Geometrically Driven Underground Camera Modeling and Calibration With Coplanarity Constraints for a Boom-Type Roadheader

The conventional calibration methods based on a perspective camera model are not suitable for the underground with two-layer glasses, which is specially designed explosion proof and dust removal in coal mine. Underground modeling algorithms urgently needed to improve precision reliability of visual measurement systems. This article presents novel geometrically driven algorithm boom-type roadheader. established under coplanarity constraints, considering explicitly impact refraction triggered by glasses deriving geometrical relationship equivalent collinearity equations. On this basis, we perform parameters model, 2D-2D correspondences between image points object coordinates planar target. A hybrid Levenberg-Marqurdt (LM) particle swarm optimization (PSO) further proposed terms dynamic combination LM PSO, optimizes results minimizing error nonlinear model. experimental demonstrate that pose errors caused glass well corrected method. accuracy cutting-head estimation has increased 55.73%, meeting requirements excavations.