分段
数学
稳健性(进化)
分割
图像分割
正规化(语言学)
算法
正多边形
图像处理
图像(数学)
人工智能
计算机科学
数学分析
几何学
基因
生物化学
化学
作者
Yutong Li,Chunlin Wu,Yuping Duan
出处
期刊:IEEE transactions on image processing
[Institute of Electrical and Electronics Engineers]
日期:2020-01-01
卷期号:29: 7061-7075
被引量:17
标识
DOI:10.1109/tip.2020.2997524
摘要
The Mumford-Shah model is an important tool for image labeling and segmentation, which pursues a piecewise smooth approximation of the original image and the boundaries with the shortest length. In contrast to previous efforts, which use the total variation regularization to measure the total length of the boundaries, we build up a novel piecewise smooth Mumford-Shah model by utilizing a non-convex ℓ p regularity term for p ∈ (0,1), which can well preserve sharp edges and eliminate geometric staircasing effects. We present optimization algorithms with convergence verification, where all subproblems can be solved by either the closed-form solution or fast Fourier transform (FFT). The method is compared to piecewise constant labeling algorithm and several state-of-the-art piecewise smooth Mumford-Shah models based on image decomposition approximations. Both labeling and segmentation results on synthetic and real images confirm the robustness and efficiency of the proposed method.
科研通智能强力驱动
Strongly Powered by AbleSci AI