Jensen, Andreas Kryger5; Savarimuthu, Thiusius Rajeeth6; Sørensen, Anders Stengaard6
1 The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, SDU2 Faculty of Engineering, SDU3 SDU Robotics, The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, SDU4 Department of Biostatistics, Department of Public Health, Det Sundhedsvidenskabelige Fakultet, SDU5 Department of Biostatistics, Department of Public Health, Det Sundhedsvidenskabelige Fakultet, SDU6 The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, SDU
We investigate methods for improving the visual quality of in vivo images of blood vessels in the human forearm. Using a near-infrared light source and a dual CCD chip camera system capable of capturing images at visual and nearinfrared spectra, we evaluate three fusion methods in terms of their capability of enhancing the blood vessels while preserving the spectral signature of the original color image. Furthermore, we investigate a possibility of removing hair in the images using a fusion rule based on the "a trous" stationary wavelet decomposition. The method with the best overall performance with both speed and quality in mind is the Intensity Injection method. Using the developed system and the methods presented in this article, it is possible to create images of high visual quality with highly emphasized blood vessels.
Ikke Angivet, 2009
Billed fusion; Wavelet; Dual Camera; Medicinsk billedbehandling; Sundhed; Infratrød billeder; Image Fusion; Medical Imaging; Health Care; Infrared Imaging
Main Research Area:
Robotics, Telematics and Applications, RTA2009, 2009