计算机工程
計算機工程
계산궤공정
COMPUTER ENGINEERING
2013年
11期
259-263
,共5页
图像分割%活动轮廓模型%水平集%分裂Bregman%灰度不均匀%边界检测函数
圖像分割%活動輪廓模型%水平集%分裂Bregman%灰度不均勻%邊界檢測函數
도상분할%활동륜곽모형%수평집%분렬Bregman%회도불균균%변계검측함수
image segmentation%active contour model%level set%split Bregman%intensity inhomogeneity%edge detection function
针对灰度不均匀图像的分割问题,提出一个基于区域的活动轮廓模型。通过构造包含图像局部信息的局部图像拟合偏差能量泛函,度量真实图像与拟合图像的偏差,并在全局凸分割的基础上,将分裂Bregman技术应用到模型能量泛函的最小化问题中,以提高分割速率。同时引入边界检测函数更加准确地探测边界位置,以提高模型的分割准确性。实验结果表明,该模型不仅可以正确分割灰度不均匀图像和受噪声干扰的图像,而且对于多目标图像以及灰度分布均值相同、方差不同的图像,也能快速、准确地得到分割结果。
針對灰度不均勻圖像的分割問題,提齣一箇基于區域的活動輪廓模型。通過構造包含圖像跼部信息的跼部圖像擬閤偏差能量汎函,度量真實圖像與擬閤圖像的偏差,併在全跼凸分割的基礎上,將分裂Bregman技術應用到模型能量汎函的最小化問題中,以提高分割速率。同時引入邊界檢測函數更加準確地探測邊界位置,以提高模型的分割準確性。實驗結果錶明,該模型不僅可以正確分割灰度不均勻圖像和受譟聲榦擾的圖像,而且對于多目標圖像以及灰度分佈均值相同、方差不同的圖像,也能快速、準確地得到分割結果。
침대회도불균균도상적분할문제,제출일개기우구역적활동륜곽모형。통과구조포함도상국부신식적국부도상의합편차능량범함,도량진실도상여의합도상적편차,병재전국철분할적기출상,장분렬Bregman기술응용도모형능량범함적최소화문제중,이제고분할속솔。동시인입변계검측함수경가준학지탐측변계위치,이제고모형적분할준학성。실험결과표명,해모형불부가이정학분할회도불균균도상화수조성간우적도상,이차대우다목표도상이급회도분포균치상동、방차불동적도상,야능쾌속、준학지득도분할결과。
In order to segment images with intensity inhomogeneity, a new region-based active contour model is proposed. By introducing the Local Image Fitting Bias(LIFB) energy function that embeds the image local information, it can measure the difference between the original image and the fitted image. Moreover, based on the globally convex segmentation method, the split Bregman technique is applied to minimize the proposed energy function more efficiently. By using an edge detection function to the proposed model, the algorithm can detect the boundaries more accurately. Experimental results show that the proposed model not only can segment images with intensity inhomogeneity and images corrupted by noise, but also can efficiently and accurately segment multi-object images and images with similar intensity means but different variances.