A new image enhancement algorithm based on nonsubsampled contourlet transform is presented. Low contrast images are transformed into multi-scale and multi-directional contour information, where a nonlinear mapping function is used to modify the contour coefficients at each level. The enhancement is achieved by amplifying weak edges and inhibiting the background noise while adjusting the dynamic range. Experiments show the proposed algorithm preserves the intrinsic geometrical information of the enhanced image and can be effectively applied to a wide range of image types under diverse illumination conditions. Introduction: Conventional histogram equalisation (CHE) based histogram modification techniques have received considerable attention due to their simplicity [1]. Although CHE and its derivatives can efficiently utilise display intensities, they tend to over enhance the image contrast if there are high peaks in the histogram, often resulting in noise and other artefacts in the output image [2]. One way to approaching this problem is to use multi-scale image decomposition, that is, processing images in each scale independently and recombining each processed image to obtain the final image. Advances in wavelet theory combined with multi-scale analysis applied to image contrast enhancement can achieve promising results. The decomposition of images into different frequency ranges permits the isolation of the frequency components introduced by 'intrinsic deformations' or 'extrinsic factors' in certain subbands [3]. In [4], singular value based image equalisation in the discrete wavelet transform domain (DWT-SVE) is proposed to equalise the illumination information in a low-low subband image and preserve the high frequency components in the other subbands. However, the 2D wavelet transform used is a separable extension of the 1D wavelet transform, which does not work well in capturing the geometry of image edges [6]. In this Letter, we propose a new image enhancement method based on the nonsubsampled contourlet transform (NSCT) [6]. The proposed algorithm enhances the dynamic range of the image while amplifying weak edges and suppressing noise by modifying the NSCT coefficients using a nonlinear mapping function in each directional subband.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.