Enhancing Dental Radiographic Images in Spline-Type Spaces

The aim of this paper is to propose a method to enhance the acquisition resolution of dental radiographic images to facilitate the clinical examination and interpretation. The algorithm is based on the approximation properties of spline-type spaces with multiple generators. These spaces are obtained by applying a discrete group of translation operators to a finite set of smooth functions, forming a Riesz basis for its closed linear span within the Hilbert space L2(R2). For computational efficiency a parallel approach to the algorithm is also proposed. The experiments show that the algorithm allows to increase the resolution of dental radiographic images to sub -- pixel levels.