CN108416742B - Sand and dust degraded image enhancement method based on color cast correction and information loss constraint - Google Patents
Sand and dust degraded image enhancement method based on color cast correction and information loss constraint Download PDFInfo
- Publication number
- CN108416742B CN108416742B CN201810065901.7A CN201810065901A CN108416742B CN 108416742 B CN108416742 B CN 108416742B CN 201810065901 A CN201810065901 A CN 201810065901A CN 108416742 B CN108416742 B CN 108416742B
- Authority
- CN
- China
- Prior art keywords
- image
- color
- value
- pixel point
- channel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012937 correction Methods 0.000 title claims abstract description 27
- 239000000428 dust Substances 0.000 title claims abstract description 25
- 239000004576 sand Substances 0.000 title claims abstract description 17
- 238000002834 transmittance Methods 0.000 claims description 35
- 238000001914 filtration Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 3
- 230000015556 catabolic process Effects 0.000 claims description 3
- 238000006731 degradation reaction Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000003321 amplification Effects 0.000 abstract description 4
- 238000003199 nucleic acid amplification method Methods 0.000 abstract description 4
- 230000002708 enhancing effect Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The invention relates to a sand dust degraded image enhancement method based on color cast correction and information loss constraint, which comprises 1) improving a color cast correction method based on a Gaussian model, and 2) improving a contrast enhancement method based on information loss constraint. The improved color cast correction method based on the Gaussian model effectively solves the problems of detail loss, dark image entirety, noise amplification, excessive stretching and the like after the original method is processed, effectively corrects the image color cast, keeps the brightness of the image and avoids the noise amplification and the detail information loss of the scenery.
Description
Technical Field
The invention relates to the technical field of enhancement of a sand dust degraded image with low contrast and yellow color under severe weather, in particular to the technical field of enhancement of a sand dust degraded image with low contrast and yellow color under the condition of sand storm weather, and specifically relates to a sand dust degraded image enhancement method based on color cast correction and information loss constraint.
Background
Most outdoor vision shooting systems, such as intelligent monitoring, autonomous navigation, vehicle tracking and monitoring systems, require clear visibility of input images. However, under severe sandstorm weather conditions, the contrast of the image is reduced, the color of the image is shifted, the quality of the image is seriously affected, and the functions of various visual systems cannot be normally performed. Therefore, the visibility enhancement method for researching the sand and dust degraded image removes the sand and dust weather influence from the image, expands the application range of the visual systems, better plays the role of the visual systems and has important practical significance. The visibility enhancement technology of the sand dust degraded image is also an important research subject in the fields of image video processing and computer vision, and the application prospect is very wide.
The technology for enhancing the sand-dust degraded image aims to solve the problems of color cast and low contrast in the sand-dust degraded image and improve the definition of the sand-dust degraded image. However, the enhancement result image of the existing sand-dust-degraded image enhancement method still has the problems of color cast, overall dark, insufficient or excessive contrast enhancement and the like. Therefore, the method for enhancing the sand-dust degraded image, which can effectively solve the problems of color cast and low contrast of the sand-dust degraded image, is a hot topic of image enhancement and pattern recognition academic research and is also a key research and development project of many computer vision field companies.
Disclosure of Invention
The invention aims to provide a sand-dust degraded image enhancement method based on color cast correction and information loss constraint.
The technical scheme adopted by the invention for solving the technical problems is as follows:
1) improved color cast correction method based on Gaussian model
The first step is as follows: calculating the color extension coefficient alpha according to the color distribution of the input sand-dust degradation image I
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue,
represents the maximum color value of the C channel in the input image I,
represents the minimum color value of the C channel in the input image I, and max () represents the maximum function.
The second step is that: calculating color cast correction image I 'according to input sand dust degraded image I'
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue, ICC-channel color value, l 'representing input image I'CC-channel color values, μ, representing a color-shift corrected image ICRepresenting the mean, σ, of the color values of the C channel in the input image ICRepresenting the standard deviation of the color values of the C channel in the input image I, and sigma representing the mean standard deviation of the color values of the input image I
α represents a color extension coefficient.
2) Improved contrast enhancement method based on information loss constraint
The first step is as follows: processing the color cast correction image I 'according to an image defogging method based on dark channel prior to obtain an initial transmittance map t'
Wherein: Ω (x) represents an image block of 15 × 15 pixels centered on a pixel point x in the color cast corrected image I ', and min () represents a minimum value function I'CAnd (y) represents the color value of a channel C of a pixel point y in the image block omega (x), C represents a color channel C epsilon { R, G, B }, R, G, B in the image respectively represent three color channels of red, green and blue, and t' (x) represents the initial transmittance value of the pixel point x.
The second step is that: initializing transmittance values within image blocks
Wherein: omega (x) represents an image block with a pixel point x as the center and the size of 15 x 15 pixels, xwRepresenting the pixels in the local block omega (x),
representing a pixel point xwThe transmittance value of (b), t' (x), represents the initial transmittance value of pixel point x.
The third step: computing a target local block
Wherein: x is the number ofwRepresents a pixel in an image block Ω (x), Ω (x) represents an image block of 15 × 15 pixels centered on the pixel x, I' (x)w) Pixel point x in image block omega (x) for representing color cast correction image IwThe color value of (a) of (b),
representing a pixel point xwThe value of the transmittance of (a) to (b),
representing target image blocks
Middle pixel point xwThe color value of (a).
The fourth step: counting target image blocks
Respectively calculate the information loss amount of
The number N of pixel points with the middle color value less than 01And the number N of pixels with color values greater than 2552Total number of pixel overflows N in image blockloss＝N1+N2。
The fifth step: setting a threshold ThIf 2, if Nloss>ThIt indicates that the information loss is too large,
not meeting our requirements, the step length delta t needs to be increased by 0.05
Namely, it is
And returning to the third step for recalculation if Nloss≤ThThen give an order
Wherein: t "(x) represents the optimized transmission value of pixel point x,
representing a pixel point xwThe transmittance value of (a).
And a sixth step: repeating the second step to the fifth step, and calculating the optimized transmittance value of each pixel point to obtain the optimized transmittance graph t ″
The seventh step: obtaining a refined transmittance graph t by using the transmittance graph t' after the guide filtering processing optimization, and obtaining a result image J by using an image enhancement method based on an atmospheric scattering model
Wherein: the step adopts a guide filtering algorithm, the input image is an optimized transmissivity image t ', the guide image is a color cast correction image I', and the filtering radius is 45 pixels. I '(x) represents a color value of a pixel x in the color cast corrected image I', t (x) represents a refined transmittance value of the pixel x, and J (x) represents a color value of the pixel x in the resulting image J.
Compared with the background technology, the invention has the beneficial effects that:
the method firstly uses an improved color cast correction method based on a Gaussian model to perform color cast correction on the sand dust degraded image, and then performs contrast enhancement on the color cast corrected image by an improved contrast enhancement method based on information loss constraint. The improved color cast correction method based on the Gaussian model effectively solves the problems of detail loss, dark image entirety, noise amplification, excessive stretching and the like after the original method is processed, effectively corrects the image color cast, keeps the brightness of the image and avoids the noise amplification and the detail information loss of the scenery. Compared with the problems of insufficient or excessive contrast enhancement, new color cast, uneven enhancement degree and the like existing after the treatment of the existing method. The invention can effectively improve the overall contrast and brightness of the image, prevent new color cast and keep the detail information of the image.
The invention is suitable for effectively enhancing the contrast of different types of dust and sand degraded images, can effectively avoid the phenomenon of image color cast, and keeps good color fidelity and proper brightness.
Drawings
Fig. 1 is a schematic structural diagram of an embodiment of the present invention.
Detailed Description
The invention relates to a sand dust degraded image enhancement method based on color cast correction and information loss constraint, which comprises the following specific implementation steps:
1) improved color cast correction method based on Gaussian model
The first step is as follows: calculating the color extension coefficient alpha according to the color distribution of the input sand-dust degradation image I
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue,
represents the maximum color value of the C channel in the input image I,
represents the minimum color value of the C channel in the input image I, and max () represents the maximum function.
The second step is that: calculating color cast correction image I 'according to input sand dust degraded image I'
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue, ICC-channel color value, l 'representing input image I'CC-channel color values, μ, representing a color-shift corrected image ICRepresenting the mean, σ, of the color values of the C channel in the input image ICRepresenting the standard deviation of the color values of the C channel in the input image I, and sigma representing the mean standard deviation of the color values of the input image I
α represents a color extension coefficient.
2) Improved contrast enhancement method based on information loss constraint
The first step is as follows: processing the color cast correction image I 'according to an image defogging method based on dark channel prior to obtain an initial transmittance map t'
Wherein: Ω (x) represents an image block of 15 × 15 pixels centered on a pixel point x in the color cast corrected image I ', and min () represents a minimum value function I'CAnd (y) represents the color value of a channel C of a pixel point y in the image block omega (x), C represents a color channel C epsilon { R, G, B }, R, G, B in the image respectively represent three color channels of red, green and blue, and t' (x) represents the initial transmittance value of the pixel point x.
The second step is that: initializing transmittance values within image blocks
Wherein: omega (x) represents an image block with a pixel point x as the center and the size of 15 x 15 pixels, xwRepresenting the pixels in the local block omega (x),
representing a pixel point xwThe transmittance value of (b), t' (x), represents the initial transmittance value of pixel point x.
The third step: computing a target local block
Wherein: x is the number ofwRepresents a pixel in an image block Ω (x), Ω (x) represents an image block of 15 × 15 pixels centered on the pixel x, I' (x)w) Pixel point x in image block omega (x) for representing color cast correction image IwThe color value of (a) of (b),
representing a pixel point xwThe value of the transmittance of (a) to (b),
representing target image blocks
Middle pixel point xwThe color value of (a).
The fourth step: counting target image blocks
Respectively calculate the information loss amount of
The number N of pixel points with the middle color value less than 01And the number N of pixels with color values greater than 2552Total number of pixel overflows N in image blockloss＝N1+N2。
The fifth step: setting a threshold ThIf 2, if Nloss>ThIt indicates that the information loss is too large,
not meeting our requirements, the step length delta t needs to be increased by 0.05
Namely, it is
And returning to the third step for recalculation if Nloss≤ThThen give an order
Wherein: t "(x) represents the optimized transmission value of pixel point x,
representing a pixel point xwThe transmittance value of (a).
And a sixth step: repeating the second step to the fifth step, and calculating the optimized transmittance value of each pixel point to obtain the optimized transmittance graph t ″
The seventh step: obtaining a refined transmittance graph t by using the transmittance graph t' after the guide filtering processing optimization, and obtaining a result image J by using an image enhancement method based on an atmospheric scattering model
Wherein: the step adopts a guide filtering algorithm, the input image is an optimized transmissivity image t ', the guide image is a color cast correction image I', and the filtering radius is 45 pixels. I '(x) represents a color value of a pixel x in the color cast corrected image I', t (x) represents a refined transmittance value of the pixel x, and J (x) represents a color value of the pixel x in the resulting image J.
Claims (1)
1. A sand and dust degraded image enhancement method based on color cast correction and information loss constraint comprises the following steps:
1) improved color cast correction method based on Gaussian distribution
The first step is as follows: calculating the color extension coefficient alpha according to the color distribution of the input sand-dust degradation image I
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue,
represents the maximum color value of the C channel in the input image I,
represents the minimum color value of the C channel in the input image I, max () represents the maximum function;
the second step is that: calculating color cast correction image I 'according to input sand dust degraded image I'
Wherein: c represents a color channel C epsilon { R, G, B } in the image, R, G, B respectively represent three color channels of red, green and blue, ICC-channel color value, l 'representing input image I'CC-channel color values, μ, representing a color-shift corrected image ICRepresenting the mean, σ, of the color values of the C channel in the input image ICObject for representing C channel color value in input image ITolerance, σ, represents the mean standard deviation of I-color values of an input image
α represents a color extension coefficient;
2) improved contrast enhancement method based on information loss constraint
The first step is as follows: processing the color cast correction image I 'according to an image defogging method based on dark channel prior to obtain an initial transmittance map t'
Wherein: Ω (x) represents an image block of 15 × 15 pixels centered on a pixel point x in the color cast corrected image I ', and min () represents a minimum value function I'C(y) represents the color value of a channel C of a pixel point y in an image block omega (x), C represents a color channel C belonging to { R, G, B }, R, G, B in the image respectively represent three color channels of red, green and blue, and t' (x) represents the initial transmittance value of the pixel point x;
the second step is that: initializing transmittance values within image blocks
Wherein: omega (x) represents an image block with a pixel point x as the center and the size of 15 x 15 pixels, xwRepresenting the pixels in the local block omega (x),
representing a pixel point xwT' (x) represents the initial transmittance value of pixel point x;
the third step: computing a target local block
Wherein: x is the number ofwRepresents a pixel in an image block Ω (x), Ω (x) represents an image block of 15 × 15 pixels centered on the pixel x, I' (x)w) Pixel point x in image block omega (x) for representing color cast correction image IwThe color value of (a) of (b),
representing a pixel point xwThe value of the transmittance of (a) to (b),
representing target image blocks
Middle pixel point xwA color value of (a);
the fourth step: counting target image blocks
Respectively calculate the information loss amount of
The number N of pixel points with the middle color value less than 01And the number N of pixels with color values greater than 2552Total number of pixel overflows N in image blockloss＝N1+N2；
The fifth step: setting a threshold ThIf 2, if Nloss＞ThIt indicates that the information loss is too large,
not meeting our requirements, the step length delta t needs to be increased by 0.05
Namely, it is
And returning to the third step for recalculation if Nloss≤ThThen give an order
Wherein: t "(x) represents the optimized transmission value of pixel point x,
representing a pixel point xwA transmittance value of (a);
and a sixth step: repeating the second step to the fifth step, and calculating the optimized transmittance value of each pixel point to obtain an optimized transmittance graph t';
the seventh step: obtaining a refined transmittance graph t by using the transmittance graph t' after the guide filtering processing optimization, and obtaining a result image J by using an image enhancement method based on an atmospheric scattering model
Wherein: the step adopts a guide filtering algorithm, an input image is an optimized transmissivity image t ', a guide image is a color cast correction image I', and the filtering radius is 45 pixels; i '(x) represents a color value of a pixel x in the color cast corrected image I', t (x) represents a refined transmittance value of the pixel x, and J (x) represents a color value of the pixel x in the resulting image J.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810065901.7A CN108416742B (en) | 2018-01-23 | 2018-01-23 | Sand and dust degraded image enhancement method based on color cast correction and information loss constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810065901.7A CN108416742B (en) | 2018-01-23 | 2018-01-23 | Sand and dust degraded image enhancement method based on color cast correction and information loss constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108416742A CN108416742A (en) | 2018-08-17 |
| CN108416742B true CN108416742B (en) | 2021-09-17 |
Family
ID=63126296
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810065901.7A Active CN108416742B (en) | 2018-01-23 | 2018-01-23 | Sand and dust degraded image enhancement method based on color cast correction and information loss constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108416742B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111833270B (en) * | 2020-07-13 | 2023-02-10 | 新疆大学 | Rapid sand-dust degradation image enhancement method |
| CN112116542B (en) * | 2020-09-24 | 2024-03-08 | 西安宇视信息科技有限公司 | Image contrast enhancement method, device, electronic equipment and storage medium |
| CN114820385B (en) * | 2022-05-20 | 2023-06-23 | 河南科技学院 | Locally adaptive underwater image color correction method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8124328B2 (en) * | 2008-05-20 | 2012-02-28 | Eastman Kodak Company | Methods for imaging and processing positive-working imageable elements |
| CN103413305A (en) * | 2013-08-01 | 2013-11-27 | 北京航空航天大学 | Single-image rapid defogging method and device and image processing system |
| CN103996178A (en) * | 2014-05-30 | 2014-08-20 | 天津大学 | Sand and dust weather color image enhancing method |
| CN105118027A (en) * | 2015-07-28 | 2015-12-02 | 北京航空航天大学 | Image defogging method |
| CN105427266A (en) * | 2016-01-04 | 2016-03-23 | 西安理工大学 | Sand and dust image clearing method according to information loss restraint |
| CN105469372A (en) * | 2015-12-30 | 2016-04-06 | 广西师范大学 | Mean filtering-based fog-degraded image sharp processing method |
| CN105574830A (en) * | 2016-02-04 | 2016-05-11 | 沈阳工业大学 | Low-quality image enhancement method under extreme weather conditions |
| CN106485681A (en) * | 2016-10-18 | 2017-03-08 | 河海大学常州校区 | Color image restoration method under water based on color correction and red channel prior |
| CN106897981A (en) * | 2017-04-12 | 2017-06-27 | 湖南源信光电科技股份有限公司 | A kind of enhancement method of low-illumination image based on guiding filtering |
-
2018
- 2018-01-23 CN CN201810065901.7A patent/CN108416742B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8124328B2 (en) * | 2008-05-20 | 2012-02-28 | Eastman Kodak Company | Methods for imaging and processing positive-working imageable elements |
| CN103413305A (en) * | 2013-08-01 | 2013-11-27 | 北京航空航天大学 | Single-image rapid defogging method and device and image processing system |
| CN103996178A (en) * | 2014-05-30 | 2014-08-20 | 天津大学 | Sand and dust weather color image enhancing method |
| CN105118027A (en) * | 2015-07-28 | 2015-12-02 | 北京航空航天大学 | Image defogging method |
| CN105469372A (en) * | 2015-12-30 | 2016-04-06 | 广西师范大学 | Mean filtering-based fog-degraded image sharp processing method |
| CN105427266A (en) * | 2016-01-04 | 2016-03-23 | 西安理工大学 | Sand and dust image clearing method according to information loss restraint |
| CN105574830A (en) * | 2016-02-04 | 2016-05-11 | 沈阳工业大学 | Low-quality image enhancement method under extreme weather conditions |
| CN106485681A (en) * | 2016-10-18 | 2017-03-08 | 河海大学常州校区 | Color image restoration method under water based on color correction and red channel prior |
| CN106897981A (en) * | 2017-04-12 | 2017-06-27 | 湖南源信光电科技股份有限公司 | A kind of enhancement method of low-illumination image based on guiding filtering |
Non-Patent Citations (5)
| Title |
|---|
| Enhancement for Dust-Sand Storm Images;Jian Wang 等;《MMM 2016》;20161231;842-849 * |
| Shunyuan Yu 等.Single sand-dust image restoration using information loss constraint.《Journal of Modern optics》.2016,第63卷(第21期),2121-2130. * |
| 极端天气条件下低质图像增强算法研究;刘振宇 等;《计算机工程与应用》;20171231;第53卷(第8期);193-198、245 * |
| 沙尘天气下降质图像的清晰化方法研究;李海燕 等;《喀什师范学院报》;20121130;第33卷(第6期);45-48 * |
| 沙尘降质图像清晰化算法;智宁 等;《中国图象图形学报》;20161231;第21卷(第12期);1585-1592 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108416742A (en) | 2018-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107767354B (en) | Image defogging algorithm based on dark channel prior | |
| CN107301623B (en) | Traffic image defogging method and system based on dark channel and image segmentation | |
| CN110148095B (en) | Underwater image enhancement method and enhancement device | |
| CN106251301B (en) | A kind of single image to the fog method based on dark primary priori | |
| CN106157267B (en) | Image defogging transmissivity optimization method based on dark channel prior | |
| CN110782407B (en) | Single image defogging method based on sky region probability segmentation | |
| CN102063706B (en) | Rapid defogging method | |
| CN111292258B (en) | Image defogging method based on dark channel prior and bright channel prior | |
| CN104036466B (en) | A kind of video defogging method and system | |
| CN108416742B (en) | Sand and dust degraded image enhancement method based on color cast correction and information loss constraint | |
| CN109087254B (en) | Unmanned aerial vehicle aerial image haze sky and white area self-adaptive processing method | |
| CN103778900B (en) | A kind of image processing method and system | |
| CN111598814B (en) | Single image defogging method based on extreme scattering channel | |
| CN107527325A (en) | A kind of monocular underweater vision Enhancement Method preferential based on dark | |
| CN109118440B (en) | Single image defogging method based on transmissivity fusion and adaptive atmospheric light estimation | |
| CN111563852A (en) | Dark channel prior defogging method based on low-complexity MF | |
| CN114693548B (en) | Dark channel defogging method based on bright area detection | |
| CN110969584B (en) | Low-illumination image enhancement method | |
| CN104331867A (en) | Image defogging method and device and mobile terminal | |
| CN106709876B (en) | Optical remote sensing image defogging method based on dark image element principle | |
| CN111598800A (en) | Single image defogging method based on space domain homomorphic filtering and dark channel prior | |
| CN109934779B (en) | Defogging method based on guided filtering optimization | |
| CN107977941A (en) | A kind of bright areas color fidelity and the image defogging method of contrast enhancing | |
| CN107451962B (en) | Image defogging method and device | |
| CN106296626B (en) | A kind of night video enhancement method based on gradient fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |