Multimed Tools Appl https://doi.org/10.1007/s11042-018-6206-z Non-rigid point set registration via global and local constraints Changcai Yang 1 · Meifang Zhang 2 · Zejun Zhang 3 · Lifang Wei 3 · Riqing Chen 3 · Huabing Zhou 4 Received: 16 September 2017 / Revised: 26 April 2018 / Accepted: 23 May 2018 © Springer Science+Business Media, LLC, part of Springer Nature 2018 Abstract Non-rigid point set registration is often encountered in meical image processing, pattern recognition, and computer vision. This paper presents a new method for non-rigid point set registration that can be used to recover the underlying coherent spatial mapping (CSM). Firstly, putative correspondences between two point sets are established by using feature descriptors. Secondly, each point is expressed as a weighted sum of several nearest neighbors and the same relation holds after the transformation. Then, this local geometri- cal constraint is combined with the global model, and the transformation problem is solved  Riqing Chen riqing.chen@fafu.edu.cn Changcai Yang changcaiyang@gmail.com Meifang Zhang zjzhang fafu@163.com Zejun Zhang mfzhang85@163.com Lifang Wei weilifang0028@fafu.edu.cn Huabing Zhou zhouhuabing@gmail.com 1 Digital Fujian Research Institute of Big Data for Agriculture and Forestry, College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou 350002, China 2 Fujian Health College, Fuzhou 350101, China 3 College of Computer and Information Sciences, Fujian Agriculture and Forestry University, Fuzhou 350002, China 4 Wuhan Institute of Technology, Wuhan 430073, China