An Extension of Reversible Image Enhancement Processing for Saturation and Brightness Contrast

Abstract

This paper proposes a reversible image processing method for color images that can independently improve saturation and enhance brightness contrast. Image processing techniques have been popularly used to obtain desired images. The existing techniques generally do not consider reversibility. Recently, many reversible image processing methods have been widely researched. Most of the previous studies have investigated reversible contrast enhancement for grayscale images based on data hiding techniques. When these techniques are simply applied to color images, hue distortion occurs. Several efficient methods have been studied for color images, but they could not guarantee complete reversibility. We previously proposed a new method that reversibly controls not only the brightness contrast, but also saturation. However, this method cannot fully control them independently. To tackle this issue, we extend our previous work without losing its advantages. The proposed method uses the HSV cone model, while our previous method uses the HSV cylinder model. The experimental results demonstrate that our method flexibly controls saturation and brightness contrast reversibly and independently.

Keywords: HSV color space; contrast enhancement; data hiding; reversibility; saturation improvement.

Figure 1

Block diagrams of previous methods. (a) Previous method [13]; (b) previous method [15].

Figure 2

Block diagrams of the proposed method. (a) Saturation improvement and contrast enhancement process; (b) recovery process.

Figure 3

Histogram transition in $Min$ for saturation improvement. (a) In case the reference bin is empty (Step 2 (Case 1)); (b) in case the reference bin is not empty (Step 2 (Case 2)).

Figure 4

Histogram transition of $Min$ and $Max$ for CE. (a) Original histograms of $Min$ and $Max$; (b) merged histogram; (c) merged histogram after preprocessing [8]; (d) histograms separated into $Min$ and $Max$; (e) $Max$ histogram after CE process and $Min$ histogram after adjustment.

Figure 5

Histogram transition of $\widehat{Max}$, $\widehat{Median}$, and $\widehat{Min}$ in the adjustment for magnitude relation. (a) Histograms of $\widehat{Max}$, $\widehat{Median}$, and $\widehat{Min}$; (b) merged histogram; (c) merged histogram after preprocessing [8]; (d) histograms separated into $\widehat{Max}$, $\widehat{Median}$ and $\widehat{Min}$; (e) histograms of $\widehat{Max}$ shifted by +1, $\widehat{Min}$ shifted by −1, and $\widehat{Median}$.

Figure 6

Images output by proposed method, where $I_S=0,10,20$ and $I_V=0,15,30$ (kodim16).

Figure 7

Original and output images and their pixel distributions for saturation and brightness, where $I_S$ = 20 and $I_V$ = 30 (kodim16). (a) Original image; (b) proposed method; (c) previous method [15]; (d) previous method [13]; (e) previous method [1].

Figure 8

Original and output images and their pixel distributions for saturation and brightness, where $I_S$ = 20 and $I_V$ = 30 (house). (a) Original image; (b) proposed method; (c) previous method [15]; (d) previous method [13]; (e) previous method [1].

Figure 9

Maximum levels of saturation improvement and CE for proposed method (kodim16).

Figure 10

Maximum values of RCE and saturation difference (Kodak). (a) RCE; (b) saturation difference.

Figure 11

Maximum values of RCE and saturation difference (SIPI). (a) RCE; (b) saturation difference.