一种基于光谱比值的细胞多光谱图像自动分割方法  

An Automatic Segmentation Scheme for Cell Multi-spectral Micrograms Based on Spectrum Ratio

作　　者：张燕[1] 曾立波[1] 王思贤[1] 吴琼水[1] 

机构地区：[1]武汉大学电子信息学院,武汉430079

出　　处：《计算机工程与应用》2005年第5期218-220,共3页Computer Engineering and Applications

基　　金：国家科技部"十五"攻关项目"光谱成象分析系统的研制与开发"(编号:2001BA210A02)

摘　　要：提出了一种基于光谱比值的细胞多光谱显微图像的自动分割方法。首先,从扣除背底后的多光谱图像中选择两个最佳波段图像相除,生成一幅比率图像,然后对该图像进行自动多阈值分割、二值形态学操作,最终获得了细胞的胞浆和胞核覆盖层。首次将光谱比值技术应用到细胞多光谱显微图像分割中,对骨髓细胞的自动分割试验证明该方法具有分割准确、分割速度快、受外界干扰少的特点。This paper presents an automatic segmentation scheme for cell multi-spectral micrograms based on spectrum ratio.Firstly,background is removed from the multi-spectral images.Secondly,two selected optimal bandwidth images are divided to make a new ratio grey image.Thirdly,the new grey image is separated with automatic threshold.Finally,the segmented regions are smoothed by morphological operations and the overlays of the plasm and the nucleus are obtained.In this paper,spectrum ratio technique is firstly implemented in the segmentation of cell multi-spectral micrograms.The experimental result of marrow cell images proves that this approach for segmenting cell images is rather accurate,fast,and robust to noise.

关 键 词：分割 多光谱 光谱比值 骨髓细胞 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]