A morphological approach of target detection on perspective plane

This paper presents a morphological approach of target detection on the perspective-distorted plane for image analysis applications. By modifying morphological operations with a dynamic-varying structuring element, distorted targets with variant contrast values. shape deformations, and different location/orientation can be detected successfully on the perspective plane even though they are in cluttered background. Two experimental simulation results are presented to show that our algorithm works successfully in detecting the targets from their blurred perspective plane image. (C) 2001 Elsevier Science B.V. All rights reserved.