Clinical monitoring of ocular physiology using digital image analysis

Aim: To examine the use of image analysis to quantify changes in ocular physiology. Method: A purpose designed computer program was written to objectively quantify bulbar hyperaemia, tarsal redness, corneal staining and tarsal staining. Thresholding, colour extraction and edge detection paradigms were investigated. The repeatability (stability) of each technique to changes in image luminance was assessed. A clinical pictorial grading scale was analysed to examine the repeatability and validity of the chosen image analysis technique. Results: Edge detection using a 3 × 3 kernel was found to be the most stable to changes in image luminance (2.6% over a +60 to -90% luminance range) and correlated well with the CCLRU scale images of bulbar hyperaemia (r = 0.96), corneal staining (r = 0.85) and the staining of palpebral roughness (r = 0.96). Extraction of the red colour plane demonstrated the best correlation-sensitivity combination for palpebral hyperaemia (r = 0.96). Repeatability variability was <0.5%. Conclusions: Digital imaging, in conjunction with computerised image analysis, allows objective, clinically valid and repeatable quantification of ocular features. It offers the possibility of improved diagnosis and monitoring of changes in ocular physiology in clinical practice. © 2003 British Contact Lens Association. Published by Elsevier Science Ltd. All rights reserved.

Methods of Color Images Procession for Further Identification of the Object

The offered paper deals with the problems of color images preliminary procession. Among these are: interference control (local ones and noise) and extraction of the object from the background on the stage preceding the process of contours extraction. It was considered for a long time that execution of smoothing in segmentation through the boundary extraction is inadmissible, but the described methods and the obtained results evidence about expedience of using the noise control methods.

Spatial ecological complexity measures in GRASS GIS

Good estimates of ecosystem complexity are essential for a number of ecological tasks: from biodiversity estimation, to forest structure variable retrieval, to feature extraction by edge detection and generation of multifractal surface as neutral models for e.g. feature change assessment. Hence, measuring ecological complexity over space becomes crucial in macroecology and geography. Many geospatial tools have been advocated in spatial ecology to estimate ecosystem complexity and its changes over space and time. Among these tools, free and open source options especially offer opportunities to guarantee the robustness of algorithms and reproducibility. In this paper we will summarize the most straightforward measures of spatial complexity available in the Free and Open Source Software GRASS GIS, relating them to key ecological patterns and processes.

Extended principal component analysis and its applications to image analysis

The objectives of this research are to analyze and develop a modified Principal Component Analysis (PCA) and to develop a two-dimensional PCA with applications in image processing. PCA is a classical multivariate technique where its mathematical treatment is purely based on the eigensystem of positive-definite symmetric matrices. Its main function is to statistically transform a set of correlated variables to a new set of uncorrelated variables over $\IR\sp{n}$ by retaining most of the variations present in the original variables.^ The variances of the Principal Components (PCs) obtained from the modified PCA form a correlation matrix of the original variables. The decomposition of this correlation matrix into a diagonal matrix produces a set of orthonormal basis that can be used to linearly transform the given PCs. It is this linear transformation that reproduces the original variables. The two-dimensional PCA can be devised as a two successive of one-dimensional PCA. It can be shown that, for an $m\times n$ matrix, the PCs obtained from the two-dimensional PCA are the singular values of that matrix.^ In this research, several applications for image analysis based on PCA are developed, i.e., edge detection, feature extraction, and multi-resolution PCA decomposition and reconstruction. ^

Algoritmi di segmentazione di immagini mediche

Nell'ambito dell'elaborazione delle immagini, si definisce segmentazione il processo atto a scomporre un'immagine nelle sue regioni costituenti o negli oggetti che la compongono. Ciò avviene sulla base di determinati criteri di appartenenza dei pixel ad una regione. Si tratta di uno degli obiettivi più difficili da perseguire, anche perché l'accuratezza del risultato dipende dal tipo di informazione che si vuole ricavare dall'immagine. Questa tesi analizza, sperimenta e raffronta alcune tecniche di elaborazione e segmentazione applicate ad immagini digitali di tipo medico. In particolare l'obiettivo di questo studio è stato quello di proporre dei possibili miglioramenti alle tecniche di segmentazione comunemente utilizzate in questo ambito, all'interno di uno specifico set di immagini: tomografie assiali computerizzate (TAC) frontali e laterali aventi per soggetto ginocchia, con ivi impiantate protesi superiore e inferiore. L’analisi sperimentale ha portato allo sviluppo di due algoritmi in grado di estrarre correttamente i contorni delle sole protesi senza rilevare falsi punti di edge, chiudere eventuali gap, il tutto a un basso costo computazionale.

Constructive Synthesis of Memory-Intensive Accelerators for FPGA From Nested Loop Kernels

Field-programmable gate arrays are ideal hosts to custom accelerators for signal, image, and data processing but de- mand manual register transfer level design if high performance and low cost are desired. High-level synthesis reduces this design burden but requires manual design of complex on-chip and off-chip memory architectures, a major limitation in applications such as video processing. This paper presents an approach to resolve this shortcoming. A constructive process is described that can derive such accelerators, including on- and off-chip memory storage from a C description such that a user-defined throughput constraint is met. By employing a novel statement-oriented approach, dataflow intermediate models are derived and used to support simple ap- proaches for on-/off-chip buffer partitioning, derivation of custom on-chip memory hierarchies and architecture transformation to ensure user-defined throughput constraints are met with minimum cost. When applied to accelerators for full search motion estima- tion, matrix multiplication, Sobel edge detection, and fast Fourier transform, it is shown how real-time performance up to an order of magnitude in advance of existing commercial HLS tools is enabled whilst including all requisite memory infrastructure. Further, op- timizations are presented that reduce the on-chip buffer capacity and physical resource cost by up to 96% and 75%, respectively, whilst maintaining real-time performance.

Shock Train/Boundary-Layer Interaction in Rectangular Scramjet Isolators

Numerous studies of the dual-mode scramjet isolator, a critical component in preventing inlet unstart and/or vehicle loss by containing a collection of flow disturbances called a shock train, have been performed since the dual-mode propulsion cycle was introduced in the 1960s. Low momentum corner flow and other three-dimensional effects inherent to rectangular isolators have, however, been largely ignored in experimental studies of the boundary layer separation driven isolator shock train dynamics. Furthermore, the use of two dimensional diagnostic techniques in past works, be it single-perspective line-of-sight schlieren/shadowgraphy or single axis wall pressure measurements, have been unable to resolve the three-dimensional flow features inside the rectangular isolator. These flow characteristics need to be thoroughly understood if robust dual-mode scramjet designs are to be fielded. The work presented in this thesis is focused on experimentally analyzing shock train/boundary layer interactions from multiple perspectives in aspect ratio 1.0, 3.0, and 6.0 rectangular isolators with inflow Mach numbers ranging from 2.4 to 2.7. Secondary steady-state Computational Fluid Dynamics studies are performed to compare to the experimental results and to provide additional perspectives of the flow field. Specific issues that remain unresolved after decades of isolator shock train studies that are addressed in this work include the three-dimensional formation of the isolator shock train front, the spatial and temporal low momentum corner flow separation scales, the transient behavior of shock train/boundary layer interaction at specific coordinates along the isolator's lateral axis, and effects of the rectangular geometry on semi-empirical relations for shock train length prediction. A novel multiplane shadowgraph technique is developed to resolve the structure of the shock train along both the minor and major duct axis simultaneously. It is shown that the shock train front is of a hybrid oblique/normal nature. Initial low momentum corner flow separation spawns the formation of oblique shock planes which interact and proceed toward the center flow region, becoming more normal in the process. The hybrid structure becomes more two-dimensional as aspect ratio is increased but corner flow separation precedes center flow separation on the order of 1 duct height for all aspect ratios considered. Additional instantaneous oil flow surface visualization shows the symmetry of the three-dimensional shock train front around the lower wall centerline. Quantitative synthetic schlieren visualization shows the density gradient magnitude approximately double between the corner oblique and center flow normal structures. Fast response pressure measurements acquired near the corner region of the duct show preliminary separation in the outer regions preceding centerline separation on the order of 2 seconds. Non-intrusive Focusing Schlieren Deflectometry Velocimeter measurements reveal that both shock train oscillation frequency and velocity component decrease as measurements are taken away from centerline and towards the side-wall region, along with confirming the more two dimensional shock train front approximation for higher aspect ratios. An updated modification to Waltrup \& Billig's original semi-empirical shock train length relation for circular ducts based on centerline pressure measurements is introduced to account for rectangular isolator aspect ratio, upstream corner separation length scale, and major- and minor-axis boundary layer momentum thickness asymmetry. The latter is derived both experimentally and computationally and it is shown that the major-axis (side-wall) boundary layer has lower momentum thickness compared to the minor-axis (nozzle bounded) boundary layer, making it more separable. Furthermore, it is shown that the updated correlation drastically improves shock train length prediction capabilities in higher aspect ratio isolators. This thesis suggests that performance analysis of rectangular confined supersonic flow fields can no longer be based on observations and measurements obtained along a single axis alone. Knowledge gained by the work performed in this study will allow for the development of more robust shock train leading edge detection techniques and isolator designs which can greatly mitigate the risk of inlet unstart and/or vehicle loss in flight.

Automatic dynamic mask extraction for PIV images containing an unsteady interface, bubbles, and a moving structure

When performing Particle Image Velocimetry (PIV) measurements in complex fluid flows with moving interfaces and a two-phase flow, it is necessary to develop a mask to remove non-physical measurements. This is the case when studying, for example, the complex bubble sweep-down phenomenon observed in oceanographic research vessels. Indeed, in such a configuration, the presence of an unsteady free surface, of a solid–liquid interface and of bubbles in the PIV frame, leads to generate numerous laser reflections and therefore spurious velocity vectors. In this note, an image masking process is developed to successively identify the boundaries of the ship and the free surface interface. As the presence of the solid hull surface induces laser reflections, the hull edge contours are simply detected in the first PIV frame and dynamically estimated for consecutive ones. As for the unsteady surface determination, a specific process is implemented like the following: i) the edge detection of the gradient magnitude in the PIV frame, ii) the extraction of the particles by filtering high-intensity large areas related to the bubbles and/or hull reflections, iii) the extraction of the rough region containing these particles and their reflections, iv) the removal of these reflections. The unsteady surface is finally obtained with a fifth-order polynomial interpolation. The resulted free surface is successfully validated from the Fourier analysis and by visualizing selected PIV images containing numerous spurious high intensity areas. This paper demonstrates how this data analysis process leads to PIV images database without reflections and an automatic detection of both the free surface and the rigid body. An application of this new mask is finally detailed, allowing a preliminary analysis of the hydrodynamic flow.

Fast Retinal Vessel Detection and Measurement Using Wavelets and Edge Location Refinement

The relationship between changes in retinal vessel morphology and the onset and progression of diseases such as diabetes, hypertension and retinopathy of prematurity (ROP) has been the subject of several large scale clinical studies. However, the difficulty of quantifying changes in retinal vessels in a sufficiently fast, accurate and repeatable manner has restricted the application of the insights gleaned from these studies to clinical practice. This paper presents a novel algorithm for the efficient detection and measurement of retinal vessels, which is general enough that it can be applied to both low and high resolution fundus photographs and fluorescein angiograms upon the adjustment of only a few intuitive parameters. Firstly, we describe the simple vessel segmentation strategy, formulated in the language of wavelets, that is used for fast vessel detection. When validated using a publicly available database of retinal images, this segmentation achieves a true positive rate of 70.27%, false positive rate of 2.83%, and accuracy score of 0.9371. Vessel edges are then more precisely localised using image profiles computed perpendicularly across a spline fit of each detected vessel centreline, so that both local and global changes in vessel diameter can be readily quantified. Using a second image database, we show that the diameters output by our algorithm display good agreement with the manual measurements made by three independent observers. We conclude that the improved speed and generality offered by our algorithm are achieved without sacrificing accuracy. The algorithm is implemented in MATLAB along with a graphical user interface, and we have made the source code freely available. 

Automated detection and measurement of isolated retinal arterioles by a combination of edge enhancement and cost analysis

Pressure myography studies have played a crucial role in our understanding of vascular physiology and pathophysiology. Such studies depend upon the reliable measurement of changes in the diameter of isolated vessel segments over time. Although several software packages are available to carry out such measurements on small arteries and veins, no such software exists to study smaller vessels (<50 µm in diameter). We provide here a new, freely available open-source algorithm, MyoTracker, to measure and track changes in the diameter of small isolated retinal arterioles. The program has been developed as an ImageJ plug-in and uses a combination of cost analysis and edge enhancement to detect the vessel walls. In tests performed on a dataset of 102 images, automatic measurements were found to be comparable to those of manual ones. The program was also able to track both fast and slow constrictions and dilations during intraluminal pressure changes and following application of several drugs. Variability in automated measurements during analysis of videos and processing times were also investigated and are reported. MyoTracker is a new software to assist during pressure myography experiments on small isolated retinal arterioles. It provides fast and accurate measurements with low levels of noise and works with both individual images and videos. Although the program was developed to work with small arterioles, it is also capable of tracking the walls of other types of microvessels, including venules and capillaries. It also works well with larger arteries, and therefore may provide an alternative to other packages developed for larger vessels when its features are considered advantageous.

Towards automatic power line detection for a UAV surveillance system using pulse coupled neural filter and an improved Hough transform

Spatial information captured from optical remote sensors on board unmanned aerial vehicles (UAVs) has great potential in automatic surveillance of electrical infrastructure. For an automatic vision-based power line inspection system, detecting power lines from a cluttered background is one of the most important and challenging tasks. In this paper, a novel method is proposed, specifically for power line detection from aerial images. A pulse coupled neural filter is developed to remove background noise and generate an edge map prior to the Hough transform being employed to detect straight lines. An improved Hough transform is used by performing knowledge-based line clustering in Hough space to refine the detection results. The experiment on real image data captured from a UAV platform demonstrates that the proposed approach is effective for automatic power line detection.

Improved simultaneous computation of motion detection and optical flow for object tracking

Object tracking systems require accurate segmentation of the objects from the background for effective tracking. Motion segmentation or optical flow can be used to segment incoming images. Whilst optical flow allows multiple moving targets to be separated based on their individual velocities, optical flow techniques are prone to errors caused by changing lighting and occlusions, both common in a surveillance environment. Motion segmentation techniques are more robust to fluctuating lighting and occlusions, but don't provide information on the direction of the motion. In this paper we propose a combined motion segmentation/optical flow algorithm for use in object tracking. The proposed algorithm uses the motion segmentation results to inform the optical flow calculations and ensure that optical flow is only calculated in regions of motion, and improve the performance of the optical flow around the edge of moving objects. Optical flow is calculated at pixel resolution and tracking of flow vectors is employed to improve performance and detect discontinuities, which can indicate the location of overlaps between objects. The algorithm is evaluated by attempting to extract a moving target within the flow images, given expected horizontal and vertical movement (i.e. the algorithms intended use for object tracking). Results show that the proposed algorithm outperforms other widely used optical flow techniques for this surveillance application.

A hidden Markov model and relative entropy rate approach to vision-based dim target detection for UAV sense-and-avoid

Uninhabited aerial vehicles (UAVs) are a cutting-edge technology that is at the forefront of aviation/aerospace research and development worldwide. Many consider their current military and defence applications as just a token of their enormous potential. Unlocking and fully exploiting this potential will see UAVs in a multitude of civilian applications and routinely operating alongside piloted aircraft. The key to realising the full potential of UAVs lies in addressing a host of regulatory, public relation, and technological challenges never encountered be- fore. Aircraft collision avoidance is considered to be one of the most important issues to be addressed, given its safety critical nature. The collision avoidance problem can be roughly organised into three areas: 1) Sense; 2) Detect; and 3) Avoid. Sensing is concerned with obtaining accurate and reliable information about other aircraft in the air; detection involves identifying potential collision threats based on available information; avoidance deals with the formulation and execution of appropriate manoeuvres to maintain safe separation. This thesis tackles the detection aspect of collision avoidance, via the development of a target detection algorithm that is capable of real-time operation onboard a UAV platform. One of the key challenges of the detection problem is the need to provide early warning. This translates to detecting potential threats whilst they are still far away, when their presence is likely to be obscured and hidden by noise. Another important consideration is the choice of sensors to capture target information, which has implications for the design and practical implementation of the detection algorithm. The main contributions of the thesis are: 1) the proposal of a dim target detection algorithm combining image morphology and hidden Markov model (HMM) filtering approaches; 2) the novel use of relative entropy rate (RER) concepts for HMM filter design; 3) the characterisation of algorithm detection performance based on simulated data as well as real in-flight target image data; and 4) the demonstration of the proposed algorithm's capacity for real-time target detection. We also consider the extension of HMM filtering techniques and the application of RER concepts for target heading angle estimation. In this thesis we propose a computer-vision based detection solution, due to the commercial-off-the-shelf (COTS) availability of camera hardware and the hardware's relatively low cost, power, and size requirements. The proposed target detection algorithm adopts a two-stage processing paradigm that begins with an image enhancement pre-processing stage followed by a track-before-detect (TBD) temporal processing stage that has been shown to be effective in dim target detection. We compare the performance of two candidate morphological filters for the image pre-processing stage, and propose a multiple hidden Markov model (MHMM) filter for the TBD temporal processing stage. The role of the morphological pre-processing stage is to exploit the spatial features of potential collision threats, while the MHMM filter serves to exploit the temporal characteristics or dynamics. The problem of optimising our proposed MHMM filter has been examined in detail. Our investigation has produced a novel design process for the MHMM filter that exploits information theory and entropy related concepts. The filter design process is posed as a mini-max optimisation problem based on a joint RER cost criterion. We provide proof that this joint RER cost criterion provides a bound on the conditional mean estimate (CME) performance of our MHMM filter, and this in turn establishes a strong theoretical basis connecting our filter design process to filter performance. Through this connection we can intelligently compare and optimise candidate filter models at the design stage, rather than having to resort to time consuming Monte Carlo simulations to gauge the relative performance of candidate designs. Moreover, the underlying entropy concepts are not constrained to any particular model type. This suggests that the RER concepts established here may be generalised to provide a useful design criterion for multiple model filtering approaches outside the class of HMM filters. In this thesis we also evaluate the performance of our proposed target detection algorithm under realistic operation conditions, and give consideration to the practical deployment of the detection algorithm onboard a UAV platform. Two fixed-wing UAVs were engaged to recreate various collision-course scenarios to capture highly realistic vision (from an onboard camera perspective) of the moments leading up to a collision. Based on this collected data, our proposed detection approach was able to detect targets out to distances ranging from about 400m to 900m. These distances, (with some assumptions about closing speeds and aircraft trajectories) translate to an advanced warning ahead of impact that approaches the 12.5 second response time recommended for human pilots. Furthermore, readily available graphic processing unit (GPU) based hardware is exploited for its parallel computing capabilities to demonstrate the practical feasibility of the proposed target detection algorithm. A prototype hardware-in- the-loop system has been found to be capable of achieving data processing rates sufficient for real-time operation. There is also scope for further improvement in performance through code optimisations. Overall, our proposed image-based target detection algorithm offers UAVs a cost-effective real-time target detection capability that is a step forward in ad- dressing the collision avoidance issue that is currently one of the most significant obstacles preventing widespread civilian applications of uninhabited aircraft. We also highlight that the algorithm development process has led to the discovery of a powerful multiple HMM filtering approach and a novel RER-based multiple filter design process. The utility of our multiple HMM filtering approach and RER concepts, however, extend beyond the target detection problem. This is demonstrated by our application of HMM filters and RER concepts to a heading angle estimation problem.

Fast power line detection and localization using steerable filter for active UAV guidance

In this paper we present a fast power line detection and localisation algorithm as well as propose a high-level guidance architecture for active vision-based Unmanned Aerial Vehicle (UAV) guidance. The detection stage is based on steerable filters for edge ridge detection, followed by a line fitting algorithm to refine candidate power lines in images. The guidance architecture assumes an UAV with an onboard Gimbal camera. We first control the position of the Gimbal such that the power line is in the field of view of the camera. Then its pose is used to generate the appropriate control commands such that the aircraft moves and flies above the lines. We present initial experimental results for the detection stage which shows that the proposed algorithm outperforms two state-of-the-art line detection algorithms for power line detection from aerial imagery.