Building Detection in Complex Scenes Thorough Effective Separation of Buildings from Trees

Awrangjeb, Mohammad; Zhang, Chunsun; Fraser, Clive S.

doi:10.14358/pers.78.7.729

Cited by 66 publications

(75 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The effectiveness of using image texture features for remote sensing data classification has been justified by several studies [39][40][41]. Similar to [42], the following image texture features are calculated in this work:…”

Section: Image Texture Featuresmentioning

confidence: 99%

High-Resolution Remote Sensing Data Classification over Urban Areas Using Random Forest Ensemble and Fully Connected Conditional Random Field

Sun

Lin

Shen

et al. 2017

IJGI

View full text Add to dashboard Cite

Abstract:As an intermediate step between raw remote sensing data and digital maps, remote sensing data classification has been a challenging and long-standing problem in the remote sensing research community. In this work, an automated and effective supervised classification framework is presented for classifying high-resolution remote sensing data. Specifically, the presented method proceeds in three main stages: feature extraction, classification, and classified result refinement. In the feature extraction stage, both multispectral images and 3D geometry data are used, which utilizes the complementary information from multisource data. In the classification stage, to tackle the problems associated with too many training samples and take full advantage of the information in the large-scale dataset, a random forest (RF) ensemble learning strategy is proposed by combining several RF classifiers together. Finally, an improved fully connected conditional random field (FCCRF) graph model is employed to derive the contextual information to refine the classification results. Experiments on the ISPRS Semantic Labeling Contest dataset show that the presented 3-stage method achieves 86.9% overall accuracy, which is a new state-of-the-art non-CNN (convolutional neural networks)-based classification method.

show abstract

Section: Image Texture Featuresmentioning

confidence: 99%

High-Resolution Remote Sensing Data Classification over Urban Areas Using Random Forest Ensemble and Fully Connected Conditional Random Field

Sun

Lin

Shen

et al. 2017

IJGI

View full text Add to dashboard Cite

show abstract

“…Figure 4b shows different classes of the extracted image lines. In terms of completeness and correctness (Awrangjeb et al, 2012) the classification performance was, 'ground': 89% and 93%, 'tree': 96% and 98%, 'roof edge': 87% and 71% and 'roof ridge': 95% and 95%. Although 'tree' and 'roof ridge' classes were identified with high accuracy, there were reasons why 'ground' and 'roof edge' classes were not always so accurate.…”

Section: Line Classificationmentioning

confidence: 98%

Automatic Reconstruction of Building Roofs Through Effective Integration of Lidar and Multispectral Imagery

Awrangjeb

Zhang

Fraser

2012

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Automatic 3D reconstruction of building roofs from remotely sensed data is important for many applications including city modeling. This paper proposes a new method for automatic 3D roof reconstruction through an effective integration of LIDAR data and multispectral imagery. Using the ground height from a DEM, the raw LIDAR points are separated into two groups. The first group contains the ground points that are exploited to constitute a 'ground mask'. The second group contains the non-ground points that are used to generate initial roof planes. The structural lines are extracted from the grey-scale version of the orthoimage and they are classified into several classes such as 'ground', 'tree', 'roof edge' and 'roof ridge' using the ground mask, the NDVI image (Normalised Difference Vegetation Index from the multi-band orthoimage) and the entropy image (from the grey-scale orthoimage). The lines from the later two classes are primarily used to fit initial planes to the neighbouring LIDAR points. Other image lines within the vicinity of an initial plane are selected to fit the boundary of the plane. Once the proper image lines are selected and others are discarded, the final plane is reconstructed using the selected lines. Experimental results show that the proposed method can handle irregular and large registration errors between the LIDAR data and orthoimagery.

show abstract

“…The second group consists of the non-ground points that represent elevated objects, such as buildings and trees with heights above T h . Many authors (e.g., [37,38]) have used h c = 2.5 m. However, for the Vaihingen data set, it was observed that this height threshold removes many low buildings and sheds. Therefore, h c has been set at 1 m for this study.…”

Section: Lidar Classification and Mask Generationmentioning

confidence: 99%

Automatic Segmentation of Raw LIDAR Data for Extraction of Building Roofs

2014

Self Cite

View full text Add to dashboard Cite

Automatic extraction of building roofs from remote sensing data is important for many applications, including 3D city modeling. This paper proposes a new method for automatic segmentation of raw LIDAR (light detection and ranging) data. Using the ground height from a DEM (digital elevation model), the raw LIDAR points are separated into two groups. The first group contains the ground points that form a "building mask". The second group contains non-ground points that are clustered using the building mask. A cluster of points usually represents an individual building or tree. During segmentation, the planar roof segments are extracted from each cluster of points and refined using rules, such as the coplanarity of points and their locality. Planes on trees are removed using information, such as area and point height difference. Experimental results on nine areas of six different data sets show that the proposed method can successfully remove vegetation and, so, offers a high success rate for building detection (about 90% correctness and completeness) and roof plane extraction (about 80% correctness and completeness), when LIDAR point density is as low as four points/m 2 . Thus, the proposed method can be exploited in various applications.

show abstract

Building Detection in Complex Scenes Thorough Effective Separation of Buildings from Trees

Cited by 66 publications

References 28 publications

High-Resolution Remote Sensing Data Classification over Urban Areas Using Random Forest Ensemble and Fully Connected Conditional Random Field

High-Resolution Remote Sensing Data Classification over Urban Areas Using Random Forest Ensemble and Fully Connected Conditional Random Field

Automatic Reconstruction of Building Roofs Through Effective Integration of Lidar and Multispectral Imagery

Automatic Segmentation of Raw LIDAR Data for Extraction of Building Roofs

Contact Info

Product

Resources

About