CN115330607B - Saturation enhancement method for color image - Google Patents

Abstract

The saturation enhancement method of color images converts the image from RGB color space to HSL color space and L*a*b* color space respectively; calculates the perceived brightness L ^** of the image in L*a*b* color space; The image in the HSL color space is saturated to obtain the enhanced saturation Based on the Helmholtz‑Kohlrausch effect, perform brightness correction on the saturation-enhanced image; calculate the perceived brightness of the brightness-corrected image. when Record the corresponding corrected brightness In HSL color space, with enhanced saturation Replaces the original saturation S with the recorded corrected brightness Instead of the original brightness L, the resulting image is converted to RGB color space. After enhancing the saturation, the present invention can keep the perceived brightness unchanged.

Description

Saturation enhancement method for color images

Technical field

The invention belongs to the technical field of image processing, and particularly relates to a saturation enhancement method for color images.

Background technique

For images with insufficient saturation, it is necessary to enhance the saturation through image processing. Color consists of hue, chroma and brightness. In color spaces such as CIE 1976L*a*b* and HSL, brightness, chroma and hue can be defined. Existing saturation enhancement methods enhance saturation while maintaining brightness and hue in the color space. However, according to the Helmholtz-Kohlrausch (H-K) effect, when the saturation of an image is enhanced, the perceived brightness increases although the brightness and hue in the color space remain unchanged. Therefore, existing methods cannot enhance saturation while keeping perceived brightness unchanged.

Contents of the invention

In order to overcome the above shortcomings of the prior art, the purpose of the present invention is to provide a saturation enhancement method for color images based on the Helmholtz-Kohlrausch effect, which keeps the perceived brightness unchanged after enhancing the saturation.

In order to achieve the above objects, the technical solution adopted by the present invention is:

The saturation enhancement method of color images is characterized by including the following steps:

Step 1, convert the image from RGB color space to HSL color space and L*a*b* color space respectively;

Step 2: Calculate the perceived brightness L ^** of the image in the L*a*b* color space;

Step 3: Perform saturation enhancement on the image in the HSL color space to obtain the enhanced saturation

Step 4: Based on the Helmholtz-Kohlrausch effect, perform brightness correction on the saturation-enhanced image;

Step 5: Calculate the perceived brightness of the brightness-corrected image When/> Record the corresponding corrected brightness

Step 6, in HSL color space to enhance the saturation Replaces the original saturation S with the recorded corrected brightness Instead of the original brightness L, the resulting image is converted to RGB color space.

In one embodiment, in step 2, the calculation formula of L ^** is:

L ^** ＝L ^* +(2.5-0.025L ^* )q(h)C ^*

Among them, L ^* , C ^* , h respectively represent the brightness (lightness), chroma (chroma) and hue (hue) in the L*a*b* color space, and q (h) represents the effect of colors of different hues on perceived brightness. Impact value:

a* and b* are chromaticity coordinates, indicating the position of the color in the L*a*b* color space.

In one embodiment, step 3, enhanced saturation The calculation formula is:

In the formula, f(S,C ^* ) is the saturation enhancement function, S is the saturation in the HSL color space, γ(C ^* ) is the sigmoid function, and C ^* represents the chromaticity in the L*a*b* color space .

In one embodiment, the value range of γ(C ^* ) is [1, γ ₀ ], and its calculation formula is:

α, β, γ ₀ are the parameters that control γ(C ^* ). The parameter β determines the slope of the sigmoid function. It is a positive real number. The parameter α is related to the chromaticity C ^* . When C ^* is smaller than α enough, γ (C ^* ) is approximately equal to 1, and the value of γ ₀ is a constant.

In one embodiment, the step 4, the brightness correction step is as follows:

Step (1) Set the maximum value of brightness L _top and the minimum value of brightness L _bottom to 1 and 0 respectively, and go to step (2);

Step (2) Calculate the average brightness value L _middle = (L _top + L _bottom )/2, and go to step (3);

Step (3) When the conditions are met When, set L _middle to correct the brightness and end; otherwise, go to (4); where ε is the precision of the setting, L ^** (H, S, L) is the original perceived brightness of the image in the HSL color space, /> Think/> The perceived brightness when replacing the original saturation S and replacing the original brightness L with L _middle ;

Step (4) When the conditions are met When, set L _middle to L _top ; otherwise, set L _middle to L _bottom , and go to step (2).

In one embodiment, the and L ^** (H,S,L) are the values in the HSL color space respectively and (H, S, L) are converted to RGB color space, then converted from RGB color space to L*a*b* color space, and calculated in L*a*b* color space.

Compared with the prior art, the beneficial effects of the present invention are:

When the existing technology enhances image saturation, while enhancing the saturation, the brightness perceived by humans will also be enhanced. The saturation enhancement method of the present invention can ensure that the perceived brightness remains unchanged while enhancing saturation.

Description of drawings

Figure 1 is a relationship diagram between hue h and q(h).

Figure 2 is a flow chart of the method of the present invention.

Figure 3 is a schematic diagram of a test result of the present invention; (a) is the original image, (b) is the image after saturation enhancement by the existing method, and (c) is the image after saturation enhancement of the present invention.

Figure 4 is a schematic diagram of another test result of the present invention; (a) is the original image, (b) is the saturation-enhanced image using the existing method, and (c) is the saturation-enhanced image of the present invention.

Detailed ways

The embodiments of the present invention will be described in detail below with reference to the drawings and examples.

In the present invention, a color image saturation enhancement method based on the Helmholtz-Kohlrausch effect is proposed.

The Helmholtz-Kohlrausch (H-K) effect is a type of human visual illusion. When the brightness and hue of a certain color remain unchanged and only the saturation of the color is adjusted, the perceived brightness perceived by people will also change. This is the H-K effect. Specifically, even if you only increase the chroma of a color, the perceived brightness of that color will increase. In order to ensure that the converted color does not exceed the color gamut, the color conversion of the present invention is performed in the HSL color space.

In the Helmholtz-Kohlrausch (HK) effect model, the perceived brightness L ^** is defined as:

L ^** (L ^* ,C ^* ,h)＝L ^* +(2.5-0.025L ^* )q(h)C ^* (1)

Among them, q(h) represents the impact of colors of different hues on perceived brightness.

a* and b* are chromaticity coordinates, indicating the position of the color in the L*a*b* color space.

The second term of formula (1), namely (2.5-0.025L ^* )q(h)C ^* , is the correction amount of HK effect. L ^* , C ^* , and h in formula (1) respectively represent the brightness (lightness), chroma (chroma), and hue (hue) in the L*a*b* color space. For achromatic colors, q(h) cannot be calculated because h is not defined. The relationship diagram between h and q(h) is shown in Figure 1.

The process of the present invention is shown in Figure 2. The goal of this invention is to enhance the perceived brightness of color images by enhancing their saturation. Same as original perceived brightness L ^** . Because compared to the 1976L*a*b* color space, the HSL color space is a regular color space, and the value range of the saturation S is [0,1], so the color gamut will not be exceeded when the saturation is enhanced.

The complete steps of the present invention can be described as:

Step 1. Convert the image from RGB color space to HSL color space and L*a*b* color space respectively. In the figure, RGB represents the image in the RGB color space, HSL represents the image in the HSL color space, and L*a*b* represents the image in the L*a*b* color space.

Step 2: Calculate the perceived brightness L ^** of the image in the L*a*b* color space. The calculation formula is the aforementioned formula (1), that is, L ^** =L ^* + (2.5-0.025L ^* )q(h)C ^* .

Step 3: Perform saturation enhancement on the image in the HSL color space to obtain the enhanced saturation with/> Substituting the original saturation, the resulting image is expressed as/>

The saturation enhancement function is defined as:

in

S in formula (5) is the saturation in the HSL color space. γ(C ^* ) is a sigmoid function, controlled by parameters α, β, and γ ₀ . The value range of γ(C ^* ) is [1, γ ₀ ]. The parameter β determines the slope of the sigmoid function, which is a positive real number. The parameter α is related to the chromaticity C ^* . When C ^* is sufficiently smaller than α, γ(C ^* ) is approximately equal to 1. The value of γ ₀ is a constant, and in this embodiment, its value is set to 3.

In the present invention, enhanced saturation is defined as:

Step 4, in order to keep the perceived brightness unchanged during the conversion process, perform brightness correction on the saturation-enhanced image based on the Helmholtz-Kohlrausch effect. The specific process can be described as follows:

Step (1) Set the maximum value of brightness L _top and the minimum value of brightness L _bottom to 1 and 0 respectively, and go to step (2).

Step (2) Calculate the average brightness value L _middle = (L _top + L _bottom )/2, and go to step (3).

Step (3) When the conditions are met When, set L _middle to correct the brightness and ends; otherwise, go to step (4); where ε is the setting accuracy, and in this embodiment, ε=0.1.

Step (4) When the conditions are met When, set L _middle to L _top ; otherwise, set L _middle to L _bottom , and go to step (2).

Among them, L ^** (H,S,L) is the original perceived brightness of the image in HSL color space, Think/> The perceived brightness when replacing the original saturation S and the original brightness L with L _middle . The two calculation methods are to divide the values in the HSL color space/> and (H, S, L) are converted to RGB color space, then converted from RGB color space to L*a*b* color space, and calculated using the aforementioned formula (1) in L*a*b* color space.

specifically:

(1) The conversion formula from HSL color space to RGB color space is:

If S=0, R, G, B are defined as

R＝G＝B＝L

If S≠0, M ₁ and M ₂ are respectively defined as

M ₁ =2L-M ₂

Process 1:

For R, set the value of h' to (H+2π/3), and go to [Processing 2]. Then, change the calculated X to R.

For G, set the value of h′ to H, and go to [Processing 2]. Then, change the calculated X to G.

For B, h′ is set to (H-2π/3), and then goes to [Processing 2]. Then turn the calculated X into B.

h′ is the temporary hue value used when calculating RGB values.

Process 2:

The calculation formula of h″ is

X is defined as

h″ is a temporary value after normalizing the range of h′ to [0,2π).

(2) The conversion formula from RGB color space to XYZ color space is:

(3) The conversion formula from XYZ color space to L*a*b* color space is:

Among them, X _n , Y _n and Z _n are the tristimulus values of standard white.

(4) Perceived brightness based on H-K effect

The perceived brightness is obtained by the aforementioned formula (1).

Step 5: Calculate the perceived brightness of the brightness-corrected image When/> Record the corresponding corrected brightness Perceived brightness/> For the calculation process, refer to /> in step 4

Step 6, in HSL color space to enhance the saturation Replaces the original saturation S with the recorded corrected brightness Replacing the original brightness L, the resulting image is expressed as/> Will/> Convert to RGB color space, and the resulting image is expressed as/> That is, the resulting image with complete saturation enhancement.

In order to facilitate the observation of changes in perceived brightness, the change in perceived brightness between the original image and the output image of each method is defined as:

in, is the perceived brightness of the i-th pixel in the original image. /> is the perceived brightness of each pixel in the image obtained by saturation enhancement. n is the total number of pixels in the original image.

It can be seen from Table 1 that the perceived brightness change of the method in the present invention Much smaller than existing technology. It can be seen from the yellow balloons shown in (a), (b), and (c) in Figure 3 and the green peppers shown in (a), (b), and (c) in Figure 4 that the existing method can There are obvious changes in the perceived brightness of balloons and green peppers; however, after the method of the present invention enhances saturation, the perceived brightness changes very little.

Table 1 for each method

Claims (4)

1. A method for enhancing saturation of a color image, comprising the steps of:

step 1, converting an image from an RGB color space to an HSL color space and an Labcolor space respectively;

step 2, calculating the perceived brightness L of the image in color space of L ^** ；

Step 3, carrying out saturation enhancement on the image in the HSL color space to obtain enhanced saturation

And 4, carrying out brightness correction on the image with enhanced saturation based on the Helmholtz-Kohlrausch effect, wherein the brightness correction steps are as follows:

step (1) maximum value L of brightness _top And a minimum value L of brightness _bottom Respectively setting to 1 and 0, and turning to the step (2);

step (2) calculating an average luminance value L _middle ＝(L _top +L _bottom ) 2, converting to the step (3);

step (3) when the condition is satisfiedWhen L is to _middle Set to correct brightness->And ends; otherwise, go to (4); wherein epsilon is the setting precision, L ^** (H, S, L) is the original perceived brightness of the image in the HSL color space,>to->Instead of the original saturation S, L _middle Perceived brightness when replacing original brightness L;

step (4) when the condition is satisfiedWhen L is to _middle Set to L _top The method comprises the steps of carrying out a first treatment on the surface of the Otherwise, L is _middle Set to L _bottom And go to step (2);

the saidAnd L ^** (H, S, L) is the value +.>And (H, S, L) to RGB color space, and then from RGB color space to L x a b color space, and performing computation in L x a b color space;

step 5, calculating the perceived brightness of the brightness-corrected imageWhen->Recording the corresponding corrected brightness->

Step 6, in HSL color space, to enhance saturationInstead of the original saturation S, to record the corrected luminance +.>Instead of the original luminance L, the resulting image is converted into RGB color space.

2. The method of enhancing saturation of a color image according to claim 1, wherein said steps 2, l ^** The calculation formula of (2) is as follows:

L ^** ＝L ^* +(2.5-0.025L ^* )q(h)C ^*

wherein L is ^* 、C ^* H represents luminance (light), chromaticity (chroma), and hue (hue) in the color space, and q (h) represents the influence value of the colors of different hues on perceived luminance:

a and b are chromaticity coordinates representing the position of a color in the L x a x b x color space.

3. The method for enhancing saturation of color image according to claim 1, wherein in step 3, the enhanced saturation isThe calculation formula of (2) is as follows:

wherein f (S, C) ^* ) As a saturation enhancement function, S is the saturation in the HSL color space, γ (C ^* ) As a sigmoid function, C ^* Representing colors in color spaceDegree.

4. A saturation enhancement method for a color image according to claim 3, wherein the gamma (C ^* ) The value range of (2) is [1, gamma ] ₀ ]The calculation formula is as follows:

α、β、γ ₀ is control gamma (C) ^* ) The parameter beta determines the slope of the sigmoid function, which is a positive real number, the parameter alpha and the chromaticity C ^* In relation to, when C ^* When the ratio α is sufficiently small, γ (C ^* ) Approximately equal to 1, gamma ₀ The value of (2) is constant.