LSDSAR, a Markovian a contrario framework for line segment detection in SAR images
作者:
Highlights:
• Line segment detection in SAR images remains a difficult problem because the line segment detectors proposed for optical images are not suitable for SAR images due to the strong multiplicative noise.
• By introducing a first order Markov chain in the a contrario model, we propose an automatic line segment detector for SAR images.
• Dealing with statistical dependencies in the a contrario framework is a crucial difficulty. The use of Markov Chain provides a simple and efficient way to overcome such issue.
摘要
•Line segment detection in SAR images remains a difficult problem because the line segment detectors proposed for optical images are not suitable for SAR images due to the strong multiplicative noise.•By introducing a first order Markov chain in the a contrario model, we propose an automatic line segment detector for SAR images.•Dealing with statistical dependencies in the a contrario framework is a crucial difficulty. The use of Markov Chain provides a simple and efficient way to overcome such issue.
论文关键词:Line segments,SAR images,A contrario models,Markov chain,Local orientations
论文评审过程:Received 18 March 2019, Revised 1 August 2019, Accepted 3 September 2019, Available online 3 September 2019, Version of Record 19 September 2019.
论文官网地址:https://doi.org/10.1016/j.patcog.2019.107034