Abstract
Abstract. Over the last decade, several automatic approaches have been proposed to extract and reconstruct 2D building footprints and 2D road profiles from ALS data, satellite images, and/or aerial imagery. Since these methods have to date been applied to various data sets and assessed through a variety of different quality indicators and ground truths, comparing the relative effectiveness of the techniques and identifying their strengths and short-comings has not been possible in a systematic way. This contest as part of IQPC15 was designed to determine pros and cons of submitted approaches in generating 2D footprint of a city region from ALS data. Specifically, participants were asked to submit 2D footprints (building outlines and road profiles) derived from ALS data from a highly dense dataset (approximately 225 points/m2) across a 1km2 of Dublin, Ireland’s city centre. The proposed evaluation strategies were designed to measure not only the capacity of each method to detect and reconstruct 2D buildings and roads but also the quality of the reconstructed building and road models in terms of shape similarity and positional accuracy.
Highlights
The availability of three-dimensional (3D) point clouds offers a potential resource for a wide range of applications
Heipke et al (1997) proposed an evaluation process to measure complete and correct road network extraction. The former aspect involve completeness, correctness and quality based on True Positive (TP), False Positive (FP) and False Negative (FN) results when compared to the reference road centrelines and extracted ones
TP, FP, and FN values were computed by comparing cell values from a pair of 2D grids represented by ground truth road (GtR) and extracted road profile (ExR), as expressed in Eq.s 8-10 and Figure 7
Summary
The availability of three-dimensional (3D) point clouds offers a potential resource for a wide range of applications (e.g. environmental planning and monitoring, computational simulation, disaster management, security, telecommunications, location-based services). Boyko and Funkhouser (2011) manually generated ground truth of a road network and proposed five comparative quantities (completeness, correctness, quality, average spill size, and prevailing spill direction) to evaluate extracted roads. This lack of consensus causes difficulty in generating a consistent comparative assessment of competing methods. This contest called for participants to submit resulting 2D footprints (building outliers and road profiles) from a provided ALS data set.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
