Abstract

Motion blur is one of the most common causes of image degradation. It is of increasing interest due to the deep penetration of digital cameras into consumer applications. In this paper, we start with a hypothesis that there is sufficient information within a blurred image and approach the deblurring problem as an optimisation process where the deblurring is to be done by satisfying a set of conditions. These conditions are derived from first principles underlying the degradation process assuming noise-free environments. We propose a novel but effective method for removing motion blur from a single blurred image via an iterative algorithm. The strength of this method is that it enables deblurring without resorting to estimation of the blur kernel or blur depth. The proposed iterative method has been tested on several images with different degrees of blur. The obtained results have been compared with state of the art techniques including those that require more than one input image. The results are consistently of high quality and comparable or superior to the existing methods which demonstrates the effectiveness of the proposed technique.

Abstract

We present a novel segmentation method to better capture the boundary of a non-homogeneous object such as the optic disk(OD), defined locally by two similar characteristic regions. Existing active contour models which utilise gradient information [12] or global region intensity [2] fail to localise aforementioned boundaries. We propose a region-based active contour model that uses local image information around each point of interest in multi-dimensional feature space. This model uses a local energy functional and level-set representation to achieve desired OD segmentation. The local energy functional defined on each image point provides sufficient information to determine a desired OD segmentation which is robust to the variations found in and around the OD region. This method has been evaluated against the segmentation provided by three medical experts on 138 retinal images. Both region and boundary-based assessment performed against two well established active contour models show strengths of the proposed method.

Abstract

The aim of a local descriptor or a feature descriptor is to efficiently represent the region detected by an interest point operator in a compact format for use in various applications related to matching. The common design principle behind most of the mainstream descriptors like SIFT, GLOH, Shape context etc is to capture the spatial distribution of features using histograms computed over a grid around interest points. Histograms provide compact representation but typically loose the spatial distribution information. In this paper, we propose to use projection-based representation to improve a descriptor’s capacity to capture spatial distribution information while retaining the invariance required. Based on this proposal, two descriptors based on the CS-LBP are introduced. The descriptors have been evaluated against known descriptors on a standard dataset and found to outperform, in most cases, the existing descriptors. The obtained results demonstrate that proposed approach has the advantages of both the statistical robustness of histogram and the capability of the projection based representation to capture spatial information.

Abstract

The presence of microaneurysms (MAs) is usually an early sign of diabetic retinopathy(DR) and their automatic detection from color retinal images is of clinical interest. In this paper, we present a new approach for automatic MA detection from digital colour fundus images. We formulate MA detection as a problem of target detection from clutter, where the probability of occurrence of target is considerably smaller compared to the clutter. A successive rejection-based strategy is proposed to progressively lower the number of clutter responses. The processing stages are designed to reject specific classes of clutter while passing majority of true MAs, using a set of specialized features. The true positives that remain after the final rejector are assigned a score which is based on its similarity to a true MA. Results of extensive evaluation of the proposed approach on three different retinal image datasets is reported, and are used to highlight the promise in the presented strategy.

Abstract

Multi-Modal image registration is the primary step in fusing complementary information contained in different imaging modalities for diagnostic purposes. We focus on two specific retinal imaging modalities namely, Color Fundus Image(CFI) and Fluroscein Fundus Angiogram(FFA). In this paper we investigate a projection based method using Radon transform for accurate matching in multi-modal retinal images. A novel Radon Transform based descriptor is proposed, which is invariant to illumination, rotation and partially to scale. Our results show that our descriptor is well suited for retinal images as it is robust to lesions, and works well even in poor quality images. The descriptor has been tested on a dataset of 126 images and compared for matching application against gradient based descriptors. The results show that Radon based descriptor outperforms the gradient based ones in both being able to discriminate between true and false matches and under presence of lesions.

Abstract

A growing number of public initiatives for screening the population for retinal disorders along with widespread availability of digital fundus (retina) cameras is leading to large accumulation of color fundus images. The ability to retrieve images based on pathologic state is a powerful functionality that has wide applications in evidence-based medicine, automated computer assisted diagnosis and in training ophthalmologists. In this paper, we propose a new methodology for content-based retrieval of retinal images showing symptoms of maculopathy. Taking the view of a disease region as one which exhibits deviation from the normal image background, a model for the image background is learnt and used to extract disease-affected image regions. These are then analysed to assess the severity level of maculopathy. Symmetry-based descriptor is derived for the macula region and employed for retrieval of images according to severity of maculopathy. The proposed approach has been tested on a publicly available dataset. The results show that background learning is successful as images with or no maculopathy are detected with a mean precision of 0.98. An aggregate precision of 0.89 is achieved for retrieval of images at three severity-levels of macular edema, demonstrating the potential offered by the proposed disease-based retrieval system.

Abstract

The shape deformation within the optic disk (OD) is an important indicator for the detection of glaucoma. In this paper, relevant disk parameters are estimated using the OD and cup boundaries. A deformable model guided by regional statistics is used to detect the OD boundary. A cup boundary detection scheme is presented based on the appearance of pallor in Lab colour space and the expected cup symmetry. The proposed scheme is tested on 170 images comprising 40 normal and 130 glaucomatous images. The proposed method gives a mean error 0.030 for normal and 0.121 for glaucomatous images in the estimation of cup-to-disk ratio which compares well with reported figures in literature.

Abstract

Interactive simulations encourage active/discovery learning in students. But developing simulations is a time consuming task. This along with the usability and scalability issues are the bottleneck in wide availability of such tools. The developer spends most of his/her time in tailoring the user interface (UI) to meet multiple constraints like pedagogy, teacher satisfaction and student learning styles. Providing an efficient way to develop UI can greatly speed up the development cycle and facilitate widespread access to high quality resources. In this paper, we describe how UI development can be separated from the simulation program to facilitate easy development of such visualization tool. We also discuss design and development of Graphical interface construction Kit (GicK) which helps in creating UI and combining it with simulation program for creating interactive simulations. A constructive physics simulation is developed as an exemplar of proposed framework. In the end we reported evaluation of constructive physics simulation by 43 higher school students and our conclusion.

Abstract

Many fields of Artificial Intelligence including natural language processing, speech processing, robotics etc., have seen explosive growth in the last decade. In this paper, we discuss the application of these advances to aid children with autism and dyslexia. This paper describes two systems we are currently developing - an alternative and augmentative communication (AAC) system for children with autism and a speech modification system for children with dyslexia. We also present results of our initial evaluation.

Abstract

Generation of upsampled tomographic images via combination of rotated lattices has been explored in [1]. In this paper, we evaluate the existing method using real phantom data. Up-sampled tomographic images are generated via combination of rotated hexagonal lattices. Sinogram data is filtered and back-projected on two hexagonal lattices which are rotated versions of each other. Samples from these lattices are interpolated to generate an up-sampled image defined on a square lattice. These results are compared with direct up-sampling method and image ISR-2 algorithm described in [10]. Two PET phantoms - NEMA and Hoffman brain phantom are used for purpose of evaluation. The results of the proposed method show considerable improvement over direct up-sampled image in terms of contrast, sharpness and imaging artifact; but when compared with ISR-2 generated image, the difference in image quality is not significant. A key advantage of the proposed method is that only two images are required for generating a high resolution image whereas ISR2 requires k low resolution images for an up-sampling factor of k.