Road Extraction Using SVM and Image Segmentation

By Mingjun Song, Daniel L. Civco. 2004

300 citations recorded by the source corpus.

Accurate road information is vital for transportation applications, including as part of geographical information systems (GIS). This article reports on the development of a two-step approach for road extraction that utilizes pixel spectral information for classification and image segmentation-derived object features. In the first step, support vector machine (SVM) was employed merely to classify the image into two groups of categories: a road group and a non-road group. For this classification, support vector machine (SVM) achieved higher accuracy than Gaussian maximum likelihood (GML). In the second step, the road group image was segmented into geometrically homogeneous objects using a region growing technique based on a similarity criterion, with higher weighting on shape factors over spectral criteria. A simple thresholding on the shape index and density features derived from these objects was performed to extract road features, which were further processed by thinning and vectorization to obtain road centerlines. The authors conclude that the proposed approach worked well with images comprised by both rural and urban area features.

Subfields: Media Technology, Ocean Engineering, Signal Processing

View Paper: https://www.semanticscholar.org/paper/e429e4874f4b3a397a754f28a919362d282b7966

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Media Technology

In Ocean Engineering

In Signal Processing

Citations in the corpus

Road Extraction by Deep Residual U-Net (2017, methodology): “For instance, Song and Civco [11] proposed a method utilizing shape index feature and support vector machine (SVM) to detect road areas.”
Automatic Road Detection and Centerline Extraction via Cascaded End-to-End Convolutional Neural Network (2017, methodology): “In general, the existing road extraction methods can be roughly classiﬁed into two categories: road detection methods [5], [7]–[9], [32]–[34] and road centerline extraction methods [10], [13], [14], [18], [19], [35]–[39].”
Support vector machines in remote sensing: A review (2011, methodology): “Another road extraction work using Ikonos imagery was published by Song and Civco (2004).”
A Survey on Object Detection in Optical Remote Sensing Images (2016, unknown): “…Ok et al., 2013; Peng and Liu, 2005; Peng et al., 2005; Senaras et al., 2013; Shufelt, 1999; Sirmaçek and Ünsalan, 2009; Sirmacek and Ünsalan, 2011; Stankov and He, 2013, 2014; Wegner et al., 2011a; Wegner et al., 2011b), trees (Haala and Brenner, 1999; Hung et al., 2012; Malek et al., 2014;…”
Deep Learning‐Based Crack Damage Detection Using Convolutional Neural Networks (2017, methodology): “This study was followed by an examination of modiﬁed edge detection problems (Nishikawa et al., 2012; Alaknanda and Kumar, 2009; Yamaguchi et al., 2008; Sinha and Fieguth, 2006; Song and Civco, 2004).”