Adaptive Statistics-Based Image Enhancement

摘要

In many applications we want to improve the visual appearance, quality, clarity and intelligibility of an image. There is no one generally accepted method of enhancing an image, which is subjective. This paper proposes to enhance an entire image by adapting the contrast and pixel intensity of blocks through statistics-based linear transformations using the desired means and standard deviations. We subdivide the image into blocks and adjust their respective means and standard deviations. This adaptively changes the contrast and pixel intensity to create an enhanced image.