Variational level set method for image segmentation with simplex constraint of landmarks
作者：
Highlights：
• The preferred image point landmarks were expressed as landmark data fidelity term with the framework of the level set method.
• The simplified SIFT descriptors were utilized to generate initial point landmarks.
• A variational image segmentation model combing preferred point landmarks with the level set framework and the simplex constraint was presented.
• A fast comprehensive alternating minimization iteration algorithm integrating the simplex projection method was designed to solve the segmentation problem.
• The segmentation results exhibit an excellent segmentation performance.
摘要
•The preferred image point landmarks were expressed as landmark data fidelity term with the framework of the level set method.•The simplified SIFT descriptors were utilized to generate initial point landmarks.•A variational image segmentation model combing preferred point landmarks with the level set framework and the simplex constraint was presented.•A fast comprehensive alternating minimization iteration algorithm integrating the simplex projection method was designed to solve the segmentation problem.•The segmentation results exhibit an excellent segmentation performance.
论文关键词：Image segmentation,Point landmarks,Simplex constraint,Alternating direction method of multipliers,Variational level set method
论文评审过程：Received 24 October 2019, Revised 9 December 2019, Accepted 11 December 2019, Available online 14 December 2019, Version of Record 31 December 2019.
论文官网地址：https://doi.org/10.1016/j.image.2019.115745
