Acoustic imaging is usually performed using a single array of omnidirectional microphones distributed on a planar, cylindrical or spherical geometry. In the case of spherical microphone arrays (SMA)s, the analysis can be done in the spherical harmonics (SH) domain, as developed in the field of Ambisonics. However, practical implementations of the latter approach are often constituted of a smaller number of microphones compared to planar arrays, which implies limited spatial resolution and bandwidth. This work deals with the use of 5 19- microphone rigid SMAs for the capture of their respective local sound field, and discusses on how to process them globally for acoustic imaging purposes. Four strategies are presented in a comparative study, by considering the distributed SMAs as: \begin{list}{ }{} \item a) an array of 5$\times$19 omnidirectional microphones in free space, not taking into account scattering from the rigid spheres, \item b) an array of 5$\times$19 microphones mounted on diffracting spheres, scattering from the rigid spheres is taken into account, \item c) 5 distributed SMAs, each of them retrieving 16 SH coefficients up to degree 3, and processed in the SH domain on all coefficients at once, \item d) 5 Distributed SMAs combined by translation of the local SH coefficients on a global origin, capable of retrieving SH coefficients up to degree 7, and processed in the SH domain. \end{list} These processing strategies are evaluated in simulations and validated in the frame of laboratory measurements based of the use of 5 SMAs distributed on a plane, at the vertices of a pentagone. Acoustic imaging results are compared with a conventional planar microphone array constituted of 81 microphones.