Efficient variational segmentation with local intensity fitting for noisy and inhomogeneous images

Po Wen Hsieh, Chung Lin Tseng, Suh Yuh Yang

研究成果: 雜誌貢獻期刊論文同行評審

摘要

This paper introduces a novel local intensity fitting energy model for segmenting noisy and intensity inhomogeneous images. A notable feature of the proposed model is its ability to simultaneously segment the image while obtaining a denoised and inhomogeneity-corrected result. The model integrates a local clustering criterion function with a denoising mechanism, in which the total energy functional comprises three key components: a local fitting energy on the denoised image, which generates a local force to attract the segmentation contour towards the expected object boundary; an edge detector-dependent smoothing term to denoise the source image, and a length regularization ensuring precise wrapping of the segmentation contour around the target object. In addition, we employ an efficient iterative convolution-thresholding method to solve the associated energy minimization problem, ensuring energy decay at each iteration. We demonstrate the efficacy and efficiency of our proposed variational image segmentation model through numerical experiments conducted on both synthetic and real images.

原文???core.languages.en_GB???
文章編號277
期刊Multimedia Systems
30
發行號5
DOIs
出版狀態已出版 - 10月 2024

指紋

深入研究「Efficient variational segmentation with local intensity fitting for noisy and inhomogeneous images」主題。共同形成了獨特的指紋。

引用此