摘要
针对PDE(partial differential equation)图像分割模型-RSF(region-scalable fitting energy)模型对初始轮廓线选择敏感问题,提出根据图像的灰度变化信息动态选择高斯核函数窗口大小的改进RSF模型.实验表明,该模型提高了RSF模型对初始轮廓线的鲁棒性.
RSF( region- scalable fitting energy) model is a famous PDE( partial differential equation) image segmentation model,which is sensitive to initialization. To address this problem,a modified RSF model whose the window size of Gaussian kernel function to each pixel in images be selected dynamically is proposed. The window sizes of Gaussian kernel functions of the model depend on the intensity of images. The experimental results show that the proposed model allows for more robustness to initialization compared to the original RSF model.
出处
《福州大学学报(自然科学版)》
CAS
北大核心
2016年第3期413-418,共6页
Journal of Fuzhou University(Natural Science Edition)
基金
国家自然科学基金资助项目(11071270)
关键词
图像分割
RSF模型
高斯核函数
边缘停止函数
image segmentation
region-scalable fitting energy model
Gaussian kernel function
edge stopping function