A critical review on diagnosis of diabetic retinopathy using machine learning and deep learning

Multimedia Tools and Applications - Diabetic Retinopathy (DR) is a health condition caused due to Diabetes Mellitus (DM). It causes vision problems and blindness due to disfigurement of human...     

Automatic diabetic retinopathy diagnosis using adjustable ophthalmoscope and multi-scale line operator

Diabetic Retinopathy (DR), the most common one of diabetic eye diseases that cause loss of vision and blindness, has become one of major health problems today. However, DR can be eased through timely treatment and periodical screening. In this paper, we proposes an automatic diabetic retinopathy diagnostic system to help patients know about their retinal conditions. We design a portable ophthalmoscope, which is composed of a retinal lens, a smartphone and a frame between them to help patients take fundus images anywhere and anytime. Then the images are transmitted to be analyzed, including localization of optic disk and macular, vessel segmentation, detection of lesions, and grading of DR. We use a multi-scale line operator to improve accuracy in segmenting small-scale vessels, a binary mask and image restoration to reduce the effect of the existence of the vessels on optic disk localization. After the analysis, the fundus image are then graded as normal, mild Non-Proliferative Diabetic Retinopathy (NPDR), moderate NPDR or severe NPDR. The grading process uses region segmentation to improve the efficiency. The final grading results are tested based on the fundus images provided by the hospitals. We evaluate our system through comparing our grading result with those graded by experts, which comes out with an overall accuracy of up to 85%.     

Adaptive machine learning classification for diabetic retinopathy

Multimedia Tools and Applications - Diabetic retinopathy is the main cause of the blindness worldwide. However, the diabetic retinopathy is hard to be detected in the initial stages, and the...     

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.     

结合域适应学习的糖尿病视网膜病变分级诊断

目的 传统的糖尿病视网膜病变(糖网)(diabetic retinopathy, DR)依赖于早期病理特征的精确检测,但由于数据集缺乏病灶标记区域导致无法有效地建立监督性分类模型,引入其他辅助数据集又会出现跨域数据异质性问题;另外,现有的糖网诊断方法大多无法直观地从语义上解释医学模型预测的结果。基于此,本文提出一种端到端式结合域适应学习的糖网自动多分类方法,该方法协同注意力机制和弱监督学习加强优化。方法 首先,利用已标记病灶区域的辅助数据训练病灶检测模型,再将目标域数据集的糖网诊断转化为弱监督学习问题,依靠多分类预测结果指导深度跨域生成对抗网络模型,提升跨域的样本图像质量,用于微调病灶检测模型,进而过滤目标域中一些无关的病灶样本,提升多分类分级诊断性能。最后,在整体模型中融合注意力机制,从医学病理诊断角度提供可解释性支持其分类决策。结果 在公开数据集Messidor上进行糖网多分类评估实验,本文方法获得了71.2%的平均准确率和80.8%的AUC(area under curve)值,相比于其他多种方法具有很大优势,可以辅助医生进行临床眼底筛查。结论 结合域适应学习的糖网分类方法在没有...     

Electric contacts inspection using machine vision

Machine vision is an excellent tool for inspecting a variety of industrial items such as textiles, printed circuit boards, electric components, labels, integrated circuits (IC), machine tools and fruits. In this paper, we propose machine vision-based inspection system for electric contact (EC), which are popularly used in switches, breakers and relays, and are important components in the electrical industry. The proposed system consists of three sub-systems, which inspect the top, side, and bottom surfaces of electric contact for different types of defects respectively. The system acquires the digital image of three views and classifies the surface defects including cracks, breaks, and scratches. For each view, this study develops different image pre-processing and feature extraction methods to enhance and detect the surface defects. The proposed system was implemented and verified using 229 samples collected from the EC production lines. Experimental results show the proposed system is effective and efficient in identifying EC defects.     

多核多示例学习的糖尿病视网膜病变诊断

目的 在传统糖尿病视网膜病变（糖网）诊断系统中,微动脉瘤和出血斑病灶检测的精确性决定了最终诊断性能。目前的检测诊断方法为了保证高敏感性而产生了大量假阳性样本,由于数据集没有标注病灶区域导致无法有效地建立监督性分类模型以去除假阳性。为了解决监督性学习在糖网诊断中的问题,提出一种基于多核多示例学习的糖网病变诊断方法。方法 首先,检测疑似的微动脉瘤和出血斑病灶区域,并将其视为多示例学习模型中的示例,而将整幅图像视为示例包,从而将糖网诊断转化为多示例学习问题;其次,提取病灶区域的特征对示例进行描述,并通过极限学习机（ELM）分类算法过滤不相关示例以提升后续多示例学习的分类性能;最后,构建多核图的多示例学习模型对健康图像和糖网病变图像进行分类,以实现糖网病变的诊断。结果 通过对国际公共数据集MESSIDOR进行糖网病变诊断评估实验,获得的准确率为90.1%,敏感性为92.4%,特异性为91.4%,ROC（receiver operating characteristic）曲线下面积为0.932,相比其他算法具有较大性能优势。结论 基于多核多示例学习方法在无需提供病灶标注的情况下,能够高效自动地对糖网病变进行诊断,从而既能避免医学图像中标注病灶的费时费力,又可以免除分类算法中假阳性去除的问题,获得较好的效果。     

Automatic target scoring system using machine vision

This paper reports on the development of a machine vision system for assessing targeting accuracy of ballistic, projectile-firing, weapon systems. The machine vision system developed here can monitor the target continuously, report precise quantitative targeting information, and simultaneously provide a color-coded display of impacts. The only other automated target scoring system on the market today is based on the analysis of acoustic signatures of projectiles using a heavily instrumented target. The proposed vision-based technique, in contrast, is inherently passive, requiring little or no modification of the existing set up.     

基于机器视觉的西瓜子外观品质检测与分类

采用机器视觉获取了西瓜子的面积、周长、最小外接矩和圆形度等外形特征,而后使用遗传算法优化的最小二乘支持向量机算法对西瓜子外观品质进行分类识别,最终实现了破损瓜子、普通瓜子和优质瓜子的区分。实验结果表明,基于最小二乘支持向量机分类的西瓜子外形检测方法能够很好地实现西瓜子外观品质的识别检测。     

Monitoring mineral wool production using real-time machine vision

This paper describes the development and use of a real-time machine vision system in the industrial production of mineral wool. Because of specific operating requirements and the very hostile environment the system employs unique hardware and software features.The main requirement is for real-time detection and control of the trajectory of molten rock as it falls onto a system of spinning wheels. The jet exhibits significant variation of its position with time, as a consequence of complex thermodynamic and hydrodynamic processes, resulting in a wide variation of quality and quantity of final product. Algorithms were developed to detect variations in the jet trajectory with time and this information is used to generate suitable control data.     

Design of fiducials for accurate registration using machine vision

The effect of the shape and size of a standard feature (fiducial) on the positional accuracy of the centroid in machine-vision applications is studied. It is shown that a circular fiducial provides superior subpixel registration accuracy among the fiducial shapes typically used today     

Quantitative automated inspection of standard parts using machine vision

On line automated visual inspection for quality and process control is becoming a very important requirement in an automated manufacturing environment. This paper examines the possibility of real-time inspection of standard part using machine vision.     

应用视觉技术的液体质量智能检测系统研究

提出了一种基于机器视觉和支持向量机集成的液体质量智能检测系统。该系统采用了专门装置采集瓶内旋转液体的序列图像。对采集到的序列图像进行运动分析,确定图像序列中的运动区域。然后提取运动区域的特征,提出多核函数支持向量机集成方法来确定运动区域是否为液体中的杂质。通过实验证明,该液体质量检测系统的检测准确率可达97%。     

结合机器视觉的车道偏离识别算法研究

首先确定车道偏离检测的目标区域,然后利用数字图像处理技术及多种算法进行处理,将推算出的左右车道线夹角作为判别车辆是否偏离车道行驶的预警依据。实验结果表明,该算法具有较好的稳定性和一定的应用价值。     

Subμm registration of fiducial marks using machine vision

Subμm registration between circularly symmetric fiducial marks can be achieved simply by finding their centroids by fitting circles to distributed edge positions determined by the second-derivative zero-crossing method. However, because of diffraction fringes, substantial errors originating in the tilt of the microscope optical axis must be minimized. This method is used for laser-fibre alignment tolerance studies     

基于机器视觉的开心果闭壳与开壳识别

为从开心果中选出开壳果和去除闭壳果以提高开心果的品质。采用以单排等间距方式在传送带上输送开心果,输送过程中用摄像头拍摄每个开心果的图像,传输至电脑,经图像处理算法识别出开心果壳是否裂开,最后在传送带末端利用气嘴吹走闭壳果实以实现开心果自动分级。开心果图像缩小至原来的0.2倍,然后寻找出二值化图像中开心果的质心,再在开心果的质心、上半部分二分之一和下半部分二分之一处各取横向像素线,分别求取3条像素线上所有像素点灰度值变化之和,最后每个开心果综合3条像素线特征判断开心果是否开壳。实验证明开心果开壳果正确识别率为93%,闭壳果正确识别率达100%。     

Automated thermal fuse inspection using machine vision and artificial neural networks

Machine vision is an excellent tool for inspecting a variety of items such as textiles, fruit, printed circuit boards, electrical components, labels, integrated circuits, machine tools, etc. This paper presents an intelligent system that incorporates machine vision with artificial intelligent networks to automatically inspect thermal fuses. An effective inspection flow is proposed to detect four commonly seen defects, including black-dot, small-head, bur, and flake during the production of thermal fuses. Backpropagation neural networks and learning vector quantization performance is compared in detecting the bur defect because of its illegibility. Different numbers of defective samples were screened out from a production line in a case study company and used to demonstrate the efficacy of the proposed system. Currently, the proposed inspection system is operating at the case study company, replacing four to six human inspectors. The system not only ensures the quality of the thermal fuses produced, but also reduced the cost of manual visual inspection.     

Identification of different stages of diabetic retinopathy using retinal optical images

Diabetes is a disease which occurs when the pancreas does not secrete enough insulin or the body is unable to process it properly. This disease affects slowly the circulatory system including that of the retina. As diabetes progresses, the vision of a patient may start to deteriorate and lead to diabetic retinopathy. In this study on different stages of diabetic retinopathy, 124 retinal photographs were analyzed. As a result, four groups were identified, viz., normal retina, moderate non-proliferative diabetic retinopathy, severe non-proliferative diabetic retinopathy and proliferative diabetic retinopathy. Classification of the four eye diseases was achieved using a three-layer feedforward neural network. The features are extracted from the raw images using the image processing techniques and fed to the classifier for classification. We demonstrate a sensitivity of more than 90% for the classifier with the specificity of 100%.     

Robotic date fruit harvesting using machine vision and a 5-DOF manipulator

Date fruits can be found at the highest point of the palm tree, so harvesting can be a dangerous task for humans. In this study, a robotic arm is presented to harvest date fruits. A manipulator, traveling on a U-shaped rail with four prismatic joints and one revolute joint, was designed and constructed to approach the fruits efficiently. The end-effector of the manipulator was an electric chainsaw capable of cutting the date bunch. Two cameras were located on the manipulator in perpendicular directions to capture images, which were then transmitted to a decision-making unit to control the speed of the manipulator links. The decision-making unit included three fuzzy inference systems that used the location data of the fruit bunch stems and the area of tree leaves in the images to determine the speed of electromotors of the trolley, horizontal link, and vertical link. An artificial palm tree with dimensions similar to a real palm tree was constructed to evaluate the performance of the robot. The performance of the robot was evaluated in various lighting conditions. The results showed that the introduced image processing algorithm could adequately detect the stems of the date bunches with accuracy, precision, and recall of 0.88, 0.92, and 0.96, respectively, in suitable lighting, that is, when there was sufficient natural lighting. Moreover, the experimental results revealed that the accuracy of the control system in cutting the stems was 87%, with a production loss of 5%. The proposed robot makes it possible to improve the capacity of using mechanized harvesting in date palm fields.     

A hybrid system for embedded machine vision using FPGAs and neural networks

This paper presents a hybrid model for embedded machine vision combining programmable hardware for the image processing tasks and a digital hardware implementation of an artificial neural network for the pattern recognition and classification tasks. A number of possible architectural implementations are compared. A prototype development system of the hybrid model has been created, and hardware details and software tools are discussed. The applicability of the hybrid design is demonstrated with the development of a vision application: real-time detection and recognition of road signs.