Featured in:
6th Iberian Conference on Pattern Recognition and Image Analysis. Madeira, Portugal
Authors:
Diogo Roxo, Nuno Gonçalves and João Barreto
Near-Lighting Endoscopes are self-illuminated cameras often used in minimal invasive surgery. Since they have a wide field of view, their images are affected by high radial distortion (RD) and reduced resolution in the periphery of the image. Perspective Shape From Shading has been used for reconstruction but suffers from the problem of resolution and high RD when endoscopes are used. We propose in this paper two improvements to the state of the art methods for PSFS in Near-Lighting Endoscopes. The first contribution is the introduction of the RD model directly in PSFS equations and the second contribution is the compensation of the reduced resolution of the image in its periphery, due to wide FOV. Tests performed in real objects and in a knee bone show that by modeling these two effects our method highly enhances the accuracy of the estimation.
© 2024 VISTeam | Made by Black Monster Media
Institute of Systems and Robotics Department of Electrical and Computers Engineering University of Coimbra