摘要
针对传统图像分割算法中存在的图像分割速度慢以及初始轮廓和参数鲁棒性差等问题,提出一种基于边力函数能量泛函的图像分割算法。通过顺序统计滤波边力函数进行数据拟合,提出边力函数能量泛函的概念,从而使得曲线快速自适应地向目标边界靠近。为了有效地调节水平集函数,引入优化后的长度项对曲线进行平滑和缩短处理,另外利用梯度下降法实现能量泛函的极小化。实验结果显示所提出的方法图像分割速度较快,分割精度较高,且对于初始轮廓与参数具有较强的鲁棒性。实验结果验证了该方法的有效性。
Aimed at the problems of slow image segmentation speed and poor robustness of initial contour and parameters in traditional image segmentation algorithms,an image segmentation algorithm based on edge force function energy functional is proposed.The data fitting was carried out by sequential statistical filtering edge force function.The concept of energy functional of edge force function was proposed,so that the curve could quickly and adaptively approach the target boundary.In order to adjust the level set function effectively,the optimized length term was introduced to smooth and shorten the curve,and the gradient descent method was used to minimize the energy functional.The experimental results show that the proposed method has fast image segmentation speed,high segmentation accuracy,and strong robustness to the initial contour and parameters.The experimental results verify the effectiveness of this method.
作者
唐银敏
李自纳
胡静
Tang Yinmin;Li Zina;Hu Jing(School of Information Engineering,Zhengzhou University,Zhengzhou 450000,Henan,China;School of Information Engineering,Nanyang Vocational College of Agriculture,Nanyang 473000,Henan,China)
出处
《计算机应用与软件》
北大核心
2023年第4期184-191,共8页
Computer Applications and Software
基金
2019年河南省职业教育教学改革研究项目(ZJC18089)。
关键词
鲁棒
快速图像分割
边力函数
能量泛函
Robust
Fast image segmentation
Edge force function
Energy functional
