Using models to recognise man-made objects

Abstract

Objects in aerial images are extracted by using a new method based on a two dimensional viewer centred model. Using this approach has advantages over existing methods because the models are easily created and it is e cient. Experiments illustrate the extraction and tracking of man made objects in reconnaissance images. In later work we intend to extend the method to allow selection between di erent object types, or between di erent views of the same object.

Man-Made Object Extraction From Remote Sensing Imagery by Graph-Based Manifold Ranking

“…Li et al (2000) utilized a 2D hidden Markov model and estimated the values of the relevant parameters by an EM algorithm. Reno and Booth (1999) extracted man-made objects by using a viewer-centred reference model with a deformable template on the basis of combining model and image information. Model-based algorithms present high pertinence and efficiency, thereby avoiding extracting numerous unnecessary features in the image.…”

Rapid, man-made object morphological segmentation for aerial images using a multi-scaled, geometric image analysis

“…These man-made object segmentations can be classified into two main categories of algorithms: (1) model-based algorithms [1][2][3] and (2) feature-based algorithms [4,5].…”