Building Change Detection Using Old Aerial Images and New LiDAR Data

Shouji Du,Zhengrong Zou,Yunsheng Zhang,Chong Fan,Yuqi Tang,Rongjun Qin,Zhihua Yang

doi:10.3390/rs8121030

Shouji Du, Zhengrong Zou + Show 5 more

Open Access

https://doi.org/10.3390/rs8121030

Copy DOI

Abstract

Building change detection is important for urban area monitoring, disaster assessment and updating geo-database. 3D information derived from image dense matching or airborne light detection and ranging (LiDAR) is very effective for building change detection. However, combining 3D data from different sources is challenging, and so far few studies have focused on building change detection using both images and LiDAR data. This study proposes an automatic method to detect building changes in urban areas using aerial images and LiDAR data. First, dense image matching is carried out to obtain dense point clouds and then co-registered LiDAR point clouds using the iterative closest point (ICP) algorithm. The registered point clouds are further resampled to a raster DSM (Digital Surface Models). In a second step, height difference and grey-scale similarity are calculated as change indicators and the graph cuts method is employed to determine changes considering the contexture information. Finally, the detected results are refined by removing the non-building changes, in which a novel method based on variance of normal direction of LiDAR points is proposed to remove vegetated areas for positive building changes (newly building or taller) and nEGI (normalized Excessive Green Index) is used for negative building changes (demolish building or lower). To evaluate the proposed method, a test area covering approximately 2.1 km2 and consisting of many different types of buildings is used for the experiment. Results indicate 93% completeness with correctness of 90.2% for positive changes, while 94% completeness with correctness of 94.1% for negative changes, which demonstrate the promising performance of the proposed method.

Highlights

The rapid process of urbanization has expedited the dynamics of the cities: numerous buildings are constructed or demolished every year in developing countries such as China
The profile is shown in Figur1e39ocf 2a2nd it iRnedmioctea2te01s6t, h9,e1s0e30two sets of point clouds are well aligned after the registration
This paper investigated the use of aerial images and light detection and ranging (LiDAR) data to determine the building changes

Summary

Introduction

The rapid process of urbanization has expedited the dynamics of the cities: numerous buildings are constructed or demolished every year in developing countries such as China. Pixel-based strategies lead to noisy outputs like isolated pixels, holes in the changed objects or jagged boundaries, since they mainly focus on the spectral values and mostly ignore the spatial context [8,9] To alleviate these problems, Hopfield neural network [10,11] and deterministic simulated annealing approach [12] have been used to consider the spatial context information to determine image changes. Hopfield neural network [10,11] and deterministic simulated annealing approach [12] have been used to consider the spatial context information to determine image changes Except from these methods, object-based methods were proposed to compare spectral values with pixel groups [13,14,15,16]. Research on change detection remains an active topic and new techniques are demanded to effectively use available data from satellite, airborne, even low-altitude platforms

Methods

Results

Discussion

Conclusion