A new surface recovery method based on hybrid reflectance model

Yanfeng Li¹ Jiquan Ma²

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12B) 9-13                                                                                       

¹Department of Information Science and Technology, Heilongjiang University, 72 Xue Fu Road, Harbin, China

²Department of Computer Science and Technology, Heilongjiang University, 72 Xue Fu Road, Harbin, China

Generally, the surface shape of a object can be recovered from a single image, that is called shape-from-shading (SFS), which relies on the assumption of Lambertian, although the specular reflectance component often exists in the original image. In order to improve the quality of the surface reconstruction in SFS, hybrid reflectance model is introduced firstly and is applied on SFS method. Firstly, the reflectance component is estimated using simulated annealing based on the distribution of the surface’s normal. Secondly, a new surface recovery algorithm is designed under a hybrid reflectance model, which is a linear model composed with Lambertian and specular reflectance. Finally, Experiments are performed on synthetic and real images.