Processing Specular Reflection Components of Chrome-plated Surface by Multi-Image Reconstruction Method

Abstract

Chrome-plated surfaces are perfectly specular, in which the diffuse reflection components are weak while the specular reflection components are so strong that their intensities saturate the CCD matrix and cause some surface features to be lost. This situation means that one image alone cannot show all of the surface features. To process specular reflection components, we proposed a novel multi-image reconstruction method (MIRM), which reconstructs a synthetic image from a set of images. In the set of images that are captured by a special ring pattern illumination system, the specular spots move in a smooth motion in the intact surface as the light source moves, and new gradient values can be obtained. Using the new gradient values a synthetic image can be reconstructed by the multi-grid method. Experimental results show that our method is effective in significantly removing specular highlights, while preserving or sharpening such features as edges and defects.