Last years have seen a big effort of research community to solve the security problems by developing video surveillance systems. Normally, developed systems have the characteristic to be executed on images acquired by autonomous cameras. In this paper, we propose a new way to develop visual surveillance systems which would not be a set of passive modules but systems that actively decide both what to see and how to see it. In particular, our idea is to act on the regulation of the acquisition parameters as consequence of what the system needs to see. The regulation strategy is based on two parameters, focus and iris, and aims to identify an optimal sequence of steps to enhance the acquisition quality of the object of interest. To this end, a hierarchy of neural networks has been employed to select first which parameter must be regulated then to adjust it.