结合卷积神经网络和超像素聚类的细胞图像分割方法

针对细胞图像尺寸大、细胞形状各异，导致从图像中分割出精准的细胞十分困难的问题，以卷积神经网络为基础，结合染色校正方法和简单线性迭代的超像素聚类算法，提出了一种新的结构来进行细胞图像分割。首先，利用染色校正方法对细胞图像进行预处理，提高图像的颜色对比度；然后利用卷积神经网络获得初步分割结果；最后再将简单线性迭代聚类获得的超像素边界信息反馈到初分割图像上进行改进提升。提出的算法可以有效地减少图像局部信息的冗余，更准确地获得目标区域的边界位置。实验表明，本文提出的算法细胞分割准确率达到了92.72%，与经典卷积神经网络、阈值分割等其他细胞分割算法相比，具有更好的分割效果。

Cell Image Segmentation Method Based on Convolution Neural Network and Super Pixel Clustering

Cell images have the characteristics of large sizes and different cell shapes. Therefore, it is very difficult to segment the precise cells from the medical images. Based on the convolution neural network, combining with the stain regularization method and the simple linear iterative clustering algorithm, this paper proposed a new structure for cell segmentation. First of all, stain regularization method can improve the color contrast of the images in preprocessing. Then the convolution neural network can get the initial cell segmentation result. At the same time, using super pixel clustering method can obtain the boundary information of cell images. Finally the boundary information is feedback to the initial cell segmentation result to improve the accuracy of the segmentation. The proposed algorithm can effectively reduce the redundant local information in image, and get the boundary of the target cell area more accurately. Experimental result shows that the segmentation accuracy of the proposed algorithm is 92.72%. Compared with the classical convolution neural network and other cell segmentation methods such as the threshold segmentation method, the proposed algorithm has better segmentation results.