为了对细胞多光谱图像进行快速、准确的分割，首先探讨了光谱比值在细胞多光谱显微图像分割中的应用，然后提出了利用多光谱图像的光谱信息，并结合传统分割方法的一种新的细胞自动分割方法。该方法首先通过从扣除背底后的多光谱图像中选择两个波段图像进行光谱比值操作来生成一幅比率图像，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。该方法首次将光谱比值技术应用到细胞多光谱显微图像分割中，对骨髓细胞图像的自动分割实验表明，该方法具有分割准确、分割速度快、受外界干扰少的特点，该方法也可以推广到其他多光谱显微图像的分割中。

In this paper, the application of spectrum ratio to the segmentation of cell multi spectral micrograms is discussed and a new cell segmentation method is presented which combines spectrum information of multi spectral images with traditional segmentation methods. Firstly, background is removed from the multi spectral images. Secondly, two selected bandwidth images are divided to make a new ratio grey image. Thirdly, the new grey image is segmented with automatic multi threshold methods. 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 introduced into the segmentation of cell multi spectral micrograms. The experiments with a diagnosis system on marrow cell images prove that this approach for segmenting cell images is rather accurate, fast, and robust to noise, and this method can be extended to the segmentation of other types of multi spectral micrograms.