In this paper we describe an implementation of the on-line validation algorithm for the analysis of occluded images, which was developed by Iliopoulos and Simpson [Iliopolous, C.S. and Simpson, J. (1997)]. The algorithm operates on images represented in one dimension as strings and assumes objects within images are of the same length. We also investigate the decomposition of a given image into the set of (perhaps partially occluded) objects occurring in it.

We first present the theoretical background to this work and sketch the main components of the analysis and validation method. Then we discuss implementation issues and the testing techniques used and present the test results.