摘要
首先用灰狼优化算法优化阈值,用优化后的阈值完成区域生长,从而准确提取视网膜血管区域。实验结果表明,该算法可以获得平均96.93%的准确率。
The threshold is optimized with gray wolf optimization algorithm for regional growth to precisely extract the diabetic retinal blood vessel region.Experiments indicate that algorithm can get an average accuracy rate of 96.93%.
引文
[1]丁蓬莉.基于深度学习的糖尿病性视网膜图像分析算法研究[D].北京:北京交通大学,2017.
[2]罗院生.从视网膜图像到糖尿病视网膜病变诊断[D].成都:电子科技大学,2017.
[3]戴百生.基于彩色视网膜图像的眼底病相关目标检测方法研究[D].哈尔滨:哈尔滨工业大学,2016.
[4]吴奎.基于特征提取和监督学习的视网膜血管分割技术研究[D].太原:太原理工大学,2016.
[5]Nayebifar B,Moghaddam H A.A novel method for retinal vessel tracking using particle filters[J].Computers in Biology and Medicine,2013,43(5):541-548.
[6]Zhao Y Q,Wang X H,Wang X F,et al.Retinal vessels segmentation based on level set and region growing[J].Pattern Recognition,2014,47(7):2437-2446.
[7]Li Q,You J,Zhang D.Vessel segmentation and width estimation in retinal images using multiscale production of matched filter responses[J].Expert Systems With Applications,2012,39(9):7600-7610.
[8]Salazar Gonzalez,Kaba D,Li Y M,et al.Segmentation of the blood vessels and optic disk in retinal images[J].IEEE Journal of Biomedical and Health Informatics,2014,18(6):1874-1886.
[9]张新明,涂强,康强,等.强化狼群等级制度的灰狼优化算法[J].数据采集与处理,2017,5(32):880-887.
[10]陈跃伟,彭道刚,夏飞,等.基于区域生长法和BP神经网络的红外图像识别[J].激光与红外,2018,48(3):402-407.
[11]刘乐,李阳,侯阿临,等.基于CT图像的肺实质分割[J].长春工业大学学报,2015,36(1):72-76.