Limits...
Visual Contrast Enhancement Algorithm Based on Histogram Equalization.

Ting CC, Wu BF, Chung ML, Chiu CC, Wu YC - Sensors (Basel) (2015)

Bottom Line: In addition, VCEA reduces the effects of the feature loss problem by using the obtained spaces.Furthermore, VCEA enhances the detailed textures of an image to generate an enhanced image with better visual quality.Experimental results show that images obtained by applying VCEA have higher contrast and are more suited to human visual perception than those processed by HE and other HE-based methods.

View Article: PubMed Central - PubMed

Affiliation: School of Defense Science, Chung Cheng Institute of Technology, National Defense University, Taoyuan 33551, Taiwan. chihchungting@gmail.com.

ABSTRACT
Image enhancement techniques primarily improve the contrast of an image to lend it a better appearance. One of the popular enhancement methods is histogram equalization (HE) because of its simplicity and effectiveness. However, it is rarely applied to consumer electronics products because it can cause excessive contrast enhancement and feature loss problems. These problems make the images processed by HE look unnatural and introduce unwanted artifacts in them. In this study, a visual contrast enhancement algorithm (VCEA) based on HE is proposed. VCEA considers the requirements of the human visual perception in order to address the drawbacks of HE. It effectively solves the excessive contrast enhancement problem by adjusting the spaces between two adjacent gray values of the HE histogram. In addition, VCEA reduces the effects of the feature loss problem by using the obtained spaces. Furthermore, VCEA enhances the detailed textures of an image to generate an enhanced image with better visual quality. Experimental results show that images obtained by applying VCEA have higher contrast and are more suited to human visual perception than those processed by HE and other HE-based methods.

No MeSH data available.


“Indoor View” processed by detailed texture enhancement (DTE). (a) DTE image; (b) DTE histogram.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541917&req=5

sensors-15-16981-f004: “Indoor View” processed by detailed texture enhancement (DTE). (a) DTE image; (b) DTE histogram.

Mentions: Following this, the DTE process determines the candidate gray values to be further enhanced. Here, DTE uses the gradient value as the basis to determine the candidate gray values in order to enhance the detailed textures. This is because a larger gradient value indicates that the relevant pixel is significantly different from adjacent pixels and is much easier to discriminate from them. On the contrary, a small gradient value indicates that the relevant pixel is similar to adjacent pixels and thus is hard to discriminate from them. To render the enhanced effect more obvious, the values of the total number of pixels of the candidate gray values cannot be small. They must be greater than the threshold value, , where M and N denote the height and width of an image, respectively. At the same time, the average gradient value of the candidate gray value has to be less than the specific value, which is the absolute value of the difference between and . Having obtained the qualified candidate gray values, the DTE process sorts them by their average gradient values, and sequentially enhances the candidate gray value with the greater average gradient until all the remaining free spaces are used up. For example, it is assumed that y is the first candidate gray value of the CPR histogram to be enhanced. The space between y − 1 and y is d gray levels, and CPRhist(y) denotes the histogram of the CPR image at gray value y. When d is greater than S(y), which is the space adjustment function introduced earlier, CPRhist(y) is shifted back by d − S(y) gray levels; conversely, when d is equal to or less than S(y),CPRhist(y) is shifted forward by S(y) − d gray levels. Once all CPRhist(y) have been sequentially shifted back or forward, the DTE image is obtained. Figure 4a is the image processed by using DTE, and it contains 119 gray values. Through the DTE process, detailed textures such as grass, trees on the left and right side, and the view behind the door in Figure 4a are enhanced. This process also makes the image appear much clearer than the CPR image, indicating that the DTE process can effectively enhance the detailed textures of images. Figure 4b is the histogram of the luminance (Y) component of Figure 4a. It is clear that the dynamic range observed in Figure 4b is wider than that observed in Figure 3b after the DTE process. Therefore, Figure 4a has better image quality. In addition, in this process, all relevant variables are automatically calculated according to the input images, and no parameters need to be tuned manually.


Visual Contrast Enhancement Algorithm Based on Histogram Equalization.

Ting CC, Wu BF, Chung ML, Chiu CC, Wu YC - Sensors (Basel) (2015)

“Indoor View” processed by detailed texture enhancement (DTE). (a) DTE image; (b) DTE histogram.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541917&req=5

sensors-15-16981-f004: “Indoor View” processed by detailed texture enhancement (DTE). (a) DTE image; (b) DTE histogram.
Mentions: Following this, the DTE process determines the candidate gray values to be further enhanced. Here, DTE uses the gradient value as the basis to determine the candidate gray values in order to enhance the detailed textures. This is because a larger gradient value indicates that the relevant pixel is significantly different from adjacent pixels and is much easier to discriminate from them. On the contrary, a small gradient value indicates that the relevant pixel is similar to adjacent pixels and thus is hard to discriminate from them. To render the enhanced effect more obvious, the values of the total number of pixels of the candidate gray values cannot be small. They must be greater than the threshold value, , where M and N denote the height and width of an image, respectively. At the same time, the average gradient value of the candidate gray value has to be less than the specific value, which is the absolute value of the difference between and . Having obtained the qualified candidate gray values, the DTE process sorts them by their average gradient values, and sequentially enhances the candidate gray value with the greater average gradient until all the remaining free spaces are used up. For example, it is assumed that y is the first candidate gray value of the CPR histogram to be enhanced. The space between y − 1 and y is d gray levels, and CPRhist(y) denotes the histogram of the CPR image at gray value y. When d is greater than S(y), which is the space adjustment function introduced earlier, CPRhist(y) is shifted back by d − S(y) gray levels; conversely, when d is equal to or less than S(y),CPRhist(y) is shifted forward by S(y) − d gray levels. Once all CPRhist(y) have been sequentially shifted back or forward, the DTE image is obtained. Figure 4a is the image processed by using DTE, and it contains 119 gray values. Through the DTE process, detailed textures such as grass, trees on the left and right side, and the view behind the door in Figure 4a are enhanced. This process also makes the image appear much clearer than the CPR image, indicating that the DTE process can effectively enhance the detailed textures of images. Figure 4b is the histogram of the luminance (Y) component of Figure 4a. It is clear that the dynamic range observed in Figure 4b is wider than that observed in Figure 3b after the DTE process. Therefore, Figure 4a has better image quality. In addition, in this process, all relevant variables are automatically calculated according to the input images, and no parameters need to be tuned manually.

Bottom Line: In addition, VCEA reduces the effects of the feature loss problem by using the obtained spaces.Furthermore, VCEA enhances the detailed textures of an image to generate an enhanced image with better visual quality.Experimental results show that images obtained by applying VCEA have higher contrast and are more suited to human visual perception than those processed by HE and other HE-based methods.

View Article: PubMed Central - PubMed

Affiliation: School of Defense Science, Chung Cheng Institute of Technology, National Defense University, Taoyuan 33551, Taiwan. chihchungting@gmail.com.

ABSTRACT
Image enhancement techniques primarily improve the contrast of an image to lend it a better appearance. One of the popular enhancement methods is histogram equalization (HE) because of its simplicity and effectiveness. However, it is rarely applied to consumer electronics products because it can cause excessive contrast enhancement and feature loss problems. These problems make the images processed by HE look unnatural and introduce unwanted artifacts in them. In this study, a visual contrast enhancement algorithm (VCEA) based on HE is proposed. VCEA considers the requirements of the human visual perception in order to address the drawbacks of HE. It effectively solves the excessive contrast enhancement problem by adjusting the spaces between two adjacent gray values of the HE histogram. In addition, VCEA reduces the effects of the feature loss problem by using the obtained spaces. Furthermore, VCEA enhances the detailed textures of an image to generate an enhanced image with better visual quality. Experimental results show that images obtained by applying VCEA have higher contrast and are more suited to human visual perception than those processed by HE and other HE-based methods.

No MeSH data available.