An Improved Random Sample Consensus Based on Density-Based Spatial Clustering of Applications with Noise for Image Mosaic
Jinda Liu, Yanyang Hou, Hongxing Pei · Pattern Recognition and Image Analysis · 2021
Abstract Image mosaic is the technique of constructing a sequence of images into a high-resolution image, which mainly includes image registration and image fusion. In this paper we propose a new method for image registration: feature vectors of matching points are formed firstly, then we use density-based spatial clustering of applications with noise to process feature vectors to improve Random Sample Consensus in the process of estimating transformation model between two images. The results show that proposed method outperforms the traditional method, which estimates the transformation model by random sample consensus only, on the spatial frequency, definition, and peak signal-to-noise ratio in images.