Contrast and Luminance Enhancement Technique for Fundus Images Using Bi-Orthogonal Wavelet Transform and Bilateral Filter

Abstract

For the identification of retinal diseases, optical images from the low power microscopic fundus camera have been used extensively by ophthalmologists. Uneven lighting and poor image contrast, however, may lead to reduced image quality, making it hard for ophthalmologists to analyze the diseases and can also lead to misinterpretation. This paper proposes a combination of the technique of Bi-orthogonal Wavelet Transform and Bilateral filtering (BWTBF) to uniformly improve visual clarity in the image field. BWT splits the image into subbands of low and high frequency, and bilateral filtering eliminates the high-frequency noise and preserves the image’s edge information. Gamma correction also tends to increase the luminance, and Contrast-Limited Adaptive Histogram Equalization (CLAHE) increases the image’s entire local contrast. Parameters such as Mean, Entropy, and Enhancement Measure (MSE) are used to evaluate this method, and the measured values are higher than the values of existing methods that confirm the efficacy of the proposed method.