CN103955902A - Weak light image enhancing method based on Retinex and Reinhard color migration - Google Patents
Weak light image enhancing method based on Retinex and Reinhard color migration Download PDFInfo
- Publication number
- CN103955902A CN103955902A CN201410192799.9A CN201410192799A CN103955902A CN 103955902 A CN103955902 A CN 103955902A CN 201410192799 A CN201410192799 A CN 201410192799A CN 103955902 A CN103955902 A CN 103955902A
- Authority
- CN
- China
- Prior art keywords
- image
- retinex
- color
- reinhard
- weak light
- 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
Landscapes
- Image Processing (AREA)
Abstract
The invention relates to a weak light image enhancing method based on Retinex and Reinhard color migration. The method comprises the following steps of (1) reading optimal scale parameters of a gaussian function, (2) enhancing a weak light image through a single-scale Retinex algorithm, (3) calculating color information of the enhanced image through a Reinhard color migration algorithm, and meanwhile reading color information data of a reference image, (4) migrating the color information of the reference image to the enhanced image, and (5) obtaining an image with the rich colors and outstanding details. Compared with the prior art, the method has the advantages of being good in image enhancement effect, enriching image color information, good in real-time performance and the like.
Description
Technical field
The present invention relates to the Enhancement Method of a kind of faint light according to video image, especially relate to a kind of Enhancement Method of the weak light image based on Retinex and Reinhard color transfer.
Background technology
Image (Image) is human transmit message's important media, and in the information of human intelligence processing, visual communication information accounts for the mankind's the more than 75% of total amount of receiving information.Image is a kind of important means of mankind's obtaining information, the perception world and then reforming world.
Power grid construction work progress relate to project management, civil engineering, very many specialties such as electric, industry cross-operation very obviously, severe, the construction safety risk of construction environment and potential safety hazard more, these are all the difficult problems existing in power grid construction work progress.With the spirit of country " three collection five large " and the demand of the information-based management and control of strengthening engineering construction key node and the safety and stability of project management transition period form contrast be, current work progress, the management and control of working-yard are indifferent, lack the technological means of security monitoring.For this reason, the present invention is taking Urban Underground construction of transformer substation as point of penetration, and in the actual conditions of power construction process, construction lighting condition in early stage in underground substation is undesirable, and the video image gray scale obtaining is very low, and visual effect is undesirable.In order to obtain image clearly, by the improvement to Retinex algorithm, realize figure image intensifying.
The development of video enhancement techniques follows hard on the development of video technique to be carried out, and because video is set of number image sequence, therefore the many digital video images that can be applied to of existing algorithm for image enhancement/software strengthen.Also there is own feature but digital video strengthens algorithm, as require the processing of high real-time, frame of video to need continuity etc.
Traditional video enhancement method does not utilize high brightness information to add in enhancing conventionally, only adopts enhancing algorithm to process for video itself.Video enhancement techniques is summarized as two large classes: airspace enhancement facture (Spatial-based Domain Enhancement) and frequency domain strengthen facture (Frequency-based Domain Enhancement).Airspace enhancement method is normally for pixel operation.Mostly the algorithm based on airspace enhancement belongs to the method for direct augmented video itself, comprises greyscale transformation, histogram transformation, filter process, fuzzy logic enhancing, based on genetic algorithm optimization etc.It is that image is carried out to certain conversion that frequency domain strengthens, and in transform domain, the coefficient after conversion is carried out to computing, and then contravariant changes to original spatial domain, and the image being enhanced is a kind of disposal route indirectly.Conventional image conversion has: Fourier transform, wavelet transformation, discrete cosine transform, Walsh transform and Hotelling transform etc.What wherein application was the most ripe is Fourier transform, it taking amendment image Fourier transform as basis, image is transformed in frequency field, again the frequency field of image is carried out to filtering processing, spatial domain is changed in last contravariant, obtains the image after strengthening, but in the time that video is processed in real time, the complicated real-time of computing is poor, can not meet the demands.
Summary of the invention
Object of the present invention is exactly the Enhancement Method that a kind of weak light image based on Retinex and Reinhard color transfer is provided in order to overcome the defect that above-mentioned prior art exists.
Object of the present invention can be achieved through the following technical solutions:
An Enhancement Method for weak light image based on Retinex and Reinhard color transfer, is characterized in that, comprises the following steps:
1) read the optimal scale parameter of Gaussian function;
2) by single scale Retinex algorithm, weak light image is strengthened;
3) calculate by Reinhard color transfer algorithm the color information that strengthens image, read the color information data of reference picture simultaneously;
4) move to and strengthen on image with reference to the color information of image;
5) obtain the image that rich color and details are outstanding.
The optimal scale calculation of parameter of described Gaussian function is as follows:
11) weak light image is carried out to single scale Retinex algorithm process;
12) scale parameter of Gaussian function in change single scale Retinex algorithm;
13) calculate the information entropy of each image;
14) ask the scale parameter of the Gaussian function that the image of information entropy maximum is corresponding;
15) using the scale parameter of the Gaussian function of trying to achieve as optimal scale parameter.
The color information of described reference picture is calculated as follows:
21) get the reference picture of a width same scene intense light irradiation image as color transfer;
22) by the color information of color transfer algorithm computing reference image;
23) the color information data of preservation reference picture.
Described Gaussian function G (x, y) is specific as follows:
C is scale parameter; λ is normalized factor, and x and y are respectively the transverse and longitudinal coordinate figure of image.
Described single scale Retinex algorithm is specially:
I (x, y) is original image, and R (x, y) is reflected image, and L (x, y) is luminance picture.
Described information entropy is calculated as follows:
Wherein p
ithe frequency that i gray level of presentation video occurs.
Compared with prior art, the present invention has the following advantages:
1, figure image intensifying is effective, asks for the optimal scale parameter of Gaussian function in single scale Retinex (SSR) by objective evaluation, now image information entropy maximum, and figure image intensifying effect is best;
2, rich image color information, strengthens image after color reference image and SSR enhancing by Reinhard color transfer;
3, real-time is good, and in order to meet the requirement of video image real-time, in the present invention, the scale parameter of single scale Retinex (SSR) Gaussian function adopts the optimal value of trying to achieve; Color transfer reference picture is from Same Scene intense light irradiation on daytime image, and this image data after pre-service are directly called after preserving.
Brief description of the drawings
Fig. 1 is specific works flow process figure of the present invention;
Fig. 2 is the optimal scale calculation of parameter process flow diagram of Gaussian function of the present invention;
Fig. 3 is the color information calculation flow chart of reference picture of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.
Embodiment
As shown in Figure 1, a kind of Enhancement Method of the weak light image based on Retinex and Reinhard color transfer, is characterized in that, comprises the following steps:
1) read the optimal scale parameter of Gaussian function;
2) by single scale Retinex algorithm, weak light image is strengthened;
3) calculate by Reinhard color transfer algorithm the color information that strengthens image, read the color information data of reference picture simultaneously;
4) move to and strengthen on image with reference to the color information of image;
5) obtain the image that rich color and details are outstanding.
As shown in Figure 2, the optimal scale calculation of parameter of described Gaussian function is as follows:
11) weak light image is carried out to single scale Retinex algorithm process;
12) scale parameter of Gaussian function in change single scale Retinex algorithm;
13) calculate the information entropy of each image;
14) ask the scale parameter of the Gaussian function that the image of information entropy maximum is corresponding;
15) using the scale parameter of the Gaussian function of trying to achieve as optimal scale parameter.
As shown in Figure 3, the color information of described reference picture is calculated as follows:
21) get the reference picture of a width same scene intense light irradiation image as color transfer;
22) by the color information of color transfer algorithm computing reference image;
23) the color information data of preservation reference picture;
In Retinex theory, image is made up of incident light and reflecting object two parts, and the image representation of its formation is
I(x,y)=R(x,y)×L(x,y)
Wherein, I (x, y) is the coloured image that observer or camera are observed, and L (x, y) is ambient brightness, and R (x, y) is the reflectivity properties of object.Object reflected light R (x, y) has determined the inwardness of image.
Gaussian function can better obtain luminance picture from original image, and this Gaussian function G (x, y) is specific as follows:
C is scale parameter; λ is normalized factor, and x and y are respectively the transverse and longitudinal coordinate figure of image.
Gaussian convolution function only has a variable element, and scale parameter c has determined the treatment effect of single scale Retinex (SSR).Hour, dynamic range compression is larger for c, and image detail is outstanding, but the loss of overall illumination, image has cross-color phenomenon; When c is larger, integral image is effective, and color fidelity is good, but dynamic range compression is less, and local detail is unintelligible.
Described single scale Retinex algorithm can be expressed as in log-domain:
I (x, y) is original image, and R (x, y) is reflected image, and L (x, y) is luminance picture.G (x, y) is Gaussian function, and luminance picture is L (x, y)=I (x, y) * G (x, y);
In SSR algorithm, convolution item is the calculating to space illumination, and physical significance is to calculate current pixel value and the ratio of the weighted mean value of neighborhood territory pixel around, thus the impact of elimination illumination change.At log space, original image deducts convolution item and is equivalent to original image and deducts low frequency part, and residue is original image HFS, has given prominence to like this edge details in original image.Single scale Retinex algorithm energization has enough been given prominence to the edge details of dark areas.
Adopt information entropy to carry out objective evaluation to image.Information entropy is the tolerance to the contained quantity of information of image.Information entropy is larger, and the complexity of presentation video texture is higher, and the contained information of image is more.
Described information entropy is calculated as follows:
Wherein p
ithe frequency that i gray level of presentation video occurs.In the time that all gray level probabilities of occurrence equate, the information entropy maximum of image; In the time that image is monochrome, the information entropy of image is zero.
Average is the average gray value of pixel, reflection image overall chiaroscuro effect, average is less, image is darker.Gray standard deviation has reflected the discrete situation of relative gray average, and standard deviation is larger, and intensity profile is overstepping the bounds of propriety loose, and picture quality is better.One width size is the image of m × n, and its brightness average and standard deviation formula are
In formula, g (i, j) is the pixel value of coordinate (i, j).
Through contrast, when entropy is maximum, average and the standard deviation of image are placed in the middle, and the observing effect of image is better.
Color transfer is divided into the color transfer between image.Adopt Reinhard color transfer to obtain good effect.
Reinhard algorithm is added up average and the standard deviation of two each Color Channels of width image at l α β color space, the wherein integral color information of average represent images, standard is the minutia of represent images, finally moves to target image with reference to the color distribution of image.
Calculate average and the standard deviation of color transfer image reference image.
Calculate average and standard deviation that single scale Retinex strengthens rear image.Average and the standard deviation of color transfer image reference image are dwindled, and ensure that the color information of image is undistorted.
The Global Information of reference picture is delivered in target image, and the average of target image adds the average of reference picture respective channel, obtains composograph.
Claims (6)
1. an Enhancement Method for the weak light image based on Retinex and Reinhard color transfer, is characterized in that, comprises the following steps:
1) read the optimal scale parameter of Gaussian function;
2) by single scale Retinex algorithm, weak light image is strengthened;
3) calculate by Reinhard color transfer algorithm the color information that strengthens image, read the color information data of reference picture simultaneously;
4) move to and strengthen on image with reference to the color information of image;
5) obtain the image that rich color and details are outstanding.
2. the Enhancement Method of a kind of weak light image based on Retinex and Reinhard color transfer according to claim 1, is characterized in that, the optimal scale calculation of parameter of described Gaussian function is as follows:
11) weak light image is carried out to single scale Retinex algorithm process;
12) scale parameter of Gaussian function in change single scale Retinex algorithm;
13) calculate the information entropy of each image;
14) ask the scale parameter of the Gaussian function that the image of information entropy maximum is corresponding;
15) using the scale parameter of the Gaussian function of trying to achieve as optimal scale parameter.
3. the Enhancement Method of a kind of weak light image based on Retinex and Reinhard color transfer according to claim 1, is characterized in that, the color information of described reference picture is calculated as follows:
21) get the reference picture of a width same scene intense light irradiation image as color transfer;
22) by the color information of color transfer algorithm computing reference image;
23) the color information data of preservation reference picture.
4. the Enhancement Method of a kind of weak light image based on Retinex and Reinhard color transfer according to claim 2, is characterized in that, described Gaussian function G (x, y) is specific as follows:
C is scale parameter; λ is normalized factor, and x and y are respectively the transverse and longitudinal coordinate figure of image.
5. the Enhancement Method of a kind of weak light image based on Retinex and Reinhard color transfer according to claim 4, is characterized in that, described single scale Retinex algorithm is specially:
I (x, y) is original image, and R (x, y) is reflected image, and L (x, y) is luminance picture.
6. the Enhancement Method of a kind of weak light image based on Retinex and Reinhard color transfer according to claim 5, is characterized in that, described information entropy is calculated as follows:
Wherein p
ithe frequency that i gray level of presentation video occurs.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410192799.9A CN103955902A (en) | 2014-05-08 | 2014-05-08 | Weak light image enhancing method based on Retinex and Reinhard color migration |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410192799.9A CN103955902A (en) | 2014-05-08 | 2014-05-08 | Weak light image enhancing method based on Retinex and Reinhard color migration |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103955902A true CN103955902A (en) | 2014-07-30 |
Family
ID=51333171
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410192799.9A Pending CN103955902A (en) | 2014-05-08 | 2014-05-08 | Weak light image enhancing method based on Retinex and Reinhard color migration |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103955902A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107038699A (en) * | 2016-11-09 | 2017-08-11 | 重庆医科大学 | Strengthen image fault rate detection method |
CN107566821A (en) * | 2017-08-27 | 2018-01-09 | 南京理工大学 | A kind of image color moving method based on multi-dimensional association rule |
CN108024062A (en) * | 2017-12-13 | 2018-05-11 | 联想(北京)有限公司 | Image processing method and image processing apparatus |
CN109712093A (en) * | 2018-12-21 | 2019-05-03 | 中国电子科技集团公司第三研究所 | A kind of color of image restoring method and device based on sky and ocean background |
CN109872331A (en) * | 2019-01-30 | 2019-06-11 | 天津大学 | A kind of remote sensing image data automatic recognition classification method based on deep learning |
CN110533741A (en) * | 2019-08-08 | 2019-12-03 | 天津工业大学 | A kind of camouflage pattern design method rapidly adapting to battlefield variation |
CN110647854A (en) * | 2019-09-27 | 2020-01-03 | 华清永安(北京)科技发展有限责任公司 | Intelligent management system for classified discharge of garbage |
CN111127377A (en) * | 2019-12-20 | 2020-05-08 | 湖北工业大学 | Weak light enhancement method based on multi-image fusion Retinex |
CN112967194A (en) * | 2021-03-04 | 2021-06-15 | Oppo广东移动通信有限公司 | Target image generation method and device, computer readable medium and electronic equipment |
CN113128433A (en) * | 2021-04-26 | 2021-07-16 | 刘秀萍 | Video monitoring image enhancement method of color migration matching characteristics |
CN113344804A (en) * | 2021-05-11 | 2021-09-03 | 湖北工业大学 | Training method of low-light image enhancement model and low-light image enhancement method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100253852A1 (en) * | 2009-04-07 | 2010-10-07 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer program |
CN103606134A (en) * | 2013-11-26 | 2014-02-26 | 国网上海市电力公司 | Enhancing method of low-light video images |
-
2014
- 2014-05-08 CN CN201410192799.9A patent/CN103955902A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100253852A1 (en) * | 2009-04-07 | 2010-10-07 | Canon Kabushiki Kaisha | Image processing apparatus, image processing method, and computer program |
CN103606134A (en) * | 2013-11-26 | 2014-02-26 | 国网上海市电力公司 | Enhancing method of low-light video images |
Non-Patent Citations (1)
Title |
---|
赵晓霞: "基于Retinex理论的视频图像增强***研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107038699A (en) * | 2016-11-09 | 2017-08-11 | 重庆医科大学 | Strengthen image fault rate detection method |
CN107038699B (en) * | 2016-11-09 | 2019-07-23 | 重庆医科大学 | Enhance image fault rate detection method |
CN107566821A (en) * | 2017-08-27 | 2018-01-09 | 南京理工大学 | A kind of image color moving method based on multi-dimensional association rule |
CN108024062A (en) * | 2017-12-13 | 2018-05-11 | 联想(北京)有限公司 | Image processing method and image processing apparatus |
CN109712093A (en) * | 2018-12-21 | 2019-05-03 | 中国电子科技集团公司第三研究所 | A kind of color of image restoring method and device based on sky and ocean background |
CN109872331A (en) * | 2019-01-30 | 2019-06-11 | 天津大学 | A kind of remote sensing image data automatic recognition classification method based on deep learning |
CN110533741A (en) * | 2019-08-08 | 2019-12-03 | 天津工业大学 | A kind of camouflage pattern design method rapidly adapting to battlefield variation |
CN110647854A (en) * | 2019-09-27 | 2020-01-03 | 华清永安(北京)科技发展有限责任公司 | Intelligent management system for classified discharge of garbage |
CN110647854B (en) * | 2019-09-27 | 2020-07-28 | 华清永安(北京)科技发展有限责任公司 | Intelligent management system for classified discharge of garbage |
CN111127377A (en) * | 2019-12-20 | 2020-05-08 | 湖北工业大学 | Weak light enhancement method based on multi-image fusion Retinex |
CN111127377B (en) * | 2019-12-20 | 2023-04-25 | 湖北工业大学 | Weak light enhancement method based on multi-image fusion Retinex |
CN112967194A (en) * | 2021-03-04 | 2021-06-15 | Oppo广东移动通信有限公司 | Target image generation method and device, computer readable medium and electronic equipment |
CN112967194B (en) * | 2021-03-04 | 2024-05-14 | Oppo广东移动通信有限公司 | Target image generation method and device, computer readable medium and electronic equipment |
CN113128433A (en) * | 2021-04-26 | 2021-07-16 | 刘秀萍 | Video monitoring image enhancement method of color migration matching characteristics |
CN113344804A (en) * | 2021-05-11 | 2021-09-03 | 湖北工业大学 | Training method of low-light image enhancement model and low-light image enhancement method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103955902A (en) | Weak light image enhancing method based on Retinex and Reinhard color migration | |
CN103606134A (en) | Enhancing method of low-light video images | |
Gao et al. | Naturalness preserved nonuniform illumination estimation for image enhancement based on retinex | |
CN101783012B (en) | Automatic image defogging method based on dark primary colour | |
Li et al. | Single image haze removal using content‐adaptive dark channel and post enhancement | |
CN103020920B (en) | Method for enhancing low-illumination images | |
CN110148093B (en) | Image defogging improvement method based on dark channel prior | |
Jang et al. | Colour image dehazing using near‐infrared fusion | |
Paul et al. | Histogram modification in adaptive bi-histogram equalization for contrast enhancement on digital images | |
Gao et al. | Single fog image restoration with multi-focus image fusion | |
Gu et al. | A Low‐Light Image Enhancement Method Based on Image Degradation Model and Pure Pixel Ratio Prior | |
Shi et al. | A joint deep neural networks-based method for single nighttime rainy image enhancement | |
Xue et al. | Video image dehazing algorithm based on multi-scale retinex with color restoration | |
Fu et al. | Multi-feature-based bilinear CNN for single image dehazing | |
Tufail et al. | Optimisation of transmission map for improved image defogging | |
Dwivedi et al. | Single image dehazing using extended local dark channel prior | |
Gan et al. | Multilevel image dehazing algorithm using conditional generative adversarial networks | |
Cui et al. | An improved dark channel defogging algorithm based on the HSI colour space | |
Si et al. | A novel method for single nighttime image haze removal based on gray space | |
Fang et al. | An Improved DCP‐Based Image Defogging Algorithm Combined with Adaptive Fusion Strategy | |
Ding et al. | Traffic image dehazing based on HSV color space | |
Baiju et al. | l 1/2 regularized joint low rank and sparse recovery technique for illumination map estimation in low light image enhancement | |
Jiang et al. | Gray‐Scale Image Dehazing Guided by Scene Depth Information | |
Yong et al. | Image enhancement algorithm research based on the archives monitoring under low illumination | |
Gao et al. | Single image dehazing based on single pixel energy minimization |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140730 |
|
RJ01 | Rejection of invention patent application after publication |