基于编解码结构的多特征融合眼底图像分割

摘要/Abstract

摘要： 为解决现有眼底图像分割方法对于细微血管存在低分割精度和低准确率的问题，提出一种基于编解码结构的U-Net改进网络模型。首先对数据进行预处理与扩充，提取绿色通道图像，并将其通过对比度限制直方图均衡化和伽马变换以增强对比度；其次训练集被输入到用于分割的神经网络中，在编码过程加入残差模块，用短跳跃连接将高、低特征信息融合，并利用空洞卷积增加感受野，解码模块加入注意力机制增加对细微血管分割精度；最后利用训练完成的分割模型进行预测得出视网膜血管分割结果。在DRIVE和CHASE-DB1眼底图像数据集上进行对比实验，模型算法的平均准确率、特异性和灵敏度分别达到96.77%和97.22%、98.74%和98.40%、80.93%和81.12%。实验结果表明该算法能够改善微细血管分割准确率及效率不高的问题，对视网膜血管可以进行更准确的分割。

关键词: 图像处理, 眼底图像, 血管分割, U型网络, 网络优化

Abstract: In order to solve the problem that the existing fundus images segmentation methods have low segmentation precision and low accuracy for micro vessels, an improved U-Net network model based on codec structure is proposed. Firstly, the data is preprocessed and expanded, the green channel image is extracted, and the contrast is enhanced by contrasting limited histogram equalization and Gamma transform; Secondly, the training set is input into the neural network for segmentation, the residual module is added in the coding process, the high and low feature information are fused by short jump connection, the receptive field is increased by hole convolution, and the attention mechanism is added in the decoding module to increase the segmentation accuracy of fine blood vessels; Finally, the trained segmentation model is used to predict the retinal vascular segmentation results. Comparative experiments on DRIVE and CHASE-DB1 fundus image data sets show that the average accuracy, specificity and sensitivity of the model algorithm are 96.77% and 97.22%, 98.74% and 98.40%, 80.93% and 81.12% respectively. The results of experiments show that the algorithm can improve the accuracy and efficiency of microvascular segmentation, and can segment retinal vessels more accurately.

Key words: image processing, fundus image, vascular segmentation, U-shaped network, network optimization

丁婉莹, 陈伟, 李昭慧. 基于编解码结构的多特征融合眼底图像分割[J]. 计算机与现代化, 2022, 0(07): 1-7.

DING Wan-ying, CHEN Wei, LI Zhao-hui. Multi-feature Fusion Fundus Image Segmentation Based on Codes Structure[J]. Computer and Modernization, 2022, 0(07): 1-7.

参考文献［30］

［1］	赵堪兴,杨培增. 眼科学［M］. 8版. 北京:人民卫生出版社, 2017.
［2］	高宏杰,邱天爽，丑远婷,等. 基于改进U型网络的眼底图像血管分割［J］. 中国生物医学工程学报, 2019,38(1):1-8.
［3］	ZHU C Z, ZOU B J, ZHAO R C, et al. Retinal vessel segmentation in colour fundus images using extreme learning machine［J］. Computerized Medical Imaging and Graphics, 2017,55:68-77.
［4］	AL-RAWI M, QUTAISHAT M, ARRAR M. An improved matched filter for blood vessel detection of digital retinal images［J］. Computers in Biology and Medicine, 2007,37(2):262-267.
［5］	YIN Y, ADEL M, BOURENNANE S. Retinal vessel segmentation using a probabilistic tracking method［J］. Pattern Recognition, 2012,45(4):1235-1244.
［6］	MIRIM S, MAHLOOJIFAR A. Retinal image analysis using curvelet transform and multistructure elements morphology by reconstruction［J］. IEEE Transactions on Biomedical Engineering, 2011,58(5):1183-1192.
［7］	WANG S L, YIN Y L, CAO G B， et al. Hierarchical retinal blood vessel segmentation based on feature and ensemble learning［J］. Neurocomputing, 2015,149:708-717.
［8］	唐明轩,李孝杰,周激流. 基于Dense Connected深度卷积神经网络的自动视网膜血管分割方法［J］. 成都信息工程大学学报, 2018,33(5):525-530.
［9］	ORANDO J I, PROKOFYEVA E, BLASCHKO M B. A discriminatively trained fully connected conditional random field model for blood vessel segmentation in fundus images［J］. IEEE Transactions on Biomedical Engineering, 2017,64(1):16-27.
［10］	GUO Y H, BUDAK U, VESPA L J,et al. A retinal vessel detection approach using convolution neural network with reinforcement sample learning strategy［J］. Measurement, 2018,125:586-591.
［11］	XIAO X, LIAN S, LUO Z M, et al. Weighted Res-UNet for high-quality retina vessel segmentation［C］// Proceedings of the 2018 9th International Conference on Information Technology in Medicine and Education. 2018:327-331.
［12］	GLOROT X, BORDES A, BENGIO Y. Deep sparse rectifier neural networks［C］// Proceedings of the 14th International Conference on Artificial Intelligence and Statistics. 2011:315-323.
［13］	IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift［C］// Proceedings of the 32nd International Conference on Machine Learning. 2015:448-456.
［14］	ROYCHOWDHURY S, KOOZEKANANI D D, PARHI K K. Blood vessel segmentation of fundus images by major vesse extraction and subimage classification［J］. IEEE Journal of Biomedical and Health Informatics, 2015,19(3):1118-1128.
［15］	郑婷月，唐晨，雷振坤.基于全卷积神经网络的多尺度视网膜血管分割［J］. 光学学报， 2019，39（2）：119-126.
［16］	LONG J, SHELHAMER E, ARRELL T D. Fully convolutional networks for semantic segmentation［C］// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2015:3431-3440.
［17］	CHEN L C, PAPANDREOU G, KOKKINOS I, et al. DeepLab: Semantic image segmentation with deep convolutionalnets, atrous convolution, and fully connected CRFs［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(4):834-848.
［18］	陈萌梦，熊兴良，张琰,等. 1种视网膜眼底图像增强的新方法［J］. 重庆医科大学学报, 2014,39(8):1087-1090.
［19］	OKTAY C, SCHLEMPER J, LE FOLGOC L, et al. Attention U-Net: Learning where to look for the pancreas［J］.arXiv preprint arXiv:1804.03999, 2018.
［20］	梁礼明，盛校棋，蓝智敏，等.自适应尺度信息的U型视网膜血管分割算法［J］.光学学报, 2019,39(8):118-132.
［21］	NGO L, HAN J H. Multi-level deep neural network for effcient segmentation of blood vessels infundus images［J］. Electronics Letters, 2017,53(16):1096-1098.
［22］	NIEMEIJER M,STAAL J,VAN GINNEKEN B,et al. Comparative study of retinal vessel segmentation methods on a new publicly available database［C］//Proceedings of SPIE:The International Society for Optical Engineering. 2004.DOI:10.1117/12.535349.
［23］	朱承璋,邹北骥,向遥,等. 彩色眼底图像视网膜血管分割方法研究进展［J］. 计算机辅助设计与图形学学报, 2015,27(11):2046-2057.
［24］	储清翠,王华彬,陶亮. 图像的局部自适应Gamma校正［J］. 计算机工程与应用, 2015,51(7):189-208.
［25］	朱辉,秦品乐. 基于多尺度特征结构的U-Net肺结节检测算法［J］. 计算机工程, 2019,45(4):254-261.
［26］	LI Q L, FENG B W, XIE L P,et al. A cross modality learning approach for vessel segmentation in retinal images［J］. IEEE Transactions on Medical Imaging, 2016, 35(1):109-118.
［27］	DASGUPTA A,S INGH S. A fully convolutional neural network based structured prediction approach towards the retinal vessel segmentation［C］// 2017 IEEE 14th International Symposium on Biomedical Imaging. 2017:248-251.
［28］	HU K, ZHANG Z Z, NIU X R, et al. Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function［J］. Neurocomputing, 2018,309:179-191.
［29］	YAN Z Q, YANG X, CHENG K T. Joint segment level and pixel-wise losses for deep learning based retinal vessel segmentation［J］. IEEE Transactions on Biomedical Engineering, 2018,65(9):1912-1923.
［30］	ALOM Z, YAKOPCIC C, HASAN M, et al. Recurrent residual U-Net for medical image segmentation［J］. Journal of Medical Imaging, 2019,6(1).DOI:10.1117/1.JMI.6.1.014006.

[1]	金安安, 丁双双, 熊卿智, 许婷婷, 习淑萍, 马继忠. 基于SWOMP算法的EPMA影像优化重构[J]. 计算机与现代化, 2020, 0(12): 104-111.
[2]	冯洋. 基于小波包变换的红外弱小目标检测[J]. 计算机与现代化, 2020, 0(12): 112-115.
[3]	张辉, 余厚云, 李克斌. 基于机械臂的涡轮壳零件表面质量视觉检测[J]. 计算机与现代化, 2020, 0(07): 121-126.
[4]	付磊,任德均,胡云起,郜明,邱吕. 基于ResNet网络的医用塑瓶制造缺陷检测方法[J]. 计算机与现代化, 2020, 0(04): 104-.
[5]	卢爱平，李盼池. 基于混沌序列的量子图像加密方案[J]. 计算机与现代化, 2020, 0(03): 86-.
[6]	买尔孜燕古丽·阿不都拉，冯向萍. 基于机器视觉的非接触式马体尺测量方法[J]. 计算机与现代化, 2019, 0(10): 108-.
[7]	王伟. 非刚性三维点云配准方法[J]. 计算机与现代化, 2018, 0(12): 116-.
[8]	敖宇,杨健晟. 欧拉视频增强技术在火焰弱信号检测中的应用[J]. 计算机与现代化, 2018, 0(07): 73-.
[9]	范新南，薛瑞阳，史朋飞，李敏，倪建军. 基于非负矩阵分解与相似性分析的运动目标检测[J]. 计算机与现代化, 2018, 0(04): 37-.
[10]	刘青松，谢晓方，曹建，张龙杰. 基于OpenCV的导引头图像处理算法性能评估软件[J]. 计算机与现代化, 2017, 0(3): 80-.
[11]	林钦壮1，钟映春1，罗唯师2. 面神经序列图像感兴趣区域识别[J]. 计算机与现代化, 2017, 0(11): 51-54.
[12]	傅天驹，郑嫦娥，田野，丘启敏，林斯俊. 复杂背景下基于深度卷积神经网络的森林火灾识别[J]. 计算机与现代化, 2016, 0(3): 52-57.
[13]	赵毅力1,2，徐丹1，张雁2. 基于ImageJ的数字图像处理课程实验教学案例[J]. 计算机与现代化, 2016, 0(3): 64-67+73.
[14]	俞向前. 基于Contourlet变换的多传感器图像融合[J]. 计算机与现代化, 2016, 0(12): 83-86.
[15]	王凡1，邢彦文2. 基于地面特征块匹配的机器视觉#br# 用于平整地面AGV行驶轨迹计算方法[J]. 计算机与现代化, 2016, 0(10): 44-48.