南京理工大学学报（自然科学版）2018，Vol.42Issue(3)：266-270,5.DOI:10.14177/j.cnki.32-1397n.2018.42.03.002
基于种子点替代像素块算法的粘连细胞图像分割
Segmentation of adherent cell image based on seed point substituting pixel block algorithm
摘要
Abstract
A seed point substitution algorithm based on pixel block scanning is proposed to separate overlapped cells with different adhesion degrees. On the basis of determining the average area of a cell,the side length of a square pixel block is determined. A two-value image is scanned by the pixel block from left to right,from top to bottom. The area matched to the pixel block is replaced with a seed point to get a seed point image. The watershed algorithm is used to get a segmented image and the cells are counted. Experiments show that the algorithm proposed here applies to adherent cell images with different adhesion degrees and nearly equal cell areas, the segmentation effect is remarkable,and the accuracy rate is above 98%.关键词
显微细胞图像/种子点/像素块/粘连细胞/细胞分割/分水岭分割算法Key words
microscopic cell images/seed points/pixel blocks/overlapped cells/cell segmentation/watershed algorithm分类
信息技术与安全科学引用本文复制引用
陈名,杨慧中..基于种子点替代像素块算法的粘连细胞图像分割[J].南京理工大学学报（自然科学版）,2018,42(3):266-270,5.基金项目
国家自然科学基金(61773181) （61773181）
中央高校基本科研业务费专项资金(JUSRP51733B) （JUSRP51733B）
