Level set framework with transcendental constraint for robust and fast image segmentation
作者：
Highlights：
• A fast and robust image segmentation method by taking the advantages of the region scalable fitting energy model and the advanced transcendental con straint term was proposed in this paper.
• Numerical results such as DICE values and computation times are presented.
• A parallel improvement is introduced into the proposed model, which makes it possible to get more accurate results efficiently.
• Comparative results show excellent performance of our model.
• The discussion on sensitivities of parameters selection shows that our model is robust.
摘要
•A fast and robust image segmentation method by taking the advantages of the region scalable fitting energy model and the advanced transcendental con straint term was proposed in this paper.•Numerical results such as DICE values and computation times are presented.•A parallel improvement is introduced into the proposed model, which makes it possible to get more accurate results efficiently.•Comparative results show excellent performance of our model.•The discussion on sensitivities of parameters selection shows that our model is robust.
论文关键词：Image segmentation,Split Bregman method,Parallel computaion,Transcendental constraint term,Level set formulation
论文评审过程：Received 13 June 2018, Revised 1 November 2019, Accepted 31 March 2021, Available online 15 April 2021, Version of Record 26 April 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.107985
