CN104881847A - Match video image enhancement method based on wavelet analysis and pseudo-color processing - Google Patents

Abstract

The present invention discloses a match video image enhancement method based on wavelet analysis and pseudo-color processing. The method comprises the steps of a, utilizing the orthogonal wavelet transformation of images to carry out the transformation processing on the football match video images; b, carrying out the football match video image enhancement processing based on the orthogonal wavelet transformation; c, carrying out the pseudo-color football match video enhancement processing on the images. The match video image enhancement method based on wavelet analysis and pseudo-color processing of the present invention can overcome the defects in the prior art of complicated operation process, long time, low reliability, etc., and realizes the advantages of simple operation process, short time and high reliability.

Description

A kind of match video image enhancing method based on wavelet analysis and Pseudo Col ored Image

Technical field

The present invention relates to technical field of video processing, particularly, relate to a kind of match video image enhancing method based on wavelet analysis and Pseudo Col ored Image.

Background technology

Conventional section of football match video image enchancing method and defect thereof

Section of football match video image enhaucament refers to the application scenario for Given Graph picture, on purpose strengthen entirety or the local characteristics of section of football match video image, original unsharp image is become clear or emphasizes some interested feature, expand the difference between different objects feature in section of football match video image, suppress uninterested feature, image and eye response characteristic are matched, improve picture quality, abundant information amount, strengthen image interpretation and recognition effect, meet the needs of some graphical analysis.

Conventional section of football match video image enchancing method can be divided into frequency domain method and space domain method.Algorithm based on frequency field in certain transform domain of section of football match video image, carries out certain to the transform coefficient values of image revise, and is that one strengthens algorithm indirectly.Frequency domain method regards a kind of 2D signal as section of football match video image, and carry out the signal Enhancement Method based on two-dimensional Fourier transform to it, its processing procedure as shown in Figure 1.Frequency domain method first requires according to concrete enhancing the correction function H (x that design is suitable, y), then Fourier transform is utilized by section of football match video image conversion to frequency field F (x, y), again certain filter correction is carried out to the frequency spectrum of section of football match video image, namely F (x is carried out, y) H (x, y) computing obtains G (x, y), finally revised image is carried out Fourier inversion to spatial domain, with this image g (x, y) that is enhanced.Adopt the noise in low pass filtering method removal figure in the process; Adopt high-pass filtering method to strengthen the high-frequency signals such as edge, make fuzzy image become clear.Frequency domain method is utilized to carry out section of football match video image enhaucament result as shown in Figure 2.

Algorithm based on spatial domain is divided into point processing algorithm and neighborhood to strengthen algorithm.Point processing algorithm and gray level correction, greyscale transformation and histogram modification etc., object makes section of football match video image imagewise uniform, or expand section of football match video dynamic range of images, expanded contrast.Neighborhood strengthens algorithm and is divided into image smoothing and sharpening two kinds.Smoothly be generally used for removal of images noise, but also easily cause the fuzzy of edge.The object of sharpening is the edge contour in order to outstanding object, is conducive to target identification.Representative spatial-domain algorithm has local averaging method and median filtering method etc., and they can be used for removing or weakening noise.Algorithms most in use has gradient method, operator, high-pass filtering, mask matching method, statistics differential technique etc.Space domain method carries out convolution algorithm operation to the pixel in image, is described below with formula:

g(x,y)＝f(x,y)*h(x,y) (4)

Wherein, f (x, y) represents original section of football match video image function; H (x, y) represents (low pass or high-pass filtering) filter impulse response space transforming function; G (x, y) represents the section of football match video image function after process.Utilize histogram modification to carry out section of football match video image enhaucament result as shown in Figure 3, utilize greyscale transformation to carry out section of football match video image enhaucament result as shown in Figure 4.

Traditional section of football match video algorithm for image enhancement is the statistic based on entire image, therefore, when calculating the conversion of entire image, low-frequency information in image, high-frequency information and noise information are converted simultaneously, thus the noise in image is enhanced while strengthening image, reduce the signal to noise ratio (S/N ratio) of image, cause the information entropy of image to decline.Be unfavorable for carrying out subsequent analysis process to interested target, do not reach the enhancing object of expection.

Realizing in process of the present invention, inventor finds at least to exist in prior art the defect such as the long and reliability of operating process complexity, spended time is low.

Summary of the invention

The object of the invention is to, for the problems referred to above, propose a kind of match video image enhancing method based on wavelet analysis and Pseudo Col ored Image, with realize operating process simple, take a short time and advantage that reliability is high.

For achieving the above object, the technical solution used in the present invention is: a kind of match video image enhancing method based on wavelet analysis and Pseudo Col ored Image, comprising:

A, the orthogonal wavelet transformation of image is utilized to carry out conversion process to section of football match video image;

B, section of football match video image enhancement processing based on orthogonal wavelet transformation;

The pseudo-colours section of football match video of c, image strengthens process.

Further, described step a, specifically comprises:

Suppose for any scale coefficient series of metric space, h ₀and h ₁be respectively low-pass filter and the Hi-pass filter coefficient of orthogonal wavelet function, these two groups of coefficients are all constant for any yardstick, and the relation between each coefficient of orthogonal wavelet transformation represents with (1):

$\left\{\begin{array}{c}{s}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\α}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\β}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_1(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\γ}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_1(m-2l)s_{k,m}^{j-1}\end{array}\right.---\left(1\right)$

The edge contour of original section of football match video image under a yardstick can be decomposed into low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts under more small scale, they are the different information of the representative original section of football match video image obtained respectively through four different wave filters, wherein be low-pass filter on space and row both direction obtains, and it corresponds to edge contour information on next yardstick; be hi-pass filter on line direction and the low-pass filter on column direction obtain, and it corresponds to detailed information in horizontal direction general picture in vertical direction; In like manner, represent detailed information in vertical direction general picture in the horizontal direction; represent to the detailed information on angular direction;

Section of football match video image, after two-dimensional orthogonal wavelets conversion is decomposed, obtains the low frequency component of image, horizontal high frequency component, vertical high frequency component and diagonal components respectively.

Further, described step b, specifically comprises:

First utilize wavelet fractal interpolation to carry out denoising to section of football match video picture signal, better removed the effect of noise;

Then use orthogonal wavelet transformation by sized by a width section of football match video picture breakdown, the position component all different with direction, can change the size of some coefficient in wavelet transformed domain before carrying out inverse transformation, amplifying decays on component interested in section of football match video image affects little component to processing result image;

Orthogonal wavelet analysis method is that a kind of spatial window and frequency window all can make adaptively changing, the empty analytical approach that frequently localizes, and two-dimentional multiple dimensioned Discrete Orthogonal Wavelets analytic definition is:

$\left\{\begin{array}{c}{S}_{j}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Φ}}_j(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{1}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^1(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{2}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^2(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{3}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^3(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\end{array}\right.---\left(2\right)$

In formula, f (x, y) is picture signal; S _jf (n, m) is the low frequency component of f (x, y); represent vertical, the diagonal sum horizontal high frequency component of f (x, y) respectively;

When formula (3) is set up, when namely high fdrequency component weight is larger, section of football match video image seems more clear:

$C_j^{\′}=\left\{\begin{array}{c}k\×C_j^H,k>1\\ C_j^L\end{array}\right.---\left(3\right)$

Wherein, for high-frequency wavelet coefficient; for low-frequency wavelet coefficients; K is high fdrequency component weight; J is the wavelet decomposition number of plies.

Further, described step c, specifically comprises:

Adopt frequency field Pseudo Col ored Image to strengthen method and Pseudo Col ored Image is carried out to section of football match video image;

Based on the section of football match video image enhaucament of wavelet analysis and Pseudo Col ored Image.

Further, described employing frequency field Pseudo Col ored Image strengthens method carries out Pseudo Col ored Image operation to section of football match video image, comprises further:

First section of football match video image through Fourier transform to frequency field, in frequency field, the wave filter of three different transmission characteristics is separated into three isolated components;

Then inverse Fourier transform is carried out to them, just obtain the monochrome image that three width represent different frequency component, then this three width image is further processed;

Finally they are added to respectively the red, green, blue display channel of color monitor as three primary colours component, realize the virtual color display of frequency field segmentation.

Further, the operation of the described section of football match video image enhaucament based on wavelet analysis and Pseudo Col ored Image, comprises further:

Step 1: use orthogonal fractal-wavelet transform process section of football match video image, obtain the image after denoising;

Step 2: utilize orthogonal wavelet transformation to be low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts more under small scale by section of football match video picture breakdown;

Step 3: amplify part interested in image, carries out Weakening treatment to information unessential in image simultaneously;

Step 4: after adopting orthogonal wavelet transformation section of football match video algorithm for image enhancement to obtain result images, be translated into frequency field, adopt the image enchancing method of Pseudo Col ored Image to process different frequency part, obtain better section of football match video image enhancement effects.

The match video image enhancing method based on wavelet analysis and Pseudo Col ored Image of various embodiments of the present invention, owing to comprising: a, utilize the orthogonal wavelet transformation of image to carry out conversion process to section of football match video image; B, section of football match video image enhancement processing based on orthogonal wavelet transformation; The pseudo-colours section of football match video of c, image strengthens process; Thus the defect that operating process in prior art is complicated, spended time is long and reliability is low can be overcome, with realize operating process simple, take a short time and advantage that reliability is high.

Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from instructions, or understand by implementing the present invention.

Below by drawings and Examples, technical scheme of the present invention is described in further detail.

Accompanying drawing explanation

Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, together with embodiments of the present invention for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:

Fig. 1 is medium frequency domain method section of football match video image enhaucament schematic diagram of the present invention;

Fig. 2 is medium frequency domain method section of football match video image enhaucament of the present invention, and (a), (b), (c) are followed successively by original image, FFT2 changing image and strengthen image;

Fig. 3 is histogram modification image enhaucament in the present invention, and (a) is original image, and (b) is the histogram of (a), and (c), for strengthening image, (d) is the histogram of (c);

Fig. 4 is grey linear transformation image enhaucament in the present invention, and (a) is original image, and (b) is for strengthening image;

Fig. 5 is wavelet decomposing schematic representation in the present invention (multi-level decomposition schematic diagram);

Fig. 6 is three layers of wavelet decomposition schematic diagram in the present invention;

Fig. 7 is orthogonal wavelet transformation section of football match video image enhaucament in the present invention, and (a) is original image, and (b) is for strengthening image;

Fig. 8 is the schematic diagram that frequency domain filter method of the present invention realizes Pseudo Col ored Image;

Fig. 9 is the image enhaucament of technical solution of the present invention put forward the methods, and the low gray level that (a) is original image strengthens image, and the high grade grey level that (b) is original image strengthens image;

Figure 10 is fuzzy section of football match video image enhaucament in the present invention, and (a) is blurred picture, and (b) is for strengthening image.

Embodiment

Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described, should be appreciated that preferred embodiment described herein is only for instruction and explanation of the present invention, is not intended to limit the present invention.

According to the embodiment of the present invention, as shown in Fig. 1-Figure 10, provide a kind of match video image enhancing method based on wavelet analysis and Pseudo Col ored Image.

Based on the section of football match video image enhaucament of wavelet analysis and Pseudo Col ored Image

2.2.2.1 the orthogonal wavelet transformation of section of football match video image

Technical solution of the present invention, in order to realize section of football match video image enhaucament, utilizes the orthogonal wavelet transformation of image to carry out conversion process to section of football match video image.

Suppose for any scale coefficient series of metric space, h ₀and h ₁be respectively low-pass filter and the Hi-pass filter coefficient of orthogonal wavelet function, these two groups of coefficients are all constant for any yardstick, and like this, the fast decoupled of orthogonal wavelet transformation can represent with Fig. 5.Relation between each coefficient of orthogonal wavelet transformation represents with (1).

$\left\{\begin{array}{c}{s}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\α}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\β}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_1(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\γ}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_1(m-2l)s_{k,m}^{j-1}\end{array}\right.---\left(1\right)$

As shown in Figure 4, the edge contour of original section of football match video image under a yardstick can be decomposed into low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts under more small scale, they are the different information of the representative original section of football match video image obtained respectively through four different wave filters, wherein be low-pass filter on space and row both direction obtains, and it corresponds to edge contour information on next yardstick; be hi-pass filter on line direction and the low-pass filter on column direction obtain, and it corresponds to detailed information in horizontal direction general picture in vertical direction; In like manner, represent detailed information in vertical direction general picture in the horizontal direction; represent to the detailed information on angular direction.Section of football match video image, after two-dimensional orthogonal wavelets conversion is decomposed, obtains the low frequency component of image, horizontal high frequency component, vertical high frequency component and diagonal components respectively.Fig. 6 is that section of football match video image is through three layers of orthogonal wavelet decomposition figure, wherein C ₃for the low frequency part of image, most of energy of image all concentrates on this region, be respectively level, vertical and diagonal components, they are all the detailed information parts of image.

2.2.2.2 based on the section of football match video image enhancement processing of orthogonal wavelet transformation

Owing to often having very strong correlativity between the wavelet coefficient of picture signal on each layer relevant position, the wavelet coefficient of noise then has weak relevant or incoherent feature.First utilize wavelet fractal interpolation to carry out denoising to section of football match video picture signal, better can be removed the effect of noise.Then use orthogonal wavelet transformation by sized by a width section of football match video picture breakdown, the position component all different with direction, can change the size of some coefficient in wavelet transformed domain before carrying out inverse transformation, so just can amplifying selectively decays on component interested in section of football match video image affects little component to processing result image.Orthogonal wavelet analysis method is that a kind of spatial window and frequency window all can make adaptively changing, the empty analytical approach that frequently localizes, and it is compared with traditional Fourier's analysis method, has better empty window features frequently.

The multiple dimensioned Discrete Orthogonal Wavelets analytic definition of two dimension is:

$\left\{\begin{array}{c}{S}_{j}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Φ}}_j(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{1}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^1(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{2}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^2(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{3}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^3(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\end{array}\right.---\left(2\right)$

In formula, f (x, y) is picture signal; S _jf (n, m) is the low frequency component of f (x, y); represent vertical, the diagonal sum horizontal high frequency component of f (x, y) respectively.

As can be seen here, after two-dimensional discrete Orthogonal wavelet analysis, section of football match video image is divided into low frequency component and vertical, diagonal sum level three high fdrequency components.Due to soft image, to be mainly manifested in three high fdrequency components less than normal, causes image detail fuzzy, therefore suitably increases by three high fdrequency components, can strengthen image detail information.So when formula (3) is set up, when namely high fdrequency component weight is larger, section of football match video image seems more clear.

$C_j^{\′}=\left\{\begin{array}{c}k\×C_j^H,k>1\\ C_j^L\end{array}\right.---\left(3\right)$

Wherein, for high-frequency wavelet coefficient; for low-frequency wavelet coefficients; K is high fdrequency component weight; J is the wavelet decomposition number of plies.Utilize experimental result that orthogonal wavelet transformation section of football match video image enchancing method obtains as shown in Figure 7.But after employing this method, owing to enhancing high fdrequency component, thus gained section of football match video image usually can be partially bright, and its contrast is poor, so also need the resolution increasing section of football match video image with Pseudo Col ored Image further.

2.2.2.3 the pseudo-colours section of football match video of image strengthens process

Technical solution of the present invention adopts frequency field Pseudo Col ored Image enhancing method to carry out Pseudo Col ored Image to section of football match video image, first section of football match video image through Fourier transform to frequency field, in frequency field, the wave filter of three different transmission characteristics is separated into three isolated components, then inverse Fourier transform is carried out to them, just the monochrome image that three width represent different frequency component is obtained, then this three width image is further processed (as histogram equalization), finally they are added to the red of color monitor respectively as three primary colours component, green, blue display channel, thus realize the virtual color display of frequency field segmentation.Its block diagram is as Fig. 8.

2.2.2.4 based on the section of football match video image enhaucament of wavelet analysis and Pseudo Col ored Image

In order to overcome the defect of the section of football match video algorithm for image enhancement based on Orthogonal wavelet analysis section of football match video algorithm for image enhancement and Pseudo Col ored Image, technical solution of the present invention proposes the section of football match video algorithm for image enhancement based on Orthogonal wavelet analysis and Pseudo Col ored Image.This algorithm both can overcome the defects such as the image of employing Orthogonal wavelet analysis section of football match video algorithm for image enhancement process is partially bright and contrast is poor, and the section of football match video algorithm for image enhancement that can overcome again Pseudo Col ored Image fully can not process the defect of some detailed information in image.The concrete implementation step of algorithm is as follows:

Step 1: use orthogonal fractal-wavelet transform process section of football match video image, obtain the image after denoising.

Step 2: utilize orthogonal wavelet transformation to be low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts more under small scale by section of football match video picture breakdown.

Step 3: in order to reach the object of section of football match video image enhaucament, amplifies part interested in image selectively, carries out Weakening treatment to information unessential in image simultaneously.

Step 4: after adopting orthogonal wavelet transformation section of football match video algorithm for image enhancement to obtain result images, be translated into frequency field, adopt the image enchancing method of Pseudo Col ored Image to process different frequency part, obtain better section of football match video image enhancement effects.

The result of the section of football match video image enchancing method process based on Orthogonal wavelet analysis and Pseudo Col ored Image utilizing technical solution of the present invention to propose as shown in Figure 9.Due to video image production and play process in may produce blurred picture, the method that technical solution of the present invention can be utilized to propose to strengthen blurred picture, as shown in Figure 10.

Experimental comparison

Technical solution of the present invention carries out the evaluation of quantitative aspect by calculating the average of section of football match video image after strengthening, information entropy and sharpness to section of football match video image enhancement effects.Formula (5) is utilized to carry out the average of computed image.

$\mathrm{mean}=\frac{1}{\mathrm{XY}}\underset{x=1}{\overset{X}{\mathrm{\Σ}}}\underset{y=1}{\overset{Y}{\mathrm{\Σ}}}{I}_e(x,y)---\left(5\right)$

Wherein, I _efor the result images after enhancing, (X, Y) is image size.For piece image, average reflects the mean flow rate of image.If average moderate (gray-scale value 128 nearby), then show that visual effect is good.

The calculating of information entropy is defined as follows:

$\mathrm{Ent}=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}{p}_i\mathrm{In}\left(p_i\right)---\left(6\right)$

Wherein, p _ifor the gray level GNL of image _icorresponding probability; L is gray level sum; I ∈ (0,1 ... L-2, L-1).Entropy is larger, reflects the quantity of information that image carries more, and therefore information entropy weighs the important indicator that image information enriches degree.

The sharpness of section of football match video image can calculate with formula (7):

Sharpness can reflect minor detail contrast in section of football match video image and texture transformation feature.Definition values is larger, illustrates that corresponding section of football match video image is more clear.The average of football match video images after 3 kinds of algorithms strengthen, information entropy and sharpness as shown in table 1.The average of each passage of RGB of football match video images after 3 kinds of algorithms strengthen, information entropy and sharpness as shown in table 2.

$\left\{\begin{array}{c}\mathrm{Def}=\frac{1}{\mathrm{XY}}\underset{x=1}{\overset{X}{\mathrm{\Σ}}}\underset{y=1}{\overset{Y}{\mathrm{\Σ}}}\sqrt{{\left(\mathrm{\Δ}{I}_x(x,y)\right)}^2+{\left(\mathrm{\Δ}{I}_y(x,y)\right)}^2}\\ \mathrm{\Δ}{I}_x(x,y)=I_e(x,y)-I_e(x-1,y)\\ \mathrm{\Δ}{I}_y(x,y)=I_e(x,y)-I_e(x,y-1)\\ \end{array}\right.---\left(7\right)$

The average of image, information entropy and sharpness after table 1 strengthens

The average of image, information entropy and sharpness after table 2 strengthens

This section introduces section of football match video Preprocessing Technique.First mean filter denoising, medium filtering denoising, the image denoising based on PDE, full variation image denoising and fractal-wavelet transform denoising five kinds of conventional image de-noising methods and defect thereof is described.Propose the Fractal Wavelet adaptive denoising algorithm based on multivariate statistical model.In denoising process, first establish a multivariate statistical model, this model can estimate various relevant information more accurately, and model parameter is improved more flexible.Then by being combined with Fractal Wavelet denoising method the image space can selecting high-quality.Nearly excellent father and son tree can be found according to piecing together distance in best subtree territory under the noise variance of appropriateness.Finally by doping muting image Fractal Wavelet coding from noise image, thus reach the object optimizing denoising.The method, while removal noise, effectively can keep edge and the textural characteristics of image, retain the fine structure of image well, achieve good denoising effect.Owing to have employed prediction Wavelet-fractal coding, optimize algorithm structure, the processing speed of algorithm is than very fast.Next describes conventional frequency field section of football match video image enhaucament and spatial domain section of football match video image enchancing method and defect thereof.Propose the section of football match video algorithm for image enhancement based on Orthogonal wavelet analysis and Pseudo Col ored Image, this algorithm both can overcome the defects such as the image of employing Orthogonal wavelet analysis section of football match video algorithm for image enhancement process is partially bright and contrast is poor, and the section of football match video algorithm for image enhancement that can overcome again Pseudo Col ored Image fully can not process the defect of some detailed information in image.

Last it is noted that the foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (6)

1., based on a match video image enhancing method for wavelet analysis and Pseudo Col ored Image, it is characterized in that, comprising:

A, the orthogonal wavelet transformation of image is utilized to carry out conversion process to section of football match video image;

B, section of football match video image enhancement processing based on orthogonal wavelet transformation;

The pseudo-colours section of football match video of c, image strengthens process.

2. the match video image enhancing method based on wavelet analysis and Pseudo Col ored Image according to claim 1, it is characterized in that, described step a, specifically comprises:

Suppose for any scale coefficient series of metric space, h ₀and h ₁be respectively low-pass filter and the Hi-pass filter coefficient of orthogonal wavelet function, these two groups of coefficients are all constant for any yardstick, and the relation between each coefficient of orthogonal wavelet transformation represents with (1):

$\left\{\begin{array}{c}{s}_{i.l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\α}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_0(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\β}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_0(k-2i)h_1(m-2l)s_{k,m}^{j-1}\\ {\mathrm{\γ}}_{i,l}^j=\underset{k,n}{\mathrm{\Σ}}{h}_1(k-2i)h_1(m-2l)s_{k,m}^{j-1}\end{array}\right.---\left(1\right)$

The edge contour of original section of football match video image under a yardstick can be decomposed into low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts under more small scale, they are the different information of the representative original section of football match video image obtained respectively through four different wave filters, wherein be low-pass filter on space and row both direction obtains, and it corresponds to edge contour information on next yardstick; be hi-pass filter on line direction and the low-pass filter on column direction obtain, and it corresponds to detailed information in horizontal direction general picture in vertical direction; In like manner, represent detailed information in vertical direction general picture in the horizontal direction; represent to the detailed information on angular direction;

Section of football match video image, after two-dimensional orthogonal wavelets conversion is decomposed, obtains the low frequency component of image, horizontal high frequency component, vertical high frequency component and diagonal components respectively.

3. the match video image enhancing method based on wavelet analysis and Pseudo Col ored Image according to claim 1 and 2, it is characterized in that, described step b, specifically comprises:

First utilize wavelet fractal interpolation to carry out denoising to section of football match video picture signal, better removed the effect of noise;

Then use orthogonal wavelet transformation by sized by a width section of football match video picture breakdown, the position component all different with direction, can change the size of some coefficient in wavelet transformed domain before carrying out inverse transformation, amplifying decays on component interested in section of football match video image affects little component to processing result image;

Orthogonal wavelet analysis method is that a kind of spatial window and frequency window all can make adaptively changing, the empty analytical approach that frequently localizes, and two-dimentional multiple dimensioned Discrete Orthogonal Wavelets analytic definition is:

$\left\{\begin{array}{c}{S}_{j}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Φ}}_j(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{1}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^1(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{2}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^2(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\\ W_j^{3}f(n,m)=\underset{R^2}{\∫\∫}f(x,y)2^{2j}{\mathrm{\Ψ}}_j^3(x-2^{-j}n,y-2^{-j}m)\mathrm{dxdy}\end{array}\right.---\left(2\right)$

In formula, f (x, y) is picture signal; S _jf (n, m) is the low frequency component of f (x, y); represent vertical, the diagonal sum horizontal high frequency component of f (x, y) respectively;

When formula (3) is set up, when namely high fdrequency component weight is larger, section of football match video image seems more clear:

$C_j^{\′}=\left\{\begin{array}{c}k\×C_j^H,k>1\\ C_j^L\end{array}\right.---\left(3\right)$

Wherein, for high-frequency wavelet coefficient; for low-frequency wavelet coefficients; K is high fdrequency component weight; J is the wavelet decomposition number of plies.

4. the match video image enhancing method based on wavelet analysis and Pseudo Col ored Image according to claim 3, it is characterized in that, described step c, specifically comprises:

Adopt frequency field Pseudo Col ored Image to strengthen method and Pseudo Col ored Image is carried out to section of football match video image;

Based on the section of football match video image enhaucament of wavelet analysis and Pseudo Col ored Image.

5. the match video image enhancing method based on wavelet analysis and Pseudo Col ored Image according to claim 4, it is characterized in that, described employing frequency field Pseudo Col ored Image strengthens method carries out Pseudo Col ored Image operation to section of football match video image, comprises further:

First section of football match video image through Fourier transform to frequency field, in frequency field, the wave filter of three different transmission characteristics is separated into three isolated components;

Then inverse Fourier transform is carried out to them, just obtain the monochrome image that three width represent different frequency component, then this three width image is further processed;

Finally they are added to respectively the red, green, blue display channel of color monitor as three primary colours component, realize the virtual color display of frequency field segmentation.

6. the match video image enhancing method based on wavelet analysis and Pseudo Col ored Image according to claim 4, is characterized in that, the operation of the described section of football match video image enhaucament based on wavelet analysis and Pseudo Col ored Image, comprises further:

Step 1: use orthogonal fractal-wavelet transform process section of football match video image, obtain the image after denoising;

Step 2: utilize orthogonal wavelet transformation to be low frequency component, horizontal high frequency component, vertical high frequency component and diagonal components four parts more under small scale by section of football match video picture breakdown;

Step 3: amplify part interested in image, carries out Weakening treatment to information unessential in image simultaneously;

Step 4: after adopting orthogonal wavelet transformation section of football match video algorithm for image enhancement to obtain result images, be translated into frequency field, adopt the image enchancing method of Pseudo Col ored Image to process different frequency part, obtain better section of football match video image enhancement effects.