A noise robust edge detector for color images using Hilbert Transform

Abstract

This paper proposes a noise robust technique to facilitate edge detection in color images contaminated with Gaussian and Speckle noises. The proposed edge detector uses the concept of Hilbert transform to perform edge sharpening and enhancement. Bilateral Filtering assists in smoothening noisy pixels without affecting high frequency edge contents. Using Bilateral Filtering as a precursor to Hilbert Transform, drastically improves the degree of noise robustness. Simulations have been carried out on medical images and the results have been validated in Gaussian and Speckle noise environment.