CN105184761A - Image rain removing method based on wavelet analysis and system - Google Patents
Image rain removing method based on wavelet analysis and system Download PDFInfo
- Publication number
- CN105184761A CN105184761A CN201510541354.1A CN201510541354A CN105184761A CN 105184761 A CN105184761 A CN 105184761A CN 201510541354 A CN201510541354 A CN 201510541354A CN 105184761 A CN105184761 A CN 105184761A
- Authority
- CN
- China
- Prior art keywords
- image
- rain
- fusion
- wavelet
- raindrop
- 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 belongs to the image processing technology field, and particularly relates to an image rain removing method based on a wavelet analysis and a system. The image rain removing method based on wavelet analysis comprises steps of performing wavelet multilayer decomposition on a video frame image according to wavelet analysis, analyzing image information of all image layers and detecting the image layer containing rain droplet noise, c determining a rain droplet contamination degree coefficient according to the brightness characteristic of the rain droplet, calculating a fusion coefficient matrix according to the rain droplet contamination degree coefficient, performing wavelet fusion on the image layer containing various image information according to the fusion coefficient matrix, and performing image reconstruction through the fusion result to obtain the rain removing image. The invention can avoid the interference of the dynamic characteristic, can more effectively and accurately remove rain droplet, can improve the usage scope of the algorithm, enables the algorithm to have a sound rain-removing effect even under the heavy rain, performs defogging operation on the rain removing image, optimizes the visual effect of the image and enables the algorithm to be more practical.
Description
Technical field
The invention belongs to technical field of image processing, particularly relate to a kind of image based on wavelet analysis and go rain method and system.
Background technology
Rain has a great impact image imaging, and image image blur and information can be caused to cover, and its direct result is that the sharpness of video image declines, the digitized processing of video image also can by this affects hydraulic performance decline.Carry out to the video image polluted by raindrop the further process that repair process is conducive to image, the performance comprised based on technology such as the target detection of image, identification, tracking, segmentation and monitoring improves.And video goes rain technology all to have wide practical use in fields such as modern military, traffic and security monitorings based on the image of wavelet analysis.
About in video image, the research of raindrop characteristic has been subject to the extensive concern of international academic community, go the research of rain algorithm also from (StarikS such as Starik in 2003, WermanM.Simulationofraininvideos [C] ProceedingofTextureWorkshop, ICCV.Nice, France:2003, median method 2:406-409) proposed starts to obtain and develops rapidly, the method of process has no longer been confined to initial simple median calculation, the degree of bias calculates, K mean cluster, Kalman filtering, dictionary learning and sparse coding, guide filtering, interframe luminance difference, HSV space, optical flow method, a lot of method such as motion segmentation also starts to be applied in raindrop in video image gradually and detects with the algorithm removed, the effect that raindrop are removed also is enhanced gradually.The interframe luminance difference that Garg etc. propose to utilize raindrop to bring at first carries out raindrop initial survey, then the rectilinearity of the raindrop feature consistent with direction is utilized to screen further, finally remove raindrop impact according to the pixel intensity of front and back frame, the raindrop that raindrop do not cover in sequential frame image situation can be met preferably and detect and remove; The influence of color that raindrop bring to pixel is taken into account by Zhang etc., thus improve raindrop detect accuracy, improve based on brightness change remove the effect of rain algorithm on coloured image; The brightness impact of raindrop and influence of color are applied in the algorithm by Liu etc. simultaneously, detect raindrop and remove with two frames; Tripathi etc. first study the probabilistic statistical characteristics of raindrop pixel intensity change, and the symmetry then utilizing raindrop pixel intensity change realizes raindrop and detects, only based on time domain and in addition the affecting of consideration locus time effect incomplete same; First Kang etc. utilize bilateral filtering that rain figure is divided into HFS and low frequency part, and process further HFS and obtain non-rain composition, obtain rain figure in conjunction with low frequency part; First Huang etc. utilize context to retrain to carry out Iamge Segmentation, and utilize context-aware to carry out single width to remove rain based on the image of wavelet analysis, and propose innovatory algorithm on this basis, first used super complete dictionary and process HFS in literary composition.
Particularly recent years, the study hotspot that video image goes rain technology to become new.How ensureing that the prerequisite of high robust is rained the accuracy rate and real-time that improve and go, it is the focus that current video image goes to rain field.In the algorithm of current existence, be applied to static scene video raindrop and detect the achievement in research having comparatively maturation with the algorithm removed, but when being applied on the video in dynamic scene, algorithm, it is considered that there is the interference that moving object brings in video, cannot reach desirable Detection results for the moving object not high with raindrop characteristic difference degree.In addition, real-time process in the occasion such as automated navigation system, safety monitoring system applied in multinomial technology has very large application demand.Often need in these application scenarios to obtain result in time, feed back to user, the delayed of Video processing likely causes user to do the judgement made mistake.Therefore in video, raindrop detection not only needs to improve precision with removal, also needs to improve processing speed, and needs to find optimal balance point therebetween.But current algorithm also cannot take into account processing speed and the precision of various scene, realize going the real-time of rain algorithm to be an important topic faced by current research.
In sum, the shortcoming that existing image goes rain technology to exist is: existing image goes rain algorithm cannot take into account processing speed and the precision of various scene; Meanwhile, existing image goes rain technology not to be very desirable for the rain effect of going of dynamic scene, and algorithm complex and algorithm real-time can not be taken into account well.
Summary of the invention
The invention provides a kind of image based on wavelet analysis and go rain method and system, being intended to solve existing image goes rain technology cannot take into account processing speed and the precision of various scene, rain effect of going for dynamic scene is not very desirable, and the technical matters that algorithm complex and algorithm real-time can not be taken into account.
The present invention is achieved in that a kind of image based on wavelet analysis goes rain method, comprising:
Step a: the decomposition of small echo multilayer is carried out to video frame images according to wavelet analysis;
Step b: the image information analyzing each layer, detects the layer comprising rain cell noise;
Step c: according to the light characteristic definition raindrop pollution level coefficient of raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image.
The technical scheme that the embodiment of the present invention is taked also comprises: in described step a, described according to wavelet analysis to video frame images carry out small echo multilayer decompose be specially: the Malla algorithm based on wavelet analysis decomposes, and the decomposition formula of Malla algorithm is:
C
i=H
cH
rC
i-1
In above-mentioned formula, H and G is the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) respectively, C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.
The technical scheme that the embodiment of the present invention is taked also comprises: in described step c, the described definition of the light characteristic according to raindrop raindrop pollution level coefficient, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image and be specially: make raindrop pollution level coefficient S=G × E, wherein, G is partial gradient, E is local energy, partial gradient and local energy two parameters are multiplied and obtain a new variable S, the larger pollution of S value is more serious, s-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, and three partial fusion results are carried out Image Reconstruction obtain rain image.
The technical scheme that the embodiment of the present invention is taked also comprises: described partial gradient G is defined as:
In above-mentioned formula, Δ xf (i, j) and Δ yf (i, j) is respectively the horizontal and vertical direction of point (i, j), M and N is respectively the length of side in region; The local energy E of described pixel is expressed as:
The technical scheme that the embodiment of the present invention is taked also comprises: also comprise after described step c: utilize and carry out mist elimination process based on dark primary mist elimination algorithm to removing rain image, finally removed rain image.
The technical scheme that the embodiment of the present invention is taked also comprises: described utilization is specially going rain image to carry out mist elimination process based on dark primary mist elimination algorithm: adopt quick bilateral filtering method to carry out mist elimination; Two-sided filter is defined as:
In above-mentioned formula, Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is:
Above-mentioned formula is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.
Another technical scheme that the embodiment of the present invention is taked is: a kind of image based on wavelet analysis goes rain system, comprises wavelet decomposition module and wavelet fusion module; Described wavelet decomposition module is used for carrying out the decomposition of small echo multilayer according to wavelet analysis to image, and analyzes the image information of each layer, detects the layer comprising rain cell noise; Described wavelet fusion module is used for the light characteristic definition raindrop pollution level coefficient according to raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image.
The technical scheme that the embodiment of the present invention is taked also comprises: described wavelet decomposition module is carried out the decomposition of small echo multilayer according to wavelet analysis to video frame images and is specially: the Malla algorithm based on wavelet analysis decomposes, and the decomposition formula of Malla algorithm is:
C
i=H
cH
rC
i-1
In above-mentioned formula, H and G is the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) respectively, C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.
The technical scheme that the embodiment of the present invention is taked also comprises: described wavelet fusion module is according to the light characteristic definition raindrop pollution level coefficient of raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image and be specially: make raindrop pollution level coefficient S=G × E, wherein, G is partial gradient, E is local energy, partial gradient and local energy two parameters are multiplied and obtain a new variable S, the larger pollution of S value is more serious, s-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, and three partial fusion results are carried out Image Reconstruction obtain rain image.
The technical scheme that the embodiment of the present invention is taked also comprises: also comprise raindrop and get rid of module, and described raindrop are got rid of module and carried out mist elimination process based on dark primary mist elimination algorithm to removing rain image for utilizing, and are finally removed rain image; Be specially: adopt quick bilateral filtering method to carry out mist elimination; Two-sided filter is defined as:
In above-mentioned formula, Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is:
Above-mentioned formula is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.
The image based on wavelet analysis of the embodiment of the present invention goes rain method and system to adopt the method for small echo multi-level decomposition and Wavelet Fusion to differentiate the aspect at rain cell noise place, based on the rule of raindrop effect definition Wavelet Fusion, and in specific aspect, carry out Wavelet Fusion to reach the object of raindrop removal, the interference being subject to dynamic perfromance can be avoided, remove raindrop more accurately and effectively, improve the usable range of algorithm, make algorithm rain also can have in the situation that the force of rain is very large and good go effect; And to going rain image to carry out defogging, optimize the visual effect of image, make algorithm more have practicality.
Accompanying drawing explanation
Fig. 1 is the process flow diagram that the image based on wavelet analysis of the embodiment of the present invention goes rain method;
Fig. 2 is wavelet decomposition schematic diagram; Wherein, Fig. 2 (a) is original image, and Fig. 2 (b) is the low-frequency information of image, and Fig. 2 (c) to Fig. 2 (l) is the high-frequency structure that the tenth of image arrives ground floor;
Fig. 3 is Wavelet decomposing and recomposing process flow diagram;
Fig. 4 is the structural representation that the image based on wavelet analysis of the embodiment of the present invention goes rain system.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Referring to Fig. 1, is the process flow diagram that the image based on wavelet analysis of the embodiment of the present invention goes rain method.Because rain has direction, intensity and the uncertain characteristic of outward appearance, so be difficult to set up unified physical model in spatial domain.But analyze on frequency domain, can not affect by these uncertain characteristics.From the angle of frequency domain, the present invention adopts the method for the decomposition of small echo multilayer and Wavelet Fusion to differentiate the aspect at rain cell noise place, based on the rule of raindrop effect definition Wavelet Fusion, and in specific aspect, carry out Wavelet Fusion to reach the object of raindrop removal.Particularly, the image based on wavelet analysis of the embodiment of the present invention goes rain method to comprise the following steps:
Step 100: input video two field picture, carries out the decomposition of small echo multilayer according to wavelet analysis to image, and analyzes the image information of each layer, detect the layer comprising rain cell noise;
In step 100, wavelet transformation is very high-quality image analysis method, and the multi-layer image of wavelet transformation decomposes the frequency range that can find image raindrop.Wavelet analysis has good temporal frequency locating features, signal decomposition can be become multiple and have different frequency sub-band, the frequency sub-band of frequency characteristic and directivity characteristics, so wavelet analysis is also referred to as school microscop.By the inspiration of picture breakdown and restructing algorithm, Mallat proposes the Malla algorithm based on wavelet analysis, i.e. the many resolution decomposition of image and reconstruct pyramid algorith.Restructing algorithm is the inverse process of decomposition algorithm, just can recover original burst through liftering.
the projection f (x, y) in space can be used for representing two dimensional image signal:
Obtain decomposition formula if the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) is H and G, Malla algorithm is respectively:
C
i=H
cH
rC
i-1
In formula (4), C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.The restructing algorithm of Malla algorithm can be expressed as:
In formula (5), H
*, G
*be respectively the inverse matrix of H, G.
Described layer comprises the layer comprising background and colouring information, the layer comprising rain cell noise and comprises the layer of image texture and object edge information.Wavelet analysis can carry out linearly image respectively, high pass and low-pass filtering.Carry out the low-pass filtering of row and column on the original image, the low frequency coefficient approximation component C of ground floor can be obtained
1, it comprises image background and colouring information.Carry out the high-pass filtering of row and column on the original image, horizontal high-frequent detail coefficients can be obtained
vertical high frequency detail coefficients
with diagonal line high frequency detail coefficient
they comprise image texture and the object edge information of different directions.Above-mentioned boundary operation repeats at C
1the enterprising line operate of low frequency component, can obtain each frequency component C of the corresponding second layer
2,
with
if above-mentioned boundary operation repeats to carry out at m-1 layer, C can be obtained
m,
with
The frequency of rain cell noise is very high, and the noise ratio raindrop of image texture and object edge are taller, and the frequency of image background and colouring information is very low.So by decomposing based on wavelet analysis multilayer by raindrop pollution image, can find that rain cell noise should be comprised in the high frequency coefficient part of low layer, larger Decomposition order is usually used to guarantee the detailed information of image after rain.In embodiments of the present invention, the Decomposition order carrying out the decomposition of small echo multilayer to image is ten layers.Specifically as shown in Figure 2, be wavelet decomposition schematic diagram; Wherein, Fig. 2 (a) is original image, and Fig. 2 (b) is the low-frequency information of image, and Fig. 2 (c) to Fig. 2 (l) is the high-frequency structure that the tenth of image arrives ground floor.From analysis, most rain cell noise concentrates in the second to the four high frequency coefficient layer, and the five to the ten high frequency coefficient layer then comprises most image background and colouring information, and ground floor comprises image texture and object edge information.
Step 200: according to the light characteristic definition raindrop pollution level coefficient of raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image;
In step 200, because the gray shade scale of the pixel covered by raindrop is larger than background gray scale, can edge effect be produced, so partial gradient can be used for measuring the change of gray scale, and partial gradient is defined as:
In formula (6), Δ xf (i, j) and Δ yf (i, j) is respectively the horizontal and vertical direction of point (i, j), M and N is respectively the length of side in region.Because raindrop brightness is substantially constant, raindrop pixel has higher and more stable energy, and the local energy of pixel can be expressed as:
Make raindrop pollution level coefficient S=G × E, namely partial gradient and local energy two parameters are multiplied and obtain a new variable S, and the larger pollution of S value is more serious.S-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, finally three partial fusion results are carried out Image Reconstruction and obtain rain image, can make like this color of image and details undistorted.In embodiments of the present invention, in order to remove raindrop, for the layer at rain cell noise place, the weights of fusion coefficients matrix should be less than 1; And for not by the layer that raindrop pollute, the weights of fusion coefficients matrix are set as and are greater than 1; Specifically as shown in Figure 3, be Wavelet decomposing and recomposing process flow diagram.
Step 300: utilize and carry out mist elimination process based on dark primary mist elimination algorithm to removing rain image, finally removed rain image;
In step 300, when Sometimes When It Rains, be first can thicken, contrast declines, so after going rain operation, carry out the mist elimination process based on dark priority algorithm improved.The present invention adopts quick bilateral filtering method, obtains clearly dark edge.Bilateral filtering can process the details fuzzy problem that medium filtering produces, and make mist elimination effect more natural, particularly, two-sided filter is defined as:
In formula (8), Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is as follows:
Formula (9) is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.Bilateral filtering not only considers spatial information, also contemplates the monochrome information of pixel, can make the quick and high-quality image mist elimination of the clear realization of image smoothing.
Step 400: output video two field picture.
Referring to Fig. 4, is the structural representation that the image based on wavelet analysis of the embodiment of the present invention goes rain system.The image based on wavelet analysis of the embodiment of the present invention goes rain system to comprise wavelet decomposition module, wavelet fusion module, image mist elimination module and image output module; Particularly:
Wavelet decomposition module is used for input video two field picture, carries out the decomposition of small echo multilayer, and analyzes the image information of each layer, detect the layer comprising rain cell noise according to wavelet analysis to image; Wherein, wavelet transformation is very high-quality image analysis method, and the multi-layer image of wavelet transformation decomposes the frequency range that can find image raindrop.Wavelet analysis has good temporal frequency locating features, signal decomposition can be become multiple and have different frequency sub-band, the frequency sub-band of frequency characteristic and directivity characteristics, so wavelet analysis is also referred to as school microscop.By the inspiration of picture breakdown and restructing algorithm, Mallat proposes the Malla algorithm based on wavelet analysis, i.e. the many resolution decomposition of image and reconstruct pyramid algorith.Restructing algorithm is the inverse process of decomposition algorithm, just can recover original burst through liftering.
the projection f (x, y) in space can be used for representing two dimensional image signal:
Obtain decomposition formula if the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) is H and G, Malla algorithm is respectively:
C
i=H
cH
rC
i-1
In formula (4), C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.The restructing algorithm of Malla algorithm can be expressed as:
In formula (5), H
*, G
*be respectively the inverse matrix of H, G.
Described layer comprises the layer comprising background and colouring information, the layer comprising rain cell noise and comprises the layer of image texture and object edge information.Wavelet analysis can carry out linearly image respectively, high pass and low-pass filtering.Carry out the low-pass filtering of row and column on the original image, the low frequency coefficient approximation component C of ground floor can be obtained
1, it comprises image background and colouring information.Carry out the high-pass filtering of row and column on the original image, horizontal high-frequent detail coefficients can be obtained
vertical high frequency detail coefficients
with diagonal line high frequency detail coefficient
they comprise image texture and the object edge information of different directions.Above-mentioned boundary operation repeats at C
1the enterprising line operate of low frequency component, can obtain each frequency component C of the corresponding second layer
2,
with
if above-mentioned boundary operation repeats to carry out at m-1 layer, C can be obtained
m,
with
The frequency of rain cell noise is very high, and the noise ratio raindrop of image texture and object edge are taller, and the frequency of image background and colouring information is very low.So by decomposing based on wavelet analysis multilayer by raindrop pollution image, can find that rain cell noise should be comprised in the high frequency coefficient part of low layer, larger Decomposition order is usually used to guarantee the detailed information of image after rain.In embodiments of the present invention, the Decomposition order carrying out the decomposition of small echo multilayer to image is ten layers.Specifically as shown in Figure 2, be wavelet decomposition schematic diagram; Wherein, Fig. 2 (a) is original image, and Fig. 2 (b) is the low-frequency information of image, and Fig. 2 (c) to Fig. 2 (l) is the high-frequency structure that the tenth of image arrives ground floor.From analysis, most rain cell noise concentrates in the second to the four high frequency coefficient layer, and the five to the ten high frequency coefficient layer then comprises most image background and colouring information, and ground floor comprises image texture and object edge information.
Wavelet fusion module is used for the light characteristic definition raindrop pollution level coefficient according to raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image; Wherein, because the gray shade scale of the pixel covered by raindrop is larger than background gray scale, can edge effect be produced, so partial gradient can be used for measuring the change of gray scale, and partial gradient is defined as:
In formula (6), Δ xf (i, j) and Δ yf (i, j) is respectively the horizontal and vertical direction of point (i, j), M and N is respectively the length of side in region.Because raindrop brightness is substantially constant, raindrop pixel has higher and more stable energy, and the local energy of pixel can be expressed as:
Make raindrop pollution level coefficient S=G × E, namely partial gradient and local energy two parameters are multiplied and obtain a new variable S, and the larger pollution of S value is more serious.S-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, finally three partial fusion results are carried out Image Reconstruction and obtain rain image, can make like this color of image and details undistorted.In embodiments of the present invention, in order to remove raindrop, for the layer at rain cell noise place, the weights of fusion coefficients matrix should be less than 1; And for not by the layer that raindrop pollute, the weights of fusion coefficients matrix are set as and are greater than 1; Specifically as shown in Figure 3, be Wavelet decomposing and recomposing process flow diagram.
Image mist elimination module is used for utilizing carries out mist elimination process based on dark primary mist elimination algorithm to removing rain image, is finally removed rain image; Wherein, when Sometimes When It Rains, be first can thicken, contrast declines, so after going rain operation, carry out the mist elimination process based on dark priority algorithm improved.The present invention adopts quick bilateral filtering method, obtains clearly dark edge.Bilateral filtering can process the details fuzzy problem that medium filtering produces, and make mist elimination effect more natural, particularly, two-sided filter is defined as:
In formula (8), Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is as follows:
Formula (9) is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.Bilateral filtering not only considers spatial information, also contemplates the monochrome information of pixel, can make the quick and high-quality image mist elimination of the clear realization of image smoothing.
Image output module is used for output video two field picture.
The image based on wavelet analysis of the embodiment of the present invention goes rain method and system to adopt the method for small echo multi-level decomposition and Wavelet Fusion to differentiate the aspect at rain cell noise place, based on the rule of raindrop effect definition Wavelet Fusion, and in specific aspect, carry out Wavelet Fusion to reach the object of raindrop removal, the interference being subject to dynamic perfromance can be avoided, remove raindrop more accurately and effectively, improve the usable range of algorithm, make algorithm rain also can have in the situation that the force of rain is very large and good go effect; And to going rain image to carry out defogging, optimize the visual effect of image, make algorithm more have practicality.
Not only there is good restoration result, but also not by the impact of noise intensity degree.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. the image based on wavelet analysis goes a rain method, comprising:
Step a: the decomposition of small echo multilayer is carried out to video frame images according to wavelet analysis;
Step b: the image information analyzing each layer, detects the layer comprising rain cell noise;
Step c: according to the light characteristic definition raindrop pollution level coefficient of raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image.
2. the image based on wavelet analysis according to claim 1 goes rain method, it is characterized in that, in described step a, described according to wavelet analysis to video frame images carry out small echo multilayer decompose be specially: the Malla algorithm based on wavelet analysis decomposes, and the decomposition formula of Malla algorithm is:
C
i=H
cH
rC
i-1
In above-mentioned formula, H and G is the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) respectively, C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.
3. the image based on wavelet analysis according to claim 1 goes rain method, it is characterized in that, in described step c, the described definition of the light characteristic according to raindrop raindrop pollution level coefficient, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image and be specially: make raindrop pollution level coefficient S=G × E, wherein, G is partial gradient, E is local energy, partial gradient and local energy two parameters are multiplied and obtain a new variable S, the larger pollution of S value is more serious, s-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, and three partial fusion results are carried out Image Reconstruction obtain rain image.
4. the image based on wavelet analysis according to claim 3 goes rain method, it is characterized in that, described partial gradient G is defined as:
In above-mentioned formula, △ xf (i, j) and △ yf (i, j) is respectively the horizontal and vertical direction of point (i, j), M and N is respectively the length of side in region; The local energy E of described pixel is expressed as:
5. the image based on wavelet analysis according to claim 1 goes rain method, it is characterized in that, also comprises after described step c: utilize and carry out mist elimination process based on dark primary mist elimination algorithm to removing rain image, finally removed rain image.
6. the image based on wavelet analysis according to claim 5 goes rain method, it is characterized in that, described utilization is specially going rain image to carry out mist elimination process based on dark primary mist elimination algorithm: adopt quick bilateral filtering method to carry out mist elimination; Two-sided filter is defined as:
In above-mentioned formula, Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is:
Above-mentioned formula is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.
7. the image based on wavelet analysis goes a rain system, it is characterized in that, comprises wavelet decomposition module and wavelet fusion module; Described wavelet decomposition module is used for carrying out the decomposition of small echo multilayer according to wavelet analysis to image, and analyzes the image information of each layer, detects the layer comprising rain cell noise; Described wavelet fusion module is used for the light characteristic definition raindrop pollution level coefficient according to raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image.
8. the image based on wavelet analysis according to claim 7 goes rain system, it is characterized in that, described wavelet decomposition module is carried out the decomposition of small echo multilayer according to wavelet analysis to video frame images and is specially: the Malla algorithm based on wavelet analysis decomposes, and the decomposition formula of Malla algorithm is:
C
i=H
cH
rC
i-1
In above-mentioned formula, H and G is the matrix of coefficients of scaling function φ (x) and wavelet function ψ (x) respectively, C
i,
with
correspondence image C respectively
i-1low frequency part, the HFS of vertical direction, the HFS of vertical direction and diagonal HFS.
9. the image based on wavelet analysis according to claim 7 goes rain system, it is characterized in that, described wavelet fusion module is according to the light characteristic definition raindrop pollution level coefficient of raindrop, according to raindrop pollution level coefficient calculations fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising different images information, and carry out Image Reconstruction by fusion results and obtain rain image and be specially: make raindrop pollution level coefficient S=G × E, wherein, G is partial gradient, E is local energy, partial gradient and local energy two parameters are multiplied and obtain a new variable S, the larger pollution of S value is more serious, s-matrix is normalized and obtains S ', matrix of coefficients and S ' matrix are used for being weighted optimization to image reconstruction algorithm and obtain fusion coefficients matrix, according to fusion coefficients matrix, respectively Wavelet Fusion is carried out to the layer comprising background and colouring information, rain cell noise and image texture and object edge information, and three partial fusion results are carried out Image Reconstruction obtain rain image.
10. the image based on wavelet analysis according to claim 7 goes rain system, it is characterized in that: also comprise raindrop and get rid of module, described raindrop are got rid of module and are carried out mist elimination process based on dark primary mist elimination algorithm to removing rain image for utilizing, and are finally removed rain image; Be specially: adopt quick bilateral filtering method to carry out mist elimination; Two-sided filter is defined as:
In above-mentioned formula, Ip and Iq is the brightness value of pixel p and q respectively, and Wp expression formula is:
Above-mentioned formula is normalized weight coefficient, σ
dand σ
reuclidean space distance respectively and the standard deviation of brightness variation range,
with
the gaussian kernel function of respective standard difference respectively.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510541354.1A CN105184761A (en) | 2015-08-28 | 2015-08-28 | Image rain removing method based on wavelet analysis and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510541354.1A CN105184761A (en) | 2015-08-28 | 2015-08-28 | Image rain removing method based on wavelet analysis and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105184761A true CN105184761A (en) | 2015-12-23 |
Family
ID=54906815
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510541354.1A Pending CN105184761A (en) | 2015-08-28 | 2015-08-28 | Image rain removing method based on wavelet analysis and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105184761A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107274369A (en) * | 2017-06-16 | 2017-10-20 | 南京信息职业技术学院 | Single image defogging acceleration method based on frequency domain decomposition |
CN109636738A (en) * | 2018-11-09 | 2019-04-16 | 温州医科大学 | The single image rain noise minimizing technology and device of double fidelity term canonical models based on wavelet transformation |
CN110111268A (en) * | 2019-04-18 | 2019-08-09 | 上海师范大学 | The removing rain based on single image method and device learnt based on dark and blurred width |
CN110246102A (en) * | 2019-06-13 | 2019-09-17 | 中国人民解放军陆军炮兵防空兵学院 | A kind of clear processing method of rainy day video |
CN111178340A (en) * | 2020-04-10 | 2020-05-19 | 支付宝(杭州)信息技术有限公司 | Image recognition method and training method of image recognition model |
CN111861957A (en) * | 2020-07-02 | 2020-10-30 | Tcl华星光电技术有限公司 | Image fusion method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103729828A (en) * | 2013-12-12 | 2014-04-16 | 中国科学院深圳先进技术研究院 | Video rain removing method |
CN104112259A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
CN104112255A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
CN104299199A (en) * | 2014-10-22 | 2015-01-21 | 中国科学院深圳先进技术研究院 | Video raindrop detection and removal method based on wavelet transform |
-
2015
- 2015-08-28 CN CN201510541354.1A patent/CN105184761A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103729828A (en) * | 2013-12-12 | 2014-04-16 | 中国科学院深圳先进技术研究院 | Video rain removing method |
CN104112259A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
CN104112255A (en) * | 2014-06-19 | 2014-10-22 | 中国科学院深圳先进技术研究院 | Rain removing method and system for single image |
CN104299199A (en) * | 2014-10-22 | 2015-01-21 | 中国科学院深圳先进技术研究院 | Video raindrop detection and removal method based on wavelet transform |
Non-Patent Citations (1)
Title |
---|
CHEN ZHEN等: "A New Algorithm of Rain (Snow) Removal in Video", 《JOURNAL OF MULTIMEDIA》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107274369A (en) * | 2017-06-16 | 2017-10-20 | 南京信息职业技术学院 | Single image defogging acceleration method based on frequency domain decomposition |
CN109636738A (en) * | 2018-11-09 | 2019-04-16 | 温州医科大学 | The single image rain noise minimizing technology and device of double fidelity term canonical models based on wavelet transformation |
CN109636738B (en) * | 2018-11-09 | 2019-10-01 | 温州医科大学 | The single image rain noise minimizing technology and device of double fidelity term canonical models based on wavelet transformation |
CN110111268A (en) * | 2019-04-18 | 2019-08-09 | 上海师范大学 | The removing rain based on single image method and device learnt based on dark and blurred width |
CN110111268B (en) * | 2019-04-18 | 2021-08-03 | 上海师范大学 | Single image rain removing method and device based on dark channel and fuzzy width learning |
CN110246102A (en) * | 2019-06-13 | 2019-09-17 | 中国人民解放军陆军炮兵防空兵学院 | A kind of clear processing method of rainy day video |
CN110246102B (en) * | 2019-06-13 | 2022-05-31 | 中国人民解放军陆军炮兵防空兵学院 | Method for clearly processing video in rainy days |
CN111178340A (en) * | 2020-04-10 | 2020-05-19 | 支付宝(杭州)信息技术有限公司 | Image recognition method and training method of image recognition model |
CN111178340B (en) * | 2020-04-10 | 2020-07-21 | 支付宝(杭州)信息技术有限公司 | Image recognition method and training method of image recognition model |
CN111861957A (en) * | 2020-07-02 | 2020-10-30 | Tcl华星光电技术有限公司 | Image fusion method and device |
CN111861957B (en) * | 2020-07-02 | 2024-03-08 | Tcl华星光电技术有限公司 | Image fusion method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105184761A (en) | Image rain removing method based on wavelet analysis and system | |
Li et al. | A comprehensive benchmark analysis of single image deraining: Current challenges and future perspectives | |
Dudhane et al. | C^ 2msnet: A novel approach for single image haze removal | |
Kang et al. | Automatic single-image-based rain streaks removal via image decomposition | |
Kuanar et al. | Night time haze and glow removal using deep dilated convolutional network | |
Chillakuru et al. | Enhancing wind power monitoring through motion deblurring with modified GoogleNet algorithm | |
Hassan et al. | Real-time image dehazing by superpixels segmentation and guidance filter | |
CN105096272B (en) | A kind of defogging method based on dual-tree complex wavelet | |
CN105913392A (en) | Degraded image overall quality improving method in complex environment | |
CN105976383A (en) | Power transmission equipment fault diagnosis method based on limit learning machine image recognition | |
Ju et al. | BDPK: Bayesian dehazing using prior knowledge | |
CN103295225B (en) | Train bogie edge detection method under the conditions of low-light | |
Patil et al. | Motion saliency based generative adversarial network for underwater moving object segmentation | |
CN111709888B (en) | Aerial image defogging method based on improved generation countermeasure network | |
CN109446913A (en) | A kind of detection method for judging vehicle bottom and whether reequiping | |
Pushpalwar et al. | Image inpainting approaches-a review | |
CN105046670A (en) | Image rain removal method and system | |
CN113362375A (en) | Moving object detection method for vehicle | |
CN106067163A (en) | A kind of image rain removing method based on wavelet analysis and system | |
Fu et al. | An anisotropic Gaussian filtering model for image de-hazing | |
Wang et al. | An efficient remote sensing image denoising method in extended discrete shearlet domain | |
CN103020905A (en) | Sparse-constraint-adaptive NLM (non-local mean) super-resolution reconstruction method aiming at character image | |
CN106023112A (en) | Image rain removing method and system based on wavelet analysis | |
CN102254306B (en) | Real-time image defogging method based on image simplified hierachical model | |
Jiang et al. | Pseudo‐Siamese residual atrous pyramid network for multi‐focus image fusion |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151223 |