Task driven perceptual organization for extraction of rooftop polygons

Abstract

A new method for extracting planar polygonal rooftops in monocular aerial imagery is proposed. Structural features are extracted and hierarchically related using perceptual grouping techniques. Top-down feature verification is used so that features, and links between the features, aTe uerijied with local information in the image and weighed in a graph. Cycles in the graph correspond to possible building rooftop hypotheses. Virtual features are hypothesised for the perceptual completion of partially occluded rooftops.Extraction of the "best1' grouping of features into a building rooftop hypothesis is posed as a graph search problem. The maximally weighted, independent set of cycles in the graph is extracted as the final set of roof boundaries.

Building Detection and Description from a Single Intensity Image

“…Additional techniques have been reported recently, some of which rely on a variety of data sources, such as range data [11][12][13][14][15][16][17][18][19][20][21][22]. Simple edge-based techniques such as contour tracing [2,4,9,10] encounter the problem of a rapidly growing search space. A more robust edge-based technique is that of perceptual grouping [4,5,8].…”

“…Simple edge-based techniques such as contour tracing [2,4,9,10] encounter the problem of a rapidly growing search space. A more robust edge-based technique is that of perceptual grouping [4,5,8]. For reconstruction of the 3-D information, most of the monocular systems use the corresponding shadow evidence of a building to infer the building height [2,3,5].…”

Signal-to-noise behavior for matches to gradient direction models of corners in images

“…They are used in problems that involve (1) extracting polygons associated with man-made objects (such as buildings) from overhead images [8,9], (2) establishing correspondences between points in overlapping images for image registration [22,21], stereovision, or sensor calibration [19,23], (3) detecting and tracking motion in video sequences [4,20], etc.…”

Autonomous Building Detection Using Edge Properties and Image Color Invariants

“…Considering both radiometry and geometry, many of these algorithms are edge-based detection techniques [6][7][8] consisting of linear feature detection and grouping for parallelogram structure hypotheses extraction [9]. In addition to linear feature detection approaches, other edge-based detection approaches rely on building polygons verification using the geometric structure of buildings [7], shadow [6,[10][11][12], and illuminating angles [6], among others.…”

“…In addition to linear feature detection approaches, other edge-based detection approaches rely on building polygons verification using the geometric structure of buildings [7], shadow [6,[10][11][12], and illuminating angles [6], among others.…”