摘要
针对岩心CT图像中岩心区域与背景区域边界模糊,现有的分割方法无法有效分割出岩心区域,影响数字岩心模型的准确性问题,论文提出了基于改进线性迭代聚类(SLIC)的岩心背景分割算法优化分割效果。首先以图像复杂度为依据,得出图像预分割的超像素个数;其次对岩心和背景区域的相似超像素进行区域合并,减少后续冗余计算;最后根据岩心像素与背景像素区域像素值差异分割岩心背景。实验结果表明,论文算法有效分割出岩心区域,避免了阈值法分割对岩心区域的破坏,减少了用户干预,在解决岩心背景分割的完整性和有效性方面表现出了良好的性能。
Aiming at the blurry boundary between core area and background area in core CT images,the existing segmenta⁃tion methods cannot effectively segment core areas and affect the accuracy of digital core models.This paper proposes core back⁃ground segmentation based on improved linear iterative clustering(SLIC).The algorithm optimizes the segmentation effect.First,based on the complexity of the image,the number of superpixels pre-segmented is obtained.Second,similar superpixels of the core and background regions are merged to reduce subsequent redundant calculations.Finally,the pixel values of the core pixels and the background pixels are different.Split core background.Experimental results show that the algorithm effectively separates the core re⁃gion,the damage of the core region is avoided by threshold segmentation,user intervention is reduced,and good performance is demonstrated in solving the integrity and effectiveness of core background segmentation.
作者
陈国军
李胜
尹鹏
滕一诺
CHEN Guojun;LI Sheng;YIN Peng;TENG Yinuo(College of Computer Science and Technology,China University of Petroleum(East China),Qingdao 266580)
出处
《计算机与数字工程》
2021年第12期2639-2642,2664,共5页
Computer & Digital Engineering
关键词
图像分割
图像复杂度
超像素
区域合并
背景分割
image segmentation
image complexity
superpixel
region merging
background segmentation
