Computer Science and Engineering, Department of


Date of this Version



Published in Proceedings of the Ninth International Conference on Pattern Recognition (November 1988): 220-222. Copyright 1988, IEEE. Used by permission.


The goal of the scene labeling problem is to identify a set of regions in a given image. There are several approaches to solve this problem, including backtracking, graph matching, etc. A new method called split-level relaxation based on discrete relaxation was proposed in. It takes care of multiple semantic constraints, by considering each of them independently. The problem is NP-complete, so it takes a long time to solve this problem. With the advent of multiprocessors, it is now imperative to see if the problem can be solved faster in the average case.

We give a framework for solving the problem in a parallel processing environment, using split-level relaxation. Experiments done on a multiprocessor show that indeed it is advantageous to use them to solve this problem. The results are also presented.