JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering 2019 Vol.20 No.8 P.1075-1086

A saliency and Gaussian net model for retinal vessel segmentation

Author(s): Lan-yan Xue, Jia-wen Lin, Xin-rong Cao, Shao-hua Zheng, Lun Yu
Affiliation(s): College of Physics and Information Engineering, Fuzhou University, Fuzhou 350108, China; more
Corresponding email(s): xuelanyan@126.com
Key Words: Retinal vessel segmentation, Saliency model, Gaussian net (GNET), Feature learning

Share this article to： More <<< Previous Article \|Next Article >>>

Abstract: retinal vessel segmentation is a significant problem in the analysis of fundus images. A novel deep learning structure called the gaussian net (GNET) model combined with a saliency model is proposed for retinal vessel segmentation. A saliency image is used as the input of the GNET model replacing the original image. The GNET model adopts a bilaterally symmetrical structure. In the left structure, the first layer is upsampling and the other layers are max-pooling. In the right structure, the final layer is max-pooling and the other layers are upsampling. The proposed approach is evaluated using the DRIVE database. Experimental results indicate that the GNET model can obtain more precise features and subtle details than the UNET models. The proposed algorithm performs well in extracting vessel networks, and is more accurate than other deep learning methods. retinal vessel segmentation can help extract vessel change characteristics and provide a basis for screening the cerebrovascular diseases.

融合显著性模型和高斯网模型的视网膜血管分割方法

摘要：视网膜血管分割是眼底图像分析的一个重要问题。本文提出一种融合显著性模型和高斯网（GNET）模型的新型深度学习结构分割视网膜血管。显著性图像替代原始图像作为GNET模型的输入。GNET模型具有双边对称结构。左边结构中，在第一层进行上采样操作，在其他层进行最大池化操作；右边结构中，在第一层进行最大池化操作，在其他层进行上采样操作。利用DRIVE数据库对所提方法进行评估。实验结果表明，与UNET模型相比，GNET模型能获得更精确的特征和更精细的细节。本文所提算法能提取准确的血管网络，与其他深度学习方法相比具有更高精确度。视网膜血管分割有助于提取血管变化特征，为脑血管疾病筛查提供依据。

关键词：视网膜血管分割；显著性模型；高斯网模型（GNET）；特征学习

Reference

[1]Achanta R, Hemami S, Estrada F, et al., 2009. Frequency- tuned salient region detection. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.1597-1604.

[2]Ayala G, Leon T, Zapater V, 2005. Different averages of a fuzzy set with an application to vessel segmentation. IEEE Trans Fuzzy Syst, 13(3):384-393.

[3]Chaudhuri S, Chatterjee S, Katz N, et al., 1989. Detection of blood vessels in retinal images using two-dimensional matched filters. IEEE Trans Med Imag, 8(3):263-269.

[4]Franklin SW, Rajan SE, 2014. Retinal vessel segmentation employing ANN technique by Gabor and moment invariants-based features. Appl Soft Comput, 22:94-100.

[5]Fu HZ, Cao XC, Tu ZW, 2013. Cluster-based co-saliency detection. IEEE Trans Imag Process, 22(10):3766-3778.

[6]Fu HZ, Xu YW, Kee DW, et al., 2016. Retinal vessel segmentation via deep learning network and fully-connected conditional random fields. Proc IEEE 13^th Int Symp on Biomedical Imaging, p.698-701.

[7]He XT, Peng YX, Zhao JJ, 2017. Fine-grained discriminative localization via saliency-guided faster R-CNN. Proc 25^th ACM Int Conf on Multimedia, p.627-635.

[8]Hu YT, Wang NN, Tao DC, et al., 2016. SERF: a simple, effective, robust, and fast image super-resolver from cascaded linear regression. IEEE Trans Imag Process, 25(9):4091-4102.

[9]Ikram MK, de Jong FJ, Bos MJ, et al., 2006. Retinal vessel diameters and risk of stroke: the Rotterdam study. Neurology, 66(9):1339-1343.

[10]Imani E, Pourreza HR, 2016. A novel method for retinal exudate segmentation using signal separation algorithm. Comput Method Program Biomed, 133:195-205.

[11]Kumar RP, Albregtsen F, Reimers M, et al., 2015. Blood vessel segmentation and centerline tracking using local structure analysis. Proc 6^th European Conf of the Int Federation for Medical and Biological Engineering, p.122-125.

[12]Liu I, Sun Y, 1993. Recursive tracking of vascular networks in angiograms based on the detection-deletion scheme. IEEE Trans Med Imag, 12(2):334-341.

[13]Liu Z, Zou WB, Li LN, et al., 2014. Co-saliency detection based on hierarchical segmentation. IEEE Signal Process Lett, 21(1):88-92.

[14]Maji D, Santara A, Ghosh S, et al., 2015. Deep neural network and random forest hybrid architecture for learning to detect retinal vessels in fundus images. Proc 37^th Annual Int Conf of the IEEE Engineering in Medicine and Biology Society, p.3029-3032.

[15]Odstrcilik J, Radim K, Attila B, et al., 2013. Retinal vessel segmentation by improved matched filtering: evaluation on a new high-resolution fundus image database. IET Image Process, 7(4):373-383.

[16]Peng YX, He XT, Zhao JJ, 2018. Object-part attention model for fine-grained image classification. IEEE Trans Imag Process, 27(3):1487-1500.

[17]Ronneberger O, Fischer P, Brox T, 2015. U-Net: convolutional networks for biomedical image segmentation. Proc 18^th Int Conf on Medical Image Computing and Computer- Assisted Intervention, p.234-241.

[18]Schmidhuber J, 2015. Deep learning in neural networks: an overview. Neur Netw, 61:85-117.

[19]Shelhamer E, Long J, Darrell T, 2017. Fully convolutional networks for semantic segmentation. IEEE Trans Patt Anal Mach Intell, 39(4):640-651.

[20]Solouma NH, Youssef ABM, Badr YA, et al., 2002. A new real-time retinal tracking system for image-guided laser treatment. IEEE Trans Biomed Eng, 49(9):1059-1067.

[21]Vese LA, Chan TF, 2002. A multiphase level set framework for image segmentation using the Mumford and Shah model. Int J Comput Vis, 50(3):271-293.

[22]Wang NN, Gao XB, Sun LY, et al., 2017. Bayesian face sketch synthesis. IEEE Trans Imag Process, 26(3):1264-1274.

[23]Wang NN, Gao XB, Sun LY, et al., 2018. Anchored neighborhood index for face sketch synthesis. IEEE Trans Circ Syst Video Technol, 28(9):2154-2163.

[24]Wang XH, Zhao YQ, Liao M, et al., 2015. Automatic segmentation for retinal vessel based on multi-scale 2D Gabor wavelet. Acta Automat Sin, 41(5):970-980 (in Chinese).

[25]Xiao TJ, Xu YC, Yang KY, et al., 2015. The application of two-level attention models in deep convolutional neural network for fine-grained image classification. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.842-850.

[26]Zana F, Klein JC, 2001. Segmentation of vessel-like patterns using mathematical morphology and curvature evaluation. IEEE Trans Imag Process, 10(7):1010-1019.

[27]Zhao YQ, Wang XH, Wang XF, et al., 2014. Retinal vessels segmentation based on level set and region growing. Patt Recog, 47(7):2437-2446.

[28]Zhu CZ, Zou BJ, Xiang Y, et al., 2015. A survey of retinal vessel segmentation in fundus images. J Comput Aided Des Comput Graph, 27(11):2046-2057 (in Chinese).

[29]Zhu CZ, Zou BJ, Xiang Y, et al., 2016. An ensemble retinal vessel segmentation based on supervised learning in fundus images. Chin J Electron, 25(3):503-511.

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Similar articles

- Go to

融合显著性模型和高斯网模型的视网膜血管分割方法

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference