首页 > 论文 > 激光与光电子学进展 > 56卷 > 6期(pp:61005--1)

基于U-Net卷积神经网络的纳米颗粒分割

Nanoparticle Segmentation Based on U-Net Convolutional Neural Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为了准确测量纳米颗粒的尺寸,依据透射电子显微镜拍摄的纳米颗粒图像,提出了一种基于U-Net卷积神经网络的颗粒自动分割方法。将U-Net部分网络结构与批量归一化层相结合,减弱了网络对初始化的依赖,提升了训练速度。对纳米颗粒图像进行半隐式偏微分方程滤波以增强图像边缘信息,利用改进的U-Net网络训练纳米颗粒个体分割模型,得到了分割结果。研究结果表明,所提方法能准确分割出图像中的纳米颗粒,对边缘模糊和强度不均的纳米颗粒的分割效果提升显著。

Abstract

In order to accurately measure the size of nanoparticles, an automatic particle segmentation method based on U-Net convolutional neural network is proposed according to the nanoparticle images captured by the transmission electron microscopy. Combined with the Batch Normalization (BN) layer, it reduces the dependence of networks on initialization and thus speeds up training. The nanoparticle image is filtered by the semi-implicit partial differential equation to enhance the image edge information. The improved U-Net network is used to train the nanoparticle individual segmentation model and the segmentation result is obtained. The research results show that the proposed method can accurately segment the nanoparticles in the image, and the segmentation effect is especially obvious for the nanoparticles with edge blurs and uneven intensities.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391

DOI:10.3788/lop56.061005

所属栏目:图像处理

基金项目:国家自然科学基金(61601325)、天津市科技重大专项(17ZXSCSY00060)、天津市高等学校创新团队培养计划(TD13-5034)

收稿日期:2018-09-18

修改稿日期:2018-09-28

网络出版日期:2018-10-17

作者单位    点击查看

张芳:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387
吴玥:天津工业大学电子与信息工程学院, 天津 300387
肖志涛:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387
耿磊:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387
吴骏:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387
刘彦北:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387
王雯:天津工业大学电子与信息工程学院, 天津 300387天津市光电检测技术与系统重点实验室, 天津 300387

联系人作者:肖志涛(xiaozhitao@tjpu.edu.cn)

【1】Pan L C, Ge B Z, Zhang F G. Laser particle size measurement based on annular sample cell[J]. Acta Optica Sinica, 2017, 37(10): 1029001.
潘林超, 葛宝臻, 张福根. 基于环形样品池的激光粒度测量方法[J]. 光学学报, 2017, 37(10): 1029001.

【2】He F L, Guo Y C, Gao C. Improved PCNN method for human target infrared image segmentation under complex environments[J]. Acta Optica Sinica, 2017, 37(2): 0215003.
贺付亮, 郭永彩, 高潮. 复杂环境下用于人体目标红外图像分割的改进PCNN方法[J]. 光学学报, 2017, 37(2): 0215003.

【3】Fang J X. Study based on the variational level set methods for image segmentation[D]. Shanghai: Shanghai Jiao Tong University, 2012: 20-28.
方江雄. 基于变分水平集的图像分割方法研究[D]. 上海: 上海交通大学, 2012: 20-28.

【4】Liu X B. The study on the segmentation of circle-like particle images[D]. Changsha: Hunan University, 2006.
刘相滨. 类圆性颗粒图像分割技术研究[D]. 长沙: 湖南大学, 2006.

【5】Ni K, Wu Y Q, Geng S. Segmentation of metallographic image based on improved CV model integrated with local fitting term[J]. Acta Optica Sinica, 2018, 38(4): 0411009.
倪康, 吴一全, 庚嵩. 基于改进区域项CV模型的金相图像分割[J]. 光学学报, 2018, 38(4): 0411009.

【6】Li C M, Kao C Y, Gore J C, et al. Minimization of region-scalable fitting energy for image segmentation[J]. IEEE Transactions on Image Processing, 2008, 17(10): 1940-1949.

【7】Zhang B, Ni K Z, Wang L J, et al. New algorithm of detecting optical surface imperfection based on background correction and image segmentation[J]. Acta Optica Sinica, 2016, 36(9): 0911004.
张博, 倪开灶, 王林军, 等. 基于背景校正和图像分割定量分析光学元件表面疵病的新算法[J]. 光学学报, 2016, 36(9): 0911004.

【8】Wang Y, Chen J, Wang G. Particle seed images segmentation method based on the improved fuzzy C-means clustering algorithm[J]. Journal of North University of China(Natural Science Edition), 2018, 39(2): 177-182.
王宇, 陈婧, 王高. 基于改进的模糊C均值聚类算法的颗粒种子图像分割方法[J]. 中北大学学报(自然科学版), 2018, 39(2): 177-182.

【9】Lü J C, Wang L P, Zhang Q Q. Research on the background segmentation method based on nano-particles images[J]. Journal of Baicheng Normal University, 2017, 31(12): 1-6.
吕镜潺, 王丽萍, 张倩倩. 基于纳米颗粒图像的背景分割方法探究[J]. 白城师范学院学报, 2017, 31(12): 1-6.

【10】Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(4): 640-651.

【11】Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.

【12】Bansal A, Chen X L, Russell B, et al. PixelNet: representation of the pixels, by the pixels, and for the pixels[EB/OL]. (2017-02-27)[2018-11-23]. https:∥arxiv.org/abs/1702.06506

【13】Ronneberger O, Fischer P, Brox T.U-Net: Convolutional networks for biomedical image segmentation[C]∥International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2015), October 5-9, Munich, Germany. Switzerland: Springer, 2015: 234-241.

【14】Xiao Z T, Yuan Q, Zhang F, et al. The oriented partial differential equation based on the discontinuities measure for electronic speckle pattern interferometry[J]. Journal of Electronics & Information Technology, 2014, 36(11): 2600-2606.
肖志涛, 袁泉, 张芳, 等. 结合非连续性测度的方向偏微分方程在电子散斑干涉中的应用[J]. 电子与信息学报, 2014, 36(11): 2600-2606.

【15】Weickert J, Scharr H. A scheme for coherence-enhancing diffusion filtering with optimized rotation invariance[J]. Journal of Visual Communication and Image Representation, 2002, 13(1/2): 103-118.

【16】Zhang Y G, Yi B S, Wu C Y, et al. Low-dose CT image denoising method based on convolutional neural network[J]. Acta Optica Sinica, 2018, 38(4): 0410003.
章云港, 易本顺, 吴晨玥, 等. 基于卷积神经网络的低剂量CT图像去噪方法[J]. 光学学报, 2018, 38(4): 0410003.

【17】Gao W, Zhou Z H. Dropout Rademacher complexity of deep neural networks[J]. Science China Information Sciences, 2016, 59(7): 072104.

引用该论文

Zhang Fang,Wu Yue,Xiao Zhitao,Geng Lei,Wu Jun,Liu Yanbei,Wang Wen. Nanoparticle Segmentation Based on U-Net Convolutional Neural Network[J]. Laser & Optoelectronics Progress, 2019, 56(6): 061005

张芳,吴玥,肖志涛,耿磊,吴骏,刘彦北,王雯. 基于U-Net卷积神经网络的纳米颗粒分割[J]. 激光与光电子学进展, 2019, 56(6): 061005

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF