Majid Mirmehdi

Segmentation of color textures

This paper describes an approach to perceptual segmentation of color image textures. A multiscale representation of the texture image, generated by a multiband smoothing algorithm based on human psychophysical measurements of color appearance is used as the input. Initial segmentation is achieved by applying a clustering algorithm to the image at the coarsest level of smoothing. The segmented clusters are then restructured in order to isolate core clusters, i.e., patches in which the pixels are definitely associated with the same region. The image pixels representing the core clusters are used to form 3D color histograms which are then used for probabilistic assignment of all other pixels to the core clusters to form larger clusters and categorise the rest of the image. The process of setting up color histograms and probabilistic reassignment of the pixels to the clusters is then propagated through finer levels of smoothing until a full segmentation is achieved at the highest level of resolution.

Automated identification of diabetic retinal exudates in digital colour images

Aim: To identify retinal exudates automatically from colour retinal images. Methods: The colour retinal images were segmented using fuzzy C-means clustering following some key preprocessing steps. To classify the segmented regions into exudates and non-exudates, an artificial neural network classifier was investigated. Results: The proposed system can achieve a diagnostic accuracy with 95.0% sensitivity and 88.9% specificity for the identification of images containing any evidence of retinopathy, where the trade off between sensitivity and specificity was appropriately balanced for this particular problem. Furthermore, it demonstrates 93.0% sensitivity and 94.1% specificity in terms of exudate based classification. Conclusions: This study indicates that automated evaluation of digital retinal images could be used to screen for exudative diabetic retinopathy.

Real-Time Detection and Recognition of Road Traffic Signs

This paper proposes a novel system for the automatic detection and recognition of traffic signs. The proposed system detects candidate regions as maximally stable extremal regions (MSERs), which offers robustness to variations in lighting conditions. Recognition is based on a cascade of support vector machine (SVM) classifiers that were trained using histogram of oriented gradient (HOG) features. The training data are generated from synthetic template images that are freely available from an online database; thus, real footage road signs are not required as training data. The proposed system is accurate at high vehicle speeds, operates under a range of weather conditions, runs at an average speed of 20 frames per second, and recognizes all classes of ideogram-based (nontext) traffic symbols from an online road sign database. Comprehensive comparative results to illustrate the performance of the system are presented.

MAC: Magnetostatic Active Contour Model

We propose an active contour model using an external force field that is based on magnetostatics and hypothesized magnetic interactions between the active contour and object boundaries. The major contribution of the method is that the interaction of its forces can greatly improve the active contour in capturing complex geometries and dealing with difficult initializations, weak edges, and broken boundaries. The proposed method is shown to achieve significant improvements when compared against six well-known and state-of-the-art shape recovery methods, including the geodesic snake, the generalized version of gradient vector flow (GVF) snake, the combined geodesic and GVF snake, and the charged particle model.

Comparison of colour spaces for optic disc localisation in retinal images

The location of the optic disc is of critical importance in retinal image analysis. In this work we improve on an approach introduced by Mendels, Heneghan and Thiran (1999) which localises an optic disc region through grey level morphology, followed by snake fitting. We propose and implement both the automatic initialisation of the snake and the application of morphology in colour space. We examine various methods of performing the morphology step (to remove the interference of blood vessels) and compare them against each other. We demonstrate that our proposed simple Lab colour morphology method is particularly suitable for the characteristics of our optic disc images. Results indicate 90.32% average accuracy in localising, the optic disc boundary.

RAGS: region-aided geometric snake

An enhanced, region-aided, geometric active contour that is more tolerant toward weak edges and noise in images is introduced. The proposed method integrates gradient flow forces with region constraints, composed of image region vector flow forces obtained through the diffusion of the region segmentation map. We refer to this as the region-aided geometric snake or RAGS. The diffused region forces can be generated from any reliable region segmentation technique, greylevel or color. This extra region force gives the snake a global complementary view of the boundary information within the image which, along with the local gradient flow, helps detect fuzzy boundaries and overcome noisy regions. The partial differential equation (PDE) resulting from this integration of image gradient flow and diffused region flow is implemented using a level set approach. We present various examples and also evaluate and compare the performance of RAGS on weak boundaries and noisy images.

Classification and Localisation of Diabetic-Related Eye Disease

Retinal exudates are a characteristic feature of many retinal diseases such as Diabetic Retinopathy. We address the development of a method to quantitatively diagnose these random yellow patches in colour retinal images automatically. After a colour normalisation and contrast enhancement preprocessing step, the colour retinal image is segmented using Fuzzy C-Means clustering. We then classify the segmented regions into two disjoint classes, exudates and non-exudates, comparing the performance of various classifiers. We also locate the optic disk both to remove it as a candidate region and to measure its boundaries accurately since it is a significant landmark feature for ophthalmologists. Three different approaches are reported for optic disk localisation based on template matching, least squares arc estimation and snakes. The system could achieve an overall diagnostic accuracy of 90.1% for identification of the exudate pathologies and 90.7% for optic disk localisation.

Comparative Exudate Classification Using Support Vector Machines and Neural Networks

After segmenting candidate exudates regions in colour retinal images we present and compare two methods for their classification. The Neural Network based approach performs marginally better than the Support Vector Machine based approach, but we show that the latter are more flexible given criteria such as control of sensitivity and specificity rates. We present classification results for different learning algorithms for the Neural Net and use both hard and soft margins for the Support Vector Machines. We also present ROC curves to examine the trade-off between the sensitivity and specificity of the classifiers.

Recognising text in real scenes

Abstract. We present two different approaches to the location and recovery of text in images of real scenes. The techniques we describe are invariant to the scale and 3D orientation of the text, and allow recovery of text in cluttered scenes. The first approach uses page edges and other rectangular boundaries around text to locate a surface containing text, and to recover a fronto-parallel view. This is performed using line detection, perceptual grouping, and comparison of potential text regions using a confidence measure. The second approach uses low-level texture measures with a neural network classifier to locate regions of text in an image. Then we recover a fronto-parallel view of each located paragraph of text by separating the individual lines of text and determining the vanishing points of the text plane. We illustrate our results using a number of images.

Detection and Tracking of Very Small Low Contrast Objects

We present a Kalman tracking algorithm that can track a number of very small, low contrast objects through an image sequence taken from a static camera. The issues that we have addressed to achieve this are twofold. Firstly, the detection of small objects comprising a few pixels only, moving slowly in the image, and secondly, tracking of multiple small targets even though they may be lost either through occlusion or in noisy signal. The approach uses a combination of wavelet filtering for detection with an interest operator for testing multiple target hypotheses based within the framework of a Kalman tracker. We demonstrate the robustness of the approach to occlusion and for multiple targets.