Open AccessObject Oriented Road Extraction from Remote Sensing Images Using Improved Watershed Segmentation
Author(s) -
Dawei Liu,
Shujing Gao
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2005/1/012077
Subject(s) - watershed , segmentation , computer science , artificial intelligence , computer vision , extraction (chemistry) , process (computing) , image segmentation , object (grammar) , scale space segmentation , pattern recognition (psychology) , remote sensing , geography , chemistry , chromatography , operating system
A novel object oriented road extraction method is presented for the road extraction from remote sensing images. Firstly, an improved watershed algorithm is adopted for image segmentation, and the spectral, texture and geometric features of the image are fully considered in the segmentation process so as to improve the segmentation accuracy. Then road knowledge base is built and various features of the road are added into the base. Finally, the road extraction purpose is reached by calculating the features of the objects and comparing them with the knowledge base. Experiment results show that the method can achieve higher accuracy and quality in road extraction.