基于YOLO算法的眼底图像视盘定位方法

视盘的各个参数是衡量眼底健康状况和病灶的重要指标,视盘的检测和定位对于观察视盘的形态尤为重要。在以往的视盘定位研究中,主要根据视盘的形状、亮度、眼底血管的走向等特征使用图像处理的方法对眼底图像中视盘进行定位。由于人为因素影响较大,特征提取时间较长,且视盘定位效率低,因此提出一种基于YOLO算法的眼底图像视盘定位方法。利用YOLO算法将眼底图像划分为N×N的格子,每个格子负责检测视盘中心点是否落入该格子中,通过多尺度的方式和残差层融合低级特征对视盘进行定位,得到不同大小的边界框,最后通过非极大抑制的方式筛选出得分最高的边界框。通过在3个公开的眼底图像数据集(DRIVE、DRISHTI-GS1和MESSIDOR)上,对所提出的视盘定位方法进行测试,定位准确率均为100%,实验同时定位出视盘的中心点坐标,与标准中心点的平均欧氏距离分别为22.36 px、2.52 px、21.42 px,验证了该方法的准确性和通用性。     

视觉显著性的眼底图像视盘检测

目的 青光眼是导致失明的主要疾病之一,视盘区域的形状、大小等参数是青光眼临床诊断的重要指标。然而眼底图像通常亮度低、对比度弱,且眼底结构复杂,各组织以及病灶干扰严重。为解决上述问题,实现视盘的精确检测,提出一种视觉显著性的眼底图像视盘检测方法。方法 首先,依据视盘区域显著的特点,采用一种基于视觉显著性的方法对视盘区域进行定位;其次,采用全卷积神经网络（fully convolutional neural network,FCN）预训练模型提取深度特征,同时计算视盘区域的平均灰度,进而提取颜色特征;最后,将深度特征、视盘区域的颜色特征和背景先验信息融合到单层元胞自动机（single-layer cellular automata,SCA）中迭代演化,实现眼底图像视盘区域的精确检测。结果 在视网膜图像公开数据集DRISHTI-GS、MESSIDOR和DRIONS-DB上对本文算法进行实验验证,平均相似度系数分别为0.965 8、0.961 6和0.971 1;杰卡德系数分别为0.934 1、0.922 4和0.937 6;召回率系数分别为0.964 8、0.958 9和0.967 4;准确度系数分别为0.996 6、0.995 3和0.996 8,在3个数据集上均可精确地检测视盘区域。实验结果表明,本文算法精确度高,鲁棒性强,运算速度快。结论 本文算法能够有效克服眼底图像亮度低、对比度弱及血管、病灶等组织干扰的影响,在多个视网膜图像公开数据集上进行验证均取得了较好的检测结果,具有较强的泛化性,可以实现视盘区域的精确检测。     

基于DWT和形态学的眼底图像小血管检测

针对眼底图像中末端小血管检测难、细节容易丢失的问题．提出一种基于离散小波变换（DWT）和形态学滤波的检测算法。通过小波变换多尺度分析眼底图像小血管系数、背景系数的不同特征．选取分量信号的系数后重构图像。同时以自适应阈值Canny算法提取小血管边缘;然后将结合小血管宽度选择适当结构元素半径,对重构图像进行灰度膨胀,实现小血管检测。结果表明,形态学结合DWT的检测算法能够准确地检测小血管．与常见边缘检测算法相比检测成功率较高。     

结合PCNN和自适应DGA算法的视网膜血管分割

为辅助诊断眼底疾病,提出一种眼底图像血管自动分割方法。首先利用对比度受限制的自适应直方图均衡化（CLAHE）技术与二维高斯匹配滤波器增强血管与背景对比度;然后利用自适应分布式遗传算法（ADGA）对PCNN参数设置自动寻优,将寻优得到的参数用于PCNN血管分割;最后采用面积滤波和区域连通性方法对分割结果进行后处理,得到优化后的血管检测结果。通过在国际上公认的彩色眼底图像库STARE中的实验结果表明,相比于利用传统的DGA算法对PCNN参数寻优,所提方法将分割的平均准确度从0.929?3提高到0.945?4,具有更高的鲁棒性、有效性和可靠性。     

多相主动轮廓模型的眼底图像杯盘分割

目的 视盘及视杯的检测对于分析眼底图像和视网膜视神经疾病计算机辅助诊断来说十分重要,利用医学眼底图像中视盘和视杯呈现椭圆形状这一特征,提出了椭圆约束下的多相主动轮廓模型,实现视盘视杯的同时精确分割。方法 该算法根据视盘视杯在灰度图像中具有不同的区域亮度,建立多相主动轮廓模型,然后将椭圆形约束内嵌于该模型中。通过对该模型的能量泛函进行求解,得到椭圆参数的演化方程。分割时首先设定两条椭圆形初始曲线,根据演化方程,驱动曲线分别向视盘和视杯方向进行移动。当轮廓线到达视盘、视杯边缘时,曲线停止演化。结果 在不同医学眼底图像中对算法进行验证,对算法抗噪性、不同初始曲线选取等进行了实验,并与多种算法进行了对比。实验结果表明,本文模型能够同时分割出视盘及视杯,与其他模型的分割结果相比,本文算法的分割结果更加准确。结论 本文算法可以精确分割医学眼底图像中的视盘和视杯,该算法不需要预处理,具有较强的鲁棒性和抗噪性。     

基于HOG的酿酒葡萄叶检测

在酿酒葡萄生长状态与病虫害自动监测中,需要在图像中检测出葡萄叶片,通过提取葡萄叶片图像的方向梯度直方图（HOG）特征投入到支持向量机（SVM）分类器中以实现对葡萄叶片的识别;结合多尺度目标定位和均值漂移算法还可以自动确定图像中葡萄叶片的位置。实验结果表明,使用线性核函数训练后的分类器对葡萄叶片和四种常见杂草的识别率达95.5%。该方法对光照和环境变化有较好的鲁棒性,自然条件下成像的叶片图像的葡萄叶片检出率达到了80%以上。     

基于直觉模糊散度的噪声图像边缘检测

针对传统图像边缘检测算法抑制噪声能力差的问题,提出一种基于直觉模糊集（Intuitionistic Fuzzy Set,IFS）的边缘检测算法。该算法设定了一个表示平坦区域的模板图像,并在图像窗口内构造了一种同时考虑了图像梯度和图像窗口的方差信息的隶属度函数,然后通过计算图像窗口与模板图像之间的模糊直觉散度（Intuitionistic Fuzzy Divergence,IFD）对边缘进行定位和输出。实验结果表明,对于被高斯噪声或均匀噪声严重污染的图像,该算法能够得到较好的检测结果。     

基于置信度计算的快速眼底图像视盘定位

眼底图像视盘定位是视盘分割的重要前提.针对视盘定位结果易受图像对比度的影响的问题,提出一种基于置信度计算的快速视盘定位方法.首先采用基于形态学变换的方法增强眼底图像中视盘、血管区域与图像背景的对比度,并根据图像增强结果中像素点的亮度特征初始定位视盘区域;然后运用局部滑动窗口扫描的方法,根据窗口内像素点亮度特征和其周围血管分布的特性计算候选区域的置信度,定位视盘区域.在不同的眼底图像公共数据上进行实验的结果表明,对于1 341幅眼底图像,该方法能准确地定位其中1 325幅图像的视盘区域,视盘定位准确率为98.8%,平均每幅图像耗时0.25 s,优于现有的视盘定位方法,适用于眼底疾病的计算机辅助诊断.     

基于二元纯位相滤波器水印的图像篡改定位

二元纯位相滤波器（BPOF）数字图像水印算法将图像离散Fourier变换的BPOF作为水印嵌入到其相应幅值的某个位平面中,较好地实现了图像自认证。研究了利用BPOF水印实现图像篡改定位。其基本原理是将图像划分为互不重叠的分块,通过在各分块中独立嵌入和检测水印,实现图像篡改检测和定位。着重讨论了以不同大小对于图像分块和以不同幅值位平面嵌入水印时,所嵌入水印的不可感知性、检测性能、图像篡改定位能力以及抗JPEG压缩性能。得出了图像分块大小、嵌入水印的幅值位平面及水印检测阈值等参数的选取策略。实验结果证明了BPOF水印可以很好地应用于图像篡改定位。     

新的眼底图像配准方法的研究

针对传统的基于特征的眼底图像配准方法配准精度不高的问题,提出了一种新的眼底图像配准方法。通过具有仿射不变性的尺度不变特征变换（Scale Invariance Feature Transform,SIFT）方法提取待配准图像的特征点匹配对。采用适合眼底图像特点的曲面变换模型,实现图像的配准,变换模型参数通过M估计获得。实验结果表明,该算法提高了配准精度,对正常眼和非正常眼的眼底图像配准都是有效的。     

视网膜图像中的黄斑中心检测

目的 在眼底图像分析中,准确的黄斑中心定位对于糖尿病性视网膜病变的计算机辅助诊断系统具有重要的意义。然而,由于光照不均匀、计算量大及病变的干扰给黄斑中心定位带来了巨大的挑战。因此,为了实现更为准确且高效的黄斑中心检测,提出一种基于血管投影和数学形态学的黄斑中心检测方法。方法 首先,基于数学形态学,提出一种自动的血管检测方法。其次,利用视盘区域的血管分布实现视盘中心的自动定位。再次,根据视盘和黄斑的解剖学结构先验信息,提取感兴趣区域。最后,在感兴趣区域内,通过数学形态学和特征提取定位黄斑中心。结果 本文提出的方法在两个标准的糖尿病视网膜病变数据库DIARETDB0和DIARETDB1上分别取得了96.92%和96.63%的成功率,且总成功率达到96.35%。此外,平均的执行时间分别为8.236 s和8.912 s。结论 实验结果表明,本文方法能快速和准确地定位黄斑中心且其性能明显地优于现有的黄斑中心检测方法。     

Combining algorithms for automatic detection of optic disc and macula in fundus images

This paper proposes an efficient combination of algorithms for the automated localization of the optic disc and macula in retinal fundus images. There is in fact no reason to assume that a single algorithm would be optimal. An ensemble of algorithms based on different principles can be more accurate than any of its individual members if the individual algorithms are doing better than random guessing. We aim to obtain an improved optic disc and macula detector by combining the prediction of multiple algorithms, benefiting from their strength and compensating their weaknesses. The location with maximum number of detectors’ outputs is formally the hotspot and is used to find the optic disc or macula center. An assessment of the performance of integrated system and detectors working separately is also presented. Our proposed combination of detectors achieved overall highest performance in detecting optic disc and fovea closest to the manually center chosen by the retinal specialist.     

Automated detection of fovea in fundus images based on vessel-free zone and adaptive Gaussian template

This study developed a computerised method for fovea centre detection in fundus images. In the method, the centre of the optic disc was localised first by the template matching method, the disc–fovea axis (a line connecting the optic disc centre and the fovea) was then determined by searching the vessel-free region, and finally the fovea centre was detected by matching the fovea template around the centre of the axis. Adaptive Gaussian templates were used to localise the centres of the optic disc and fovea for the images with different resolutions. The proposed method was evaluated using three publicly available databases (DIARETDB0, DIARETDB1 and MESSIDOR), which consisted of a total of 1419 fundus images with different resolutions. The proposed method obtained the fovea detection accuracies of 93.1%, 92.1% and 97.8% for the DIARETDB0, DIARETDB1 and MESSIDOR databases, respectively. The overall accuracy of the proposed method was 97.0% in this study.     

A novel optic disc detection scheme on retinal images

Robust and effective optic disc detection is a necessary processing component in automatic retinal screening systems. In this paper, optic disc localization is achieved by a novel illumination correction operation, and contour segmentation is completed by a supervised gradient vector flow snake (SGVF snake) model. Conventional GVF snake is not sufficient to segment contour due to vessel occlusion and fuzzy disc boundaries. In view of this reason, the SGVF snake is extended in each time of deformation iteration, so that the contour points can be classified and updated according to their corresponding feature information. The classification relies on the feature vector extraction and the statistical information generated from training images. This approach is evaluated by means of two publicly available databases, Digital Retinal Images for Vessel Extraction (DRIVE) database and Structured Analysis of the Retina (STARE) database, of color retinal images. The experimental results show that the overall performance is with 95% correct optic disc localization from the two databases and 91% disc boundaries are correctly segmented by the SGVF snake algorithm.     

Retinal optic disc localization using convergence tracking of blood vessels

Optic disc localization is of great diagnostic value related to retinal diseases, such as glaucoma and diabetic retinopathy. However, the detection process is quite challenging because positions of optic discs vary from image to image, and moreover, pathological changes, like hard exudates or neovascularization, may alter optic disc appearance. In this paper, we propose a robust approach to accurately detect the optic disc region and locate the optic disc center in color retinal images. The proposed technique employs a kernelized least-squares classifier to decide the area that contains optic disc. Then connected-component labeling and lumination information are used together to find the convergence of blood vessels, which is thought to be optic disc center. The proposed method has been evaluated over two datasets: the Digital Retinal Images for Vessel Extraction (DRIVE), and the Non-fluorescein Images for Vessel Extraction (NIVE) datasets. Experimental results have shown that our method outperforms existing methods, achieving a competitive accuracy (97.52 %) and efficiency (1.1577s).     

Automatic segmentation of optic disc in retinal fundus images using semi-supervised deep learning

Diseases of the eye require manual segmentation and examination of the optic disc by ophthalmologists. Though, image segmentation using deep learning techniques is achieving remarkable results, it leverages on large-scale labeled datasets. But, in the field of medical imaging, it is challenging to acquire large labeled datasets. Hence, this article proposes a novel deep learning model to automatically segment the optic disc in retinal fundus images by using the concepts of semi-supervised learning and transfer learning. Initially, a convolutional autoencoder (CAE) is trained to automatically learn features from a large number of unlabeled fundus images available from the Kaggle’s diabetic retinopathy (DR) dataset. The autoencoder (AE) learns the features from the unlabeled images by reconstructing the input images and becomes a pre-trained network (model). After this, the pre-trained autoencoder network is converted into a segmentation network. Later, using transfer learning, the segmentation network is trained with retinal fundus images along with their corresponding optic disc ground truth images from the DRISHTI GS1 and RIM-ONE datasets. The trained segmentation network is then tested on retinal fundus images from the test set of DRISHTI GS1 and RIM-ONE datasets. The experimental results show that the proposed method performs on par with the state-of-the-art methods achieving a 0.967 and 0.902 dice score coefficient on the test set of the DRISHTI GS1 and RIM-ONE datasets respectively. The proposed method also shows that transfer learning and semi-supervised learning overcomes the barrier imposed by the large labeled dataset. The proposed segmentation model can be used in automatic retinal image processing systems for diagnosing diseases of the eye.

     

Ant Colony Optimization-based method for optic cup segmentation in retinal images

An accurate detection of the cup region in retinal images is necessary to obtain relevant measurements for glaucoma detection. In this work, we present an Ant Colony Optimization-based method for optic cup segmentation in retinal fundus images. The artificial agents will construct their solutions influenced by a heuristic that combines the intensity gradient of the optic disc area and the curvature of the vessels. On their own, the exploration capabilities of the agents are limited; however, by sharing the experience of the entire colony, they are capable of obtaining accurate cup segmentations, even in images with a weak or non-obvious pallor. This method has been tested with the RIM-ONE dataset, yielding an average overlapping error of 24.3% of the cup segmentation and an area under the curve (AUC) of 0.7957 using the cup to disc ratio for glaucoma assessment.     

An automated decision-support system for non-proliferative diabetic retinopathy disease based on MAs and HAs detection

Diabetic retinopathy (DR) has become a serious threat in our society, which causes 45% of the legal blindness in diabetes patients. Early detection as well as the periodic screening of DR helps in reducing the progress of this disease and in preventing the subsequent loss of visual capability. This paper provides an automated diagnosis system for DR integrated with a user-friendly interface. The grading of the severity level of DR is based on detecting and analyzing the early clinical signs associated with the disease, such as microaneurysms (MAs) and hemorrhages (HAs). The system extracts some retinal features, such as optic disc, fovea, and retinal tissue for easier segmentation of dark spot lesions in the fundus images. That is followed by the classification of the correctly segmented spots into MAs and HAs. Based on the number and location of MAs and HAs, the system quantifies the severity level of DR. A database of 98 color images is used in order to evaluate the performance of the developed system. From the experimental results, it is found that the proposed system achieves 84.31% and 87.53% values in terms of sensitivity for the detection of MAs and HAs respectively. In terms of specificity, the system achieves 93.63% and 95.08% values for the detection of MAs and HAs respectively. Also, the proposed system achieves 68.98% and 74.91% values in terms of kappa coefficient for the detection of MAs and HAs respectively. Moreover, the system yields sensitivity and specificity values of 89.47% and 95.65% for the classification of DR versus normal.     

Detection and Classification of Hemorrhages in Retinal Images

Damage of the blood vessels in retina due to diabetes is called diabetic retinopathy (DR). Hemorrhages is the first clinically visible symptoms of DR. This paper presents a new technique to extract and classify the hemorrhages in fundus images. The normal objects such as blood vessels, fovea and optic disc inside retinal images are masked to distinguish them from hemorrhages. For masking blood vessels, thresholding that separates blood vessels and background intensity followed by a new filter to extract the border of vessels based on orientations of vessels are used. For masking optic disc, the image is divided into sub-images then the brightest window with maximum variance in intensity is selected. Then the candidate dark regions are extracted based on adaptive thresholding and top-hat morphological techniques. Features are extracted from each candidate region based on ophthalmologist selection such as color and size and pattern recognition techniques such as texture and wavelet features. Three different types of Support Vector Machine (SVM), Linear SVM, Quadratic SVM and Cubic SVM classifier are applied to classify the candidate dark regions as either hemorrhages or healthy. The efficacy of the proposed method is demonstrated using the standard benchmark DIARETDB1 database and by comparing the results with methods in silico. The performance of the method is measured based on average sensitivity, specificity, F-score and accuracy. Experimental results show the Linear SVM classifier gives better results than Cubic SVM and Quadratic SVM with respect to sensitivity and accuracy and with respect to specificity Quadratic SVM gives better result as compared to other SVMs.     

视盘和视杯分割在计算机辅助青光眼诊断中的应用综述

青光眼是以视神经损伤、特征性视野损伤为特点的一类眼病,在早期很难诊断,尽早发现可更好地遏制青光眼病症的恶化,降低致盲率。视盘和视杯的比值是评价青光眼诊断中的重要指标之一,视盘和视杯的分割是青光眼诊断的关键步骤。但眼底彩照中的渗出物、不均匀照明区域等特征使其可能出现相似的亮度区域,导致视盘和视杯的分割非常困难。因此本文对现有眼底彩照中视盘和视杯的分割方法进行了总结,并将其分为5大类:水平集法、模态法、能量泛函法、划分法以及基于机器学习的混合法。系统地梳理了各类算法的代表性方法,以及基本思想、理论基础、关键技术、框架流程和优缺点等。同时,概括了适用于青光眼诊断的各种数据集,包括数据集的名称、来源以及详细内容,并总结了在各种数据集中不同视盘和视杯分割结果和诊断青光眼的量化指标及其相关结果。在现有的视盘和视杯分割方法中,许多图像处理和机器学习技术得到广泛应用。通过对该领域研究算法进行综述,清晰直观地总结了各类算法之间的特点及联系,有助于推动视盘和视杯分割在青光眼疾病临床诊断中的应用。可以在很大程度上提高临床医生的工作效率,为临床诊断青光眼提供了重要的理论研究意义和价值。