目的:介绍了一种自动提取淋巴瘤病理图像中空泡状细胞核方法。方法:以空泡状细胞核为目标,采用在CMYK模式下的多层分割法分割病理图像;提出变尺寸特征窗口模糊分类法,对空泡状细胞核进行识别提取;并用近似最短连接法对破碎的空泡核进行修补连接。结论:实验结果表明,该方法可以有效对淋巴瘤病理图像进行分割,较准确地提取空泡状细胞核,并具有一定程度的自动修补连接破碎细胞核能力。 OBJECTIVE: To introduce a method of automatically extracting vacuolar cell nuclei from lymphoma pathological images. Methods: The vacuolar nucleus was taken as the target, and the pathological image was segmented by the multi-layer segmentation method in CMYK mode. A fuzzy classification method based on the feature window of variable size was proposed to identify and extract the vacuolar nucleus. The approximate shortest connection Cavitation nucleus repair connection. Conclusion: The experimental results show that this method can effectively divide the pathological image of lymphoma, extract the vacuolar nucleus more accurately and have a certain degree of ability to automatically repair the broken nucleus.