CN111028184A - Image enhancement method and system - Google Patents
Image enhancement method and system Download PDFInfo
- Publication number
- CN111028184A CN111028184A CN202010155066.3A CN202010155066A CN111028184A CN 111028184 A CN111028184 A CN 111028184A CN 202010155066 A CN202010155066 A CN 202010155066A CN 111028184 A CN111028184 A CN 111028184A
- Authority
- CN
- China
- Prior art keywords
- image
- low
- wavelet
- defogged
- frequency sub
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 43
- 230000004927 fusion Effects 0.000 claims abstract description 21
- 238000002834 transmittance Methods 0.000 claims description 18
- 239000000126 substance Substances 0.000 claims description 6
- 230000000694 effects Effects 0.000 abstract description 20
- 230000008859 change Effects 0.000 abstract description 12
- 230000008569 process Effects 0.000 description 11
- 230000009466 transformation Effects 0.000 description 10
- 230000000007 visual effect Effects 0.000 description 8
- 239000000284 extract Substances 0.000 description 5
- 238000001914 filtration Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000006467 substitution reaction 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
- G06T5/73—Deblurring; Sharpening
-
- 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
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image enhancement method and system, wherein the method comprises the following steps: s1, carrying out wavelet decomposition on the collected fog-containing image to obtain a low-frequency component; s2, performing wavelet reconstruction on the low-frequency component to obtain a low-frequency sub-image; s3, estimating an atmospheric dissipation function, an atmospheric light value and medium transmissivity based on the low-frequency sub-image; s4, generating a defogging low-frequency sub-image based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity; s5, obtaining a first defogged image containing the defogged image by adopting a dark channel prior defogging method; s6, carrying out wavelet decomposition on the first defogged image to obtain a high-frequency component; s7, performing wavelet reconstruction on the high-frequency component to obtain a high-frequency sub-image corresponding to the first defogged image; and S8, performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image. The second defogged image generated by the invention can simultaneously realize detail enhancement, and the brightness change effect of the image is natural.
Description
Technical Field
The invention relates to the field of image processing, in particular to an image enhancement method and system.
Background
Unmanned aerial vehicle aerial photography is widely applied to many outdoor vision systems such as traffic monitoring and environmental monitoring. Because the weather environment in general mountain area is comparatively abominable, fog environment possibility is very high, and the picture of directly shooting from unmanned aerial vehicle is comparatively unclear, has image noise.
The invention patent application with publication number CN 108460743 a discloses an unmanned aerial vehicle aerial image defogging algorithm based on a dark channel, which obtains a dark channel image according to a color foggy image, obtains a global atmospheric light component a from the dark channel image, obtains a coarse transmittance map according to the dark channel image, down-samples the coarse transmittance map, guides filtering, up-samples the coarse transmittance map to obtain a fine transmittance map, obtains an initial defogged image according to the fine transmittance map and the global atmospheric light component a, and finally performs local contrast enhancement on the initial defogged image by using a contrast-limiting adaptive histogram equalization, so that the problem that the image is darker after defogging in the dark channel is solved, and local information of an original image is highlighted for subsequent image processing work.
The image is defogged by adopting a dark primary color prior method, the influence of fog is essentially removed, and the detail information of the image is enhanced, but the brightness of a target scene is usually not as good as that of atmospheric light, so that the overall restored image is darker and the visual effect is poor. The application improves the image brightness through local contrast enhancement, and the effect is relatively harsh. In order to retain image edge and detail information and enable the processed image effect to be more obvious, the prior art provides an image defogging method combining wavelet transformation and an atmospheric dissipation function by utilizing the conclusion that fog mainly affects the low-frequency region of an image. Firstly, performing first-level wavelet decomposition on a fog-containing image by utilizing the characteristic that fog mainly affects a low-frequency region of the image, and performing defogging treatment on a low-frequency sub-image; and then, taking the defogged image as a new low-frequency sub-image, and performing wavelet fusion with the high-frequency sub-image to obtain a final defogged image. The method can well eliminate the white edge phenomenon of the field depth mutation area and has a good processing effect. However, the method does not process the high-frequency area, so that the detail information of the image is not enhanced, and the detail processing effect is poor.
Therefore, how to implement detail enhancement and natural-effect image enhancement processing is a problem to be solved in the field.
Disclosure of Invention
The invention aims to provide an image enhancement method and system aiming at the defects of the prior art. The method effectively solves the problems that the restored image is dark in whole and poor in visual effect by the existing dark primary color prior method, and the detail information of the image defogging method of wavelet transformation and atmospheric dissipation function is not enhanced and the detail processing effect is poor, and the generated defogged image can simultaneously realize detail enhancement and the brightness change effect of the image is natural.
In order to achieve the purpose, the invention adopts the following technical scheme:
an image enhancement method, comprising:
s1, carrying out wavelet decomposition on the collected fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
s2, performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
s3, estimating an atmospheric dissipation function based on the low-frequency sub-image, and estimating an atmospheric light value and medium transmittance based on the atmospheric dissipation function;
s4, generating a defogging low-frequency sub-image based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
s5, obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
s6, carrying out wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
s7, performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
and S8, performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
wherein the content of the first and second substances,is a low-frequency sub-image corresponding to the fog-containing image,in order to be the medium transmittance,for the atmospheric dissipation function, A is the atmospheric light value and x is a certain image pixel.
Further, the basis functions of the wavelet decomposition include a Haar wavelet, a Daubechies wavelet, a Symlets wavelet.
Further, the atmospheric light value is an average of the first 0.2% brightness values of the atmospheric dissipation function.
Further, the wavelet fusion rules include a coefficient absolute value greater method, a weighted average method and a local variance criterion.
The invention also proposes an image enhancement system comprising:
the first wavelet decomposition module is used for performing wavelet decomposition on the acquired fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
the first wavelet reconstruction module is used for performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
the estimation module is used for estimating an atmospheric dissipation function based on the low-frequency sub-image and estimating an atmospheric light value and medium transmissivity based on the atmospheric dissipation function;
the first generation module is used for generating defogging low-frequency sub-images based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
the second generation module is used for obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
the second wavelet decomposition module is used for performing wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
the second wavelet reconstruction module is used for performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
and the fusion module is used for performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
Further, the first generating module comprises: the defogging low-frequency subimageIs calculated as:
wherein the content of the first and second substances,is a low-frequency sub-image corresponding to the fog-containing image,in order to be the medium transmittance,for the atmospheric dissipation function, A is the atmospheric light value and x is a certain image pixel.
Further, the basis functions of the wavelet decomposition include a Haar wavelet, a Daubechies wavelet, a Symlets wavelet.
Further, the atmospheric light value is an average of the first 0.2% brightness values of the atmospheric dissipation function.
Further, the wavelet fusion rules include a coefficient absolute value greater method, a weighted average method and a local variance criterion.
The invention provides an image enhancement method and system, which respectively generate a low-frequency sub-image through wavelet transformation and an atmospheric dissipation function, extract a high-frequency sub-image through a dark primary color prior method, perform wavelet fusion on the low-frequency sub-image and the high-frequency sub-image, and generate a second defogged image. Because the low-frequency sub-image stores the brightness change characteristic of the image and the high-frequency sub-image is subjected to information enhancement, the second defogged image generated by the invention can simultaneously realize detail enhancement and has natural brightness change effect. The invention effectively solves the problems that the restored image by the prior dark channel prior method is dark as a whole and has poor visual effect, and the image defogging method of wavelet transformation and atmospheric dissipation function has no enhanced detail information and poor detail processing effect. In the atmosphere dissipation defogging process, the characteristic that fog mainly influences the low-frequency part of an image is fully utilized, the low-frequency component of the image containing fog is processed, the image defogging effect is ensured, the data processing capacity is reduced, and the processing efficiency is improved. Meanwhile, the maximum brightness value is not directly selected as the atmospheric light value, and the average value of the first 0.2% brightness value of the atmospheric dissipation function is selected as the atmospheric light value A, so that the influence of highlight noise is effectively avoided.
Drawings
FIG. 1 is a flowchart of an image enhancement method according to an embodiment;
fig. 2 is a structural diagram of an image enhancement system according to a second embodiment.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
Example one
As shown in fig. 1, the present embodiment proposes an image enhancement method, including:
s1, carrying out wavelet decomposition on the collected fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
for images containing fog, the fog affects mainly the low frequency parts of the image. Therefore, the invention firstly carries out wavelet decomposition on the collected fog-containing image to obtain the low-frequency component corresponding to the fog-containing image, and processes the low-frequency component of the fog-containing image, thereby reducing the data processing amount and improving the processing efficiency while ensuring the image defogging effect. The wavelet decomposition process of the fog-containing image mainly carries out filtering processing on the horizontal direction and the vertical direction of the image, and then samples the image, thereby obtaining the result of image decomposition. The wavelet basis functions include haar (haar) wavelets, daubechies (dbn) wavelets, symlets (symn) wavelets, etc., and the present invention is not limited to specific wavelet decompositions.
S2, performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
the invention carries out defogging on the low-frequency part of the fog-containing image, so that after the low-frequency part of the fog-containing image is obtained through wavelet decomposition, the low-frequency part is processed through wavelet reconstruction, and a low-frequency sub-image corresponding to the fog-containing image containing the low-frequency part is reconstructed.
S3, estimating an atmospheric dissipation function based on the low-frequency sub-image, and estimating an atmospheric light value and medium transmittance based on the atmospheric dissipation function;
for the low-frequency sub-image containing fog, the optical model is as follows:
wherein the content of the first and second substances,is a low-frequency sub-image containing fog,is composed ofThe generated defogged image is displayed on the display screen,in order to be the medium transmittance,for the atmospheric dissipation function, A is the atmospheric light value and x is a certain image pixel.
Defogging the fog-containing image substantially byComputation generationTherefore, the invention firstly estimates the atmospheric dissipation function based on the low-frequency sub-image, and estimates the atmospheric light value and the medium transmissivity based on the atmospheric dissipation function.
Function of atmospheric dissipationThe estimation of (2) can adopt any existing estimation method, and is not limited herein. For the atmospheric light value, in the region with the most dense fog or the sky region in the atmospheric dissipation function graph, the brightness value of the atmospheric dissipation function is approximate to the estimated value of the atmospheric light value, so that the maximum brightness value of the atmospheric dissipation function is the atmospheric light value. Meanwhile, in order to avoid the influence of noise, the maximum brightness value is not directly selected, but the average of the first 0.2% brightness value is selectedThe average value is taken as the atmospheric light value A. At the same time, utilizeCalculating medium transmittance。
S4, generating a defogging low-frequency sub-image based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
therefore, after the atmospheric dissipation function, the atmospheric light value and the medium transmittance are estimated, the defogging low-frequency sub-image is generated according to the calculation of the company.
S5, obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
in a local area, mostly less non-sky, there are always some small blocks (at least one) of pixels, whose luminance is very low in one or several color channels, close to 0, called dark primaries. The influence of fog is essentially removed by dark channel prior defogging, and the image detail information is enhanced, so that the defogging low-frequency sub-image is generated by utilizing wavelet decomposition and an atmospheric dissipation function, and the first defogged image of the defogged image is obtained by adopting a dark channel prior defogging method, namely, all information in the defogged image is enhanced by the dark channel prior defogging method.
S6, carrying out wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
for the first defogged image, it includes enhanced image detail information. The high-frequency sub-images reflect detailed information such as edges and textures of the images, the low-frequency sub-images correspond to regions with slow brightness change in the images and reflect overall information of the images, and a common dark primary color prior method has the problems that the overall restored images are dark and the visual effect is poor. By processing the high-frequency component of the first defogged image, the enhanced detail information such as edges and textures is saved, and meanwhile, the influence of the dark whole restored image caused by the low-frequency component is avoided. The wavelet decomposition of the first defogged image may be the same process as the wavelet decomposition of the fog-containing image, or may be a different process, and is not limited herein.
S7, performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
the invention extracts the high-frequency part of the first defogged image, so that after the high-frequency part of the first defogged image is obtained through wavelet decomposition, the high-frequency part is processed through wavelet reconstruction, and a high-frequency sub-image corresponding to the first defogged image containing the high-frequency part is reconstructed.
And S8, performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
In order to solve the problems that the overall restored image is dark and poor in visual effect by the existing dark primary color prior method, and the details of the image defogging method of wavelet transformation and atmospheric dissipation function are not enhanced and poor in detail processing effect, the invention respectively generates low-frequency sub-images by the wavelet transformation and the atmospheric dissipation function, extracts high-frequency sub-images by the dark primary color prior method, and performs wavelet fusion on the low-frequency sub-images and the high-frequency sub-images to generate a second defogged image. Because the low-frequency sub-image stores the brightness change characteristic of the image and the high-frequency sub-image is subjected to information enhancement, the second defogged image generated by the invention can simultaneously realize detail enhancement and the brightness change effect of the image is natural.
The wavelet fusion rules include a method of relatively large absolute value of coefficient, a weighted average method, a local variance criterion, and the like.
Example two
As shown in fig. 2, the present embodiment proposes an image enhancement system including:
the first wavelet decomposition module is used for performing wavelet decomposition on the acquired fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
for images containing fog, the fog affects mainly the low frequency parts of the image. Therefore, the invention firstly carries out wavelet decomposition on the collected fog-containing image to obtain the low-frequency component corresponding to the fog-containing image, and processes the low-frequency component of the fog-containing image, thereby reducing the data processing amount and improving the processing efficiency while ensuring the image defogging effect. The wavelet decomposition process of the fog-containing image mainly carries out filtering processing on the horizontal direction and the vertical direction of the image, and then samples the image, thereby obtaining the result of image decomposition. The wavelet basis functions include haar (haar) wavelets, daubechies (dbn) wavelets, symlets (symn) wavelets, etc., and the present invention is not limited to specific wavelet decompositions.
The first wavelet reconstruction module is used for performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
the invention carries out defogging on the low-frequency part of the fog-containing image, so that after the low-frequency part of the fog-containing image is obtained through wavelet decomposition, the low-frequency part is processed through wavelet reconstruction, and a low-frequency sub-image corresponding to the fog-containing image containing the low-frequency part is reconstructed.
The estimation module is used for estimating an atmospheric dissipation function based on the low-frequency sub-image and estimating an atmospheric light value and medium transmissivity based on the atmospheric dissipation function;
for the low-frequency sub-image containing fog, the optical model is as follows:
wherein the content of the first and second substances,is a low-frequency sub-image containing fog,is composed ofThe generated defogged image is displayed on the display screen,in order to be the medium transmittance,for the atmospheric dissipation function, A is the atmospheric light value and x is a certain image pixel.
Defogging the fog-containing image substantially byComputation generationTherefore, the invention firstly estimates the atmospheric dissipation function based on the low-frequency sub-image, and estimates the atmospheric light value and the medium transmissivity based on the atmospheric dissipation function.
Function of atmospheric dissipationThe estimation of (2) can adopt any existing estimation method, and is not limited herein. For the atmospheric light value, in the region with the most dense fog or the sky region in the atmospheric dissipation function graph, the brightness value of the atmospheric dissipation function is approximate to the estimated value of the atmospheric light value, so that the maximum brightness value of the atmospheric dissipation function is the atmospheric light value. Meanwhile, in order to avoid the influence of noise, the present invention does not directly select the maximum luminance value, but selects the average value of the first 0.2% luminance values as the atmospheric light value a. At the same time, utilizeCalculating medium transmittance。
The first generation module is used for generating defogging low-frequency sub-images based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
therefore, after the atmospheric dissipation function, the atmospheric light value and the medium transmittance are estimated, the defogging low-frequency sub-image is generated according to the calculation of the company.
The second generation module is used for obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
in a local area, mostly less non-sky, there are always some small blocks (at least one) of pixels, whose luminance is very low in one or several color channels, close to 0, called dark primaries. The influence of fog is essentially removed by dark channel prior defogging, and the image detail information is enhanced, so that the defogging low-frequency sub-image is generated by utilizing wavelet decomposition and an atmospheric dissipation function, and the first defogged image of the defogged image is obtained by adopting a dark channel prior defogging method, namely, all information in the defogged image is enhanced by the dark channel prior defogging method.
The second wavelet decomposition module is used for performing wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
for the first defogged image, it includes enhanced image detail information. The high-frequency sub-images reflect detailed information such as edges and textures of the images, the low-frequency sub-images correspond to regions with slow brightness change in the images and reflect overall information of the images, and a common dark primary color prior method has the problems that the overall restored images are dark and the visual effect is poor. By processing the high-frequency component of the first defogged image, the enhanced detail information such as edges and textures is saved, and meanwhile, the influence of the dark whole restored image caused by the low-frequency component is avoided. The wavelet decomposition of the first defogged image may be the same process as the wavelet decomposition of the fog-containing image, or may be a different process, and is not limited herein.
The second wavelet reconstruction module is used for performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
the invention extracts the high-frequency part of the first defogged image, so that after the high-frequency part of the first defogged image is obtained through wavelet decomposition, the high-frequency part is processed through wavelet reconstruction, and a high-frequency sub-image corresponding to the first defogged image containing the high-frequency part is reconstructed.
And the fusion module is used for performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
In order to solve the problems that the overall restored image is dark and poor in visual effect by the existing dark primary color prior method, and the details of the image defogging method of wavelet transformation and atmospheric dissipation function are not enhanced and poor in detail processing effect, the invention respectively generates low-frequency sub-images by the wavelet transformation and the atmospheric dissipation function, extracts high-frequency sub-images by the dark primary color prior method, and performs wavelet fusion on the low-frequency sub-images and the high-frequency sub-images to generate a second defogged image. Because the low-frequency sub-image stores the brightness change characteristic of the image and the high-frequency sub-image is subjected to information enhancement, the second defogged image generated by the invention can simultaneously realize detail enhancement and the brightness change effect of the image is natural.
The wavelet fusion rules include a method of relatively large absolute value of coefficient, a weighted average method, a local variance criterion, and the like.
Therefore, according to the image enhancement method and the image enhancement system provided by the invention, the low-frequency sub-image is generated through wavelet transformation and an atmospheric dissipation function, the high-frequency sub-image is extracted through a dark primary color prior method, and the low-frequency sub-image and the high-frequency sub-image are subjected to wavelet fusion to generate the second defogged image. Because the low-frequency sub-image stores the brightness change characteristic of the image and the high-frequency sub-image is subjected to information enhancement, the second defogged image generated by the invention can simultaneously realize detail enhancement and has natural brightness change effect. The invention effectively solves the problems that the restored image by the prior dark channel prior method is dark as a whole and has poor visual effect, and the image defogging method of wavelet transformation and atmospheric dissipation function has no enhanced detail information and poor detail processing effect. In the atmosphere dissipation defogging process, the characteristic that fog mainly influences the low-frequency part of an image is fully utilized, the low-frequency component of the image containing fog is processed, the image defogging effect is ensured, the data processing capacity is reduced, and the processing efficiency is improved. Meanwhile, the maximum brightness value is not directly selected as the atmospheric light value, and the average value of the first 0.2% brightness value of the atmospheric dissipation function is selected as the atmospheric light value A, so that the influence of highlight noise is effectively avoided.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. An image enhancement method, comprising:
s1, carrying out wavelet decomposition on the collected fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
s2, performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
s3, estimating an atmospheric dissipation function based on the low-frequency sub-image, and estimating an atmospheric light value and medium transmittance based on the atmospheric dissipation function;
s4, generating a defogging low-frequency sub-image based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
s5, obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
s6, carrying out wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
s7, performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
and S8, performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
2. The image enhancement method according to claim 1, wherein the step S4 includes: the defogging low-frequency subimageIs calculated as:
3. The image enhancement method of claim 1 wherein the basis functions of the wavelet decomposition comprise a Haar wavelet, a Daubechies wavelet, a Symlets wavelet.
4. The image enhancement method of claim 1 wherein the atmospheric light value is an average of the top 0.2% brightness values of the atmospheric dissipation function.
5. The image enhancement method according to claim 1, wherein the rules of wavelet fusion include a larger coefficient absolute value method, a weighted average method, a local variance criterion.
6. An image enhancement system, comprising:
the first wavelet decomposition module is used for performing wavelet decomposition on the acquired fog-containing image to obtain a low-frequency component corresponding to the fog-containing image;
the first wavelet reconstruction module is used for performing wavelet reconstruction on the low-frequency component corresponding to the fog-containing image to obtain a low-frequency sub-image corresponding to the fog-containing image;
the estimation module is used for estimating an atmospheric dissipation function based on the low-frequency sub-image and estimating an atmospheric light value and medium transmissivity based on the atmospheric dissipation function;
the first generation module is used for generating defogging low-frequency sub-images based on the atmospheric dissipation function, the atmospheric light value and the medium transmissivity;
the second generation module is used for obtaining a first defogged image of the defogged image by adopting a dark channel prior defogging method;
the second wavelet decomposition module is used for performing wavelet decomposition on the first defogged image to obtain a high-frequency component corresponding to the first defogged image;
the second wavelet reconstruction module is used for performing wavelet reconstruction on the high-frequency component corresponding to the first defogged image to obtain a high-frequency sub-image corresponding to the first defogged image;
and the fusion module is used for performing wavelet fusion on the low-frequency sub-image and the high-frequency sub-image to generate a second defogged image.
7. The image enhancement system of claim 6, wherein the first generation module comprises: the defogging low-frequency subimageIs calculated as:
8. The image enhancement system of claim 6 wherein the basis functions of the wavelet decomposition include a Haar wavelet, a Daubechies wavelet, a Symlets wavelet.
9. The image enhancement system of claim 6 wherein the atmospheric light value is an average of the top 0.2% brightness values of the atmospheric dissipation function.
10. The image enhancement system of claim 6 wherein the rules of wavelet fusion include a larger coefficient absolute value method, a weighted average method, a local variance criterion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010155066.3A CN111028184B (en) | 2020-03-09 | 2020-03-09 | Image enhancement method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010155066.3A CN111028184B (en) | 2020-03-09 | 2020-03-09 | Image enhancement method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111028184A true CN111028184A (en) | 2020-04-17 |
CN111028184B CN111028184B (en) | 2020-06-12 |
Family
ID=70199432
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010155066.3A Active CN111028184B (en) | 2020-03-09 | 2020-03-09 | Image enhancement method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111028184B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113379632A (en) * | 2021-06-15 | 2021-09-10 | 深圳市赛蓝科技有限公司 | Image defogging method and system based on wavelet transmissivity optimization |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049888A (en) * | 2012-12-07 | 2013-04-17 | 西安电子科技大学 | Image/video demisting method based on combination of dark primary color of atmospheric scattered light |
-
2020
- 2020-03-09 CN CN202010155066.3A patent/CN111028184B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049888A (en) * | 2012-12-07 | 2013-04-17 | 西安电子科技大学 | Image/video demisting method based on combination of dark primary color of atmospheric scattered light |
Non-Patent Citations (1)
Title |
---|
周慧娟: "无人机图像去雾算法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113379632A (en) * | 2021-06-15 | 2021-09-10 | 深圳市赛蓝科技有限公司 | Image defogging method and system based on wavelet transmissivity optimization |
Also Published As
Publication number | Publication date |
---|---|
CN111028184B (en) | 2020-06-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110148095B (en) | Underwater image enhancement method and enhancement device | |
US9361670B2 (en) | Method and system for image haze removal based on hybrid dark channel prior | |
CN108734670B (en) | Method for restoring single night weak-illumination haze image | |
EP2568438B1 (en) | Image defogging method and system | |
CN103020920B (en) | Method for enhancing low-illumination images | |
CN104021532B (en) | A kind of image detail enhancement method of infrared image | |
WO2016206087A1 (en) | Low-illumination image processing method and device | |
CN106530257A (en) | Remote sensing image de-fogging method based on dark channel prior model | |
US20140140619A1 (en) | Method and System for Removal of Fog, Mist, or Haze from Images and Videos | |
CN108564597B (en) | Video foreground object extraction method fusing Gaussian mixture model and H-S optical flow method | |
CN107292830B (en) | Low-illumination image enhancement and evaluation method | |
CN110544213A (en) | Image defogging method based on global and local feature fusion | |
CN110827218B (en) | Airborne image defogging method based on weighted correction of HSV (hue, saturation, value) transmissivity of image | |
CN109389569B (en) | Monitoring video real-time defogging method based on improved DehazeNet | |
CN108154492B (en) | A kind of image based on non-local mean filtering goes haze method | |
CN111353955A (en) | Image processing method, device, equipment and storage medium | |
CN116823628A (en) | Image processing method and image processing device | |
CN109345479B (en) | Real-time preprocessing method and storage medium for video monitoring data | |
CN108898561B (en) | Defogging method, server and system for foggy image containing sky area | |
CN111028184B (en) | Image enhancement method and system | |
Gadnayak et al. | A survey on image dehazing methods | |
CN110738624B (en) | Area-adaptive image defogging system and method | |
Yang et al. | A novel single image dehazing method | |
Negru et al. | Exponential image enhancement in daytime fog conditions | |
Adidela et al. | Single image and video Dehazing: A dark channel prior (DCP)-based approach |
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 |