LAUSR

Insufficient luminosity and poor local contrast are the major hurdles affecting the visual quality of the fundus images. A suitable framework is proposed for the enhanced visual perception of color fundus images based on a hybrid approach that combines gamma correction and singular value equalization for luminosity enhancement and contrast-limited adaptive histogram equalization (CLAHE) for local contrast enhancement. Luminosity enhancement is done by performing singular value equalization of the low-frequency component of the original value channel of the image in hue, saturation, and value color space using the low-frequency component of the gamma-corrected value channel of the same image. Discrete wavelet transform is applied for extracting the corresponding low-frequency components from the original and gamma-corrected value channels. Local contrast enhancement is achieved using CLAHE performed on the luminosity channel in $$L^*a^*b^*$$L∗a∗b∗ color space. The performance of the proposed method is analyzed qualitatively based on visual assessment and quantitatively with the parameters such as peak signal-to-noise ratio, absolute mean brightness error, discrete entropy and measure of enhancement. Experiments conducted on the color fundus images show improved results with sufficient detail preservation and enhanced visual perception compared to the existing methods.