Colour to Grey by Maximum Signed Gradient
暂无分享,去创建一个
While the magnitude of colour gradients is well defined, its direction is not. The ambiguity in the direction is due to our inability to decide if red colour is greater than green in the same way we can say that white is brighter than grey. In this paper, we present a method to calculate the maximum signed gradient and integrate it to a grey image. Experiments on natural photos show that seventy per cent of edges are due to luminance change, i.e., all the channels change in the same direction. The other gradients are due to material change such as yellow to blue or green to orange. The algorithm can be used to decompose images into luminance and chrominance. It produces noticeably better local contrast than normal luminance conversion.
[1] Reinhard Klette,et al. Height from Gradient Using Surface Curvature and Area Constraints , 2002, ICVGIP.
[2] Rama Chellappa,et al. An algebraic approach to surface reconstruction from gradient fields , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.
[3] Graham D. Finlayson,et al. Spherical sampling methods for the calculation of metamer mismatch volumes , 2019, Journal of the Optical Society of America. A, Optics, image science, and vision.