基于U-net网络的细胞核检测方法  

Cell detection method based on U-net network

作　　者：秦晨阳 应捷[1] 杨海马[1] QIN Chenyang;YING Jie;YANG Haima(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093

出　　处：《光学仪器》2021年第1期1-5,共5页Optical Instruments

摘　　要：为节省时间和研究人员的精力,采用计算机辅助对细胞或细胞核进行检测。利用卷积神经网络的U-net衍生网络并结合图像处理过程开展对细胞核的检测。研究结果表明,该检测方法的检测精确度为0.82,召回率为0.83,F指标为0.83,具有较好的识别和分割效果。In order to save time and researchers'energy,computers are used to assist in the detection of cells or nuclei.In this paper,a derivative network U-net of convolution neural network is combined with image processing to detect nucleus.The result shows that the accuracy is 0.82,the recall rate is 0.83,and the F index is 0.83,which has good recognition and segmentation effect.

关 键 词：神经网络 图像处理 细胞核检测 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]