CN102629323B - Color image binarization and hiding and recovery method of binarization color image - Google Patents

Abstract

A binarization method of a color image comprises the following steps: (1) selecting a source image, acquiring chromatic values of red, green and blue three components of each pixel point of the source image and calculating a brightness value of the source image; (2) calculating an average brightness of the source image; (3) using a Zadeh-X transformation method to carry out binarization transformation on the red, green and blue three components of the source image so as to generate the transformed color image, and calculating the average brightness of the transformed color image; (4) comparing the average brightness of the source image with the average brightness of the transformed color image and determining a result image of the binarization transformation. The method of the invention has the following advantages that: binarization processing can be performed based on the color image so as to obtain the actual binarization color image; and compared to the source image, the binarization color image has little change in brightness.

Description

Hiding and restoration methods of the binaryzation of coloured image and binaryzation coloured image

Technical field

The invention belongs to technical field of image processing, is to make coloured image in the situation that the constant method of coloured image being carried out to binary conversion treatment of brightness.

Background technology

Binary image refers to the image that two gray/color values are only arranged.Image binaryzation refers to that the image conversion that gray/color is abundant becomes to only have the process of two gray/color values.Because binary image is simple, data volume is little, can highlight the profile of interested target, therefore occupy very important status in Digital Image Processing, be widely used in every field, as: optical character recognition technology OCR (optical character recognition), the processing again of document etc.

And the world is riot of color, so most of image that people obtain is all colored.The coloured image binaryzation has two kinds of thinkings always: the first is to carry out binary conversion treatment after the coloured image gray processing again, and the second is directly based on coloured image, to carry out.Because the color image information amount is large, how the coloured image of binaryzation defines the also unified saying of neither one, so the second way does not obtain good effect.The technology of main flow remains the employing first method at present, but carrying out the image that binary conversion treatment obtains after gray processing is black white image again, it is not traditional coloured image, and gray processing diminishes, after gray processing, the quality of image has also determined that next step carries out the effect of binary conversion treatment.Up till now, there is no the technological means of the second thinking, can't realize the directly binary conversion treatment based on coloured image.Because binary image often contains crucial information, need in communication, hide and Recovery processing, for hiding of binaryzation coloured image, be therefore also the focus of association area researchist research.

Summary of the invention

One of purpose of the present invention is to propose a kind of directly coloured image binarization method based on coloured image, and the binary image that adopts the method to obtain remains coloured image, and after conversion, the brightness of image is constant.

The binaryzation coloured image related in the present invention refers to that three color components of red, green, blue all have the coloured image of two identical chromatic values.Fig. 7 is a) a width binaryzation coloured image, Fig. 7 b) be Fig. 7 colourity spectrum a).

To achieve the object of the present invention, the present invention proposes a kind of binarization method of coloured image, and its key is to comprise the following steps:

(1) step 1, selecting coloured image is source images Si, obtains chromatic value R (x, y), G (x, y), the B (x, y) of three kinds of components of red, green, blue of each pixel of source images, and calculates the brightness value Gray (x, y) of source images;

In practical application, brightness value is equal to gray-scale value.

(2) step 2, the mean flow rate AGSi of calculating source images, described mean flow rate AGSi can be obtained by following formula:

$\mathrm{AGSi}=\frac{1}{M\×N}\underset{y=0}{\overset{N-1}{\mathrm{\Σ}}}\underset{x=0}{\overset{M-1}{\mathrm{\Σ}}}\mathrm{Gray}(x,y)$

In above formula, the brightness value that Gray (x, y) is source image pixels point (x, y), M, N are respectively the pixel count of source images in X, Y-direction;

(3) step 3, utilize the Zadeh-X transform method to carry out the binaryzation conversion to three components of red, green, blue of source images, the mean flow rate AGDi of coloured image Di after generating the coloured image Di after converting and calculating this conversion;

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is: $T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>k\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; T (k, x, y) means the chromatic value of three kinds of components of red, green, blue of the rear coloured image Di of conversion; Delta=1, the integer that k is 1～255, the integer that Theta is 0～255;

The value of different k, can obtain the wide Binary color image image of different colourity bands of a spectrum, and when getting k=255, the coloured image Di after the conversion of generation has maximum bandwidth, can obtain like this bianry image of best quality.

In the situation that k is definite value, in the value of 0～255 scope interscan Theta, Theta often gets a value, the mean flow rate AGDi of coloured image Di after a corresponding conversion;

(4) step 4, determine the result images of protecting the conversion of brightness binaryzation, and its concrete grammar is as follows:

The mean flow rate AGDi of coloured image Di and the mean flow rate AGSi of source images after each conversion are compared, the mean flow rate AGDi of coloured image Di after the immediate conversion of mean flow rate AGSi of search and source images, and definite Theta value corresponding to this AGDi, coloured image Di after the corresponding conversion of this Theta, for protecting the result images of brightness binaryzation conversion, i.e. binaryzation coloured image.

Adopt technical scheme of the present invention to process the image obtained and remain coloured image, three color components of the red, green, blue of this coloured image all have two identical chromatic values.With source images, compare, the coloured image brightness after binaryzation remains unchanged substantially, extremely approaches source images therefore to people's visual effect.

Carry out again the binaryzation conversion after the coloured image gray processing, make the color level of image from 2 ²⁴dropped to 2, and the conversion of guarantor's brightness binaryzation of coloured image makes the color level of image from 2 ²⁴dropped to 2 ³, having retained colored the time and also improved contrast, key message is preserved, and has increased a kind of new picture format: 2 ³the position bitmap format.

When scanning Theta, can adopt the method for exhaustion, scan each Theta value, obtain coloured image Di after 256 conversion, corresponding 256 mean flow rate AGDi, wherein the binaryzation coloured image corresponding with the AGDi of source images mean flow rate AGSi difference minimum be considered as meeting the binaryzation coloured image of protecting luminance transformation.

In order to accelerate the sweep velocity of Theta, in described step 3 and step 4, can scan in the following order Theta: take 127 as starting point, from the off by increase at every turn 1 or 1 the mode of subtracting change, every variation once, after the described conversion of acquisition correspondence, the mean flow rate AGDi of coloured image Di, compare the mean flow rate AGSi of this AGDi and source images, determines the Theta direction of scanning that approaches source images mean flow rate AGSi.

Concrete operations are:

The threshold value A GT of suitable luminance difference of image after a given source images and conversion makes after conversion the mean flow rate AGDi of coloured image satisfied with the difference P of the mean flow rate AGSi of source images:

P＝|AGDi-AGSi|≤AGT

To a scanning direction time, P is greater than AGT, and P increases along with the variation of Theta, abandons the scanning of this direction.

Two of purpose of the present invention is to provide a kind of hiding and restoration methods that the binaryzation coloured image is carried out, can hide and recover the binaryzation coloured image generated.

For achieving the above object, the binaryzation coloured image that the present invention is directed to above-mentioned generation proposes a kind of hiding and restoration methods, and its key is to carry out according to the following steps:

Step 1, selecting the binaryzation coloured image is source images, obtain the chromatic value R (x of three kinds of components of red, green, blue of each pixel of source images, y), G (x, y), B (x, y), and determine that the maximum of three kinds of component chromatic values of red, green, blue of described source images is b, reckling is a;

Step 2, utilize the Zadeh-X transform method respectively to chromatic value R (x, y), the G (x of three kinds of components of red, green, blue of each pixel of source images, y), B (x, y) converted, realization is hidden source images, obtains and hides the rear unresolvable coloured image of human vision;

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is:

$T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>k\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; The chromatic value of three kinds of components of red, green, blue of binaryzation coloured image after T (k, x, y) means to hide; The transformation parameter that Delat, k, Theta are round numbers, Delta=1,0<k≤10, a≤Theta<b;

In described Zadeh-X conversion, the transformation parameter of each pixel is identical;

Step 3, utilize described Zadeh-X transfer pair to hide rear coloured image and recovered, and obtains the image after recovering, and the image after this recovery is identical with source images;

When carrying out described Zadeh-X conversion, transformation parameter value respectively is: Delta=1, k=255, the left side dividing value of image after Theta=hides.

Remarkable result of the present invention is: a kind of method that coloured image binarization method based on coloured image is provided and the binary image generated is hidden and recovered, brightness and the source images of the binaryzation coloured image that utilizes this binarization method to generate are very approaching, and be coloured image truly, also realized hiding and excavating this binaryzation coloured image on the basis of Zadeh-X conversion.

the accompanying drawing explanation

Fig. 1 is the source images in embodiment 1, comprises two width source images (a), (b), and wherein the mean flow rate of (a) is 110.0884, and mean flow rate (b) is 49.0451;

Fig. 2 carries out the result images after binary conversion treatment to Fig. 1, the result images that wherein (a) is Fig. 1 (a), and mean flow rate is 110.7047; (b) be the result images of Fig. 1 (b), mean flow rate is 49.5710;

Fig. 3 is the process flow diagram of embodiment 1;

Fig. 4 (a) is the image after Fig. 2 (a) is hidden;

Fig. 4 (b) is the image after Fig. 2 (b) is hidden;

Fig. 5 (a) is the image after Fig. 4 (a) is recovered;

Fig. 5 (b) is the image after Fig. 4 (b) is recovered;

Fig. 6 is the process flow diagram of embodiment 2;

Fig. 7 is the colourity spectrum of binaryzation coloured image and this binary image, wherein, a) is a width binaryzation coloured image, is b) the colourity spectrum of three color components of red, green, blue a) from top to bottom successively.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.

Embodiment 1: a kind of binarization method of coloured image, comprise the following steps, as shown in Figure 3:

(1) step 1, selecting two width coloured images shown in Fig. 1 is source images Si, obtains the chromatic value R (x of three kinds of components of red, green, blue of each pixel of source images, y), G (x, y), B (x, y), and the brightness value Gray (x, y) of calculating source images;

(2) step 2, the mean flow rate AGSi of calculating source images, described mean flow rate AGSi can be obtained by following formula:

$\mathrm{AGSi}=\frac{1}{M\&times;N}\underset{y=0}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{x=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\mathrm{Gray}(x,y)$

In above formula, the brightness value that Gray (x, y) is source image pixels point (x, y), M, N are respectively the pixel count of source images in X, Y-direction;

(3) step 3, utilize the Zadeh-X transform method to carry out the binaryzation conversion to three components of red, green, blue of source images, the mean flow rate AGDi of coloured image Di after generating the coloured image Di after converting and calculating this conversion;

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is:

$T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000671485900051.png"\; state="\{\{state\}\}">k</figure-callout>}\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; T (k, x, y) means the chromatic value of three kinds of components of red, green, blue of the rear coloured image Di of conversion; Delta=1, the integer that k is 1～255, the integer that Theta is 0～255;

The value of different k, can obtain the wide Binary color image image of different colourity bands of a spectrum, and when getting k=255, the coloured image Di after the conversion of generation has maximum bandwidth, and the visual quality of the binary image of gained is best.K=255 in the present embodiment.

In the situation that k is definite value, in the value of 0～255 scope interscan Theta, Theta often gets a value, the mean flow rate AGDi of coloured image Di after a corresponding conversion;

(4) step 4, determine the result images of protecting the conversion of brightness binaryzation, and its concrete grammar is as follows:

The mean flow rate AGDi of coloured image Di and the mean flow rate AGSi of source images after each conversion are compared, the mean flow rate AGDi of coloured image Di after the immediate conversion of mean flow rate AGSi of search and source images, and definite Theta value corresponding to this AGDi, coloured image Di after the corresponding conversion of this Theta, for protecting the result images of brightness binaryzation conversion, i.e. binaryzation coloured image.

In order to accelerate the sweep velocity of Theta, in described step 3 and step 4, can scan in the following order Theta: take 127 as starting point, from the off by increase at every turn 1 or 1 the mode of subtracting change, every variation once, after the described conversion of acquisition correspondence, the mean flow rate AGDi of coloured image Di, compare the mean flow rate AGSi of this AGDi and source images, determines the Theta direction of scanning that approaches source images mean flow rate AGSi.

Concrete operations are:

The threshold value A GT of suitable luminance difference of image after a given source images and conversion makes after conversion the mean flow rate AGDi of coloured image satisfied with the difference P of the mean flow rate AGSi of source images:

P＝|AGDi-AGSi|≤AGT

To a scanning direction time, P is greater than AGT, and P increases along with the variation of Theta, abandons the scanning of this direction.

The result images of the two width source images of Fig. 1 after the binaryzation conversion is shown in Fig. 2, and the Theta value of their correspondences is respectively 121,92.With source images, compare, the variation of the coloured image brightness after binaryzation all, in 1 brightness level range, extremely approaches source images therefore to people's visual effect.

Embodiment 2: hiding and restoration methods of a kind of binaryzation coloured image, as shown in Figure 6, carry out according to the following steps:

Step 1, select Fig. 2 (a) and (b) shown in the binaryzation coloured image be source images, obtain the chromatic value R (x of three kinds of components of red, green, blue of each pixel of source images, y), G (x, y), B (x, y), and determine that the maximum of three kinds of component chromatic values of red, green, blue of described source images is b, reckling is a; The a=0 of Fig. 2 (a) and Fig. 2 (b), b=255.

Step 2, utilize the Zadeh-X transform method respectively to chromatic value R (x, y), the G (x of three kinds of components of red, green, blue of each pixel of source images, y), B (x, y) converted, realization is hidden source images, obtains and hides the rear invisible coloured image of human vision; Image after hiding is as shown in Fig. 4 (a) and Fig. 4 (b).

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is:

$T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>\mathrm{<figure-callout\; id="0"\; label="k"\; filenames="HSA00000671485900051.png"\; state="\{\{state\}\}">k</figure-callout>}\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; The chromatic value of three kinds of components of red, green, blue of binaryzation coloured image after T (k, x, y) means to hide; The transformation parameter that Delat, k, Theta are round numbers, Delta=1,0<k≤10, a≤Theta<b;

In the present embodiment, the transformation parameter k=1 of Fig. 4 (a) and Fig. 4 (b), Theta=0.

In described Zadeh-X conversion, the transformation parameter of each pixel is identical;

Step 3, utilize described Zadeh-X transfer pair to hide rear coloured image and recovered, and obtains the image after recovering, and as shown in Fig. 5 (a) and Fig. 5 (b), the image after this recovery is identical with source images;

When carrying out described Zadeh-X conversion, transformation parameter value respectively is: Delta=1, k=255, the left side dividing value of image, i.e. Theta=0 after Theta=hides.

Claims (2)

1. the binarization method of a coloured image, is characterized in that comprising the following steps:

(1) step 1, selecting coloured image is source images Si, obtains chromatic value R (x, y), G (x, y), the B (x, y) of three kinds of components of red, green, blue of each pixel of source images, and calculates the brightness value Gray (x, y) of source images;

(2) step 2, the mean flow rate AGSi of calculating source images, described mean flow rate AGSi can be obtained by following formula:

$\mathrm{AGSi}=\frac{1}{M\&times;N}\underset{y=0}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{x=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\mathrm{Gray}(x,y)$

In above formula, the brightness value that Gray (x, y) is source image pixels point (x, y), M, N are respectively the pixel count of source images in X, Y-direction;

(3) step 3, utilize the Zadeh-X transform method to carry out the binaryzation conversion to three components of red, green, blue of source images, the mean flow rate AGDi of coloured image Di after generating the coloured image Di after converting and calculating this conversion;

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is:

$T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>k\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; T (k, x, y) means the chromatic value of three kinds of components of red, green, blue of the rear coloured image Di of conversion; Delta=1, the integer that k is 1～255, the integer that Theta is 0～255;

In the situation that k is definite value, in the value of 0～255 scope interscan Theta, Theta often gets a value, the mean flow rate AGDi of coloured image Di after a corresponding conversion;

(4) step 4, determine the result images of protecting the conversion of brightness binaryzation, and its concrete grammar is as follows:

The mean flow rate AGDi of coloured image Di and the mean flow rate AGSi of source images after each conversion are compared, the mean flow rate AGDi of coloured image Di after the immediate conversion of mean flow rate AGSi of search and source images, and definite Theta value corresponding to this AGDi, coloured image Di after the corresponding conversion of this Theta, for protecting the result images of brightness binaryzation conversion, i.e. binaryzation coloured image;

In described step 3 and step 4, can scan in the following order Theta: take 127 as starting point, from the off by increase at every turn 1 or 1 the mode of subtracting change, every variation once, the mean flow rate AGDi of coloured image Di after the described conversion of acquisition correspondence, the mean flow rate AGSi of this AGDi and source images is compared, determine the Theta direction of scanning that approaches source images mean flow rate AGSi;

Concrete operations are:

The threshold value A GT of suitable luminance difference of image after a given source images and conversion makes after conversion the mean flow rate AGDi of coloured image satisfied with the difference P of the mean flow rate AGSi of source images:

P＝|AGDi-AGSi|≤AGT

To a scanning direction time, P is greater than AGT, and P increases along with the variation of Theta, abandons the scanning of this direction.

2. one kind to the hiding and restoration methods of binaryzation coloured image claimed in claim 1, it is characterized in that carrying out according to the following steps:

Step 1, selecting the binaryzation coloured image is source images, obtain the chromatic value R (x of three kinds of components of red, green, blue of each pixel of source images, y), G (x, y), B (x, y), and determine that the maximum of three kinds of component chromatic values of red, green, blue of described source images is b, reckling is a;

Step 2, utilize the Zadeh-X transform method respectively to chromatic value R (x, y), the G (x of three kinds of components of red, green, blue of each pixel of source images, y), B (x, y) converted, realization is hidden source images, obtains and hides the rear unresolvable coloured image of human vision;

Described Zadeh-X transform method is as follows:

$T(k,x,y)=k\frac{O(x,y)-\mathrm{Theta}}{\mathrm{Delta}}$

Constraint condition is:

$T(k,x,y)=\left\{\begin{array}{cc}k,& T(k,x,y)>k\\ 0,& T(k,x,y)<0\end{array}\right.$

In above formula, O (x, y) means the chromatic value of three kinds of components of red, green, blue of source images; The chromatic value of three kinds of components of red, green, blue of binaryzation coloured image after T (k, x, y) means to hide; The transformation parameter that Delat, k, Theta are round numbers, Delta=1,0<k≤10, a≤Theta<b;

In described Zadeh-X conversion, the transformation parameter of each pixel is identical;

Step 3, utilize described Zadeh-X transfer pair to hide rear coloured image and recovered, and obtains the image after recovering, and the image after this recovery is identical with source images;

When carrying out described Zadeh-X conversion, transformation parameter value respectively is: Delta=1, k=255, the left side dividing value of image after Theta=hides.

Images