Various infant studies suggest that statistical regularities in the speech stream (e.g. transitional probabilities) are one of the first speech segmentation cues available. Statistical learning may serve as a mechanism for learning various language specific segmentation cues (e.g. stress segmentation by English speakers). To test this possibility we exposed adults to an artificial language in which all words had a novel acoustic cue on the final syllable. Subjects were presented with a continuous stream of synthesized speech in which the words were repeated in random order. Subjects were then given a new set of words to see if they had learned the acoustic cue and generalized it to new stimuli. Finally, subjects were exposed to a competition stream in which the transitional probability and novel acoustic cues conflicted to see which cue they preferred to use for segmentation. Results on the word-learning test suggest that subjects were able to segment the first exposure stream, however, on the cue transfer test they did not display any evidence of learning the relationship between word boundaries and the novel acoustic cue. Subjects were able to learn statistical words from the competition stream despite extra intervening syllables.

Magnetic resonance imaging (MRI) is a powerful imaging tool for diagnostics, which AMRA uses to segment and quantify certain anatomical regions. This thesis investigate the possibilities of using deep learning for the particular task of AMRAs segmentation, both for ordinary regions (fat and muscle regions) and injured muscles.The main approach performs muscle and fat segmentation separately, and compares results for three approaches; a full resolution approach, a down-sample approach (trained on down-sampled images) and an ensemble approach (uses voting among the 7 best networks).The results shows that deep learning segmentation is possible for the task, with satisfactory results. The down-sampled approach works best for fat segmentation, which can be related to the inconsistently over-segmented ground truth fat masks. It is therefore unnecessary with the additional resolution, which might only impair the performance. The down-sampled approach achieves better results also for muscle segmentation. Ensemble learning does in general not improve the neither the segmentation dice score nor the biomarker predictions. Injured muscles are more difficult to predict due to smaller muscles in the particular used dataset, and an increased data versatility. As a summary, deep learning shows great potential for the task. The results are overall satisfactory (mostly for a down-sampled approach), but further work needs to be done for injured muscles in order to make it clinically useful.

Indiana University-Purdue University Indianapolis (IUPUI) / The vein structure in the sclera, the white and opaque outer protective covering of the eye, is anecdotally stable over time and unique to each person. As a result, it is well suited for use as a biometric for human identification. A few researchers have performed sclera vein pattern recognition and have reported promising, but low accuracy, initial results. Sclera recognition poses several challenges: the vein structure moves and deforms with the movement of the eye and its surrounding tissues; images of sclera patterns are often defocused and/or saturated; and, most importantly, the vein structure in the sclera is multi-layered and has complex non-linear deformation. The previous approaches in sclera recognition have treated the sclera patterns as a one-layered vein structure, and, as a result, their sclera recognition accuracy is not high. In this thesis, we propose a new method for sclera recognition with the following contributions: First, we developed a color-based sclera region estimation scheme for sclera segmentation. Second, we designed a Gabor wavelet based sclera pattern enhancement method, and an adaptive thresholding method to emphasize and binarize the sclera vein patterns. Third, we proposed a line descriptor based feature extraction, registration, and matching method that is scale-, orientation-, and deformation-invariant, and can mitigate the multi-layered deformation effects and tolerate segmentation error. It is empirically verified using the UBIRIS and IUPUI multi-wavelength databases that the proposed method can perform accurate sclera recognition. In addition, the recognition results are compared to iris recognition algorithms, with very comparable results.

No description available.

No description available.

Video segmentation is an application of computer vision aimed at automating the extraction of an object from a series of video frames. However, it is a difficult problem, especially to compute at real-time, interactive rates. Although general application to video is difficult because of the wide range of image scenarios, user interaction can help to reduce the problem space and speed up the computation. This thesis presents a fast object-tracking tool that selects an object from a series of frames based on minimal user input. Our Intelligent Rotoscoping tool aims for increased speed and accuracy over other video segmentation tools, while maintaining reproducibility of results. For speed, the tool stays ahead of the user in selecting frames and responding to feedback. For accuracy, it interprets user input such that the user does not have to edit in every frame. For reproducibility, it maintains results for multiple iterations. Realization of these goals comes from the following process. After selecting a single frame, the user watches a speedy propagation of the initial selection with minor nudges where the selection misses its mark. This allows the user to “mold” the selection in certain frames while the tool is propagating the fixes to neighboring frames. It has a simple interface, minimal preprocessing, and minimal user input. It takes in any sort of film and exploits the spatial-temporal coherence of the object to be segmented. The tool allows artistic freedom without demanding intensive sequential processing. This thesis includes three specific extensions to Intelligent Scissors for application to video: 1. Leapfrogging, a robust method to propagate a user's single-frame selection over multiple frames by snapping each selection to its neighboring frame. 2. Histogram snapping, a method for training each frame's cost map based on previous user selections by measuring proximity to pixels in a training set and snapping to the most similar pixel's cost. 3. A real-time feedback and correction loop that provides an intuitive interface for a user to watch and control the selection propagation, with which input the algorithm updates the training data.

This thesis presents a method for interactive part selection for mesh and point set surface models that combines scribble-based selection methods with hierarchically accelerated graph-cut segmentation. Using graph-cut segmentation to determine optimal intuitive part boundaries enables easy part selection on complex geometries and allows for a simple, scribble-based interface that focuses on selecting within visible parts instead of precisely defining part boundaries that may be in difficult or occluded regions. Hierarchical acceleration is used to maintain interactive speeds with large models and to determine connectivity when extending the technique to point set models.

Historical topographic maps are valuable resources for studying past landscapes, but two-dimensional cartographic features are unsuitable for geospatial analysis. They must be extracted and converted into digital formats. This has been accomplished by researchers using sophisticated image processing and pattern recognition techniques, and more recently, artificial intelligence. While these methods are sometimes successful, they require a high level of technical expertise, limiting their accessibility. This research presents a straightforward method practitioners can use to create digital representations of historical topographic data within commercially available Geographic Information Systems (GIS) software. This study uses convolutional neural networks to extract elevation contour lines from a 1940 United States Geological Survey (USGS) topographic map in Sevier County, TN, ultimately producing a Digital Elevation Model (DEM). The topographically derived DEM (TOPO-DEM) is compared to a modern LiDAR-derived DEM to analyze its quality and utility. GIS-capable historians, archaeologists, geographers, and others can use this method in their research and land management practices.

Over the years, cancer has increasingly become a global health problem [12]. For successful treatment, early detection and diagnosis is critical. Radiomics is the use of CT, PET, MRI or Ultrasound imaging as input data, extracting features from image-based data, and then using machine learning for quantitative analysis and disease prediction [23, 14, 19, 1]. Feature reduction is critical as most quantitative features can have unnecessary redundant characteristics. The objective of this research is to use machine learning techniques in reducing the number of dimensions, thereby rendering the data manageable. Radiomics steps include Imaging, segmentation, feature extraction, and analysis. For this research, a large-scale CT data for Lung cancer diagnosis collected by scholars from Medical University in China is used to illustrate the dimension reduction techniques via R, SAS, and Python softwares. The proposed reduction and analysis techniques were PCA, Clustering, and Manifold-based algorithms. The results indicated the texture-based features

Geo-specific road database development is important to a driving simulation system and a very labor intensive process. Road databases for driving simulation need high resolution and accuracy. Even though commercial software is available on the market, a lot of manual work still has to be done when the road crosssectional profile is not uniform. This research deals with geo-specific road databases development, especially for roads with non-uniform cross sections. In this research, the United States Geographical Survey (USGS) road information is used with aerial photos to accurately extract road boundaries, using image segmentation and data compression techniques. Image segmentation plays an important role in extracting road boundary information. There are numerous methods developed for image segmentation. Six methods have been tried for the purpose of road image segmentation. The major problems with road segmentation are due to the large variety of road appearances and the many linear features in roads. A method that does not require a database of sample images is desired. Furthermore, this method should be able to handle the complexity of road appearances. The proposed method for road segmentation is based on the mean-shift clustering algorithm and it yields a high accuracy. In the phase of building road databases and visual databases based on road segmentation results, the Linde-Buzo-Gray (LBG) vector quantization algorithm is used to identify repeatable cross section profiles. In the phase of texture mapping, five major uniform textures are considered - pavement, white marker, yellow marker, concrete and grass. They are automatically mapped to polygons. In the chapter of results, snapshots of road/visual database are presented.