Road Extraction Using Topological Derivative and Mathematical Morphology
- PDF / 552,121 Bytes
- 6 Pages / 547.087 x 737.008 pts Page_size
- 58 Downloads / 199 Views
SHORT NOTES
Road Extraction Using Topological Derivative and Mathematical Morphology P. N. Anil · S. Natarajan
Received: 26 December 2011 / Accepted: 25 July 2012 © Indian Society of Remote Sensing 2013
Abstract Road extraction from remotely sensed images has always been a challenging problem. In this paper we present an approach for road extraction based on topological derivative and mathematical morphology. The road extraction scheme has three main steps: image segmentation using topological derivative, road cluster identification and road cluster filtering using mathematical morphology. Keywords Topological derivative · Road cluster · Mathematical morphology
Introduction Road extraction from remotely sensed imagery has been an active research area in computer vision and digital photogrammetry. There are many
P. N. Anil (B) Dept. of Mathematics, Global Academy of Technology, Rajarajeswari Nagar Ideal Home Township, Bangalore 560098, India e-mail: [email protected] S. Natarajan Dept. of Information Science and Engineering, PES Institute of Technology, 100 Feet Ring Road, Bangalore, India e-mail: [email protected]
algorithms for road extraction, differing in their input data, their goals and the methods used to attain these goals. Most of the works published in literature on road extraction from remotely sensed imagery can be classified in to two categories: Semi-automatic and automated extraction. Semi-automatic road extraction is an interactive process between an operator and computer algorithms. In such methods, an operator selects initial seed point(s) and a direction for road tracking algorithm. Vosselman and de Knecht (1995) proposed a semi-automatic road tracking algorithm which combines least square matching of grey value profiles with the Kalman filter. A dynamic programming based approach was presented in Gruen and Li (1997). Agouris et al. (2001) proposed a road extraction from digital imagery using deformable contour models. A semi automatic road extraction algorithm using template matching was proposed by Park and Kim (2001). Zhou et al. (2005) present a semiautomatic road tracking system based on particle filtering and human–computer interactions. Matthew et al. (2007) present a road extraction method by using a family of cooperating snakes. An approach for road extraction based on active contour model is presented by Matthew et al. (2007) as well as Anil and Natarajan (2010b). On the other hand, automatic road extraction is fully computer dependent without any human intervention and has been always an interesting
J Indian Soc Remote Sens
subject for researchers. Chiang et al. (2001) proposed an automatic road extraction from aerial images based on Kalman filtering and snakes. A method to detect road network from high resolution image using combination of fuzzy system and mathematical morphology is presented in Mohammadzadeh et al. (2004). Zhang and Couloigner (2004) present a wavelet approach to road extraction from high spatial resolution remotely sensed imagery. Doucette et al. (2004) pr
Data Loading...
