In this paper, we describe the development of a rapidly reconfigurable system in which the users’ tacit knowledge and requirements are

elicited via a process of Interactive Evolution, finding the image processing parameters to achieve the required goals without any need for

specialised knowledge of the machine vision system. We show that the resulting segmentation can be quickly and easily evolved from

scratch, and achieves detection rates comparable to those of a hand-tuned system on a hot-rolled steel defect recognition problem.-In this paper, we describe the development of a rapidly reconfigurable system in which the users tacit knowledge and requirements areelicited via a process of Interactive Evolution, finding the image processing parameters to achieve the required goals without any need forspecialised knowledge of the machine vision system. We show that the resulting segmentation can be quickly and easily evolved fromscratch, and achieves detection rates comparable to those of a hand-tuned system on a hot-rolled steel defect recognition problem.