CN105139358A - Video raindrop removing method and system based on combination of morphology and fuzzy C clustering - Google Patents

Abstract

The invention provides a video raindrop removing method and system based on the combination of morphology and fuzzy C clustering, comprising the steps of: first, obtaining the gray scale intensity sequence at a video frame sequence corresponding pixel position in a video image; according to the gray scale intensity sequence, dividing pixels into a raindrop kind and a background kind by utilizing a fuzzy C-means algorithm; calculating length-width ratios corresponding to rain lines in the raindrop kind and the background kind by utilizing raindrop morphological characteristics; distinguishing a raindrop area and a moving target area according to the length-width ratios, determining a pixel to be the raindrop area if the length-width ratio of pixels of the background kind is greater, or else determining the pixel to be the moving target area; further removing moving objects; obtaining polluted pixels according to a detection result; and finally replacing the pixels detected to be polluted by raindrops with the mixed value between the raindrops and a background color to remove raindrops, thereby accurately and effectively removing raindrops, and improving the adaptability of a rain-removing algorithm.

Description

The video raindrop minimizing technology of a kind of combining form and fuzzy C-means clustering and system thereof

Technical field

The present invention relates to technical field of image processing, particularly relate to video raindrop minimizing technology and the system thereof of a kind of combining form and fuzzy C-means clustering.

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 image goes rain technology all to have wide practical use in fields such as modern military, traffic and security monitorings.

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 image to remove rain, 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.So we propose on this basis a kind of utilize gauss hybrid models to improve effectively go rain method.Program image goes rain algorithm to have very high accuracy, has more robustness.

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.

Summary of the invention

Based on this, the invention provides the video raindrop minimizing technology of a kind of combining form and fuzzy C-means clustering, effectively to solve prior art Problems existing.

The video raindrop minimizing technology providing a kind of combining form and fuzzy C-means clustering can not be invented, comprise the steps:

Obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image;

According to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes;

Raindrop morphological characteristic is utilized to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes;

According to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district;

Raindrop district obtained above is deducted moving target district and obtains pure raindrop region;

The pixel polluted by raindrop is extracted in described pure raindrop region;

Detect that the pixel polluted by raindrop realizes the removal of raindrop by replacing by the mixed number between raindrop and background colour.

In certain embodiments, wherein, obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image, comprise the steps:

What produce when building formula quantitative description raindrop whereabouts is fuzzy,

$\begin{array}{c}{I}_r(x,y)={\∫}_0^{\mathrm{\τ}}{E}_r(x,y)dt+{\∫}_{\mathrm{\τ}}^T{E}_b(x,y)dt\\ ={\mathrm{\αI}}_E(x,y)+(1-\mathrm{\α})I_b(x,y)\end{array}$

I _r(x, y) represents the pixel intensity of location of pixels (x, y), and τ represents that raindrop fell through the time required for location of pixels (x, y), and T is the camera exposure time, E _r(x, y) represents the irradiance of raindrop through location of pixels (x, y), E _bthe average irradiance that (x, y) is background pixel;

According to described pixel intensity, the pixel of image sequence in described video image is divided into background classes and rain belt class.

In certain embodiments, wherein, according to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes, specifically comprise the steps:

Objective definition function is:

$J(U,V)=\underset{i=1}{\overset{n}{\Σ}}\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}^m{d}^2(x_i,v_j)$

Wherein, smoothing factor m controls and shares degree between fuzzy class, and the optimum valuing range of m is [1.5,2.5], for subordinated-degree matrix, meet $\begin{array}{cc}\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}=1,& V=\{v_1,v_2,...v_c\}\end{array}$ For the set of C cluster centre, d (x _i, v _j) be the Euclidean distance of i-th sample to a jth cluster, the criterion of cluster generally gets the minimal value of objective function J (U, V);

Lagrange multiplier is adopted to solve above-mentioned formula as follows:

$F=\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}^m\·d_{ij}^2+\mathrm{\λ}(\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}-1)$

Adopt optimization first order necessary condition, following iteration optimization formula can be obtained:

$\left\{\begin{array}{c}{\mathrm{\μ}}_{ij}={\left(d_{ij}^2\right)}^{-1/(m-1)}/\underset{k=1}{\overset{c}{\mathrm{\Σ}}}{\left(d_{kj}^2\right)}^{-1/(m-1)},\\ {\mathrm{\ν}}_j=k=\underset{k=1}{\overset{c}{\mathrm{\Σ}}}{\mathrm{\μ}}_{ik}^m{x}_k/\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\μ}}_{ij}^m\end{array}\right.$

Described optimization first order necessary condition is:

Based on above-mentioned iterationization excellent formula, pixel is divided into raindrop class and background classes.

In certain embodiments, wherein, utilize raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes, be specially:

Build the final speed v of raindrop and the relational expression of diameter d, be designated as:

v(d)＝-0.2+0.5d-0.9d ²+0.1d ³

Wherein, $\begin{array}{cc}b(d,z)=\frac{df}{z},& l(d,z)=\frac{v\left(d\right)ef}{z}+\frac{df}{z}\end{array};$

According to above-mentioned relation formula, calculate the scope of length breadth ratio corresponding to rain line:

$AR\left(d\right)=\frac{l(d,z)}{b(d,z)}=\frac{v\left(d\right)e}{d}+1$

Wherein, the diameter of raindrop is d, and the distance of raindrop and camera is the focal length that the parameter list of z and camera shows the length l of a rain line and width b, f and represents camera, and e represents the time shutter of camera.

In certain embodiments, in described pure raindrop region, extract the pixel polluted by raindrop, be specially:

In described pure raindrop region, come to distinguish with other fast moving objects according to the rectilinearity of raindrop, finally draw the pixel polluted by raindrop.

In certain embodiments, detecting that the pixel polluted by raindrop realizes the removal of raindrop by replacing by the mixed number between raindrop and background colour, being specially:

The pixel of alternative raindrop is set to I _mix, make I _mix=α I _bcenter+ (1-α) I _rcenter, wherein α is the weight that in background colour B, number accounts for total number k.

The video raindrop that present invention also offers a kind of combining form and fuzzy C-means clustering on the other hand remove system, comprising:

Gray-scale intensity sequence units, obtains the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image;

Fuzzy C-means clustering unit, according to described gray-scale intensity sequence, utilizes Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes;

Raindrop length breadth ratio computing unit, utilizes raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes;

Pixel discrimination unit, according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district;

Pure raindrop unit, the raindrop district obtained by above-mentioned initial survey deducts moving target district and obtains pure raindrop region;

Pixel extraction unit, extracts the pixel polluted by raindrop in described pure raindrop region;

By replacing by the mixed number between raindrop and background colour, raindrop removal unit, detects that the pixel polluted by raindrop realizes the removal of raindrop.

The present invention adopts technique scheme to have following beneficial effect:

The video raindrop minimizing technology of a kind of combining form provided by the invention and fuzzy C-means clustering and system, first the gray-scale intensity sequence in video image on sequence of frames of video respective pixel position is obtained, according to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes, raindrop morphological characteristic is utilized to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes, according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, then be judged as raindrop district, otherwise this pixel is moving target district, thus get rid of moving object further, contaminated pixel is obtained again according to testing result, detect that the pixel polluted by raindrop realizes the removal of raindrop finally by replacing by the mixed number between raindrop and background colour, thus raindrop can be removed accurately and effectively, improve the adaptability of rain algorithm.

In addition, the video raindrop minimizing technology of a kind of combining form provided by the invention and fuzzy C-means clustering and system, utilize α to mix and carry out raindrop removal, ensure that raindrop can not leave sharp-pointed edge after removing, add the robustness of algorithm.

Accompanying drawing explanation

The flow chart of steps of the video raindrop minimizing technology of the combining form that Fig. 1 provides for the embodiment of the present invention and fuzzy C-means clustering;

Fig. 2 is video image grey level histogram;

The video raindrop of the combining form that Fig. 3 provides for the embodiment of the present invention and fuzzy C-means clustering remove the structural representation of system.

Embodiment

For the ease of understanding the present invention, below with reference to relevant drawings, the present invention is described more fully.Better embodiment of the present invention is given in accompanying drawing.These are only the preferred embodiments of the present invention; not thereby the scope of the claims of the present invention is limited; every utilize instructions of the present invention and accompanying drawing content to do equivalent structure or equivalent flow process conversion; or be directly or indirectly used in other relevant technical fields, be all in like manner included in scope of patent protection of the present invention.

Unless otherwise defined, all technology used herein and scientific terminology are identical with belonging to the implication that those skilled in the art of the present invention understand usually.The object of term used in the description of the invention herein just in order to describe concrete embodiment, is not intended to be restriction the present invention.Term as used herein " and/or " comprise arbitrary and all combinations of one or more relevant Listed Items.

Refer to Fig. 1, the flow chart of steps of the video raindrop minimizing technology of the combining form that the embodiment of the present invention provides and fuzzy C-means clustering, comprises the steps:

Step S110: obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image;

Preferably, obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image, comprise the steps:

What produce when building formula quantitative description raindrop whereabouts is fuzzy,

$\begin{array}{c}{I}_r(x,y)={\∫}_0^{\mathrm{\τ}}{E}_r(x,y)dt+{\∫}_{\mathrm{\τ}}^T{E}_b(x,y)dt\\ ={\mathrm{\αI}}_E(x,y)+(1-\mathrm{\α})I_b(x,y)\end{array}$

I _r(x, y) represents the pixel intensity of location of pixels (x, y), and τ represents that raindrop fell through the time required for location of pixels (x, y), and T is the camera exposure time, E _r(x, y) represents the irradiance of raindrop through location of pixels (x, y), E _bthe average irradiance that (x, y) is background pixel;

According to described pixel intensity, the pixel of image sequence in described video image is divided into background classes and rain belt class.

Be appreciated that, because raindrop falling speed is fast, under normal exposure speed, image does not observe spherical raindrop, but the rain line that raindrop are formed due to rapid movement. under physical environment, above-mentioned formula can describe the physics imaging process of raindrop, and can quantitative description raindrop fall time produce fuzzy.

Be appreciated that the time domain specification according to raindrop, by the background light intensity that rain band covers, much larger than the light intensity of background itself.Therefore, in the static scene of Still Camera picked-up, if certain pixel was covered by rain, its light intensity histogram, refer to Fig. 2 and just there are two peaks, a light distribution (the dotted line left side) representing background, a light distribution (on the right of dotted line) representing raindrop.And if a certain pixel is not covered by raindrop in whole video, its light intensity histogram only has a peak representing background light distribution.For a certain pixel, according to light intensity property of the histogram, just the pixel of image sequence in whole video can be divided into two classes: background classes and rain belt class.

Step S120: according to described gray-scale intensity sequence, utilizes Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes;

Preferably, above-mentionedly specifically to comprise the steps:

Objective definition function is:

$J(U,V)=\underset{i=1}{\overset{n}{\Σ}}\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}^m{d}^2(x_i,v_j)$

Wherein, smoothing factor m controls and shares degree between fuzzy class, and the span of m is [1.5,2.5], for subordinated-degree matrix, meet $\begin{array}{cc}\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}=1,& V=\{v_1,v_2,...v_c\}\end{array}$ For the set of C cluster centre, d (x _i, v _j) be the Euclidean distance of i-th sample to a jth cluster, the criterion of cluster generally gets the minimal value of objective function J (U, V);

Lagrange multiplier is adopted to solve above-mentioned formula as follows:

$F=\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}^m\·d_{ij}^2+\mathrm{\λ}(\underset{j=1}{\overset{c}{\Σ}}{\mathrm{\μ}}_{ij}-1)$

Adopt optimization first order necessary condition, following iteration optimization formula can be obtained:

$\left\{\begin{array}{c}{\mathrm{\μ}}_{ij}={\left(d_{ij}^2\right)}^{-1/(m-1)}/\underset{k=1}{\overset{c}{\mathrm{\Σ}}}{\left(d_{kj}^2\right)}^{-1/(m-1)},\\ {\mathrm{\ν}}_j=k=\underset{k=1}{\overset{c}{\mathrm{\Σ}}}{\mathrm{\μ}}_{ik}^m{x}_k/\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\μ}}_{ij}^m.\end{array}\right.$

Described optimization first order necessary condition is:

Based on above-mentioned iterationization excellent formula, pixel is divided into raindrop class and background classes.

Be appreciated that FCM Algorithms is the fuzzy clustering method of based target function, be summed up as the nonlinear programming problem of a belt restraining by cluster, obtained the fuzzy division of data set by Optimization Solution; The whole computation process of C fuzzy clustering algorithm is exactly improvement cluster centre repeatedly and cluster subordinated-degree matrix, is finally summed up as and solves a nonlinear optimal problem.

Step S130: utilize raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes;

Preferably, utilize raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes, be specially:

Build the final speed v of raindrop and the relational expression of diameter d, be designated as:

v(d)＝-0.2+0.5d-0.9d ²+0.1d ³

Wherein, $\begin{array}{cc}b(d,z)=\frac{df}{z},& l(d,z)=\frac{v\left(d\right)ef}{z}+\frac{df}{z}\end{array};$

According to above-mentioned relation formula, calculate the scope of length breadth ratio corresponding to rain line:

$AR\left(d\right)=\frac{l(d,z)}{b(d,z)}=\frac{v\left(d\right)e}{d}+1$

Wherein, the diameter of raindrop is d, and the distance of raindrop and camera is the focal length that the parameter list of z and camera shows the length l of a rain line and width b, f and represents camera, and e represents the time shutter of camera.

Can find out, under rain the ratio of width to height of line only depend on diameter and the time shutter of raindrop, this also shows that the shorter camera exposure time can cause the stretching that visible rain line is lower, has nothing to do with the distance of raindrop and camera, other parameter of camera.

In practice, suppose concerning camera in the same circumstances minimum the and full-size of visible raindrop be respectively 0.1mm and 3mm, the time shutter of camera is 16ms, and so we can obtain the aspect ratio range 47<AR<52 of a rain line.If the time shutter of camera is unknown, we can suppose that it is between conventional camera exposure time range 1-40ms, so can draw the aspect ratio range 3.5<AR<95 of rain line.

Step S140: according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district;

Be appreciated that and can find out that raindrop have larger the ratio of width to height according to the shape attribute of raindrop.Object and the raindrop of motion can be distinguished based on this point, thus the scene containing moving object can be processed.If the length breadth ratio of foreground pixel is comparatively large, be then judged as raindrop district, otherwise this pixel is moving target district.The object that the method can well reduce motion detects and the impact of removing raindrop.

Step S150: raindrop district obtained above is deducted moving target district and obtains pure raindrop region;

Be appreciated that the raindrop region optics of frame difference method and raindrop and chromatic characteristic initial survey obtained, deduct the moving object region that optical flow method detects and obtain pure raindrop region.

Step S160: extract the pixel polluted by raindrop in described pure raindrop region;

Be appreciated that the fall trajectory due to raindrop is straight line, appear at projection plane when imaging with the form of rain line, therefore can come to distinguish with snow or other fast moving objects such as spray by the rectilinearity of raindrop, finally draw the pixel polluted by raindrop.

Step S170: detect that the pixel polluted by raindrop realizes the removal of raindrop by replacing by the mixed number between raindrop and background colour.

Preferably, detecting that the pixel polluted by raindrop realizes the removal of raindrop by replacing by the mixed number between raindrop and background colour, being specially:

The pixel of alternative raindrop is set to I _mix, make I _mix=α I _bcenter+ (1-α) I _rcenter, wherein α is the weight that in background colour B, number accounts for total number k.

Be appreciated that after successfully detecting rain belt, removal effect can be reached by the pixel replacing the raindrop detected by the mixed number between raindrop and background colour.According to statistical property, the pixel of alternative raindrop is set to I _mix, make I _mix=α I _bcenter+ (1-α) I _rcenter, wherein α is the weight that in background classes B, number accounts for total number k.Because at reservation background intensity values I _bcenterbasis on, be mixed with the average intensity value I of raindrop _rcenter, ensure that raindrop can not leave sharp-pointed edge after removing, serve level and smooth effect.

Refer to Fig. 3, the video raindrop that present invention also offers a kind of combining form and fuzzy C-means clustering remove system, comprising: gray-scale intensity sequence units 310, obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image; Fuzzy C-means clustering unit 320, according to described gray-scale intensity sequence, utilizes Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes; Raindrop length breadth ratio computing unit 330, utilizes raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes; Pixel discrimination unit 340, according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district; Pure raindrop unit 350, the raindrop district obtained by above-mentioned initial survey deducts moving target district and obtains pure raindrop region; Pixel extraction unit 360, extracts the pixel polluted by raindrop in described pure raindrop region; By replacing by the mixed number between raindrop and background colour, raindrop removal unit 370, detects that the pixel polluted by raindrop realizes the removal of raindrop.The detailed performing step of said system described in vide supra, here can repeat no more.

The video raindrop minimizing technology of a kind of combining form provided by the invention and fuzzy C-means clustering and system, first the gray-scale intensity sequence in video image on sequence of frames of video respective pixel position is obtained, according to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes, raindrop morphological characteristic is utilized to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes, according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, then be judged as raindrop district, otherwise this pixel is moving target district, thus get rid of moving object further, contaminated pixel is obtained again according to testing result, detect that the pixel polluted by raindrop realizes the removal of raindrop finally by replacing by the mixed number between raindrop and background colour, thus raindrop can be removed accurately and effectively, improve the adaptability of rain algorithm.

The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but therefore can not be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.

Claims (7)

1. a video raindrop minimizing technology for combining form and fuzzy C-means clustering, is characterized in that, comprise the steps:

Obtain the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image;

According to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes;

Raindrop morphological characteristic is utilized to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes;

According to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district;

Raindrop district obtained above is deducted moving target district and obtains pure raindrop region;

The pixel polluted by raindrop is extracted in described pure raindrop region;

Detect that the pixel polluted by raindrop realizes the removal of raindrop by replacing by the mixed number between raindrop and background colour.

2. the video raindrop minimizing technology of combining form according to claim 1 and fuzzy C-means clustering, is characterized in that, wherein, obtains the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image, comprises the steps:

What produce when building formula quantitative description raindrop whereabouts is fuzzy,

$\begin{array}{c}{I}_r(x,y)={\&Integral;}_0^{\mathrm{\&tau;}}{E}_r(x,y)dt+{\&Integral;}_{\mathrm{\&tau;}}^T{E}_b(x,y)dt\\ ={\mathrm{\&alpha;I}}_E(x,y)+(1-\mathrm{\&alpha;})I_b(x,y)\end{array}$

I _r(x, y) represents the pixel intensity of location of pixels (x, y), and τ represents that raindrop fell through the time required for location of pixels (x, y), and T is the camera exposure time, E _r(x, y) represents the irradiance of raindrop through location of pixels (x, y), E _bthe average irradiance that (x, y) is background pixel;

According to described pixel intensity, the pixel of image sequence in described video image is divided into background classes and rain belt class.

3. the video raindrop minimizing technology of combining form as claimed in claim 1 and fuzzy C-means clustering, it is characterized in that, wherein, according to described gray-scale intensity sequence, utilize Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes, specifically comprise the steps:

Objective definition function is:

$J(U,V)=\underset{i=1}{\overset{n}{\&Sigma;}}\underset{j=1}{\overset{c}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}^m{d}^2(x_i,v_j)$

Wherein, smoothing factor m controls and shares degree between fuzzy class, and the span of m is [1.5,2.5], for subordinated-degree matrix, meet for the set of C cluster centre, d (x _i, v _j) be the Euclidean distance of i-th sample to a jth cluster, the criterion of cluster generally gets the minimal value of objective function J (U, V);

Lagrange multiplier is adopted to solve above-mentioned formula as follows:

$F=\underset{j=1}{\overset{c}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}^m\&CenterDot;d_{ij}^2+\mathrm{\&lambda;}(\underset{j=1}{\overset{c}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}-1)$

Adopt optimization first order necessary condition, following iteration optimization formula can be obtained:

$\left\{\begin{array}{c}{\mathrm{\&mu;}}_{ij}={\left(d_{ij}^2\right)}^{-1/(m-1)}/\underset{k=1}{\overset{c}{\mathrm{\&Sigma;}}}{\left(d_{kj}^2\right)}^{-1/(m-1)},\\ {\mathrm{\&nu;}}_j=k=\underset{k=1}{\overset{c}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{ik}^m{x}_k/\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{ij}^m\end{array}\right.$

Described optimization first order necessary condition is: $\left\{\begin{array}{c}{\mathrm{\&mu;}}_{ij}={\left(d_{ij}^2\right)}^{-1/(m-1)}/\underset{k=1}{\overset{c^{-}}{\mathrm{\&Sigma;}}}{\left(d_{kj}^2\right)}^{-1/(m-1)},\\ {\mathrm{\&nu;}}_j=k=\underset{k=1}{\overset{c}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{ik}^m{x}_k/\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{ij}^m,j=1,2,...,c.\end{array}\right.;$

Based on above-mentioned iterationization excellent formula, pixel is divided into raindrop class and background classes.

4. the video raindrop minimizing technology of combining form as claimed in claim 1 and fuzzy C-means clustering, is characterized in that, wherein, utilizes raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes, is specially:

Build the final speed v of raindrop and the relational expression of diameter d, be designated as:

v(d)＝-0.2+0.5d-0.9d ²+0.1d ³

Wherein, $\begin{array}{cc}b(d,z)=\frac{df}{z},& l(d,z)=\frac{v\left(d\right)ef}{z}+\frac{df}{z};\end{array}$

According to above-mentioned relation formula, calculate the scope of length breadth ratio corresponding to rain line:

$AR\left(d\right)=\frac{l(d,z)}{b(d,z)}=\frac{v\left(d\right)e}{d}+1$

Wherein, the diameter of raindrop is d, and the distance of raindrop and camera is the focal length that the parameter list of z and camera shows the length l of a rain line and width b, f and represents camera, and e represents the time shutter of camera.

5. the video raindrop minimizing technology of combining form as claimed in claim 1 and fuzzy C-means clustering, is characterized in that, extracts the pixel polluted by raindrop, be specially in described pure raindrop region:

In described pure raindrop region, come to distinguish with other fast moving objects according to the rectilinearity of raindrop, finally draw the pixel polluted by raindrop.

6. the video raindrop minimizing technology of combining form as claimed in claim 1 and fuzzy C-means clustering, is characterized in that, detecting that the pixel polluted by raindrop realizes the removal of raindrop, being specially by replacing by the mixed number between raindrop and background colour:

The pixel of alternative raindrop is set to I _mix, make I _mix=α I _bcenter+ (1-α) I _rcenter, wherein α is the weight that in background colour B, number accounts for total number k.

7. the video raindrop of combining form and fuzzy C-means clustering remove a system, it is characterized in that, comprising:

Gray-scale intensity sequence units, obtains the gray-scale intensity sequence on sequence of frames of video respective pixel position in video image;

Fuzzy C-means clustering unit, according to described gray-scale intensity sequence, utilizes Fuzzy C-Means Cluster Algorithm that pixel is divided into raindrop class and background classes;

Raindrop length breadth ratio computing unit, utilizes raindrop morphological characteristic to calculate length breadth ratio corresponding to rain line in described raindrop class and background classes;

Pixel discrimination unit, according to described length breadth ratio difference raindrop district and moving target district, if the length breadth ratio of the pixel of background classes is large, be then judged as raindrop district, otherwise this pixel is moving target district;

Pure raindrop unit, the raindrop district obtained by above-mentioned initial survey deducts moving target district and obtains pure raindrop region;

Pixel extraction unit, extracts the pixel polluted by raindrop in described pure raindrop region;

By replacing by the mixed number between raindrop and background colour, raindrop removal unit, detects that the pixel polluted by raindrop realizes the removal of raindrop.