Abstract
Abstract. Automatic detection and extraction of buildings from aerial images are considerable challenges in many applications, including disaster management, navigation, urbanization monitoring, emergency responses, 3D city mapping and reconstruction. However, the most important problem is to precisely localize buildings from single aerial images where there is no additional information such as LiDAR point cloud data or high resolution Digital Surface Models (DSMs). In this paper, a Deep Learning (DL)-based approach is proposed to localize buildings, estimate the relative height information, and extract the buildings’ boundaries using a single aerial image. In order to detect buildings and extract the bounding boxes, a Fully Connected Convolutional Neural Network (FC-CNN) is trained to classify building and non-building objects. We also introduced a novel Multi-Scale Convolutional-Deconvolutional Network (MS-CDN) including skip connection layers to predict normalized DSMs (nDSMs) from a single image. The extracted bounding boxes as well as predicted nDSMs are then employed by an Active Contour Model (ACM) to provide precise boundaries of buildings. The experiments show that, even having noises in the predicted nDSMs, the proposed method performs well on single aerial images with different building shapes. The quality rate for building detection is about 86% and the RMSE for nDSM prediction is about 4 m. Also, the accuracy of boundary extraction is about 68%. Since the proposed framework is based on a single image, it could be employed for real time applications.
Highlights
One of the most important applications of remotely sensed data focuses on the detection/extraction, identification, localization and characterization of man-made structures including buildings
* Corresponding author (Manno-Kovacs and Ozgun Ok, 2015), a combination of Gaussian Mixture Model (GMM) clustering and Conditional Random Field (CRF) classification algorithms (Li et al, 2015), a self-supervised decision fusion framework (Senaras and Yarman Vural, 2015), and a supervised segmentation algorithm based on the image descriptors (Dornaika et al, 2016)
Compared to those traditional methods applied to building detection, the Deep Learning methods such as Convolution Neural Networks (CNNs) are recently employed for urban image classification (Alidoost and Arefi, 2016; Makantasis et al, 2015; Saito and Aoki, 2015; Vakalopoulou et al, 2015; Yuan, 2016; Zhang et al, 2016)
Summary
One of the most important applications of remotely sensed data focuses on the detection/extraction, identification, localization and characterization of man-made structures including buildings. Some of these recent algorithms include an energy based optimization algorithm using the Local Gradient Orientation Density (LGOD) (Benedek et al, 2012), a graphbased algorithm using shadow information of buildings (Izadi and Saeedi, 2012; Ok et al, 2013), a combination of the k-means clustering algorithm and a Purposive FastICA model (Ghaffarian and Ghaffarian, 2014), the multi label graph partitioning strategy (Manno-Kovacs and Ozgun Ok, 2015), a combination of Gaussian Mixture Model (GMM) clustering and Conditional Random Field (CRF) classification algorithms (Li et al, 2015), a self-supervised decision fusion framework (Senaras and Yarman Vural, 2015), and a supervised segmentation algorithm based on the image descriptors (Dornaika et al, 2016) Compared to those traditional methods applied to building detection, the Deep Learning methods such as Convolution Neural Networks (CNNs) are recently employed for urban image classification (Alidoost and Arefi, 2016; Makantasis et al, 2015; Saito and Aoki, 2015; Vakalopoulou et al, 2015; Yuan, 2016; Zhang et al, 2016). The CNNs-based classification is employed to detect buildings automatically resulting the initial boundaries of buildings (e.g. bounding boxes), while the CNNs-based regression is used to
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
