In the process industry, machine vision systems have been shown to be beneficial for the characterization of several products whose quality can be related to some visual features such as color, color uniformity, surface roughness, and presence of surface defects. With respect to the visual inspection traditionally carried out by a panel of trained experts, artificial vision systems can return a quick, accurate, and objective indication of the quality of the manufactured end product. However, reproducibility of the image analysis results is ensured only as long as the conditions, under which the images used for the calibration of the quality assessment model were collected, do not change during normal operation of the machine vision system. In this paper (Part II of this series following Part I, Ind. Eng. Chem. Res., 2013, 52, 12309-12318), we discuss a technology transfer problem for a machine vision system, namely the problem arising when replacing the camera, on which the system is centered, with a different one. In order to compensate for the differences between cameras, two strategies are tested. The :first one, which is borrowed from the machine vision literature, aims at Matching the color spaces of the two cameras through a linear or a nonlinear transformation. The second one, which was proposed in Part I to manage light alterations, aims at matching the color spaces of the cameras by synchronizing their projection onto a principal component analysis model. The superior performance of the latter strategy is demonstrated through two case studies involving images of pharmaceutical tablets and calibration standards of different colors, as well as images of film-coated tablets with different percentages of applied coating material.