Contrast Enhancement of Medical Images Using Multiscale Edge Representation

Experience suggests the existence of a connection between the contrast of a grayscale image and the gradient magnitude of intensity edges in the neighborhood where the contrast is measured. This observation motivates the development of edge-based contrast enhancement techniques. In this paper, we present a simple and e ective method for image contrast enhancement based on the multiscale edge representation of images. The contrast of an image can be enhanced simply by stretching or upscaling the multiscale gradient maxima of the image. This method o ers exibility to selectively enhance features of di erent sizes and ability to control noise magni cation. We present some experimental results from enhancing medical images and discuss the advantages of this wavelet approach over other edge-based techniques.