An adaptive enhancement method for low illumination color images

Abstract

In order to effectively improve the visual effect and image quality of color images under low illumination conditions, we propose an image enhancement method based on HSV and CIEL*a*b* color spaces for adaptively enhancing color image under low illumination conditions. The proposed method takes into account the characteristics of low illumination color images, and has the strategies of contrast, brightness enhancement, and color saturation correction. We utilize our proposed adaptive chaotic particle swarm optimization algorithm in this paper combined with gamma correction to improve the overall brightness of the image, and generate the best brightness adjustment effect in the proposed algorithm. In addition, our improved adaptive stretching function is used to enhance the image saturation. The experimental results show that compared with other traditional and latest color image enhancement algorithms, the proposed algorithm significantly enhances the visual effect of the low illumination color images. It can not only improve the contrast of low illumination color images and avoid color distortion, but also effectively improve the brightness of the image and provide more detail enhancement while maintaining the naturalness of the image.

Appendices

Appendix A

The L*a*b* color space is also called CIELAB, and it is devoted to the perceptual uniformity of human vision. The Euclidean distance is used to illustrate the difference between colors. L* component represents the brightness of pixels, and its extends from 0 (pure black) to 100 (pure white). The range of values of both a* and b* is [-128, 127], and a* component indicates the range from red to green, and b* component is the range from yellow to blue. L* component closely matches the brightness perception of human, and it is linear with human brightness perception. That is to say, if the L* value of one color is 1.5 times that of another color, then the brightness of the first color is 1.5 times that of the second color in visual perception. However, there is no direct conversion formula between RGB color space and L*a*b* color space. The image in RGB color space is firstly converted to the CIEXYZ color space, and then CIEXYZ color space is transformed into CIEL*a*b* color space. The specific conversion steps of RGB color space and L*a*b* color space are as follows:

(1) RGB to CIEXYZ color space

$$ \left[ \begin{matrix} X \\ Y \\ Z \\ \end{matrix} \right] =\left[ \begin{matrix} 0.412453 & 0.357580 & 0.180423 \\ 0.212671 & 0.715160 & 0.072169 \\ 0.019334& 0.119193 & 0.950227 \\ \end{matrix} \right]\left[ \begin{matrix} R \\ G \\ B \\ \end{matrix} \right] $$

(19)

(2) CIEXYZ to L*a*b* color space

$$ L^{*}=116f(Y/Y_{n})-16 $$

(20)

$$ a^{*}=500[f(X/X_{n})-f(Y/Y_{n})] $$

(21)

$$ b^{*}=200[f(Y/Y_{n})-f(Z/Z_{n})] $$

(22)

$$ f(u)=\left\{\begin{matrix} u^{\frac{1}{3}},&u>(\frac{6}{29})^{3}\\ & \\ \frac{1}{3}(\frac{29}{6})^{2}u+\frac{16}{116},&u\leq (\frac{6}{29})^{3}\end{matrix}\right. $$

(23)

where X_n, Y_n and Z_n are the CIE XYZ tristimulus values of standard D65 lighting white spot, take the value of X_n = 0.950456, Y_n = 1.000000, Z_n = 1.088754.

(3) L*a*b* to CIEXYZ color space

$$ \hat{X}=X_{W}\times g(\frac{L^{*}+16}{116}+\frac{a^{*}}{500}) $$

(24)

$$ \hat{Y}=Y_{W}\times g(\frac{L^{*}+16}{116}) $$

(25)

$$ \hat{Z}=Z_{W}\times g(\frac{L^{*}+16}{116}-\frac{b^{*}}{200}) $$

(26)

where g(u) is the inverse function f^− 1(u) of f(u).

$$ g(t)=\left\{\begin{matrix} u^{3},&u> \frac{6}{29}\\ & \\ 3(\frac{6}{29})^{2} (u-\frac{4}{29}), & u\leq \frac{6}{29} \end{matrix}\right. $$

(27)

(4) CIEXYZ to RGB color spac

$$ \left[ \begin{array}{c} \hat{R} \\ \hat{G} \\ \hat{B} \\ \end{array} \right] =\left[ \begin{array}{ccc} 3.240479 & -1.537150 & -0.498535 \\ -0.96256 & 1.875992 & 0.041556 \\ 0.055648 & -0.204043 & 1.057311 \\ \end{array} \right]\left[ \begin{array}{c} \hat{X} \\ \hat{Y} \\ \hat{Z} \\ \end{array} \right] $$

(28)

Appendix B

HSV color space is composed of three attributes: H(hue), S(saturation) and V(value). The H stands for hue and displays the true color attribute, such as cyan, blue, magenta, red, green and blue and so on. The S stands for saturation and represents the dilution of the true color attribute with the white color. It shows that the saturation will be higher if the value of the true color attribute is more than the white, otherwise the saturation will be smaller. The V stands for the value of the color intensity or brightness in the image. Let R, G, B denotes red, green and blue colors. The conversion formula of RGB color space to HSV color space is as follows:

$$ H=arccos\frac{(2R-G-B)}{2\sqrt{(R-G)^{2}-(R-B)(G-B)}} $$

(29)

$$ S=\frac{max(R,G,B)-min(R,G,B)}{max(R,G,B)} $$

(30)

$$ V=max(R,G,B) $$

(31)

where H stands for hue, S stands for saturation, V stands for value.

Then the HSV color space is reversed to RGB color space, and the conversion formula is as follows:

$$ \left\{\begin{matrix} C=V\times S\\ H=\frac{H}{60^{0}}\\ X=C\times(1-|Hmod2-1) \end{matrix}\right. $$

(32)

$$ (R_{1},G_{1},B_{1})\left\{\begin{matrix} (C,X,0)& 0\leq H\leq 1\\ (X,C,0)& 1\leq H\leq 2\\ (0,C,X)& 2\leq H\leq 3\\ (0,X,C)& 3\leq H\leq 4\\ (X,0,C)& 4\leq H\leq 5\\ (C,0,X)& 5\leq H\leq 6 \end{matrix}\right. $$

(33)

$$ \left\{\begin{matrix} m=V-C\\ (R,G,B)=(R_{1}+m,G_{1}+m,B_{1}+m) \end{matrix}\right. $$

(34)

where C is the chromaticity, X is the median of the second largest component with this color.