基于显著性检测和改进局部高斯分布拟合模型的眼底图像视盘边界自动提取
|
摘要
正确的视盘(OD)定位和分割是糖尿病视网膜病变自动筛选系统中的两个主要步骤.鉴于此,提出一种基于显著性目标检测和改进局部高斯分布拟合(LGDF)模型的视神经盘分割方法.该方法主要包含两个阶段:第一阶段,将显著性检测技术应用到增强的视网膜图像中实现视盘的自动定位;第二阶段,通过增加椭圆约束信息来改进局部高斯分布拟合(LGDF)模型分割视盘边界.使用公开数据库Diaretdbq对所提出方法的性能进行测试,并与其他先进的方法进行对比,结果验证了所提出方法的优越性和有效性.
Accurate optic disc(OD) localization and segmentation are two main steps when designing automated screening systems for diabetic retinopathy. In this paper, an OD segmentation approach based on the saliency object detection and modified local Gaussian distribution fitting model(LGDF) is proposed. This approah consists of two stages: In the first stage, the saliency detection technique is introduced to the enhanced retinal image with the aim of locating the OD; in the second stage, the OD boundary is extracted by the modified LGDF model with oval-shaped constrain. The performance of proposed approach is tested on the public DIARETDB1 database. Compared to the state-of-the-art approaches, the experimental results show the superiority and effectiveness of the proposed approach.
Accurate optic disc(OD) localization and segmentation are two main steps when designing automated screening systems for diabetic retinopathy. In this paper, an OD segmentation approach based on the saliency object detection and modified local Gaussian distribution fitting model(LGDF) is proposed. This approah consists of two stages: In the first stage, the saliency detection technique is introduced to the enhanced retinal image with the aim of locating the OD; in the second stage, the OD boundary is extracted by the modified LGDF model with oval-shaped constrain. The performance of proposed approach is tested on the public DIARETDB1 database. Compared to the state-of-the-art approaches, the experimental results show the superiority and effectiveness of the proposed approach.
引文
[1] Hatanaka Y, Noudo A, Muramatsu C, et al. Automatic measurement of cup to disc ratio based on line profile analysis in retinal images[C]. Proc of the 33rd Annual Int Conf of the IEEE Engineering in Medicine and Biology Society. Boston:IEEE, 2011:3387-3390.
[2] Esmaeili M, Rabbani H, Dehnavi A M. Automatic optic disk boundary extraction by the use of curvelet transform and deformable variational level set model[J]. Pattern Recognition, 2012, 45(7):2832-2842.
[3] Hsiao H K, Liu C C, Yu C Y. A novel optic disc detection scheme on retinal images[J]. Expert Systems with Applications An Int J, 2012, 39(12):10600-10606.
[4] Joshi G D, Sivaswamy J, Krishnadas S R. Optic disk and cup segmentation from monocular color retinal images for glaucoma assessment[J]. IEEE Trans on Medical Imaging, 2011, 30(6):1192–1205.
[5] Lalonde M, Beaulieu M, Gagnon L. Fast and robust optic disc detection using pyramidal decomposition and hausdorff based template matching[J]. IEEE Trans on Medical Imaging, 2001, 20(11):1193–1200.
[6] Chrástek R, Wolf M, Donath K. Optic disc segmentation in retinal images[J]. Bildverarbeitung für die Medizin,2002, 2002:263–266.
[7] Mary M C V S, Rajsingh E B, Jacob J K K, et al.An empirical study on optic disc segmentation using an active contour model[J]. Biomedical Signal Processing and Control, 2015, 18:19-29.
[8] Sánchez I, Hornero R, López M I. A novel automatic image processing algorithm for detection of hard exudates based on retinal image analysis[J]. Medical Engineering&Physics, 2008, 30(3):350-357.
[9] Bharath R, Nicholas L Z J, Xiang C. Scalable scene understanding using saliency-guided object localization[C]. IEEE Int Conf on Control and Automation. Hangzhou:IEEE, 2013:1503-1508.
[10] Kao E F, Lin P C, Chou M C, et al. Automated detection of fovea in fundus images based on vessel-free zone and adaptive Gaussian template[J]. Computer Methods&Programs in Biomedicine, 2014, 117(2):92-103.
[11] Wang L, He L, Mishra A, et al. Active contours driven by local Gaussian distribution fitting energy[J]. Signal Processing, 2009, 89(12):2435-2447.
[12] Kauppi T, Kalesnykiene V, Kamarainen J K.DIARETDB1 diabetic retinopathy database and evaluation protocol[J]. Proc of the Medical Image Understanding and Analysis, 2007:61-65.
[13] Mithun N C, Das S, Fattah S A. Automated detection of optic disc and blood vessel in retinal image using morphological, edge detection and feature extraction technique[C]. Int Conf on Computer and Information Technology. Bangladesh Khulna:IEEE, 2014:98-102.
[14] Welfer D, Scharcanski J, Marinho D R. A morphologic two-stage approach for automated optic disk detection in color eye fundus images[J]. Pattern Recognition Letters,2013, 34(5):476-485.
[15] Bharkad S. Automatic segmentation of optic disk in retinal images[J]. Biomedical Signal Processing&Control, 2017, 31:483-498.
[2] Esmaeili M, Rabbani H, Dehnavi A M. Automatic optic disk boundary extraction by the use of curvelet transform and deformable variational level set model[J]. Pattern Recognition, 2012, 45(7):2832-2842.
[3] Hsiao H K, Liu C C, Yu C Y. A novel optic disc detection scheme on retinal images[J]. Expert Systems with Applications An Int J, 2012, 39(12):10600-10606.
[4] Joshi G D, Sivaswamy J, Krishnadas S R. Optic disk and cup segmentation from monocular color retinal images for glaucoma assessment[J]. IEEE Trans on Medical Imaging, 2011, 30(6):1192–1205.
[5] Lalonde M, Beaulieu M, Gagnon L. Fast and robust optic disc detection using pyramidal decomposition and hausdorff based template matching[J]. IEEE Trans on Medical Imaging, 2001, 20(11):1193–1200.
[6] Chrástek R, Wolf M, Donath K. Optic disc segmentation in retinal images[J]. Bildverarbeitung für die Medizin,2002, 2002:263–266.
[7] Mary M C V S, Rajsingh E B, Jacob J K K, et al.An empirical study on optic disc segmentation using an active contour model[J]. Biomedical Signal Processing and Control, 2015, 18:19-29.
[8] Sánchez I, Hornero R, López M I. A novel automatic image processing algorithm for detection of hard exudates based on retinal image analysis[J]. Medical Engineering&Physics, 2008, 30(3):350-357.
[9] Bharath R, Nicholas L Z J, Xiang C. Scalable scene understanding using saliency-guided object localization[C]. IEEE Int Conf on Control and Automation. Hangzhou:IEEE, 2013:1503-1508.
[10] Kao E F, Lin P C, Chou M C, et al. Automated detection of fovea in fundus images based on vessel-free zone and adaptive Gaussian template[J]. Computer Methods&Programs in Biomedicine, 2014, 117(2):92-103.
[11] Wang L, He L, Mishra A, et al. Active contours driven by local Gaussian distribution fitting energy[J]. Signal Processing, 2009, 89(12):2435-2447.
[12] Kauppi T, Kalesnykiene V, Kamarainen J K.DIARETDB1 diabetic retinopathy database and evaluation protocol[J]. Proc of the Medical Image Understanding and Analysis, 2007:61-65.
[13] Mithun N C, Das S, Fattah S A. Automated detection of optic disc and blood vessel in retinal image using morphological, edge detection and feature extraction technique[C]. Int Conf on Computer and Information Technology. Bangladesh Khulna:IEEE, 2014:98-102.
[14] Welfer D, Scharcanski J, Marinho D R. A morphologic two-stage approach for automated optic disk detection in color eye fundus images[J]. Pattern Recognition Letters,2013, 34(5):476-485.
[15] Bharkad S. Automatic segmentation of optic disk in retinal images[J]. Biomedical Signal Processing&Control, 2017, 31:483-498.