血细胞图像的自适应分割  

AUTO-SEGMENTATION OF BLOOD CORPUSCLE IMAGE

作　　者：王兆仲[1] 赵宇[1] 周付根[1] 姜志国[1] 

机构地区：[1]北京航空航天大学图像中心,北京100083

出　　处：《中国体视学与图像分析》2002年第1期49-52,共4页Chinese Journal of Stereology and Image Analysis

摘　　要：目的 :对血细胞图像进行自适应分割。方法 :首先将RGB图像变换到HSV(Hue/Saturation/Value)空间 ,清除图像中不具有饱和度信息的杂质 ;然后利用形态学重建滤波器作滤波 ,在保持目标边缘性质的同时滤掉脉冲干扰 ,并消除图像照度不均匀的影响 ;最后用Watershed变换完成图像分割。结果 :实现了低质量血细胞图像的自适应分割。结论 :该方法具有很强的抗干扰能力 ,分割结果令人满意。Aim: To segment the blood corpuscles automatically from their images. Methods: The RGB image was transferred to HSV model with the impurities of their small S value erased. Then the pulse noise and uniform illumination of the image were filtered. Finally the blood corpuscles were picked up from the image by using watershed transformation. Results: Auto-segmentation of poor quality blood corpuscle image was achieved. Conclusions: The method has a powerful anfi-interference effect and can achieve satisfactory segmentation.

关 键 词：图像分割 医学图像处理 HSV彩色模型 数学形态学 血细胞图像 

分 类 号：R446.1[医药卫生—诊断学]