Abstract
Abstract Complex surfaces merging multiple scales of features are often measured on multi-sensor systems, which requires sophisticated data modelling and fusion methods. A Gaussian process-based Bayesian inference method is presented to model the multi-scale surface geometries by designing composite covariance kernel functions. The statistical nature of the Gaussian process makes the method generic for different kinds of surfaces, and capable of giving credibility to the established model, which can further be used as a critical criterion to perform active data sampling and fusion in multi-sensor systems. Experimental work concerning the validity and application of this method is presented.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
