结合染色校正和卷积神经网络的细胞图像识别方法
首发时间:2016-11-29
摘要:细胞图像识别是医学图像处理中的重要组成部分。由于医学图像具有尺寸较大,细胞形状各异等特性导致细胞识别工作通常费时费力。因此,使用机器学习算法代替手工进行细胞图像识别工作变得十分的必要。作为深度学习的重要组成部分,卷积神经网络已成为目前机器学习领域的研究热点。本文以卷积神经网络为基础,结合染色校正方法,提出了一种具有多个输入层的卷积神经网络算法。实验表明,本文提出的算法与其它分类算法相比,具有更高的准确率与更优的识别效果。
For information in English, please click here
Cell Image Recognition Method Based on Stain Correction and Convolution Neural Network
Abstract:Cell image recognition is an important part of medical image processing. Because of the large size of medical image and the different cell shapes, the cell recognition is usually time-consuming and laborious. Therefore, it is necessary to use machine learning algorithms to replace manual cell image recognition. As an important part of deep learning, convolutional neural network has become a hot research topic in the field of machine learning. In this paper, a convolution neural network algorithm with multiple input layers is proposed, which is based on convolutional neural network combined with the method of stain correction. Experimental results show that the proposed algorithm has higher accuracy and better recognition performance compared with other recognition algorithms.
Keywords: cell recognition convolution neural networkkey stain correction
论文图表:
引用
No.4711112116357614****
同行评议
共计0人参与
勘误表
结合染色校正和卷积神经网络的细胞图像识别方法
评论
全部评论