Global–local optimizations by hierarchical cuts and climbing energies
作者：
Highlights：
• Theoretical results on partition selection from, hierarchies of segmentations.
• Framework: input image, hierarchy of segmentations, positive energy, optimization.
• Optimization by parent–child class comparisons, children composition laws.
• Dynamic programs, segmentation results on Berkeley database, indigenous dataset.
• Euclidean formulation, partial optimization, equivalence with max-flow.
摘要
Highlights•Theoretical results on partition selection from, hierarchies of segmentations.•Framework: input image, hierarchy of segmentations, positive energy, optimization.•Optimization by parent–child class comparisons, children composition laws.•Dynamic programs, segmentation results on Berkeley database, indigenous dataset.•Euclidean formulation, partial optimization, equivalence with max-flow.
论文关键词：Hierarchical segmentation,Climbing optimization,Mathematical morphology,Energy minimization,Dynamic programming
论文评审过程：Available online 22 May 2013.
论文官网地址：https://doi.org/10.1016/j.patcog.2013.05.012
