixel in window;xiFor the numerical value of every bit in local window;NmeanFor light radiation field average value in window.This
The size of local window is 9x9 in embodiment, therefore n=81.
(C2) standard deviation normalized:Nstd2=Nstd 2/(k3+Nstd 2), k3For the control parameter of setting, control normalization
The degree of crook of curve, k in the present embodiment3=1000.
(C3) adaptive window size N is asked for:N=2*floor (k4*Nstd2)+1, wherein, floor () is to take downwards
The operation of integer;K in the present embodiment4=11.
2nd, processing and restored image are filtered to low-light (level) image.
(D1) light radiation field initial value is done and handled for the first time
WhereinFor convolution operator;avefilter1(N,x,y)
For the mean filter template previously prepared;W (x, y) is light radiation field initial value;
(D2) second of processing is done to light radiation field initial value
WhereinFor convolution operator;
Avefilter2 (2*N+1, x, y) is the mean filter template previously prepared;W (x, y) is initial for light radiation field
Value;
(D3) fine estimation of light radiation field is obtained
U (x, y)=ave1 (x, y) * (1-Nstd2)+ave2(x,y)*Nstd2, wherein Nstd2Previously to prepare mean filter
Required normalized light radiation field standard deviation;Ave1 (x, y) and ave2 (x, y) is respectively to carry out two to light radiation field initial value
It is secondary to filter the value tried to achieve;U (x, y) is defined as the fine estimation of light radiation field;
(D4) transmissivity distribution function s (x, y) is obtained:
Wherein ω is the adaption brightness reparative factor previously prepared;B=
255-A;U (x, y) is the light radiation field estimate in previous step.
(D5) image after being restored according to air photon diffusion models
Because original low-light (level) image is coloured image, such as Fig. 2, then according to recovery formula J (x, y)=(Ioriginal(x,y)-
A) * s (x, y)+A, respectively by the value I of triple channeloriginal_R(x,y)、Ioriginal_G(x,y)、Ioriginal_B(x, y) and basis
Light intensity A, transmissivity distribution function s (x, y) substitute into the disengaging value J that R, G, B triple channel is calculatedR(x,y)、JG(x,y)、JB(x,
Y), you can image J (x, y) after being restored;
Table 2 is corresponding 8 pixel values of Fig. 2 after luminance raising
The above description is merely a specific embodiment, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained
Cover within protection scope of the present invention.Therefore, protection scope of the present invention answers the scope of the claims of being subject to.
Claims (2)
- A kind of 1. low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering, it is characterised in that:According to following step It is rapid to realize:1) the light radiation field of low-light (level) image is defined, takes the maximum I of the triple channel of each pixel of low-light (level) imageoriginal, Obtain light radiation field initial value w (x, y);2) by the estimation to low-light (level) brightness, adaptive brightness reparative factor ω is prepared:Obtain the average value of w (x, y): NAmean=mean (w (x, y)), wherein, NAmeanFor the average of w (x, y), mean () is the operation averaged, its formula isWherein na is the sum of pixel in image;x′iFor the brightness value of certain point in image;W (x, y) is light The initial value of radiation field;Adaption brightness reparative factor ω=k0-(NAmean/k1)2, wherein NAmeanReflect the bright of entire image Spend average value, k1For the control parameter of setting, which reflects the bright-dark degree of image;3) by the estimation to low-light (level) brightness, basic light intensity A is prepared:The size for defining entire image is imagesize= Height*Width, wherein Height are that the high Width of image is the width of image, the Nogata of statistics light radiation field initial value w (x, y) Scheme Histogram_w;Calculate the cumulative of histogram andHistogram_w, when Corresponding j1 is basic light intensity A, wherein k2For the control parameter of setting;4) two adaptive mean filter avefilter1 (N, x, y), avefilter2 (N, x, y) are prepared, define local window Mean filter is avefilter (N, x, y), its formula is: Wherein p, q be two dimensional image in pixel coordinate, the length of side N=2r+1 of square window;The standard deviation of light radiation field in node-by-node algorithm low-light (level) image local window:Nstd=std (w (x, y)), wherein, w (x, y) For the initial value of light radiation field;Std () is the operation for asking for standard deviation, its formula isIts Middle n is the sum of pixel in 2r+1 length of side square windows;xiFor the pixel value of every bit in image;NmeanFor the 2r+1 length of sides Brightness of image average value in square window;NstdFor the standard deviation of light radiation field in 2r+1 length of side square windows, office is reflected The difference of portion's image in window;Normalized is done to the standard deviation of light radiation field:Nstd2=Nstd 2/(k3+Nstd 2), k3For the control parameter of setting, control The degree of crook of normalized curve;Ask for adaptive window size N:N=2*floor (k4*Nstd2)+1, wherein, floor () is the behaviour of downward round numbers Make, k4For the control parameter of setting, the window of wave filter can be controlled, its value is bigger, and filtering window is bigger, filtered figure As more smooth;5) luminance raising processing is carried out to low-light (level) image:Light radiation field initial value is done and is handled for the first time:WhereinFor convolution Operator;Avefilter1 (N, x, y) is the mean filter template previously prepared;W (x, y) is light radiation field initial value;Second of processing is done to light radiation field initial value,Wherein For convolution operator;Avefilter2 (2*N+1, x, y) is the mean filter template previously prepared;W (x, y) is light radiation field Initial value:The fine estimation of light radiation field:U (x, y)=ave1 (x, y) * (1-Nstd2)+ave2(x,y)*Nstd2, wherein Nstd2For Previously prepare the normalized light radiation field standard deviation required by mean filter;Ave1 (x, y) and ave2 (x, y) is respectively to light Radiation field initial value is filtered the value tried to achieve twice;U (x, y) is defined as the fine estimation of light radiation field;Transmissivity distribution function s (x, y):Wherein ω is adaptive previously to have prepared Answer brightness reparative factor;B=255-A;U (x, y) is calculated light radiation field fine estimation;According to recovery formula J (x, y)=(Ioriginal(x, y)-A) * s (x, y)+A, by gray level image brightness value IoriginalAnd base Plinth light intensity A, transmissivity distribution function s (x, y), which are substituted into, is calculated value J (x, y) after recovery, you can image J after being restored (x, y)。
- 2. a kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering according to claim 1, its It is characterized in that:Parameter k1:Adjustable control parameter, scope are 0.7≤k1≤0.95。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410266769.8A CN104112253B (en) | 2014-06-16 | 2014-06-16 | A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410266769.8A CN104112253B (en) | 2014-06-16 | 2014-06-16 | A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104112253A CN104112253A (en) | 2014-10-22 |
CN104112253B true CN104112253B (en) | 2018-05-11 |
Family
ID=51709033
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410266769.8A Active CN104112253B (en) | 2014-06-16 | 2014-06-16 | A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104112253B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104463800B (en) * | 2014-12-04 | 2017-09-29 | 无锡日联科技股份有限公司 | A kind of gradation of image Enhancement Method |
CN104574318A (en) * | 2015-01-21 | 2015-04-29 | 宁夏大学 | Method and device for enhancing night vision image in self-adaption mode |
CN105046663B (en) * | 2015-07-10 | 2017-08-04 | 西南科技大学 | A kind of adaptive enhancement method of low-illumination image for simulating human visual perception |
CN105427265B (en) * | 2015-12-25 | 2018-05-29 | 武汉鸿瑞达信息技术有限公司 | A kind of method for enhancing color image contrast ratio and system |
CN106327439B (en) * | 2016-08-16 | 2019-01-01 | 华侨大学 | A kind of quick haze sky image clarification method |
CN106327450A (en) * | 2016-11-07 | 2017-01-11 | 湖南源信光电科技有限公司 | Method for enhancing low-light video image based on space-time accumulation and image degradation model |
CN107016659A (en) * | 2017-05-11 | 2017-08-04 | 国网青海省电力公司海南供电公司 | For the enhanced method of picture clarity in unmanned plane electric inspection process |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020914A (en) * | 2012-12-18 | 2013-04-03 | 华侨大学 | Rapid image defogging method based on spatial continuity principle |
CN104050636A (en) * | 2014-06-05 | 2014-09-17 | 华侨大学 | Method for enhancing granularity-controllable low-illuminance image |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7590303B2 (en) * | 2005-09-29 | 2009-09-15 | Samsung Electronics Co., Ltd. | Image enhancement method using local illumination correction |
-
2014
- 2014-06-16 CN CN201410266769.8A patent/CN104112253B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103020914A (en) * | 2012-12-18 | 2013-04-03 | 华侨大学 | Rapid image defogging method based on spatial continuity principle |
CN104050636A (en) * | 2014-06-05 | 2014-09-17 | 华侨大学 | Method for enhancing granularity-controllable low-illuminance image |
Non-Patent Citations (1)
Title |
---|
基于照度分割的局部多尺度Retinex算法;汪荣贵等;《电子学报》;20100531;第38卷(第5期);第1181-1186页 * |
Also Published As
Publication number | Publication date |
---|---|
CN104112253A (en) | 2014-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104112253B (en) | A kind of low-light (level) image/video Enhancement Method based on self-adapting multi-dimension filtering | |
CN106504212A (en) | A kind of improved HSI spatial informations low-luminance color algorithm for image enhancement | |
CN105046658B (en) | A kind of low-light (level) image processing method and device | |
CN105809643B (en) | A kind of image enchancing method based on adaptive block channel extrusion | |
CN109816608B (en) | Low-illumination image self-adaptive brightness enhancement method based on noise suppression | |
CN110232670B (en) | Method for enhancing visual effect of image based on high-low frequency separation | |
CN105205794A (en) | Synchronous enhancement de-noising method of low-illumination image | |
CN111968041A (en) | Self-adaptive image enhancement method | |
CN107862672B (en) | Image defogging method and device | |
CN106651817A (en) | Non-sampling contourlet-based image enhancement method | |
US7379561B2 (en) | Method and system for differentially and regularly modifying a digital image by pixel | |
CN113962898A (en) | Low-illumination image enhancement method based on illumination map optimization and adaptive gamma correction | |
CN105931206A (en) | Method for enhancing sharpness of color image with color constancy | |
CN111105359A (en) | Tone mapping method for high dynamic range image | |
CN115984535A (en) | Low-illumination image judgment and image enhancement method for drilling operation site | |
CN116188339A (en) | Retinex and image fusion-based scotopic vision image enhancement method | |
CN103514588B (en) | Image enchancing method and system | |
CN104050636B (en) | Method for enhancing granularity-controllable low-illuminance image | |
CN114187222A (en) | Low-illumination image enhancement method and system and storage medium | |
CN107358592B (en) | Iterative global adaptive image enhancement method | |
CN107451962A (en) | A kind of image defogging method and device | |
CN109859138B (en) | Infrared image enhancement method based on human visual characteristics | |
CN109003241B (en) | Linear image enhancement method with double control function | |
CN101478690B (en) | Image irradiation correcting method based on color domain mapping | |
CN108765310B (en) | Adaptive transmissivity restoration image defogging method based on multi-scale window |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C41 | Transfer of patent application or patent right or utility model | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20151019 Address after: 430000, Hubei Wuhan New Technology Development Zone, East Lake, Optics Valley Road, 35, silver long science and Technology Industrial Park, phase 4, 3, 3 Applicant after: WUHAN HONGRUIDA INFORMATION TECHNOLOGY CO., LTD. Address before: 1, No. 430000, Optics Valley International Chamber of Commerce building, 2-2 Kanto Road, East Lake New Technology Development Zone, Hubei, Wuhan, B1610 Applicant before: Wuhan Bo Ruida Information Technology Co., Ltd |
|
GR01 | Patent grant | ||
GR01 | Patent grant |