摘要： 针对重叠藻细胞显微图像的计数问题，提出一种基于活动轮廓模型的重叠藻细胞自动计数方法。对重叠藻细胞显微图像进行预处理，标记连通域，提取并处理局部图像，采用活动轮廓模型算法提取重叠藻细胞边缘，同时进行边缘分析和计数。与传统分水岭算法相比，该计数方法不直接进行重叠藻细胞分割，可以消除因直接分割时过分割与欠分割导致的计数误差，提高重叠藻细胞计算准确度。实验结果表明，该方法对重叠藻细胞的计数准确率高于90%。

关键词: 活动轮廓模型, 显微图像, 重叠藻细胞, 计数, 边缘检测, 分水岭算法

Abstract: A novel counting method based on active contour model is proposed to solve the difficulty in overlapped algae counting. A series of preprocessing methods are applied to obtain a binary image. The active contour model algorithm is used in edge detection after extracting each connected component region in the original image. The counting result comes out according to edge analysis. Compared to watershed algorithm, this method improves counting accuracy of overlapped algae cells by eliminating the error caused by overlapped particles segmentation. Experimental result shows this algorithm has good performance in overlapped algae counting with accuracy over 90%.

Key words: active contour model, microscopy image, overlapped algae cell, counting, edge detection, watershed algorithm

中图分类号: 

• TP391