CERES :: Browsing by Author "Zhao, Yitian"

Browsing by Author "Zhao, Yitian"

Now showing 1 - 20 of 23

Open Access
Angle-closure assessment in anterior segment OCT images via deep learning
(Elsevier, 2021-01-07) Hao, Huaying; Zhao, Yitian; Yan, Qifeng; Higashita, Risa; Zhang, Jiong; Zhao, Yifan; Xu, Yanwu; Li, Fei; Zhang, Xiulan; Liu, Jiang
Precise characterization and analysis of anterior chamber angle (ACA) are of great importance in facilitating clinical examination and diagnosis of angle-closure disease. Currently, the gold standard for diagnostic angle assessment is observation of ACA by gonioscopy. However, gonioscopy requires direct contact between the gonioscope and patients’ eye, which is uncomfortable for patients and may deform the ACA, leading to false results. To this end, in this paper, we explore a potential way for grading ACAs into open-, appositional- and synechial angles by Anterior Segment Optical Coherence Tomography (AS-OCT), rather than the conventional gonioscopic examination. The proposed classification schema can be beneficial to clinicians who seek to better understand the progression of the spectrum of angle-closure disease types, so as to further assist the assessment and required treatment at different stages of angle-closure disease. To be more specific, we first use an image alignment method to generate sequences of AS-OCT images. The ACA region is then localized automatically by segmenting an important biomarker - the iris - as this is a primary structural cue in identifying angle-closure disease. Finally, the AS-OCT images acquired in both dark and bright illumination conditions are fed into our Multi-Sequence Deep Network (MSDN) architecture, in which a convolutional neural network (CNN) module is applied to extract feature representations, and a novel ConvLSTM-TC module is employed to study the spatial state of these representations. In addition, a novel time-weighted cross-entropy loss (TC) is proposed to optimize the output of the ConvLSTM, and the extracted features are further aggregated for the purposes of classification. The proposed method is evaluated across 66 eyes, which include 1584 AS-OCT sequences, and a total of 16,896 images. The experimental results show that the proposed method outperforms existing state-of-the-art methods in applicability, effectiveness, and accuracy.
Open Access
Automated tortuosity analysis of nerve fibers in corneal confocal microscopy
(IEEE, 2020-02-17) Zhao, Yitian; Zhang, Jiong; Pereira, Ella; Zheng, Yalin; Su, Pan; Xie, Jianyang; Zhao, Yifan; Shi, Yonggang; Qi, Hong; Liu, Jiang; Liu, Yonghuai
Precise characterization and analysis of corneal nerve fiber tortuosity are of great importance in facilitating examination and diagnosis of many eye-related diseases. In this paper we propose a fully automated method for image-level tortuosity estimation, comprising image enhancement, exponential curvature estimation, and tortuosity level classification. The image enhancement component is based on an extended Retinex model, which not only corrects imbalanced illumination and improves image contrast in an image, but also models noise explicitly to aid removal of imaging noise. Afterwards, we take advantage of exponential curvature estimation in the 3D space of positions and orientations to directly measure curvature based on the enhanced images, rather than relying on the explicit segmentation and skeletonization steps in a conventional pipeline usually with accumulated pre-processing errors. The proposed method has been applied over two corneal nerve microscopy datasets for the estimation of a tortuosity level for each image. The experimental results show that it performs better than several selected state-of-the-art methods. Furthermore, we have performed manual gradings at tortuosity level of four hundred and three corneal nerve microscopic images, and this dataset has been released for public access to facilitate other researchers in the community in carrying out further research on the same and related topics.
Open Access
Automatic 2-D/3-D vessel enhancement in multiple modality images using a weighted symmetry filter
(IEEE, 2017-09-26) Zhao, Yitian; Zhao, Yitian; Zheng, Yalin; Liu, Yonghuai; Zhao, Yifan; Luo, Lingling; Yang, Siyuan; Na, Tong; Wang, Yongtian; Liu, Jiang
Automated detection of vascular structures is of great importance in understanding the mechanism, diagnosis and treatment of many vascular pathologies. However, automatic vascular detection continues to be an open issue because of difficulties posed by multiple factors such as poor contrast, inhomogeneous backgrounds, anatomical variations, and the presence of noise during image acquisition. In this paper, we propose a novel 2D/3D symmetry filter to tackle these challenging issues for enhancing vessels from different imaging modalities. The proposed filter not only considers local phase features by using a quadrature filter to distinguish between lines and edges, but also uses the weighted geometric mean of the blurred and shifted responses of the quadrature filter, which allows more tolerance of vessels with irregular appearance. As a result, this filter shows a strong response to the vascular features under typical imaging conditions. Results based on 8 publicly available datasets (six 2D datasets, one 3D dataset and one 3D synthetic dataset) demonstrate its superior performance to other state-ofthe- art methods.
Open Access
A compactness based saliency approach for leakages detection in fluorescein angiogram
(Springer Verlag, 2016-07-26) Zhao, Yitian; Su, Pan; Yang, Jian; Zhao, Yifan; Zheng, Yalin; Wang, Yongtian
This study has developed a novel saliency detection method based on compactness feature for detecting three common types of leakage in retinal fluorescein angiogram: large focal, punctate focal, and vessel segment leakage. Leakage from retinal vessels occurs in a wide range of retinal diseases, such as diabetic maculopathy and paediatric malarial retinopathy. The proposed framework consists of three major steps: saliency detection, saliency refinement and leakage detection. First, the Retinex theory is adapted to address the illumination inhomogeneity problem. Then two saliency cues, intensity and compactness, are proposed for the estimation of the saliency map of each individual superpixel at each level. The saliency maps at different levels over the same cues are fused using an averaging operator. Finally, the leaking sites can be detected by masking the vessel and optic disc regions. The effectiveness of this framework has been evaluated by applying it to different types of leakage images with cerebral malaria. The sensitivity in detecting large focal, punctate focal and vessel segment leakage is 98.1, 88.2 and 82.7 %, respectively, when compared to a reference standard of manual annotations by expert human observers. The developed framework will become a new powerful tool for studying retinal conditions involving retinal leakage.
Open Access
COSTA: a multi-center TOF-MRA dataset and a style self-consistency network for cerebrovascular segmentation
(IEEE, 2024-12) Mou, Lei; Yan, Qifeng; Lin, Jinghui; Zhao, Yifan; Liu, Yonghuai; Ma, Shaodong; Zhang, Jiong; Lv, Wenhao; Zhou, Tao; Frangi, Alejandro F.; Zhao, Yitian
Time-of-flight magnetic resonance angiography (TOF-MRA) is the least invasive and ionizing radiation-free approach for cerebrovascular imaging, but variations in imaging artifacts across different clinical centers and imaging vendors result in inter-site and inter-vendor heterogeneity, making its accurate and robust cerebrovascular segmentation challenging. Moreover, the limited availability and quality of annotated data pose further challenges for segmentation methods to generalize well to unseen datasets. In this paper, we construct the largest and most diverse TOF-MRA dataset (COSTA) from 8 individual imaging centers, with all the volumes manually annotated. Then we propose a novel network for cerebrovascular segmentation, namely CESAR, with the ability to tackle feature granularity and image style heterogeneity issues. Specifically, a coarse-to-fine architecture is implemented to refine cerebrovascular segmentation in an iterative manner. An automatic feature selection module is proposed to selectively fuse global long-range dependencies and local contextual information of cerebrovascular structures. A style self-consistency loss is then introduced to explicitly align diverse styles of TOF-MRA images to a standardized one. Extensive experimental results on the COSTA dataset demonstrate the effectiveness of our CESAR network against state-of-the-art methods. We have made 6 subsets of COSTA with the source code online available, in order to promote relevant research in the community.
Open Access
A dementia classification framework using frequency and time-frequency features based on EEG signals
(IEEE, 2019-04-04) Durongbhan, Pholpat; Zhao, Yifan; Chen, Liangyu; Zis, Panagiotis; De Marco, Matteo; Unwin, Zoe C.; Venneri, Annalena; He, Xiongxiong; Li, Sheng; Zhao, Yitian; Blackburn, Daniel J.; Sarrigiannis, Ptolemaios G.
Alzheimer’s disease (AD) accounts for 60%–70% of all dementia cases, and clinical diagnosis at its early stage is extremely difficult. As several new drugs aiming to modify disease progression or alleviate symptoms are being developed, to assess their efficacy, novel robust biomarkers of brain function are urgently required. This paper aims to explore a routine to gain such biomarkers using the quantitative analysis of electroencephalography (QEEG). This paper proposes a supervised classification framework that uses EEG signals to classify healthy controls (HC) and AD participants. The framework consists of data augmentation, feature extraction, K-nearest neighbor (KNN) classification, quantitative evaluation, and topographic visualization. Considering the human brain either as a stationary or a dynamical system, both the frequency-based and time–frequency-based features were tested in 40 participants. The results show that: 1) the proposed method can achieve up to a 99% classification accuracy on short (4s) eyes open EEG epochs, with the KNN algorithm that has best performance when compared with alternative machine learning approaches; 2) the features extracted using the wavelet transform produced better classification performance in comparison to the features based on FFT; and 3) in the spatial domain, the temporal and parietal areas offer the best distinction between healthy controls and AD. The proposed framework can effectively classify HC and AD participants with high accuracy, meanwhile offering identification and the localization of significant QEEG features. These important findings and the proposed classification framework could be used for the development of a biomarker for the diagnosis and monitoring of disease progression in AD.
Open Access
Early detection of dementia through retinal imaging and trustworthy AI
(Springer , 2024-10-04) Hao, Jinkui; Kwapong, William R.; Shen, Ting; Fu, Huazhu; Xu, Yanwu; Lu, Qinkang; Liu, Shouyue; Zhang, Jiong; Liu, Yonghuai; Zhao, Yifan; Zheng, Yalin; Frangi, Alejandro F.; Zhang, Shuting; Qi, Hong; Zhao, Yitian
Alzheimer's disease (AD) is a global healthcare challenge lacking a simple and affordable detection method. We propose a novel deep learning framework, Eye-AD, to detect Early-onset Alzheimer's Disease (EOAD) and Mild Cognitive Impairment (MCI) using OCTA images of retinal microvasculature and choriocapillaris. Eye-AD employs a multilevel graph representation to analyze intra- and inter-instance relationships in retinal layers. Using 5751 OCTA images from 1671 participants in a multi-center study, our model demonstrated superior performance in EOAD (internal data: AUC = 0.9355, external data: AUC = 0.9007) and MCI detection (internal data: AUC = 0.8630, external data: AUC = 0.8037). Furthermore, we explored the associations between retinal structural biomarkers in OCTA images and EOAD/MCI, and the results align well with the conclusions drawn from our deep learning interpretability analysis. Our findings provide further evidence that retinal OCTA imaging, coupled with artificial intelligence, will serve as a rapid, noninvasive, and affordable dementia detection.
Open Access
An edge detection method using outer totalistic cellular automata
(Elsevier, 2016-05-20) Amrogowicz, Sebastian; Zhao, Yitian; Zhao, Yifan
A number of Cellular Automata (CA)-based edge detectors have been developed recently due to the simplicity of the model and the potential for simultaneous removal of different types of noise in the process of detection. This paper introduced a novel edge detector using Outer Totalistic Cellular Automata. Its performance has been compared with other recently developed CA-based edge detectors, in addition to some classic methods, through testing images from a public library. Visual and quantitative measurement of similarity with manually marked correct edges confirmed the superiority of the proposed method over conventional and state-of-the-art CA-based edge detectors.
Open Access
Engaging students for the learning and assessment of the advanced computer graphics module using the latest technologies
(inScience Press, 2017-07) Liu, Yonghuai; Yang, Longzhi; Han, Jiwan; Lu, Bin; Yuen, Peter W. T.; Zhao, Yitian; Song, Ran
The advanced computer graphics has been one of the most basic and landmark modules in the field of computer science. It usually covers such topics as core mathematics, lighting and shading, texture mapping, colour and depth, and advanced modeling. All such topics involve mathematics for object modeling and transformation, and programming for object visualization and interaction. While some students are not as good in either mathematics or programming, it is usually a challenge to teach computer graphics to these students effectively. This is because it is difficult for students to link mathematics and programming with what they used to see in video games and the TV advertisements for example and thus they can easily be put off. In this paper, we investigate how the latest technologies can help alleviate the teaching and learning tasks. Instead of selecting the low level programming languages for demonstration and assignment such as Java, Java 3D, C++, or OpenGL, we selected Three.js, which is one of the latest and freely accessible 3D graphics libraries. It has a unique advantage that it provides a seamless interface between the main stream web browsers and 2D/3D graphics. The developed code can be run on a web browser such as Firefox, Chrome, or Safari for testing, debugging and visualization without code changing. The unique design patterns and objectives of Three.js can be very attractive to third party software houses to develop auxiliary functions, methods and tutorials and to make them freely available for the public. Such a unique property of Three.js and its widely available supporting resources are especially helpful to engage students, inspire their learning and facilitate teaching. To evaluate the effectiveness for using Three.js in teaching computer graphics we have set up an assignment for scene modeling in the last 4 years with focuses on the quality of the simulated scene (50%) and the quality of the assignment report (50%). We have evaluated different assessment forms of the module that we taught in the last four years: in 2013-2014 the module consisted of 20% assignment and 80% exam based on Java 3D; in 2014-2015 the same proportion of assignment/exam but based on WebGL, in 2015-2016 the module was 50-50% of assignment and exam but based on Three.js; and in this year the module is 100% assignment based on Three.js. The effectiveness of the module delivery has been evaluated both qualitatively and quantitatively from five aspects: a) average marks of students, b) moderator report, c) module evaluation questionnaire, d) external examiner’s comments and e) examination board recommendations. The results have shown that Three.js is indeed more successful in engaging students for learning and the 100% assignment assessment enables students to focus more on the design and development. This four year result is really encouraging to us as an educational institute to embrace the latest technologies for the delivery of such challenging modules as computer graphics and machine learning.
Open Access
Imaging of nonlinear and dynamic functional brain connectivity based on EEG recordings with the application on the diagnosis of Alzheimer's disease
(IEEE, 2019-11-14) Zhao, Yifan; Zhao, Yitian; Durongbhan, Pholpat; Chen, Liangyu; Liu, Jiang; Billings, S. A.; Zis, Panagiotis; Unwin, Zoe C.; De Marco, Matteo; Venneri, Annalena; Blackburn, Daniel J.; Sarrigiannis, Ptolemaios G.
Since age is the most significant risk factor for the development of Alzheimer’s disease (AD), it is important to understand the effect of normal ageing on brain network characteristics before we can accurately diagnose the condition based on information derived from resting state electroencephalogram (EEG) recordings, aiming to detect brain network disruption. This paper proposes a novel brain functional connectivity imaging method, particularly targeting the contribution of nonlinear dynamics of functional connectivity, on distinguishing participants with AD from healthy controls (HC). We describe a parametric method established upon a Nonlinear Finite Impulse Response model, and a revised orthogonal least squares algorithm used to estimate the linear, nonlinear and combined connectivity between any two EEG channels without fitting a full model. This approach, where linear and non-linear interactions and their spatial distribution and dynamics can be estimated independently, offered us the means to dissect the dynamic brain network disruption in AD from a new perspective and to gain some insight into the dynamic behaviour of brain networks in two age groups (above and below 70) with normal cognitive function. Although linear and stationary connectivity dominates the classification contributions, quantitative results have demonstrated that nonlinear and dynamic connectivity can significantly improve the classification.
Open Access
Intensity and compactness enabled saliency estimation for leakage detection in diabetic and malarial retinopathy
(Institute of Electrical and Electronics Engineers, 2016-07-21) Zhao, Yitian; Zheng, Yalin; Liu, Yonghuai; Yang, Jian; Zhao, Yifan; Chen, Duanduan; Wang, Yongtian
Leakage in retinal angiography currently is a key feature for confirming the activities of lesions in the management of a wide range of retinal diseases, such as diabetic maculopathy and paediatric malarial retinopathy. This paper proposes a new saliency-based method for the detection of leakage in fluorescein angiography. A superpixel approach is firstly employed to divide the image into meaningful patches (or superpixels) at different levels. Two saliency cues, intensity and compactness, are then proposed for the estimation of the saliency map of each individual superpixel at each level. The saliency maps at different levels over the same cues are fused using an averaging operator. The two saliency maps over different cues are fused using a pixel-wise multiplication operator. Leaking regions are finally detected by thresholding the saliency map followed by a graph-cut segmentation. The proposed method has been validated using the only two publicly available datasets: one for malarial retinopathy and the other for diabetic retinopathy. The experimental results show that it outperforms one of the latest competitors and performs as well as a human expert for leakage detection and outperforms several state-of-the-art methods for saliency detection.
Open Access
A miniaturised active thermography system to inspect composite laminates
(IEEE, 2020-10-13) Du, Weixiang; Liu, Haochen; Zhao, Yitian; Sirikham, Adisorn; Addepalli, Sri; Zhao, Yifan
With the rapid increase of the integration and complexity of industrial components, the inaccessibility and inapplicability of existing Non-destructive testing devices have become a bottleneck for in-situ inspection of these objects. This paper introduces a miniaturised active thermography system featured with a small size, low resolution and low-cost thermal sensor, where two optional excitation sources including flash and laser are integrated. Dedicated data analysis approaches to evaluate defects are proposed considering the degraded signal quality. Three carbon fibre reinforced polymer laminates with a variety of defects are evaluated quantitatively and qualitatively using the proposed system by comparing with two existing non-miniaturised inspection systems. The results show that the proposed system can work effectively for the degradation assessment of composite laminates. Even with the technical limitations that affect the detectability, for instance, the low pixel resolution, this technique will play an important role to inspect components featured with geometrically intricate space
Open Access
An optimal factor analysis approach to improve the wavelet-based image resolution enhancement techniques
(Global Journal of Computer Science and Technology, 2016-12-17) Witwit, Wasnaa; Zhao, Yitian; Jenkins, Karl W.; Zhao, Yifan
The existing wavelet-based image resolution enhancement techniques have many assumptions, such as limitation of the way to generate low-resolution images and the selection of wavelet functions, which limits their applications in different fields. This paper initially identifies the factors that effectively affect the performance of these techniques and quantitatively evaluates the impact of the existing assumptions. An approach called Optimal Factor Analysis employing the genetic algorithm is then introduced to increase the applicability and fidelity of the existing methods. Moreover, a new Figure of Merit is proposed to assist the selection of parameters and better measure the overall performance. The experimental results show that the proposed approach improves the performance of the selected image resolution enhancement methods and has potential to be extended to other methods.
Open Access
Randomness-restricted diffusion model for ocular surface structure segmentation
(IEEE, 2025-03) Guo, Xinyu; Wen, Han; Hao, Huaying; Zhao, Yifan; Meng, Yanda; Liu, Jiang; Zheng, Yalin; Chen, Wei; Zhao, Yitian
Ocular surface diseases affect a significant portion of the population worldwide. Accurate segmentation and quantification of different ocular surface structures are crucial for the understanding of these diseases and clinical decision-making. However, the automated segmentation of the ocular surface structure is relatively unexplored and faces several challenges. Ocular surface structure boundaries are often inconspicuous and obscured by glare from reflections. In addition, the segmentation of different ocular structures always requires training of multiple individual models. Thus, developing a one-model-fits-all segmentation approach is desirable. In this paper, we introduce a randomness-restricted diffusion model for multiple ocular surface structure segmentation. First, a time-controlled fusion-attention module (TFM) is proposed to dynamically adjust the information flow within the diffusion model, based on the temporal relationships between the network’s input and time. TFM enables the network to effectively utilize image features to constrain the randomness of the generation process. We further propose a low-frequency consistency filter and a new loss to alleviate model uncertainty and error accumulation caused by the multi-step denoising process. Extensive experiments have shown that our approach can segment seven different ocular surface structures. Our method performs better than both dedicated ocular surface segmentation methods and general medical image segmentation methods. We further validated the proposed method over two clinical datasets, and the results demonstrated that it is beneficial to clinical applications, such as the meibomian gland dysfunction grading and aqueous deficient dry eye diagnosis.
Open Access
Range images
(Wiley, 2019-05-10) Liu, Yonghuai; Yuen, Peter W. T.; Pang, Yanwei; Zhao, Yitian; Rosin, Paul L.
This article gives an overview of range‐imaging techniques with an aim to let the reader better understand how the difficult issue, such as the registration of overlapping range images, can be approached and solved. It firstly introduces the characteristics of range images and highlights examples of 3D image visualizations, associated technical issues, applications, and the differences of range imaging with respect to the traditional digital broadband imaging. Subsequently, one of the most popular feature extraction and matching methods, the signature of histograms of orientations (SHOT) method, is then outlined. However, the “matched” points generated by SHOT usually generate high proportion of false positives due to various factors such as imaging noise, lack of features, and cluttered backgrounds. Thus, the article discusses more about image‐matching issues, particularly to emphasize how the widely employed range image alignment technique, the random sample consensus (RANSAC) method, is compared with a simple, yet effective, technique based on normalized error penalization (NEP). This simple NEP method utilizes a strategy to penalize point matches whose errors are far away from the majority. The capability of the method for the evaluation of point matches between overlapping range images is illustrated by experiments using real range image data sets. Interestingly enough, these range images appear to be easier to register than expected. Finally, some conclusions have been drawn and further readings for other fundamental techniques and concepts have been suggested.
Open Access
Region-based saliency estimation for 3D shape analysis and understanding
(Elsevier, 2016-02-01) Zhao, Yitian; Liu, Yonghuai; Wang, Yongjun; Wei, Baogang; Yang, Jian; Zhao, Yifan; Wang, Yongtian
The detection of salient regions is an important pre-processing step for many 3D shape analysis and understanding tasks. This paper proposes a novel method for saliency detection in 3D free form shapes. Firstly, we smooth the surface normals by a bilateral filter. Such a method is capable of smoothing the surfaces and retaining the local details. Secondly, a novel method is proposed for the estimation of the saliency value of each vertex. To this end, two new features are defined: Retinex-based Importance Feature (RIF) and Relative Normal Distance (RND). They are based on the human visual perception characteristics and surface geometry respectively. Since the vertex based method cannot guarantee that the detected salient regions are semantically continuous and complete, we propose to refine such values based on surface patches. The detected saliency is finally used to guide the existing techniques for mesh simplification, interest point detection, and overlapping point cloud registration. The comparative studies based on real data from three publicly accessible databases show that the proposed method usually outperforms five selected state of the art ones both qualitatively and quantitatively for saliency detection and 3D shape analysis and understanding.
Open Access
Retinal vascular network topology reconstruction and artery/vein classification via dominant set clustering
(IEEE, 2019-07-03) Zhao, Yitian; Xie, Jianyang; Zhang, Huaizhong; Zheng, Yalin; Zhao, Yifan; Qi, Hong; Zhao, Yangchun; Su, Pan; Liu, Jiang; Liu, Yonghuai
The estimation of vascular network topology in complex networks is important in understanding the relationship between vascular changes and a wide spectrum of diseases. Automatic classification of the retinal vascular trees into arteries and veins is of direct assistance to the ophthalmologist in terms of diagnosis and treatment of eye disease. However, it is challenging due to their projective ambiguity and subtle changes in appearance, contrast and geometry in the imaging process. In this paper, we propose a novel method that is capable of making the artery/vein (A/V) distinction in retinal color fundus images based on vascular network topological properties. To this end, we adapt the concept of dominant set clustering and formalize the retinal blood vessel topology estimation and the A/V classification as a pairwise clustering problem. The graph is constructed through image segmentation, skeletonization and identification of significant nodes. The edge weight is defined as the inverse Euclidean distance between its two end points in the feature space of intensity, orientation, curvature, diameter, and entropy. The reconstructed vascular network is classified into arteries and veins based on their intensity and morphology. The proposed approach has been applied to five public databases, INSPIRE, IOSTAR, VICAVR, DRIVE and WIDE, and achieved high accuracies of 95.1%, 94.2%, 93.8%, 91.1%, and 91.0%, respectively. Furthermore, we have made manual annotations of the blood vessel topologies for INSPIRE, IOSTAR, VICAVR, and DRIVE datasets, and these annotations are released for public access so as to facilitate researchers in the community.
Open Access
Retinal vascular segmentation using superpixel-based line operator and its application to vascular topology estimation
(2018-05-09) Na, Tong; Xie, Jianyang; Zhao, Yitian; Zhao, Yifan; Liu, Yue; Wang, Yongtian; Liu, Jiang
Purpose: Automatic methods of analyzing of retinal vascular networks, such as retinal blood vessel detection, vascular network topology estimation, and arteries / veins classi cation are of great assistance to the ophthalmologist in terms of diagnosis and treatment of a wide spectrum of diseases. Methods: We propose a new framework for precisely segmenting retinal vasculatures, constructing retinal vascular network topology, and separating the arteries and veins. A non-local total variation inspired Retinex model is employed to remove the image intensity inhomogeneities and relatively poor contrast. For better generalizability and segmentation performance, a superpixel based line operator is proposed as to distinguish between lines and the edges, thus allowing more tolerance in the position of the respective contours. The concept of dominant sets clustering is adopted to estimate retinal vessel topology and classify the vessel network into arteries and veins. Results: The proposed segmentation method yields competitive results on three pub- lic datasets (STARE, DRIVE, and IOSTAR), and it has superior performance when com- pared with unsupervised segmentation methods, with accuracy of 0.954, 0.957, and 0.964, respectively. The topology estimation approach has been applied to ve public databases 1 (DRIVE,STARE, INSPIRE, IOSTAR, and VICAVR) and achieved high accuracy of 0.830, 0.910, 0.915, 0.928, and 0.889, respectively. The accuracies of arteries / veins classi cation based on the estimated vascular topology on three public databases (INSPIRE, DRIVE and VICAVR) are 0.90.9, 0.910, and 0.907, respectively. Conclusions: The experimental results show that the proposed framework has e ectively addressed crossover problem, a bottleneck issue in segmentation and vascular topology recon- struction. The vascular topology information signi cantly improves the accuracy on arteries / veins classi cation.
Open Access
Saliency driven vasculature segmentation with infinite perimeter active contour model
(Elsevier, 2017-02-22) Zhao, Yitian; Zhao, Jingliang; Yang, Jian; Liu, Yonghuai; Zhao, Yifan; Zheng, Yalin; Xia, Likun; Wang, Yongtian
Automated detection of retinal blood vessels plays an important role in advancing the understanding of the mechanism, diagnosis and treatment of cardiovascular disease and many systemic diseases, such as diabetic retinopathy and age-related macular degeneration. Here, we propose a new framework for precisely segmenting retinal vasculatures. The proposed framework consists of three steps. A non-local total variation model is adapted to the Retinex theory, which aims to address challenges presented by intensity inhomogeneities, and the relatively low contrast of thin vessels compared to the background. The image is then divided into superpixels, and a compactness-based saliency detection method is proposed to locate the object of interest. For better general segmentation performance, we then make use of a new infinite active contour model to segment the vessels in each superpixel. The proposed framework has wide applications, and the results show that our model outperforms its competitors.
Open Access
Satellite image resolution enhancement using discrete wavelet transform and new edge-directed interpolation
(Society of Photo-optical Instrumentation Engineers (SPIE), 2017-03-30) Witwit, Wasnaa; Zhao, Yifan; Jenkins, Karl W.; Zhao, Yitian
An image resolution enhancement approach based on discrete wavelet transform (DWT) and new edge-directed interpolation (NEDI) for degraded satellite images by geometric distortion to correct the errors in image geometry and recover the edge details of directional high-frequency subbands is proposed. The observed image is decomposed into four frequency subbands through DWT, and then the three high-frequency subbands and the observed image are processed with NEDI. To better preserve the edges and remove potential noise in the estimated high-frequency subbands, an adaptive threshold is applied to process the estimated wavelet coefficients. Finally, the enhanced image is reconstructed by applying inverse DWT. Four criteria are introduced, aiming to better assess the overall performance of the proposed approach for different types of satellite images. A public satellite images data set is selected for the validation purpose. The visual and quantitative results show the superiority of the proposed approach over the conventional and state-of-the-art image resolution enhancement.