Directional Segmentation Based on Shear Transform and Shape Features for Road Centerlines Extraction from High Resolutio
Accurate extraction of road networks from high resolution remote sensing images is a problem not satisfactorily solved by existing approaches, especially when the color of road is close to that of background. This paper studies a new road networks extract
- PDF / 1,497,176 Bytes
- 9 Pages / 439.37 x 666.14 pts Page_size
- 61 Downloads / 147 Views
stract. Accurate extraction of road networks from high resolution remote sensing images is a problem not satisfactorily solved by existing approaches, especially when the color of road is close to that of background. This paper studies a new road networks extraction from remote sensing images based on the shear transform, the directional segmentation, shape features and a skeletonization algorithm. The proposed method includes the following steps. Firstly, we combine shear transform with directional segmentation to get road regions. Secondly, road shape features filtering are used to extract reliable road segments. Finally, the road centerlines are extracted by a skeletonization algorithm. Road networks are then generated by post-processing. Experimental results show that this method is efficient in road centerlines extraction from remote sensing images. Keywords: Road centerlines extraction · Shear transform · Directional segmentation · Shape features
1
Introduction
Roads are the backbone and essential modes of transportation, providing many different supports for human civilization. Road extraction plays a very important role in vehicle navigation system, urban planning, disaster management system and traffic management system. Due to the improvement of image resolution, the image has all sorts of detailed information to obtain very good reflection, but these details characteristics are interference for the extraction of road. Also, high-resolution satellite images have serious shadows, particularly in urban areas, which have an impact on road extraction, so the road extraction from high-resolution satellite images has a great scientific significance. In recent years, various road extraction algorithms have been proposed. A variety of road detection techniques[1]include knowledge based methods[2], mathematical morphology[3],[4], snakes[5]–[6], classification[7]–[10], differential geometry[11], region competition [12], active testing [13], perceptual grouping [14], and dynamic programming [15]. Mena [16] and Fortier et al. [17] provide extensive surveys of the literature on road extraction technique. Although the above methods show a good performance in road extraction, it is difficult to get a satisfactory result [18], and we need do some further research. © Springer-Verlag Berlin Heidelberg 2015 H. Zha et al. (Eds.): CCCV 2015, Part II, CCIS 547, pp. 1–9, 2015. DOI: 10.1007/978-3-662-48570-5_1
2
R. Liu et al.
Chaudhuri et al. [19] proposed a semi-automatic road detection method. In this method there were only a small set of directions to be used to detect the road segment. Thus some road segments are not detected. In order to solve the above problem, we would like to develop an efficient algorithm. This paper proposes a method based on shear transform, the directional segmentation, shape features and a skeletonization algorithm. The organization of this paper is as follows: In Section 2, the new method is described. In Section 3, we compare the experimental result with a semi-automatic road detection method
Data Loading...
