Image enhancement with preservation of brightness and details using multiscale top-hat transform

Abstract

Image enhancement is a widely used technique to enhance the visual quality of images. However, this process introduces distortions and loss of detail in the images. In this article, we present a scheme that extracts the brightness and darkness features of an image in multiple scales. This extraction process is done by multiscale top-hat transform, which uses two structuring elements in the basic operations of mathematical morphology. Image enhancement consists of adding regions of brightness and removing regions of darkness from the original image. Tests were performed using 200 images from a public database. Our approach was compared with state-of-the-art algorithms to verify its relative performance. Experimental results show that our proposal improves images in terms of contrast, detail, preserves mean brightness and also introduces less distortion than the compared algorithms.