Detecting texture paths and patterns in aerial images

Abstract

This paper presents a methodology for the extraction, recognition and characterization of paths in aerial images. We present a sequential algorithm that creates a linear approximation of the underlying path. The path is divided into straight line segments in a sequential manner. It takes under consideration the local spatial structure of the patterns in order to extract the straight line segment and the overall global path is sequentially formed from each segment. We present the preprocessing steps used to extract patterns of interest from an image and the results from two case study examples. The methodology has applications for the autonomous path planning of aerial vehicle following a specified structure (river, road) and in cases where that we are interested in the spatial relationships of the underlying patterns in the image and want to detect changes in the spatial structure of the patterns.