3-D color histogram equalization by principal component analysis

•The proposed method uses a one dimensional histogram for color HE.•The proposed method applies equalization on a transformation of input image.•The proposed method uses the projection of image on the first component of PCA.•The resulted images of the proposed method are free from brightening.•The resulted mean brightness differences are not very significant.

Color histogram equalization is a method for improving visual appearance of images by enhancing image contrast. Color histogram equalization methods are mostly faced with problems like over-enhancement and brightening. In this paper a new color histogram equalization method is proposed which defines a new three dimensional cumulative distribution function based on a one-dimensional histogram. This one-dimensional histogram is calculated by taking into account the correlation between color channels using PCA. Over-enhancement and brightening are solved by this method because of applying the equalization on a transformed image instead of image itself.