Constant elasticity of substitution function based RANSAC for image stitching

Feature matching is very important in image stitching. RANSAC algorithm a representative for feature matching. However, still has many shortcomings such as large number of iterations, computational complexity and cannot completely eliminate mismatches. To address above problem, this paper, we propose novel method termed constant elasticity substitution function based (CES-RANSAC) Specifically, CES-RANSAC improves the by constructing utility function, optimizing boundary calculating Cobb-Douglas coefficients. It also introduces Lindahl equilibrium to derive return value t help Experiments show that compared with traditional algorithm, improved accuracy increased efficiency, which further efficiency algorithm.