A constraint satisfaction approach to the resolution of uncertainty in image interpretation

Abstract

A technique for image interpretation based on the constraint satisfaction methodology is described. The technique uses an intelligent backtracking algorithm to solve the constraint satisfaction problem and also analyzes failures of the backtracking routine to suggest modifications to help locate a solution. This technique is able to overcome uncertainties in image interpretation by generating partial solutions and inferring values for the missing objects. An outline of this processing is presented and an application of this technique is also given. The technique is compared to another approach in the same domain. Preliminary results on over 1000 test images are included.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>