CAJ | 학술논문

针对图像分割中的灰度不均匀和轮廓初始化问题，提出一种基于区域的活动轮廓模型。将图像的全局信息和局部信息作为能量项驱动活动轮廓向目标边缘演化，以有效分割灰度不均匀图像，为保证图像分割的速度和精度，在能量方程中加入长度项和惩罚项，并采用梯度下降法得到该模型的最小化能量方程。实验结果表明，和局部二值拟合模型、局部图像拟合模型相比，该模型能分割灰度不均匀的图像，对初始轮廓曲线大小和位置更不敏感，且分割图像所需的迭代次数、迭代时间更少。
침대도상분할중적회도불균균화륜곽초시화문제，제출일충기우구역적활동륜곽모형。장도상적전국신식화국부신식작위능량항구동활동륜곽향목표변연연화，이유효분할회도불균균도상，위보증도상분할적속도화정도，재능량방정중가입장도항화징벌항，병채용제도하강법득도해모형적최소화능량방정。실험결과표명，화국부이치의합모형、국부도상의합모형상비，해모형능분할회도불균균적도상，대초시륜곽곡선대소화위치경불민감，차분할도상소수적질대차수、질대시간경소。
In order to overcome the problem of weak boundary and intensity inhomogeneity, a region-based active contour model for image segmentation is proposed in this paper. Using the global and local image information as the energy term driving evolution of active contour to the objective boundaries can effectively segment images with intensity inhomogeneity. In order to segment the image fast and accurately,the length term and penalty are incorporated into the energy equation. By adopting gradient descent method, the minimization of the energy equation can be given. Segmentation tests demonstrate that the proposed method can segment images with intensity inhomogeneity, needs less iteration and few iteration times,and is less sensitive to the location and size of the initial contour when is compared with the Local Binary Fitting( LBF) model and the Local Image Fitting( LIF) model.