Fluorescence imaging is a tomography technique which provides tissue functional information such as oxygenation, and tumour detection. Propagation of near-infrared light in tissue is modelled by a coupled system of equations, one for the excitation light (light irradiated through the tissue by a laser) and one for the emission light (light which comes from the fluorophore). We used the radiative transfer equation and its diffusion approximation equation to model the fluorescence process. This talk particularly addresses the feasibility of a two-step scheme for reconstruction of a fluorophore target embedded in a semi-infinite medium by neglecting the presence of the fluorophore target for the excitation light and used an analytical solution of the homogeneous semi-infinite medium for the forward problem. In the first step of this reconstruction scheme, we implemented a pixel-based reconstruction using the Landweber method. The second step uses this result as an initial guess for solving the shape and contrast value reconstruction problem using the level set method. Numerical experiments using Monte Carlo data measurements, show that the proposed scheme provides reconstructions of shape, location and contrast value of the target with rather good accuracy.