<p>Two novel contributions to Content Based Image Retrieval are presented and discussed. The first is a search engine for font recognition. The intended usage is the search in very large font databases. The input to the search engine is an image of a text line, and the output is the name of the font used when printing the text. After pre-processing and segmentation of the input image, a local approach is used, where features are calculated for individual characters. The method is based on eigenimages calculated from edge filtered character images, which enables compact feature vectors that can be computed rapidly. A system for visualizing the entire font database is also proposed. Applying geometry preserving linear- and non-linear manifold learning methods, the structure of the high-dimensional feature space is mapped to a two-dimensional representation, which can be reorganized into a grid-based display. The performance of the search engine and the visualization tool is illustrated with a large database containing more than 2700 fonts.</p><p>The second contribution is the inclusion of color-based emotion-related properties in image retrieval. The color emotion metric used is derived from psychophysical experiments and uses three scales: <em>activity</em>, <em>weight </em>and <em>heat</em>. It was originally designed for single-color combinations and later extended to include pairs of colors. A modified approach for statistical analysis of color emotions in images, involving transformations of ordinary RGB-histograms, is used for image classification and retrieval. The methods are very fast in feature extraction, and descriptor vectors are very short. This is essential in our application where the intended use is the search in huge image databases containing millions or billions of images. The proposed method is evaluated in psychophysical experiments, using both category scaling and interval scaling. The results show that people in general perceive color emotions for multi-colored images in similar ways, and that observer judgments correlate with derived values.</p><p>Both the font search engine and the emotion based retrieval system are implemented in publicly available search engines. User statistics gathered during a period of 20 respectively 14 months are presented and discussed.</p>

<p>Image segmentation is the problem of partitioning an image into meaningful parts, often consisting of an object and background. As an important part of many imaging applications, e.g. face recognition, tracking of moving cars and people etc, it is of general interest to design robust and fast segmentation algorithms. However, it is well accepted that there is no general method for solving all segmentation problems. Instead, the algorithms have to be highly adapted to the application in order to achieve good performance. In this thesis, we will study segmentation methods for blood vessels in medical images. The need for accurate segmentation tools in medical applications is driven by the increased capacity of the imaging devices. Common modalities such as CT and MRI generate images which simply cannot be examined manually, due to high resolutions and a large number of image slices. Furthermore, it is very difficult to visualize complex structures in three-dimensional image volumes without cutting away large portions of, perhaps important, data. Tools, such as segmentation, can aid the medical staff in browsing through such large images by highlighting objects of particular importance. In addition, segmentation in particular can output models of organs, tumors, and other structures for further analysis, quantification or simulation.</p><p>We have divided the segmentation of blood vessels into two parts. First, we model the vessels as a collection of lines and edges (linear structures) and use filtering techniques to detect such structures in an image. Second, the output from this filtering is used as input for segmentation tools. Our contributions mainly lie in the design of a multi-scale filtering and integration scheme for de- tecting vessels of varying widths and the modification of optimization schemes for finding better segmentations than traditional methods do. We validate our ideas on synthetical images mimicking typical blood vessel structures, and show proof-of-concept results on real medical images.</p>

There are many imaging techniques that generate three-dimensional volume images today. With higher precision in the image acquisition equipment, storing and processing these images require increasing amount of data processing capacity. Traditionally, three-dimensional images are represented by cubic (or cuboid) picture elements on a cubic grid. The two-dimensional hexagonal grid has some advantages over the traditionally used square grid. For example, less samples are needed to get the same reconstruction quality, it is less rotational dependent, and each picture element has only one type of neighbor which simplifies many algorithms. The corresponding three-dimensional grids are the face-centered cubic (fcc) grid and the body-centered cubic (bcc) grids. In this thesis, image representations using non-standard grids is examined. The focus is on the fcc and bcc grids and tools for processing images on these grids, but distance functions and related algorithms (distance transforms and various representations of objects) are defined in a general framework allowing any point-lattice in any dimension. Formulas for point-to-point distance and conditions for metricity are given in the general case and parameter optimization is presented for the fcc and bcc grids. Some image acquisition and visualization techniques for the fcc and bcc grids are also presented. More theoretical results define distance functions for grids of arbitrary dimensions. Less samples are needed to represent images on non-standard grids. Thus, the huge amount of data generated by for example computerized tomography can be reduced by representating the images on non-standard grids such as the fcc or bcc grids. The thesis gives a tool-box that can be used to acquire, process, and visualize images on high-dimensional, non-standard grids.

This dissertation presents results of a medical image analysis project leading towards development of a comprehensive set of methods and tools for computational image analysis of dynamic contrast-enhanced (DCE) breast magnetic resonance image (MRI), with the aim to aid the physician in interpreting DCE breast MRI examinations. Toward this goal, we developed image analysis methods that would be needed in a breast MRI computer aided diagnosis (CADx) system. A novel contribution of this dissertation is the performance evaluation for each of the major algorithm components developed in this dissertation project. This dissertation begins with reviewing breast imaging techniques, including routinely used modalities in current clinical practice and emerging techniques still in development. We discuss at length the principles of DCE breast MRI, a very sensitive breast imaging modality that has been increasingly used in clinical practice. Then we review the diagnostic guidelines for interpreting DCE breast MRI, and explain the needs and challenges that arise in developing computational image analysis system for breast MRI applications. In this dissertation project, both the morphological and kinetic features of the lesion are automatically extracted for diagnostic purpose. In order to extract morphological features from the segmented lesions, the lesion needs to be accurately segmented out from its surrounding tissues. We utilized a probabilistic method to obtain an optimal segmentation map based on several algorithmic segmentation outputs. In evaluating the performance of segmentation algorithms, we compared the algorithmic segmentation results against manually segmented lesions, and further assessed the segmentation impact on subsequent classification stage. In order to extract accurate kinetic information, the motion needs to be compensated across image volumes acquired sequentially. In this dissertation, we comparatively assessed the similarity metric in registering DCE breast MR images. The performance of cross correlation(CC) coefficient, and mutual information (MI) were studied in both rigid and non-rigid registration schemes. Numerical results and statistical properties were reported. The resultant image quality after registration is discussed both qualitatively and quantitatively. In this dissertation we implemented a classification system based upon quantitative morphological and kinetic features in improving the specificity of breast MRI. Morphological and kinetic features of the lesion were extracted automatically, and then the feature selection step was utilized to select the most relevant features to maximize the classifier performance. In our study, the area under the receiver operating curve (AUC) is used as the performance metric of the classifier, and our results are competitive with those of previous studies. The dissertation concludes by summarizing the contribution of this project and suggesting the future directions of quantitative and highly automated approaches to breast MR image analysis. / text

Changes in retinal structure are related to different eye diseases. Various retinal imaging techniques, such as fundus imaging and optical coherence tomography (OCT) imaging modalities, have been developed for non-intrusive ophthalmology diagnoses according to the vasculature changes. However, it is time consuming or even impossible for ophthalmologists to manually label all the retinal structures from fundus images and OCT images. Therefore, computer aided diagnosis system for retinal imaging plays an important role in the assessment of ophthalmologic diseases and cardiovascular disorders. The aim of this PhD thesis is to develop segmentation methods to extract clinically useful information from these retinal images, which are acquired from different imaging modalities. In other words, we built the segmentation methods to extract important structures from both 2D fundus images and 3D OCT images. In the first part of my PhD project, two novel level set based methods were proposed for detecting the blood vessels and optic discs from fundus images. The first one integrates Chan-Vese's energy minimizing active contour method with the edge constraint term and Gaussian Mixture Model based term for blood vessels segmentation, while the second method combines the edge constraint term, the distance regularisation term and the shape-prior term for locating the optic disc. Both methods include the pre-processing stage, used for removing noise and enhancing the contrast between the object and the background. Three automated layer segmentation methods were built for segmenting intra-retinal layers from 3D OCT macular and optic nerve head images in the second part of my PhD project. The first two methods combine different methods according to the data characteristics. First, eight boundaries of the intra-retinal layers were detected from the 3D OCT macular images and the thickness maps of the seven layers were produced. Second, four boundaries of the intra-retinal layers were located from 3D optic nerve head images and the thickness maps of the Retinal Nerve Fiber Layer (RNFL) were plotted. Finally, the choroidal layer segmentation method based on the Level Set framework was designed, which embedded with the distance regularisation term, edge constraint term and Markov Random Field modelled region term. The thickness map of the choroidal layer was calculated and shown.

Texture can be broadly stated as spatial variation of image intensities. Texture analysis and classification is a well researched area for its importance to many computer vision applications. Consequently, much research has focussed on deriving powerful and efficient texture descriptors. Local binary patterns (LBP) and its variants are simple yet powerful texture descriptors. LBP features describe the texture neighbourhood of a pixel using simple comparison operators, and are often calculated based on varying neighbourhood radii to provide multi-resolution texture descriptions. A comprehensive evaluation of different LBP variants on a common benchmark dataset is missing in the literature. This thesis presents the performance for different LBP variants on texture classification and retrieval tasks. The results show that multi-scale local binary pattern variance (LBPV) gives the best performance over eight benchmarked datasets. Furthermore, improvements to the Dominant LBP (D-LBP) by ranking dominant patterns over complete training set and Compound LBP (CM-LBP) by considering 16 bits binary codes are suggested which are shown to outperform their original counterparts. The main contribution of the thesis is the introduction of multi-dimensional LBP features, which preserve the relationships between different scales by building a multi-dimensional histogram. The results on benchmarked classification and retrieval datasets clearly show that the multi-dimensional LBP (MD-LBP) improves the results compared to conventional multi-scale LBP. The same principle is applied to LBPV (MD-LBPV), again leading to improved performance. The proposed variants result in relatively large feature lengths which is addressed using three different feature length reduction techniques. Principle component analysis (PCA) is shown to give the best performance when the feature length is reduced to match that of conventional multi-scale LBP. The proposed multi-dimensional LBP variants are applied for medical image analysis application. The first application is nailfold capillary (NC) image classification. Performance of MD-LBPV on NC images is highest, whereas for second application, HEp-2 cell classification, performance of MD-LBP is highest. It is observed that the proposed texture descriptors gives improved texture classification accuracy.

Borehole imaging is one of the fastest and most precise methods for collecting subsurface data that provides high resolution information on layering, texture and dips, permitting a core-like description of the subsurface. Although the range of information recoverable from this technology is widely acknowledged, image logs are still used in a strictly qualitative manner. Interpreting image logs manually is cumbersome, time consuming and is subjective based on the experience of the interpreter. This thesis outlines new methods that automate image log interpretation and extract subsurface lithofacies information in a quantitative manner. We developed two methodologies based on advanced image analysis techniques successfully employed in remote sensing and medical imaging. The first one is a pixelbased pattern recognition technique applying textural analysis to quantify image textural properties. These properties together with standard logs and core-derived lithofacies information are used to train a back propagation Neural Network. In principle the trained and tested Neural Network is applicable for automated borehole image interpretation from similar geological settings. However, this pixel-based approach fails to make use explicitly of the spatial characteristics of a high resolution image. TAT second methodology is introduced which groups identical neighbouring pixels into objects. The resultant spectrally and spatially consistent objects are then related to geologically meaningful groups such as lithofacies by employing fuzzy classifiers. This method showed better results and is applied to outcrop photos, core photos and image logs, including a ‘difficult’ data set from a deviated well. The latter image log did not distinguish some of the conductive and resistive regions, as observed from standard logs and core photos. This is overcome by marking bed boundaries using standard logs. Bed orientations were estimated using an automated sinusoid fitting algorithm within a formal uncertainty framework in order to distinguish dipping beds and horizontal stratification. Integration of these derived logs in the methodology yields a complete automated lithofacies identification, even from the difficult dataset. The results were validated through the interpretation of cored intervals by a geologist. This is a supervised classification method which incorporates the expertise of one or several geologists, and hence includes human logic, reasoning, and current knowledge of the field heterogeneity. By including multiple geologists in the training, the results become less dependent on each individual’s subjectivity and prior experience. The method is also easily adaptable to other geological settings. In addition, it is applicable to several kinds of borehole images, for example wireline electrical borehole wall images, core photographs, and logging-while-drilling (LWD) images. Thus, the theme of this dissertation is the development of methodologies which makes image log interpretation simpler, faster, less subjective, and efficient such that it can be applied to large quantities of data.

Quantitative microscopy deals with the extraction of quantitative measurements from samples observed under a microscope. Recent developments in microscopy systems, sample preparation and handling techniques have enabled high throughput biological experiments resulting in large amounts of image data, at biological scales ranging from subcellular structures such as fluorescently tagged nucleic acid sequences to whole organisms such as zebrafish embryos. Consequently, methods and algorithms for automated quantitative analysis of these images have become increasingly important. These methods range from traditional image analysis techniques to use of deep learning architectures. Many biomedical microscopy assays result in fluorescent spots. Robust detection and precise localization of these spots are two important, albeit sometimes overlapping, areas for application of quantitative image analysis. We demonstrate the use of popular deep learning architectures for spot detection and compare them against more traditional parametric model-based approaches. Moreover, we quantify the effect of pre-training and change in the size of training sets on detection performance. Thereafter, we determine the potential of training deep networks on synthetic and semi-synthetic datasets and their comparison with networks trained on manually annotated real data. In addition, we present a two-alternative forced-choice based tool for assisting in manual annotation of real image data. On a spot localization track, we parallelize a popular compressed sensing based localization method and evaluate its performance in conjunction with different optimizers, noise conditions and spot densities. We investigate its sensitivity to different point spread function estimates. Zebrafish is an important model organism, attractive for whole-organism image-based assays for drug discovery campaigns. The effect of drug-induced neuronal damage may be expressed in the form of zebrafish shape deformation. First, we present an automated method for accurate quantification of tail deformations in multi-fish micro-plate wells using image analysis techniques such as illumination correction, segmentation, generation of branch-free skeletons of partial tail-segments and their fusion to generate complete tails. Later, we demonstrate the use of a deep learning-based pipeline for classifying micro-plate wells as either drug-affected or negative controls, resulting in competitive performance, and compare the performance from deep learning against that from traditional image analysis approaches.

Objectives: The overall aim of this thesis was to investigate the precision error, annual changes, and monitoring time intervals of muscle and fat outcomes measured by peripheral quantitative computed tomography (pQCT), as well as explore the strength of their associations with fall status in older adults. Methods: Participants aged >60 years old (N=190) were recruited from the Saskatoon Cohort of the Canadian Multicentre Osteoporosis Study (CaMOs). The precision error (Root Mean Squared Co-efficient of Variation, CV%RMS) of soft-tissue outcomes from previously reported pQCT image analysis protocols (n=6) were calculated and compared using repeat forearm and lower leg scans collected from a random sub-sample of women (n=35). Prospective scans were collected with 1 and/or 2 years of follow-up (n=97) to estimate annual changes and monitoring time intervals for pQCT-derived muscle and fat outcomes in women. Imaging data and responses from a retrospective fall status questionnaire were analyzed to investigate the associations of muscle density, functional mobility, and health- related factors to fall status for both men and women (n=183). Results: Precision errors of muscle and fat outcomes ranged from 0.7 to 6.4% in older women, however not all protocols were equally precise. Muscle cross-sectional area decreased by 0.8 to 1.2% per year, with greater losses in the lower limb. Biological changes in muscle area and density may be detected with 80 and 95% certainty within monitoring time intervals of 4 to 9 years. The odds of having reported a fall increased by 17% for every unit decrease in muscle density (mean 70.2, SD 2.6mg/cm3) after adjusting for age, sex, body mass index, general health status, diabetes, the number of comorbidities, and functional mobility. Discussion: This dissertation demonstrated the potential for pQCT to study changes in muscle and fat outcomes in older adults. Both muscle area and density can be precisely measured. Observed annual changes in soft-tissue outcomes were small in older adults; highlighting the importance of precise measurements to detect changes beyond measurement error. Together with the estimated monitoring time intervals, these findings can assist the planning of prospective investigations of musculoskeletal health in aging. Furthermore, based on the observed independent association between muscle density and fall status, monitoring muscle density may further complement the study of musculoskeletal health and fall risk in community-dwelling older adults.

<p>The pictorial language of the Swedish children is characterized by the idea that a “good” drawing should be in the right perspective and as photographically realistic as possible. This is a study about the pictorial language of the children in the Gambia and Senegal. Is the pictorial language different with the children living in a culture that has a stronger tradition of spoken word and visual communication than the children living in the western civilisation?</p><p>With the help of different theories concerning children’s creating of art, this study is trying to sort out the differences. It is also explaining about different theories when it comes to development stages in the children’s drawings and how the culture, tradition and conventions influence both the pictorial grammar and the ideal image.</p><p>The study is based on drawings collected in schools in The Gambia and Senegal and the drawings are analysed with the help of theories in Karin Aronssons “Barns världar – barns bilder”.</p><p>The study is also based on observations and interviews with children and teachers in a school in the Gambia.</p>