Semiautomatic Extraction of Belt-like Roads from High Spatial Resolution Remotely Sensed Imagery based on Self-adaptively Variable Width of Template Matching Metho

A novel semiautomatic ribbon roads extraction method for high spatial resolution remotely sensed (HSRRS) imagery based on self-adaptively variable bandwidth template matching is proposed in this letter. At first, a human operator inputs three seed points to initialize the position of the road, including the starting points, road width and road direction. Then, a rectangular template is used to estimate the best way forward point by the template matching, so the automatic extraction is triggered. According to the actual road conditions, the method can increase or decrease the width of the rectangle template self-adaptively to extract complex roads by building four small rectangle templates. To finish the whole road network, the above tracking process is repeated. UCX aerial imagery are used to validate our methods, and the results show that compared with traditional profile template matching algorithm, our method not only decreases the number of tracing errors and manual intervention times significantly, but also improves the accuracy and efficiency of the semiautomatic extraction road.

• Publication year

  2019

• Venue

  2019 3rd International Conference on Robotics and Automation Sciences (ICRAS)

• Publication date

  2019-06-01

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1109/ICRAS.2019.8809027
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Automatic Road Detection and Centerline Extraction via Cascaded End-to-End Convolutional Neural Network

  2017cited by this paper
• An Integrated Method for Urban Main-Road Centerline Extraction From Optical Remotely Sensed Imagery

  2014cited by this paper
• Semi-automatic extraction of road networks by least squares interlaced template matching in urban areas

  2011influential reference
• Semi-automatic Extraction of Ribbon Roads from High Resolution Remotely Sensed Imagery Based on Angular Texture Signature and Profile Match

  2008cited by this paper
• Online Learning With Novelty Detection in Human-Guided Road Tracking

  2007cited by this paper
• A pixel shape index coupled with spectral information for classification of high spatial resolution remotely sensed imagery

  2006cited by this paper
• The recognition of road network from high‐resolution satellite remotely sensed data using image morphological characteristics

  2005cited by this paper
• Towards an operational system for automated updating of road databases by integration of imagery and geodata

  2004cited by this paper
• A Robust Method for Semi-Automatic Extraction of Road Centerlines Using a Piecewise Parabolic Model and Least Square Template Matching

  2004cited by this paper
• Road extraction from aerial and satellite images by dynamic programming

  1995cited by this paper
• Semiautomatic extraction of roads from aerial images

  1994cited by this paper
• Automatic finding of main roads in aerial images by using geometric-stochastic models and estimation

  1993cited by this paper
• Cooperative methods for road tracking in aerial imagery

  1988cited by this paper
• SEMI-AUTOMATIC ROAD EXTRACTION FROM HIGH-RESOLUTION SATELLITE IMAGE

  year unknowncited by this paper

• DRCNet: Road Extraction From Remote Sensing Images Using DenseNet With Recurrent Criss-Cross Attention and Convolutional Block Attention Module

  2023cites this paper