基于过渡区提取的视网膜血管分割方法

 针对现有视网膜血管分割方法对于小血管和低对比度血管分割效果差的问题,提出了一种基于过渡区提取的视网膜血管分割方法.该方法首先采用二维高斯匹配滤波预处理以增强血管,然后采用基于最佳熵的方法提取主血管、采用基于分布式遗传算法和Otsu相结合的方法提取过渡区,最后利用区域连通性分析所提取的主血管和过渡区,分割出最终的血管.通过在Hoover眼底图像库中的实验,结果表明该方法在小血管的提取、连通性和有效性方面均优于Hoover算法,另外由于迁移策略的分布式遗传算法的引入,使得算法效率也明显提高.     

Detection of anatomic structures in human retinal imagery

The widespread availability of electronic imaging devices throughout the medical community is leading to a growing body of research on image processing and analysis to diagnose retinal disease such as diabetic retinopathy (DR). Productive computer-based screening of large, at-risk populations at low cost requires robust, automated image analysis. In this paper we present results for the automatic detection of the optic nerve and localization of the macula using digital red-free fundus photography. Our method relies on the accurate segmentation of the vasculature of the retina followed by the determination of spatial features describing the density, average thickness, and average orientation of the vasculature in relation to the position of the optic nerve. Localization of the macula follows using knowledge of the optic nerve location to detect the horizontal raphe of the retina using a geometric model of the vasculature. We report 90.4% detection performance for the optic nerve and 92.5% localization performance for the macula for red-free fundus images representing a population of 345 images corresponding to 269 patients with 18 different pathologies associated with DR and other common retinal diseases such as age-related macular degeneration.     

Retinal image analysis using curvelet transform and multistructure elements morphology by reconstruction

Retinal images can be used in several applications, such as ocular fundus operations as well as human recognition. Also, they play important roles in detection of some diseases in early stages, such as diabetes, which can be performed by comparison of the states of retinal blood vessels. Intrinsic characteristics of retinal images make the blood vessel detection process difficult. Here, we proposed a new algorithm to detect the retinal blood vessels effectively. Due to the high ability of the curvelet transform in representing the edges, modification of curvelet transform coefficients to enhance the retinal image edges better prepares the image for the segmentation part. The directionality feature of the multistructure elements method makes it an effective tool in edge detection. Hence, morphology operators using multistructure elements are applied to the enhanced image in order to find the retinal image ridges. Afterward, morphological operators by reconstruction eliminate the ridges not belonging to the vessel tree while trying to preserve the thin vessels unchanged. In order to increase the efficiency of the morphological operators by reconstruction, they were applied using multistructure elements. A simple thresholding method along with connected components analysis (CCA) indicates the remained ridges belonging to vessels. In order to utilize CCA more efficiently, we locally applied the CCA and length filtering instead of considering the whole image. Experimental results on a known database, DRIVE, and achieving to more than 94% accuracy in about 50 s for blood vessel detection, proved that the blood vessels can be effectively detected by applying our method on the retinal images.     

A high performance hardware architecture for portable,low-power retinal vessel segmentation

The retina of the human eye and more particularly the retinal blood vasculature can be used in several medical and biometric applications. The use of retinal images in such applications however, is computationally intensive, due to the high complexity of the algorithms used to extract the vessels from the retina. In addition, the emergence of portable biometric authentication applications, as well as onsite biomedical diagnostics raises the need for real-time, power-efficient implementations of such algorithms that can also satisfy the performance and accuracy requirements of portable systems that use retinal images. In an attempt to meet those requirements, this work presents a VLSI implementation of a retina vessel segmentation system while exploring various parameters that affect the power consumption, the accuracy and performance of the system. The proposed design implements an unsupervised vessel segmentation algorithm which utilizes matched filtering with signed integers to enhance the difference between the blood vessels and the rest of the retina. The design accelerates the process of obtaining a binary map of the vessels tree by using parallel processing and efficient resource sharing, achieving real-time performance. The design has been verified on a commercial FPGA platform and exhibits significant performance improvements (up to 90×) when compared to other existing hardware and software implementations, with an overall accuracy of 92.4%. Furthermore, the low power consumption of the proposed VLSI implementation enables the proposed architecture to be used in portable systems, as it achieves an efficient balance between performance, power consumption and accuracy.     

Automated microaneurysm detection using local contrast normalization and local vessel detection

Screening programs using retinal photography for the detection of diabetic eye disease are being introduced in the UK and elsewhere. Automatic grading of the images is being considered by health boards so that the human grading task is reduced. Microaneurysms (MAs) are the earliest sign of this disease and so are very important for classifying whether images show signs of retinopathy. This paper describes automatic methods for MA detection and shows how image contrast normalization can improve the ability to distinguish between MAs and other dots that occur on the retina. Various methods for contrast normalization are compared. Best results were obtained with a method that uses the watershed transform to derive a region that contains no vessels or other lesions. Dots within vessels are handled successfully using a local vessel detection technique. Results are presented for detection of individual MAs and for detection of images containing MAs. Images containing MAs are detected with sensitivity 85.4% and specificity 83.1%.     

Quantitative Television Fluoroangiography?The Optical Measurement of Dye Concentrations and Estimation of Retinal Blood Flow

The development of a system for the measurement of dye concentrations from single retinal vessels during retinal fluorescein angiography is presented and discussed. The system uses a fundus camera modified for TV viewing. Video gating techniques define the areas of the retina to be studied, and video peak detection yields dye concentrations from retinal vessels. The time course of dye concentration is presented and blood flow into the retina is estimated by a time of transit technique.     

High speed,low power VLSI CMOS vision sensor for geometric centre object tracking

In this article an object tracking CMOS sensor is presented. The architecture incorporates photo detection devices and pixel level processing elements for capturing and processing the image data and extracting the object's coordinates. The edges of the image scene are extracted by in-pixel edge detectors and the region (object) of interest, selected by the user, is segmented using a switch network. Coordinates of the desired region are obtained by extracting the geometric centre of the region. Tracking of the selected object is then performed by automatic reselection of the region using the updated coordinates. The proposed design presents less sensitivity to threshold adjustments than binarisation techniques. The sensor has been designed as a 64 × 64 pixel VLSI CMOS chip in the 0.35 μm standard CMOS technology. The proposed structure is analysed with regard to its operation in the presence of mismatches and noise. Features of the sensor are reported and compared with some previous object tracking designs. Because the power dissipation is small, the chip is ideal for low-power applications.     

Streaming an image through the eye: The retina seen as a dithered scalable image coder

We propose the design of an original scalable image coder/decoder that is inspired from the mammalians retina. Our coder accounts for the time-dependent and also non-deterministic behavior of the actual retina. The present work brings two main contributions: As a first step, (i) we design a deterministic image coder mimicking most of the retinal processing stages and then (ii) we introduce a retinal noise in the coding process, that we model here as a dither signal, to gain interesting perceptual features. Regarding our first contribution, our main source of inspiration will be the biologically plausible model of the retina called Virtual Retina. The coder that we propose has two stages. The first stage is an image transform which is performed by the outer layers of the retina. Here we model it by filtering the image with a bank of difference of Gaussians with time-delays. The second stage is a time-dependent analog-to-digital conversion which is performed by the inner layers of the retina. The main novelty of this coder is to show that the time-dependent behavior of the retina cells could ensure, in an implicit way, scalability and bit allocation. Regarding our second contribution, we reconsider the inner layers of the retina. We emit a possible interpretation for the non-determinism observed by neurophysiologists in their output. For this sake, we model the retinal noise that occurs in these layers by a dither signal. The dithering process that we propose adds several interesting features to our image coder. The dither noise whitens the reconstruction error and decorrelates it from the input stimuli. Furthermore, integrating the dither noise in our coder allows a faster recognition of the fine details of the image during the decoding process. Our present paper goal is twofold. First, we aim at mimicking as closely as possible the retina for the design of a novel image coder while keeping encouraging performances. Second, we bring a new insight concerning the non-deterministic behavior of the retina.     

An ensemble classification-based approach applied to retinal blood vessel segmentation

This paper presents a new supervised method for segmentation of blood vessels in retinal photographs. This method uses an ensemble system of bagged and boosted decision trees and utilizes a feature vector based on the orientation analysis of gradient vector field, morphological transformation, line strength measures, and Gabor filter responses. The feature vector encodes information to handle the healthy as well as the pathological retinal image. The method is evaluated on the publicly available DRIVE and STARE databases, frequently used for this purpose and also on a new public retinal vessel reference dataset CHASE_DB1?which is a subset of retinal images of multiethnic children from the Child Heart and Health Study in England (CHASE) dataset. The performance of the ensemble system is evaluated in detail and the incurred accuracy, speed, robustness, and simplicity make the algorithm a suitable tool for automated retinal image analysis.     

基于多模型融合和区域迭代生长的视网膜血管自动分割

糖尿病视网膜病变是成年人致盲首因,视网膜血管分割是诊断糖尿病视网膜病变的基础.为提高视网膜血管分割准确性,提出一种基于多模型融合和区域迭代生长的视网膜血管自动分割算法.首先,预处理后分别构建数学形态学、匹配滤波器、尺度空间分析、多尺度线检测和神经网络模型初步分割视网膜血管,为减少噪声取五个分割结果的均值作为初步输出.其次,设计掩膜分离渗出物和视盘,将数学形态学模型分割结果替换掩膜白色区域,并融合初步输出生成组合结果.最后,考虑视网膜血管先验知识,对组合结果阈值分割和区域迭代生长后获取最终结果.实验结果表明,该算法分割DRIVE和STARE眼底图像库视网膜血管的检测精度、敏感度和特异性分别为0.9457、0.7843、0.9815以及0.9472、0.7826、0.9803,优于多数经典算法.     

Iterative Channel Decoding for PC-Based Software Radio DVB-T Receiver

The previous scheme only modifies the Viterbi decoder input positions where the Reed-Solomon decoder makes sure they are correct. Although it is simple and can get perfect image quality in static channels, it still suffer image quality degradation in mobile channels. Besides that, we propose to change the parameter Q of the metrics of the Viterbi decoder input positions where the Reed-Solomon decoder makes sure they are incorrect. The Q is defined as the norm of channel state information (CSI), Q = |CSI|². We propose to change Q to maximum for the Viterbi decoder input positions where the Reed-Solomon decoder makes sure they are incorrect, so the incorrect paths are unlikely to be selected in the Viterbi algorithm. The field test results in 100 km/h moving speed show that the proposed scheme with Q modification outperforms the previous scheme without Q modification by 0.36 % (7.45–>7.09 %) in symbol error rate and noticeable image quality difference.     

多尺度特征融合双U型视网膜分割算法

视网膜血管形态结构是反映人体健康的重要指标 ,针对现有视网膜血管分割存在主 血管模糊、微细血管断裂和视盘误分割等问题,提出多尺度特征融合双U型视网膜分割算 法。首先,利用低层U-Net高效循环残差模块对眼底图像进行粗粒度分割,得到视网膜血 管 初步轮廓。其次,将粗分割图与原始特征图像素相乘送入高层U-Net,利用其缩放宽残差 模 块进行细粒度图像解码,丰富视网膜血管细节信息。同时利用3路径注意力机制复合性连接 双网络的编码层与解码层,实现特征映射跨网络传播,减小上下文语义差异。最后,融合双 层网络输出提取血管区域,双U 型网络能够更深层次提取血管像素,精准分割出视网膜细 节。在DRIVE与STARE数据集上进行实验,其准确率分别为96.45%和97.02%,敏感度分 别为83.35%和81.40%,特异性分别为98.38%和 98.83%,总体性能优于现有算法。     

基于视网膜血管模型的图像分割与血管提取

本文采用非线性图像处理方法,对眼底图像进行去噪增强和分析,提取血管中心线,以便于对视网膜血管网络进行定量描述和定量分析。实验表明该方法能好地从非荧光眼底图像中提取出膜血管网络。     

Detection of optic disc in retinal images by means of a geometrical model of vessel structure

We present here a new method to identify the position of the optic disc (OD) in retinal fundus images. The method is based on the preliminary detection of the main retinal vessels. All retinal vessels originate from the OD and their path follows a similar directional pattern (parabolic course) in all images. To describe the general direction of retinal vessels at any given position in the image, a geometrical parametric model was proposed, where two of the model parameters are the coordinates of the OD center. Using as experimental data samples of vessel centerline points and corresponding vessel directions, provided by any vessel identification procedure, model parameters were identified by means of a simulated annealing optimization technique. These estimated values provide the coordinates of the center of OD. A Matlab prototype implementing this method was developed. An evaluation of the proposed procedure was performed using the set of 81 images from the STARE project, containing images from both normal and pathological subjects. The OD position was correctly identified in 79 out of 81 images (98%), even in rather difficult pathological situations.     

基于多尺度多路径FCN的视网膜血管分割

视网膜血管的形态结构信息可以为糖尿病、高血压等疾病提供诊断依据。提出了一种基于多尺度多路径的全卷积神经网络的视网膜血管分割方法。首先,利用空洞卷积代替池化层和上采样操作,在不增加参数的情况下增加感受野,避免了细节信息的丢失;其次,通过使用不同空洞率的空洞卷积实现图像数据的多尺度特征提取,充分学习图像的多尺度特征,避免网络过深,并提升了细小血管的提取能力;同时,利用跳层结构在网络中建立多条信息流通路径,通过多路径信息流充分传递多尺度特征信息,提高网络预测效果。实验结果表明,该算法在DRIVE数据集上的平均准确度、灵敏度和特异性分别为95.46%.81.24%,97.77%,取得了较好的视网膜血管的分割效果。     

Structural study of degraded ZnMgSSe blue light emitters

Using electroluminescence (EL) topography and transmission electron microscopy (TEM), we investigated the nonluminescent regions which form while current is being injected into ZnMgSSe/ZnSSe/ZnCdSe-based blue light emitters. Small dark spots were observed just after turn-on and spread out forming rough nonluminescent triangles in the <100> directions in the EL image of the active region. TEM studies showed that the small dark spots are pre-existing stacking faults originating at the substrate/epitaxial layer interface. The nonluminescent triangles were found to be a dense region of dislocation dipoles and dislocation loops. Each dipole was aligned along two <110> directions in the {111} planes. The Burgers vectors were of the type a/2<011> inclined at 45° to the (001) junction plane.     

一种新的视网膜血管网络自动分割方法

提出了一种基于脉冲耦合神经网络(PCNN)和分布式遗传算法(DGA)的视网膜血管自动分割方法.首先采用二维高斯匹配滤波器预处理以增强血管,然后采用DGA快速搜索出PCNN的最佳参数设置值并运用PCNN分割出增强图像的血管网络,最后对分割得到的血管网络结合区域连通性特征,采用面积滤波算子滤除噪声,提取出最终的血管网络.通过在国际上公开的Hoover眼底图像库中的实验,结果表明,该方法在血管分支提取和算法有效性方面明显优于Hoover算法,具有较高的临床应用价值.     

Zernike moments and LDP-weighted patches for content-based image retrieval

This paper proposes a new algorithm using global and local features for content-based image retrieval. Global features are extracted using the magnitude of Zernike moments (ZMs). Local features are obtained through local directional pattern (LDP). Generally, LDP is used to extract texture-based features from an image. In this paper, LDP is used to encode both texture and shape information of an image to represent more meaningful features. To encode texture-based features, original image is used to compute the LDP features. To extract shape information from an image, dual-tree complex wavelet transform (DT-CWT) is applied on image which generates six directional wavelets. These six directional wavelets are superimposed in order to obtain shape-encoded image. LDP is then applied on this wavelet-based shape-encoded image. Further, to enhance retrieval accuracy, LDP features are extracted from patches of both original and shape-encoded images. These patches are assigned with weights based on average discrimination capability of features in a patch. Experiments are performed using three different standard databases with various variations such as pose, distortion, partial occlusion and complex structure. The proposed technique achieves 96.4 and 98.76 % retrieval accuracy at a recall of 50 %, for Kimia-99 and COIL-100 databases, respectively. For MPEG-7 CE-2 shape database, retrieval accuracy of 61.93 % is achieved in terms of average Bull’s eye performance (BEP). The proposed technique is also tested on Springer medical image database to explore its scope in other areas, wherein it attains average BEP of 69.68 % in comparison with 61.52 % with ZMs. It is observed that the proposed technique outperforms other well-known existing methods of image retrieval.     

A ΣΔ micro accelerometer with 6 µg/√Hz resolution and 130 dB dynamic range

This paper reports the implementation of a low noise, high dynamic-range ΣΔ readout for low cost capacitive Micro-Electro-Mechanical Systems (MEMS) accelerometers. The readout circuit sets the bandwidth of the ΣΔ loop through an extra feedback path, and hence allows the closed loop system to operate with the low noise characteristics similar to a second-order ΣΔ analog-to-digital converter. A thorough noise analysis of the proposed accelerometer shows that the mechanical noise is the most significant source and quantization noise is mostly eliminated. Dynamic range (DR) of the system is improved by minimizing the circuit noise and increasing the full scale range (FSR) by high-voltage pulse feedback. Utilization of these techniques allows the implementation of low cost, low noise, and high DR navigation-grade accelerometers, by eliminating the need for large proof mass, large area MEMS sensors. The proposed system can achieve a minimum of 6.0 µg/√Hz noise floor, 3.2 µg bias instability, and a maximum of 130 dB DR at 1 Hz. A FSR of ±20 g is reported for 6.2 µg/√Hz noise floor. This range can be increased up to ±40 g at the cost of noise performance and DR.     

Architectured Materials to Improve the Reliability of Power Electronics Modules: Substrate and Lead-Free Solder

Power electronics modules (>100 A, >500 V) are essential components for the development of electrical and hybrid vehicles. These modules are formed from silicon chips (transistors and diodes) assembled on copper substrates by soldering. Owing to the fact that the assembly is heterogeneous, and because of thermal gradients, shear stresses are generated in the solders and cause premature damage to such electronics modules. This work focuses on architectured materials for the substrate and on lead-free solders to reduce the mechanical effects of differential expansion, improve the reliability of the assembly, and achieve a suitable operating temperature (<175°C). These materials are composites whose thermomechanical properties have been optimized by numerical simulation and validated experimentally. The substrates have good thermal conductivity (>280 W m^?1 K^?1) and a macroscopic coefficient of thermal expansion intermediate between those of Cu and Si, as well as limited structural evolution in service conditions. An approach combining design, optimization, and manufacturing of new materials has been followed in this study, leading to improved thermal cycling behavior of the component.