流域算法对血及骨髓涂片的彩色图像分割

为了开发血及骨髓涂片中血细胞自动分类及计算机辅助诊断系统，提出 了一个彩色流域算法，以完成对有核血细胞（胞核和胞浆）的分割和识别。通过彩色空间变换，有效地利用了图像中血细胞的颜色信息及先验知识，并且为了抑制过度分割，合理地对背景、成熟红细胞及有核细胞进行了标记，在彩色梯度图像上完成了基于淹没的流域算法，其中淹没过程通过多级循环队列数据结构完成，胞核和胞浆的分割正确率分别为91.9%和88.4%，研究结果表明，该算法对彩色血细胞图像的分割是有效的，它必将提高血细胞自动分类系统的性能。

Applying Watershed Algorithm to Segmentation of Color Imag es from Blood and BoneMarrow Smears

To develop an automatic classifying and diagnostic system for the hematopoietic cells from the blood and bone marrow smears stained with Wright, a color watershed algorithm is proposed for segmentation and recognition of the nucleated hematopoietic cells (including nucleus and cytoplasm). In order to prevent over segmentation, color imformation and prior knowledge is used to mark the background, erythrocytes and nucleated cells. The present algorithm is implemented in red green blue (RGB) color gradient images and based on an immersion process analogy, in which the flooding of water in the image is effciently simulated by hierarchical cycle queues of pixels. The average success rate is 91.9% for nuclei and 88.4% for cytoplasm. The result shows that the method is valid and efficient to segment color images from blood and bone marrow smears.