CN102186089A - Simple-component video image rain field removing method - Google Patents

Abstract

The invention discloses a simple-component video image rain field removing method, which is characterized by comprising the following steps of: first converting an image into a YCbCr color space in combination with color space conversion according to the color attributes of raindrops; then extracting a Y component, and performing raindrop removal on the Y component by using the conventional rain field removing method; and finally converting the image into a red, green and blue (RGB) color space by combining Cb and Cr components. By the method, simple-component rain field removal is realized; and compared with the conventional three-component rain removing method, the invention saves the processing of two components and improves the real-time performance.

Description

Simple component video image rain field removal method

Technical field

The present invention relates to the image rain field removal method in the fields such as a kind of computer vision, belong to technical field of image processing.

Background technology

Computer vision system has obtained extensive use in military field.Yet but there is a critical problem-very responsive to bad weather in vision system.Rainy weather tends to cause atmospheric visibility to reduce, and the image quality of vision system descends, and brings huge difficulty to the monitoring that depends on vision system, navigational guidance, target following, recognition system etc.Recovery rainy weather hypograph has great significance to all weather operations of vision system.

Find that through literature search the patent of invention of removing about video image rain field does not have only domestic and international small part scholar to go the rain technology to carry out preliminary research to image to prior art.

Garg etc. (Garg et al., IEEE Conference on Computer Visionand Pattern Recognition 2004:528-535) at first utilizes raindrop optical model Preliminary detection raindrop; Secondly, utilize raindrop pixel grey scale changing value and background gray levels to present this constraints of linear relationship, remove the raindrop of flase drop in the first step; Then, based on there being these characteristics of stronger space time correlation once more raindrop to be discerned on the raindrop direction of motion; At last, utilize averaging method to remove rain field in the image.This method has obtained certain effect, but needs 31 two field pictures, and real-time is relatively poor.

(Zhang et al., IEEE International Conference onMultimedia and Expo 2006:461-464) at first carry out K-means cluster (K=2) to detect raindrop to each pixel along time-axis direction to Zhang etc.; Then, the raindrop pixel value of detection uses background color to replace, and realizes that image removes rain.This method detects raindrop need utilize all videos frame, and real-time is relatively poor.

Brewer etc. (Brewer et al., Lecture Notes in Computer Science, 2008,5342:451-458) at first utilize the optical model Preliminary detection raindrop of Garg; Then, utilize the length-width ratio of raindrop and the raindrop that direction constrain is removed the previous step flase drop; At last, utilizing averaging method to carry out raindrop equally removes.Only need three two field pictures when this method is handled, real-time is better.

Present existing image rain field removal method all is to carry out at rgb color space, realizes that image goes rain need remove rain field on R, G, three components of B respectively.If algorithm itself is consuming time more, the processing of three components can strengthen time cost.

Summary of the invention

Technical problem to be solved by this invention provides a kind of real-time video image rain field removal method.

For solving the problems of the technologies described above, the present invention takes following technical scheme to realize:

A kind of simple component video image rain field removal method is characterized in that, may further comprise the steps:

(1) according to the color attribute of raindrop, transforms, image is transformed into the YCbCr color space in conjunction with color space;

(2) extract the Y component, use existing rain field removal method that it is carried out raindrop and remove;

(3) in conjunction with two components of untreated Cb, Cr, image is transformed into rgb color space.

Aforesaid simple component video image rain field removal method is characterized in that: in described step (1), it mainly is that the interior light of Wide-angle scope causes because refraction action has converged more that the brightness of raindrop is higher than background luminance.Raindrop are to the incidence angle θ of red, green, blue _R, θ _G, θ _BApproximately equal, and intensity variations directly determines the brightness of pixel to change.Therefore, the Strength Changes amount Δ R of the R that refraction action causes, G, three components of B, Δ G, Δ B also answer approximately equal, the color attribute of Here it is raindrop.

It is as follows that rgb color space is converted into the formula of color space of YCbCr:

$\left[\begin{array}{c}Y\\ \mathrm{Cb}\\ \mathrm{Cr}\end{array}\right]=\left[\begin{array}{c}16\\ 128\\ 128\end{array}\right]+\left[\begin{array}{ccc}0.2568& 0.5041& 0.0980\\ -0.1482& -0.2910& 0.4392\\ 0.4392& -0.3678& -0.0714\end{array}\right]\left[\begin{array}{c}R\\ G\\ B\end{array}\right]---\left(1\right)$

For the rainy video image of a frame, the intensity level of each pixel on R, G, B component can be thought and be made up of two parts, a part is the background intensity value that is not covered by raindrop, another part be since the raindrop that cause of refraction action with respect to the variable quantity of background.Extract Cb and the Cr component in the formula (1) and be rewritten as following form:

Cb＝128-0.1482(R _bg+ΔR)-0.2910(G _bg+ΔG)+0.4392(B _bg+ΔB) (2)

Cr＝128+0.4392(R _bg+ΔR)-0.3678(G _bg+ΔG)-0.0714(B _bg+ΔB)

Wherein, R _Bg, G _Bg, B _BgBe respectively the background intensity value that three color components of red, green, blue are not covered by raindrop, Δ R, Δ G, Δ B are the variable quantity of background intensity on R, G, B component that raindrop cause.If pixel do not cover by raindrop, Δ R then, Δ G, Δ B are zero; Otherwise, all non-vanishing.In conjunction with the color attribute of raindrop, Δ R, Δ G, Δ B approximately equal, again because of Δ R, the coefficient sum before the Δ G, Δ B component is zero, so Cb and Cr component are not influenced by raindrop.Formula (2) can be written as form as follows:

Cb＝128-0.1482R _bg-0.2910G _bg+0.4392B _bg

Cr＝128+0.4392R _bg-0.3678G _bg-0.0714B _bg

(3)

After being transformed into the YCbCr color space, video image only Y component contains raindrop, Cb and two automatic cancellations of component of Cr the Strength Changes parts that cause of raindrop, make it not be subjected to the influence of raindrop.Therefore, remove video image rain field only needs to remove Y component moderate rain field, can reach the purpose of removing whole video image rain field.This also is theory analysis and demonstration that simple component removes rain.

Aforesaid simple component video image rain field removal method is characterized in that: in described step (2), extract the Y component, use existing rain field removal method to carry out raindrop to it and remove.

Aforesaid simple component video image rain field removal method is characterized in that: in described step (3), go Y component behind the rain in conjunction with two components of Cb, Cr, utilize formula (1) reverse conversion to return rgb color space, can remove the rain field in the coloured image.

So far, simple component video image rain field removal process is finished.

Simple component video image rain of the present invention field removal method, at first the color attribute color combining space according to raindrop transforms, and image is converted into the YCbCr color space; Extract the Y component then, use the existing raindrop that go among the rain method removal Y; In conjunction with untreated Cb, Cr component image is transformed back rgb color space at last, make the existing rain method of going be converted into the simple component processing, improved real-time from the three-component processing.The present invention transforms according to the color attribute color combining space of raindrop, and it is that simple component is handled that method improvement is removed in existing three-component rain field, has realized the removal of image moderate rain field apace, to heavy rain, less than, dynamically and various scenes such as static state all be suitable for.

Description of drawings

Fig. 1 removes algorithm flow chart for simple component video image rain of the present invention field;

Fig. 2 is the refraction analysis diagrams of raindrop to red, green, blue;

Embodiment

The present invention is described in further detail below in conjunction with embodiment.

With reference to Fig. 1, simple component video image rain field is removed and be may further comprise the steps:

The first step according to the color attribute of raindrop, transforms in conjunction with color space, and image is transformed into the YCbCr color space;

Second step, extract the Y component, use existing rain field removal method that it is carried out raindrop and remove;

The 3rd step in conjunction with two components of untreated Cb, Cr, was transformed into rgb color space with image, finished the rain field and removed.

With reference to Fig. 2, blue light compares ruddiness and green glow has the bigger refraction angle of visual field, and this is that wavelength is long more owing to the refractive index difference of raindrop to ruddiness, green glow, blue light causes, and refractive index is more little.But raindrop are very little to the refractive index difference of red, green, blue, can regard approximately equal as.It mainly is that the interior light of Wide-angle scope causes because refraction action has converged more that the brightness of raindrop is higher than background luminance, and intensity variations directly determines the brightness of pixel to change.Therefore, R, the G, the Strength Changes amount Δ R of three components of B, Δ G, the Δ B that cause of refraction action also answers approximately equal.

In sum, the present invention transforms, the existing rain method of going is converted into the simple component processing from the three-component processing according to the color attribute color combining space of raindrop, has reduced the port number of handling, and has improved real-time.

Above-mentioned embodiment does not limit technical scheme of the present invention in any form, and the technical scheme that mode obtained that every employing is equal to replacement or equivalent transformation all drops on protection scope of the present invention.

Claims (4)

1. a simple component video image rain field removal method is characterized in that, may further comprise the steps:

(1) according to the color attribute of raindrop, transforms, image is converted into the YCbCr color space in conjunction with color space;

(2) extract the Y component, use existing rain field removal method that it is carried out raindrop and remove;

(3) in conjunction with two components of untreated Cb, Cr, image is transformed into rgb color space.

2. simple component video image rain according to claim 1 field removal method is characterized in that: in described step (1), specifically may further comprise the steps:

11) color attribute of described raindrop is R, G, Strength Changes amount Δ R, the Δ G of three components of B, the Δ B approximately equal that refraction action causes; By formula (1) is converted into rgb color space the color space of YCbCr:

$\left[\begin{array}{c}Y\\ \mathrm{Cb}\\ \mathrm{Cr}\end{array}\right]=\left[\begin{array}{c}16\\ 128\\ 128\end{array}\right]+\left[\begin{array}{ccc}0.2568& 0.5041& 0.0980\\ -0.1482& -0.2910& 0.4392\\ 0.4392& -0.3678& -0.0714\end{array}\right]\left[\begin{array}{c}R\\ G\\ B\end{array}\right]---\left(1\right)$

12) extract Cb and the Cr component in the formula (1) and be rewritten as following form:

Cb＝128-0.1482(R _bg+ΔR)-0.2910(G _bg+ΔG)+0.4392(B _bg+ΔB) (2)

Cr＝128+0.4392(R _bg+ΔR)-0.3678(G _bg+ΔG)-0.0714(B _bg+ΔB)

Wherein, R _Bg, G _Bg, B _BgBe respectively the background intensity value that three color components of red, green, blue are not covered by raindrop, Δ R, Δ G, Δ B are the variable quantity of background intensity on R, G, B component that raindrop cause;

13) if pixel is not covered by raindrop, Δ R then, Δ G, Δ B is zero, otherwise, all non-vanishing, in conjunction with the color attribute of raindrop, Δ R, Δ G, Δ B approximately equal, again because of Δ R, Δ G, coefficient sum before the Δ B component is zero, so Cb and Cr component are not influenced by raindrop, formula (2) is written as form as follows:

Cb＝128-0.1482R _bg-0.2910G _bg+0.4392B _bg

Cr＝128+0.4392R _bg-0.3678G _bg-0.0714B _bg

(3)。

3. simple component video image rain according to claim 1 field removal method is characterized in that: in described step (2), extract the Y component, use existing rain field removal method to carry out raindrop to it and remove.

4. simple component video image rain according to claim 1 field removal method, it is characterized in that: in described step (3), go Y component behind the rain in conjunction with two components of Cb, Cr, utilize formula (1) reverse conversion to return rgb color space, can remove the rain field in the coloured image.

Images