【摘要】 在尿沉渣图像中经常会出现细胞重叠和粘连的现象。将粘连的细胞分离是尿沉渣自动检测系统中一项重要而困难的任务。针对这一情况,提出一种基于组合思想的分割算法。该算法首先利用改进的自适应阈值分割算法结合形态学处理对细胞进行定位得到子图像,然后再利用基于距离变换的分水岭算法对子图像进行粘连细胞分离。实验结果证明,该算法能准确、稳定地分割尿沉渣粘连细胞。同时,该算法明显地提高了尿沉渣自动检测系统的性能。更多还原

【Abstract】 The cell overlapping and adhesion phenomena often appear in image of urinary sediment.Separating overlapped cells to individual ones is a great important and difficult task in automatic urinary sediment examination system.To solve this problem,an efficient segmenting algorithm based on combination strategy is proposed in this paper.The algorithm first locates the overlapped cells by the improved adaptive thresholding segmentation combining with mathematical morphology to get the sub-image,and then adopts the watersheds algorithm based on distance transform to segment the overlapping cells on sub-image.The experimental results prove that the algorithm can segment the overlapping cells in urinary sediment precisely and stably.Furthermore,the algorithm apparently increases the performance of the automatic urinary sediment examination system.更多还原